diff --git a/.github/PULL_REQUEST_TEMPLATE.md b/.github/PULL_REQUEST_TEMPLATE.md index e045170561d..3dcce68ab46 100644 --- a/.github/PULL_REQUEST_TEMPLATE.md +++ b/.github/PULL_REQUEST_TEMPLATE.md @@ -59,6 +59,9 @@ At a minimum, the following information should be added (but add more as needed) - [ ] Exclude: All with TSAN, MSAN, UBSAN, Coverage - [ ] Exclude: All with aarch64, release, debug --- +- [ ] Run only fuzzers related jobs (libFuzzer fuzzers, AST fuzzers, etc.) +- [ ] Exclude: AST fuzzers +--- - [ ] Do not test - [ ] Woolen Wolfdog - [ ] Upload binaries for special builds diff --git a/.github/actions/debug/action.yml b/.github/actions/debug/action.yml new file mode 100644 index 00000000000..e1fe3f28024 --- /dev/null +++ b/.github/actions/debug/action.yml @@ -0,0 +1,18 @@ +name: DebugInfo +description: Prints workflow debug info + +runs: + using: "composite" + steps: + - name: Print envs + shell: bash + run: | + echo "::group::Envs" + env + echo "::endgroup::" + - name: Print Event.json + shell: bash + run: | + echo "::group::Event.json" + python3 -m json.tool "$GITHUB_EVENT_PATH" + echo "::endgroup::" diff --git a/.github/workflows/auto_releases.yml b/.github/workflows/auto_releases.yml new file mode 100644 index 00000000000..c159907187c --- /dev/null +++ b/.github/workflows/auto_releases.yml @@ -0,0 +1,109 @@ +name: AutoReleases + +env: + PYTHONUNBUFFERED: 1 + +concurrency: + group: autoreleases + +on: + # schedule: + # - cron: '0 9 * * *' + workflow_dispatch: + inputs: + dry-run: + description: 'Dry run' + required: false + default: true + type: boolean + +jobs: + AutoReleaseInfo: + runs-on: [self-hosted, style-checker-aarch64] + outputs: + data: ${{ steps.info.outputs.AUTO_RELEASE_PARAMS }} + dry_run: ${{ steps.info.outputs.DRY_RUN }} + steps: + - name: Debug Info + uses: ./.github/actions/debug + - name: Set envs + run: | + cat >> "$GITHUB_ENV" << 'EOF' + ROBOT_CLICKHOUSE_SSH_KEY<> "$GITHUB_ENV" + - name: Check out repository code + uses: ClickHouse/checkout@v1 + - name: Prepare Info + id: info + run: | + cd "$GITHUB_WORKSPACE/tests/ci" + python3 auto_release.py --prepare + echo "::group::Auto Release Info" + python3 -m json.tool /tmp/autorelease_info.json + echo "::endgroup::" + { + echo 'AUTO_RELEASE_PARAMS<> "$GITHUB_ENV" + { + echo 'AUTO_RELEASE_PARAMS<> "$GITHUB_OUTPUT" + echo "DRY_RUN=true" >> "$GITHUB_OUTPUT" + - name: Post Release Branch statuses + run: | + cd "$GITHUB_WORKSPACE/tests/ci" + python3 auto_release.py --post-status + - name: Clean up + uses: ./.github/actions/clean + + Release_0: + needs: AutoReleaseInfo + name: Release ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[0].release_branch }} + if: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[0] && fromJson(needs.AutoReleaseInfo.outputs.data).releases[0].ready }} + uses: ./.github/workflows/create_release.yml + with: + ref: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[0].commit_sha }} + type: patch + dry-run: ${{ needs.AutoReleaseInfo.outputs.dry_run }} +# +# Release_1: +# needs: [AutoReleaseInfo, Release_0] +# name: Release ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[1].release_branch }} +# if: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[1] && fromJson(needs.AutoReleaseInfo.outputs.data).releases[1].ready }} +# uses: ./.github/workflows/create_release.yml +# with: +# ref: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[1].commit_sha }} +# type: patch +# dry-run: ${{ env.DRY_RUN }} +# +# Release_2: +# needs: [AutoReleaseInfo, Release_1] +# name: Release ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[2].release_branch }} +# if: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[0] && fromJson(needs.AutoReleaseInfo.outputs.data).releases[2].ready }} +# uses: ./.github/workflow/create_release.yml +# with: +# ref: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[0].commit_sha }} +# type: patch +# dry-run: ${{ env.DRY_RUN }} +# +# Release_3: +# needs: [AutoReleaseInfo, Release_2] +# name: Release ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[3].release_branch }} +# if: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[3] && fromJson(needs.AutoReleaseInfo.outputs.data).releases[3].ready }} +# uses: ./.github/workflow/create_release.yml +# with: +# ref: ${{ fromJson(needs.AutoReleaseInfo.outputs.data).releases[3].commit_sha }} +# type: patch +# dry-run: ${{ env.DRY_RUN }} + +# - name: Post Slack Message +# if: ${{ !cancelled() }} +# run: | +# cd "$GITHUB_WORKSPACE/tests/ci" +# python3 auto_release.py --post-auto-release-complete --wf-status ${{ job.status }} diff --git a/.github/workflows/backport_branches.yml b/.github/workflows/backport_branches.yml index 322946ac77b..23744dc7f8f 100644 --- a/.github/workflows/backport_branches.yml +++ b/.github/workflows/backport_branches.yml @@ -260,13 +260,18 @@ jobs: - name: Finish label if: ${{ !failure() }} run: | + export WORKFLOW_RESULT_FILE="/tmp/workflow_results.json" + cat > "$WORKFLOW_RESULT_FILE" << 'EOF' + ${{ toJson(needs) }} + EOF cd "$GITHUB_WORKSPACE/tests/ci" # update mergeable check - python3 merge_pr.py --set-ci-status --wf-status ${{ contains(needs.*.result, 'failure') && 'failure' || 'success' }} + python3 merge_pr.py --set-ci-status # update overall ci report python3 finish_check.py --wf-status ${{ contains(needs.*.result, 'failure') && 'failure' || 'success' }} python3 merge_pr.py - name: Check Workflow results + if: ${{ !cancelled() }} run: | export WORKFLOW_RESULT_FILE="/tmp/workflow_results.json" cat > "$WORKFLOW_RESULT_FILE" << 'EOF' diff --git a/.github/workflows/create_release.yml b/.github/workflows/create_release.yml index d4993b373df..1553d689227 100644 --- a/.github/workflows/create_release.yml +++ b/.github/workflows/create_release.yml @@ -2,6 +2,7 @@ name: CreateRelease concurrency: group: release + 'on': workflow_dispatch: inputs: @@ -26,6 +27,26 @@ concurrency: required: false default: false type: boolean + workflow_call: + 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: string + only-repo: + description: 'Run only repos updates including docker (repo-recovery, tests)' + required: false + default: false + type: boolean + dry-run: + description: 'Dry run' + required: false + default: false + type: boolean jobs: CreateRelease: @@ -101,6 +122,7 @@ jobs: --volume=".:/wd" --workdir="/wd" \ clickhouse/style-test \ ./tests/ci/changelog.py -v --debug-helpers \ + --gh-user-or-token ${{ secrets.ROBOT_CLICKHOUSE_COMMIT_TOKEN }} \ --jobs=5 \ --output="./docs/changelogs/${{ env.RELEASE_TAG }}.md" ${{ env.RELEASE_TAG }} git add ./docs/changelogs/${{ env.RELEASE_TAG }}.md @@ -129,9 +151,9 @@ jobs: if: ${{ inputs.type == 'patch' && ! inputs.only-repo }} shell: bash run: | - python3 ./tests/ci/create_release.py --set-progress-completed git reset --hard HEAD git checkout "$GITHUB_REF_NAME" + python3 ./tests/ci/create_release.py --set-progress-completed - name: Create GH Release if: ${{ inputs.type == 'patch' && ! inputs.only-repo }} shell: bash diff --git a/.github/workflows/jepsen.yml b/.github/workflows/jepsen.yml index 035ba2e5b98..ecafde9e4cb 100644 --- a/.github/workflows/jepsen.yml +++ b/.github/workflows/jepsen.yml @@ -64,9 +64,10 @@ jobs: - name: Check out repository code uses: ClickHouse/checkout@v1 - name: Check Workflow results + if: ${{ !cancelled() }} run: | export WORKFLOW_RESULT_FILE="/tmp/workflow_results.json" - cat >> "$WORKFLOW_RESULT_FILE" << 'EOF' + 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 64083668719..629cf79770e 100644 --- a/.github/workflows/merge_queue.yml +++ b/.github/workflows/merge_queue.yml @@ -103,9 +103,14 @@ jobs: - name: Check and set merge status if: ${{ needs.StyleCheck.result == 'success' }} run: | + export WORKFLOW_RESULT_FILE="/tmp/workflow_results.json" + cat > "$WORKFLOW_RESULT_FILE" << 'EOF' + ${{ toJson(needs) }} + EOF cd "$GITHUB_WORKSPACE/tests/ci" - python3 merge_pr.py --set-ci-status --wf-status ${{ contains(needs.*.result, 'failure') && 'failure' || 'success' }} + python3 merge_pr.py --set-ci-status - name: Check Workflow results + if: ${{ !cancelled() }} run: | export WORKFLOW_RESULT_FILE="/tmp/workflow_results.json" cat > "$WORKFLOW_RESULT_FILE" << 'EOF' diff --git a/.github/workflows/nightly.yml b/.github/workflows/nightly.yml index ea9c125db70..36fea39686f 100644 --- a/.github/workflows/nightly.yml +++ b/.github/workflows/nightly.yml @@ -52,6 +52,7 @@ jobs: - name: Check out repository code uses: ClickHouse/checkout@v1 - name: Check Workflow results + if: ${{ !cancelled() }} run: | export WORKFLOW_RESULT_FILE="/tmp/workflow_results.json" cat > "$WORKFLOW_RESULT_FILE" << 'EOF' diff --git a/.github/workflows/pull_request.yml b/.github/workflows/pull_request.yml index 071f0f1e20a..dbc740ebc1b 100644 --- a/.github/workflows/pull_request.yml +++ b/.github/workflows/pull_request.yml @@ -170,7 +170,11 @@ jobs: if: ${{ needs.StyleCheck.result == 'success' }} run: | cd "$GITHUB_WORKSPACE/tests/ci" - python3 merge_pr.py --set-ci-status --wf-status ${{ contains(needs.*.result, 'failure') && 'failure' || 'success' }} + export WORKFLOW_RESULT_FILE="/tmp/workflow_results.json" + cat > "$WORKFLOW_RESULT_FILE" << 'EOF' + ${{ toJson(needs) }} + EOF + python3 merge_pr.py --set-ci-status - name: Check Workflow results uses: ./.github/actions/check_workflow with: diff --git a/.github/workflows/release_branches.yml b/.github/workflows/release_branches.yml index b79208b03a6..a5cd6321e8c 100644 --- a/.github/workflows/release_branches.yml +++ b/.github/workflows/release_branches.yml @@ -481,12 +481,10 @@ jobs: - 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 + if: ${{ !cancelled() }} run: | export WORKFLOW_RESULT_FILE="/tmp/workflow_results.json" cat > "$WORKFLOW_RESULT_FILE" << 'EOF' diff --git a/.gitmodules b/.gitmodules index 7e0b4df4ad1..cdee6a43ad8 100644 --- a/.gitmodules +++ b/.gitmodules @@ -230,9 +230,6 @@ [submodule "contrib/minizip-ng"] path = contrib/minizip-ng url = https://github.com/zlib-ng/minizip-ng -[submodule "contrib/annoy"] - path = contrib/annoy - url = https://github.com/ClickHouse/annoy [submodule "contrib/qpl"] path = contrib/qpl url = https://github.com/intel/qpl @@ -341,16 +338,13 @@ url = https://github.com/graphitemaster/incbin.git [submodule "contrib/usearch"] path = contrib/usearch - url = https://github.com/unum-cloud/usearch.git + url = https://github.com/ClickHouse/usearch.git [submodule "contrib/SimSIMD"] path = contrib/SimSIMD url = https://github.com/ashvardanian/SimSIMD.git [submodule "contrib/FP16"] path = contrib/FP16 url = https://github.com/Maratyszcza/FP16.git -[submodule "contrib/robin-map"] - path = contrib/robin-map - url = https://github.com/Tessil/robin-map.git [submodule "contrib/aklomp-base64"] path = contrib/aklomp-base64 url = https://github.com/aklomp/base64.git diff --git a/CHANGELOG.md b/CHANGELOG.md index 5cd4200d9ad..1793fd14ccd 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -64,6 +64,7 @@ * 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)). * 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)). diff --git a/CMakeLists.txt b/CMakeLists.txt index f796e6c4616..6fa91fa002f 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -187,14 +187,6 @@ else () set(NO_WHOLE_ARCHIVE --no-whole-archive) endif () -if (NOT CMAKE_BUILD_TYPE_UC STREQUAL "RELEASE") - # Can be lld or ld-lld or lld-13 or /path/to/lld. - if (LINKER_NAME MATCHES "lld") - set (CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -Wl,--gdb-index") - message (STATUS "Adding .gdb-index via --gdb-index linker option.") - endif () -endif() - if (NOT (SANITIZE_COVERAGE OR WITH_COVERAGE) AND (CMAKE_BUILD_TYPE_UC STREQUAL "RELEASE" OR CMAKE_BUILD_TYPE_UC STREQUAL "RELWITHDEBINFO" @@ -330,17 +322,21 @@ if (DISABLE_OMIT_FRAME_POINTER) set (CMAKE_ASM_FLAGS_ADD "${CMAKE_ASM_FLAGS_ADD} -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer") endif() +# Before you start hating your debugger because it refuses to show variables (''), try building with -DDEBUG_O_LEVEL="0" +# https://stackoverflow.com/questions/63386189/whats-the-difference-between-a-compilers-o0-option-and-og-option/63386263#63386263 +set(DEBUG_O_LEVEL "g" CACHE STRING "The -Ox level used for debug builds") + 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} -Og ${DEBUG_INFO_FLAGS} ${CMAKE_CXX_FLAGS_ADD}") +set (CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} -O${DEBUG_O_LEVEL} ${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} -Og ${DEBUG_INFO_FLAGS} ${CMAKE_C_FLAGS_ADD}") +set (CMAKE_C_FLAGS_DEBUG "${CMAKE_C_FLAGS_DEBUG} -O${DEBUG_O_LEVEL} ${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} -Og ${DEBUG_INFO_FLAGS} ${CMAKE_ASM_FLAGS_ADD}") +set (CMAKE_ASM_FLAGS_DEBUG "${CMAKE_ASM_FLAGS_DEBUG} -O${DEBUG_O_LEVEL} ${DEBUG_INFO_FLAGS} ${CMAKE_ASM_FLAGS_ADD}") if (OS_DARWIN) set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -stdlib=libc++") @@ -402,7 +398,7 @@ if ((NOT OS_LINUX AND NOT OS_ANDROID) OR (CMAKE_BUILD_TYPE_UC STREQUAL "DEBUG") set(ENABLE_GWP_ASAN OFF) endif () -option (ENABLE_FIU "Enable Fiu" ON) +option (ENABLE_LIBFIU "Enable libfiu" ON) option(WERROR "Enable -Werror compiler option" ON) @@ -428,12 +424,17 @@ if (NOT SANITIZE) set (CMAKE_POSITION_INDEPENDENT_CODE OFF) endif() -if (OS_LINUX AND NOT (ARCH_AARCH64 OR ARCH_S390X) AND NOT SANITIZE) - # Slightly more efficient code can be generated - # It's disabled for ARM because otherwise ClickHouse cannot run on Android. +if (NOT OS_ANDROID AND OS_LINUX AND NOT ARCH_S390X AND NOT SANITIZE) + # Using '-no-pie' builds executables with fixed addresses, resulting in slightly more efficient code + # and keeping binary addresses constant even with ASLR enabled. + # Disabled on Android as it requires PIE: https://source.android.com/docs/security/enhancements#android-5 + # Disabled on IBM S390X due to build issues with 'no-pie' + # Disabled with sanitizers to avoid issues with maximum relocation size: https://github.com/ClickHouse/ClickHouse/pull/49145 set (CMAKE_CXX_FLAGS_RELWITHDEBINFO "${CMAKE_CXX_FLAGS_RELWITHDEBINFO} -fno-pie") set (CMAKE_C_FLAGS_RELWITHDEBINFO "${CMAKE_C_FLAGS_RELWITHDEBINFO} -fno-pie") set (CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -no-pie -Wl,-no-pie") +else () + message (WARNING "ClickHouse is built as PIE, system.trace_log will contain invalid addresses after server restart.") endif () if (ENABLE_TESTS) diff --git a/base/base/CMakeLists.txt b/base/base/CMakeLists.txt index 341c92d3042..247028b96e0 100644 --- a/base/base/CMakeLists.txt +++ b/base/base/CMakeLists.txt @@ -1,4 +1,4 @@ -add_compile_options($<$,$>:${COVERAGE_FLAGS}>) +add_compile_options("$<$,$>:${COVERAGE_FLAGS}>") if (USE_CLANG_TIDY) set (CMAKE_CXX_CLANG_TIDY "${CLANG_TIDY_PATH}") diff --git a/base/base/cgroupsv2.cpp b/base/base/cgroupsv2.cpp index 87f62bf377d..e0e37c8729b 100644 --- a/base/base/cgroupsv2.cpp +++ b/base/base/cgroupsv2.cpp @@ -27,27 +27,6 @@ bool cgroupsV2Enabled() #endif } -bool cgroupsV2MemoryControllerEnabled() -{ -#if defined(OS_LINUX) - chassert(cgroupsV2Enabled()); - /// 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. - 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; - std::string controllers; - std::getline(controllers_file, controllers); - return controllers.find("memory") != std::string::npos; -#else - return false; -#endif -} - fs::path cgroupV2PathOfProcess() { #if defined(OS_LINUX) @@ -71,3 +50,28 @@ fs::path cgroupV2PathOfProcess() return {}; #endif } + +std::optional getCgroupsV2PathContainingFile([[maybe_unused]] std::string_view file_name) +{ +#if defined(OS_LINUX) + if (!cgroupsV2Enabled()) + return {}; + + fs::path current_cgroup = cgroupV2PathOfProcess(); + if (current_cgroup.empty()) + return {}; + + /// Return the bottom-most nested file. If there is no such file at the current + /// level, try again at the parent level as settings are inherited. + while (current_cgroup != default_cgroups_mount.parent_path()) + { + const auto path = current_cgroup / file_name; + if (fs::exists(path)) + return {current_cgroup}; + current_cgroup = current_cgroup.parent_path(); + } + return {}; +#else + return {}; +#endif +} diff --git a/base/base/cgroupsv2.h b/base/base/cgroupsv2.h index cfb916ff358..a6276474254 100644 --- a/base/base/cgroupsv2.h +++ b/base/base/cgroupsv2.h @@ -1,6 +1,7 @@ #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). @@ -11,11 +12,11 @@ static inline const std::filesystem::path default_cgroups_mount = "/sys/fs/cgrou /// Is cgroups v2 enabled on the system? bool cgroupsV2Enabled(); -/// Is the memory controller of cgroups v2 enabled on the system? -/// Assumes that cgroupsV2Enabled() is enabled. -bool cgroupsV2MemoryControllerEnabled(); - /// 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::filesystem::path cgroupV2PathOfProcess(); + +/// Returns the most nested cgroup dir containing the specified file. +/// If cgroups v2 is not enabled - returns an empty optional. +std::optional getCgroupsV2PathContainingFile([[maybe_unused]] std::string_view file_name); diff --git a/base/base/getMemoryAmount.cpp b/base/base/getMemoryAmount.cpp index 03aab1eac72..bbfbecdbffd 100644 --- a/base/base/getMemoryAmount.cpp +++ b/base/base/getMemoryAmount.cpp @@ -19,9 +19,6 @@ std::optional getCgroupsV2MemoryLimit() if (!cgroupsV2Enabled()) return {}; - if (!cgroupsV2MemoryControllerEnabled()) - return {}; - std::filesystem::path current_cgroup = cgroupV2PathOfProcess(); if (current_cgroup.empty()) return {}; diff --git a/base/poco/Foundation/include/Poco/ErrorHandler.h b/base/poco/Foundation/include/Poco/ErrorHandler.h index c0b5bf9621e..f774f2ccf5e 100644 --- a/base/poco/Foundation/include/Poco/ErrorHandler.h +++ b/base/poco/Foundation/include/Poco/ErrorHandler.h @@ -21,6 +21,7 @@ #include "Poco/Exception.h" #include "Poco/Foundation.h" #include "Poco/Mutex.h" +#include "Poco/Message.h" namespace Poco @@ -78,6 +79,10 @@ public: /// /// The default implementation just breaks into the debugger. + virtual void logMessageImpl(Message::Priority priority, const std::string & msg) {} + /// Write a messages to the log + /// Useful for logging from Poco + static void handle(const Exception & exc); /// Invokes the currently registered ErrorHandler. @@ -87,6 +92,9 @@ public: static void handle(); /// Invokes the currently registered ErrorHandler. + static void logMessage(Message::Priority priority, const std::string & msg); + /// Invokes the currently registered ErrorHandler to log a message. + static ErrorHandler * set(ErrorHandler * pHandler); /// Registers the given handler as the current error handler. /// diff --git a/base/poco/Foundation/src/ErrorHandler.cpp b/base/poco/Foundation/src/ErrorHandler.cpp index d0af8ea8a12..1d0a41b77ec 100644 --- a/base/poco/Foundation/src/ErrorHandler.cpp +++ b/base/poco/Foundation/src/ErrorHandler.cpp @@ -8,7 +8,7 @@ // Copyright (c) 2005-2006, Applied Informatics Software Engineering GmbH. // and Contributors. // -// SPDX-License-Identifier: BSL-1.0 +// SPDX-License-Identifier: BSL-1.0 // @@ -35,79 +35,91 @@ ErrorHandler::~ErrorHandler() void ErrorHandler::exception(const Exception& exc) { - poco_debugger_msg(exc.what()); + poco_debugger_msg(exc.what()); } - + void ErrorHandler::exception(const std::exception& exc) { - poco_debugger_msg(exc.what()); + poco_debugger_msg(exc.what()); } void ErrorHandler::exception() { - poco_debugger_msg("unknown exception"); + poco_debugger_msg("unknown exception"); } void ErrorHandler::handle(const Exception& exc) { - FastMutex::ScopedLock lock(_mutex); - try - { - _pHandler->exception(exc); - } - catch (...) - { - } + FastMutex::ScopedLock lock(_mutex); + try + { + _pHandler->exception(exc); + } + catch (...) + { + } } - + void ErrorHandler::handle(const std::exception& exc) { - FastMutex::ScopedLock lock(_mutex); - try - { - _pHandler->exception(exc); - } - catch (...) - { - } + FastMutex::ScopedLock lock(_mutex); + try + { + _pHandler->exception(exc); + } + catch (...) + { + } } void ErrorHandler::handle() { - FastMutex::ScopedLock lock(_mutex); - try - { - _pHandler->exception(); - } - catch (...) - { - } + FastMutex::ScopedLock lock(_mutex); + try + { + _pHandler->exception(); + } + catch (...) + { + } +} + +void ErrorHandler::logMessage(Message::Priority priority, const std::string & msg) +{ + FastMutex::ScopedLock lock(_mutex); + try + { + _pHandler->logMessageImpl(priority, msg); + } + catch (...) + { + } } ErrorHandler* ErrorHandler::set(ErrorHandler* pHandler) { - poco_check_ptr(pHandler); + poco_check_ptr(pHandler); - FastMutex::ScopedLock lock(_mutex); - ErrorHandler* pOld = _pHandler; - _pHandler = pHandler; - return pOld; + FastMutex::ScopedLock lock(_mutex); + ErrorHandler* pOld = _pHandler; + _pHandler = pHandler; + return pOld; } ErrorHandler* ErrorHandler::defaultHandler() { - // NOTE: Since this is called to initialize the static _pHandler - // variable, sh has to be a local static, otherwise we run - // into static initialization order issues. - static SingletonHolder sh; - return sh.get(); + // NOTE: Since this is called to initialize the static _pHandler + // variable, sh has to be a local static, otherwise we run + // into static initialization order issues. + static SingletonHolder sh; + return sh.get(); } diff --git a/base/poco/Net/include/Poco/Net/HTTPServerSession.h b/base/poco/Net/include/Poco/Net/HTTPServerSession.h index 3df7995509a..b0659ca405c 100644 --- a/base/poco/Net/include/Poco/Net/HTTPServerSession.h +++ b/base/poco/Net/include/Poco/Net/HTTPServerSession.h @@ -58,6 +58,10 @@ namespace Net void setKeepAliveTimeout(Poco::Timespan keepAliveTimeout); + size_t getKeepAliveTimeout() const { return _keepAliveTimeout.totalSeconds(); } + + size_t getMaxKeepAliveRequests() const { return _maxKeepAliveRequests; } + private: bool _firstRequest; Poco::Timespan _keepAliveTimeout; diff --git a/base/poco/Net/src/HTTPServerSession.cpp b/base/poco/Net/src/HTTPServerSession.cpp index f67a63a9e0e..8eec3e14872 100644 --- a/base/poco/Net/src/HTTPServerSession.cpp +++ b/base/poco/Net/src/HTTPServerSession.cpp @@ -19,11 +19,11 @@ namespace Poco { namespace Net { -HTTPServerSession::HTTPServerSession(const StreamSocket& socket, HTTPServerParams::Ptr pParams): - HTTPSession(socket, pParams->getKeepAlive()), - _firstRequest(true), - _keepAliveTimeout(pParams->getKeepAliveTimeout()), - _maxKeepAliveRequests(pParams->getMaxKeepAliveRequests()) +HTTPServerSession::HTTPServerSession(const StreamSocket & socket, HTTPServerParams::Ptr pParams) + : HTTPSession(socket, pParams->getKeepAlive()) + , _firstRequest(true) + , _keepAliveTimeout(pParams->getKeepAliveTimeout()) + , _maxKeepAliveRequests(pParams->getMaxKeepAliveRequests()) { setTimeout(pParams->getTimeout()); } @@ -52,11 +52,12 @@ bool HTTPServerSession::hasMoreRequests() } else if (_maxKeepAliveRequests != 0 && getKeepAlive()) { - if (_maxKeepAliveRequests > 0) - --_maxKeepAliveRequests; - return buffered() > 0 || socket().poll(_keepAliveTimeout, Socket::SELECT_READ); - } - else return false; + if (_maxKeepAliveRequests > 0) + --_maxKeepAliveRequests; + return buffered() > 0 || socket().poll(_keepAliveTimeout, Socket::SELECT_READ); + } + else + return false; } diff --git a/base/poco/Net/src/SocketImpl.cpp b/base/poco/Net/src/SocketImpl.cpp index 484b8cfeec3..13a655d153d 100644 --- a/base/poco/Net/src/SocketImpl.cpp +++ b/base/poco/Net/src/SocketImpl.cpp @@ -17,6 +17,7 @@ #include "Poco/Net/StreamSocketImpl.h" #include "Poco/NumberFormatter.h" #include "Poco/Timestamp.h" +#include "Poco/ErrorHandler.h" #include // FD_SET needs memset on some platforms, so we can't use diff --git a/base/poco/Net/src/TCPServer.cpp b/base/poco/Net/src/TCPServer.cpp index 9bdae900bd6..b957829fb7d 100644 --- a/base/poco/Net/src/TCPServer.cpp +++ b/base/poco/Net/src/TCPServer.cpp @@ -8,7 +8,7 @@ // Copyright (c) 2005-2006, Applied Informatics Software Engineering GmbH. // and Contributors. // -// SPDX-License-Identifier: BSL-1.0 +// SPDX-License-Identifier: BSL-1.0 // @@ -44,190 +44,194 @@ TCPServerConnectionFilter::~TCPServerConnectionFilter() TCPServer::TCPServer(TCPServerConnectionFactory::Ptr pFactory, Poco::UInt16 portNumber, TCPServerParams::Ptr pParams): - _socket(ServerSocket(portNumber)), - _thread(threadName(_socket)), - _stopped(true) -{ - Poco::ThreadPool& pool = Poco::ThreadPool::defaultPool(); - if (pParams) - { - int toAdd = pParams->getMaxThreads() - pool.capacity(); - if (toAdd > 0) pool.addCapacity(toAdd); - } - _pDispatcher = new TCPServerDispatcher(pFactory, pool, pParams); - + _socket(ServerSocket(portNumber)), + _thread(threadName(_socket)), + _stopped(true) +{ + Poco::ThreadPool& pool = Poco::ThreadPool::defaultPool(); + if (pParams) + { + int toAdd = pParams->getMaxThreads() - pool.capacity(); + if (toAdd > 0) pool.addCapacity(toAdd); + } + _pDispatcher = new TCPServerDispatcher(pFactory, pool, pParams); + } TCPServer::TCPServer(TCPServerConnectionFactory::Ptr pFactory, const ServerSocket& socket, TCPServerParams::Ptr pParams): - _socket(socket), - _thread(threadName(socket)), - _stopped(true) + _socket(socket), + _thread(threadName(socket)), + _stopped(true) { - Poco::ThreadPool& pool = Poco::ThreadPool::defaultPool(); - if (pParams) - { - int toAdd = pParams->getMaxThreads() - pool.capacity(); - if (toAdd > 0) pool.addCapacity(toAdd); - } - _pDispatcher = new TCPServerDispatcher(pFactory, pool, pParams); + Poco::ThreadPool& pool = Poco::ThreadPool::defaultPool(); + if (pParams) + { + int toAdd = pParams->getMaxThreads() - pool.capacity(); + if (toAdd > 0) pool.addCapacity(toAdd); + } + _pDispatcher = new TCPServerDispatcher(pFactory, pool, pParams); } TCPServer::TCPServer(TCPServerConnectionFactory::Ptr pFactory, Poco::ThreadPool& threadPool, const ServerSocket& socket, TCPServerParams::Ptr pParams): - _socket(socket), - _pDispatcher(new TCPServerDispatcher(pFactory, threadPool, pParams)), - _thread(threadName(socket)), - _stopped(true) + _socket(socket), + _pDispatcher(new TCPServerDispatcher(pFactory, threadPool, pParams)), + _thread(threadName(socket)), + _stopped(true) { } TCPServer::~TCPServer() { - try - { - stop(); - _pDispatcher->release(); - } - catch (...) - { - poco_unexpected(); - } + try + { + stop(); + _pDispatcher->release(); + } + catch (...) + { + poco_unexpected(); + } } const TCPServerParams& TCPServer::params() const { - return _pDispatcher->params(); + return _pDispatcher->params(); } void TCPServer::start() { - poco_assert (_stopped); + poco_assert (_stopped); - _stopped = false; - _thread.start(*this); + _stopped = false; + _thread.start(*this); } - + void TCPServer::stop() { - if (!_stopped) - { - _stopped = true; - _thread.join(); - _pDispatcher->stop(); - } + if (!_stopped) + { + _stopped = true; + _thread.join(); + _pDispatcher->stop(); + } } void TCPServer::run() { - while (!_stopped) - { - Poco::Timespan timeout(250000); - try - { - if (_socket.poll(timeout, Socket::SELECT_READ)) - { - try - { - StreamSocket ss = _socket.acceptConnection(); - - if (!_pConnectionFilter || _pConnectionFilter->accept(ss)) - { - // enable nodelay per default: OSX really needs that + while (!_stopped) + { + Poco::Timespan timeout(250000); + try + { + if (_socket.poll(timeout, Socket::SELECT_READ)) + { + try + { + StreamSocket ss = _socket.acceptConnection(); + + if (!_pConnectionFilter || _pConnectionFilter->accept(ss)) + { + // enable nodelay per default: OSX really needs that #if defined(POCO_OS_FAMILY_UNIX) - if (ss.address().family() != AddressFamily::UNIX_LOCAL) + if (ss.address().family() != AddressFamily::UNIX_LOCAL) #endif - { - ss.setNoDelay(true); - } - _pDispatcher->enqueue(ss); - } - } - catch (Poco::Exception& exc) - { - ErrorHandler::handle(exc); - } - catch (std::exception& exc) - { - ErrorHandler::handle(exc); - } - catch (...) - { - ErrorHandler::handle(); - } - } - } - catch (Poco::Exception& exc) - { - ErrorHandler::handle(exc); - // possibly a resource issue since poll() failed; - // give some time to recover before trying again - Poco::Thread::sleep(50); - } - } + { + ss.setNoDelay(true); + } + _pDispatcher->enqueue(ss); + } + else + { + ErrorHandler::logMessage(Message::PRIO_WARNING, "Filtered out connection from " + ss.peerAddress().toString()); + } + } + catch (Poco::Exception& exc) + { + ErrorHandler::handle(exc); + } + catch (std::exception& exc) + { + ErrorHandler::handle(exc); + } + catch (...) + { + ErrorHandler::handle(); + } + } + } + catch (Poco::Exception& exc) + { + ErrorHandler::handle(exc); + // possibly a resource issue since poll() failed; + // give some time to recover before trying again + Poco::Thread::sleep(50); + } + } } int TCPServer::currentThreads() const { - return _pDispatcher->currentThreads(); + return _pDispatcher->currentThreads(); } int TCPServer::maxThreads() const { - return _pDispatcher->maxThreads(); + return _pDispatcher->maxThreads(); } - + int TCPServer::totalConnections() const { - return _pDispatcher->totalConnections(); + return _pDispatcher->totalConnections(); } int TCPServer::currentConnections() const { - return _pDispatcher->currentConnections(); + return _pDispatcher->currentConnections(); } int TCPServer::maxConcurrentConnections() const { - return _pDispatcher->maxConcurrentConnections(); + return _pDispatcher->maxConcurrentConnections(); } - + int TCPServer::queuedConnections() const { - return _pDispatcher->queuedConnections(); + return _pDispatcher->queuedConnections(); } int TCPServer::refusedConnections() const { - return _pDispatcher->refusedConnections(); + return _pDispatcher->refusedConnections(); } void TCPServer::setConnectionFilter(const TCPServerConnectionFilter::Ptr& pConnectionFilter) { - poco_assert (_stopped); + poco_assert (_stopped); - _pConnectionFilter = pConnectionFilter; + _pConnectionFilter = pConnectionFilter; } std::string TCPServer::threadName(const ServerSocket& socket) { - std::string name("TCPServer: "); - name.append(socket.address().toString()); - return name; + std::string name("TCPServer: "); + name.append(socket.address().toString()); + return name; } diff --git a/base/poco/Net/src/TCPServerDispatcher.cpp b/base/poco/Net/src/TCPServerDispatcher.cpp index 7f9f9a20ee7..9ca7c271e63 100644 --- a/base/poco/Net/src/TCPServerDispatcher.cpp +++ b/base/poco/Net/src/TCPServerDispatcher.cpp @@ -8,7 +8,7 @@ // Copyright (c) 2005-2007, Applied Informatics Software Engineering GmbH. // and Contributors. // -// SPDX-License-Identifier: BSL-1.0 +// SPDX-License-Identifier: BSL-1.0 // @@ -33,44 +33,44 @@ namespace Net { class TCPConnectionNotification: public Notification { public: - TCPConnectionNotification(const StreamSocket& socket): - _socket(socket) - { - } - - ~TCPConnectionNotification() - { - } - - const StreamSocket& socket() const - { - return _socket; - } + TCPConnectionNotification(const StreamSocket& socket): + _socket(socket) + { + } + + ~TCPConnectionNotification() + { + } + + const StreamSocket& socket() const + { + return _socket; + } private: - StreamSocket _socket; + StreamSocket _socket; }; TCPServerDispatcher::TCPServerDispatcher(TCPServerConnectionFactory::Ptr pFactory, Poco::ThreadPool& threadPool, TCPServerParams::Ptr pParams): - _rc(1), - _pParams(pParams), - _currentThreads(0), - _totalConnections(0), - _currentConnections(0), - _maxConcurrentConnections(0), - _refusedConnections(0), - _stopped(false), - _pConnectionFactory(pFactory), - _threadPool(threadPool) + _rc(1), + _pParams(pParams), + _currentThreads(0), + _totalConnections(0), + _currentConnections(0), + _maxConcurrentConnections(0), + _refusedConnections(0), + _stopped(false), + _pConnectionFactory(pFactory), + _threadPool(threadPool) { - poco_check_ptr (pFactory); + poco_check_ptr (pFactory); - if (!_pParams) - _pParams = new TCPServerParams; - - if (_pParams->getMaxThreads() == 0) - _pParams->setMaxThreads(threadPool.capacity()); + if (!_pParams) + _pParams = new TCPServerParams; + + if (_pParams->getMaxThreads() == 0) + _pParams->setMaxThreads(threadPool.capacity()); } @@ -81,161 +81,184 @@ TCPServerDispatcher::~TCPServerDispatcher() void TCPServerDispatcher::duplicate() { - ++_rc; + ++_rc; } void TCPServerDispatcher::release() { - if (--_rc == 0) delete this; + if (--_rc == 0) delete this; } void TCPServerDispatcher::run() { - AutoPtr guard(this); // ensure object stays alive + AutoPtr guard(this); // ensure object stays alive - int idleTime = (int) _pParams->getThreadIdleTime().totalMilliseconds(); + int idleTime = (int) _pParams->getThreadIdleTime().totalMilliseconds(); - for (;;) - { - try - { - AutoPtr pNf = _queue.waitDequeueNotification(idleTime); - if (pNf && !_stopped) - { - TCPConnectionNotification* pCNf = dynamic_cast(pNf.get()); - if (pCNf) - { - beginConnection(); - if (!_stopped) - { - std::unique_ptr pConnection(_pConnectionFactory->createConnection(pCNf->socket())); - poco_check_ptr(pConnection.get()); - pConnection->start(); - } - /// endConnection() should be called after destroying TCPServerConnection, - /// otherwise currentConnections() could become zero while some connections are yet still alive. - endConnection(); - } - } - } - catch (Poco::Exception &exc) { ErrorHandler::handle(exc); } - catch (std::exception &exc) { ErrorHandler::handle(exc); } - catch (...) { ErrorHandler::handle(); } - FastMutex::ScopedLock lock(_mutex); - if (_stopped || (_currentThreads > 1 && _queue.empty())) - { - --_currentThreads; - break; - } - } + for (;;) + { + try + { + AutoPtr pNf = _queue.waitDequeueNotification(idleTime); + if (pNf && !_stopped) + { + TCPConnectionNotification* pCNf = dynamic_cast(pNf.get()); + if (pCNf) + { + beginConnection(); + if (!_stopped) + { + std::unique_ptr pConnection(_pConnectionFactory->createConnection(pCNf->socket())); + poco_check_ptr(pConnection.get()); + pConnection->start(); + } + /// endConnection() should be called after destroying TCPServerConnection, + /// otherwise currentConnections() could become zero while some connections are yet still alive. + endConnection(); + } + } + } + catch (Poco::Exception &exc) { ErrorHandler::handle(exc); } + catch (std::exception &exc) { ErrorHandler::handle(exc); } + catch (...) { ErrorHandler::handle(); } + FastMutex::ScopedLock lock(_mutex); + if (_stopped || (_currentThreads > 1 && _queue.empty())) + { + --_currentThreads; + break; + } + } } namespace { - static const std::string threadName("TCPServerConnection"); + static const std::string threadName("TCPServerConnection"); } - + void TCPServerDispatcher::enqueue(const StreamSocket& socket) { - FastMutex::ScopedLock lock(_mutex); + FastMutex::ScopedLock lock(_mutex); - if (_queue.size() < _pParams->getMaxQueued()) - { - if (!_queue.hasIdleThreads() && _currentThreads < _pParams->getMaxThreads()) - { - try - { + ErrorHandler::logMessage(Message::PRIO_TEST, "Queue size: " + std::to_string(_queue.size()) + + ", current threads: " + std::to_string(_currentThreads) + + ", threads in pool: " + std::to_string(_threadPool.allocated()) + + ", current connections: " + std::to_string(_currentConnections)); + + + if (_queue.size() < _pParams->getMaxQueued()) + { + /// NOTE: the condition below is wrong. + /// Since the thread pool is shared between multiple servers/TCPServerDispatchers, + /// _currentThreads < _pParams->getMaxThreads() will be true when the pool is actually saturated. + /// As a result, queue is useless and connections never wait in queue. + /// Instead, we (mistakenly) think that we can create a thread for this connection, but we fail to create it + /// and the connection get rejected. + /// We could check _currentThreads < _threadPool.allocated() to make it work, + /// but it's not clear if we want to make it work + /// because it may be better to reject connection immediately if we don't have resources to handle it. + if (!_queue.hasIdleThreads() && _currentThreads < _pParams->getMaxThreads()) + { + try + { this->duplicate(); - _threadPool.startWithPriority(_pParams->getThreadPriority(), *this, threadName); - ++_currentThreads; - } - catch (Poco::Exception& exc) - { + _threadPool.startWithPriority(_pParams->getThreadPriority(), *this, threadName); + ++_currentThreads; + } + catch (Poco::Exception& exc) + { + ErrorHandler::logMessage(Message::PRIO_WARNING, "Got an exception while starting thread for connection from " + + socket.peerAddress().toString()); + ErrorHandler::handle(exc); this->release(); - ++_refusedConnections; - std::cerr << "Got exception while starting thread for connection. Error code: " - << exc.code() << ", message: '" << exc.displayText() << "'" << std::endl; - return; - } - } - _queue.enqueueNotification(new TCPConnectionNotification(socket)); - } - else - { - ++_refusedConnections; - } + ++_refusedConnections; + return; + } + } + else if (!_queue.hasIdleThreads()) + { + ErrorHandler::logMessage(Message::PRIO_TRACE, "Don't have idle threads, adding connection from " + + socket.peerAddress().toString() + " to the queue, size: " + std::to_string(_queue.size())); + } + _queue.enqueueNotification(new TCPConnectionNotification(socket)); + } + else + { + ErrorHandler::logMessage(Message::PRIO_WARNING, "Refusing connection from " + socket.peerAddress().toString() + + ", reached max queue size " + std::to_string(_pParams->getMaxQueued())); + ++_refusedConnections; + } } void TCPServerDispatcher::stop() { - _stopped = true; - _queue.clear(); - _queue.wakeUpAll(); + _stopped = true; + _queue.clear(); + _queue.wakeUpAll(); } int TCPServerDispatcher::currentThreads() const { - return _currentThreads; + return _currentThreads; } int TCPServerDispatcher::maxThreads() const { - FastMutex::ScopedLock lock(_mutex); - - return _threadPool.capacity(); + FastMutex::ScopedLock lock(_mutex); + + return _threadPool.capacity(); } int TCPServerDispatcher::totalConnections() const { - return _totalConnections; + return _totalConnections; } int TCPServerDispatcher::currentConnections() const { - return _currentConnections; + return _currentConnections; } int TCPServerDispatcher::maxConcurrentConnections() const { - return _maxConcurrentConnections; + return _maxConcurrentConnections; } int TCPServerDispatcher::queuedConnections() const { - return _queue.size(); + return _queue.size(); } int TCPServerDispatcher::refusedConnections() const { - return _refusedConnections; + return _refusedConnections; } void TCPServerDispatcher::beginConnection() { - FastMutex::ScopedLock lock(_mutex); + FastMutex::ScopedLock lock(_mutex); - ++_totalConnections; - ++_currentConnections; - if (_currentConnections > _maxConcurrentConnections) - _maxConcurrentConnections.store(_currentConnections); + ++_totalConnections; + ++_currentConnections; + if (_currentConnections > _maxConcurrentConnections) + _maxConcurrentConnections.store(_currentConnections); } void TCPServerDispatcher::endConnection() { - --_currentConnections; + --_currentConnections; } diff --git a/cmake/autogenerated_versions.txt b/cmake/autogenerated_versions.txt index d69646d3694..c82038804fe 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 54489) +SET(VERSION_REVISION 54490) SET(VERSION_MAJOR 24) -SET(VERSION_MINOR 8) +SET(VERSION_MINOR 9) SET(VERSION_PATCH 1) -SET(VERSION_GITHASH 3f8b27d7accd2b5ec4afe7d0dd459115323304af) -SET(VERSION_DESCRIBE v24.8.1.1-testing) -SET(VERSION_STRING 24.8.1.1) +SET(VERSION_GITHASH e02b434d2fc0c4fbee29ca675deab7474d274608) +SET(VERSION_DESCRIBE v24.9.1.1-testing) +SET(VERSION_STRING 24.9.1.1) # end of autochange diff --git a/cmake/sanitize.cmake b/cmake/sanitize.cmake index d8d211b9329..091e20e1387 100644 --- a/cmake/sanitize.cmake +++ b/cmake/sanitize.cmake @@ -57,8 +57,8 @@ option(WITH_COVERAGE "Instrumentation for code coverage with default implementat if (WITH_COVERAGE) message (STATUS "Enabled instrumentation for code coverage") - set(COVERAGE_FLAGS "SHELL:-fprofile-instr-generate -fcoverage-mapping") - set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -fprofile-instr-generate -fcoverage-mapping") + set (COVERAGE_FLAGS -fprofile-instr-generate -fcoverage-mapping) + set (CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -fprofile-instr-generate -fcoverage-mapping") endif() option (SANITIZE_COVERAGE "Instrumentation for code coverage with custom callbacks" OFF) diff --git a/contrib/CMakeLists.txt b/contrib/CMakeLists.txt index 977efda15ff..d7489bc5c0e 100644 --- a/contrib/CMakeLists.txt +++ b/contrib/CMakeLists.txt @@ -71,7 +71,6 @@ add_contrib (zlib-ng-cmake zlib-ng) add_contrib (bzip2-cmake bzip2) add_contrib (minizip-ng-cmake minizip-ng) add_contrib (snappy-cmake snappy) -add_contrib (rocksdb-cmake rocksdb) add_contrib (thrift-cmake thrift) # parquet/arrow/orc add_contrib (arrow-cmake arrow) # requires: snappy, thrift, double-conversion @@ -148,6 +147,7 @@ add_contrib (hive-metastore-cmake hive-metastore) # requires: thrift, avro, arro add_contrib (cppkafka-cmake cppkafka) add_contrib (libpqxx-cmake libpqxx) add_contrib (libpq-cmake libpq) +add_contrib (rocksdb-cmake rocksdb) # requires: jemalloc, snappy, zlib, lz4, zstd, liburing add_contrib (nuraft-cmake NuRaft) add_contrib (fast_float-cmake fast_float) add_contrib (idna-cmake idna) @@ -179,7 +179,7 @@ else() message(STATUS "Not using QPL") endif () -if (OS_LINUX AND ARCH_AMD64) +if (OS_LINUX AND ARCH_AMD64 AND NOT NO_SSE3_OR_HIGHER) option (ENABLE_QATLIB "Enable IntelĀ® QuickAssist Technology Library (QATlib)" ${ENABLE_LIBRARIES}) elseif(ENABLE_QATLIB) message (${RECONFIGURE_MESSAGE_LEVEL} "QATLib is only supported on x86_64") @@ -205,14 +205,12 @@ add_contrib (morton-nd-cmake morton-nd) if (ARCH_S390X) add_contrib(crc32-s390x-cmake crc32-s390x) endif() -add_contrib (annoy-cmake annoy) -option(ENABLE_USEARCH "Enable USearch (Approximate Neighborhood Search, HNSW) support" ${ENABLE_LIBRARIES}) +option(ENABLE_USEARCH "Enable USearch" ${ENABLE_LIBRARIES}) if (ENABLE_USEARCH) add_contrib (FP16-cmake FP16) - add_contrib (robin-map-cmake robin-map) add_contrib (SimSIMD-cmake SimSIMD) - add_contrib (usearch-cmake usearch) # requires: FP16, robin-map, SimdSIMD + add_contrib (usearch-cmake usearch) # requires: FP16, SimdSIMD else () message(STATUS "Not using USearch") endif () diff --git a/contrib/QAT-ZSTD-Plugin-cmake/CMakeLists.txt b/contrib/QAT-ZSTD-Plugin-cmake/CMakeLists.txt index 72d21a8572b..fc18092f574 100644 --- a/contrib/QAT-ZSTD-Plugin-cmake/CMakeLists.txt +++ b/contrib/QAT-ZSTD-Plugin-cmake/CMakeLists.txt @@ -27,7 +27,7 @@ if (ENABLE_QAT_OUT_OF_TREE_BUILD) ${QAT_AL_INCLUDE_DIR} ${QAT_USDM_INCLUDE_DIR} ${ZSTD_LIBRARY_DIR}) - target_compile_definitions(_qatzstd_plugin PRIVATE -DDEBUGLEVEL=0 PUBLIC -DENABLE_ZSTD_QAT_CODEC) + target_compile_definitions(_qatzstd_plugin PRIVATE -DDEBUGLEVEL=0) add_library (ch_contrib::qatzstd_plugin ALIAS _qatzstd_plugin) else () # In-tree build message(STATUS "Intel QATZSTD in-tree build") @@ -78,7 +78,7 @@ else () # In-tree build ${QAT_USDM_INCLUDE_DIR} ${ZSTD_LIBRARY_DIR} ${LIBQAT_HEADER_DIR}) - target_compile_definitions(_qatzstd_plugin PRIVATE -DDEBUGLEVEL=0 PUBLIC -DENABLE_ZSTD_QAT_CODEC -DINTREE) + target_compile_definitions(_qatzstd_plugin PRIVATE -DDEBUGLEVEL=0 PUBLIC -DINTREE) target_include_directories(_qatzstd_plugin SYSTEM PUBLIC $ $) add_library (ch_contrib::qatzstd_plugin ALIAS _qatzstd_plugin) endif () diff --git a/contrib/SimSIMD b/contrib/SimSIMD index de2cb75b9e9..91a76d1ac51 160000 --- a/contrib/SimSIMD +++ b/contrib/SimSIMD @@ -1 +1 @@ -Subproject commit de2cb75b9e9e3389d5e1e51fd9f8ed151f3c17cf +Subproject commit 91a76d1ac519b3b9dc8957734a3dabd985f00c26 diff --git a/contrib/annoy b/contrib/annoy deleted file mode 160000 index f2ac8e7b48f..00000000000 --- a/contrib/annoy +++ /dev/null @@ -1 +0,0 @@ -Subproject commit f2ac8e7b48f9a9cf676d3b58286e5455aba8e956 diff --git a/contrib/annoy-cmake/CMakeLists.txt b/contrib/annoy-cmake/CMakeLists.txt deleted file mode 100644 index bdef7d92132..00000000000 --- a/contrib/annoy-cmake/CMakeLists.txt +++ /dev/null @@ -1,24 +0,0 @@ -option(ENABLE_ANNOY "Enable Annoy index support" ${ENABLE_LIBRARIES}) - -# Annoy index should be disabled with undefined sanitizer. Because of memory storage optimizations -# (https://github.com/ClickHouse/annoy/blob/9d8a603a4cd252448589e84c9846f94368d5a289/src/annoylib.h#L442-L463) -# UBSan fails and leads to crash. Simmilar issue is already opened in Annoy repo -# https://github.com/spotify/annoy/issues/456 -# Problem with aligment can lead to errors like -# (https://stackoverflow.com/questions/46790550/c-undefined-behavior-strict-aliasing-rule-or-incorrect-alignment) -# or will lead to crash on arm https://developer.arm.com/documentation/ka003038/latest -# This issues should be resolved before annoy became non-experimental (--> setting "allow_experimental_annoy_index") -if ((NOT ENABLE_ANNOY) OR (SANITIZE STREQUAL "undefined") OR (ARCH_AARCH64)) - message (STATUS "Not using annoy") - return() -endif() - -set(ANNOY_PROJECT_DIR "${ClickHouse_SOURCE_DIR}/contrib/annoy") -set(ANNOY_SOURCE_DIR "${ANNOY_PROJECT_DIR}/src") - -add_library(_annoy INTERFACE) -target_include_directories(_annoy SYSTEM INTERFACE ${ANNOY_SOURCE_DIR}) - -add_library(ch_contrib::annoy ALIAS _annoy) -target_compile_definitions(_annoy INTERFACE ENABLE_ANNOY) -target_compile_definitions(_annoy INTERFACE ANNOYLIB_MULTITHREADED_BUILD) diff --git a/contrib/libfiu-cmake/CMakeLists.txt b/contrib/libfiu-cmake/CMakeLists.txt index e805491edbb..eab55087c98 100644 --- a/contrib/libfiu-cmake/CMakeLists.txt +++ b/contrib/libfiu-cmake/CMakeLists.txt @@ -1,20 +1,21 @@ -if (NOT ENABLE_FIU) - message (STATUS "Not using fiu") +if (NOT ENABLE_LIBFIU) + message (STATUS "Not using libfiu") return () endif () -set(FIU_DIR "${ClickHouse_SOURCE_DIR}/contrib/libfiu/") +set(LIBFIU_DIR "${ClickHouse_SOURCE_DIR}/contrib/libfiu/") -set(FIU_SOURCES - ${FIU_DIR}/libfiu/fiu.c - ${FIU_DIR}/libfiu/fiu-rc.c - ${FIU_DIR}/libfiu/backtrace.c - ${FIU_DIR}/libfiu/wtable.c +set(LIBFIU_SOURCES + ${LIBFIU_DIR}/libfiu/fiu.c + ${LIBFIU_DIR}/libfiu/fiu-rc.c + ${LIBFIU_DIR}/libfiu/backtrace.c + ${LIBFIU_DIR}/libfiu/wtable.c ) -set(FIU_HEADERS "${FIU_DIR}/libfiu") +set(LIBFIU_HEADERS "${LIBFIU_DIR}/libfiu") -add_library(_fiu ${FIU_SOURCES}) -target_compile_definitions(_fiu PUBLIC DUMMY_BACKTRACE) -target_include_directories(_fiu PUBLIC ${FIU_HEADERS}) -add_library(ch_contrib::fiu ALIAS _fiu) +add_library(_libfiu ${LIBFIU_SOURCES}) +target_compile_definitions(_libfiu PUBLIC DUMMY_BACKTRACE) +target_compile_definitions(_libfiu PUBLIC FIU_ENABLE) +target_include_directories(_libfiu PUBLIC ${LIBFIU_HEADERS}) +add_library(ch_contrib::libfiu ALIAS _libfiu) diff --git a/contrib/libprotobuf-mutator b/contrib/libprotobuf-mutator index 1f95f808306..b922c8ab900 160000 --- a/contrib/libprotobuf-mutator +++ b/contrib/libprotobuf-mutator @@ -1 +1 @@ -Subproject commit 1f95f8083066f5b38fd2db172e7e7f9aa7c49d2d +Subproject commit b922c8ab9004ef9944982e4f165e2747b13223fa diff --git a/contrib/librdkafka b/contrib/librdkafka index 2d2aab6f5b7..39d4ed49ccf 160000 --- a/contrib/librdkafka +++ b/contrib/librdkafka @@ -1 +1 @@ -Subproject commit 2d2aab6f5b79db1cfca15d7bf0dee75d00d82082 +Subproject commit 39d4ed49ccf3406e2bf825d5d7b0903b5a290782 diff --git a/contrib/libunwind b/contrib/libunwind index a89d904befe..601db0b0e03 160000 --- a/contrib/libunwind +++ b/contrib/libunwind @@ -1 +1 @@ -Subproject commit a89d904befea07814628c6ce0b44083c4e149c62 +Subproject commit 601db0b0e03018c01710470a37703b618f9cf08b diff --git a/contrib/qpl b/contrib/qpl index d4715e0e798..c2ced94c53c 160000 --- a/contrib/qpl +++ b/contrib/qpl @@ -1 +1 @@ -Subproject commit d4715e0e79896b85612158e135ee1a85f3b3e04d +Subproject commit c2ced94c53c1ee22191201a59878e9280bc9b9b8 diff --git a/contrib/qpl-cmake/CMakeLists.txt b/contrib/qpl-cmake/CMakeLists.txt index 7a84048e16b..89332ae0f7a 100644 --- a/contrib/qpl-cmake/CMakeLists.txt +++ b/contrib/qpl-cmake/CMakeLists.txt @@ -4,7 +4,6 @@ set (QPL_PROJECT_DIR "${ClickHouse_SOURCE_DIR}/contrib/qpl") set (QPL_SRC_DIR "${ClickHouse_SOURCE_DIR}/contrib/qpl/sources") set (QPL_BINARY_DIR "${ClickHouse_BINARY_DIR}/build/contrib/qpl") set (EFFICIENT_WAIT OFF) -set (BLOCK_ON_FAULT ON) set (LOG_HW_INIT OFF) set (SANITIZE_MEMORY OFF) set (SANITIZE_THREADS OFF) @@ -16,16 +15,20 @@ function(GetLibraryVersion _content _outputVar) SET(${_outputVar} ${CMAKE_MATCH_1} PARENT_SCOPE) endfunction() -set (QPL_VERSION 1.2.0) +set (QPL_VERSION 1.6.0) message(STATUS "Intel QPL version: ${QPL_VERSION}") -# There are 5 source subdirectories under $QPL_SRC_DIR: isal, c_api, core-sw, middle-layer, c_api. -# Generate 8 library targets: middle_layer_lib, isal, isal_asm, qplcore_px, qplcore_avx512, qplcore_sw_dispatcher, core_iaa, middle_layer_lib. +# There are 5 source subdirectories under $QPL_SRC_DIR: c_api, core-iaa, core-sw, middle-layer and isal. +# Generate 8 library targets: qpl_c_api, core_iaa, qplcore_px, qplcore_avx512, qplcore_sw_dispatcher, middle_layer_lib, isal and isal_asm, +# which are then combined into static or shared qpl. # Output ch_contrib::qpl by linking with 8 library targets. -# The qpl submodule comes with its own version of isal. It contains code which does not exist in upstream isal. It would be nice to link -# only upstream isal (ch_contrib::isal) but at this point we can't. +# Note, QPL has integrated a customized version of ISA-L to meet specific needs. +# This version has been significantly modified and there are no plans to maintain compatibility with the upstream version +# or upgrade the current copy. + +## cmake/CompileOptions.cmake and automatic wrappers generation # ========================================================================== # Copyright (C) 2022 Intel Corporation @@ -442,6 +445,7 @@ function(generate_unpack_kernel_arrays current_directory PLATFORMS_LIST) endforeach() endfunction() +# [SUBDIR]isal enable_language(ASM_NASM) @@ -479,7 +483,6 @@ set(ISAL_ASM_SRC ${QPL_SRC_DIR}/isal/igzip/igzip_body.asm ${QPL_SRC_DIR}/isal/igzip/igzip_set_long_icf_fg_04.asm ${QPL_SRC_DIR}/isal/igzip/igzip_set_long_icf_fg_06.asm ${QPL_SRC_DIR}/isal/igzip/igzip_multibinary.asm - ${QPL_SRC_DIR}/isal/igzip/stdmac.asm ${QPL_SRC_DIR}/isal/crc/crc_multibinary.asm ${QPL_SRC_DIR}/isal/crc/crc32_gzip_refl_by8.asm ${QPL_SRC_DIR}/isal/crc/crc32_gzip_refl_by8_02.asm @@ -505,7 +508,6 @@ set_property(GLOBAL APPEND PROPERTY QPL_LIB_DEPS # Setting external and internal interfaces for ISA-L library target_include_directories(isal PUBLIC $ - PRIVATE ${QPL_SRC_DIR}/isal/include PUBLIC ${QPL_SRC_DIR}/isal/igzip) set_target_properties(isal PROPERTIES @@ -617,12 +619,9 @@ target_compile_options(qplcore_sw_dispatcher # [SUBDIR]core-iaa file(GLOB HW_PATH_SRC ${QPL_SRC_DIR}/core-iaa/sources/aecs/*.c - ${QPL_SRC_DIR}/core-iaa/sources/aecs/*.cpp ${QPL_SRC_DIR}/core-iaa/sources/driver_loader/*.c - ${QPL_SRC_DIR}/core-iaa/sources/driver_loader/*.cpp ${QPL_SRC_DIR}/core-iaa/sources/descriptors/*.c - ${QPL_SRC_DIR}/core-iaa/sources/descriptors/*.cpp - ${QPL_SRC_DIR}/core-iaa/sources/bit_rev.c) + ${QPL_SRC_DIR}/core-iaa/sources/*.c) # Create library add_library(core_iaa OBJECT ${HW_PATH_SRC}) @@ -634,31 +633,27 @@ target_include_directories(core_iaa PRIVATE ${UUID_DIR} PUBLIC $ PUBLIC $ - PRIVATE $ # status.h in own_checkers.h - PRIVATE $ # own_checkers.h + PRIVATE $ # status.h in own_checkers.h + PRIVATE $ # for own_checkers.h PRIVATE $) target_compile_features(core_iaa PRIVATE c_std_11) target_compile_definitions(core_iaa PRIVATE QPL_BADARG_CHECK - PRIVATE $<$: BLOCK_ON_FAULT_ENABLED> PRIVATE $<$:LOG_HW_INIT> PRIVATE $<$:DYNAMIC_LOADING_LIBACCEL_CONFIG>) # [SUBDIR]middle-layer file(GLOB MIDDLE_LAYER_SRC - ${QPL_SRC_DIR}/middle-layer/analytics/*.cpp - ${QPL_SRC_DIR}/middle-layer/c_wrapper/*.cpp - ${QPL_SRC_DIR}/middle-layer/checksum/*.cpp + ${QPL_SRC_DIR}/middle-layer/accelerator/*.cpp + ${QPL_SRC_DIR}/middle-layer/analytics/*.cpp ${QPL_SRC_DIR}/middle-layer/common/*.cpp ${QPL_SRC_DIR}/middle-layer/compression/*.cpp ${QPL_SRC_DIR}/middle-layer/compression/*/*.cpp ${QPL_SRC_DIR}/middle-layer/compression/*/*/*.cpp ${QPL_SRC_DIR}/middle-layer/dispatcher/*.cpp ${QPL_SRC_DIR}/middle-layer/other/*.cpp - ${QPL_SRC_DIR}/middle-layer/util/*.cpp - ${QPL_SRC_DIR}/middle-layer/inflate/*.cpp - ${QPL_SRC_DIR}/core-iaa/sources/accelerator/*.cpp) # todo + ${QPL_SRC_DIR}/middle-layer/util/*.cpp) add_library(middle_layer_lib OBJECT ${MIDDLE_LAYER_SRC}) @@ -667,6 +662,7 @@ set_property(GLOBAL APPEND PROPERTY QPL_LIB_DEPS $) target_compile_options(middle_layer_lib + PRIVATE $<$:$<$:-O3;-U_FORTIFY_SOURCE;-D_FORTIFY_SOURCE=2>> PRIVATE ${QPL_LINUX_TOOLCHAIN_CPP_EMBEDDED_FLAGS}) target_compile_definitions(middle_layer_lib @@ -682,6 +678,7 @@ target_include_directories(middle_layer_lib PRIVATE ${UUID_DIR} PUBLIC $ PUBLIC $ + PRIVATE $ PUBLIC $ PUBLIC $ PUBLIC $) @@ -689,31 +686,50 @@ target_include_directories(middle_layer_lib target_compile_definitions(middle_layer_lib PUBLIC -DQPL_LIB) # [SUBDIR]c_api -file(GLOB_RECURSE QPL_C_API_SRC - ${QPL_SRC_DIR}/c_api/*.c - ${QPL_SRC_DIR}/c_api/*.cpp) +file(GLOB QPL_C_API_SRC + ${QPL_SRC_DIR}/c_api/compression_operations/*.c + ${QPL_SRC_DIR}/c_api/compression_operations/*.cpp + ${QPL_SRC_DIR}/c_api/filter_operations/*.cpp + ${QPL_SRC_DIR}/c_api/legacy_hw_path/*.c + ${QPL_SRC_DIR}/c_api/legacy_hw_path/*.cpp + ${QPL_SRC_DIR}/c_api/other_operations/*.cpp + ${QPL_SRC_DIR}/c_api/serialization/*.cpp + ${QPL_SRC_DIR}/c_api/*.cpp) + +add_library(qpl_c_api OBJECT ${QPL_C_API_SRC}) + +target_include_directories(qpl_c_api + PUBLIC $ + PUBLIC $ $ + PRIVATE $) + +set_target_properties(qpl_c_api PROPERTIES + $<$:C_STANDARD 17 + CXX_STANDARD 17) + +target_compile_options(qpl_c_api + PRIVATE $<$:$<$:-O3;-U_FORTIFY_SOURCE;-D_FORTIFY_SOURCE=2>> + PRIVATE $<$:${QPL_LINUX_TOOLCHAIN_CPP_EMBEDDED_FLAGS}>) + +target_compile_definitions(qpl_c_api + PUBLIC -DQPL_BADARG_CHECK # own_checkers.h + PUBLIC -DQPL_LIB # needed for middle_layer_lib + PUBLIC $<$:LOG_HW_INIT>) # needed for middle_layer_lib + +set_property(GLOBAL APPEND PROPERTY QPL_LIB_DEPS + $) + +# Final _qpl target get_property(LIB_DEPS GLOBAL PROPERTY QPL_LIB_DEPS) -add_library(_qpl STATIC ${QPL_C_API_SRC} ${LIB_DEPS}) +add_library(_qpl STATIC ${LIB_DEPS}) target_include_directories(_qpl - PUBLIC $ $ - PRIVATE $ - PRIVATE $) - -target_compile_options(_qpl - PRIVATE ${QPL_LINUX_TOOLCHAIN_CPP_EMBEDDED_FLAGS}) - -target_compile_definitions(_qpl - PRIVATE -DQPL_LIB - PRIVATE -DQPL_BADARG_CHECK - PRIVATE $<$:DYNAMIC_LOADING_LIBACCEL_CONFIG> - PUBLIC -DENABLE_QPL_COMPRESSION) + PUBLIC $ $) target_link_libraries(_qpl - PRIVATE ch_contrib::accel-config - PRIVATE ch_contrib::isal) + PRIVATE ch_contrib::accel-config) target_include_directories(_qpl SYSTEM BEFORE PUBLIC "${QPL_PROJECT_DIR}/include" diff --git a/contrib/robin-map b/contrib/robin-map deleted file mode 160000 index 851a59e0e30..00000000000 --- a/contrib/robin-map +++ /dev/null @@ -1 +0,0 @@ -Subproject commit 851a59e0e3063ee0e23089062090a73fd3de482d diff --git a/contrib/robin-map-cmake/CMakeLists.txt b/contrib/robin-map-cmake/CMakeLists.txt deleted file mode 100644 index f82ad705dcc..00000000000 --- a/contrib/robin-map-cmake/CMakeLists.txt +++ /dev/null @@ -1 +0,0 @@ -# See contrib/usearch-cmake/CMakeLists.txt diff --git a/contrib/rocksdb b/contrib/rocksdb index 01e43568fa9..5f003e4a22d 160000 --- a/contrib/rocksdb +++ b/contrib/rocksdb @@ -1 +1 @@ -Subproject commit 01e43568fa9f3f7bf107b2b66c00b286b456f33e +Subproject commit 5f003e4a22d2e48e37c98d9620241237cd30dd24 diff --git a/contrib/rocksdb-cmake/CMakeLists.txt b/contrib/rocksdb-cmake/CMakeLists.txt index 98790158baa..44aa7494607 100644 --- a/contrib/rocksdb-cmake/CMakeLists.txt +++ b/contrib/rocksdb-cmake/CMakeLists.txt @@ -1,51 +1,46 @@ option (ENABLE_ROCKSDB "Enable RocksDB" ${ENABLE_LIBRARIES}) -if (NOT ENABLE_ROCKSDB) +if (NOT ENABLE_ROCKSDB OR NO_SSE3_OR_HIGHER) # assumes SSE4.2 and PCLMUL message (STATUS "Not using RocksDB") return() endif() -# not in original build system, otherwise xxHash.cc fails to compile with ClickHouse C++23 default -set (CMAKE_CXX_STANDARD 20) - -# Always disable jemalloc for rocksdb by default because it introduces non-standard jemalloc APIs -option(WITH_JEMALLOC "build with JeMalloc" OFF) - -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) -if(WITH_SNAPPY) +if (ENABLE_JEMALLOC AND OS_LINUX) # gives compile errors with jemalloc enabled for rocksdb on non-Linux + add_definitions(-DROCKSDB_JEMALLOC -DJEMALLOC_NO_DEMANGLE) + list (APPEND THIRDPARTY_LIBS ch_contrib::jemalloc) +endif () + +if (ENABLE_LIBURING) + add_definitions(-DROCKSDB_IOURING_PRESENT) + list (APPEND THIRDPARTY_LIBS ch_contrib::liburing) +endif () + +if (WITH_SNAPPY) add_definitions(-DSNAPPY) list(APPEND THIRDPARTY_LIBS ch_contrib::snappy) endif() -if(WITH_ZLIB) +if (WITH_ZLIB) add_definitions(-DZLIB) list(APPEND THIRDPARTY_LIBS ch_contrib::zlib) endif() -if(WITH_LZ4) +if (WITH_LZ4) add_definitions(-DLZ4) list(APPEND THIRDPARTY_LIBS ch_contrib::lz4) endif() -if(WITH_ZSTD) +if (WITH_ZSTD) add_definitions(-DZSTD) list(APPEND THIRDPARTY_LIBS ch_contrib::zstd) endif() -add_definitions(-DROCKSDB_PORTABLE) - -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 @@ -91,8 +86,11 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/cache/compressed_secondary_cache.cc ${ROCKSDB_SOURCE_DIR}/cache/lru_cache.cc ${ROCKSDB_SOURCE_DIR}/cache/secondary_cache.cc + ${ROCKSDB_SOURCE_DIR}/cache/secondary_cache_adapter.cc ${ROCKSDB_SOURCE_DIR}/cache/sharded_cache.cc + ${ROCKSDB_SOURCE_DIR}/cache/tiered_secondary_cache.cc ${ROCKSDB_SOURCE_DIR}/db/arena_wrapped_db_iter.cc + ${ROCKSDB_SOURCE_DIR}/db/attribute_group_iterator_impl.cc ${ROCKSDB_SOURCE_DIR}/db/blob/blob_contents.cc ${ROCKSDB_SOURCE_DIR}/db/blob/blob_fetcher.cc ${ROCKSDB_SOURCE_DIR}/db/blob/blob_file_addition.cc @@ -109,6 +107,7 @@ set(SOURCES ${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/coalescing_iterator.cc ${ROCKSDB_SOURCE_DIR}/db/column_family.cc ${ROCKSDB_SOURCE_DIR}/db/compaction/compaction.cc ${ROCKSDB_SOURCE_DIR}/db/compaction/compaction_iterator.cc @@ -129,6 +128,7 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/db/db_impl/db_impl_write.cc ${ROCKSDB_SOURCE_DIR}/db/db_impl/db_impl_compaction_flush.cc ${ROCKSDB_SOURCE_DIR}/db/db_impl/db_impl_files.cc + ${ROCKSDB_SOURCE_DIR}/db/db_impl/db_impl_follower.cc ${ROCKSDB_SOURCE_DIR}/db/db_impl/db_impl_open.cc ${ROCKSDB_SOURCE_DIR}/db/db_impl/db_impl_debug.cc ${ROCKSDB_SOURCE_DIR}/db/db_impl/db_impl_experimental.cc @@ -174,9 +174,11 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/db/wal_manager.cc ${ROCKSDB_SOURCE_DIR}/db/wide/wide_column_serialization.cc ${ROCKSDB_SOURCE_DIR}/db/wide/wide_columns.cc + ${ROCKSDB_SOURCE_DIR}/db/wide/wide_columns_helper.cc ${ROCKSDB_SOURCE_DIR}/db/write_batch.cc ${ROCKSDB_SOURCE_DIR}/db/write_batch_base.cc ${ROCKSDB_SOURCE_DIR}/db/write_controller.cc + ${ROCKSDB_SOURCE_DIR}/db/write_stall_stats.cc ${ROCKSDB_SOURCE_DIR}/db/write_thread.cc ${ROCKSDB_SOURCE_DIR}/env/composite_env.cc ${ROCKSDB_SOURCE_DIR}/env/env.cc @@ -184,6 +186,7 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/env/env_encryption.cc ${ROCKSDB_SOURCE_DIR}/env/file_system.cc ${ROCKSDB_SOURCE_DIR}/env/file_system_tracer.cc + ${ROCKSDB_SOURCE_DIR}/env/fs_on_demand.cc ${ROCKSDB_SOURCE_DIR}/env/fs_remap.cc ${ROCKSDB_SOURCE_DIR}/env/mock_env.cc ${ROCKSDB_SOURCE_DIR}/env/unique_id_gen.cc @@ -229,6 +232,7 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/options/configurable.cc ${ROCKSDB_SOURCE_DIR}/options/customizable.cc ${ROCKSDB_SOURCE_DIR}/options/db_options.cc + ${ROCKSDB_SOURCE_DIR}/options/offpeak_time_info.cc ${ROCKSDB_SOURCE_DIR}/options/options.cc ${ROCKSDB_SOURCE_DIR}/options/options_helper.cc ${ROCKSDB_SOURCE_DIR}/options/options_parser.cc @@ -268,6 +272,7 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/table/get_context.cc ${ROCKSDB_SOURCE_DIR}/table/iterator.cc ${ROCKSDB_SOURCE_DIR}/table/merging_iterator.cc + ${ROCKSDB_SOURCE_DIR}/table/compaction_merging_iterator.cc ${ROCKSDB_SOURCE_DIR}/table/meta_blocks.cc ${ROCKSDB_SOURCE_DIR}/table/persistent_cache_helper.cc ${ROCKSDB_SOURCE_DIR}/table/plain/plain_table_bloom.cc @@ -309,6 +314,7 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/util/compression_context_cache.cc ${ROCKSDB_SOURCE_DIR}/util/concurrent_task_limiter_impl.cc ${ROCKSDB_SOURCE_DIR}/util/crc32c.cc + ${ROCKSDB_SOURCE_DIR}/util/data_structure.cc ${ROCKSDB_SOURCE_DIR}/util/dynamic_bloom.cc ${ROCKSDB_SOURCE_DIR}/util/hash.cc ${ROCKSDB_SOURCE_DIR}/util/murmurhash.cc @@ -322,6 +328,8 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/util/string_util.cc ${ROCKSDB_SOURCE_DIR}/util/thread_local.cc ${ROCKSDB_SOURCE_DIR}/util/threadpool_imp.cc + ${ROCKSDB_SOURCE_DIR}/util/udt_util.cc + ${ROCKSDB_SOURCE_DIR}/util/write_batch_util.cc ${ROCKSDB_SOURCE_DIR}/util/xxhash.cc ${ROCKSDB_SOURCE_DIR}/utilities/agg_merge/agg_merge.cc ${ROCKSDB_SOURCE_DIR}/utilities/backup/backup_engine.cc @@ -366,6 +374,7 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/utilities/persistent_cache/volatile_tier_impl.cc ${ROCKSDB_SOURCE_DIR}/utilities/simulator_cache/cache_simulator.cc ${ROCKSDB_SOURCE_DIR}/utilities/simulator_cache/sim_cache.cc + ${ROCKSDB_SOURCE_DIR}/utilities/table_properties_collectors/compact_for_tiering_collector.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 @@ -386,6 +395,7 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/utilities/transactions/write_prepared_txn_db.cc ${ROCKSDB_SOURCE_DIR}/utilities/transactions/write_unprepared_txn.cc ${ROCKSDB_SOURCE_DIR}/utilities/transactions/write_unprepared_txn_db.cc + ${ROCKSDB_SOURCE_DIR}/utilities/types_util.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 @@ -404,12 +414,6 @@ set(SOURCES ${ROCKSDB_SOURCE_DIR}/utilities/transactions/lock/range/range_tree/lib/util/memarena.cc build_version.cc) # generated by hand -if(ENABLE_SSE42 AND ENABLE_PCLMULQDQ) - set_source_files_properties( - "${ROCKSDB_SOURCE_DIR}/util/crc32c.cc" - PROPERTIES COMPILE_FLAGS "-msse4.2 -mpclmul") -endif() - if(CMAKE_SYSTEM_PROCESSOR MATCHES "^(powerpc|ppc)64") list(APPEND SOURCES "${ROCKSDB_SOURCE_DIR}/util/crc32c_ppc.c" @@ -422,14 +426,18 @@ if(HAS_ARMV8_CRC) endif(HAS_ARMV8_CRC) list(APPEND SOURCES - "${ROCKSDB_SOURCE_DIR}/port/port_posix.cc" - "${ROCKSDB_SOURCE_DIR}/env/env_posix.cc" - "${ROCKSDB_SOURCE_DIR}/env/fs_posix.cc" - "${ROCKSDB_SOURCE_DIR}/env/io_posix.cc") + ${ROCKSDB_SOURCE_DIR}/port/port_posix.cc + ${ROCKSDB_SOURCE_DIR}/env/env_posix.cc + ${ROCKSDB_SOURCE_DIR}/env/fs_posix.cc + ${ROCKSDB_SOURCE_DIR}/env/io_posix.cc) add_library(_rocksdb ${SOURCES}) add_library(ch_contrib::rocksdb ALIAS _rocksdb) target_link_libraries(_rocksdb PRIVATE ${THIRDPARTY_LIBS} ${SYSTEM_LIBS}) +# Not in the native build system but useful anyways: +# Make all functions in xxHash.h inline. Beneficial for performance: https://github.com/Cyan4973/xxHash/tree/v0.8.2#build-modifiers +target_compile_definitions (_rocksdb PRIVATE XXH_INLINE_ALL) + # SYSTEM is required to overcome some issues target_include_directories(_rocksdb SYSTEM BEFORE INTERFACE "${ROCKSDB_SOURCE_DIR}/include") diff --git a/contrib/usearch b/contrib/usearch index 955c6f9c11a..e21a5778a0d 160000 --- a/contrib/usearch +++ b/contrib/usearch @@ -1 +1 @@ -Subproject commit 955c6f9c11adfd89c912e0d1643d160b4e9e543f +Subproject commit e21a5778a0d4469ddaf38c94b7be0196bb701ee4 diff --git a/contrib/usearch-cmake/CMakeLists.txt b/contrib/usearch-cmake/CMakeLists.txt index 29fbe57106c..83221e3810f 100644 --- a/contrib/usearch-cmake/CMakeLists.txt +++ b/contrib/usearch-cmake/CMakeLists.txt @@ -1,17 +1,12 @@ -set(USEARCH_PROJECT_DIR "${ClickHouse_SOURCE_DIR}/contrib/usearch") -set(USEARCH_SOURCE_DIR "${USEARCH_PROJECT_DIR}/include") - set(FP16_PROJECT_DIR "${ClickHouse_SOURCE_DIR}/contrib/FP16") -set(ROBIN_MAP_PROJECT_DIR "${ClickHouse_SOURCE_DIR}/contrib/robin-map") -set(SIMSIMD_PROJECT_DIR "${ClickHouse_SOURCE_DIR}/contrib/SimSIMD-map") +set(SIMSIMD_PROJECT_DIR "${ClickHouse_SOURCE_DIR}/contrib/SimSIMD") +set(USEARCH_PROJECT_DIR "${ClickHouse_SOURCE_DIR}/contrib/usearch") add_library(_usearch INTERFACE) target_include_directories(_usearch SYSTEM INTERFACE ${FP16_PROJECT_DIR}/include - ${ROBIN_MAP_PROJECT_DIR}/include ${SIMSIMD_PROJECT_DIR}/include - ${USEARCH_SOURCE_DIR}) + ${USEARCH_PROJECT_DIR}/include) add_library(ch_contrib::usearch ALIAS _usearch) -target_compile_definitions(_usearch INTERFACE ENABLE_USEARCH) diff --git a/contrib/zlib-ng b/contrib/zlib-ng index 50f0eae1a41..a2fbeffdc30 160000 --- a/contrib/zlib-ng +++ b/contrib/zlib-ng @@ -1 +1 @@ -Subproject commit 50f0eae1a411764cd6d1e85b3ce471438acd3c1c +Subproject commit a2fbeffdc30a8b0ce6d54ee31208e2688eac4c9f diff --git a/contrib/zlib-ng-cmake/CMakeLists.txt b/contrib/zlib-ng-cmake/CMakeLists.txt index 79f343bfc75..91a8eb2e0a3 100644 --- a/contrib/zlib-ng-cmake/CMakeLists.txt +++ b/contrib/zlib-ng-cmake/CMakeLists.txt @@ -14,6 +14,8 @@ add_definitions(-DHAVE_VISIBILITY_HIDDEN) add_definitions(-DHAVE_VISIBILITY_INTERNAL) add_definitions(-DHAVE_BUILTIN_CTZ) add_definitions(-DHAVE_BUILTIN_CTZLL) +add_definitions(-DHAVE_ATTRIBUTE_ALIGNED) +add_definitions(-DHAVE_POSIX_MEMALIGN) set(ZLIB_ARCH_SRCS) set(ZLIB_ARCH_HDRS) @@ -24,67 +26,74 @@ if(ARCH_AARCH64) set(ARCHDIR "${SOURCE_DIR}/arch/arm") add_definitions(-DARM_FEATURES) + add_definitions(-DHAVE_SYS_AUXV_H) add_definitions(-DARM_AUXV_HAS_CRC32 -DARM_ASM_HWCAP) add_definitions(-DARM_AUXV_HAS_NEON) - add_definitions(-DARM_ACLE_CRC_HASH) - add_definitions(-DARM_NEON_ADLER32 -DARM_NEON_CHUNKSET -DARM_NEON_SLIDEHASH) + add_definitions(-DARM_ACLE) + add_definitions(-DHAVE_ARM_ACLE_H) + add_definitions(-DARM_NEON) + add_definitions(-DARM_NEON_HASLD4) - list(APPEND ZLIB_ARCH_HDRS ${ARCHDIR}/arm.h) - list(APPEND ZLIB_ARCH_SRCS ${ARCHDIR}/armfeature.c) + list(APPEND ZLIB_ARCH_HDRS ${ARCHDIR}/arm_features.h) + list(APPEND ZLIB_ARCH_SRCS ${ARCHDIR}/arm_features.c) set(ACLE_SRCS ${ARCHDIR}/crc32_acle.c ${ARCHDIR}/insert_string_acle.c) list(APPEND ZLIB_ARCH_SRCS ${ACLE_SRCS}) - set(NEON_SRCS ${ARCHDIR}/adler32_neon.c ${ARCHDIR}/chunkset_neon.c ${ARCHDIR}/slide_neon.c) + set(NEON_SRCS ${ARCHDIR}/adler32_neon.c ${ARCHDIR}/chunkset_neon.c + ${ARCHDIR}/compare256_neon.c ${ARCHDIR}/slide_hash_neon.c) list(APPEND ZLIB_ARCH_SRCS ${NEON_SRCS}) elseif(ARCH_PPC64LE) set(ARCHDIR "${SOURCE_DIR}/arch/power") - add_definitions(-DPOWER8) add_definitions(-DPOWER_FEATURES) - add_definitions(-DPOWER8_VSX_ADLER32) - add_definitions(-DPOWER8_VSX_SLIDEHASH) + add_definitions(-DHAVE_SYS_AUXV_H) - list(APPEND ZLIB_ARCH_HDRS ${ARCHDIR}/power.h) - list(APPEND ZLIB_ARCH_SRCS ${ARCHDIR}/power.c) - set(POWER8_SRCS ${ARCHDIR}/adler32_power8.c ${ARCHDIR}/slide_hash_power8.c) + if(POWER9) + add_definitions(-DPOWER9) + else() + add_definitions(-DPOWER8) + add_definitions(-DPOWER8_VSX) + add_definitions(-DPOWER8_VSX_CRC32) + endif() + + list(APPEND ZLIB_ARCH_HDRS ${ARCHDIR}/power_features.h) + list(APPEND ZLIB_ARCH_SRCS ${ARCHDIR}/power_features.c) + set(POWER8_SRCS ${ARCHDIR}/adler32_power8.c ${ARCHDIR}/chunkset_power8.c ${ARCHDIR}/slide_hash_power8.c) + list(APPEND POWER8_SRCS ${ARCHDIR}/crc32_power8.c) list(APPEND ZLIB_ARCH_SRCS ${POWER8_SRCS}) elseif(ARCH_AMD64) set(ARCHDIR "${SOURCE_DIR}/arch/x86") add_definitions(-DX86_FEATURES) - list(APPEND ZLIB_ARCH_HDRS ${ARCHDIR}/x86.h) - list(APPEND ZLIB_ARCH_SRCS ${ARCHDIR}/x86.c) + list(APPEND ZLIB_ARCH_HDRS ${ARCHDIR}/x86_features.h) + list(APPEND ZLIB_ARCH_SRCS ${ARCHDIR}/x86_features.c) if(ENABLE_AVX2) - add_definitions(-DX86_AVX2 -DX86_AVX2_ADLER32 -DX86_AVX_CHUNKSET) - set(AVX2_SRCS ${ARCHDIR}/slide_avx.c) - list(APPEND AVX2_SRCS ${ARCHDIR}/chunkset_avx.c) - list(APPEND AVX2_SRCS ${ARCHDIR}/compare258_avx.c) - list(APPEND AVX2_SRCS ${ARCHDIR}/adler32_avx.c) + add_definitions(-DX86_AVX2) + set(AVX2_SRCS ${ARCHDIR}/slide_hash_avx2.c) + list(APPEND AVX2_SRCS ${ARCHDIR}/chunkset_avx2.c) + list(APPEND AVX2_SRCS ${ARCHDIR}/compare256_avx2.c) + list(APPEND AVX2_SRCS ${ARCHDIR}/adler32_avx2.c) list(APPEND ZLIB_ARCH_SRCS ${AVX2_SRCS}) endif() if(ENABLE_SSE42) - add_definitions(-DX86_SSE42_CRC_HASH) - set(SSE42_SRCS ${ARCHDIR}/insert_string_sse.c) - list(APPEND ZLIB_ARCH_SRCS ${SSE42_SRCS}) - add_definitions(-DX86_SSE42_CRC_INTRIN) - add_definitions(-DX86_SSE42_CMP_STR) - set(SSE42_SRCS ${ARCHDIR}/compare258_sse.c) + add_definitions(-DX86_SSE42) + set(SSE42_SRCS ${ARCHDIR}/adler32_sse42.c ${ARCHDIR}/insert_string_sse42.c) list(APPEND ZLIB_ARCH_SRCS ${SSE42_SRCS}) endif() if(ENABLE_SSSE3) - add_definitions(-DX86_SSSE3 -DX86_SSSE3_ADLER32) - set(SSSE3_SRCS ${ARCHDIR}/adler32_ssse3.c) + add_definitions(-DX86_SSSE3) + set(SSSE3_SRCS ${ARCHDIR}/adler32_ssse3.c ${ARCHDIR}/chunkset_ssse3.c) list(APPEND ZLIB_ARCH_SRCS ${SSSE3_SRCS}) endif() if(ENABLE_PCLMULQDQ) add_definitions(-DX86_PCLMULQDQ_CRC) - set(PCLMULQDQ_SRCS ${ARCHDIR}/crc_folding.c) + set(PCLMULQDQ_SRCS ${ARCHDIR}/crc32_pclmulqdq.c) list(APPEND ZLIB_ARCH_SRCS ${PCLMULQDQ_SRCS}) endif() - add_definitions(-DX86_SSE2 -DX86_SSE2_CHUNKSET -DX86_SSE2_SLIDEHASH) - set(SSE2_SRCS ${ARCHDIR}/chunkset_sse.c ${ARCHDIR}/slide_sse.c) + add_definitions(-DX86_SSE2) + set(SSE2_SRCS ${ARCHDIR}/chunkset_sse2.c ${ARCHDIR}/compare256_sse2.c ${ARCHDIR}/slide_hash_sse2.c) list(APPEND ZLIB_ARCH_SRCS ${SSE2_SRCS}) add_definitions(-DX86_NOCHECK_SSE2) endif () @@ -106,39 +115,45 @@ generate_cmakein(${SOURCE_DIR}/zconf.h.in ${CMAKE_CURRENT_BINARY_DIR}/zconf.h.cm set(ZLIB_SRCS ${SOURCE_DIR}/adler32.c + ${SOURCE_DIR}/adler32_fold.c ${SOURCE_DIR}/chunkset.c - ${SOURCE_DIR}/compare258.c + ${SOURCE_DIR}/compare256.c ${SOURCE_DIR}/compress.c - ${SOURCE_DIR}/crc32.c - ${SOURCE_DIR}/crc32_comb.c + ${SOURCE_DIR}/cpu_features.c + ${SOURCE_DIR}/crc32_braid.c + ${SOURCE_DIR}/crc32_braid_comb.c + ${SOURCE_DIR}/crc32_fold.c ${SOURCE_DIR}/deflate.c ${SOURCE_DIR}/deflate_fast.c + ${SOURCE_DIR}/deflate_huff.c ${SOURCE_DIR}/deflate_medium.c ${SOURCE_DIR}/deflate_quick.c + ${SOURCE_DIR}/deflate_rle.c ${SOURCE_DIR}/deflate_slow.c + ${SOURCE_DIR}/deflate_stored.c ${SOURCE_DIR}/functable.c ${SOURCE_DIR}/infback.c - ${SOURCE_DIR}/inffast.c ${SOURCE_DIR}/inflate.c ${SOURCE_DIR}/inftrees.c ${SOURCE_DIR}/insert_string.c + ${SOURCE_DIR}/insert_string_roll.c + ${SOURCE_DIR}/slide_hash.c ${SOURCE_DIR}/trees.c ${SOURCE_DIR}/uncompr.c ${SOURCE_DIR}/zutil.c +) + +set(ZLIB_GZFILE_SRCS ${SOURCE_DIR}/gzlib.c - ${SOURCE_DIR}/gzread.c + ${CMAKE_CURRENT_BINARY_DIR}/gzread.c ${SOURCE_DIR}/gzwrite.c ) -set(ZLIB_ALL_SRCS ${ZLIB_SRCS} ${ZLIB_ARCH_SRCS}) +set(ZLIB_ALL_SRCS ${ZLIB_SRCS} ${ZLIB_ARCH_SRCS} ${ZLIB_GZFILE_SRCS}) add_library(_zlib ${ZLIB_ALL_SRCS}) add_library(ch_contrib::zlib ALIAS _zlib) -# https://github.com/zlib-ng/zlib-ng/pull/733 -# This is disabed by default -add_compile_definitions(Z_TLS=__thread) - if(HAVE_UNISTD_H) SET(ZCONF_UNISTD_LINE "#if 1 /* was set to #if 1 by configure/cmake/etc */") else() @@ -153,6 +168,9 @@ endif() set(ZLIB_PC ${CMAKE_CURRENT_BINARY_DIR}/zlib.pc) configure_file(${SOURCE_DIR}/zlib.pc.cmakein ${ZLIB_PC} @ONLY) configure_file(${CMAKE_CURRENT_BINARY_DIR}/zconf.h.cmakein ${CMAKE_CURRENT_BINARY_DIR}/zconf.h @ONLY) +configure_file(${SOURCE_DIR}/zlib.h.in ${CMAKE_CURRENT_BINARY_DIR}/zlib.h @ONLY) +configure_file(${SOURCE_DIR}/zlib_name_mangling.h.in ${CMAKE_CURRENT_BINARY_DIR}/zlib_name_mangling.h @ONLY) +configure_file(${SOURCE_DIR}/gzread.c.in ${CMAKE_CURRENT_BINARY_DIR}/gzread.c @ONLY) # We should use same defines when including zlib.h as used when zlib compiled target_compile_definitions (_zlib PUBLIC ZLIB_COMPAT WITH_GZFILEOP) diff --git a/docker/keeper/Dockerfile b/docker/keeper/Dockerfile index 94603763572..a44664259fb 100644 --- a/docker/keeper/Dockerfile +++ b/docker/keeper/Dockerfile @@ -34,7 +34,7 @@ RUN arch=${TARGETARCH:-amd64} \ # lts / testing / prestable / etc ARG REPO_CHANNEL="stable" ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}" -ARG VERSION="24.7.2.13" +ARG VERSION="24.7.3.42" ARG PACKAGES="clickhouse-keeper" ARG DIRECT_DOWNLOAD_URLS="" diff --git a/docker/keeper/entrypoint.sh b/docker/keeper/entrypoint.sh index 1390ad9ce74..68bd0ef9d87 100644 --- a/docker/keeper/entrypoint.sh +++ b/docker/keeper/entrypoint.sh @@ -40,8 +40,6 @@ fi DATA_DIR="${CLICKHOUSE_DATA_DIR:-/var/lib/clickhouse}" LOG_DIR="${LOG_DIR:-/var/log/clickhouse-keeper}" -LOG_PATH="${LOG_DIR}/clickhouse-keeper.log" -ERROR_LOG_PATH="${LOG_DIR}/clickhouse-keeper.err.log" COORDINATION_DIR="${DATA_DIR}/coordination" COORDINATION_LOG_DIR="${DATA_DIR}/coordination/log" COORDINATION_SNAPSHOT_DIR="${DATA_DIR}/coordination/snapshots" @@ -84,7 +82,7 @@ if [[ $# -lt 1 ]] || [[ "$1" == "--"* ]]; then # There is a config file. It is already tested with gosu (if it is readably by keeper user) if [ -f "$KEEPER_CONFIG" ]; then - exec $gosu /usr/bin/clickhouse-keeper --config-file="$KEEPER_CONFIG" --log-file="$LOG_PATH" --errorlog-file="$ERROR_LOG_PATH" "$@" + exec $gosu /usr/bin/clickhouse-keeper --config-file="$KEEPER_CONFIG" "$@" fi # There is no config file. Will use embedded one diff --git a/docker/packager/binary-builder/build.sh b/docker/packager/binary-builder/build.sh index bd5f2fe8466..ba9311c6880 100755 --- a/docker/packager/binary-builder/build.sh +++ b/docker/packager/binary-builder/build.sh @@ -108,7 +108,8 @@ if [ -n "$MAKE_DEB" ]; then bash -x /build/packages/build fi -mv ./programs/clickhouse* /output || mv ./programs/*_fuzzer /output +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 diff --git a/docker/packager/cctools/Dockerfile b/docker/packager/cctools/Dockerfile index d986c6a3c86..570a42d42d5 100644 --- a/docker/packager/cctools/Dockerfile +++ b/docker/packager/cctools/Dockerfile @@ -1,3 +1,5 @@ +# docker build -t clickhouse/cctools . + # This is a hack to significantly reduce the build time of the clickhouse/binary-builder # It's based on the assumption that we don't care of the cctools version so much # It event does not depend on the clickhouse/fasttest in the `docker/images.json` @@ -30,5 +32,29 @@ RUN git clone https://github.com/tpoechtrager/cctools-port.git \ && cd ../.. \ && rm -rf cctools-port +# +# GDB +# +# ld from binutils is 2.38, which has the following error: +# +# DWARF error: invalid or unhandled FORM value: 0x23 +# +ENV LD=ld.lld-${LLVM_VERSION} +ARG GDB_VERSION=15.1 +RUN apt-get update \ + && apt-get install --yes \ + libgmp-dev \ + libmpfr-dev \ + && apt-get clean \ + && rm -rf /var/lib/apt/lists/* /var/cache/debconf /tmp/* +RUN wget https://sourceware.org/pub/gdb/releases/gdb-$GDB_VERSION.tar.gz \ + && tar -xvf gdb-$GDB_VERSION.tar.gz \ + && cd gdb-$GDB_VERSION \ + && ./configure --prefix=/opt/gdb \ + && make -j $(nproc) \ + && make install \ + && rm -fr gdb-$GDB_VERSION gdb-$GDB_VERSION.tar.gz + FROM scratch COPY --from=builder /cctools /cctools +COPY --from=builder /opt/gdb /opt/gdb diff --git a/docker/server/Dockerfile.alpine b/docker/server/Dockerfile.alpine index f40118c7b06..2565828c846 100644 --- a/docker/server/Dockerfile.alpine +++ b/docker/server/Dockerfile.alpine @@ -32,7 +32,7 @@ RUN arch=${TARGETARCH:-amd64} \ # lts / testing / prestable / etc ARG REPO_CHANNEL="stable" ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}" -ARG VERSION="24.7.2.13" +ARG VERSION="24.7.3.42" ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static" ARG DIRECT_DOWNLOAD_URLS="" diff --git a/docker/server/Dockerfile.ubuntu b/docker/server/Dockerfile.ubuntu index 032aa862e4a..5ac8a58afea 100644 --- a/docker/server/Dockerfile.ubuntu +++ b/docker/server/Dockerfile.ubuntu @@ -28,7 +28,7 @@ RUN sed -i "s|http://archive.ubuntu.com|${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="24.7.2.13" +ARG VERSION="24.7.3.42" 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 e7ac62604dc..848da4ef5b0 100644 --- a/docker/test/base/Dockerfile +++ b/docker/test/base/Dockerfile @@ -28,12 +28,14 @@ RUN echo "TSAN_OPTIONS='verbosity=1000 halt_on_error=1 abort_on_error=1 history_ 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 +RUN echo "ASAN_OPTIONS='halt_on_error=1 abort_on_error=1'" >> /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 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' +ENV ASAN_OPTIONS='halt_on_error=1 abort_on_error=1' # for external_symbolizer_path, and also ensure that llvm-symbolizer really # exists (since you don't want to fallback to addr2line, it is very slow) diff --git a/docker/test/base/setup_export_logs.sh b/docker/test/base/setup_export_logs.sh index 0c869a95db2..e544397dd0c 100755 --- a/docker/test/base/setup_export_logs.sh +++ b/docker/test/base/setup_export_logs.sh @@ -218,6 +218,6 @@ function stop_logs_replication clickhouse-client --query "select database||'.'||table from system.tables where database = 'system' and (table like '%_sender' or table like '%_watcher')" | { tee /dev/stderr } | { - xargs -n1 -r -i clickhouse-client --query "drop table {}" + timeout --preserve-status --signal TERM --kill-after 5m 15m xargs -n1 -r -i clickhouse-client --query "drop table {}" } } diff --git a/docker/test/fasttest/Dockerfile b/docker/test/fasttest/Dockerfile index 4cac2ee6135..5d311c673a4 100644 --- a/docker/test/fasttest/Dockerfile +++ b/docker/test/fasttest/Dockerfile @@ -83,7 +83,7 @@ RUN arch=${TARGETARCH:-amd64} \ # Give suid to gdb to grant it attach permissions # chmod 777 to make the container user independent -RUN chmod u+s /usr/bin/gdb \ +RUN chmod u+s /opt/gdb/bin/gdb \ && mkdir -p /var/lib/clickhouse \ && chmod 777 /var/lib/clickhouse diff --git a/docker/test/fasttest/run.sh b/docker/test/fasttest/run.sh index 26283afc86a..394d31addb1 100755 --- a/docker/test/fasttest/run.sh +++ b/docker/test/fasttest/run.sh @@ -41,7 +41,7 @@ export FASTTEST_WORKSPACE export FASTTEST_SOURCE export FASTTEST_BUILD export FASTTEST_DATA -export FASTTEST_OUT +export FASTTEST_OUTPUT export PATH function ccache_status diff --git a/docker/test/fuzzer/query-fuzzer-tweaks-users.xml b/docker/test/fuzzer/query-fuzzer-tweaks-users.xml index d5b876a4c85..476464e9cc2 100644 --- a/docker/test/fuzzer/query-fuzzer-tweaks-users.xml +++ b/docker/test/fuzzer/query-fuzzer-tweaks-users.xml @@ -28,9 +28,9 @@ - + - + diff --git a/docker/test/fuzzer/run-fuzzer.sh b/docker/test/fuzzer/run-fuzzer.sh index b8f967ed9c2..ae1b9e94bed 100755 --- a/docker/test/fuzzer/run-fuzzer.sh +++ b/docker/test/fuzzer/run-fuzzer.sh @@ -193,53 +193,60 @@ function fuzz kill -0 $server_pid - # Set follow-fork-mode to parent, because we attach to clickhouse-server, not to watchdog - # and clickhouse-server can do fork-exec, for example, to run some bridge. - # Do not set nostop noprint for all signals, because some it may cause gdb to hang, - # explicitly ignore non-fatal signals that are used by server. - # Number of SIGRTMIN can be determined only in runtime. - RTMIN=$(kill -l SIGRTMIN) - echo " -set follow-fork-mode parent -handle SIGHUP nostop noprint pass -handle SIGINT nostop noprint pass -handle SIGQUIT nostop noprint pass -handle SIGPIPE nostop noprint pass -handle SIGTERM nostop noprint pass -handle SIGUSR1 nostop noprint pass -handle SIGUSR2 nostop noprint pass -handle SIG$RTMIN nostop noprint pass -info signals -continue -backtrace full -thread apply all backtrace full -info registers -disassemble /s -up -disassemble /s -up -disassemble /s -p \"done\" -detach -quit -" > script.gdb + IS_ASAN=$(clickhouse-client --query "SELECT count() FROM system.build_options WHERE name = 'CXX_FLAGS' AND position('sanitize=address' IN value)") + if [[ "$IS_ASAN" = "1" ]]; + then + echo "ASAN build detected. Not using gdb since it disables LeakSanitizer detections" + else + # Set follow-fork-mode to parent, because we attach to clickhouse-server, not to watchdog + # and clickhouse-server can do fork-exec, for example, to run some bridge. + # Do not set nostop noprint for all signals, because some it may cause gdb to hang, + # explicitly ignore non-fatal signals that are used by server. + # Number of SIGRTMIN can be determined only in runtime. + RTMIN=$(kill -l SIGRTMIN) + echo " + set follow-fork-mode parent + handle SIGHUP nostop noprint pass + handle SIGINT nostop noprint pass + handle SIGQUIT nostop noprint pass + handle SIGPIPE nostop noprint pass + handle SIGTERM nostop noprint pass + handle SIGUSR1 nostop noprint pass + handle SIGUSR2 nostop noprint pass + handle SIG$RTMIN nostop noprint pass + info signals + continue + backtrace full + thread apply all backtrace full + info registers + disassemble /s + up + disassemble /s + up + disassemble /s + p \"done\" + detach + quit + " > script.gdb - gdb -batch -command script.gdb -p $server_pid & - sleep 5 - # gdb will send SIGSTOP, spend some time loading debug info, and then send SIGCONT, wait for it (up to send_timeout, 300s) - time clickhouse-client --query "SELECT 'Connected to clickhouse-server after attaching gdb'" ||: + gdb -batch -command script.gdb -p $server_pid & + sleep 5 + # gdb will send SIGSTOP, spend some time loading debug info, and then send SIGCONT, wait for it (up to send_timeout, 300s) + time clickhouse-client --query "SELECT 'Connected to clickhouse-server after attaching gdb'" ||: + + # Check connectivity after we attach gdb, because it might cause the server + # to freeze, and the fuzzer will fail. In debug build, it can take a lot of time. + for _ in {1..180} + do + if clickhouse-client --query "select 1" + then + break + fi + sleep 1 + done + kill -0 $server_pid # This checks that it is our server that is started and not some other one + fi - # Check connectivity after we attach gdb, because it might cause the server - # to freeze, and the fuzzer will fail. In debug build, it can take a lot of time. - for _ in {1..180} - do - if clickhouse-client --query "select 1" - then - break - fi - sleep 1 - done - kill -0 $server_pid # This checks that it is our server that is started and not some other one echo 'Server started and responded.' setup_logs_replication @@ -264,8 +271,13 @@ quit # The fuzzer_pid belongs to the timeout process. actual_fuzzer_pid=$(ps -o pid= --ppid "$fuzzer_pid") - echo "Attaching gdb to the fuzzer itself" - gdb -batch -command script.gdb -p $actual_fuzzer_pid & + if [[ "$IS_ASAN" = "1" ]]; + then + echo "ASAN build detected. Not using gdb since it disables LeakSanitizer detections" + else + echo "Attaching gdb to the fuzzer itself" + gdb -batch -command script.gdb -p $actual_fuzzer_pid & + fi # Wait for the fuzzer to complete. # Note that the 'wait || ...' thing is required so that the script doesn't diff --git a/docker/test/integration/base/Dockerfile b/docker/test/integration/base/Dockerfile index 469251f648c..dc4d470a262 100644 --- a/docker/test/integration/base/Dockerfile +++ b/docker/test/integration/base/Dockerfile @@ -11,7 +11,6 @@ RUN apt-get update \ curl \ default-jre \ g++ \ - gdb \ iproute2 \ krb5-user \ libicu-dev \ @@ -73,3 +72,6 @@ maxClientCnxns=80' > /opt/zookeeper/conf/zoo.cfg && \ ENV TZ=Etc/UTC RUN ln -snf /usr/share/zoneinfo/$TZ /etc/localtime && echo $TZ > /etc/timezone + +COPY --from=clickhouse/cctools:0d6b90a7a490 /opt/gdb /opt/gdb +ENV PATH="/opt/gdb/bin:${PATH}" diff --git a/docker/test/integration/runner/Dockerfile b/docker/test/integration/runner/Dockerfile index d250b746e7d..d62009f1be3 100644 --- a/docker/test/integration/runner/Dockerfile +++ b/docker/test/integration/runner/Dockerfile @@ -30,7 +30,6 @@ RUN apt-get update \ luajit \ libssl-dev \ libcurl4-openssl-dev \ - gdb \ default-jdk \ software-properties-common \ libkrb5-dev \ @@ -87,6 +86,8 @@ COPY modprobe.sh /usr/local/bin/modprobe COPY dockerd-entrypoint.sh /usr/local/bin/ COPY misc/ /misc/ +COPY --from=clickhouse/cctools:0d6b90a7a490 /opt/gdb /opt/gdb +ENV PATH="/opt/gdb/bin:${PATH}" # Same options as in test/base/Dockerfile # (in case you need to override them in tests) diff --git a/docker/test/integration/runner/requirements.txt b/docker/test/integration/runner/requirements.txt index 8a77d8abf77..428986b5562 100644 --- a/docker/test/integration/runner/requirements.txt +++ b/docker/test/integration/runner/requirements.txt @@ -74,6 +74,7 @@ protobuf==4.25.2 psycopg2-binary==2.9.6 py4j==0.10.9.5 py==1.11.0 +pyarrow==17.0.0 pycparser==2.22 pycryptodome==3.20.0 pymongo==3.11.0 diff --git a/docker/test/performance-comparison/Dockerfile b/docker/test/performance-comparison/Dockerfile index c68a39f6f70..f7139275282 100644 --- a/docker/test/performance-comparison/Dockerfile +++ b/docker/test/performance-comparison/Dockerfile @@ -9,7 +9,6 @@ RUN apt-get update \ curl \ dmidecode \ g++ \ - gdb \ git \ gnuplot \ imagemagick \ @@ -42,6 +41,9 @@ RUN pip3 --no-cache-dir install -r requirements.txt COPY run.sh / +COPY --from=clickhouse/cctools:0d6b90a7a490 /opt/gdb /opt/gdb +ENV PATH="/opt/gdb/bin:${PATH}" + CMD ["bash", "/run.sh"] # docker run --network=host --volume :/workspace --volume=:/output -e PR_TO_TEST=<> -e SHA_TO_TEST=<> clickhouse/performance-comparison diff --git a/docker/test/stateful/run.sh b/docker/test/stateful/run.sh index 8e2f1890f89..3a4f0d97993 100755 --- a/docker/test/stateful/run.sh +++ b/docker/test/stateful/run.sh @@ -232,15 +232,26 @@ function run_tests() set +e + TEST_ARGS=( + -j 2 + --testname + --shard + --zookeeper + --check-zookeeper-session + --no-stateless + --hung-check + --print-time + --capture-client-stacktrace + "${ADDITIONAL_OPTIONS[@]}" + "$SKIP_TESTS_OPTION" + ) if [[ -n "$USE_PARALLEL_REPLICAS" ]] && [[ "$USE_PARALLEL_REPLICAS" -eq 1 ]]; then - clickhouse-test --client="clickhouse-client --allow_experimental_parallel_reading_from_replicas=1 --parallel_replicas_for_non_replicated_merge_tree=1 \ - --max_parallel_replicas=100 --cluster_for_parallel_replicas='parallel_replicas'" \ - -j 2 --testname --shard --zookeeper --check-zookeeper-session --no-stateless --no-parallel-replicas --hung-check --print-time "${ADDITIONAL_OPTIONS[@]}" \ - "$SKIP_TESTS_OPTION" 2>&1 | ts '%Y-%m-%d %H:%M:%S' | tee test_output/test_result.txt - else - clickhouse-test -j 2 --testname --shard --zookeeper --check-zookeeper-session --no-stateless --hung-check --print-time "${ADDITIONAL_OPTIONS[@]}" \ - "$SKIP_TESTS_OPTION" 2>&1 | ts '%Y-%m-%d %H:%M:%S' | tee test_output/test_result.txt + TEST_ARGS+=( + --client="clickhouse-client --allow_experimental_parallel_reading_from_replicas=1 --parallel_replicas_for_non_replicated_merge_tree=1 --max_parallel_replicas=100 --cluster_for_parallel_replicas='parallel_replicas'" + --no-parallel-replicas + ) fi + clickhouse-test "${TEST_ARGS[@]}" 2>&1 | ts '%Y-%m-%d %H:%M:%S' | tee test_output/test_result.txt set -e } diff --git a/docker/test/stateless/Dockerfile b/docker/test/stateless/Dockerfile index a0e5513a3a2..d8eb072328f 100644 --- a/docker/test/stateless/Dockerfile +++ b/docker/test/stateless/Dockerfile @@ -69,8 +69,8 @@ ENV MAX_RUN_TIME=0 # Unrelated to vars in setup_minio.sh, but should be the same there # to have the same binaries for local running scenario -ARG MINIO_SERVER_VERSION=2022-01-03T18-22-58Z -ARG MINIO_CLIENT_VERSION=2022-01-05T23-52-51Z +ARG MINIO_SERVER_VERSION=2024-08-03T04-33-23Z +ARG MINIO_CLIENT_VERSION=2024-07-31T15-58-33Z ARG TARGETARCH # Download Minio-related binaries diff --git a/docker/test/stateless/attach_gdb.lib b/docker/test/stateless/attach_gdb.lib index d288288bb17..2f1375a2f0f 100644 --- a/docker/test/stateless/attach_gdb.lib +++ b/docker/test/stateless/attach_gdb.lib @@ -5,43 +5,53 @@ source /utils.lib function attach_gdb_to_clickhouse() { - # Set follow-fork-mode to parent, because we attach to clickhouse-server, not to watchdog - # and clickhouse-server can do fork-exec, for example, to run some bridge. - # Do not set nostop noprint for all signals, because some it may cause gdb to hang, - # explicitly ignore non-fatal signals that are used by server. - # Number of SIGRTMIN can be determined only in runtime. - RTMIN=$(kill -l SIGRTMIN) - echo " -set follow-fork-mode parent -handle SIGHUP nostop noprint pass -handle SIGINT nostop noprint pass -handle SIGQUIT nostop noprint pass -handle SIGPIPE nostop noprint pass -handle SIGTERM nostop noprint pass -handle SIGUSR1 nostop noprint pass -handle SIGUSR2 nostop noprint pass -handle SIG$RTMIN nostop noprint pass -info signals -continue -backtrace full -thread apply all backtrace full -info registers -disassemble /s -up -disassemble /s -up -disassemble /s -p \"done\" -detach -quit -" > script.gdb + IS_ASAN=$(clickhouse-client --query "SELECT count() FROM system.build_options WHERE name = 'CXX_FLAGS' AND position('sanitize=address' IN value)") + if [[ "$IS_ASAN" = "1" ]]; + then + echo "ASAN build detected. Not using gdb since it disables LeakSanitizer detections" + else + # Set follow-fork-mode to parent, because we attach to clickhouse-server, not to watchdog + # and clickhouse-server can do fork-exec, for example, to run some bridge. + # Do not set nostop noprint for all signals, because some it may cause gdb to hang, + # explicitly ignore non-fatal signals that are used by server. + # Number of SIGRTMIN can be determined only in runtime. + RTMIN=$(kill -l SIGRTMIN) + # shellcheck disable=SC2016 + echo " + set follow-fork-mode parent + handle SIGHUP nostop noprint pass + handle SIGINT nostop noprint pass + handle SIGQUIT nostop noprint pass + handle SIGPIPE nostop noprint pass + handle SIGTERM nostop noprint pass + handle SIGUSR1 nostop noprint pass + handle SIGUSR2 nostop noprint pass + handle SIG$RTMIN nostop noprint pass + info signals + continue + backtrace full + info registers + p "top 1 KiB of the stack:" + p/x *(uint64_t[128]*)"'$sp'" + maintenance info sections + thread apply all backtrace full + disassemble /s + up + disassemble /s + up + disassemble /s + p \"done\" + detach + quit + " > script.gdb - # FIXME Hung check may work incorrectly because of attached gdb - # We cannot attach another gdb to get stacktraces if some queries hung - gdb -batch -command script.gdb -p "$(cat /var/run/clickhouse-server/clickhouse-server.pid)" | ts '%Y-%m-%d %H:%M:%S' >> /test_output/gdb.log & - sleep 5 - # gdb will send SIGSTOP, spend some time loading debug info and then send SIGCONT, wait for it (up to send_timeout, 300s) - run_with_retry 60 clickhouse-client --query "SELECT 'Connected to clickhouse-server after attaching gdb'" + # FIXME Hung check may work incorrectly because of attached gdb + # We cannot attach another gdb to get stacktraces if some queries hung + gdb -batch -command script.gdb -p "$(cat /var/run/clickhouse-server/clickhouse-server.pid)" | ts '%Y-%m-%d %H:%M:%S' >> /test_output/gdb.log & + sleep 5 + # gdb will send SIGSTOP, spend some time loading debug info and then send SIGCONT, wait for it (up to send_timeout, 300s) + run_with_retry 60 clickhouse-client --query "SELECT 'Connected to clickhouse-server after attaching gdb'" + fi } # vi: ft=bash diff --git a/docker/test/stateless/run.sh b/docker/test/stateless/run.sh index ea32df23af0..c70cbe1fe45 100755 --- a/docker/test/stateless/run.sh +++ b/docker/test/stateless/run.sh @@ -54,8 +54,6 @@ 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 @@ -174,7 +172,56 @@ do done setup_logs_replication -attach_gdb_to_clickhouse || true # FIXME: to not break old builds, clean on 2023-09-01 +attach_gdb_to_clickhouse + +# create tables for minio log webhooks +clickhouse-client --query "CREATE TABLE minio_audit_logs +( + log String, + event_time DateTime64(9) MATERIALIZED parseDateTime64BestEffortOrZero(trim(BOTH '\"' FROM JSONExtractRaw(log, 'time')), 9, 'UTC') +) +ENGINE = MergeTree +ORDER BY tuple()" + +clickhouse-client --query "CREATE TABLE minio_server_logs +( + log String, + event_time DateTime64(9) MATERIALIZED parseDateTime64BestEffortOrZero(trim(BOTH '\"' FROM JSONExtractRaw(log, 'time')), 9, 'UTC') +) +ENGINE = MergeTree +ORDER BY tuple()" + +# create minio log webhooks for both audit and server logs +# use async inserts to avoid creating too many parts +./mc admin config set clickminio logger_webhook:ch_server_webhook endpoint="http://localhost:8123/?async_insert=1&wait_for_async_insert=0&async_insert_busy_timeout_min_ms=5000&async_insert_busy_timeout_max_ms=5000&async_insert_max_query_number=1000&async_insert_max_data_size=10485760&query=INSERT%20INTO%20minio_server_logs%20FORMAT%20LineAsString" queue_size=1000000 batch_size=500 +./mc admin config set clickminio audit_webhook:ch_audit_webhook endpoint="http://localhost:8123/?async_insert=1&wait_for_async_insert=0&async_insert_busy_timeout_min_ms=5000&async_insert_busy_timeout_max_ms=5000&async_insert_max_query_number=1000&async_insert_max_data_size=10485760&query=INSERT%20INTO%20minio_audit_logs%20FORMAT%20LineAsString" queue_size=1000000 batch_size=500 + +max_retries=100 +retry=1 +while [ $retry -le $max_retries ]; do + echo "clickminio restart attempt $retry:" + + output=$(./mc admin service restart clickminio --wait --json 2>&1 | jq -r .status) + echo "Output of restart status: $output" + + expected_output="success +success" + if [ "$output" = "$expected_output" ]; then + echo "Restarted clickminio successfully." + break + fi + + sleep 1 + + retry=$((retry + 1)) +done + +if [ $retry -gt $max_retries ]; then + echo "Failed to restart clickminio after $max_retries attempts." +fi + +./mc admin trace clickminio > /test_output/minio.log & +MC_ADMIN_PID=$! function fn_exists() { declare -F "$1" > /dev/null; @@ -264,11 +311,22 @@ function run_tests() TIMEOUT=$((MAX_RUN_TIME - 800 > 8400 ? 8400 : MAX_RUN_TIME - 800)) START_TIME=${SECONDS} set +e - 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" "${ADDITIONAL_OPTIONS[@]}" 2>&1 \ - | ts '%Y-%m-%d %H:%M:%S' \ - | tee -a test_output/test_result.txt + + TEST_ARGS=( + --testname + --shard + --zookeeper + --check-zookeeper-session + --hung-check + --print-time + --no-drop-if-fail + --capture-client-stacktrace + --test-runs "$NUM_TRIES" + "${ADDITIONAL_OPTIONS[@]}" + ) + timeout --preserve-status --signal TERM --kill-after 60m ${TIMEOUT}s clickhouse-test "${TEST_ARGS[@]}" 2>&1 \ + | ts '%Y-%m-%d %H:%M:%S' \ + | tee -a test_output/test_result.txt set -e DURATION=$((SECONDS - START_TIME)) @@ -328,6 +386,14 @@ do fi done + +# collect minio audit and server logs +# wait for minio to flush its batch if it has any +sleep 1 +clickhouse-client -q "SYSTEM FLUSH ASYNC INSERT QUEUE" +clickhouse-client -q "SELECT log FROM minio_audit_logs ORDER BY event_time INTO OUTFILE '/test_output/minio_audit_logs.jsonl.zst' FORMAT JSONEachRow" +clickhouse-client -q "SELECT log FROM minio_server_logs ORDER BY event_time INTO OUTFILE '/test_output/minio_server_logs.jsonl.zst' FORMAT JSONEachRow" + # Stop server so we can safely read data with clickhouse-local. # Why do we read data with clickhouse-local? # Because it's the simplest way to read it when server has crashed. diff --git a/docker/test/stateless/setup_minio.sh b/docker/test/stateless/setup_minio.sh index 2b9433edd20..c1508df7e82 100755 --- a/docker/test/stateless/setup_minio.sh +++ b/docker/test/stateless/setup_minio.sh @@ -59,8 +59,8 @@ find_os() { download_minio() { local os local arch - local minio_server_version=${MINIO_SERVER_VERSION:-2022-09-07T22-25-02Z} - local minio_client_version=${MINIO_CLIENT_VERSION:-2022-08-28T20-08-11Z} + local minio_server_version=${MINIO_SERVER_VERSION:-2024-08-03T04-33-23Z} + local minio_client_version=${MINIO_CLIENT_VERSION:-2024-07-31T15-58-33Z} os=$(find_os) arch=$(find_arch) @@ -82,10 +82,10 @@ setup_minio() { local test_type=$1 ./mc alias set clickminio http://localhost:11111 clickhouse clickhouse ./mc admin user add clickminio test testtest - ./mc admin policy set clickminio readwrite user=test + ./mc admin policy attach clickminio readwrite --user=test ./mc mb --ignore-existing clickminio/test if [ "$test_type" = "stateless" ]; then - ./mc policy set public clickminio/test + ./mc anonymous set public clickminio/test fi } @@ -99,10 +99,9 @@ upload_data() { # iterating over globs will cause redundant file variable to be # a path to a file, not a filename # shellcheck disable=SC2045 - for file in $(ls "${data_path}"); do - echo "${file}"; - ./mc cp "${data_path}"/"${file}" clickminio/test/"${file}"; - done + if [ -d "${data_path}" ]; then + ./mc cp --recursive "${data_path}"/ clickminio/test/ + fi } setup_aws_credentials() { @@ -148,4 +147,4 @@ main() { setup_aws_credentials } -main "$@" \ No newline at end of file +main "$@" diff --git a/docker/test/stateless/stress_tests.lib b/docker/test/stateless/stress_tests.lib index 36782101fa7..51aa299f7a6 100644 --- a/docker/test/stateless/stress_tests.lib +++ b/docker/test/stateless/stress_tests.lib @@ -139,9 +139,9 @@ EOL - + - + @@ -308,7 +308,8 @@ function collect_query_and_trace_logs() { for table in query_log trace_log metric_log do - clickhouse-local --config-file=/etc/clickhouse-server/config.xml --only-system-tables -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.tsv.zst ||: + # Don't ignore errors here, it leads to ignore sanitizer reports when running clickhouse-local + clickhouse-local --config-file=/etc/clickhouse-server/config.xml --only-system-tables -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.tsv.zst done } diff --git a/docker/test/unit/Dockerfile b/docker/test/unit/Dockerfile index af44dc930b2..9f4b86aa0ca 100644 --- a/docker/test/unit/Dockerfile +++ b/docker/test/unit/Dockerfile @@ -4,4 +4,5 @@ ARG FROM_TAG=latest FROM clickhouse/test-base:$FROM_TAG COPY run.sh / -CMD ["/bin/bash", "/run.sh"] +RUN chmod +x run.sh +ENTRYPOINT ["/run.sh"] diff --git a/docker/test/unit/run.sh b/docker/test/unit/run.sh index 7323c384d9c..210f31170a5 100644 --- a/docker/test/unit/run.sh +++ b/docker/test/unit/run.sh @@ -1,5 +1,27 @@ #!/bin/bash set -x +# Need to keep error from tests after `tee`. Otherwise we don't alert on asan errors +set -o pipefail +set -e -timeout 40m gdb -q -ex 'set print inferior-events off' -ex 'set confirm off' -ex 'set print thread-events off' -ex run -ex bt -ex quit --args ./unit_tests_dbms --gtest_output='json:test_output/test_result.json' | tee test_output/test_result.txt +if [ "$#" -ne 1 ]; then + echo "Expected exactly one argument" + exit 1 +fi + +if [ "$1" = "GDB" ]; +then + timeout 40m \ + gdb -q -ex "set print inferior-events off" -ex "set confirm off" -ex "set print thread-events off" -ex run -ex bt -ex quit --args \ + ./unit_tests_dbms --gtest_output='json:test_output/test_result.json' \ + | tee test_output/test_result.txt +elif [ "$1" = "NO_GDB" ]; +then + timeout 40m \ + ./unit_tests_dbms --gtest_output='json:test_output/test_result.json' \ + | tee test_output/test_result.txt +else + echo "Unknown argument: $1" + exit 1 +fi diff --git a/docker/test/upgrade/run.sh b/docker/test/upgrade/run.sh index a4c4c75e5b3..e9768873f6a 100644 --- a/docker/test/upgrade/run.sh +++ b/docker/test/upgrade/run.sh @@ -129,6 +129,7 @@ configure # Check that all new/changed setting were added in settings changes history. # Some settings can be different for builds with sanitizers, so we check +# Also the automatic value of 'max_threads' and similar was displayed as "'auto(...)'" in previous versions instead of "auto(...)". # settings changes only for non-sanitizer builds. IS_SANITIZED=$(clickhouse-local --query "SELECT value LIKE '%-fsanitize=%' FROM system.build_options WHERE name = 'CXX_FLAGS'") if [ "${IS_SANITIZED}" -eq "0" ] @@ -145,7 +146,9 @@ then old_settings.value AS old_value FROM new_settings LEFT JOIN old_settings ON new_settings.name = old_settings.name - WHERE (new_settings.value != old_settings.value) AND (name NOT IN ( + WHERE (new_value != old_value) + AND NOT (startsWith(new_value, 'auto(') AND old_value LIKE '%auto(%') + AND (name NOT IN ( SELECT arrayJoin(tupleElement(changes, 'name')) FROM ( @@ -177,7 +180,7 @@ then if [ -s changed_settings.txt ] then mv changed_settings.txt /test_output/ - echo -e "Changed settings are not reflected in settings changes history (see changed_settings.txt)$FAIL$(head_escaped /test_output/changed_settings.txt)" >> /test_output/test_results.tsv + echo -e "Changed settings are not reflected in the settings changes history (see changed_settings.txt)$FAIL$(head_escaped /test_output/changed_settings.txt)" >> /test_output/test_results.tsv else echo -e "There are no changed settings or they are reflected in settings changes history$OK" >> /test_output/test_results.tsv fi diff --git a/docker/test/util/Dockerfile b/docker/test/util/Dockerfile index dc928ba7195..8b949ed95db 100644 --- a/docker/test/util/Dockerfile +++ b/docker/test/util/Dockerfile @@ -44,7 +44,6 @@ RUN apt-get update \ bash \ bsdmainutils \ build-essential \ - gdb \ git \ gperf \ moreutils \ @@ -58,3 +57,6 @@ RUN apt-get update \ && rm -rf /var/lib/apt/lists/* /var/cache/debconf /tmp/* COPY process_functional_tests_result.py / + +COPY --from=clickhouse/cctools:0d6b90a7a490 /opt/gdb /opt/gdb +ENV PATH="/opt/gdb/bin:${PATH}" diff --git a/docs/changelogs/v24.3.7.30-lts.md b/docs/changelogs/v24.3.7.30-lts.md new file mode 100644 index 00000000000..f945a54840f --- /dev/null +++ b/docs/changelogs/v24.3.7.30-lts.md @@ -0,0 +1,29 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.3.7.30-lts (c8a28cf4331) FIXME as compared to v24.3.6.48-lts (b2d33c3c45d) + +#### Improvement +* Backported in [#68103](https://github.com/ClickHouse/ClickHouse/issues/68103): Distinguish booleans and integers while parsing values for custom settings: ``` SET custom_a = true; SET custom_b = 1; ```. [#62206](https://github.com/ClickHouse/ClickHouse/pull/62206) ([Vitaly Baranov](https://github.com/vitlibar)). + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Backported in [#67931](https://github.com/ClickHouse/ClickHouse/issues/67931): 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)). +* Backported in [#68062](https://github.com/ClickHouse/ClickHouse/issues/68062): Fix boolean literals in query sent to external database (for engines like `PostgreSQL`). [#66282](https://github.com/ClickHouse/ClickHouse/pull/66282) ([vdimir](https://github.com/vdimir)). +* Backported in [#67812](https://github.com/ClickHouse/ClickHouse/issues/67812): Only relevant to the experimental Variant data type. Fix crash with Variant + AggregateFunction type. [#67122](https://github.com/ClickHouse/ClickHouse/pull/67122) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#67848](https://github.com/ClickHouse/ClickHouse/issues/67848): Fixes [#66026](https://github.com/ClickHouse/ClickHouse/issues/66026). Avoid unresolved table function arguments traversal in `ReplaceTableNodeToDummyVisitor`. [#67522](https://github.com/ClickHouse/ClickHouse/pull/67522) ([Dmitry Novik](https://github.com/novikd)). +* Backported in [#68271](https://github.com/ClickHouse/ClickHouse/issues/68271): Fix inserting into stream like engines (Kafka, RabbitMQ, NATS) through HTTP interface. [#67554](https://github.com/ClickHouse/ClickHouse/pull/67554) ([JĆ”nos Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#67806](https://github.com/ClickHouse/ClickHouse/issues/67806): Fix reloading SQL UDFs with UNION. Previously, restarting the server could make UDF invalid. [#67665](https://github.com/ClickHouse/ClickHouse/pull/67665) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#67834](https://github.com/ClickHouse/ClickHouse/issues/67834): Fix potential stack overflow in `JSONMergePatch` function. Renamed this function from `jsonMergePatch` to `JSONMergePatch` because the previous name was wrong. The previous name is still kept for compatibility. Improved diagnostic of errors in the function. This closes [#67304](https://github.com/ClickHouse/ClickHouse/issues/67304). [#67756](https://github.com/ClickHouse/ClickHouse/pull/67756) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Backported in [#68206](https://github.com/ClickHouse/ClickHouse/issues/68206): Fix wrong `count()` result when there is non-deterministic function in predicate. [#67922](https://github.com/ClickHouse/ClickHouse/pull/67922) ([JĆ”nos Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#68089](https://github.com/ClickHouse/ClickHouse/issues/68089): Fixed the calculation of the maximum thread soft limit in containerized environments where the usable CPU count is limited. [#67963](https://github.com/ClickHouse/ClickHouse/pull/67963) ([Robert Schulze](https://github.com/rschu1ze)). +* Backported in [#68120](https://github.com/ClickHouse/ClickHouse/issues/68120): Fixed skipping of untouched parts in mutations with new analyzer. Previously with enabled analyzer data in part could be rewritten by mutation even if mutation doesn't affect this part according to predicate. [#68052](https://github.com/ClickHouse/ClickHouse/pull/68052) ([Anton Popov](https://github.com/CurtizJ)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Update version after release. [#67676](https://github.com/ClickHouse/ClickHouse/pull/67676) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Backported in [#68074](https://github.com/ClickHouse/ClickHouse/issues/68074): Add an explicit error for `ALTER MODIFY SQL SECURITY` on non-view tables. [#67953](https://github.com/ClickHouse/ClickHouse/pull/67953) ([pufit](https://github.com/pufit)). + diff --git a/docs/changelogs/v24.6.3.95-stable.md b/docs/changelogs/v24.6.3.95-stable.md new file mode 100644 index 00000000000..b90b7346e40 --- /dev/null +++ b/docs/changelogs/v24.6.3.95-stable.md @@ -0,0 +1,67 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.6.3.95-stable (8325c920d11) FIXME as compared to v24.6.2.17-stable (5710a8b5c0c) + +#### Improvement +* Backported in [#66770](https://github.com/ClickHouse/ClickHouse/issues/66770): 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)). + +#### Critical Bug Fix (crash, LOGICAL_ERROR, data loss, RBAC) +* Backported in [#66885](https://github.com/ClickHouse/ClickHouse/issues/66885): Fix unexpeced size of low cardinality column in function calls. [#65298](https://github.com/ClickHouse/ClickHouse/pull/65298) ([RaĆŗl MarĆ­n](https://github.com/Algunenano)). +* Backported in [#66693](https://github.com/ClickHouse/ClickHouse/issues/66693): Fix the VALID UNTIL clause in the user definition resetting after a restart. Closes [#66405](https://github.com/ClickHouse/ClickHouse/issues/66405). [#66409](https://github.com/ClickHouse/ClickHouse/pull/66409) ([Nikolay Degterinsky](https://github.com/evillique)). +* Backported in [#67816](https://github.com/ClickHouse/ClickHouse/issues/67816): Only relevant to the experimental Variant data type. Fix crash with Variant + AggregateFunction type. [#67122](https://github.com/ClickHouse/ClickHouse/pull/67122) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#67503](https://github.com/ClickHouse/ClickHouse/issues/67503): Fix crash in DistributedAsyncInsert when connection is empty. [#67219](https://github.com/ClickHouse/ClickHouse/pull/67219) ([Pablo Marcos](https://github.com/pamarcos)). +* Backported in [#67852](https://github.com/ClickHouse/ClickHouse/issues/67852): Fixes [#66026](https://github.com/ClickHouse/ClickHouse/issues/66026). Avoid unresolved table function arguments traversal in `ReplaceTableNodeToDummyVisitor`. [#67522](https://github.com/ClickHouse/ClickHouse/pull/67522) ([Dmitry Novik](https://github.com/novikd)). +* Backported in [#67838](https://github.com/ClickHouse/ClickHouse/issues/67838): Fix potential stack overflow in `JSONMergePatch` function. Renamed this function from `jsonMergePatch` to `JSONMergePatch` because the previous name was wrong. The previous name is still kept for compatibility. Improved diagnostic of errors in the function. This closes [#67304](https://github.com/ClickHouse/ClickHouse/issues/67304). [#67756](https://github.com/ClickHouse/ClickHouse/pull/67756) ([Alexey Milovidov](https://github.com/alexey-milovidov)). + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Backported in [#66303](https://github.com/ClickHouse/ClickHouse/issues/66303): 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)). +* Backported in [#66330](https://github.com/ClickHouse/ClickHouse/issues/66330): 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)). +* Backported in [#66157](https://github.com/ClickHouse/ClickHouse/issues/66157): Fixed buffer overflow bug in `unbin`/`unhex` implementation. [#66106](https://github.com/ClickHouse/ClickHouse/pull/66106) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#66210](https://github.com/ClickHouse/ClickHouse/issues/66210): 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)). +* Backported in [#66456](https://github.com/ClickHouse/ClickHouse/issues/66456): 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)). +* Backported in [#66228](https://github.com/ClickHouse/ClickHouse/issues/66228): Fix issue in SumIfToCountIfVisitor and signed integers. [#66146](https://github.com/ClickHouse/ClickHouse/pull/66146) ([RaĆŗl MarĆ­n](https://github.com/Algunenano)). +* Backported in [#66183](https://github.com/ClickHouse/ClickHouse/issues/66183): Fix rare case with missing data in the result of distributed query, close [#61432](https://github.com/ClickHouse/ClickHouse/issues/61432). [#66174](https://github.com/ClickHouse/ClickHouse/pull/66174) ([vdimir](https://github.com/vdimir)). +* Backported in [#66271](https://github.com/ClickHouse/ClickHouse/issues/66271): 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)). +* Backported in [#66682](https://github.com/ClickHouse/ClickHouse/issues/66682): Fix handling limit for `system.numbers_mt` when no index can be used. [#66231](https://github.com/ClickHouse/ClickHouse/pull/66231) ([JĆ”nos Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#66587](https://github.com/ClickHouse/ClickHouse/issues/66587): 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)). +* Backported in [#66362](https://github.com/ClickHouse/ClickHouse/issues/66362): 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)). +* Backported in [#66613](https://github.com/ClickHouse/ClickHouse/issues/66613): 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)). +* Backported in [#66577](https://github.com/ClickHouse/ClickHouse/issues/66577): 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)). +* Backported in [#66721](https://github.com/ClickHouse/ClickHouse/issues/66721): Correctly track memory for `Allocator::realloc`. [#66548](https://github.com/ClickHouse/ClickHouse/pull/66548) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#66670](https://github.com/ClickHouse/ClickHouse/issues/66670): Fix reading of uninitialized memory when hashing empty tuples. This closes [#66559](https://github.com/ClickHouse/ClickHouse/issues/66559). [#66562](https://github.com/ClickHouse/ClickHouse/pull/66562) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Backported in [#66952](https://github.com/ClickHouse/ClickHouse/issues/66952): 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)). +* Backported in [#66956](https://github.com/ClickHouse/ClickHouse/issues/66956): Fix removing named collections in local storage. [#66599](https://github.com/ClickHouse/ClickHouse/pull/66599) ([JĆ”nos Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#66716](https://github.com/ClickHouse/ClickHouse/issues/66716): Fix removing named collections in local storage. [#66599](https://github.com/ClickHouse/ClickHouse/pull/66599) ([JĆ”nos Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Backported in [#66759](https://github.com/ClickHouse/ClickHouse/issues/66759): 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)). +* Backported in [#66751](https://github.com/ClickHouse/ClickHouse/issues/66751): 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)). +* Backported in [#67635](https://github.com/ClickHouse/ClickHouse/issues/67635): Fix for occasional deadlock in Context::getDDLWorker. [#66843](https://github.com/ClickHouse/ClickHouse/pull/66843) ([Alexander Gololobov](https://github.com/davenger)). +* Backported in [#67482](https://github.com/ClickHouse/ClickHouse/issues/67482): In rare cases ClickHouse could consider parts as broken because of some unexpected projections on disk. Now it's fixed. [#66898](https://github.com/ClickHouse/ClickHouse/pull/66898) ([alesapin](https://github.com/alesapin)). +* Backported in [#67199](https://github.com/ClickHouse/ClickHouse/issues/67199): TRUNCATE DATABASE used to stop replication as if it was a DROP DATABASE query, it's fixed. [#67129](https://github.com/ClickHouse/ClickHouse/pull/67129) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Backported in [#67381](https://github.com/ClickHouse/ClickHouse/issues/67381): Fix error `Cannot convert column because it is non constant in source stream but must be constant in result.` for a query that reads from the `Merge` table over the `Distriburted` table with one shard. [#67146](https://github.com/ClickHouse/ClickHouse/pull/67146) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#67244](https://github.com/ClickHouse/ClickHouse/issues/67244): This closes [#67156](https://github.com/ClickHouse/ClickHouse/issues/67156). This closes [#66447](https://github.com/ClickHouse/ClickHouse/issues/66447). The bug was introduced in https://github.com/ClickHouse/ClickHouse/pull/62907. [#67178](https://github.com/ClickHouse/ClickHouse/pull/67178) ([Maksim Kita](https://github.com/kitaisreal)). +* Backported in [#67578](https://github.com/ClickHouse/ClickHouse/issues/67578): Fix execution of nested short-circuit functions. [#67520](https://github.com/ClickHouse/ClickHouse/pull/67520) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#67808](https://github.com/ClickHouse/ClickHouse/issues/67808): Fix reloading SQL UDFs with UNION. Previously, restarting the server could make UDF invalid. [#67665](https://github.com/ClickHouse/ClickHouse/pull/67665) ([Antonio Andelic](https://github.com/antonio2368)). + +#### NO CL ENTRY + +* NO CL ENTRY: 'Revert "Backport [#66599](https://github.com/ClickHouse/ClickHouse/issues/66599) to 24.6: Fix dropping named collection in local storage"'. [#66922](https://github.com/ClickHouse/ClickHouse/pull/66922) ([JĆ”nos Benjamin Antal](https://github.com/antaljanosbenjamin)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Backported in [#66332](https://github.com/ClickHouse/ClickHouse/issues/66332): Do not raise a NOT_IMPLEMENTED error when getting s3 metrics with a multiple disk configuration. [#65403](https://github.com/ClickHouse/ClickHouse/pull/65403) ([Elena TorrĆ³](https://github.com/elenatorro)). +* Backported in [#66142](https://github.com/ClickHouse/ClickHouse/issues/66142): Fix flaky test_storage_s3_queue tests. [#66009](https://github.com/ClickHouse/ClickHouse/pull/66009) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Backported in [#66389](https://github.com/ClickHouse/ClickHouse/issues/66389): Disable broken cases from 02911_join_on_nullsafe_optimization. [#66310](https://github.com/ClickHouse/ClickHouse/pull/66310) ([vdimir](https://github.com/vdimir)). +* Backported in [#66428](https://github.com/ClickHouse/ClickHouse/issues/66428): Ignore subquery for IN in DDLLoadingDependencyVisitor. [#66395](https://github.com/ClickHouse/ClickHouse/pull/66395) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Backported in [#66546](https://github.com/ClickHouse/ClickHouse/issues/66546): Add additional log masking in CI. [#66523](https://github.com/ClickHouse/ClickHouse/pull/66523) ([RaĆŗl MarĆ­n](https://github.com/Algunenano)). +* Backported in [#66861](https://github.com/ClickHouse/ClickHouse/issues/66861): Fix data race in S3::ClientCache. [#66644](https://github.com/ClickHouse/ClickHouse/pull/66644) ([Konstantin Morozov](https://github.com/k-morozov)). +* Backported in [#66877](https://github.com/ClickHouse/ClickHouse/issues/66877): Support one more case in JOIN ON ... IS NULL. [#66725](https://github.com/ClickHouse/ClickHouse/pull/66725) ([vdimir](https://github.com/vdimir)). +* Backported in [#67061](https://github.com/ClickHouse/ClickHouse/issues/67061): Increase asio pool size in case the server is tiny. [#66761](https://github.com/ClickHouse/ClickHouse/pull/66761) ([alesapin](https://github.com/alesapin)). +* Backported in [#66940](https://github.com/ClickHouse/ClickHouse/issues/66940): Small fix in realloc memory tracking. [#66820](https://github.com/ClickHouse/ClickHouse/pull/66820) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#67254](https://github.com/ClickHouse/ClickHouse/issues/67254): Followup [#66725](https://github.com/ClickHouse/ClickHouse/issues/66725). [#66869](https://github.com/ClickHouse/ClickHouse/pull/66869) ([vdimir](https://github.com/vdimir)). +* Backported in [#67414](https://github.com/ClickHouse/ClickHouse/issues/67414): CI: Fix build results for release branches. [#67402](https://github.com/ClickHouse/ClickHouse/pull/67402) ([Max K.](https://github.com/maxknv)). + diff --git a/docs/changelogs/v24.7.3.42-stable.md b/docs/changelogs/v24.7.3.42-stable.md new file mode 100644 index 00000000000..48f6e301f3c --- /dev/null +++ b/docs/changelogs/v24.7.3.42-stable.md @@ -0,0 +1,37 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.7.3.42-stable (63730bc4293) FIXME as compared to v24.7.2.13-stable (6e41f601b2f) + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Backported in [#67969](https://github.com/ClickHouse/ClickHouse/issues/67969): Fixed reading of subcolumns after `ALTER ADD COLUMN` query. [#66243](https://github.com/ClickHouse/ClickHouse/pull/66243) ([Anton Popov](https://github.com/CurtizJ)). +* Backported in [#67637](https://github.com/ClickHouse/ClickHouse/issues/67637): Fix for occasional deadlock in Context::getDDLWorker. [#66843](https://github.com/ClickHouse/ClickHouse/pull/66843) ([Alexander Gololobov](https://github.com/davenger)). +* Backported in [#67820](https://github.com/ClickHouse/ClickHouse/issues/67820): Fix possible deadlock on query cancel with parallel replicas. [#66905](https://github.com/ClickHouse/ClickHouse/pull/66905) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#67881](https://github.com/ClickHouse/ClickHouse/issues/67881): Correctly parse file name/URI containing `::` if it's not an archive. [#67433](https://github.com/ClickHouse/ClickHouse/pull/67433) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#67713](https://github.com/ClickHouse/ClickHouse/issues/67713): Fix reloading SQL UDFs with UNION. Previously, restarting the server could make UDF invalid. [#67665](https://github.com/ClickHouse/ClickHouse/pull/67665) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#67995](https://github.com/ClickHouse/ClickHouse/issues/67995): Validate experimental/suspicious data types in ALTER ADD/MODIFY COLUMN. [#67911](https://github.com/ClickHouse/ClickHouse/pull/67911) ([Kruglov Pavel](https://github.com/Avogar)). + +#### Critical Bug Fix (crash, LOGICAL_ERROR, data loss, RBAC) + +* Backported in [#67818](https://github.com/ClickHouse/ClickHouse/issues/67818): Only relevant to the experimental Variant data type. Fix crash with Variant + AggregateFunction type. [#67122](https://github.com/ClickHouse/ClickHouse/pull/67122) ([Kruglov Pavel](https://github.com/Avogar)). +* Backported in [#67766](https://github.com/ClickHouse/ClickHouse/issues/67766): Fix crash of `uniq` and `uniqTheta ` with `tuple()` argument. Closes [#67303](https://github.com/ClickHouse/ClickHouse/issues/67303). [#67306](https://github.com/ClickHouse/ClickHouse/pull/67306) ([flynn](https://github.com/ucasfl)). +* Backported in [#67854](https://github.com/ClickHouse/ClickHouse/issues/67854): Fixes [#66026](https://github.com/ClickHouse/ClickHouse/issues/66026). Avoid unresolved table function arguments traversal in `ReplaceTableNodeToDummyVisitor`. [#67522](https://github.com/ClickHouse/ClickHouse/pull/67522) ([Dmitry Novik](https://github.com/novikd)). +* Backported in [#67840](https://github.com/ClickHouse/ClickHouse/issues/67840): Fix potential stack overflow in `JSONMergePatch` function. Renamed this function from `jsonMergePatch` to `JSONMergePatch` because the previous name was wrong. The previous name is still kept for compatibility. Improved diagnostic of errors in the function. This closes [#67304](https://github.com/ClickHouse/ClickHouse/issues/67304). [#67756](https://github.com/ClickHouse/ClickHouse/pull/67756) ([Alexey Milovidov](https://github.com/alexey-milovidov)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Backported in [#67518](https://github.com/ClickHouse/ClickHouse/issues/67518): Split slow test 03036_dynamic_read_subcolumns. [#66954](https://github.com/ClickHouse/ClickHouse/pull/66954) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#67516](https://github.com/ClickHouse/ClickHouse/issues/67516): Split 01508_partition_pruning_long. [#66983](https://github.com/ClickHouse/ClickHouse/pull/66983) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#67529](https://github.com/ClickHouse/ClickHouse/issues/67529): Reduce max time of 00763_long_lock_buffer_alter_destination_table. [#67185](https://github.com/ClickHouse/ClickHouse/pull/67185) ([RaĆŗl MarĆ­n](https://github.com/Algunenano)). +* Backported in [#67643](https://github.com/ClickHouse/ClickHouse/issues/67643): [Green CI] Fix potentially flaky test_mask_sensitive_info integration test. [#67506](https://github.com/ClickHouse/ClickHouse/pull/67506) ([Alexey Katsman](https://github.com/alexkats)). +* Backported in [#67609](https://github.com/ClickHouse/ClickHouse/issues/67609): Fix test_zookeeper_config_load_balancing after adding the xdist worker name to the instance. [#67590](https://github.com/ClickHouse/ClickHouse/pull/67590) ([Pablo Marcos](https://github.com/pamarcos)). +* Backported in [#67871](https://github.com/ClickHouse/ClickHouse/issues/67871): Fix 02434_cancel_insert_when_client_dies. [#67600](https://github.com/ClickHouse/ClickHouse/pull/67600) ([vdimir](https://github.com/vdimir)). +* Backported in [#67704](https://github.com/ClickHouse/ClickHouse/issues/67704): Fix 02910_bad_logs_level_in_local in fast tests. [#67603](https://github.com/ClickHouse/ClickHouse/pull/67603) ([RaĆŗl MarĆ­n](https://github.com/Algunenano)). +* Backported in [#67689](https://github.com/ClickHouse/ClickHouse/issues/67689): Fix 01605_adaptive_granularity_block_borders. [#67605](https://github.com/ClickHouse/ClickHouse/pull/67605) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#67827](https://github.com/ClickHouse/ClickHouse/issues/67827): Try fix 03143_asof_join_ddb_long. [#67620](https://github.com/ClickHouse/ClickHouse/pull/67620) ([Nikita Taranov](https://github.com/nickitat)). +* Backported in [#67892](https://github.com/ClickHouse/ClickHouse/issues/67892): Revert "Merge pull request [#66510](https://github.com/ClickHouse/ClickHouse/issues/66510) from canhld94/fix_trivial_count_non_deterministic_func". [#67800](https://github.com/ClickHouse/ClickHouse/pull/67800) ([JĆ”nos Benjamin Antal](https://github.com/antaljanosbenjamin)). + diff --git a/docs/en/development/architecture.md b/docs/en/development/architecture.md index 6428c0e90d5..a1a5901f859 100644 --- a/docs/en/development/architecture.md +++ b/docs/en/development/architecture.md @@ -118,7 +118,7 @@ And the result of interpreting the `INSERT SELECT` query is a "completed" `Query `InterpreterSelectQuery` uses `ExpressionAnalyzer` and `ExpressionActions` machinery for query analysis and transformations. This is where most rule-based query optimizations are performed. `ExpressionAnalyzer` is quite messy and should be rewritten: various query transformations and optimizations should be extracted into separate classes to allow for modular transformations of the query. -To address current problems that exist in interpreters, a new `InterpreterSelectQueryAnalyzer` is being developed. It is a new version of `InterpreterSelectQuery` that does not use `ExpressionAnalyzer` and introduces an additional abstraction level between `AST` and `QueryPipeline` called `QueryTree`. It is not production-ready yet, but it can be tested with the `allow_experimental_analyzer` flag. +To address problems that exist in interpreters, a new `InterpreterSelectQueryAnalyzer` has been developed. This is a new version of the `InterpreterSelectQuery`, which does not use the `ExpressionAnalyzer` and introduces an additional layer of abstraction between `AST` and `QueryPipeline`, called `QueryTree'. It is fully ready for use in production, but just in case it can be turned off by setting the value of the `enable_analyzer` setting to `false`. ## Functions {#functions} diff --git a/docs/en/development/contrib.md b/docs/en/development/contrib.md index db3eabaecfc..c49492c1cb4 100644 --- a/docs/en/development/contrib.md +++ b/docs/en/development/contrib.md @@ -27,23 +27,23 @@ Avoid dumping copies of external code into the library directory. Instead create a Git submodule to pull third-party code from an external upstream repository. All submodules used by ClickHouse are listed in the `.gitmodule` file. -If the library can be used as-is (the default case), you can reference the upstream repository directly. -If the library needs patching, create a fork of the upstream repository in the [ClickHouse organization on GitHub](https://github.com/ClickHouse). +- If the library can be used as-is (the default case), you can reference the upstream repository directly. +- If the library needs patching, create a fork of the upstream repository in the [ClickHouse organization on GitHub](https://github.com/ClickHouse). In the latter case, we aim to isolate custom patches as much as possible from upstream commits. -To that end, create a branch with prefix `clickhouse/` from the branch or tag you want to integrate, e.g. `clickhouse/master` (for branch `master`) or `clickhouse/release/vX.Y.Z` (for tag `release/vX.Y.Z`). -This ensures that pulls from the upstream repository into the fork will leave custom `clickhouse/` branches unaffected. -Submodules in `contrib/` must only track `clickhouse/` branches of forked third-party repositories. +To that end, create a branch with prefix `ClickHouse/` from the branch or tag you want to integrate, e.g. `ClickHouse/2024_2` (for branch `2024_2`) or `ClickHouse/release/vX.Y.Z` (for tag `release/vX.Y.Z`). +Avoid following upstream development branches `master`/ `main` / `dev` (i.e., prefix branches `ClickHouse/master` / `ClickHouse/main` / `ClickHouse/dev` in the fork repository). +Such branches are moving targets which make proper versioning harder. +"Prefix branches" ensure that pulls from the upstream repository into the fork will leave custom `ClickHouse/` branches unaffected. +Submodules in `contrib/` must only track `ClickHouse/` branches of forked third-party repositories. -Patches are only applied against `clickhouse/` branches of external libraries. -For that, push the patch as a branch with `clickhouse/`, e.g. `clickhouse/fix-some-desaster`. -Then create a PR from the new branch against the custom tracking branch with `clickhouse/` prefix, (e.g. `clickhouse/master` or `clickhouse/release/vX.Y.Z`) and merge the patch. +Patches are only applied against `ClickHouse/` branches of external libraries. + +There are two ways to do that: +- you like to make a new fix against a `ClickHouse/`-prefix branch in the forked repository, e.g. a sanitizer fix. In that case, push the fix as a branch with `ClickHouse/` prefix, e.g. `ClickHouse/fix-sanitizer-disaster`. Then create a PR from the new branch against the custom tracking branch, e.g. `ClickHouse/2024_2 <-- ClickHouse/fix-sanitizer-disaster` and merge the PR. +- you update the submodule and need to re-apply earlier patches. In this case, re-creating old PRs is overkill. Instead, simply cherry-pick older commits into the new `ClickHouse/` branch (corresponding to the new version). Feel free to squash commits of PRs that had multiple commits. In the best case, we did contribute custom patches back to upstream and can omit patches in the new version. + +Once the submodule has been updated, bump the submodule in ClickHouse to point to the new hash in the fork. Create patches of third-party libraries with the official repository in mind and consider contributing the patch back to the upstream repository. This makes sure that others will also benefit from the patch and it will not be a maintenance burden for the ClickHouse team. - -To pull upstream changes into the submodule, you can use two methods: -- (less work but less clean): merge upstream `master` into the corresponding `clickhouse/` tracking branch in the forked repository. You will need to resolve merge conflicts with previous custom patches. This method can be used when the `clickhouse/` branch tracks an upstream development branch like `master`, `main`, `dev`, etc. -- (more work but cleaner): create a new branch with `clickhouse/` prefix from the upstream commit or tag you like to integrate. Then re-apply all existing patches using new PRs (or squash them into a single PR). This method can be used when the `clickhouse/` branch tracks a specific upstream version branch or tag. It is cleaner in the sense that custom patches and upstream changes are better isolated from each other. - -Once the submodule has been updated, bump the submodule in ClickHouse to point to the new hash in the fork. diff --git a/docs/en/development/tests.md b/docs/en/development/tests.md index 269995a1a96..6cb36e2049b 100644 --- a/docs/en/development/tests.md +++ b/docs/en/development/tests.md @@ -14,7 +14,7 @@ Each functional test sends one or multiple queries to the running ClickHouse ser Tests are located in `queries` directory. There are two subdirectories: `stateless` and `stateful`. Stateless tests run queries without any preloaded test data - they often create small synthetic datasets on the fly, within the test itself. Stateful tests require preloaded test data from ClickHouse and it is available to general public. -Each test can be one of two types: `.sql` and `.sh`. `.sql` test is the simple SQL script that is piped to `clickhouse-client --multiquery`. `.sh` test is a script that is run by itself. SQL tests are generally preferable to `.sh` tests. You should use `.sh` tests only when you have to test some feature that cannot be exercised from pure SQL, such as piping some input data into `clickhouse-client` or testing `clickhouse-local`. +Each test can be one of two types: `.sql` and `.sh`. `.sql` test is the simple SQL script that is piped to `clickhouse-client`. `.sh` test is a script that is run by itself. SQL tests are generally preferable to `.sh` tests. You should use `.sh` tests only when you have to test some feature that cannot be exercised from pure SQL, such as piping some input data into `clickhouse-client` or testing `clickhouse-local`. :::note A common mistake when testing data types `DateTime` and `DateTime64` is assuming that the server uses a specific time zone (e.g. "UTC"). This is not the case, time zones in CI test runs @@ -38,7 +38,7 @@ For more options, see `tests/clickhouse-test --help`. You can simply run all tes ### Adding a New Test -To add new test, create a `.sql` or `.sh` file in `queries/0_stateless` directory, check it manually and then generate `.reference` file in the following way: `clickhouse-client --multiquery < 00000_test.sql > 00000_test.reference` or `./00000_test.sh > ./00000_test.reference`. +To add new test, create a `.sql` or `.sh` file in `queries/0_stateless` directory, check it manually and then generate `.reference` file in the following way: `clickhouse-client < 00000_test.sql > 00000_test.reference` or `./00000_test.sh > ./00000_test.reference`. Tests should use (create, drop, etc) only tables in `test` database that is assumed to be created beforehand; also tests can use temporary tables. diff --git a/docs/en/engines/table-engines/index.md b/docs/en/engines/table-engines/index.md index 5e81eacc937..20c7c511aa9 100644 --- a/docs/en/engines/table-engines/index.md +++ b/docs/en/engines/table-engines/index.md @@ -61,6 +61,7 @@ Engines in the family: - [RabbitMQ](../../engines/table-engines/integrations/rabbitmq.md) - [PostgreSQL](../../engines/table-engines/integrations/postgresql.md) - [S3Queue](../../engines/table-engines/integrations/s3queue.md) +- [TimeSeries](../../engines/table-engines/integrations/time-series.md) ### Special Engines {#special-engines} diff --git a/docs/en/engines/table-engines/integrations/kafka.md b/docs/en/engines/table-engines/integrations/kafka.md index 7bdc856c9fd..de6492e8ea7 100644 --- a/docs/en/engines/table-engines/integrations/kafka.md +++ b/docs/en/engines/table-engines/integrations/kafka.md @@ -251,6 +251,44 @@ The number of rows in one Kafka message depends on whether the format is row-bas - For row-based formats the number of rows in one Kafka message can be controlled by setting `kafka_max_rows_per_message`. - For block-based formats we cannot divide block into smaller parts, but the number of rows in one block can be controlled by general setting [max_block_size](../../../operations/settings/settings.md#setting-max_block_size). +## Experimental engine to store committed offsets in ClickHouse Keeper + +If `allow_experimental_kafka_offsets_storage_in_keeper` is enabled, then two more settings can be specified to the Kafka table engine: + - `kafka_keeper_path` specifies the path to the table in ClickHouse Keeper + - `kafka_replica_name` specifies the replica name in ClickHouse Keeper + +Either both of the settings must be specified or neither of them. When both of them are specified, then a new, experimental Kafka engine will be used. The new engine doesn't depend on storing the committed offsets in Kafka, but stores them in ClickHouse Keeper. It still tries to commit the offsets to Kafka, but it only depends on those offsets when the table is created. In any other circumstances (table is restarted, or recovered after some error) the offsets stored in ClickHouse Keeper will be used as an offset to continue consuming messages from. Apart from the committed offset, it also stores how many messages were consumed in the last batch, so if the insert fails, the same amount of messages will be consumed, thus enabling deduplication if necessary. + +Example: + +``` sql +CREATE TABLE experimental_kafka (key UInt64, value UInt64) +ENGINE = Kafka('localhost:19092', 'my-topic', 'my-consumer', 'JSONEachRow') +SETTINGS + kafka_keeper_path = '/clickhouse/{database}/experimental_kafka', + kafka_replica_name = 'r1' +SETTINGS allow_experimental_kafka_offsets_storage_in_keeper=1; +``` + +Or to utilize the `uuid` and `replica` macros similarly to ReplicatedMergeTree: + +``` sql +CREATE TABLE experimental_kafka (key UInt64, value UInt64) +ENGINE = Kafka('localhost:19092', 'my-topic', 'my-consumer', 'JSONEachRow') +SETTINGS + kafka_keeper_path = '/clickhouse/{database}/{uuid}', + kafka_replica_name = '{replica}' +SETTINGS allow_experimental_kafka_offsets_storage_in_keeper=1; +``` + +### Known limitations + +As the new engine is experimental, it is not production ready yet. There are few known limitations of the implementation: + - The biggest limitation is the engine doesn't support direct reading. Reading from the engine using materialized views and writing to the engine work, but direct reading doesn't. As a result, all direct `SELECT` queries will fail. + - Rapidly dropping and recreating the table or specifying the same ClickHouse Keeper path to different engines might cause issues. As best practice you can use the `{uuid}` in `kafka_keeper_path` to avoid clashing paths. + - To make repeatable reads, messages cannot be consumed from multiple partitions on a single thread. On the other hand, the Kafka consumers have to be polled regularly to keep them alive. As a result of these two objectives, we decided to only allow creating multiple consumers if `kafka_thread_per_consumer` is enabled, otherwise it is too complicated to avoid issues regarding polling consumers regularly. + - Consumers created by the new storage engine do not show up in [`system.kafka_consumers`](../../../operations/system-tables/kafka_consumers.md) table. + **See Also** - [Virtual columns](../../../engines/table-engines/index.md#table_engines-virtual_columns) diff --git a/docs/en/engines/table-engines/integrations/s3.md b/docs/en/engines/table-engines/integrations/s3.md index 93f4a187656..d664c37bd0f 100644 --- a/docs/en/engines/table-engines/integrations/s3.md +++ b/docs/en/engines/table-engines/integrations/s3.md @@ -146,6 +146,7 @@ Code: 48. DB::Exception: Received from localhost:9000. DB::Exception: Reading fr - `_file` ā€” Name of the file. Type: `LowCardinalty(String)`. - `_size` ā€” Size of the file in bytes. Type: `Nullable(UInt64)`. If the size is unknown, the value is `NULL`. - `_time` ā€” Last modified time of the file. Type: `Nullable(DateTime)`. If the time is unknown, the value is `NULL`. +- `_etag` ā€” ETag of the file. Type: `LowCardinalty(String)`. If the etag is unknown, the value is `NULL`. For more information about virtual columns see [here](../../../engines/table-engines/index.md#table_engines-virtual_columns). diff --git a/docs/en/engines/table-engines/integrations/time-series.md b/docs/en/engines/table-engines/integrations/time-series.md new file mode 100644 index 00000000000..4830fd61d27 --- /dev/null +++ b/docs/en/engines/table-engines/integrations/time-series.md @@ -0,0 +1,295 @@ +--- +slug: /en/engines/table-engines/special/time_series +sidebar_position: 60 +sidebar_label: TimeSeries +--- + +# TimeSeries Engine [Experimental] + +A table engine storing time series, i.e. a set of values associated with timestamps and tags (or labels): + +``` +metric_name1[tag1=value1, tag2=value2, ...] = {timestamp1: value1, timestamp2: value2, ...} +metric_name2[...] = ... +``` + +:::info +This is an experimental feature that may change in backwards-incompatible ways in the future releases. +Enable usage of the TimeSeries table engine +with [allow_experimental_time_series_table](../../../operations/settings/settings.md#allow-experimental-time-series-table) setting. +Input the command `set allow_experimental_time_series_table = 1`. +::: + +## Syntax {#syntax} + +``` sql +CREATE TABLE name [(columns)] ENGINE=TimeSeries +[SETTINGS var1=value1, ...] +[DATA db.data_table_name | DATA ENGINE data_table_engine(arguments)] +[TAGS db.tags_table_name | TAGS ENGINE tags_table_engine(arguments)] +[METRICS db.metrics_table_name | METRICS ENGINE metrics_table_engine(arguments)] +``` + +## Usage {#usage} + +It's easier to start with everything set by default (it's allowed to create a `TimeSeries` table without specifying a list of columns): + +``` sql +CREATE TABLE my_table ENGINE=TimeSeries +``` + +Then this table can be used with the following protocols (a port must be assigned in the server configuration): +- [prometheus remote-write](../../../interfaces/prometheus.md#remote-write) +- [prometheus remote-read](../../../interfaces/prometheus.md#remote-read) + +## Target tables {#target-tables} + +A `TimeSeries` table doesn't have its own data, everything is stored in its target tables. +This is similar to how a [materialized view](../../../sql-reference/statements/create/view#materialized-view) works, +with the difference that a materialized view has one target table +whereas a `TimeSeries` table has three target tables named [data]{#data-table}, [tags]{#tags-table], and [metrics]{#metrics-table}. + +The target tables can be either specified explicitly in the `CREATE TABLE` query +or the `TimeSeries` table engine can generate inner target tables automatically. + +The target tables are the following: +1. The _data_ table {#data-table} contains time series associated with some identifier. +The _data_ table must have columns: + +| Name | Mandatory? | Default type | Possible types | Description | +|---|---|---|---|---| +| `id` | [x] | `UUID` | any | Identifies a combination of a metric names and tags | +| `timestamp` | [x] | `DateTime64(3)` | `DateTime64(X)` | A time point | +| `value` | [x] | `Float64` | `Float32` or `Float64` | A value associated with the `timestamp` | + +2. The _tags_ table {#tags-table} contains identifiers calculated for each combination of a metric name and tags. +The _tags_ table must have columns: + +| Name | Mandatory? | Default type | Possible types | Description | +|---|---|---|---|---| +| `id` | [x] | `UUID` | any (must match the type of `id` in the [data]{#data-table} table) | An `id` identifies a combination of a metric name and tags. The DEFAULT expression specifies how to calculate such an identifier | +| `metric_name` | [x] | `LowCardinality(String)` | `String` or `LowCardinality(String)` | The name of a metric | +| `` | [ ] | `String` | `String` or `LowCardinality(String)` or `LowCardinality(Nullable(String))` | The value of a specific tag, the tag's name and the name of a corresponding column are specified in the [tags_to_columns](#settings) setting | +| `tags` | [x] | `Map(LowCardinality(String), String)` | `Map(String, String)` or `Map(LowCardinality(String), String)` or `Map(LowCardinality(String), LowCardinality(String))` | Map of tags excluding the tag `__name__` containing the name of a metric and excluding tags with names enumerated in the [tags_to_columns](#settings) setting | +| `all_tags` | [ ] | `Map(String, String)` | `Map(String, String)` or `Map(LowCardinality(String), String)` or `Map(LowCardinality(String), LowCardinality(String))` | Ephemeral column, each row is a map of all the tags excluding only the tag `__name__` containing the name of a metric. The only purpose of that column is to be used while calculating `id` | +| `min_time` | [ ] | `Nullable(DateTime64(3))` | `DateTime64(X)` or `Nullable(DateTime64(X))` | Minimum timestamp of time series with that `id`. The column is created if [store_min_time_and_max_time](#settings) is `true` | +| `max_time` | [ ] | `Nullable(DateTime64(3))` | `DateTime64(X)` or `Nullable(DateTime64(X))` | Maximum timestamp of time series with that `id`. The column is created if [store_min_time_and_max_time](#settings) is `true` | + +3. The _metrics_ table {#metrics-table} contains some information about metrics been collected, the types of those metrics and their descriptions. +The _metrics_ table must have columns: + +| Name | Mandatory? | Default type | Possible types | Description | +|---|---|---|---|---| +| `metric_family_name` | [x] | `String` | `String` or `LowCardinality(String)` | The name of a metric family | +| `type` | [x] | `String` | `String` or `LowCardinality(String)` | The type of a metric family, one of "counter", "gauge", "summary", "stateset", "histogram", "gaugehistogram" | +| `unit` | [x] | `String` | `String` or `LowCardinality(String)` | The unit used in a metric | +| `help` | [x] | `String` | `String` or `LowCardinality(String)` | The description of a metric | + +Any row inserted into a `TimeSeries` table will be in fact stored in those three target tables. +A `TimeSeries` table contains all those columns from the [data]{#data-table}, [tags]{#tags-table}, [metrics]{#metrics-table} tables. + +## Creation {#creation} + +There are multiple ways to create a table with the `TimeSeries` table engine. +The simplest statement + +``` sql +CREATE TABLE my_table ENGINE=TimeSeries +``` + +will actually create the following table (you can see that by executing `SHOW CREATE TABLE my_table`): + +``` sql +CREATE TABLE my_table +( + `id` UUID DEFAULT reinterpretAsUUID(sipHash128(metric_name, all_tags)), + `timestamp` DateTime64(3), + `value` Float64, + `metric_name` LowCardinality(String), + `tags` Map(LowCardinality(String), String), + `all_tags` Map(String, String), + `min_time` Nullable(DateTime64(3)), + `max_time` Nullable(DateTime64(3)), + `metric_family_name` String, + `type` String, + `unit` String, + `help` String +) +ENGINE = TimeSeries +DATA ENGINE = MergeTree ORDER BY (id, timestamp) +DATA INNER UUID '01234567-89ab-cdef-0123-456789abcdef' +TAGS ENGINE = AggregatingMergeTree PRIMARY KEY metric_name ORDER BY (metric_name, id) +TAGS INNER UUID '01234567-89ab-cdef-0123-456789abcdef' +METRICS ENGINE = ReplacingMergeTree ORDER BY metric_family_name +METRICS INNER UUID '01234567-89ab-cdef-0123-456789abcdef' +``` + +So the columns were generated automatically and also there are three inner UUIDs in this statement - +one per each inner target table that was created. +(Inner UUIDs are not shown normally until setting +[show_table_uuid_in_table_create_query_if_not_nil](../../../operations/settings/settings#show_table_uuid_in_table_create_query_if_not_nil) +is set.) + +Inner target tables have names like `.inner_id.data.xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx`, +`.inner_id.tags.xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx`, `.inner_id.metrics.xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx` +and each target table has columns which is a subset of the columns of the main `TimeSeries` table: + +``` sql +CREATE TABLE default.`.inner_id.data.xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx` +( + `id` UUID, + `timestamp` DateTime64(3), + `value` Float64 +) +ENGINE = MergeTree +ORDER BY (id, timestamp) +``` + +``` sql +CREATE TABLE default.`.inner_id.tags.xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx` +( + `id` UUID DEFAULT reinterpretAsUUID(sipHash128(metric_name, all_tags)), + `metric_name` LowCardinality(String), + `tags` Map(LowCardinality(String), String), + `all_tags` Map(String, String) EPHEMERAL, + `min_time` SimpleAggregateFunction(min, Nullable(DateTime64(3))), + `max_time` SimpleAggregateFunction(max, Nullable(DateTime64(3))) +) +ENGINE = AggregatingMergeTree +PRIMARY KEY metric_name +ORDER BY (metric_name, id) +``` + +``` sql +CREATE TABLE default.`.inner_id.metrics.xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx` +( + `metric_family_name` String, + `type` String, + `unit` String, + `help` String +) +ENGINE = ReplacingMergeTree +ORDER BY metric_family_name +``` + +## Adjusting types of columns {#adjusting-column-types} + +You can adjust the types of almost any column of the inner target tables by specifying them explicitly +while defining the main table. For example, + +``` sql +CREATE TABLE my_table +( + timestamp DateTime64(6) +) ENGINE=TimeSeries +``` + +will make the inner [data]{#data-table} table store timestamp in microseconds instead of milliseconds: + +``` sql +CREATE TABLE default.`.inner_id.data.xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx` +( + `id` UUID, + `timestamp` DateTime64(6), + `value` Float64 +) +ENGINE = MergeTree +ORDER BY (id, timestamp) +``` + +## The `id` column {#id-column} + +The `id` column contains identifiers, every identifier is calculated for a combination of a metric name and tags. +The DEFAULT expression for the `id` column is an expression which will be used to calculate such identifiers. +Both the type of the `id` column and that expression can be adjusted by specifying them explicitly: + +``` sql +CREATE TABLE my_table +( + id UInt64 DEFAULT sipHash64(metric_name, all_tags) +) ENGINE=TimeSeries +``` + +## The `tags` and `all_tags` columns {#tags-and-all-tags} + +There are two columns containing maps of tags - `tags` and `all_tags`. In this example they mean the same, however they can be different +if setting `tags_to_columns` is used. This setting allows to specify that a specific tag should be stored in a separate column instead of storing +in a map inside the `tags` column: + +``` sql +CREATE TABLE my_table ENGINE=TimeSeries SETTINGS = {'instance': 'instance', 'job': 'job'} +``` + +This statement will add columns +``` + `instance` String, + `job` String +``` +to the definition of both `my_table` and its inner [tags]{#tags-table} target table. In this case the `tags` column will not contain tags `instance` and `job`, +but the `all_tags` column will contain them. The `all_tags` column is ephemeral and its only purpose to be used in the DEFAULT expression +for the `id` column. + +The types of columns can be adjusted by specifying them explicitly: + +``` sql +CREATE TABLE my_table (instance LowCardinality(String), job LowCardinality(Nullable(String))) +ENGINE=TimeSeries SETTINGS = {'instance': 'instance', 'job': 'job'} +``` + +## Table engines of inner target tables {#inner-table-engines} + +By default inner target tables use the following table engines: +- the [data]{#data-table} table uses [MergeTree](../mergetree-family/mergetree); +- the [tags]{#tags-table} table uses [AggregatingMergeTree](../mergetree-family/aggregatingmergetree) because the same data is often inserted multiple times to this table so we need a way +to remove duplicates, and also because it's required to do aggregation for columns `min_time` and `max_time`; +- the [metrics]{#metrics-table} table uses [ReplacingMergeTree](../mergetree-family/replacingmergetree) because the same data is often inserted multiple times to this table so we need a way +to remove duplicates. + +Other table engines also can be used for inner target tables if it's specified so: + +``` sql +CREATE TABLE my_table ENGINE=TimeSeries +DATA ENGINE=ReplicatedMergeTree +TAGS ENGINE=ReplicatedAggregatingMergeTree +METRICS ENGINE=ReplicatedReplacingMergeTree +``` + +## External target tables {#external-target-tables} + +It's possible to make a `TimeSeries` table use a manually created table: + +``` sql +CREATE TABLE data_for_my_table +( + `id` UUID, + `timestamp` DateTime64(3), + `value` Float64 +) +ENGINE = MergeTree +ORDER BY (id, timestamp); + +CREATE TABLE tags_for_my_table ... + +CREATE TABLE metrics_for_my_table ... + +CREATE TABLE my_table ENGINE=TimeSeries DATA data_for_my_table TAGS tags_for_my_table METRICS metrics_for_my_table; +``` + +## Settings {#settings} + +Here is a list of settings which can be specified while defining a `TimeSeries` table: + +| Name | Type | Default | Description | +|---|---|---|---| +| `tags_to_columns` | Map | {} | Map specifying which tags should be put to separate columns in the [tags]{#tags-table} table. Syntax: `{'tag1': 'column1', 'tag2' : column2, ...}` | +| `use_all_tags_column_to_generate_id` | Bool | true | When generating an expression to calculate an identifier of a time series, this flag enables using the `all_tags` column in that calculation | +| `store_min_time_and_max_time` | Bool | true | If set to true then the table will store `min_time` and `max_time` for each time series | +| `aggregate_min_time_and_max_time` | Bool | true | When creating an inner target `tags` table, this flag enables using `SimpleAggregateFunction(min, Nullable(DateTime64(3)))` instead of just `Nullable(DateTime64(3))` as the type of the `min_time` column, and the same for the `max_time` column | +| `filter_by_min_time_and_max_time` | Bool | true | If set to true then the table will use the `min_time` and `max_time` columns for filtering time series | + +# Functions {#functions} + +Here is a list of functions supporting a `TimeSeries` table as an argument: +- [timeSeriesData](../../../sql-reference/table-functions/timeSeriesData.md) +- [timeSeriesTags](../../../sql-reference/table-functions/timeSeriesTags.md) +- [timeSeriesMetrics](../../../sql-reference/table-functions/timeSeriesMetrics.md) diff --git a/docs/en/engines/table-engines/mergetree-family/annindexes.md b/docs/en/engines/table-engines/mergetree-family/annindexes.md index 5a81313f62e..097b0f5850a 100644 --- a/docs/en/engines/table-engines/mergetree-family/annindexes.md +++ b/docs/en/engines/table-engines/mergetree-family/annindexes.md @@ -17,7 +17,7 @@ In terms of SQL, the nearest neighborhood problem can be expressed as follows: ``` sql SELECT * -FROM table_with_ann_index +FROM table ORDER BY Distance(vectors, Point) LIMIT N ``` @@ -27,75 +27,111 @@ Function `Distance` computes the distance between two vectors. Often, the Euclid distance functions](/docs/en/sql-reference/functions/distance-functions.md) are also possible. `Point` is the reference point, e.g. `(0.17, 0.33, ...)`, and `N` limits the number of search results. -An alternative formulation of the nearest neighborhood search problem looks as follows: +This query returns the top-`N` closest points to the reference point. Parameter `N` limits the number of returned values which is useful for +situations where `MaxDistance` is difficult to determine in advance. -``` sql -SELECT * -FROM table_with_ann_index -WHERE Distance(vectors, Point) < MaxDistance -LIMIT N -``` - -While the first query returns the top-`N` closest points to the reference point, the second query returns all points closer to the reference -point than a maximally allowed radius `MaxDistance`. Parameter `N` limits the number of returned values which is useful for situations where -`MaxDistance` is difficult to determine in advance. - -With brute force search, both queries are expensive (linear in the number of points) because the distance between all points in `vectors` and +With brute force search, the query is expensive (linear in the number of points) because the distance between all points in `vectors` and `Point` must be computed. To speed this process up, Approximate Nearest Neighbor Search Indexes (ANN indexes) store a compact representation of the search space (using clustering, search trees, etc.) which allows to compute an approximate answer much quicker (in sub-linear time). -# Creating and Using ANN Indexes {#creating_using_ann_indexes} +# Creating and Using Vector Similarity Indexes -Syntax to create an ANN index over an [Array(Float32)](../../../sql-reference/data-types/array.md) column: +Syntax to create a vector similarity index over an [Array(Float32)](../../../sql-reference/data-types/array.md) column: ```sql -CREATE TABLE table_with_ann_index +CREATE TABLE table ( - `id` Int64, - `vectors` Array(Float32), - INDEX [ann_index_name vectors TYPE [ann_index_type]([ann_index_parameters]) [GRANULARITY [N]] + id Int64, + vectors Array(Float32), + INDEX index_name vectors TYPE vector_similarity(method, distance_function[, quantization, connectivity, expansion_add, expansion_search]) [GRANULARITY N] ) ENGINE = MergeTree ORDER BY id; ``` +Parameters: +- `method`: Supports currently only `hnsw`. +- `distance_function`: either `L2Distance` (the [Euclidean distance](https://en.wikipedia.org/wiki/Euclidean_distance) - the length of a + line between two points in Euclidean space), or `cosineDistance` (the [cosine + distance](https://en.wikipedia.org/wiki/Cosine_similarity#Cosine_distance)- the angle between two non-zero vectors). +- `quantization`: either `f32`, `f16`, or `i8` for storing the vector with reduced precision (optional, default: `f32`) +- `m`: the number of neighbors per graph node (optional, default: 16) +- `ef_construction`: (optional, default: 128) +- `ef_search`: (optional, default: 64) + +Value 0 for parameters `m`, `ef_construction`, and `ef_search` refers to the default value. + +Example: + +```sql +CREATE TABLE table +( + id Int64, + vectors Array(Float32), + INDEX idx vectors TYPE vector_similarity('hnsw', 'L2Distance') -- Alternative syntax: TYPE vector_similarity(hnsw, L2Distance) +) +ENGINE = MergeTree +ORDER BY id; +``` + +Vector similarity indexes are based on the [USearch library](https://github.com/unum-cloud/usearch), which implements the [HNSW +algorithm](https://arxiv.org/abs/1603.09320), i.e., a hierarchical graph where each point represents a vector and the edges represent +similarity. Such hierarchical structures can be very efficient on large collections. They may often fetch 0.05% or less data from the +overall dataset, while still providing 99% recall. This is especially useful when working with high-dimensional vectors, that are expensive +to load and compare. The library also has several hardware-specific SIMD optimizations to accelerate further distance computations on modern +Arm (NEON and SVE) and x86 (AVX2 and AVX-512) CPUs and OS-specific optimizations to allow efficient navigation around immutable persistent +files, without loading them into RAM. + +USearch indexes are currently experimental, to use them you first need to `SET allow_experimental_vector_similarity_index = 1`. + +Vector similarity indexes currently support two distance functions: +- `L2Distance`, also called Euclidean distance, is the length of a line segment between two points in Euclidean space + ([Wikipedia](https://en.wikipedia.org/wiki/Euclidean_distance)). +- `cosineDistance`, also called cosine similarity, is the cosine of the angle between two (non-zero) vectors + ([Wikipedia](https://en.wikipedia.org/wiki/Cosine_similarity)). + +Vector similarity indexes allows storing the vectors in reduced precision formats. Supported scalar kinds are `f64`, `f32`, `f16` or `i8`. +If no scalar kind was specified during index creation, `f16` is used as default. + +For normalized data, `L2Distance` is usually a better choice, otherwise `cosineDistance` is recommended to compensate for scale. If no +distance function was specified during index creation, `L2Distance` is used as default. + +:::note +All arrays must have same length. To avoid errors, you can use a +[CONSTRAINT](/docs/en/sql-reference/statements/create/table.md#constraints), for example, `CONSTRAINT constraint_name_1 CHECK +length(vectors) = 256`. Also, empty `Arrays` and unspecified `Array` values in INSERT statements (i.e. default values) are not supported. +::: + +:::note +The vector similarity index currently does not work with per-table, non-default `index_granularity` settings (see +[here](https://github.com/ClickHouse/ClickHouse/pull/51325#issuecomment-1605920475)). If necessary, the value must be changed in config.xml. +::: + ANN indexes are built during column insertion and merge. As a result, `INSERT` and `OPTIMIZE` statements will be slower than for ordinary tables. ANNIndexes are ideally used only with immutable or rarely changed data, respectively when are far more read requests than write requests. -ANN indexes support two types of queries: - -- ORDER BY queries: +ANN indexes support these queries: ``` sql SELECT * - FROM table_with_ann_index + FROM table [WHERE ...] ORDER BY Distance(vectors, Point) LIMIT N ``` -- WHERE queries: - - ``` sql - SELECT * - FROM table_with_ann_index - WHERE Distance(vectors, Point) < MaxDistance - LIMIT N - ``` - :::tip To avoid writing out large vectors, you can use [query parameters](/docs/en/interfaces/cli.md#queries-with-parameters-cli-queries-with-parameters), e.g. ```bash -clickhouse-client --param_vec='hello' --query="SELECT * FROM table_with_ann_index WHERE L2Distance(vectors, {vec: Array(Float32)}) < 1.0" +clickhouse-client --param_vec='hello' --query="SELECT * FROM table WHERE L2Distance(vectors, {vec: Array(Float32)}) < 1.0" ``` ::: -**Restrictions**: Queries that contain both a `WHERE Distance(vectors, Point) < MaxDistance` and an `ORDER BY Distance(vectors, Point)` -clause cannot use ANN indexes. Also, the approximate algorithms used to determine the nearest neighbors require a limit, hence queries -without `LIMIT` clause cannot utilize ANN indexes. Also, ANN indexes are only used if the query has a `LIMIT` value smaller than setting +**Restrictions**: Approximate algorithms used to determine the nearest neighbors require a limit, hence queries without `LIMIT` clause +cannot utilize ANN indexes. Also, ANN indexes are only used if the query has a `LIMIT` value smaller than setting `max_limit_for_ann_queries` (default: 1 million rows). This is a safeguard to prevent large memory allocations by external libraries for approximate neighbor search. @@ -122,128 +158,3 @@ brute-force distance calculation over all rows of the granules. With a small `GR equally good, only the processing performance differs. It is generally recommended to use a large `GRANULARITY` for ANN indexes and fall back to a smaller `GRANULARITY` values only in case of problems like excessive memory consumption of the ANN structures. If no `GRANULARITY` was specified for ANN indexes, the default value is 100 million. - - -# Available ANN Indexes {#available_ann_indexes} - -- [Annoy](/docs/en/engines/table-engines/mergetree-family/annindexes.md#annoy-annoy) - -- [USearch](/docs/en/engines/table-engines/mergetree-family/annindexes.md#usearch-usearch) - -## Annoy {#annoy} - -Annoy indexes are currently experimental, to use them you first need to `SET allow_experimental_annoy_index = 1`. They are also currently -disabled on ARM due to memory safety problems with the algorithm. - -This type of ANN index is based on the [Annoy library](https://github.com/spotify/annoy) which recursively divides the space into random -linear surfaces (lines in 2D, planes in 3D etc.). - -
- -
- -Syntax to create an Annoy index over an [Array(Float32)](../../../sql-reference/data-types/array.md) column: - -```sql -CREATE TABLE table_with_annoy_index -( - id Int64, - vectors Array(Float32), - INDEX [ann_index_name] vectors TYPE annoy([Distance[, NumTrees]]) [GRANULARITY N] -) -ENGINE = MergeTree -ORDER BY id; -``` - -Annoy currently supports two distance functions: -- `L2Distance`, also called Euclidean distance, is the length of a line segment between two points in Euclidean space - ([Wikipedia](https://en.wikipedia.org/wiki/Euclidean_distance)). -- `cosineDistance`, also called cosine similarity, is the cosine of the angle between two (non-zero) vectors - ([Wikipedia](https://en.wikipedia.org/wiki/Cosine_similarity)). - -For normalized data, `L2Distance` is usually a better choice, otherwise `cosineDistance` is recommended to compensate for scale. If no -distance function was specified during index creation, `L2Distance` is used as default. - -Parameter `NumTrees` is the number of trees which the algorithm creates (default if not specified: 100). Higher values of `NumTree` mean -more accurate search results but slower index creation / query times (approximately linearly) as well as larger index sizes. - -:::note -All arrays must have same length. To avoid errors, you can use a -[CONSTRAINT](/docs/en/sql-reference/statements/create/table.md#constraints), for example, `CONSTRAINT constraint_name_1 CHECK -length(vectors) = 256`. Also, empty `Arrays` and unspecified `Array` values in INSERT statements (i.e. default values) are not supported. -::: - -The creation of Annoy indexes (whenever a new part is build, e.g. at the end of a merge) is a relatively slow process. You can increase -setting `max_threads_for_annoy_index_creation` (default: 4) which controls how many threads are used to create an Annoy index. Please be -careful with this setting, it is possible that multiple indexes are created in parallel in which case there can be overparallelization. - -Setting `annoy_index_search_k_nodes` (default: `NumTrees * LIMIT`) determines how many tree nodes are inspected during SELECTs. Larger -values mean more accurate results at the cost of longer query runtime: - -```sql -SELECT * -FROM table_name -ORDER BY L2Distance(vectors, Point) -LIMIT N -SETTINGS annoy_index_search_k_nodes=100; -``` - -:::note -The Annoy index currently does not work with per-table, non-default `index_granularity` settings (see -[here](https://github.com/ClickHouse/ClickHouse/pull/51325#issuecomment-1605920475)). If necessary, the value must be changed in config.xml. -::: - -## USearch {#usearch} - -This type of ANN index is based on the [USearch library](https://github.com/unum-cloud/usearch), which implements the [HNSW -algorithm](https://arxiv.org/abs/1603.09320), i.e., builds a hierarchical graph where each point represents a vector and the edges represent -similarity. Such hierarchical structures can be very efficient on large collections. They may often fetch 0.05% or less data from the -overall dataset, while still providing 99% recall. This is especially useful when working with high-dimensional vectors, -that are expensive to load and compare. The library also has several hardware-specific SIMD optimizations to accelerate further -distance computations on modern Arm (NEON and SVE) and x86 (AVX2 and AVX-512) CPUs and OS-specific optimizations to allow efficient -navigation around immutable persistent files, without loading them into RAM. - -
- -
- -Syntax to create an USearch index over an [Array](../../../sql-reference/data-types/array.md) column: - -```sql -CREATE TABLE table_with_usearch_index -( - id Int64, - vectors Array(Float32), - INDEX [ann_index_name] vectors TYPE usearch([Distance[, ScalarKind]]) [GRANULARITY N] -) -ENGINE = MergeTree -ORDER BY id; -``` - -USearch currently supports two distance functions: -- `L2Distance`, also called Euclidean distance, is the length of a line segment between two points in Euclidean space - ([Wikipedia](https://en.wikipedia.org/wiki/Euclidean_distance)). -- `cosineDistance`, also called cosine similarity, is the cosine of the angle between two (non-zero) vectors - ([Wikipedia](https://en.wikipedia.org/wiki/Cosine_similarity)). - -USearch allows storing the vectors in reduced precision formats. Supported scalar kinds are `f64`, `f32`, `f16` or `i8`. If no scalar kind -was specified during index creation, `f16` is used as default. - -For normalized data, `L2Distance` is usually a better choice, otherwise `cosineDistance` is recommended to compensate for scale. If no -distance function was specified during index creation, `L2Distance` is used as default. diff --git a/docs/en/engines/table-engines/mergetree-family/mergetree.md b/docs/en/engines/table-engines/mergetree-family/mergetree.md index 7ffbd9a5bae..183b94f4641 100644 --- a/docs/en/engines/table-engines/mergetree-family/mergetree.md +++ b/docs/en/engines/table-engines/mergetree-family/mergetree.md @@ -1005,7 +1005,7 @@ They can be used for prewhere optimization only if we enable `set allow_statisti ## Column-level Settings {#column-level-settings} -Certain MergeTree settings can be override at column level: +Certain MergeTree settings can be overridden at column level: - `max_compress_block_size` ā€” Maximum size of blocks of uncompressed data before compressing for writing to a table. - `min_compress_block_size` ā€” Minimum size of blocks of uncompressed data required for compression when writing the next mark. diff --git a/docs/en/interfaces/http.md b/docs/en/interfaces/http.md index f5b6326fa96..03fdfa048c8 100644 --- a/docs/en/interfaces/http.md +++ b/docs/en/interfaces/http.md @@ -379,7 +379,7 @@ You can mitigate this problem by enabling `wait_end_of_query=1` ([Response Buffe However, this does not completely solve the problem because the result must still fit within the `http_response_buffer_size`, and other settings like `send_progress_in_http_headers` can interfere with the delay of the header. The only way to catch all errors is to analyze the HTTP body before parsing it using the required format. -### Queries with Parameters {#cli-queries-with-parameters} +## Queries with Parameters {#cli-queries-with-parameters} You can create a query with parameters and pass values for them from the corresponding HTTP request parameters. For more information, see [Queries with Parameters for CLI](../interfaces/cli.md#cli-queries-with-parameters). diff --git a/docs/en/interfaces/prometheus.md b/docs/en/interfaces/prometheus.md new file mode 100644 index 00000000000..8e7023cc51f --- /dev/null +++ b/docs/en/interfaces/prometheus.md @@ -0,0 +1,160 @@ +--- +slug: /en/interfaces/prometheus +sidebar_position: 19 +sidebar_label: Prometheus protocols +--- + +# Prometheus protocols + +## Exposing metrics {#expose} + +:::note +ClickHouse Cloud does not currently support connecting to Prometheus. To be notified when this feature is supported, please contact support@clickhouse.com. +::: + +ClickHouse can expose its own metrics for scraping from Prometheus: + +```xml + + 9363 + /metrics + true + true + true + true + + +Section `` can be used to make more extended handlers. +This section is similar to [](/en/interfaces/http) but works for prometheus protocols: + +```xml + + 9363 + + + /metrics + + expose_metrics + true + true + true + true + + + + +``` + +Settings: + +| Name | Default | Description | +|---|---|---|---| +| `port` | none | Port for serving the exposing metrics protocol. | +| `endpoint` | `/metrics` | HTTP endpoint for scraping metrics by prometheus server. Starts with `/`. Should not be used with the `` section. | +| `url` / `headers` / `method` | none | Filters used to find a matching handler for a request. Similar to the fields with the same names in the [](/en/interfaces/http) section. | +| `metrics` | true | Expose metrics from the [system.metrics](/en/operations/system-tables/metrics) table. | +| `asynchronous_metrics` | true | Expose current metrics values from the [system.asynchronous_metrics](/en/operations/system-tables/asynchronous_metrics) table. | +| `events` | true | Expose metrics from the [system.events](/en/operations/system-tables/events) table. | +| `errors` | true | Expose the number of errors by error codes occurred since the last server restart. This information could be obtained from the [system.errors](/en/operations/system-tables/errors) as well. | + +Check (replace `127.0.0.1` with the IP addr or hostname of your ClickHouse server): +```bash +curl 127.0.0.1:9363/metrics +``` + +## Remote-write protocol {#remote-write} + +ClickHouse supports the [remote-write](https://prometheus.io/docs/specs/remote_write_spec/) protocol. +Data are received by this protocol and written to a [TimeSeries](/en/engines/table-engines/special/time_series) table +(which should be created beforehand). + +```xml + + 9363 + + + /write + + remote_write + db_name + time_series_table
+ + + + +``` + +Settings: + +| Name | Default | Description | +|---|---|---|---| +| `port` | none | Port for serving the `remote-write` protocol. | +| `url` / `headers` / `method` | none | Filters used to find a matching handler for a request. Similar to the fields with the same names in the [](/en/interfaces/http) section. | +| `table` | none | The name of a [TimeSeries](/en/engines/table-engines/special/time_series) table to write data received by the `remote-write` protocol. This name can optionally contain the name of a database too. | +| `database` | none | The name of a database where the table specified in the `table` setting is located if it's not specified in the `table` setting. | + +## Remote-read protocol {#remote-read} + +ClickHouse supports the [remote-read](https://prometheus.io/docs/prometheus/latest/querying/remote_read_api/) protocol. +Data are read from a [TimeSeries](/en/engines/table-engines/special/time_series) table and sent via this protocol. + +```xml + + 9363 + + + /read + + remote_read + db_name + time_series_table
+ + + + +``` + +Settings: + +| Name | Default | Description | +|---|---|---|---| +| `port` | none | Port for serving the `remote-read` protocol. | +| `url` / `headers` / `method` | none | Filters used to find a matching handler for a request. Similar to the fields with the same names in the [](/en/interfaces/http) section. | +| `table` | none | The name of a [TimeSeries](/en/engines/table-engines/special/time_series) table to read data to send by the `remote-read` protocol. This name can optionally contain the name of a database too. | +| `database` | none | The name of a database where the table specified in the `table` setting is located if it's not specified in the `table` setting. | + +## Configuration for multiple protocols {#multiple-protocols} + +Multiple protocols can be specified together in one place: + +```xml + + 9363 + + + /metrics + + expose_metrics + true + true + true + true + + + + /write + + remote_write + db_name.time_series_table
+ + + + /read + + remote_read + db_name.time_series_table
+ + + + +``` diff --git a/docs/en/interfaces/schema-inference.md b/docs/en/interfaces/schema-inference.md index 05fae994cbe..4afba20d76c 100644 --- a/docs/en/interfaces/schema-inference.md +++ b/docs/en/interfaces/schema-inference.md @@ -359,13 +359,14 @@ DESC format(JSONEachRow, '{"int" : 42, "float" : 42.42, "string" : "Hello, World Dates, DateTimes: ```sql -DESC format(JSONEachRow, '{"date" : "2022-01-01", "datetime" : "2022-01-01 00:00:00"}') +DESC format(JSONEachRow, '{"date" : "2022-01-01", "datetime" : "2022-01-01 00:00:00", "datetime64" : "2022-01-01 00:00:00.000"}') ``` ```response -ā”Œā”€nameā”€ā”€ā”€ā”€ā”€ā”¬ā”€typeā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€default_typeā”€ā”¬ā”€default_expressionā”€ā”¬ā”€commentā”€ā”¬ā”€codec_expressionā”€ā”¬ā”€ttl_expressionā”€ā” -ā”‚ date ā”‚ Nullable(Date) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ -ā”‚ datetime ā”‚ Nullable(DateTime64(9)) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +ā”Œā”€nameā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€typeā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€default_typeā”€ā”¬ā”€default_expressionā”€ā”¬ā”€commentā”€ā”¬ā”€codec_expressionā”€ā”¬ā”€ttl_expressionā”€ā” +ā”‚ date ā”‚ Nullable(Date) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā”‚ datetime ā”‚ Nullable(DateTime) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā”‚ datetime64 ā”‚ Nullable(DateTime64(9)) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` Arrays: @@ -759,12 +760,13 @@ DESC format(CSV, 'Hello world!,World hello!') Dates, DateTimes: ```sql -DESC format(CSV, '"2020-01-01","2020-01-01 00:00:00"') +DESC format(CSV, '"2020-01-01","2020-01-01 00:00:00","2022-01-01 00:00:00.000"') ``` ```response ā”Œā”€nameā”€ā”¬ā”€typeā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€default_typeā”€ā”¬ā”€default_expressionā”€ā”¬ā”€commentā”€ā”¬ā”€codec_expressionā”€ā”¬ā”€ttl_expressionā”€ā” ā”‚ c1 ā”‚ Nullable(Date) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ -ā”‚ c2 ā”‚ Nullable(DateTime64(9)) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā”‚ c2 ā”‚ Nullable(DateTime) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā”‚ c3 ā”‚ Nullable(DateTime64(9)) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` @@ -956,12 +958,13 @@ DESC format(TSKV, 'int=42 float=42.42 bool=true string=Hello,World!\n') Dates, DateTimes: ```sql -DESC format(TSV, '2020-01-01 2020-01-01 00:00:00') +DESC format(TSV, '2020-01-01 2020-01-01 00:00:00 2022-01-01 00:00:00.000') ``` ```response ā”Œā”€nameā”€ā”¬ā”€typeā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€default_typeā”€ā”¬ā”€default_expressionā”€ā”¬ā”€commentā”€ā”¬ā”€codec_expressionā”€ā”¬ā”€ttl_expressionā”€ā” ā”‚ c1 ā”‚ Nullable(Date) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ -ā”‚ c2 ā”‚ Nullable(DateTime64(9)) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā”‚ c2 ā”‚ Nullable(DateTime) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā”‚ c3 ā”‚ Nullable(DateTime64(9)) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` @@ -1126,12 +1129,13 @@ DESC format(Values, $$(42, 42.42, true, 'Hello,World!')$$) Dates, DateTimes: ```sql -DESC format(Values, $$('2020-01-01', '2020-01-01 00:00:00')$$) -``` + DESC format(Values, $$('2020-01-01', '2020-01-01 00:00:00', '2022-01-01 00:00:00.000')$$) + ``` ```response ā”Œā”€nameā”€ā”¬ā”€typeā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€default_typeā”€ā”¬ā”€default_expressionā”€ā”¬ā”€commentā”€ā”¬ā”€codec_expressionā”€ā”¬ā”€ttl_expressionā”€ā” ā”‚ c1 ā”‚ Nullable(Date) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ -ā”‚ c2 ā”‚ Nullable(DateTime64(9)) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā”‚ c2 ā”‚ Nullable(DateTime) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā”‚ c3 ā”‚ Nullable(DateTime64(9)) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` @@ -1504,8 +1508,8 @@ DESC format(JSONEachRow, $$ #### input_format_try_infer_datetimes -If enabled, ClickHouse will try to infer type `DateTime64` from string fields in schema inference for text formats. -If all fields from a column in sample data were successfully parsed as datetimes, the result type will be `DateTime64(9)`, +If enabled, ClickHouse will try to infer type `DateTime` or `DateTime64` from string fields in schema inference for text formats. +If all fields from a column in sample data were successfully parsed as datetimes, the result type will be `DateTime` or `DateTime64(9)` (if any datetime had fractional part), if at least one field was not parsed as datetime, the result type will be `String`. Enabled by default. @@ -1513,39 +1517,66 @@ Enabled by default. **Examples** ```sql -SET input_format_try_infer_datetimes = 0 +SET input_format_try_infer_datetimes = 0; DESC format(JSONEachRow, $$ - {"datetime" : "2021-01-01 00:00:00.000"} - {"datetime" : "2022-01-01 00:00:00.000"} + {"datetime" : "2021-01-01 00:00:00", "datetime64" : "2021-01-01 00:00:00.000"} + {"datetime" : "2022-01-01 00:00:00", "datetime64" : "2022-01-01 00:00:00.000"} $$) ``` ```response -ā”Œā”€nameā”€ā”€ā”€ā”€ā”€ā”¬ā”€typeā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€default_typeā”€ā”¬ā”€default_expressionā”€ā”¬ā”€commentā”€ā”¬ā”€codec_expressionā”€ā”¬ā”€ttl_expressionā”€ā” -ā”‚ datetime ā”‚ Nullable(String) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +ā”Œā”€nameā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€typeā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€default_typeā”€ā”¬ā”€default_expressionā”€ā”¬ā”€commentā”€ā”¬ā”€codec_expressionā”€ā”¬ā”€ttl_expressionā”€ā” +ā”‚ datetime ā”‚ Nullable(String) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā”‚ datetime64 ā”‚ Nullable(String) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` ```sql -SET input_format_try_infer_datetimes = 1 +SET input_format_try_infer_datetimes = 1; DESC format(JSONEachRow, $$ - {"datetime" : "2021-01-01 00:00:00.000"} - {"datetime" : "2022-01-01 00:00:00.000"} + {"datetime" : "2021-01-01 00:00:00", "datetime64" : "2021-01-01 00:00:00.000"} + {"datetime" : "2022-01-01 00:00:00", "datetime64" : "2022-01-01 00:00:00.000"} $$) ``` ```response -ā”Œā”€nameā”€ā”€ā”€ā”€ā”€ā”¬ā”€typeā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€default_typeā”€ā”¬ā”€default_expressionā”€ā”¬ā”€commentā”€ā”¬ā”€codec_expressionā”€ā”¬ā”€ttl_expressionā”€ā” -ā”‚ datetime ā”‚ Nullable(DateTime64(9)) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +ā”Œā”€nameā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€typeā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€default_typeā”€ā”¬ā”€default_expressionā”€ā”¬ā”€commentā”€ā”¬ā”€codec_expressionā”€ā”¬ā”€ttl_expressionā”€ā” +ā”‚ datetime ā”‚ Nullable(DateTime) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā”‚ datetime64 ā”‚ Nullable(DateTime64(9)) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` ```sql DESC format(JSONEachRow, $$ - {"datetime" : "2021-01-01 00:00:00.000"} - {"datetime" : "unknown"} + {"datetime" : "2021-01-01 00:00:00", "datetime64" : "2021-01-01 00:00:00.000"} + {"datetime" : "unknown", "datetime64" : "unknown"} $$) ``` ```response -ā”Œā”€nameā”€ā”€ā”€ā”€ā”€ā”¬ā”€typeā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€default_typeā”€ā”¬ā”€default_expressionā”€ā”¬ā”€commentā”€ā”¬ā”€codec_expressionā”€ā”¬ā”€ttl_expressionā”€ā” -ā”‚ datetime ā”‚ Nullable(String) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +ā”Œā”€nameā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€typeā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€default_typeā”€ā”¬ā”€default_expressionā”€ā”¬ā”€commentā”€ā”¬ā”€codec_expressionā”€ā”¬ā”€ttl_expressionā”€ā” +ā”‚ datetime ā”‚ Nullable(String) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā”‚ datetime64 ā”‚ Nullable(String) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +``` + +#### input_format_try_infer_datetimes_only_datetime64 + +If enabled, ClickHouse will always infer `DateTime64(9)` when `input_format_try_infer_datetimes` is enabled even if datetime values don't contain fractional part. + +Disabled by default. + +**Examples** + +```sql +SET input_format_try_infer_datetimes = 1; +SET input_format_try_infer_datetimes_only_datetime64 = 1; +DESC format(JSONEachRow, $$ + {"datetime" : "2021-01-01 00:00:00", "datetime64" : "2021-01-01 00:00:00.000"} + {"datetime" : "2022-01-01 00:00:00", "datetime64" : "2022-01-01 00:00:00.000"} + $$) +``` + +```text +ā”Œā”€nameā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€typeā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€default_typeā”€ā”¬ā”€default_expressionā”€ā”¬ā”€commentā”€ā”¬ā”€codec_expressionā”€ā”¬ā”€ttl_expressionā”€ā” +ā”‚ datetime ā”‚ Nullable(DateTime64(9)) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā”‚ datetime64 ā”‚ Nullable(DateTime64(9)) ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` Note: Parsing datetimes during schema inference respect setting [date_time_input_format](/docs/en/operations/settings/settings-formats.md#date_time_input_format) diff --git a/docs/en/operations/analyzer.md b/docs/en/operations/analyzer.md index 298c6dacd06..c9b3c30d30d 100644 --- a/docs/en/operations/analyzer.md +++ b/docs/en/operations/analyzer.md @@ -123,7 +123,7 @@ To ensure consistent and expected results, especially when migrating old queries In the new version of the analyzer, the rules for determining the common supertype for columns specified in the `USING` clause have been standardized to produce more predictable outcomes, especially when dealing with type modifiers like `LowCardinality` and `Nullable`. - `LowCardinality(T)` and `T`: When a column of type `LowCardinality(T)` is joined with a column of type `T`, the resulting common supertype will be `T`, effectively discarding the `LowCardinality` modifier. - + - `Nullable(T)` and `T`: When a column of type `Nullable(T)` is joined with a column of type `T`, the resulting common supertype will be `Nullable(T)`, ensuring that the nullable property is preserved. **Example:** @@ -144,7 +144,7 @@ During projection names computation, aliases are not substituted. SELECT 1 + 1 AS x, x + 1 -SETTINGS allow_experimental_analyzer = 0 +SETTINGS enable_analyzer = 0 FORMAT PrettyCompact ā”Œā”€xā”€ā”¬ā”€plus(plus(1, 1), 1)ā”€ā” @@ -154,7 +154,7 @@ FORMAT PrettyCompact SELECT 1 + 1 AS x, x + 1 -SETTINGS allow_experimental_analyzer = 1 +SETTINGS enable_analyzer = 1 FORMAT PrettyCompact ā”Œā”€xā”€ā”¬ā”€plus(x, 1)ā”€ā” @@ -177,7 +177,7 @@ SELECT toTypeName(if(0, [2, 3, 4], 'String')) ### Heterogeneous clusters -The new analyzer significantly changed the communication protocol between servers in the cluster. Thus, it's impossible to run distributed queries on servers with different `allow_experimental_analyzer` setting values. +The new analyzer significantly changed the communication protocol between servers in the cluster. Thus, it's impossible to run distributed queries on servers with different `enable_analyzer` setting values. ### Mutations are interpreted by previous analyzer diff --git a/docs/en/operations/query-cache.md b/docs/en/operations/query-cache.md index 7a920671fc2..384938e28f6 100644 --- a/docs/en/operations/query-cache.md +++ b/docs/en/operations/query-cache.md @@ -143,6 +143,18 @@ value can be specified at session, profile or query level using setting [query_c Entries in the query cache are compressed by default. This reduces the overall memory consumption at the cost of slower writes into / reads from the query cache. To disable compression, use setting [query_cache_compress_entries](settings/settings.md#query-cache-compress-entries). +Sometimes it is useful to keep multiple results for the same query cached. This can be achieved using setting +[query_cache_tag](settings/settings.md#query-cache-tag) that acts as as a label (or namespace) for a query cache entries. The query cache +considers results of the same query with different tags different. + +Example for creating three different query cache entries for the same query: + +```sql +SELECT 1 SETTINGS use_query_cache = true; -- query_cache_tag is implicitly '' (empty string) +SELECT 1 SETTINGS use_query_cache = true, query_cache_tag = 'tag 1'; +SELECT 1 SETTINGS use_query_cache = true, query_cache_tag = 'tag 2'; +``` + ClickHouse reads table data in blocks of [max_block_size](settings/settings.md#setting-max_block_size) rows. Due to filtering, aggregation, etc., result blocks are typically much smaller than 'max_block_size' but there are also cases where they are much bigger. Setting [query_cache_squash_partial_results](settings/settings.md#query-cache-squash-partial-results) (enabled by default) controls if result blocks diff --git a/docs/en/operations/server-configuration-parameters/settings.md b/docs/en/operations/server-configuration-parameters/settings.md index a1e3c292b04..9fce83a0dc4 100644 --- a/docs/en/operations/server-configuration-parameters/settings.md +++ b/docs/en/operations/server-configuration-parameters/settings.md @@ -1400,6 +1400,16 @@ The number of seconds that ClickHouse waits for incoming requests before closing 10 ``` +## max_keep_alive_requests {#max-keep-alive-requests} + +Maximal number of requests through a single keep-alive connection until it will be closed by ClickHouse server. Default to 10000. + +**Example** + +``` xml +10 +``` + ## listen_host {#listen_host} Restriction on hosts that requests can come from. If you want the server to answer all of them, specify `::`. @@ -2112,48 +2122,6 @@ The trailing slash is mandatory. /var/lib/clickhouse/ ``` -## Prometheus {#prometheus} - -:::note -ClickHouse Cloud does not currently support connecting to Prometheus. To be notified when this feature is supported, please contact support@clickhouse.com. -::: - -Exposing metrics data for scraping from [Prometheus](https://prometheus.io). - -Settings: - -- `endpoint` ā€“ HTTP endpoint for scraping metrics by prometheus server. Start from ā€˜/ā€™. -- `port` ā€“ Port for `endpoint`. -- `metrics` ā€“ Expose metrics from the [system.metrics](../../operations/system-tables/metrics.md#system_tables-metrics) table. -- `events` ā€“ Expose metrics from the [system.events](../../operations/system-tables/events.md#system_tables-events) table. -- `asynchronous_metrics` ā€“ Expose current metrics values from the [system.asynchronous_metrics](../../operations/system-tables/asynchronous_metrics.md#system_tables-asynchronous_metrics) table. -- `errors` - Expose the number of errors by error codes occurred since the last server restart. This information could be obtained from the [system.errors](../../operations/system-tables/asynchronous_metrics.md#system_tables-errors) as well. - -**Example** - -``` xml - - 0.0.0.0 - 8123 - 9000 - - - /metrics - 9363 - true - true - true - true - - - -``` - -Check (replace `127.0.0.1` with the IP addr or hostname of your ClickHouse server): -```bash -curl 127.0.0.1:9363/metrics -``` - ## query_log {#query-log} Setting for logging queries received with the [log_queries=1](../../operations/settings/settings.md) setting. diff --git a/docs/en/operations/settings/merge-tree-settings.md b/docs/en/operations/settings/merge-tree-settings.md index 67fa45c20cd..a13aacc76e6 100644 --- a/docs/en/operations/settings/merge-tree-settings.md +++ b/docs/en/operations/settings/merge-tree-settings.md @@ -1041,3 +1041,27 @@ Compression rates of LZ4 or ZSTD improve on average by 20-40%. This setting works best for tables with no primary key or a low-cardinality primary key, i.e. a table with only few distinct primary key values. High-cardinality primary keys, e.g. involving timestamp columns of type `DateTime64`, are not expected to benefit from this setting. + +## lightweight_mutation_projection_mode + +By default, lightweight delete `DELETE` does not work for tables with projections. This is because rows in a projection may be affected by a `DELETE` operation. So the default value would be `throw`. +However, this option can change the behavior. With the value either `drop` or `rebuild`, deletes will work with projections. `drop` would delete the projection so it might be fast in the current query as projection gets deleted but slow in future queries as no projection attached. +`rebuild` would rebuild the projection which might affect the performance of the current query, but might speedup for future queries. A good thing is that these options would only work in the part level, +which means projections in the part that don't get touched would stay intact instead of triggering any action like drop or rebuild. + +Possible values: + +- throw, drop, rebuild + +Default value: throw + +## deduplicate_merge_projection_mode + +Whether to allow create projection for the table with non-classic MergeTree, that is not (Replicated, Shared) MergeTree. If allowed, what is the action when merge projections, either drop or rebuild. So classic MergeTree would ignore this setting. +It also controls `OPTIMIZE DEDUPLICATE` as well, but has effect on all MergeTree family members. Similar to the option `lightweight_mutation_projection_mode`, it is also part level. + +Possible values: + +- throw, drop, rebuild + +Default value: throw \ No newline at end of file diff --git a/docs/en/operations/settings/settings.md b/docs/en/operations/settings/settings.md index 143ce836beb..de601fe02dc 100644 --- a/docs/en/operations/settings/settings.md +++ b/docs/en/operations/settings/settings.md @@ -1800,6 +1800,17 @@ Possible values: Default value: `0`. +## query_cache_tag {#query-cache-tag} + +A string which acts as a label for [query cache](../query-cache.md) entries. +The same queries with different tags are considered different by the query cache. + +Possible values: + +- Any string + +Default value: `''` + ## query_cache_max_size_in_bytes {#query-cache-max-size-in-bytes} The maximum amount of memory (in bytes) the current user may allocate in the [query cache](../query-cache.md). 0 means unlimited. @@ -4051,7 +4062,7 @@ Rewrite aggregate functions with if expression as argument when logically equiva For example, `avg(if(cond, col, null))` can be rewritten to `avgOrNullIf(cond, col)`. It may improve performance. :::note -Supported only with experimental analyzer (`allow_experimental_analyzer = 1`). +Supported only with experimental analyzer (`enable_analyzer = 1`). ::: ## database_replicated_initial_query_timeout_sec {#database_replicated_initial_query_timeout_sec} @@ -4629,8 +4640,8 @@ Default Value: 5. ## memory_overcommit_ratio_denominator {#memory_overcommit_ratio_denominator} -It represents soft memory limit in case when hard limit is reached on user level. -This value is used to compute overcommit ratio for the query. +It represents the soft memory limit when the hard limit is reached on the global level. +This value is used to compute the overcommit ratio for the query. Zero means skip the query. Read more about [memory overcommit](memory-overcommit.md). @@ -4646,8 +4657,8 @@ Default value: `5000000`. ## memory_overcommit_ratio_denominator_for_user {#memory_overcommit_ratio_denominator_for_user} -It represents soft memory limit in case when hard limit is reached on global level. -This value is used to compute overcommit ratio for the query. +It represents the soft memory limit when the hard limit is reached on the user level. +This value is used to compute the overcommit ratio for the query. Zero means skip the query. Read more about [memory overcommit](memory-overcommit.md). @@ -5609,8 +5620,43 @@ Minimal size of block to compress in CROSS JOIN. Zero value means - disable this Default value: `1GiB`. +## restore_replace_external_engines_to_null + +For testing purposes. Replaces all external engines to Null to not initiate external connections. + +Default value: `False` + +## restore_replace_external_table_functions_to_null + +For testing purposes. Replaces all external table functions to Null to not initiate external connections. + +Default value: `False` + ## disable_insertion_and_mutation Disable all insert and mutations (alter table update / alter table delete / alter table drop partition). Set to true, can make this node focus on reading queries. Default value: `false`. + +## use_hive_partitioning + +When enabled, ClickHouse will detect Hive-style partitioning in path (`/name=value/`) in file-like table engines [File](../../engines/table-engines/special/file.md#hive-style-partitioning)/[S3](../../engines/table-engines/integrations/s3.md#hive-style-partitioning)/[URL](../../engines/table-engines/special/url.md#hive-style-partitioning)/[HDFS](../../engines/table-engines/integrations/hdfs.md#hive-style-partitioning)/[AzureBlobStorage](../../engines/table-engines/integrations/azureBlobStorage.md#hive-style-partitioning) and will allow to use partition columns as virtual columns in the query. These virtual columns will have the same names as in the partitioned path, but starting with `_`. + +Default value: `false`. + +## allow_experimental_time_series_table {#allow-experimental-time-series-table} + +Allows creation of tables with the [TimeSeries](../../engines/table-engines/integrations/time-series.md) table engine. + +Possible values: + +- 0 ā€” the [TimeSeries](../../engines/table-engines/integrations/time-series.md) table engine is disabled. +- 1 ā€” the [TimeSeries](../../engines/table-engines/integrations/time-series.md) table engine is enabled. + +Default value: `0`. + +## create_if_not_exists + +Enable `IF NOT EXISTS` for `CREATE` statement by default. If either this setting or `IF NOT EXISTS` is specified and a table with the provided name already exists, no exception will be thrown. + +Default value: `false`. diff --git a/docs/en/operations/system-tables/kafka_consumers.md b/docs/en/operations/system-tables/kafka_consumers.md index 7e28a251e26..d58c9f754fd 100644 --- a/docs/en/operations/system-tables/kafka_consumers.md +++ b/docs/en/operations/system-tables/kafka_consumers.md @@ -24,6 +24,7 @@ Columns: - `num_rebalance_revocations`, (UInt64) - number of times the consumer was revoked its partitions - `num_rebalance_assignments`, (UInt64) - number of times the consumer was assigned to Kafka cluster - `is_currently_used`, (UInt8) - consumer is in use +- `last_used`, (UInt64) - last time this consumer was in use, unix time in microseconds - `rdkafka_stat` (String) - library internal statistic. See https://github.com/ClickHouse/librdkafka/blob/master/STATISTICS.md . Set `statistics_interval_ms` to 0 disable, default is 3000 (once in three seconds). Example: diff --git a/docs/en/operations/system-tables/query_cache.md b/docs/en/operations/system-tables/query_cache.md index a9f86f5fc2b..9c48574a329 100644 --- a/docs/en/operations/system-tables/query_cache.md +++ b/docs/en/operations/system-tables/query_cache.md @@ -9,6 +9,7 @@ Columns: - `query` ([String](../../sql-reference/data-types/string.md)) ā€” Query string. - `result_size` ([UInt64](../../sql-reference/data-types/int-uint.md#uint-ranges)) ā€” Size of the query cache entry. +- `tag` ([LowCardinality(String)](../../sql-reference/data-types/lowcardinality.md)) ā€” Tag of the query cache entry. - `stale` ([UInt8](../../sql-reference/data-types/int-uint.md)) ā€” If the query cache entry is stale. - `shared` ([UInt8](../../sql-reference/data-types/int-uint.md)) ā€” If the query cache entry is shared between multiple users. - `compressed` ([UInt8](../../sql-reference/data-types/int-uint.md)) ā€” If the query cache entry is compressed. @@ -26,6 +27,7 @@ Row 1: ā”€ā”€ā”€ā”€ā”€ā”€ query: SELECT 1 SETTINGS use_query_cache = 1 result_size: 128 +tag: stale: 0 shared: 0 compressed: 1 diff --git a/docs/en/operations/system-tables/trace_log.md b/docs/en/operations/system-tables/trace_log.md index 5adc33de37f..a60de2a08d1 100644 --- a/docs/en/operations/system-tables/trace_log.md +++ b/docs/en/operations/system-tables/trace_log.md @@ -3,7 +3,7 @@ slug: /en/operations/system-tables/trace_log --- # trace_log -Contains stack traces collected by the sampling query profiler. +Contains stack traces collected by the [sampling query profiler](../../operations/optimizing-performance/sampling-query-profiler.md). ClickHouse creates this table when the [trace_log](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-trace_log) server configuration section is set. Also see settings: [query_profiler_real_time_period_ns](../../operations/settings/settings.md#query_profiler_real_time_period_ns), [query_profiler_cpu_time_period_ns](../../operations/settings/settings.md#query_profiler_cpu_time_period_ns), [memory_profiler_step](../../operations/settings/settings.md#memory_profiler_step), [memory_profiler_sample_probability](../../operations/settings/settings.md#memory_profiler_sample_probability), [trace_profile_events](../../operations/settings/settings.md#trace_profile_events). diff --git a/docs/en/operations/utilities/clickhouse-keeper-client.md b/docs/en/operations/utilities/clickhouse-keeper-client.md index 6407c66783b..a66ecbc1372 100644 --- a/docs/en/operations/utilities/clickhouse-keeper-client.md +++ b/docs/en/operations/utilities/clickhouse-keeper-client.md @@ -28,39 +28,39 @@ A client application to interact with clickhouse-keeper by its native protocol. Connected to ZooKeeper at [::1]:9181 with session_id 137 / :) ls keeper foo bar -/ :) cd keeper +/ :) cd 'keeper' /keeper :) ls api_version -/keeper :) cd api_version +/keeper :) cd 'api_version' /keeper/api_version :) ls -/keeper/api_version :) cd xyz +/keeper/api_version :) cd 'xyz' Path /keeper/api_version/xyz does not exist /keeper/api_version :) cd ../../ / :) ls keeper foo bar -/ :) get keeper/api_version +/ :) get 'keeper/api_version' 2 ``` ## Commands {#clickhouse-keeper-client-commands} -- `ls [path]` -- Lists the nodes for the given path (default: cwd) -- `cd [path]` -- Changes the working path (default `.`) -- `exists ` -- Returns `1` if node exists, `0` otherwise -- `set [version]` -- Updates the node's value. Only updates if version matches (default: -1) -- `create [mode]` -- Creates new node with the set value -- `touch ` -- Creates new node with an empty string as value. Doesn't throw an exception if the node already exists -- `get ` -- Returns the node's value -- `rm [version]` -- Removes the node only if version matches (default: -1) -- `rmr ` -- Recursively deletes path. Confirmation required +- `ls '[path]'` -- Lists the nodes for the given path (default: cwd) +- `cd '[path]'` -- Changes the working path (default `.`) +- `exists ''` -- Returns `1` if node exists, `0` otherwise +- `set '' [version]` -- Updates the node's value. Only updates if version matches (default: -1) +- `create '' [mode]` -- Creates new node with the set value +- `touch ''` -- Creates new node with an empty string as value. Doesn't throw an exception if the node already exists +- `get ''` -- Returns the node's value +- `rm '' [version]` -- Removes the node only if version matches (default: -1) +- `rmr ''` -- Recursively deletes path. Confirmation required - `flwc ` -- Executes four-letter-word command - `help` -- Prints this message -- `get_direct_children_number [path]` -- Get numbers of direct children nodes under a specific path -- `get_all_children_number [path]` -- Get all numbers of children nodes under a specific path -- `get_stat [path]` -- Returns the node's stat (default `.`) -- `find_super_nodes [path]` -- Finds nodes with number of children larger than some threshold for the given path (default `.`) +- `get_direct_children_number '[path]'` -- Get numbers of direct children nodes under a specific path +- `get_all_children_number '[path]'` -- Get all numbers of children nodes under a specific path +- `get_stat '[path]'` -- Returns the node's stat (default `.`) +- `find_super_nodes '[path]'` -- Finds nodes with number of children larger than some threshold for the given path (default `.`) - `delete_stale_backups` -- Deletes ClickHouse nodes used for backups that are now inactive - `find_big_family [path] [n]` -- Returns the top n nodes with the biggest family in the subtree (default path = `.` and n = 10) -- `sync ` -- Synchronizes node between processes and leader +- `sync ''` -- Synchronizes node between processes and leader - `reconfig "" [version]` -- Reconfigure Keeper cluster. See https://clickhouse.com/docs/en/guides/sre/keeper/clickhouse-keeper#reconfiguration diff --git a/docs/en/sql-reference/aggregate-functions/reference/groupconcat.md b/docs/en/sql-reference/aggregate-functions/reference/groupconcat.md index 072252de8c9..6a24aa244bf 100644 --- a/docs/en/sql-reference/aggregate-functions/reference/groupconcat.md +++ b/docs/en/sql-reference/aggregate-functions/reference/groupconcat.md @@ -10,7 +10,7 @@ Calculates a concatenated string from a group of strings, optionally separated b **Syntax** ``` sql -groupConcat(expression [, delimiter] [, limit]); +groupConcat[(delimiter [, limit])](expression); ``` **Arguments** @@ -20,7 +20,7 @@ groupConcat(expression [, delimiter] [, limit]); - `limit` ā€” A positive [integer](../../../sql-reference/data-types/int-uint.md) specifying the maximum number of elements to concatenate. If more elements are present, excess elements are ignored. This parameter is optional. :::note -If delimiter is specified without limit, it must be the first parameter following the expression. If both delimiter and limit are specified, delimiter must precede limit. +If delimiter is specified without limit, it must be the first parameter. If both delimiter and limit are specified, delimiter must precede limit. ::: **Returned value** @@ -61,7 +61,7 @@ This concatenates all names into one continuous string without any separator. Query: ``` sql -SELECT groupConcat(Name, ', ', 2) FROM Employees; +SELECT groupConcat(', ')(Name) FROM Employees; ``` Result: @@ -78,7 +78,7 @@ This output shows the names separated by a comma followed by a space. Query: ``` sql -SELECT groupConcat(Name, ', ', 2) FROM Employees; +SELECT groupConcat(', ', 2)(Name) FROM Employees; ``` Result: diff --git a/docs/en/sql-reference/data-types/dynamic.md b/docs/en/sql-reference/data-types/dynamic.md index 8be81471377..f9befd166fe 100644 --- a/docs/en/sql-reference/data-types/dynamic.md +++ b/docs/en/sql-reference/data-types/dynamic.md @@ -14,7 +14,7 @@ To declare a column of `Dynamic` type, use the following syntax: Dynamic(max_types=N) ``` -Where `N` is an optional parameter between `1` and `255` indicating how many different data types can be stored inside a column with type `Dynamic` across single block of data that is stored separately (for example across single data part for MergeTree table). If this limit is exceeded, all new types will be converted to type `String`. Default value of `max_types` is `32`. +Where `N` is an optional parameter between `0` and `254` indicating how many different data types can be stored as separate subcolumns inside a column with type `Dynamic` across single block of data that is stored separately (for example across single data part for MergeTree table). If this limit is exceeded, all values with new types will be stored together in a special shared data structure in binary form. Default value of `max_types` is `32`. :::note The Dynamic data type is an experimental feature. To use it, set `allow_experimental_dynamic_type = 1`. @@ -224,41 +224,43 @@ SELECT d::Dynamic(max_types=5) as d2, dynamicType(d2) FROM test; ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` -If `K < N`, then the values with the rarest types are converted to `String`: +If `K < N`, then the values with the rarest types will be inserted into a single special subcolumn, but still will be accessible: ```text CREATE TABLE test (d Dynamic(max_types=4)) ENGINE = Memory; INSERT INTO test VALUES (NULL), (42), (43), ('42.42'), (true), ([1, 2, 3]); -SELECT d, dynamicType(d), d::Dynamic(max_types=2) as d2, dynamicType(d2) FROM test; +SELECT d, dynamicType(d), d::Dynamic(max_types=2) as d2, dynamicType(d2), isDynamicElementInSharedData(d2) FROM test; ``` ```text -ā”Œā”€dā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€dynamicType(d)ā”€ā”¬ā”€d2ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€dynamicType(d2)ā”€ā” -ā”‚ į“ŗįµį“øį“ø ā”‚ None ā”‚ į“ŗįµį“øį“ø ā”‚ None ā”‚ -ā”‚ 42 ā”‚ Int64 ā”‚ 42 ā”‚ Int64 ā”‚ -ā”‚ 43 ā”‚ Int64 ā”‚ 43 ā”‚ Int64 ā”‚ -ā”‚ 42.42 ā”‚ String ā”‚ 42.42 ā”‚ String ā”‚ -ā”‚ true ā”‚ Bool ā”‚ true ā”‚ String ā”‚ -ā”‚ [1,2,3] ā”‚ Array(Int64) ā”‚ [1,2,3] ā”‚ String ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +ā”Œā”€dā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€dynamicType(d)ā”€ā”¬ā”€d2ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€dynamicType(d2)ā”€ā”¬ā”€isDynamicElementInSharedData(d2)ā”€ā” +ā”‚ į“ŗįµį“øį“ø ā”‚ None ā”‚ į“ŗįµį“øį“ø ā”‚ None ā”‚ false ā”‚ +ā”‚ 42 ā”‚ Int64 ā”‚ 42 ā”‚ Int64 ā”‚ false ā”‚ +ā”‚ 43 ā”‚ Int64 ā”‚ 43 ā”‚ Int64 ā”‚ false ā”‚ +ā”‚ 42.42 ā”‚ String ā”‚ 42.42 ā”‚ String ā”‚ false ā”‚ +ā”‚ true ā”‚ Bool ā”‚ true ā”‚ Bool ā”‚ true ā”‚ +ā”‚ [1,2,3] ā”‚ Array(Int64) ā”‚ [1,2,3] ā”‚ Array(Int64) ā”‚ true ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` -If `K=1`, all types are converted to `String`: +Functions `isDynamicElementInSharedData` returns `true` for rows that are stored in a special shared data structure inside `Dynamic` and as we can see, resulting column contains only 2 types that are not stored in shared data structure. + +If `K=0`, all types will be inserted into single special subcolumn: ```text CREATE TABLE test (d Dynamic(max_types=4)) ENGINE = Memory; INSERT INTO test VALUES (NULL), (42), (43), ('42.42'), (true), ([1, 2, 3]); -SELECT d, dynamicType(d), d::Dynamic(max_types=1) as d2, dynamicType(d2) FROM test; +SELECT d, dynamicType(d), d::Dynamic(max_types=0) as d2, dynamicType(d2), isDynamicElementInSharedData(d2) FROM test; ``` ```text -ā”Œā”€dā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€dynamicType(d)ā”€ā”¬ā”€d2ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€dynamicType(d2)ā”€ā” -ā”‚ į“ŗįµį“øį“ø ā”‚ None ā”‚ į“ŗįµį“øį“ø ā”‚ None ā”‚ -ā”‚ 42 ā”‚ Int64 ā”‚ 42 ā”‚ String ā”‚ -ā”‚ 43 ā”‚ Int64 ā”‚ 43 ā”‚ String ā”‚ -ā”‚ 42.42 ā”‚ String ā”‚ 42.42 ā”‚ String ā”‚ -ā”‚ true ā”‚ Bool ā”‚ true ā”‚ String ā”‚ -ā”‚ [1,2,3] ā”‚ Array(Int64) ā”‚ [1,2,3] ā”‚ String ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +ā”Œā”€dā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€dynamicType(d)ā”€ā”¬ā”€d2ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€dynamicType(d2)ā”€ā”¬ā”€isDynamicElementInSharedData(d2)ā”€ā” +ā”‚ į“ŗįµį“øį“ø ā”‚ None ā”‚ į“ŗįµį“øį“ø ā”‚ None ā”‚ false ā”‚ +ā”‚ 42 ā”‚ Int64 ā”‚ 42 ā”‚ Int64 ā”‚ true ā”‚ +ā”‚ 43 ā”‚ Int64 ā”‚ 43 ā”‚ Int64 ā”‚ true ā”‚ +ā”‚ 42.42 ā”‚ String ā”‚ 42.42 ā”‚ String ā”‚ true ā”‚ +ā”‚ true ā”‚ Bool ā”‚ true ā”‚ Bool ā”‚ true ā”‚ +ā”‚ [1,2,3] ā”‚ Array(Int64) ā”‚ [1,2,3] ā”‚ Array(Int64) ā”‚ true ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` ## Reading Dynamic type from the data @@ -411,17 +413,17 @@ SELECT d, dynamicType(d) FROM test ORDER by d; ## Reaching the limit in number of different data types stored inside Dynamic -`Dynamic` data type can store only limited number of different data types inside. By default, this limit is 32, but you can change it in type declaration using syntax `Dynamic(max_types=N)` where N is between 1 and 255 (due to implementation details, it's impossible to have more than 255 different data types inside Dynamic). -When the limit is reached, all new data types inserted to `Dynamic` column will be casted to `String` and stored as `String` values. +`Dynamic` data type can store only limited number of different data types as separate subcolumns. By default, this limit is 32, but you can change it in type declaration using syntax `Dynamic(max_types=N)` where N is between 0 and 254 (due to implementation details, it's impossible to have more than 254 different data types that can be stored as separate subcolumns inside Dynamic). +When the limit is reached, all new data types inserted to `Dynamic` column will be inserted into a single shared data structure that stores values with different data types in binary form. Let's see what happens when the limit is reached in different scenarios. ### Reaching the limit during data parsing -During parsing of `Dynamic` values from the data, when the limit is reached for current block of data, all new values will be inserted as `String` values: +During parsing of `Dynamic` values from the data, when the limit is reached for current block of data, all new values will be inserted into shared data structure: ```sql -SELECT d, dynamicType(d) FROM format(JSONEachRow, 'd Dynamic(max_types=3)', ' +SELECT d, dynamicType(d), isDynamicElementInSharedData(d) FROM format(JSONEachRow, 'd Dynamic(max_types=3)', ' {"d" : 42} {"d" : [1, 2, 3]} {"d" : "Hello, World!"} @@ -432,22 +434,22 @@ SELECT d, dynamicType(d) FROM format(JSONEachRow, 'd Dynamic(max_types=3)', ' ``` ```text -ā”Œā”€dā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€dynamicType(d)ā”€ā” -ā”‚ 42 ā”‚ Int64 ā”‚ -ā”‚ [1,2,3] ā”‚ Array(Int64) ā”‚ -ā”‚ Hello, World! ā”‚ String ā”‚ -ā”‚ 2020-01-01 ā”‚ String ā”‚ -ā”‚ ["str1", "str2", "str3"] ā”‚ String ā”‚ -ā”‚ {"a" : 1, "b" : [1, 2, 3]} ā”‚ String ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +ā”Œā”€dā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€dynamicType(d)ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€isDynamicElementInSharedData(d)ā”€ā” +ā”‚ 42 ā”‚ Int64 ā”‚ false ā”‚ +ā”‚ [1,2,3] ā”‚ Array(Int64) ā”‚ false ā”‚ +ā”‚ Hello, World! ā”‚ String ā”‚ false ā”‚ +ā”‚ 2020-01-01 ā”‚ Date ā”‚ true ā”‚ +ā”‚ ['str1','str2','str3'] ā”‚ Array(String) ā”‚ true ā”‚ +ā”‚ (1,[1,2,3]) ā”‚ Tuple(a Int64, b Array(Int64)) ā”‚ true ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` -As we can see, after inserting 3 different data types `Int64`, `Array(Int64)` and `String` all new types were converted to `String`. +As we can see, after inserting 3 different data types `Int64`, `Array(Int64)` and `String` all new types were inserted into special shared data structure. ### During merges of data parts in MergeTree table engines -During merge of several data parts in MergeTree table the `Dynamic` column in the resulting data part can reach the limit of different data types inside and won't be able to store all types from source parts. -In this case ClickHouse chooses what types will remain after merge and what types will be casted to `String`. In most cases ClickHouse tries to keep the most frequent types and cast the rarest types to `String`, but it depends on the implementation. +During merge of several data parts in MergeTree table the `Dynamic` column in the resulting data part can reach the limit of different data types that can be stored in separate subcolumns inside and won't be able to store all types as subcolumns from source parts. +In this case ClickHouse chooses what types will remain as separate subcolumns after merge and what types will be inserted into shared data structure. In most cases ClickHouse tries to keep the most frequent types and store the rarest types in shared data structure, but it depends on the implementation. Let's see an example of such merge. First, let's create a table with `Dynamic` column, set the limit of different data types to `3` and insert values with `5` different types: @@ -463,17 +465,17 @@ INSERT INTO test SELECT number, 'str_' || toString(number) FROM numbers(1); Each insert will create a separate data pert with `Dynamic` column containing single type: ```sql -SELECT count(), dynamicType(d), _part FROM test GROUP BY _part, dynamicType(d) ORDER BY _part; +SELECT count(), dynamicType(d), isDynamicElementInSharedData(d), _part FROM test GROUP BY _part, dynamicType(d), isDynamicElementInSharedData(d) ORDER BY _part, count(); ``` ```text -ā”Œā”€count()ā”€ā”¬ā”€dynamicType(d)ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€_partā”€ā”€ā”€ā”€ā”€ā” -ā”‚ 5 ā”‚ UInt64 ā”‚ all_1_1_0 ā”‚ -ā”‚ 4 ā”‚ Array(UInt64) ā”‚ all_2_2_0 ā”‚ -ā”‚ 3 ā”‚ Date ā”‚ all_3_3_0 ā”‚ -ā”‚ 2 ā”‚ Map(UInt64, UInt64) ā”‚ all_4_4_0 ā”‚ -ā”‚ 1 ā”‚ String ā”‚ all_5_5_0 ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +ā”Œā”€count()ā”€ā”¬ā”€dynamicType(d)ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€isDynamicElementInSharedData(d)ā”€ā”¬ā”€_partā”€ā”€ā”€ā”€ā”€ā” +ā”‚ 5 ā”‚ UInt64 ā”‚ false ā”‚ all_1_1_0 ā”‚ +ā”‚ 4 ā”‚ Array(UInt64) ā”‚ false ā”‚ all_2_2_0 ā”‚ +ā”‚ 3 ā”‚ Date ā”‚ false ā”‚ all_3_3_0 ā”‚ +ā”‚ 2 ā”‚ Map(UInt64, UInt64) ā”‚ false ā”‚ all_4_4_0 ā”‚ +ā”‚ 1 ā”‚ String ā”‚ false ā”‚ all_5_5_0 ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` Now, let's merge all parts into one and see what will happen: @@ -481,18 +483,20 @@ Now, let's merge all parts into one and see what will happen: ```sql SYSTEM START MERGES test; OPTIMIZE TABLE test FINAL; -SELECT count(), dynamicType(d), _part FROM test GROUP BY _part, dynamicType(d) ORDER BY _part; +SELECT count(), dynamicType(d), isDynamicElementInSharedData(d), _part FROM test GROUP BY _part, dynamicType(d), isDynamicElementInSharedData(d) ORDER BY _part, count() desc; ``` ```text -ā”Œā”€count()ā”€ā”¬ā”€dynamicType(d)ā”€ā”¬ā”€_partā”€ā”€ā”€ā”€ā”€ā” -ā”‚ 5 ā”‚ UInt64 ā”‚ all_1_5_2 ā”‚ -ā”‚ 6 ā”‚ String ā”‚ all_1_5_2 ā”‚ -ā”‚ 4 ā”‚ Array(UInt64) ā”‚ all_1_5_2 ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +ā”Œā”€count()ā”€ā”¬ā”€dynamicType(d)ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€isDynamicElementInSharedData(d)ā”€ā”¬ā”€_partā”€ā”€ā”€ā”€ā”€ā” +ā”‚ 5 ā”‚ UInt64 ā”‚ false ā”‚ all_1_5_2 ā”‚ +ā”‚ 4 ā”‚ Array(UInt64) ā”‚ false ā”‚ all_1_5_2 ā”‚ +ā”‚ 3 ā”‚ Date ā”‚ false ā”‚ all_1_5_2 ā”‚ +ā”‚ 2 ā”‚ Map(UInt64, UInt64) ā”‚ true ā”‚ all_1_5_2 ā”‚ +ā”‚ 1 ā”‚ String ā”‚ true ā”‚ all_1_5_2 ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` -As we can see, ClickHouse kept the most frequent types `UInt64` and `Array(UInt64)` and casted all other types to `String`. +As we can see, ClickHouse kept the most frequent types `UInt64` and `Array(UInt64)` as subcolumns and inserted all other types into shared data. ## JSONExtract functions with Dynamic @@ -509,22 +513,23 @@ SELECT JSONExtract('{"a" : [1, 2, 3]}', 'a', 'Dynamic') AS dynamic, dynamicType( ``` ```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``` +SELECT JSONExtract('{"obj" : {"a" : 42, "b" : "Hello", "c" : [1,2,3]}}', 'obj', 'Map(String, Dynamic)') AS map_of_dynamics, mapApply((k, v) -> (k, dynamicType(v)), map_of_dynamics) AS map_of_dynamic_types +``` ```text -ā”Œā”€map_of_dynamicsā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€map_of_dynamic_typesā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā” -ā”‚ {'a':42,'b':'Hello','c':[1,2,3]} ā”‚ {'a':'UInt32','b':'String','c':'Array(UInt32)'} ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +ā”Œā”€map_of_dynamicsā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€map_of_dynamic_typesā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā” +ā”‚ {'a':42,'b':'Hello','c':[1,2,3]} ā”‚ {'a':'Int64','b':'String','c':'Array(Nullable(Int64))'} ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` ```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``` +SELECT JSONExtractKeysAndValues('{"a" : 42, "b" : "Hello", "c" : [1,2,3]}', 'Dynamic') AS dynamics, arrayMap(x -> (x.1, dynamicType(x.2)), dynamics) AS dynamic_types``` ``` ```text -ā”Œā”€dynamicsā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€dynamic_typesā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā” -ā”‚ [('a',42),('b','Hello'),('c',[1,2,3])] ā”‚ [('a','UInt32'),('b','String'),('c','Array(UInt32)')] ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +ā”Œā”€dynamicsā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€dynamic_typesā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā” +ā”‚ [('a',42),('b','Hello'),('c',[1,2,3])] ā”‚ [('a','Int64'),('b','String'),('c','Array(Nullable(Int64))')] ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` ### Binary output format diff --git a/docs/en/sql-reference/data-types/geo.md b/docs/en/sql-reference/data-types/geo.md index 7ffc7447d96..8ce53bb2ef2 100644 --- a/docs/en/sql-reference/data-types/geo.md +++ b/docs/en/sql-reference/data-types/geo.md @@ -52,6 +52,48 @@ Result: ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` +## LineString + +`LineString` is a line stored as an array of points: [Array](array.md)([Point](#point)). + +**Example** + +Query: + +```sql +CREATE TABLE geo_linestring (l LineString) ENGINE = Memory(); +INSERT INTO geo_linestring VALUES([(0, 0), (10, 0), (10, 10), (0, 10)]); +SELECT l, toTypeName(l) FROM geo_linestring; +``` +Result: + +``` text +ā”Œā”€rā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€toTypeName(r)ā”€ā” +ā”‚ [(0,0),(10,0),(10,10),(0,10)] ā”‚ LineString ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +``` + +## MultiLineString + +`MultiLineString` is multiple lines stored as an array of `LineString`: [Array](array.md)([LineString](#linestring)). + +**Example** + +Query: + +```sql +CREATE TABLE geo_multilinestring (l MultiLineString) ENGINE = Memory(); +INSERT INTO geo_multilinestring VALUES([[(0, 0), (10, 0), (10, 10), (0, 10)], [(1, 1), (2, 2), (3, 3)]]); +SELECT l, toTypeName(l) FROM geo_multilinestring; +``` +Result: + +``` text +ā”Œā”€lā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€toTypeName(l)ā”€ā”€ā”€ā” +ā”‚ [[(0,0),(10,0),(10,10),(0,10)],[(1,1),(2,2),(3,3)]] ā”‚ MultiLineString ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +``` + ## Polygon `Polygon` is a polygon with holes stored as an array of rings: [Array](array.md)([Ring](#ring)). First element of outer array is the outer shape of polygon and all the following elements are holes. diff --git a/docs/en/sql-reference/functions/geo/polygon.md b/docs/en/sql-reference/functions/geo/polygon.md index 25a7a1fac8e..be9e9810626 100644 --- a/docs/en/sql-reference/functions/geo/polygon.md +++ b/docs/en/sql-reference/functions/geo/polygon.md @@ -6,11 +6,13 @@ title: "Functions for Working with Polygons" ## WKT -Returns a WKT (Well Known Text) geometric object from various [Geo Data Types](../../data-types/geo.md). Supported WKT objects are: +Returns a WKT (Well Known Text) geometric object from various [Geo Data Types](../../data-types/geo.md). Supported WKT objects are: - POINT - POLYGON - MULTIPOLYGON +- LINESTRING +- MULTILINESTRING **Syntax** @@ -26,12 +28,16 @@ WKT(geo_data) - [Ring](../../data-types/geo.md#ring) - [Polygon](../../data-types/geo.md#polygon) - [MultiPolygon](../../data-types/geo.md#multipolygon) +- [LineString](../../data-types/geo.md#linestring) +- [MultiLineString](../../data-types/geo.md#multilinestring) **Returned value** - WKT geometric object `POINT` is returned for a Point. - WKT geometric object `POLYGON` is returned for a Polygon -- WKT geometric object `MULTIPOLYGON` is returned for a MultiPolygon. +- WKT geometric object `MULTIPOLYGON` is returned for a MultiPolygon. +- WKT geometric object `LINESTRING` is returned for a LineString. +- WKT geometric object `MULTILINESTRING` is returned for a MultiLineString. **Examples** @@ -84,7 +90,7 @@ SELECT ### Input parameters -String starting with `MULTIPOLYGON` +String starting with `MULTIPOLYGON` ### Returned value @@ -170,6 +176,34 @@ SELECT readWKTLineString('LINESTRING (1 1, 2 2, 3 3, 1 1)'); [(1,1),(2,2),(3,3),(1,1)] ``` +## readWKTMultiLineString + +Parses a Well-Known Text (WKT) representation of a MultiLineString geometry and returns it in the internal ClickHouse format. + +### Syntax + +```sql +readWKTMultiLineString(wkt_string) +``` + +### Arguments + +- `wkt_string`: The input WKT string representing a MultiLineString geometry. + +### Returned value + +The function returns a ClickHouse internal representation of the multilinestring geometry. + +### Example + +```sql +SELECT readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3), (4 4, 5 5, 6 6))'); +``` + +```response +[[(1,1),(2,2),(3,3)],[(4,4),(5,5),(6,6)]] +``` + ## readWKTRing Parses a Well-Known Text (WKT) representation of a Polygon geometry and returns a ring (closed linestring) in the internal ClickHouse format. @@ -219,7 +253,7 @@ UInt8, 0 for false, 1 for true ## polygonsDistanceSpherical -Calculates the minimal distance between two points where one point belongs to the first polygon and the second to another polygon. Spherical means that coordinates are interpreted as coordinates on a pure and ideal sphere, which is not true for the Earth. Using this type of coordinate system speeds up execution, but of course is not precise. +Calculates the minimal distance between two points where one point belongs to the first polygon and the second to another polygon. Spherical means that coordinates are interpreted as coordinates on a pure and ideal sphere, which is not true for the Earth. Using this type of coordinate system speeds up execution, but of course is not precise. ### Example diff --git a/docs/en/sql-reference/functions/other-functions.md b/docs/en/sql-reference/functions/other-functions.md index afd779ca8cd..1b50104da52 100644 --- a/docs/en/sql-reference/functions/other-functions.md +++ b/docs/en/sql-reference/functions/other-functions.md @@ -4189,3 +4189,94 @@ Result: ā”‚ 32 ā”‚ ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` + +## getSubcolumn + +Takes a table expression or identifier and constant string with the name of the sub-column, and returns the requested sub-column extracted from the expression. + +**Syntax** + +```sql +getSubcolumn(col_name, subcol_name) +``` + +**Arguments** + +- `col_name` ā€” Table expression or identifier. [Expression](../syntax.md/#expressions), [Identifier](../syntax.md/#identifiers). +- `subcol_name` ā€” The name of the sub-column. [String](../data-types/string.md). + +**Returned value** + +- Returns the extracted sub-column. + +**Example** + +Query: + +```sql +CREATE TABLE t_arr (arr Array(Tuple(subcolumn1 UInt32, subcolumn2 String))) ENGINE = MergeTree ORDER BY tuple(); +INSERT INTO t_arr VALUES ([(1, 'Hello'), (2, 'World')]), ([(3, 'This'), (4, 'is'), (5, 'subcolumn')]); +SELECT getSubcolumn(arr, 'subcolumn1'), getSubcolumn(arr, 'subcolumn2') FROM t_arr; +``` + +Result: + +```response + ā”Œā”€getSubcolumn(arr, 'subcolumn1')ā”€ā”¬ā”€getSubcolumn(arr, 'subcolumn2')ā”€ā” +1. ā”‚ [1,2] ā”‚ ['Hello','World'] ā”‚ +2. ā”‚ [3,4,5] ā”‚ ['This','is','subcolumn'] ā”‚ + ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +``` + +## getTypeSerializationStreams + +Enumerates stream paths of a data type. + +:::note +This function is intended for use by developers. +::: + +**Syntax** + +```sql +getTypeSerializationStreams(col) +``` + +**Arguments** + +- `col` ā€” Column or string representation of a data-type from which the data type will be detected. + +**Returned value** + +- Returns an array with all the serialization sub-stream paths.[Array](../data-types/array.md)([String](../data-types/string.md)). + +**Examples** + +Query: + +```sql +SELECT getTypeSerializationStreams(tuple('a', 1, 'b', 2)); +``` + +Result: + +```response + ā”Œā”€getTypeSerializationStreams(('a', 1, 'b', 2))ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā” +1. ā”‚ ['{TupleElement(1), Regular}','{TupleElement(2), Regular}','{TupleElement(3), Regular}','{TupleElement(4), Regular}'] ā”‚ + ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +``` + +Query: + +```sql +SELECT getTypeSerializationStreams('Map(String, Int64)'); +``` + +Result: + +```response + ā”Œā”€getTypeSerializationStreams('Map(String, Int64)')ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā” +1. ā”‚ ['{ArraySizes}','{ArrayElements, TupleElement(keys), Regular}','{ArrayElements, TupleElement(values), Regular}'] ā”‚ + ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +``` + diff --git a/docs/en/sql-reference/functions/type-conversion-functions.md b/docs/en/sql-reference/functions/type-conversion-functions.md index 1e618b8cdab..9e4b39e95e4 100644 --- a/docs/en/sql-reference/functions/type-conversion-functions.md +++ b/docs/en/sql-reference/functions/type-conversion-functions.md @@ -64,9 +64,8 @@ toInt8(expr) - `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Unsupported arguments: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -95,7 +94,7 @@ SELECT toInt8(-8), toInt8(-8.8), toInt8('-8') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -156,7 +155,7 @@ Query: SELECT toInt8OrZero('-8'), toInt8OrZero('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -216,7 +215,7 @@ Query: SELECT toInt8OrNull('-8'), toInt8OrNull('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -251,9 +250,8 @@ toInt8OrDefault(expr[, default]) - `default` (optional) ā€” The default value to return if parsing to type `Int8` is unsuccessful. [Int8](../data-types/int-uint.md). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Arguments for which the default value is returned: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -281,7 +279,7 @@ Query: SELECT toInt8OrDefault('-8', CAST('-1', 'Int8')), toInt8OrDefault('abc', CAST('-1', 'Int8')) -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -314,9 +312,8 @@ toInt16(expr) - `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Unsupported arguments: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -345,7 +342,7 @@ SELECT toInt16(-16), toInt16(-16.16), toInt16('-16') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -406,7 +403,7 @@ Query: SELECT toInt16OrZero('-16'), toInt16OrZero('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -466,7 +463,7 @@ Query: SELECT toInt16OrNull('-16'), toInt16OrNull('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -501,9 +498,8 @@ toInt16OrDefault(expr[, default]) - `default` (optional) ā€” The default value to return if parsing to type `Int16` is unsuccessful. [Int16](../data-types/int-uint.md). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Arguments for which the default value is returned: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -531,7 +527,7 @@ Query: SELECT toInt16OrDefault('-16', CAST('-1', 'Int16')), toInt16OrDefault('abc', CAST('-1', 'Int16')) -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -564,9 +560,8 @@ toInt32(expr) - `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Unsupported arguments: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -595,7 +590,7 @@ SELECT toInt32(-32), toInt32(-32.32), toInt32('-32') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -656,7 +651,7 @@ Query: SELECT toInt32OrZero('-32'), toInt32OrZero('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -715,7 +710,7 @@ Query: SELECT toInt32OrNull('-32'), toInt32OrNull('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -750,9 +745,8 @@ toInt32OrDefault(expr[, default]) - `default` (optional) ā€” The default value to return if parsing to type `Int32` is unsuccessful. [Int32](../data-types/int-uint.md). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Arguments for which the default value is returned: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -780,7 +774,7 @@ Query: SELECT toInt32OrDefault('-32', CAST('-1', 'Int32')), toInt32OrDefault('abc', CAST('-1', 'Int32')) -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -813,9 +807,8 @@ toInt64(expr) - `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Unsupported types: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -844,7 +837,7 @@ SELECT toInt64(-64), toInt64(-64.64), toInt64('-64') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -905,7 +898,7 @@ Query: SELECT toInt64OrZero('-64'), toInt64OrZero('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -965,7 +958,7 @@ Query: SELECT toInt64OrNull('-64'), toInt64OrNull('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1000,9 +993,8 @@ toInt64OrDefault(expr[, default]) - `default` (optional) ā€” The default value to return if parsing to type `Int64` is unsuccessful. [Int64](../data-types/int-uint.md). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Arguments for which the default value is returned: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -1030,7 +1022,7 @@ Query: SELECT toInt64OrDefault('-64', CAST('-1', 'Int64')), toInt64OrDefault('abc', CAST('-1', 'Int64')) -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1063,9 +1055,8 @@ toInt128(expr) - `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Unsupported arguments: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -1093,7 +1084,7 @@ SELECT toInt128(-128), toInt128(-128.8), toInt128('-128') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1154,7 +1145,7 @@ Query: SELECT toInt128OrZero('-128'), toInt128OrZero('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1214,7 +1205,7 @@ Query: SELECT toInt128OrNull('-128'), toInt128OrNull('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1279,7 +1270,7 @@ Query: SELECT toInt128OrDefault('-128', CAST('-1', 'Int128')), toInt128OrDefault('abc', CAST('-1', 'Int128')) -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1312,9 +1303,8 @@ toInt256(expr) - `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Unsupported arguments: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -1342,7 +1332,7 @@ SELECT toInt256(-256), toInt256(-256.256), toInt256('-256') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1403,7 +1393,7 @@ Query: SELECT toInt256OrZero('-256'), toInt256OrZero('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1463,7 +1453,7 @@ Query: SELECT toInt256OrNull('-256'), toInt256OrNull('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1498,9 +1488,8 @@ toInt256OrDefault(expr[, default]) - `default` (optional) ā€” The default value to return if parsing to type `Int256` is unsuccessful. [Int256](../data-types/int-uint.md). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Arguments for which the default value is returned: - String representations of Float32/64 values, including `NaN` and `Inf` @@ -1528,7 +1517,7 @@ Query: SELECT toInt256OrDefault('-256', CAST('-1', 'Int256')), toInt256OrDefault('abc', CAST('-1', 'Int256')) -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1546,7 +1535,7 @@ toInt256OrDefault('abc', CAST('-1', 'Int256')): -1 - [`toInt256OrZero`](#toint256orzero). - [`toInt256OrNull`](#toint256ornull). -# toUInt8 +## toUInt8 Converts an input value to a value of type [`UInt8`](../data-types/int-uint.md). Throws an exception in case of an error. @@ -1561,9 +1550,8 @@ toUInt8(expr) - `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Unsupported arguments: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -1592,7 +1580,7 @@ SELECT toUInt8(8), toUInt8(8.8), toUInt8('8') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1653,7 +1641,7 @@ Query: SELECT toUInt8OrZero('-8'), toUInt8OrZero('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1713,7 +1701,7 @@ Query: SELECT toUInt8OrNull('8'), toUInt8OrNull('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1748,9 +1736,8 @@ toUInt8OrDefault(expr[, default]) - `default` (optional) ā€” The default value to return if parsing to type `UInt8` is unsuccessful. [UInt8](../data-types/int-uint.md). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Arguments for which the default value is returned: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -1778,7 +1765,7 @@ Query: SELECT toUInt8OrDefault('8', CAST('0', 'UInt8')), toUInt8OrDefault('abc', CAST('0', 'UInt8')) -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1811,9 +1798,8 @@ toUInt16(expr) - `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Unsupported arguments: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -1842,7 +1828,7 @@ SELECT toUInt16(16), toUInt16(16.16), toUInt16('16') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1903,7 +1889,7 @@ Query: SELECT toUInt16OrZero('16'), toUInt16OrZero('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1963,7 +1949,7 @@ Query: SELECT toUInt16OrNull('16'), toUInt16OrNull('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -1998,9 +1984,8 @@ toUInt16OrDefault(expr[, default]) - `default` (optional) ā€” The default value to return if parsing to type `UInt16` is unsuccessful. [UInt16](../data-types/int-uint.md). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Arguments for which the default value is returned: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -2028,7 +2013,7 @@ Query: SELECT toUInt16OrDefault('16', CAST('0', 'UInt16')), toUInt16OrDefault('abc', CAST('0', 'UInt16')) -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2061,9 +2046,8 @@ toUInt32(expr) - `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Unsupported arguments: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -2092,7 +2076,7 @@ SELECT toUInt32(32), toUInt32(32.32), toUInt32('32') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2154,7 +2138,7 @@ Query: SELECT toUInt32OrZero('32'), toUInt32OrZero('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2214,7 +2198,7 @@ Query: SELECT toUInt32OrNull('32'), toUInt32OrNull('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2249,9 +2233,8 @@ toUInt32OrDefault(expr[, default]) - `default` (optional) ā€” The default value to return if parsing to type `UInt32` is unsuccessful. [UInt32](../data-types/int-uint.md). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Arguments for which the default value is returned: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -2279,7 +2262,7 @@ Query: SELECT toUInt32OrDefault('32', CAST('0', 'UInt32')), toUInt32OrDefault('abc', CAST('0', 'UInt32')) -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2312,9 +2295,8 @@ toUInt64(expr) - `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Unsupported types: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -2343,7 +2325,7 @@ SELECT toUInt64(64), toUInt64(64.64), toUInt64('64') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2362,7 +2344,7 @@ toUInt64('64'): 64 - [`toUInt64OrNull`](#touint64ornull). - [`toUInt64OrDefault`](#touint64ordefault). -## toInt64OrZero +## toUInt64OrZero Like [`toUInt64`](#touint64), this function converts an input value to a value of type [UInt64](../data-types/int-uint.md) but returns `0` in case of an error. @@ -2404,7 +2386,7 @@ Query: SELECT toUInt64OrZero('64'), toUInt64OrZero('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2464,7 +2446,7 @@ Query: SELECT toUInt64OrNull('64'), toUInt64OrNull('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2499,9 +2481,8 @@ toUInt64OrDefault(expr[, default]) - `defauult` (optional) ā€” The default value to return if parsing to type `UInt64` is unsuccessful. [UInt64](../data-types/int-uint.md). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Arguments for which the default value is returned: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -2529,7 +2510,7 @@ Query: SELECT toUInt64OrDefault('64', CAST('0', 'UInt64')), toUInt64OrDefault('abc', CAST('0', 'UInt64')) -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2562,9 +2543,8 @@ toUInt128(expr) - `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Unsupported arguments: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -2592,7 +2572,7 @@ SELECT toUInt128(128), toUInt128(128.8), toUInt128('128') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2653,7 +2633,7 @@ Query: SELECT toUInt128OrZero('128'), toUInt128OrZero('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2713,7 +2693,7 @@ Query: SELECT toUInt128OrNull('128'), toUInt128OrNull('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2778,7 +2758,7 @@ Query: SELECT toUInt128OrDefault('128', CAST('0', 'UInt128')), toUInt128OrDefault('abc', CAST('0', 'UInt128')) -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2811,9 +2791,8 @@ toUInt256(expr) - `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Unsupported arguments: - String representations of Float32/64 values, including `NaN` and `Inf`. @@ -2841,7 +2820,7 @@ SELECT toUInt256(256), toUInt256(256.256), toUInt256('256') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2902,7 +2881,7 @@ Query: SELECT toUInt256OrZero('256'), toUInt256OrZero('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2962,7 +2941,7 @@ Query: SELECT toUInt256OrNull('256'), toUInt256OrNull('abc') -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -2997,9 +2976,8 @@ toUInt256OrDefault(expr[, default]) - `default` (optional) ā€” The default value to return if parsing to type `UInt256` is unsuccessful. [UInt256](../data-types/int-uint.md). Supported arguments: -- Values of type (U)Int8/16/32/64/128/256. +- Values or string representations of type (U)Int8/16/32/64/128/256. - Values of type Float32/64. -- String representations of (U)Int8/16/32/128/256. Arguments for which the default value is returned: - String representations of Float32/64 values, including `NaN` and `Inf` @@ -3027,7 +3005,7 @@ Query: SELECT toUInt256OrDefault('-256', CAST('0', 'UInt256')), toUInt256OrDefault('abc', CAST('0', 'UInt256')) -FORMAT vertical; +FORMAT Vertical; ``` Result: @@ -3045,13 +3023,425 @@ toUInt256OrDefault('abc', CAST('0', 'UInt256')): 0 - [`toUInt256OrZero`](#touint256orzero). - [`toUInt256OrNull`](#touint256ornull). -## toFloat(32\|64) +## toFloat32 -## toFloat(32\|64)OrZero +Converts an input value to a value of type [`Float32`](../data-types/float.md). Throws an exception in case of an error. -## toFloat(32\|64)OrNull +**Syntax** -## toFloat(32\|64)OrDefault +```sql +toFloat32(expr) +``` + +**Arguments** + +- `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). + +Supported arguments: +- Values of type (U)Int8/16/32/64/128/256. +- String representations of (U)Int8/16/32/128/256. +- Values of type Float32/64, including `NaN` and `Inf`. +- String representations of Float32/64, including `NaN` and `Inf` (case-insensitive). + +Unsupported arguments: +- String representations of binary and hexadecimal values, e.g. `SELECT toFloat32('0xc0fe');`. + +**Returned value** + +- 32-bit floating point value. [Float32](../data-types/float.md). + +**Example** + +Query: + +```sql +SELECT + toFloat32(42.7), + toFloat32('42.7'), + toFloat32('NaN') +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +toFloat32(42.7): 42.7 +toFloat32('42.7'): 42.7 +toFloat32('NaN'): nan +``` + +**See also** + +- [`toFloat32OrZero`](#tofloat32orzero). +- [`toFloat32OrNull`](#tofloat32ornull). +- [`toFloat32OrDefault`](#tofloat32ordefault). + +## toFloat32OrZero + +Like [`toFloat32`](#tofloat32), this function converts an input value to a value of type [Float32](../data-types/float.md) but returns `0` in case of an error. + +**Syntax** + +```sql +toFloat32OrZero(x) +``` + +**Arguments** + +- `x` ā€” A String representation of a number. [String](../data-types/string.md). + +Supported arguments: +- String representations of (U)Int8/16/32/128/256, Float32/64. + +Unsupported arguments (return `0`): +- String representations of binary and hexadecimal values, e.g. `SELECT toFloat32OrZero('0xc0fe');`. + +**Returned value** + +- 32-bit Float value if successful, otherwise `0`. [Float32](../data-types/float.md). + +**Example** + +Query: + +```sql +SELECT + toFloat32OrZero('42.7'), + toFloat32OrZero('abc') +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +toFloat32OrZero('42.7'): 42.7 +toFloat32OrZero('abc'): 0 +``` + +**See also** + +- [`toFloat32`](#tofloat32). +- [`toFloat32OrNull`](#tofloat32ornull). +- [`toFloat32OrDefault`](#tofloat32ordefault). + +## toFloat32OrNull + +Like [`toFloat32`](#tofloat32), this function converts an input value to a value of type [Float32](../data-types/float.md) but returns `NULL` in case of an error. + +**Syntax** + +```sql +toFloat32OrNull(x) +``` + +**Arguments** + +- `x` ā€” A String representation of a number. [String](../data-types/string.md). + +Supported arguments: +- String representations of (U)Int8/16/32/128/256, Float32/64. + +Unsupported arguments (return `\N`): +- String representations of binary and hexadecimal values, e.g. `SELECT toFloat32OrNull('0xc0fe');`. + +**Returned value** + +- 32-bit Float value if successful, otherwise `\N`. [Float32](../data-types/float.md). + +**Example** + +Query: + +```sql +SELECT + toFloat32OrNull('42.7'), + toFloat32OrNull('abc') +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +toFloat32OrNull('42.7'): 42.7 +toFloat32OrNull('abc'): į“ŗįµį“øį“ø +``` + +**See also** + +- [`toFloat32`](#tofloat32). +- [`toFloat32OrZero`](#tofloat32orzero). +- [`toFloat32OrDefault`](#tofloat32ordefault). + +## toFloat32OrDefault + +Like [`toFloat32`](#tofloat32), this function converts an input value to a value of type [Float32](../data-types/float.md) but returns the default value in case of an error. +If no `default` value is passed then `0` is returned in case of an error. + +**Syntax** + +```sql +toFloat32OrDefault(expr[, default]) +``` + +**Arguments** + +- `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions) / [String](../data-types/string.md). +- `default` (optional) ā€” The default value to return if parsing to type `Float32` is unsuccessful. [Float32](../data-types/float.md). + +Supported arguments: +- Values of type (U)Int8/16/32/64/128/256. +- String representations of (U)Int8/16/32/128/256. +- Values of type Float32/64, including `NaN` and `Inf`. +- String representations of Float32/64, including `NaN` and `Inf` (case-insensitive). + +Arguments for which the default value is returned: +- String representations of binary and hexadecimal values, e.g. `SELECT toFloat32OrDefault('0xc0fe', CAST('0', 'Float32'));`. + +**Returned value** + +- 32-bit Float value if successful, otherwise returns the default value if passed or `0` if not. [Float32](../data-types/float.md). + +**Example** + +Query: + +```sql +SELECT + toFloat32OrDefault('8', CAST('0', 'Float32')), + toFloat32OrDefault('abc', CAST('0', 'Float32')) +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +toFloat32OrDefault('8', CAST('0', 'Float32')): 8 +toFloat32OrDefault('abc', CAST('0', 'Float32')): 0 +``` + +**See also** + +- [`toFloat32`](#tofloat32). +- [`toFloat32OrZero`](#tofloat32orzero). +- [`toFloat32OrNull`](#tofloat32ornull). + +## toFloat64 + +Converts an input value to a value of type [`Float64`](../data-types/float.md). Throws an exception in case of an error. + +**Syntax** + +```sql +toFloat64(expr) +``` + +**Arguments** + +- `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). + +Supported arguments: +- Values of type (U)Int8/16/32/64/128/256. +- String representations of (U)Int8/16/32/128/256. +- Values of type Float32/64, including `NaN` and `Inf`. +- String representations of type Float32/64, including `NaN` and `Inf` (case-insensitive). + +Unsupported arguments: +- String representations of binary and hexadecimal values, e.g. `SELECT toFloat64('0xc0fe');`. + +**Returned value** + +- 64-bit floating point value. [Float64](../data-types/float.md). + +**Example** + +Query: + +```sql +SELECT + toFloat64(42.7), + toFloat64('42.7'), + toFloat64('NaN') +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +toFloat64(42.7): 42.7 +toFloat64('42.7'): 42.7 +toFloat64('NaN'): nan +``` + +**See also** + +- [`toFloat64OrZero`](#tofloat64orzero). +- [`toFloat64OrNull`](#tofloat64ornull). +- [`toFloat64OrDefault`](#tofloat64ordefault). + +## toFloat64OrZero + +Like [`toFloat64`](#tofloat64), this function converts an input value to a value of type [Float64](../data-types/float.md) but returns `0` in case of an error. + +**Syntax** + +```sql +toFloat64OrZero(x) +``` + +**Arguments** + +- `x` ā€” A String representation of a number. [String](../data-types/string.md). + +Supported arguments: +- String representations of (U)Int8/16/32/128/256, Float32/64. + +Unsupported arguments (return `0`): +- String representations of binary and hexadecimal values, e.g. `SELECT toFloat64OrZero('0xc0fe');`. + +**Returned value** + +- 64-bit Float value if successful, otherwise `0`. [Float64](../data-types/float.md). + +**Example** + +Query: + +```sql +SELECT + toFloat64OrZero('42.7'), + toFloat64OrZero('abc') +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +toFloat64OrZero('42.7'): 42.7 +toFloat64OrZero('abc'): 0 +``` + +**See also** + +- [`toFloat64`](#tofloat64). +- [`toFloat64OrNull`](#tofloat64ornull). +- [`toFloat64OrDefault`](#tofloat64ordefault). + +## toFloat64OrNull + +Like [`toFloat64`](#tofloat64), this function converts an input value to a value of type [Float64](../data-types/float.md) but returns `NULL` in case of an error. + +**Syntax** + +```sql +toFloat64OrNull(x) +``` + +**Arguments** + +- `x` ā€” A String representation of a number. [String](../data-types/string.md). + +Supported arguments: +- String representations of (U)Int8/16/32/128/256, Float32/64. + +Unsupported arguments (return `\N`): +- String representations of binary and hexadecimal values, e.g. `SELECT toFloat64OrNull('0xc0fe');`. + +**Returned value** + +- 64-bit Float value if successful, otherwise `\N`. [Float64](../data-types/float.md). + +**Example** + +Query: + +```sql +SELECT + toFloat64OrNull('42.7'), + toFloat64OrNull('abc') +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +toFloat64OrNull('42.7'): 42.7 +toFloat64OrNull('abc'): į“ŗįµį“øį“ø +``` + +**See also** + +- [`toFloat64`](#tofloat64). +- [`toFloat64OrZero`](#tofloat64orzero). +- [`toFloat64OrDefault`](#tofloat64ordefault). + +## toFloat64OrDefault + +Like [`toFloat64`](#tofloat64), this function converts an input value to a value of type [Float64](../data-types/float.md) but returns the default value in case of an error. +If no `default` value is passed then `0` is returned in case of an error. + +**Syntax** + +```sql +toFloat64OrDefault(expr[, default]) +``` + +**Arguments** + +- `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions) / [String](../data-types/string.md). +- `default` (optional) ā€” The default value to return if parsing to type `Float64` is unsuccessful. [Float64](../data-types/float.md). + +Supported arguments: +- Values of type (U)Int8/16/32/64/128/256. +- String representations of (U)Int8/16/32/128/256. +- Values of type Float32/64, including `NaN` and `Inf`. +- String representations of Float32/64, including `NaN` and `Inf` (case-insensitive). + +Arguments for which the default value is returned: +- String representations of binary and hexadecimal values, e.g. `SELECT toFloat64OrDefault('0xc0fe', CAST('0', 'Float64'));`. + +**Returned value** + +- 64-bit Float value if successful, otherwise returns the default value if passed or `0` if not. [Float64](../data-types/float.md). + +**Example** + +Query: + +```sql +SELECT + toFloat64OrDefault('8', CAST('0', 'Float64')), + toFloat64OrDefault('abc', CAST('0', 'Float64')) +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +toFloat64OrDefault('8', CAST('0', 'Float64')): 8 +toFloat64OrDefault('abc', CAST('0', 'Float64')): 0 +``` + +**See also** + +- [`toFloat64`](#tofloat64). +- [`toFloat64OrZero`](#tofloat64orzero). +- [`toFloat64OrNull`](#tofloat64ornull). ## toDate @@ -3542,175 +3932,1022 @@ SELECT toDateTime64('2019-01-01 00:00:00', 3, 'Asia/Istanbul') AS value, toTypeN ## toDateTime64OrDefault -## toDecimal(32\|64\|128\|256) +## toDecimal32 -Converts `value` to the [Decimal](../data-types/decimal.md) data type with precision of `S`. The `value` can be a number or a string. The `S` (scale) parameter specifies the number of decimal places. +Converts an input value to a value of type [`Decimal(9, S)`](../data-types/decimal.md) with scale of `S`. Throws an exception in case of an error. -- `toDecimal32(value, S)` -- `toDecimal64(value, S)` -- `toDecimal128(value, S)` -- `toDecimal256(value, S)` +**Syntax** -## toDecimal(32\|64\|128\|256)OrNull - -Converts an input string to a [Nullable(Decimal(P,S))](../data-types/decimal.md) data type value. This family of functions includes: - -- `toDecimal32OrNull(expr, S)` ā€” Results in `Nullable(Decimal32(S))` data type. -- `toDecimal64OrNull(expr, S)` ā€” Results in `Nullable(Decimal64(S))` data type. -- `toDecimal128OrNull(expr, S)` ā€” Results in `Nullable(Decimal128(S))` data type. -- `toDecimal256OrNull(expr, S)` ā€” Results in `Nullable(Decimal256(S))` data type. - -These functions should be used instead of `toDecimal*()` functions, if you prefer to get a `NULL` value instead of an exception in the event of an input value parsing error. +```sql +toDecimal32(expr, S) +``` **Arguments** -- `expr` ā€” [Expression](../syntax.md/#syntax-expressions), returns a value in the [String](../data-types/string.md) data type. ClickHouse expects the textual representation of the decimal number. For example, `'1.111'`. -- `S` ā€” Scale, the number of decimal places in the resulting value. +- `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). +- `S` ā€” Scale parameter between 0 and 9, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). + +Supported arguments: +- Values or string representations of type (U)Int8/16/32/64/128/256. +- Values or string representations of type Float32/64. + +Unsupported arguments: +- Values or string representations of Float32/64 values `NaN` and `Inf` (case-insensitive). +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal32('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal32`: `( -1 * 10^(9 - S), 1 * 10^(9 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an exception. +::: **Returned value** -A value in the `Nullable(Decimal(P,S))` data type. The value contains: +- Value of type `Decimal(9, S)`. [Decimal32(S)](../data-types/int-uint.md). -- Number with `S` decimal places, if ClickHouse interprets the input string as a number. -- `NULL`, if ClickHouse canā€™t interpret the input string as a number or if the input number contains more than `S` decimal places. - -**Examples** +**Example** Query: -``` sql -SELECT toDecimal32OrNull(toString(-1.111), 5) AS val, toTypeName(val); +```sql +SELECT + toDecimal32(2, 1) AS a, toTypeName(a) AS type_a, + toDecimal32(4.2, 2) AS b, toTypeName(b) AS type_b, + toDecimal32('4.2', 3) AS c, toTypeName(c) AS type_c +FORMAT Vertical; ``` Result: ```response -ā”Œā”€ā”€ā”€ā”€valā”€ā”¬ā”€toTypeName(toDecimal32OrNull(toString(-1.111), 5))ā”€ā” -ā”‚ -1.111 ā”‚ Nullable(Decimal(9, 5)) ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: 2 +type_a: Decimal(9, 1) +b: 4.2 +type_b: Decimal(9, 2) +c: 4.2 +type_c: Decimal(9, 3) ``` -Query: +**See also** -``` sql -SELECT toDecimal32OrNull(toString(-1.111), 2) AS val, toTypeName(val); +- [`toDecimal32OrZero`](#todecimal32orzero). +- [`toDecimal32OrNull`](#todecimal32ornull). +- [`toDecimal32OrDefault`](#todecimal32ordefault). + +## toDecimal32OrZero + +Like [`toDecimal32`](#todecimal32), this function converts an input value to a value of type [Decimal(9, S)](../data-types/decimal.md) but returns `0` in case of an error. + +**Syntax** + +```sql +toDecimal32OrZero(expr, S) ``` -Result: - -```response -ā”Œā”€ā”€valā”€ā”¬ā”€toTypeName(toDecimal32OrNull(toString(-1.111), 2))ā”€ā” -ā”‚ į“ŗįµį“øį“ø ā”‚ Nullable(Decimal(9, 2)) ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ -``` - - -## toDecimal(32\|64\|128\|256)OrDefault - -Converts an input string to a [Decimal(P,S)](../data-types/decimal.md) data type value. This family of functions includes: - -- `toDecimal32OrDefault(expr, S)` ā€” Results in `Decimal32(S)` data type. -- `toDecimal64OrDefault(expr, S)` ā€” Results in `Decimal64(S)` data type. -- `toDecimal128OrDefault(expr, S)` ā€” Results in `Decimal128(S)` data type. -- `toDecimal256OrDefault(expr, S)` ā€” Results in `Decimal256(S)` data type. - -These functions should be used instead of `toDecimal*()` functions, if you prefer to get a default value instead of an exception in the event of an input value parsing error. - **Arguments** -- `expr` ā€” [Expression](../syntax.md/#syntax-expressions), returns a value in the [String](../data-types/string.md) data type. ClickHouse expects the textual representation of the decimal number. For example, `'1.111'`. -- `S` ā€” Scale, the number of decimal places in the resulting value. +- `expr` ā€” A String representation of a number. [String](../data-types/string.md). +- `S` ā€” Scale parameter between 0 and 9, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). + +Supported arguments: +- String representations of type (U)Int8/16/32/64/128/256. +- String representations of type Float32/64. + +Unsupported arguments: +- String representations of Float32/64 values `NaN` and `Inf`. +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal32OrZero('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal32`: `( -1 * 10^(9 - S), 1 * 10^(9 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an error. +::: **Returned value** -A value in the `Decimal(P,S)` data type. The value contains: - -- Number with `S` decimal places, if ClickHouse interprets the input string as a number. -- Default `Decimal(P,S)` data type value, if ClickHouse canā€™t interpret the input string as a number or if the input number contains more than `S` decimal places. - -**Examples** - -Query: - -``` sql -SELECT toDecimal32OrDefault(toString(-1.111), 5) AS val, toTypeName(val); -``` - -Result: - -```response -ā”Œā”€ā”€ā”€ā”€valā”€ā”¬ā”€toTypeName(toDecimal32OrDefault(toString(-1.111), 5))ā”€ā” -ā”‚ -1.111 ā”‚ Decimal(9, 5) ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ -``` - -Query: - -``` sql -SELECT toDecimal32OrDefault(toString(-1.111), 2) AS val, toTypeName(val); -``` - -Result: - -```response -ā”Œā”€valā”€ā”¬ā”€toTypeName(toDecimal32OrDefault(toString(-1.111), 2))ā”€ā” -ā”‚ 0 ā”‚ Decimal(9, 2) ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ -``` - -## toDecimal(32\|64\|128\|256)OrZero - -Converts an input value to the [Decimal(P,S)](../data-types/decimal.md) data type. This family of functions includes: - -- `toDecimal32OrZero( expr, S)` ā€” Results in `Decimal32(S)` data type. -- `toDecimal64OrZero( expr, S)` ā€” Results in `Decimal64(S)` data type. -- `toDecimal128OrZero( expr, S)` ā€” Results in `Decimal128(S)` data type. -- `toDecimal256OrZero( expr, S)` ā€” Results in `Decimal256(S)` data type. - -These functions should be used instead of `toDecimal*()` functions, if you prefer to get a `0` value instead of an exception in the event of an input value parsing error. - -**Arguments** - -- `expr` ā€” [Expression](../syntax.md/#syntax-expressions), returns a value in the [String](../data-types/string.md) data type. ClickHouse expects the textual representation of the decimal number. For example, `'1.111'`. -- `S` ā€” Scale, the number of decimal places in the resulting value. - -**Returned value** - -A value in the `Nullable(Decimal(P,S))` data type. The value contains: - -- Number with `S` decimal places, if ClickHouse interprets the input string as a number. -- 0 with `S` decimal places, if ClickHouse canā€™t interpret the input string as a number or if the input number contains more than `S` decimal places. +- Value of type `Decimal(9, S)` if successful, otherwise `0` with `S` decimal places. [Decimal32(S)](../data-types/decimal.md). **Example** Query: ``` sql -SELECT toDecimal32OrZero(toString(-1.111), 5) AS val, toTypeName(val); +SELECT + toDecimal32OrZero(toString(-1.111), 5) AS a, + toTypeName(a), + toDecimal32OrZero(toString('Inf'), 5) as b, + toTypeName(b) +FORMAT Vertical; ``` Result: ```response -ā”Œā”€ā”€ā”€ā”€valā”€ā”¬ā”€toTypeName(toDecimal32OrZero(toString(-1.111), 5))ā”€ā” -ā”‚ -1.111 ā”‚ Decimal(9, 5) ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: -1.111 +toTypeName(a): Decimal(9, 5) +b: 0 +toTypeName(b): Decimal(9, 5) ``` +**See also** + +- [`toDecimal32`](#todecimal32). +- [`toDecimal32OrNull`](#todecimal32ornull). +- [`toDecimal32OrDefault`](#todecimal32ordefault). + +## toDecimal32OrNull + +Like [`toDecimal32`](#todecimal32), this function converts an input value to a value of type [Nullable(Decimal(9, S))](../data-types/decimal.md) but returns `0` in case of an error. + +**Syntax** + +```sql +toDecimal32OrNull(expr, S) +``` + +**Arguments** + +- `expr` ā€” A String representation of a number. [String](../data-types/string.md). +- `S` ā€” Scale parameter between 0 and 9, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). + +Supported arguments: +- String representations of type (U)Int8/16/32/64/128/256. +- String representations of type Float32/64. + +Unsupported arguments: +- String representations of Float32/64 values `NaN` and `Inf`. +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal32OrNull('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal32`: `( -1 * 10^(9 - S), 1 * 10^(9 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an error. +::: + +**Returned value** + +- Value of type `Nullable(Decimal(9, S))` if successful, otherwise value `NULL` of the same type. [Decimal32(S)](../data-types/decimal.md). + +**Examples** + Query: ``` sql -SELECT toDecimal32OrZero(toString(-1.111), 2) AS val, toTypeName(val); +SELECT + toDecimal32OrNull(toString(-1.111), 5) AS a, + toTypeName(a), + toDecimal32OrNull(toString('Inf'), 5) as b, + toTypeName(b) +FORMAT Vertical; ``` Result: ```response -ā”Œā”€ā”€valā”€ā”¬ā”€toTypeName(toDecimal32OrZero(toString(-1.111), 2))ā”€ā” -ā”‚ 0.00 ā”‚ Decimal(9, 2) ā”‚ -ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: -1.111 +toTypeName(a): Nullable(Decimal(9, 5)) +b: į“ŗįµį“øį“ø +toTypeName(b): Nullable(Decimal(9, 5)) ``` +**See also** + +- [`toDecimal32`](#todecimal32). +- [`toDecimal32OrZero`](#todecimal32orzero). +- [`toDecimal32OrDefault`](#todecimal32ordefault). + +## toDecimal32OrDefault + +Like [`toDecimal32`](#todecimal32), this function converts an input value to a value of type [Decimal(9, S)](../data-types/decimal.md) but returns the default value in case of an error. + +**Syntax** + +```sql +toDecimal32OrDefault(expr, S[, default]) +``` + +**Arguments** + +- `expr` ā€” A String representation of a number. [String](../data-types/string.md). +- `S` ā€” Scale parameter between 0 and 9, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). +- `default` (optional) ā€” The default value to return if parsing to type `Decimal32(S)` is unsuccessful. [Decimal32(S)](../data-types/decimal.md). + +Supported arguments: +- String representations of type (U)Int8/16/32/64/128/256. +- String representations of type Float32/64. + +Unsupported arguments: +- String representations of Float32/64 values `NaN` and `Inf`. +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal32OrDefault('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal32`: `( -1 * 10^(9 - S), 1 * 10^(9 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an error. +::: + +**Returned value** + +- Value of type `Decimal(9, S)` if successful, otherwise returns the default value if passed or `0` if not. [Decimal32(S)](../data-types/decimal.md). + +**Examples** + +Query: + +``` sql +SELECT + toDecimal32OrDefault(toString(0.0001), 5) AS a, + toTypeName(a), + toDecimal32OrDefault('Inf', 0, CAST('-1', 'Decimal32(0)')) AS b, + toTypeName(b) +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: 0.0001 +toTypeName(a): Decimal(9, 5) +b: -1 +toTypeName(b): Decimal(9, 0) +``` + +**See also** + +- [`toDecimal32`](#todecimal32). +- [`toDecimal32OrZero`](#todecimal32orzero). +- [`toDecimal32OrNull`](#todecimal32ornull). + +## toDecimal64 + +Converts an input value to a value of type [`Decimal(18, S)`](../data-types/decimal.md) with scale of `S`. Throws an exception in case of an error. + +**Syntax** + +```sql +toDecimal64(expr, S) +``` + +**Arguments** + +- `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). +- `S` ā€” Scale parameter between 0 and 18, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). + +Supported arguments: +- Values or string representations of type (U)Int8/16/32/64/128/256. +- Values or string representations of type Float32/64. + +Unsupported arguments: +- Values or string representations of Float32/64 values `NaN` and `Inf` (case-insensitive). +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal64('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal64`: `( -1 * 10^(18 - S), 1 * 10^(18 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an exception. +::: + +**Returned value** + +- Value of type `Decimal(18, S)`. [Decimal64(S)](../data-types/int-uint.md). + +**Example** + +Query: + +```sql +SELECT + toDecimal64(2, 1) AS a, toTypeName(a) AS type_a, + toDecimal64(4.2, 2) AS b, toTypeName(b) AS type_b, + toDecimal64('4.2', 3) AS c, toTypeName(c) AS type_c +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: 2 +type_a: Decimal(18, 1) +b: 4.2 +type_b: Decimal(18, 2) +c: 4.2 +type_c: Decimal(18, 3) +``` + +**See also** + +- [`toDecimal64OrZero`](#todecimal64orzero). +- [`toDecimal64OrNull`](#todecimal64ornull). +- [`toDecimal64OrDefault`](#todecimal64ordefault). + +## toDecimal64OrZero + +Like [`toDecimal64`](#todecimal64), this function converts an input value to a value of type [Decimal(18, S)](../data-types/decimal.md) but returns `0` in case of an error. + +**Syntax** + +```sql +toDecimal64OrZero(expr, S) +``` + +**Arguments** + +- `expr` ā€” A String representation of a number. [String](../data-types/string.md). +- `S` ā€” Scale parameter between 0 and 18, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). + +Supported arguments: +- String representations of type (U)Int8/16/32/64/128/256. +- String representations of type Float32/64. + +Unsupported arguments: +- String representations of Float32/64 values `NaN` and `Inf`. +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal64OrZero('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal64`: `( -1 * 10^(18 - S), 1 * 10^(18 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an error. +::: + +**Returned value** + +- Value of type `Decimal(18, S)` if successful, otherwise `0` with `S` decimal places. [Decimal64(S)](../data-types/decimal.md). + +**Example** + +Query: + +``` sql +SELECT + toDecimal64OrZero(toString(0.0001), 18) AS a, + toTypeName(a), + toDecimal64OrZero(toString('Inf'), 18) as b, + toTypeName(b) +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: 0.0001 +toTypeName(a): Decimal(18, 18) +b: 0 +toTypeName(b): Decimal(18, 18) +``` + +**See also** + +- [`toDecimal64`](#todecimal64). +- [`toDecimal64OrNull`](#todecimal64ornull). +- [`toDecimal64OrDefault`](#todecimal64ordefault). + +## toDecimal64OrNull + +Like [`toDecimal64`](#todecimal64), this function converts an input value to a value of type [Nullable(Decimal(18, S))](../data-types/decimal.md) but returns `0` in case of an error. + +**Syntax** + +```sql +toDecimal64OrNull(expr, S) +``` + +**Arguments** + +- `expr` ā€” A String representation of a number. [String](../data-types/string.md). +- `S` ā€” Scale parameter between 0 and 18, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). + +Supported arguments: +- String representations of type (U)Int8/16/32/64/128/256. +- String representations of type Float32/64. + +Unsupported arguments: +- String representations of Float32/64 values `NaN` and `Inf`. +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal64OrNull('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal64`: `( -1 * 10^(18 - S), 1 * 10^(18 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an error. +::: + +**Returned value** + +- Value of type `Nullable(Decimal(18, S))` if successful, otherwise value `NULL` of the same type. [Decimal64(S)](../data-types/decimal.md). + +**Examples** + +Query: + +``` sql +SELECT + toDecimal64OrNull(toString(0.0001), 18) AS a, + toTypeName(a), + toDecimal64OrNull(toString('Inf'), 18) as b, + toTypeName(b) +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: 0.0001 +toTypeName(a): Nullable(Decimal(18, 18)) +b: į“ŗįµį“øį“ø +toTypeName(b): Nullable(Decimal(18, 18)) +``` + +**See also** + +- [`toDecimal64`](#todecimal64). +- [`toDecimal64OrZero`](#todecimal64orzero). +- [`toDecimal64OrDefault`](#todecimal64ordefault). + +## toDecimal64OrDefault + +Like [`toDecimal64`](#todecimal64), this function converts an input value to a value of type [Decimal(18, S)](../data-types/decimal.md) but returns the default value in case of an error. + +**Syntax** + +```sql +toDecimal64OrDefault(expr, S[, default]) +``` + +**Arguments** + +- `expr` ā€” A String representation of a number. [String](../data-types/string.md). +- `S` ā€” Scale parameter between 0 and 18, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). +- `default` (optional) ā€” The default value to return if parsing to type `Decimal64(S)` is unsuccessful. [Decimal64(S)](../data-types/decimal.md). + +Supported arguments: +- String representations of type (U)Int8/16/32/64/128/256. +- String representations of type Float32/64. + +Unsupported arguments: +- String representations of Float32/64 values `NaN` and `Inf`. +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal64OrDefault('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal64`: `( -1 * 10^(18 - S), 1 * 10^(18 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an error. +::: + +**Returned value** + +- Value of type `Decimal(18, S)` if successful, otherwise returns the default value if passed or `0` if not. [Decimal64(S)](../data-types/decimal.md). + +**Examples** + +Query: + +``` sql +SELECT + toDecimal64OrDefault(toString(0.0001), 18) AS a, + toTypeName(a), + toDecimal64OrDefault('Inf', 0, CAST('-1', 'Decimal64(0)')) AS b, + toTypeName(b) +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: 0.0001 +toTypeName(a): Decimal(18, 18) +b: -1 +toTypeName(b): Decimal(18, 0) +``` + +**See also** + +- [`toDecimal64`](#todecimal64). +- [`toDecimal64OrZero`](#todecimal64orzero). +- [`toDecimal64OrNull`](#todecimal64ornull). + +## toDecimal128 + +Converts an input value to a value of type [`Decimal(38, S)`](../data-types/decimal.md) with scale of `S`. Throws an exception in case of an error. + +**Syntax** + +```sql +toDecimal128(expr, S) +``` + +**Arguments** + +- `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). +- `S` ā€” Scale parameter between 0 and 38, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). + +Supported arguments: +- Values or string representations of type (U)Int8/16/32/64/128/256. +- Values or string representations of type Float32/64. + +Unsupported arguments: +- Values or string representations of Float32/64 values `NaN` and `Inf` (case-insensitive). +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal128('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal128`: `( -1 * 10^(38 - S), 1 * 10^(38 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an exception. +::: + +**Returned value** + +- Value of type `Decimal(38, S)`. [Decimal128(S)](../data-types/int-uint.md). + +**Example** + +Query: + +```sql +SELECT + toDecimal128(99, 1) AS a, toTypeName(a) AS type_a, + toDecimal128(99.67, 2) AS b, toTypeName(b) AS type_b, + toDecimal128('99.67', 3) AS c, toTypeName(c) AS type_c +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: 99 +type_a: Decimal(38, 1) +b: 99.67 +type_b: Decimal(38, 2) +c: 99.67 +type_c: Decimal(38, 3) +``` + +**See also** + +- [`toDecimal128OrZero`](#todecimal128orzero). +- [`toDecimal128OrNull`](#todecimal128ornull). +- [`toDecimal128OrDefault`](#todecimal128ordefault). + +## toDecimal128OrZero + +Like [`toDecimal128`](#todecimal128), this function converts an input value to a value of type [Decimal(38, S)](../data-types/decimal.md) but returns `0` in case of an error. + +**Syntax** + +```sql +toDecimal128OrZero(expr, S) +``` + +**Arguments** + +- `expr` ā€” A String representation of a number. [String](../data-types/string.md). +- `S` ā€” Scale parameter between 0 and 38, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). + +Supported arguments: +- String representations of type (U)Int8/16/32/64/128/256. +- String representations of type Float32/64. + +Unsupported arguments: +- String representations of Float32/64 values `NaN` and `Inf`. +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal128OrZero('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal128`: `( -1 * 10^(38 - S), 1 * 10^(38 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an error. +::: + +**Returned value** + +- Value of type `Decimal(38, S)` if successful, otherwise `0` with `S` decimal places. [Decimal128(S)](../data-types/decimal.md). + +**Example** + +Query: + +``` sql +SELECT + toDecimal128OrZero(toString(0.0001), 38) AS a, + toTypeName(a), + toDecimal128OrZero(toString('Inf'), 38) as b, + toTypeName(b) +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: 0.0001 +toTypeName(a): Decimal(38, 38) +b: 0 +toTypeName(b): Decimal(38, 38) +``` + +**See also** + +- [`toDecimal128`](#todecimal128). +- [`toDecimal128OrNull`](#todecimal128ornull). +- [`toDecimal128OrDefault`](#todecimal128ordefault). + +## toDecimal128OrNull + +Like [`toDecimal128`](#todecimal128), this function converts an input value to a value of type [Nullable(Decimal(38, S))](../data-types/decimal.md) but returns `0` in case of an error. + +**Syntax** + +```sql +toDecimal128OrNull(expr, S) +``` + +**Arguments** + +- `expr` ā€” A String representation of a number. [String](../data-types/string.md). +- `S` ā€” Scale parameter between 0 and 38, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). + +Supported arguments: +- String representations of type (U)Int8/16/32/64/128/256. +- String representations of type Float32/64. + +Unsupported arguments: +- String representations of Float32/64 values `NaN` and `Inf`. +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal128OrNull('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal128`: `( -1 * 10^(38 - S), 1 * 10^(38 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an error. +::: + +**Returned value** + +- Value of type `Nullable(Decimal(38, S))` if successful, otherwise value `NULL` of the same type. [Decimal128(S)](../data-types/decimal.md). + +**Examples** + +Query: + +``` sql +SELECT + toDecimal128OrNull(toString(1/42), 38) AS a, + toTypeName(a), + toDecimal128OrNull(toString('Inf'), 38) as b, + toTypeName(b) +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: 0.023809523809523808 +toTypeName(a): Nullable(Decimal(38, 38)) +b: į“ŗįµį“øį“ø +toTypeName(b): Nullable(Decimal(38, 38)) +``` + +**See also** + +- [`toDecimal128`](#todecimal128). +- [`toDecimal128OrZero`](#todecimal128orzero). +- [`toDecimal128OrDefault`](#todecimal128ordefault). + +## toDecimal128OrDefault + +Like [`toDecimal128`](#todecimal128), this function converts an input value to a value of type [Decimal(38, S)](../data-types/decimal.md) but returns the default value in case of an error. + +**Syntax** + +```sql +toDecimal128OrDefault(expr, S[, default]) +``` + +**Arguments** + +- `expr` ā€” A String representation of a number. [String](../data-types/string.md). +- `S` ā€” Scale parameter between 0 and 38, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). +- `default` (optional) ā€” The default value to return if parsing to type `Decimal128(S)` is unsuccessful. [Decimal128(S)](../data-types/decimal.md). + +Supported arguments: +- String representations of type (U)Int8/16/32/64/128/256. +- String representations of type Float32/64. + +Unsupported arguments: +- String representations of Float32/64 values `NaN` and `Inf`. +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal128OrDefault('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal128`: `( -1 * 10^(38 - S), 1 * 10^(38 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an error. +::: + +**Returned value** + +- Value of type `Decimal(38, S)` if successful, otherwise returns the default value if passed or `0` if not. [Decimal128(S)](../data-types/decimal.md). + +**Examples** + +Query: + +``` sql +SELECT + toDecimal128OrDefault(toString(1/42), 18) AS a, + toTypeName(a), + toDecimal128OrDefault('Inf', 0, CAST('-1', 'Decimal128(0)')) AS b, + toTypeName(b) +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: 0.023809523809523808 +toTypeName(a): Decimal(38, 18) +b: -1 +toTypeName(b): Decimal(38, 0) +``` + +**See also** + +- [`toDecimal128`](#todecimal128). +- [`toDecimal128OrZero`](#todecimal128orzero). +- [`toDecimal128OrNull`](#todecimal128ornull). + +## toDecimal256 + +Converts an input value to a value of type [`Decimal(76, S)`](../data-types/decimal.md) with scale of `S`. Throws an exception in case of an error. + +**Syntax** + +```sql +toDecimal256(expr, S) +``` + +**Arguments** + +- `expr` ā€” Expression returning a number or a string representation of a number. [Expression](../syntax.md/#syntax-expressions). +- `S` ā€” Scale parameter between 0 and 76, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). + +Supported arguments: +- Values or string representations of type (U)Int8/16/32/64/128/256. +- Values or string representations of type Float32/64. + +Unsupported arguments: +- Values or string representations of Float32/64 values `NaN` and `Inf` (case-insensitive). +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal256('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal256`: `( -1 * 10^(76 - S), 1 * 10^(76 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an exception. +::: + +**Returned value** + +- Value of type `Decimal(76, S)`. [Decimal256(S)](../data-types/int-uint.md). + +**Example** + +Query: + +```sql +SELECT + toDecimal256(99, 1) AS a, toTypeName(a) AS type_a, + toDecimal256(99.67, 2) AS b, toTypeName(b) AS type_b, + toDecimal256('99.67', 3) AS c, toTypeName(c) AS type_c +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: 99 +type_a: Decimal(76, 1) +b: 99.67 +type_b: Decimal(76, 2) +c: 99.67 +type_c: Decimal(76, 3) +``` + +**See also** + +- [`toDecimal256OrZero`](#todecimal256orzero). +- [`toDecimal256OrNull`](#todecimal256ornull). +- [`toDecimal256OrDefault`](#todecimal256ordefault). + +## toDecimal256OrZero + +Like [`toDecimal256`](#todecimal256), this function converts an input value to a value of type [Decimal(76, S)](../data-types/decimal.md) but returns `0` in case of an error. + +**Syntax** + +```sql +toDecimal256OrZero(expr, S) +``` + +**Arguments** + +- `expr` ā€” A String representation of a number. [String](../data-types/string.md). +- `S` ā€” Scale parameter between 0 and 76, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). + +Supported arguments: +- String representations of type (U)Int8/16/32/64/128/256. +- String representations of type Float32/64. + +Unsupported arguments: +- String representations of Float32/64 values `NaN` and `Inf`. +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal256OrZero('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal256`: `( -1 * 10^(76 - S), 1 * 10^(76 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an error. +::: + +**Returned value** + +- Value of type `Decimal(76, S)` if successful, otherwise `0` with `S` decimal places. [Decimal256(S)](../data-types/decimal.md). + +**Example** + +Query: + +``` sql +SELECT + toDecimal256OrZero(toString(0.0001), 76) AS a, + toTypeName(a), + toDecimal256OrZero(toString('Inf'), 76) as b, + toTypeName(b) +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: 0.0001 +toTypeName(a): Decimal(76, 76) +b: 0 +toTypeName(b): Decimal(76, 76) +``` + +**See also** + +- [`toDecimal256`](#todecimal256). +- [`toDecimal256OrNull`](#todecimal256ornull). +- [`toDecimal256OrDefault`](#todecimal256ordefault). + +## toDecimal256OrNull + +Like [`toDecimal256`](#todecimal256), this function converts an input value to a value of type [Nullable(Decimal(76, S))](../data-types/decimal.md) but returns `0` in case of an error. + +**Syntax** + +```sql +toDecimal256OrNull(expr, S) +``` + +**Arguments** + +- `expr` ā€” A String representation of a number. [String](../data-types/string.md). +- `S` ā€” Scale parameter between 0 and 76, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). + +Supported arguments: +- String representations of type (U)Int8/16/32/64/128/256. +- String representations of type Float32/64. + +Unsupported arguments: +- String representations of Float32/64 values `NaN` and `Inf`. +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal256OrNull('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal256`: `( -1 * 10^(76 - S), 1 * 10^(76 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an error. +::: + +**Returned value** + +- Value of type `Nullable(Decimal(76, S))` if successful, otherwise value `NULL` of the same type. [Decimal256(S)](../data-types/decimal.md). + +**Examples** + +Query: + +``` sql +SELECT + toDecimal256OrNull(toString(1/42), 76) AS a, + toTypeName(a), + toDecimal256OrNull(toString('Inf'), 76) as b, + toTypeName(b) +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: 0.023809523809523808 +toTypeName(a): Nullable(Decimal(76, 76)) +b: į“ŗįµį“øį“ø +toTypeName(b): Nullable(Decimal(76, 76)) +``` + +**See also** + +- [`toDecimal256`](#todecimal256). +- [`toDecimal256OrZero`](#todecimal256orzero). +- [`toDecimal256OrDefault`](#todecimal256ordefault). + +## toDecimal256OrDefault + +Like [`toDecimal256`](#todecimal256), this function converts an input value to a value of type [Decimal(76, S)](../data-types/decimal.md) but returns the default value in case of an error. + +**Syntax** + +```sql +toDecimal256OrDefault(expr, S[, default]) +``` + +**Arguments** + +- `expr` ā€” A String representation of a number. [String](../data-types/string.md). +- `S` ā€” Scale parameter between 0 and 76, specifying how many digits the fractional part of a number can have. [UInt8](../data-types/int-uint.md). +- `default` (optional) ā€” The default value to return if parsing to type `Decimal256(S)` is unsuccessful. [Decimal256(S)](../data-types/decimal.md). + +Supported arguments: +- String representations of type (U)Int8/16/32/64/128/256. +- String representations of type Float32/64. + +Unsupported arguments: +- String representations of Float32/64 values `NaN` and `Inf`. +- String representations of binary and hexadecimal values, e.g. `SELECT toDecimal256OrDefault('0xc0fe', 1);`. + +:::note +An overflow can occur if the value of `expr` exceeds the bounds of `Decimal256`: `( -1 * 10^(76 - S), 1 * 10^(76 - S) )`. +Excessive digits in a fraction are discarded (not rounded). +Excessive digits in the integer part will lead to an error. +::: + +**Returned value** + +- Value of type `Decimal(76, S)` if successful, otherwise returns the default value if passed or `0` if not. [Decimal256(S)](../data-types/decimal.md). + +**Examples** + +Query: + +``` sql +SELECT + toDecimal256OrDefault(toString(1/42), 76) AS a, + toTypeName(a), + toDecimal256OrDefault('Inf', 0, CAST('-1', 'Decimal256(0)')) AS b, + toTypeName(b) +FORMAT Vertical; +``` + +Result: + +```response +Row 1: +ā”€ā”€ā”€ā”€ā”€ā”€ +a: 0.023809523809523808 +toTypeName(a): Decimal(76, 76) +b: -1 +toTypeName(b): Decimal(76, 0) +``` + +**See also** + +- [`toDecimal256`](#todecimal256). +- [`toDecimal256OrZero`](#todecimal256orzero). +- [`toDecimal256OrNull`](#todecimal256ornull). + ## toString Functions for converting between numbers, strings (but not fixed strings), dates, and dates with times. diff --git a/docs/en/sql-reference/statements/alter/partition.md b/docs/en/sql-reference/statements/alter/partition.md index 778816f8934..1bb7817364a 100644 --- a/docs/en/sql-reference/statements/alter/partition.md +++ b/docs/en/sql-reference/statements/alter/partition.md @@ -9,6 +9,7 @@ The following operations with [partitions](/docs/en/engines/table-engines/merget - [DETACH PARTITION\|PART](#detach-partitionpart) ā€” Moves a partition or part to the `detached` directory and forget it. - [DROP PARTITION\|PART](#drop-partitionpart) ā€” Deletes a partition or part. +- [DROP DETACHED PARTITION\|PART](#drop-detached-partitionpart) - Delete a part or all parts of a partition from `detached`. - [FORGET PARTITION](#forget-partition) ā€” Deletes a partition metadata from zookeeper if it's empty. - [ATTACH PARTITION\|PART](#attach-partitionpart) ā€” Adds a partition or part from the `detached` directory to the table. - [ATTACH PARTITION FROM](#attach-partition-from) ā€” Copies the data partition from one table to another and adds. @@ -68,7 +69,7 @@ ALTER TABLE mt DROP PART 'all_4_4_0'; ## DROP DETACHED PARTITION\|PART ``` sql -ALTER TABLE table_name [ON CLUSTER cluster] DROP DETACHED PARTITION|PART partition_expr +ALTER TABLE table_name [ON CLUSTER cluster] DROP DETACHED PARTITION|PART ALL|partition_expr ``` Removes the specified part or all parts of the specified partition from `detached`. diff --git a/docs/en/sql-reference/statements/alter/statistics.md b/docs/en/sql-reference/statements/alter/statistics.md index 6880cef0e5c..7a1774a01b5 100644 --- a/docs/en/sql-reference/statements/alter/statistics.md +++ b/docs/en/sql-reference/statements/alter/statistics.md @@ -8,26 +8,28 @@ sidebar_label: STATISTICS The following operations are available: -- `ALTER TABLE [db].table ADD STATISTICS (columns list) TYPE (type list)` - Adds statistic description to tables metadata. +- `ALTER TABLE [db].table ADD STATISTICS [IF NOT EXISTS] (column list) TYPE (type list)` - Adds statistic description to tables metadata. -- `ALTER TABLE [db].table MODIFY STATISTICS (columns list) TYPE (type list)` - Modifies statistic description to tables metadata. +- `ALTER TABLE [db].table MODIFY STATISTICS (column list) TYPE (type list)` - Modifies statistic description to tables metadata. -- `ALTER TABLE [db].table DROP STATISTICS (columns list)` - Removes statistics from the metadata of the specified columns and deletes all statistics objects in all parts for the specified columns. +- `ALTER TABLE [db].table DROP STATISTICS [IF EXISTS] (column list)` - Removes statistics from the metadata of the specified columns and deletes all statistics objects in all parts for the specified columns. -- `ALTER TABLE [db].table CLEAR STATISTICS (columns list)` - Deletes all statistics objects in all parts for the specified columns. Statistics objects can be rebuild using `ALTER TABLE MATERIALIZE STATISTICS`. +- `ALTER TABLE [db].table CLEAR STATISTICS [IF EXISTS] (column list)` - Deletes all statistics objects in all parts for the specified columns. Statistics objects can be rebuild using `ALTER TABLE MATERIALIZE STATISTICS`. -- `ALTER TABLE [db.]table MATERIALIZE STATISTICS (columns list)` - Rebuilds the statistic for columns. Implemented as a [mutation](../../../sql-reference/statements/alter/index.md#mutations). +- `ALTER TABLE [db.]table MATERIALIZE STATISTICS [IF EXISTS] (column list)` - Rebuilds the statistic for columns. Implemented as a [mutation](../../../sql-reference/statements/alter/index.md#mutations). The first two commands are lightweight in a sense that they only change metadata or remove files. Also, they are replicated, syncing statistics metadata via ZooKeeper. -There is an example adding two statistics types to two columns: +## Example: + +Adding two statistics types to two columns: ``` ALTER TABLE t1 MODIFY STATISTICS c, d TYPE TDigest, Uniq; ``` :::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). +Statistic are supported only for [`*MergeTree`](../../../engines/table-engines/mergetree-family/mergetree.md) engine tables (including [replicated](../../../engines/table-engines/mergetree-family/replication.md) variants). ::: diff --git a/docs/en/sql-reference/statements/create/table.md b/docs/en/sql-reference/statements/create/table.md index 9c8984d698f..7428e6cd6ca 100644 --- a/docs/en/sql-reference/statements/create/table.md +++ b/docs/en/sql-reference/statements/create/table.md @@ -241,12 +241,12 @@ CREATE OR REPLACE TABLE test ( id UInt64, size_bytes Int64, - size String Alias formatReadableSize(size_bytes) + size String ALIAS formatReadableSize(size_bytes) ) ENGINE = MergeTree ORDER BY id; -INSERT INTO test Values (1, 4678899); +INSERT INTO test VALUES (1, 4678899); SELECT id, size_bytes, size FROM test; ā”Œā”€idā”€ā”¬ā”€size_bytesā”€ā”¬ā”€sizeā”€ā”€ā”€ā”€ā”€ā” @@ -497,7 +497,7 @@ If you perform a SELECT query mentioning a specific value in an encrypted column ```sql CREATE TABLE mytable ( - x String Codec(AES_128_GCM_SIV) + x String CODEC(AES_128_GCM_SIV) ) ENGINE = MergeTree ORDER BY x; ``` diff --git a/docs/en/sql-reference/statements/delete.md b/docs/en/sql-reference/statements/delete.md index a52b7204c30..78142f880fe 100644 --- a/docs/en/sql-reference/statements/delete.md +++ b/docs/en/sql-reference/statements/delete.md @@ -36,9 +36,9 @@ If you anticipate frequent deletes, consider using a [custom partitioning key](/ ## Limitations of lightweight `DELETE` -### Lightweight `DELETE`s do not work with projections +### Lightweight `DELETE`s with projections -Currently, `DELETE` does not work for tables with projections. This is because rows in a projection may be affected by a `DELETE` operation and may require the projection to be rebuilt, negatively affecting `DELETE` performance. +By default, `DELETE` does not work for tables with projections. This is because rows in a projection may be affected by a `DELETE` operation. But there is a [MergeTree setting](https://clickhouse.com/docs/en/operations/settings/merge-tree-settings) `lightweight_mutation_projection_mode` can change the behavior. ## Performance considerations when using lightweight `DELETE` diff --git a/docs/en/sql-reference/statements/select/join.md b/docs/en/sql-reference/statements/select/join.md index 96d9d26977d..b228f7025c4 100644 --- a/docs/en/sql-reference/statements/select/join.md +++ b/docs/en/sql-reference/statements/select/join.md @@ -186,7 +186,7 @@ Otherwise, you'll get `INVALID_JOIN_ON_EXPRESSION`. ::: -Clickhouse currently supports `ALL INNER/LEFT/RIGHT/FULL JOIN` with inequality conditions in addition to equality conditions. The inequality conditions are supported only for `hash` and `grace_hash` join algorithms. The inequality conditions are not supported with `join_use_nulls`. +Clickhouse currently supports `ALL/ANY/SEMI/ANTI INNER/LEFT/RIGHT/FULL JOIN` with inequality conditions in addition to equality conditions. The inequality conditions are supported only for `hash` and `grace_hash` join algorithms. The inequality conditions are not supported with `join_use_nulls`. **Example** diff --git a/docs/en/sql-reference/table-functions/azureBlobStorage.md b/docs/en/sql-reference/table-functions/azureBlobStorage.md index f59fedeb3a2..6936c807f96 100644 --- a/docs/en/sql-reference/table-functions/azureBlobStorage.md +++ b/docs/en/sql-reference/table-functions/azureBlobStorage.md @@ -77,3 +77,16 @@ SELECT count(*) FROM azureBlobStorage('DefaultEndpointsProtocol=https;AccountNam **See Also** - [AzureBlobStorage Table Engine](/docs/en/engines/table-engines/integrations/azureBlobStorage.md) + +## Hive-style partitioning {#hive-style-partitioning} + +When setting `use_hive_partitioning` is set to 1, ClickHouse will detect Hive-style partitioning in the path (`/name=value/`) and will allow to use partition columns as virtual columns in the query. These virtual columns will have the same names as in the partitioned path, but starting with `_`. + +**Example** + +Use virtual column, created with Hive-style partitioning + +``` sql +SET use_hive_partitioning = 1; +SELECT * from azureBlobStorage(config, storage_account_url='...', container='...', blob_path='http://data/path/date=*/country=*/code=*/*.parquet') where _date > '2020-01-01' and _country = 'Netherlands' and _code = 42; +``` diff --git a/docs/en/sql-reference/table-functions/file.md b/docs/en/sql-reference/table-functions/file.md index 44b1b50620a..3243e6cf569 100644 --- a/docs/en/sql-reference/table-functions/file.md +++ b/docs/en/sql-reference/table-functions/file.md @@ -103,7 +103,7 @@ LIMIT 2; ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ ``` -### Inserting data from a file into a table: +### Inserting data from a file into a table ``` sql INSERT INTO FUNCTION @@ -206,6 +206,19 @@ SELECT count(*) FROM file('big_dir/**/file002', 'CSV', 'name String, value UInt3 - `_size` ā€” Size of the file in bytes. Type: `Nullable(UInt64)`. If the file size is unknown, the value is `NULL`. - `_time` ā€” Last modified time of the file. Type: `Nullable(DateTime)`. If the time is unknown, the value is `NULL`. +## Hive-style partitioning {#hive-style-partitioning} + +When setting `use_hive_partitioning` is set to 1, ClickHouse will detect Hive-style partitioning in the path (`/name=value/`) and will allow to use partition columns as virtual columns in the query. These virtual columns will have the same names as in the partitioned path, but starting with `_`. + +**Example** + +Use virtual column, created with Hive-style partitioning + +``` sql +SET use_hive_partitioning = 1; +SELECT * from file('data/path/date=*/country=*/code=*/*.parquet') where _date > '2020-01-01' and _country = 'Netherlands' and _code = 42; +``` + ## Settings {#settings} - [engine_file_empty_if_not_exists](/docs/en/operations/settings/settings.md#engine-file-empty_if-not-exists) - allows to select empty data from a file that doesn't exist. Disabled by default. diff --git a/docs/en/sql-reference/table-functions/fuzzQuery.md b/docs/en/sql-reference/table-functions/fuzzQuery.md new file mode 100644 index 00000000000..e15f8a40156 --- /dev/null +++ b/docs/en/sql-reference/table-functions/fuzzQuery.md @@ -0,0 +1,36 @@ +--- +slug: /en/sql-reference/table-functions/fuzzQuery +sidebar_position: 75 +sidebar_label: fuzzQuery +--- + +# fuzzQuery + +Perturbs the given query string with random variations. + +``` sql +fuzzQuery(query[, max_query_length[, random_seed]]) +``` + +**Arguments** + +- `query` (String) - The source query to perform the fuzzing on. +- `max_query_length` (UInt64) - A maximum length the query can get during the fuzzing process. +- `random_seed` (UInt64) - A random seed for producing stable results. + +**Returned Value** + +A table object with a single column containing perturbed query strings. + +## Usage Example + +``` sql +SELECT * FROM fuzzQuery('SELECT materialize(\'a\' AS key) GROUP BY key') LIMIT 2; +``` + +``` + ā”Œā”€queryā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā” +1. ā”‚ SELECT 'a' AS key GROUP BY key ā”‚ +2. ā”‚ EXPLAIN PIPELINE compact = true SELECT 'a' AS key GROUP BY key ā”‚ + ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +``` diff --git a/docs/en/sql-reference/table-functions/hdfs.md b/docs/en/sql-reference/table-functions/hdfs.md index 28cba5ccc6a..f96e48d914d 100644 --- a/docs/en/sql-reference/table-functions/hdfs.md +++ b/docs/en/sql-reference/table-functions/hdfs.md @@ -44,6 +44,7 @@ LIMIT 2 Paths may use globbing. Files must match the whole path pattern, not only the suffix or prefix. - `*` ā€” Represents arbitrarily many characters except `/` but including the empty string. +- `**` ā€” Represents all files inside a folder recursively. - `?` ā€” Represents an arbitrary single character. - `{some_string,another_string,yet_another_one}` ā€” Substitutes any of strings `'some_string', 'another_string', 'yet_another_one'`. The strings can contain the `/` symbol. - `{N..M}` ā€” Represents any number `>= N` and `<= M`. @@ -99,6 +100,19 @@ FROM hdfs('hdfs://hdfs1:9000/big_dir/file{0..9}{0..9}{0..9}', 'CSV', 'name Strin - `_size` ā€” Size of the file in bytes. Type: `Nullable(UInt64)`. If the size is unknown, the value is `NULL`. - `_time` ā€” Last modified time of the file. Type: `Nullable(DateTime)`. If the time is unknown, the value is `NULL`. +## Hive-style partitioning {#hive-style-partitioning} + +When setting `use_hive_partitioning` is set to 1, ClickHouse will detect Hive-style partitioning in the path (`/name=value/`) and will allow to use partition columns as virtual columns in the query. These virtual columns will have the same names as in the partitioned path, but starting with `_`. + +**Example** + +Use virtual column, created with Hive-style partitioning + +``` sql +SET use_hive_partitioning = 1; +SELECT * from HDFS('hdfs://hdfs1:9000/data/path/date=*/country=*/code=*/*.parquet') where _date > '2020-01-01' and _country = 'Netherlands' and _code = 42; +``` + ## Storage Settings {#storage-settings} - [hdfs_truncate_on_insert](/docs/en/operations/settings/settings.md#hdfs_truncate_on_insert) - allows to truncate file before insert into it. Disabled by default. diff --git a/docs/en/sql-reference/table-functions/s3.md b/docs/en/sql-reference/table-functions/s3.md index 35e5d86034c..1bd9f38517e 100644 --- a/docs/en/sql-reference/table-functions/s3.md +++ b/docs/en/sql-reference/table-functions/s3.md @@ -274,6 +274,19 @@ FROM s3( - `_size` ā€” 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` ā€” Last modified time of the file. Type: `Nullable(DateTime)`. If the time is unknown, the value is `NULL`. +## Hive-style partitioning {#hive-style-partitioning} + +When setting `use_hive_partitioning` is set to 1, ClickHouse will detect Hive-style partitioning in the path (`/name=value/`) and will allow to use partition columns as virtual columns in the query. These virtual columns will have the same names as in the partitioned path, but starting with `_`. + +**Example** + +Use virtual column, created with Hive-style partitioning + +``` sql +SET use_hive_partitioning = 1; +SELECT * from s3('s3://data/path/date=*/country=*/code=*/*.parquet') where _date > '2020-01-01' and _country = 'Netherlands' and _code = 42; +``` + ## Storage Settings {#storage-settings} - [s3_truncate_on_insert](/docs/en/operations/settings/settings.md#s3_truncate_on_insert) - allows to truncate file before insert into it. Disabled by default. diff --git a/docs/en/sql-reference/table-functions/timeSeriesData.md b/docs/en/sql-reference/table-functions/timeSeriesData.md new file mode 100644 index 00000000000..aa7a9d30c2a --- /dev/null +++ b/docs/en/sql-reference/table-functions/timeSeriesData.md @@ -0,0 +1,28 @@ +--- +slug: /en/sql-reference/table-functions/timeSeriesData +sidebar_position: 145 +sidebar_label: timeSeriesData +--- + +# timeSeriesData + +`timeSeriesData(db_name.time_series_table)` - Returns the [data](../../engines/table-engines/integrations/time-series.md#data-table) table +used by table `db_name.time_series_table` which table engine is [TimeSeries](../../engines/table-engines/integrations/time-series.md): + +``` sql +CREATE TABLE db_name.time_series_table ENGINE=TimeSeries DATA data_table +``` + +The function also works if the _data_ table is inner: + +``` sql +CREATE TABLE db_name.time_series_table ENGINE=TimeSeries DATA INNER UUID '01234567-89ab-cdef-0123-456789abcdef' +``` + +The following queries are equivalent: + +``` sql +SELECT * FROM timeSeriesData(db_name.time_series_table); +SELECT * FROM timeSeriesData('db_name.time_series_table'); +SELECT * FROM timeSeriesData('db_name', 'time_series_table'); +``` diff --git a/docs/en/sql-reference/table-functions/timeSeriesMetrics.md b/docs/en/sql-reference/table-functions/timeSeriesMetrics.md new file mode 100644 index 00000000000..913f1185bca --- /dev/null +++ b/docs/en/sql-reference/table-functions/timeSeriesMetrics.md @@ -0,0 +1,28 @@ +--- +slug: /en/sql-reference/table-functions/timeSeriesMetrics +sidebar_position: 145 +sidebar_label: timeSeriesMetrics +--- + +# timeSeriesMetrics + +`timeSeriesMetrics(db_name.time_series_table)` - Returns the [metrics](../../engines/table-engines/integrations/time-series.md#metrics-table) table +used by table `db_name.time_series_table` which table engine is [TimeSeries](../../engines/table-engines/integrations/time-series.md): + +``` sql +CREATE TABLE db_name.time_series_table ENGINE=TimeSeries METRICS metrics_table +``` + +The function also works if the _metrics_ table is inner: + +``` sql +CREATE TABLE db_name.time_series_table ENGINE=TimeSeries METRICS INNER UUID '01234567-89ab-cdef-0123-456789abcdef' +``` + +The following queries are equivalent: + +``` sql +SELECT * FROM timeSeriesMetrics(db_name.time_series_table); +SELECT * FROM timeSeriesMetrics('db_name.time_series_table'); +SELECT * FROM timeSeriesMetrics('db_name', 'time_series_table'); +``` diff --git a/docs/en/sql-reference/table-functions/timeSeriesTags.md b/docs/en/sql-reference/table-functions/timeSeriesTags.md new file mode 100644 index 00000000000..663a7dc6ac8 --- /dev/null +++ b/docs/en/sql-reference/table-functions/timeSeriesTags.md @@ -0,0 +1,28 @@ +--- +slug: /en/sql-reference/table-functions/timeSeriesTags +sidebar_position: 145 +sidebar_label: timeSeriesTags +--- + +# timeSeriesTags + +`timeSeriesTags(db_name.time_series_table)` - Returns the [tags](../../engines/table-engines/integrations/time-series.md#tags-table) table +used by table `db_name.time_series_table` which table engine is [TimeSeries](../../engines/table-engines/integrations/time-series.md): + +``` sql +CREATE TABLE db_name.time_series_table ENGINE=TimeSeries TAGS tags_table +``` + +The function also works if the _tags_ table is inner: + +``` sql +CREATE TABLE db_name.time_series_table ENGINE=TimeSeries TAGS INNER UUID '01234567-89ab-cdef-0123-456789abcdef' +``` + +The following queries are equivalent: + +``` sql +SELECT * FROM timeSeriesTags(db_name.time_series_table); +SELECT * FROM timeSeriesTags('db_name.time_series_table'); +SELECT * FROM timeSeriesTags('db_name', 'time_series_table'); +``` diff --git a/docs/en/sql-reference/table-functions/url.md b/docs/en/sql-reference/table-functions/url.md index 3bb7aff53a7..b4027594e7c 100644 --- a/docs/en/sql-reference/table-functions/url.md +++ b/docs/en/sql-reference/table-functions/url.md @@ -55,6 +55,19 @@ Character `|` inside patterns is used to specify failover addresses. They are it - `_size` ā€” Size of the resource in bytes. Type: `Nullable(UInt64)`. If the size is unknown, the value is `NULL`. - `_time` ā€” Last modified time of the file. Type: `Nullable(DateTime)`. If the time is unknown, the value is `NULL`. +## Hive-style partitioning {#hive-style-partitioning} + +When setting `use_hive_partitioning` is set to 1, ClickHouse will detect Hive-style partitioning in the path (`/name=value/`) and will allow to use partition columns as virtual columns in the query. These virtual columns will have the same names as in the partitioned path, but starting with `_`. + +**Example** + +Use virtual column, created with Hive-style partitioning + +``` sql +SET use_hive_partitioning = 1; +SELECT * from url('http://data/path/date=*/country=*/code=*/*.parquet') where _date > '2020-01-01' and _country = 'Netherlands' and _code = 42; +``` + ## Storage Settings {#storage-settings} - [engine_url_skip_empty_files](/docs/en/operations/settings/settings.md#engine_url_skip_empty_files) - allows to skip empty files while reading. Disabled by default. diff --git a/docs/en/sql-reference/window-functions/lagInFrame.md b/docs/en/sql-reference/window-functions/lagInFrame.md index de6e9005baa..01bf809e76e 100644 --- a/docs/en/sql-reference/window-functions/lagInFrame.md +++ b/docs/en/sql-reference/window-functions/lagInFrame.md @@ -1,7 +1,7 @@ --- slug: /en/sql-reference/window-functions/lagInFrame sidebar_label: lagInFrame -sidebar_position: 8 +sidebar_position: 9 --- # lagInFrame diff --git a/docs/en/sql-reference/window-functions/leadInFrame.md b/docs/en/sql-reference/window-functions/leadInFrame.md index 4a82c03f6e6..dae4353b582 100644 --- a/docs/en/sql-reference/window-functions/leadInFrame.md +++ b/docs/en/sql-reference/window-functions/leadInFrame.md @@ -1,7 +1,7 @@ --- slug: /en/sql-reference/window-functions/leadInFrame sidebar_label: leadInFrame -sidebar_position: 9 +sidebar_position: 10 --- # leadInFrame diff --git a/docs/en/sql-reference/window-functions/percent_rank.md b/docs/en/sql-reference/window-functions/percent_rank.md new file mode 100644 index 00000000000..2e348f2a333 --- /dev/null +++ b/docs/en/sql-reference/window-functions/percent_rank.md @@ -0,0 +1,72 @@ +--- +slug: /en/sql-reference/window-functions/percent_rank +sidebar_label: percent_rank +sidebar_position: 8 +--- + +# percent_rank + +returns the relative rank (i.e. percentile) of rows within a window partition. + +**Syntax** + +Alias: `percentRank` (case-sensitive) + +```sql +percent_rank (column_name) + OVER ([[PARTITION BY grouping_column] [ORDER BY sorting_column] + [RANGE BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING]] | [window_name]) +FROM table_name +WINDOW window_name as ([PARTITION BY grouping_column] [ORDER BY sorting_column] RANGE BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) +``` + +The default and required window frame definition is `RANGE BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING`. + +For more detail on window function syntax see: [Window Functions - Syntax](./index.md/#syntax). + +**Example** + + +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, + percent_rank() OVER (ORDER BY salary DESC) AS percent_rank +FROM salaries; +``` + +Result: + +```response + + ā”Œā”€playerā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€salaryā”€ā”¬ā”€ā”€ā”€ā”€ā”€ā”€ā”€percent_rankā”€ā” +1. ā”‚ Gary Chen ā”‚ 195000 ā”‚ 0 ā”‚ +2. ā”‚ Robert George ā”‚ 195000 ā”‚ 0 ā”‚ +3. ā”‚ Charles Juarez ā”‚ 190000 ā”‚ 0.3333333333333333 ā”‚ +4. ā”‚ Michael Stanley ā”‚ 150000 ā”‚ 0.5 ā”‚ +5. ā”‚ Scott Harrison ā”‚ 150000 ā”‚ 0.5 ā”‚ +6. ā”‚ Douglas Benson ā”‚ 150000 ā”‚ 0.5 ā”‚ +7. ā”‚ James Henderson ā”‚ 140000 ā”‚ 1 ā”‚ + ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ + +``` diff --git a/docs/ru/development/architecture.md b/docs/ru/development/architecture.md index 575799cccc4..d2afbf233b8 100644 --- a/docs/ru/development/architecture.md +++ b/docs/ru/development/architecture.md @@ -115,7 +115,7 @@ ClickHouse ā€” ŠæŠ¾Š»Š½Š¾Ń†ŠµŠ½Š½Š°Ń стŠ¾Š»Š±Ń†Š¾Š²Š°Ń Š”Š£Š‘Š”. Š”Š°Š½Š½Ń‹ `InterpreterSelectQuery` ŠøсŠæŠ¾Š»ŃŒŠ·ŃƒŠµŃ‚ `ExpressionAnalyzer` Šø `ExpressionActions` Š¼ŠµŃ…Š°Š½ŠøŠ·Š¼Ń‹ Š“Š»Ń Š°Š½Š°Š»ŠøŠ·Š° Š·Š°ŠæрŠ¾ŃŠ¾Š² Šø ŠæрŠµŠ¾Š±Ń€Š°Š·Š¾Š²Š°Š½ŠøŠ¹. Š˜Š¼ŠµŠ½Š½Š¾ Š·Š“ŠµŃŃŒ Š²Ń‹ŠæŠ¾Š»Š½ŃŠµŃ‚ся Š±Š¾Š»ŃŒŃˆŠøŠ½ŃŃ‚Š²Š¾ Š¾ŠæтŠøŠ¼ŠøŠ·Š°Ń†ŠøŠ¹ Š·Š°ŠæрŠ¾ŃŠ¾Š² Š½Š° Š¾ŃŠ½Š¾Š²Šµ ŠæрŠ°Š²ŠøŠ». `ExpressionAnalyzer` Š½Š°ŠæŠøсŠ°Š½ Š“Š¾Š²Š¾Š»ŃŒŠ½Š¾ Š³Ń€ŃŠ·Š½Š¾ Šø Š“Š¾Š»Š¶ŠµŠ½ Š±Ń‹Ń‚ŃŒ ŠæŠµŃ€ŠµŠæŠøсŠ°Š½: рŠ°Š·Š»ŠøчŠ½Ń‹Šµ ŠæрŠµŠ¾Š±Ń€Š°Š·Š¾Š²Š°Š½Šøя Š·Š°ŠæрŠ¾ŃŠ¾Š² Šø Š¾ŠæтŠøŠ¼ŠøŠ·Š°Ń†ŠøŠø Š“Š¾Š»Š¶Š½Ń‹ Š±Ń‹Ń‚ŃŒ ŠøŠ·Š²Š»ŠµŃ‡ŠµŠ½Ń‹ Š² Š¾Ń‚Š“ŠµŠ»ŃŒŠ½Ń‹Šµ ŠŗŠ»Š°ŃŃŃ‹, чтŠ¾Š±Ń‹ ŠæŠ¾Š·Š²Š¾Š»Šøть Š¼Š¾Š“уŠ»ŃŒŠ½Ń‹Šµ ŠæрŠµŠ¾Š±Ń€Š°Š·Š¾Š²Š°Š½Šøя ŠøŠ»Šø Š·Š°ŠæрŠ¾ŃŃ‹. -Š”Š»Ń рŠµŃˆŠµŠ½Šøя тŠµŠŗущŠøх ŠæрŠ¾Š±Š»ŠµŠ¼, сущŠµŃŃ‚Š²ŃƒŃŽŃ‰Šøх Š² ŠøŠ½Ń‚ŠµŃ€ŠæрŠµŃ‚Š°Ń‚Š¾Ń€Š°Ń…, рŠ°Š·Ń€Š°Š±Š°Ń‚Ń‹Š²Š°ŠµŃ‚ся Š½Š¾Š²Ń‹Š¹ `InterpreterSelectQueryAnalyzer`. Š­Ń‚Š¾ Š½Š¾Š²Š°Ń Š²ŠµŃ€ŃŠøя `InterpreterSelectQuery`, ŠŗŠ¾Ń‚Š¾Ń€Š°Ń Š½Šµ ŠøсŠæŠ¾Š»ŃŒŠ·ŃƒŠµŃ‚ `ExpressionAnalyzer` Šø Š²Š²Š¾Š“Šøт Š“Š¾ŠæŠ¾Š»Š½ŠøтŠµŠ»ŃŒŠ½Ń‹Š¹ урŠ¾Š²ŠµŠ½ŃŒ Š°Š±ŃŃ‚Ń€Š°ŠŗцŠøŠø Š¼ŠµŠ¶Š“у `AST` Šø `QueryPipeline`, Š½Š°Š·Ń‹Š²Š°ŠµŠ¼Ń‹Š¹ `QueryTree`. ŠžŠ½ ŠµŃ‰Šµ Š½Šµ Š³Š¾Ń‚Š¾Š² Šŗ ŠøсŠæŠ¾Š»ŃŒŠ·Š¾Š²Š°Š½Šøю Š² ŠæрŠ¾Š“Š°ŠŗшŠµŠ½Šµ, Š½Š¾ ŠµŠ³Š¾ Š¼Š¾Š¶Š½Š¾ ŠæрŠ¾Ń‚ŠµŃŃ‚ŠøрŠ¾Š²Š°Ń‚ŃŒ с ŠæŠ¾Š¼Š¾Ń‰ŃŒŃŽ фŠ»Š°Š³Š° `allow_experimental_analyzer`. +Š”Š»Ń рŠµŃˆŠµŠ½Šøя ŠæрŠ¾Š±Š»ŠµŠ¼, сущŠµŃŃ‚Š²ŃƒŃŽŃ‰Šøх Š² ŠøŠ½Ń‚ŠµŃ€ŠæрŠµŃ‚Š°Ń‚Š¾Ń€Š°Ń…, Š±Ń‹Š» рŠ°Š·Ń€Š°Š±Š¾Ń‚Š°Š½ Š½Š¾Š²Ń‹Š¹ `InterpreterSelectQueryAnalyzer`. Š­Ń‚Š¾ Š½Š¾Š²Š°Ń Š²ŠµŃ€ŃŠøя `InterpreterSelectQuery`, ŠŗŠ¾Ń‚Š¾Ń€Š°Ń Š½Šµ ŠøсŠæŠ¾Š»ŃŒŠ·ŃƒŠµŃ‚ `ExpressionAnalyzer` Šø Š²Š²Š¾Š“Šøт Š“Š¾ŠæŠ¾Š»Š½ŠøтŠµŠ»ŃŒŠ½Ń‹Š¹ урŠ¾Š²ŠµŠ½ŃŒ Š°Š±ŃŃ‚Ń€Š°ŠŗцŠøŠø Š¼ŠµŠ¶Š“у `AST` Šø `QueryPipeline`, Š½Š°Š·Ń‹Š²Š°ŠµŠ¼Ń‹Š¹ `QueryTree`. ŠžŠ½ ŠæŠ¾Š»Š½Š¾ŃŃ‚ŃŒŃŽ Š³Š¾Ń‚Š¾Š² Šŗ ŠøсŠæŠ¾Š»ŃŒŠ·Š¾Š²Š°Š½Šøю Š² ŠæрŠ¾Š“Š°ŠŗшŠµŠ½Šµ, Š½Š¾ Š½Š° Š²ŃŃŠŗŠøŠ¹ сŠ»ŃƒŃ‡Š°Š¹ ŠµŠ³Š¾ Š¼Š¾Š¶Š½Š¾ Š²Ń‹ŠŗŠ»ŃŽŃ‡Šøть, устŠ°Š½Š¾Š²ŠøŠ² Š·Š½Š°Ń‡ŠµŠ½ŠøŠµ Š½Š°ŃŃ‚Ń€Š¾Š¹ŠŗŠø `enable_analyzer` Š² `false`. ## Š¤ŃƒŠ½ŠŗцŠøŠø {#functions} diff --git a/docs/ru/sql-reference/table-functions/file.md b/docs/ru/sql-reference/table-functions/file.md index 546a674d41a..f3e8b0f46b7 100644 --- a/docs/ru/sql-reference/table-functions/file.md +++ b/docs/ru/sql-reference/table-functions/file.md @@ -81,6 +81,7 @@ SELECT * FROM file('test.csv', 'CSV', 'column1 UInt32, column2 UInt32, column3 U ŠžŠ±Ń€Š°Š±Š°Ń‚Ń‹Š²Š°Ń‚ŃŒŃŃ Š±ŃƒŠ“ут тŠµ Šø тŠ¾Š»ŃŒŠŗŠ¾ тŠµ фŠ°Š¹Š»Ń‹, ŠŗŠ¾Ń‚Š¾Ń€Ń‹Šµ сущŠµŃŃ‚Š²ŃƒŃŽŃ‚ Š² фŠ°Š¹Š»Š¾Š²Š¾Š¹ сŠøстŠµŠ¼Šµ Šø уŠ“Š¾Š²Š»ŠµŃ‚Š²Š¾Ń€ŃŃŽŃ‚ Š²ŃŠµŠ¼Ńƒ шŠ°Š±Š»Š¾Š½Ńƒ ŠæутŠø. - `*` ā€” Š·Š°Š¼ŠµŠ½ŃŠµŃ‚ Š»ŃŽŠ±Š¾Šµ ŠŗŠ¾Š»ŠøчŠµŃŃ‚Š²Š¾ Š»ŃŽŠ±Ń‹Ń… сŠøŠ¼Š²Š¾Š»Š¾Š² ŠŗрŠ¾Š¼Šµ `/`, Š²ŠŗŠ»ŃŽŃ‡Š°Ń Š¾Ń‚ŃŃƒŃ‚ŃŃ‚Š²ŠøŠµ сŠøŠ¼Š²Š¾Š»Š¾Š². +- `**` ā€” Š—Š°Š¼ŠµŠ½ŃŠµŃ‚ Š»ŃŽŠ±Š¾Šµ ŠŗŠ¾Š»ŠøчŠµŃŃ‚Š²Š¾ Š»ŃŽŠ±Ń‹Ń… сŠøŠ¼Š²Š¾Š»Š¾Š², Š²ŠŗŠ»ŃŽŃ‡Š°Ń `/`, тŠ¾ ŠµŃŃ‚ŃŒ Š¾ŃŃƒŃ‰ŠµŃŃ‚Š²Š»ŃŠµŃ‚ рŠµŠŗурсŠøŠ²Š½Ń‹Š¹ ŠæŠ¾ŠøсŠŗ ŠæŠ¾ Š²Š»Š¾Š¶ŠµŠ½Š½Ń‹Š¼ Š“ŠøрŠµŠŗтŠ¾Ń€ŠøяŠ¼. - `?` ā€” Š·Š°Š¼ŠµŠ½ŃŠµŃ‚ рŠ¾Š²Š½Š¾ Š¾Š“ŠøŠ½ Š»ŃŽŠ±Š¾Š¹ сŠøŠ¼Š²Š¾Š». - `{some_string,another_string,yet_another_one}` ā€” Š·Š°Š¼ŠµŠ½ŃŠµŃ‚ Š»ŃŽŠ±ŃƒŃŽ ŠøŠ· стрŠ¾Šŗ `'some_string', 'another_string', 'yet_another_one'`. Š­Ń‚Šø стрŠ¾ŠŗŠø тŠ°ŠŗŠ¶Šµ Š¼Š¾Š³ŃƒŃ‚ сŠ¾Š“ŠµŃ€Š¶Š°Ń‚ŃŒ сŠøŠ¼Š²Š¾Š» `/`. - `{N..M}` ā€” Š·Š°Š¼ŠµŠ½ŃŠµŃ‚ Š»ŃŽŠ±Š¾Šµ чŠøсŠ»Š¾ Š² ŠøŠ½Ń‚ŠµŃ€Š²Š°Š»Šµ Š¾Ń‚ `N` Š“Š¾ `M` Š²ŠŗŠ»ŃŽŃ‡ŠøтŠµŠ»ŃŒŠ½Š¾ (Š¼Š¾Š¶ŠµŃ‚ сŠ¾Š“ŠµŃ€Š¶Š°Ń‚ŃŒ Š²ŠµŠ“ущŠøŠµ Š½ŃƒŠ»Šø). diff --git a/docs/ru/sql-reference/table-functions/hdfs.md b/docs/ru/sql-reference/table-functions/hdfs.md index 6dcb1a21791..13f1bdc43af 100644 --- a/docs/ru/sql-reference/table-functions/hdfs.md +++ b/docs/ru/sql-reference/table-functions/hdfs.md @@ -47,6 +47,7 @@ LIMIT 2 - `*` ā€” Š—Š°Š¼ŠµŠ½ŃŠµŃ‚ Š»ŃŽŠ±Š¾Šµ ŠŗŠ¾Š»ŠøчŠµŃŃ‚Š²Š¾ Š»ŃŽŠ±Ń‹Ń… сŠøŠ¼Š²Š¾Š»Š¾Š² (ŠŗрŠ¾Š¼Šµ `/`), Š²ŠŗŠ»ŃŽŃ‡Š°Ń Š¾Ń‚ŃŃƒŃ‚ŃŃ‚Š²ŠøŠµ сŠøŠ¼Š²Š¾Š»Š¾Š². +- `**` ā€” Š—Š°Š¼ŠµŠ½ŃŠµŃ‚ Š»ŃŽŠ±Š¾Šµ ŠŗŠ¾Š»ŠøчŠµŃŃ‚Š²Š¾ Š»ŃŽŠ±Ń‹Ń… сŠøŠ¼Š²Š¾Š»Š¾Š², Š²ŠŗŠ»ŃŽŃ‡Š°Ń `/`, тŠ¾ ŠµŃŃ‚ŃŒ Š¾ŃŃƒŃ‰ŠµŃŃ‚Š²Š»ŃŠµŃ‚ рŠµŠŗурсŠøŠ²Š½Ń‹Š¹ ŠæŠ¾ŠøсŠŗ ŠæŠ¾ Š²Š»Š¾Š¶ŠµŠ½Š½Ń‹Š¼ Š“ŠøрŠµŠŗтŠ¾Ń€ŠøяŠ¼. - `?` ā€” Š—Š°Š¼ŠµŠ½ŃŠµŃ‚ рŠ¾Š²Š½Š¾ Š¾Š“ŠøŠ½ Š»ŃŽŠ±Š¾Š¹ сŠøŠ¼Š²Š¾Š». - `{some_string,another_string,yet_another_one}` ā€” Š—Š°Š¼ŠµŠ½ŃŠµŃ‚ Š»ŃŽŠ±ŃƒŃŽ ŠøŠ· стрŠ¾Šŗ `'some_string', 'another_string', 'yet_another_one'`. Š­Ń‚Šø стрŠ¾ŠŗŠø тŠ°ŠŗŠ¶Šµ Š¼Š¾Š³ŃƒŃ‚ сŠ¾Š“ŠµŃ€Š¶Š°Ń‚ŃŒ сŠøŠ¼Š²Š¾Š» `/`. - `{N..M}` ā€” Š—Š°Š¼ŠµŠ½ŃŠµŃ‚ Š»ŃŽŠ±Š¾Šµ чŠøсŠ»Š¾ Š² ŠøŠ½Ń‚ŠµŃ€Š²Š°Š»Šµ Š¾Ń‚ `N` Š“Š¾ `M` Š²ŠŗŠ»ŃŽŃ‡ŠøтŠµŠ»ŃŒŠ½Š¾ (Š¼Š¾Š¶ŠµŃ‚ сŠ¾Š“ŠµŃ€Š¶Š°Ń‚ŃŒ Š²ŠµŠ“ущŠøŠµ Š½ŃƒŠ»Šø). diff --git a/docs/zh/operations/external-authenticators/kerberos.md b/docs/zh/operations/external-authenticators/kerberos.md index 649a0b9bd48..d1a39bbc952 100644 --- a/docs/zh/operations/external-authenticators/kerberos.md +++ b/docs/zh/operations/external-authenticators/kerberos.md @@ -23,30 +23,30 @@ slug: /zh/operations/external-authenticators/kerberos ē¤ŗ例 (čæ›å…„ `config.xml`): ```xml - + - + ``` äø»ä½“č§„čŒƒ: ```xml - + HTTP/clickhouse.example.com@EXAMPLE.COM - + ``` ęŒ‰é¢†åŸŸčæ‡ę»¤: ```xml - + EXAMPLE.COM - + ``` !!! warning "ę³Øꄏ" @@ -74,7 +74,7 @@ Kerberosäø»ä½“名ē§°ę ¼å¼é€šåøø遵å¾Ŗ仄äø‹ęؔ式: ē¤ŗ例 (čæ›å…„ `users.xml`): ``` - + @@ -85,7 +85,7 @@ Kerberosäø»ä½“名ē§°ę ¼å¼é€šåøø遵å¾Ŗ仄äø‹ęؔ式: - + ``` !!! warning "č­¦å‘Š" diff --git a/programs/CMakeLists.txt b/programs/CMakeLists.txt index ce3a4659e0e..3add371b30f 100644 --- a/programs/CMakeLists.txt +++ b/programs/CMakeLists.txt @@ -1,4 +1,4 @@ -add_compile_options($<$,$>:${COVERAGE_FLAGS}>) +add_compile_options("$<$,$>:${COVERAGE_FLAGS}>") if (USE_CLANG_TIDY) set (CMAKE_CXX_CLANG_TIDY "${CLANG_TIDY_PATH}") diff --git a/programs/benchmark/Benchmark.cpp b/programs/benchmark/Benchmark.cpp index 48dca82eb2b..36f774a3c12 100644 --- a/programs/benchmark/Benchmark.cpp +++ b/programs/benchmark/Benchmark.cpp @@ -75,6 +75,8 @@ public: const String & default_database_, const String & user_, const String & password_, + const String & proto_send_chunked_, + const String & proto_recv_chunked_, const String & quota_key_, const String & stage, bool randomize_, @@ -128,7 +130,9 @@ public: connections.emplace_back(std::make_unique( concurrency, cur_host, cur_port, - default_database_, user_, password_, quota_key_, + default_database_, user_, password_, + proto_send_chunked_, proto_recv_chunked_, + quota_key_, /* cluster_= */ "", /* cluster_secret_= */ "", /* client_name_= */ std::string(DEFAULT_CLIENT_NAME), @@ -662,6 +666,50 @@ int mainEntryClickHouseBenchmark(int argc, char ** argv) Strings hosts = options.count("host") ? options["host"].as() : Strings({"localhost"}); + String proto_send_chunked {"notchunked"}; + String proto_recv_chunked {"notchunked"}; + + if (options.count("proto_caps")) + { + std::string proto_caps_str = options["proto_caps"].as(); + + std::vector proto_caps; + splitInto<','>(proto_caps, proto_caps_str); + + for (auto cap_str : proto_caps) + { + std::string direction; + + if (cap_str.starts_with("send_")) + { + direction = "send"; + cap_str = cap_str.substr(std::string_view("send_").size()); + } + else if (cap_str.starts_with("recv_")) + { + direction = "recv"; + cap_str = cap_str.substr(std::string_view("recv_").size()); + } + + if (cap_str != "chunked" && cap_str != "notchunked" && cap_str != "chunked_optional" && cap_str != "notchunked_optional") + throw Exception(ErrorCodes::BAD_ARGUMENTS, "proto_caps option is incorrect ({})", proto_caps_str); + + if (direction.empty()) + { + proto_send_chunked = cap_str; + proto_recv_chunked = cap_str; + } + else + { + if (direction == "send") + proto_send_chunked = cap_str; + else + proto_recv_chunked = cap_str; + } + } + } + + Benchmark benchmark( options["concurrency"].as(), options["delay"].as(), @@ -673,6 +721,8 @@ int mainEntryClickHouseBenchmark(int argc, char ** argv) options["database"].as(), options["user"].as(), options["password"].as(), + proto_send_chunked, + proto_recv_chunked, options["quota_key"].as(), options["stage"].as(), options.count("randomize"), diff --git a/programs/client/Client.cpp b/programs/client/Client.cpp index 1d99d223ee9..25c94c56aa6 100644 --- a/programs/client/Client.cpp +++ b/programs/client/Client.cpp @@ -223,7 +223,7 @@ std::vector Client::loadWarningMessages() size_t rows = packet.block.rows(); for (size_t i = 0; i < rows; ++i) - messages.emplace_back(column[i].get()); + messages.emplace_back(column[i].safeGet()); } continue; diff --git a/programs/client/Client.h b/programs/client/Client.h index 7fdf77031ab..07a8e293b1a 100644 --- a/programs/client/Client.h +++ b/programs/client/Client.h @@ -11,7 +11,10 @@ class Client : public ClientApplicationBase public: using Arguments = ClientApplicationBase::Arguments; - Client() = default; + Client() + { + fuzzer = QueryFuzzer(randomSeed(), &std::cout, &std::cerr); + } void initialize(Poco::Util::Application & self) override; diff --git a/programs/client/clickhouse-client.xml b/programs/client/clickhouse-client.xml index c32b63413e9..9ce7d1cb223 100644 --- a/programs/client/clickhouse-client.xml +++ b/programs/client/clickhouse-client.xml @@ -38,6 +38,21 @@ {display_name} \e[1;31m:)\e[0m + + + 9000 + + + diff --git a/programs/server/fuzzers/tcp_protocol_fuzzer.cpp b/programs/server/fuzzers/tcp_protocol_fuzzer.cpp index 950ea09669a..7cebdc2ad65 100644 --- a/programs/server/fuzzers/tcp_protocol_fuzzer.cpp +++ b/programs/server/fuzzers/tcp_protocol_fuzzer.cpp @@ -10,6 +10,7 @@ #include #include +#include #include @@ -25,6 +26,12 @@ static int64_t port = 9000; using namespace std::chrono_literals; +void on_exit() +{ + BaseDaemon::terminate(); + main_app.wait(); +} + extern "C" int LLVMFuzzerInitialize(int * argc, char ***argv) { @@ -60,6 +67,8 @@ int LLVMFuzzerInitialize(int * argc, char ***argv) exit(-1); } + atexit(on_exit); + return 0; } diff --git a/programs/server/play.html b/programs/server/play.html index 9590a65524c..0d76a01cf7e 100644 --- a/programs/server/play.html +++ b/programs/server/play.html @@ -522,6 +522,9 @@ const current_url = new URL(window.location); const opened_locally = location.protocol == 'file:'; + /// Run query instantly after page is loaded if the run parameter is present. + const run_immediately = current_url.searchParams.has("run"); + const server_address = current_url.searchParams.get('url'); if (server_address) { document.getElementById('url').value = server_address; @@ -599,6 +602,9 @@ const title = "ClickHouse Query: " + query; let history_url = window.location.pathname + '?user=' + encodeURIComponent(user); + if (run_immediately) { + history_url += "&run=1"; + } if (server_address != location.origin) { /// Save server's address in URL if it's not identical to the address of the play UI. history_url += '&url=' + encodeURIComponent(server_address); @@ -1160,6 +1166,10 @@ }); } + if (run_immediately) { + post(); + } + document.getElementById('toggle-light').onclick = function() { setColorTheme('light', true); } diff --git a/src/Access/AccessBackup.cpp b/src/Access/AccessBackup.cpp index 90effdab70f..d9ee89b45ce 100644 --- a/src/Access/AccessBackup.cpp +++ b/src/Access/AccessBackup.cpp @@ -93,7 +93,7 @@ namespace break; } - UUID id = parse(line); + UUID id = parse(line.substr(0, line.find('\t'))); line.clear(); String queries; diff --git a/src/AggregateFunctions/AggregateFunctionGroupArray.cpp b/src/AggregateFunctions/AggregateFunctionGroupArray.cpp index 7034e6373b1..5cc9f725b46 100644 --- a/src/AggregateFunctions/AggregateFunctionGroupArray.cpp +++ b/src/AggregateFunctions/AggregateFunctionGroupArray.cpp @@ -780,12 +780,12 @@ AggregateFunctionPtr createAggregateFunctionGroupArray( if (type != Field::Types::Int64 && type != Field::Types::UInt64) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Parameter for aggregate function {} should be positive number", name); - if ((type == Field::Types::Int64 && parameters[0].get() < 0) || - (type == Field::Types::UInt64 && parameters[0].get() == 0)) + if ((type == Field::Types::Int64 && parameters[0].safeGet() < 0) || + (type == Field::Types::UInt64 && parameters[0].safeGet() == 0)) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Parameter for aggregate function {} should be positive number", name); has_limit = true; - max_elems = parameters[0].get(); + max_elems = parameters[0].safeGet(); } else throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, @@ -816,11 +816,11 @@ AggregateFunctionPtr createAggregateFunctionGroupArraySample( if (type != Field::Types::Int64 && type != Field::Types::UInt64) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Parameter for aggregate function {} should be positive number", name); - if ((type == Field::Types::Int64 && parameters[i].get() < 0) || - (type == Field::Types::UInt64 && parameters[i].get() == 0)) + if ((type == Field::Types::Int64 && parameters[i].safeGet() < 0) || + (type == Field::Types::UInt64 && parameters[i].safeGet() == 0)) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Parameter for aggregate function {} should be positive number", name); - return parameters[i].get(); + return parameters[i].safeGet(); }; UInt64 max_elems = get_parameter(0); diff --git a/src/AggregateFunctions/AggregateFunctionGroupArrayIntersect.cpp b/src/AggregateFunctions/AggregateFunctionGroupArrayIntersect.cpp index 38f2fcb9fb9..36d00b1d9ec 100644 --- a/src/AggregateFunctions/AggregateFunctionGroupArrayIntersect.cpp +++ b/src/AggregateFunctions/AggregateFunctionGroupArrayIntersect.cpp @@ -83,16 +83,16 @@ public: if (version == 1) { for (size_t i = 0; i < arr_size; ++i) - set.insert(static_cast((*data_column)[offset + i].get())); + set.insert(static_cast((*data_column)[offset + i].safeGet())); } else if (!set.empty()) { typename State::Set new_set; for (size_t i = 0; i < arr_size; ++i) { - typename State::Set::LookupResult set_value = set.find(static_cast((*data_column)[offset + i].get())); + typename State::Set::LookupResult set_value = set.find(static_cast((*data_column)[offset + i].safeGet())); if (set_value != nullptr) - new_set.insert(static_cast((*data_column)[offset + i].get())); + new_set.insert(static_cast((*data_column)[offset + i].safeGet())); } set = std::move(new_set); } diff --git a/src/AggregateFunctions/AggregateFunctionGroupArrayMoving.cpp b/src/AggregateFunctions/AggregateFunctionGroupArrayMoving.cpp index 026b8d1956f..2c3ac7f883e 100644 --- a/src/AggregateFunctions/AggregateFunctionGroupArrayMoving.cpp +++ b/src/AggregateFunctions/AggregateFunctionGroupArrayMoving.cpp @@ -269,12 +269,12 @@ AggregateFunctionPtr createAggregateFunctionMoving( if (type != Field::Types::Int64 && type != Field::Types::UInt64) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Parameter for aggregate function {} should be positive integer", name); - if ((type == Field::Types::Int64 && parameters[0].get() <= 0) || - (type == Field::Types::UInt64 && parameters[0].get() == 0)) + if ((type == Field::Types::Int64 && parameters[0].safeGet() <= 0) || + (type == Field::Types::UInt64 && parameters[0].safeGet() == 0)) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Parameter for aggregate function {} should be positive integer", name); limit_size = true; - max_elems = parameters[0].get(); + max_elems = parameters[0].safeGet(); } else throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, diff --git a/src/AggregateFunctions/AggregateFunctionGroupArraySorted.cpp b/src/AggregateFunctions/AggregateFunctionGroupArraySorted.cpp index d41d743e17a..27043ed6aa6 100644 --- a/src/AggregateFunctions/AggregateFunctionGroupArraySorted.cpp +++ b/src/AggregateFunctions/AggregateFunctionGroupArraySorted.cpp @@ -397,11 +397,11 @@ AggregateFunctionPtr createAggregateFunctionGroupArray( if (type != Field::Types::Int64 && type != Field::Types::UInt64) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Parameter for aggregate function {} should be positive number", name); - if ((type == Field::Types::Int64 && parameters[0].get() < 0) || - (type == Field::Types::UInt64 && parameters[0].get() == 0)) + if ((type == Field::Types::Int64 && parameters[0].safeGet() < 0) || + (type == Field::Types::UInt64 && parameters[0].safeGet() == 0)) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Parameter for aggregate function {} should be positive number", name); - max_elems = parameters[0].get(); + max_elems = parameters[0].safeGet(); } else throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, diff --git a/src/AggregateFunctions/AggregateFunctionGroupConcat.cpp b/src/AggregateFunctions/AggregateFunctionGroupConcat.cpp index 5494ef74705..636ac80e350 100644 --- a/src/AggregateFunctions/AggregateFunctionGroupConcat.cpp +++ b/src/AggregateFunctions/AggregateFunctionGroupConcat.cpp @@ -247,7 +247,7 @@ AggregateFunctionPtr createAggregateFunctionGroupConcat( if (type != Field::Types::String) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "First parameter for aggregate function {} should be string", name); - delimiter = parameters[0].get(); + delimiter = parameters[0].safeGet(); } if (parameters.size() == 2) { @@ -256,12 +256,12 @@ AggregateFunctionPtr createAggregateFunctionGroupConcat( if (type != Field::Types::Int64 && type != Field::Types::UInt64) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Second parameter for aggregate function {} should be a positive number", name); - if ((type == Field::Types::Int64 && parameters[1].get() <= 0) || - (type == Field::Types::UInt64 && parameters[1].get() == 0)) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Second parameter for aggregate function {} should be a positive number, got: {}", name, parameters[1].get()); + if ((type == Field::Types::Int64 && parameters[1].safeGet() <= 0) || + (type == Field::Types::UInt64 && parameters[1].safeGet() == 0)) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Second parameter for aggregate function {} should be a positive number, got: {}", name, parameters[1].safeGet()); has_limit = true; - limit = parameters[1].get(); + limit = parameters[1].safeGet(); } if (has_limit) diff --git a/src/AggregateFunctions/AggregateFunctionGroupUniqArray.cpp b/src/AggregateFunctions/AggregateFunctionGroupUniqArray.cpp index 7b4300b3568..5cbf449c946 100644 --- a/src/AggregateFunctions/AggregateFunctionGroupUniqArray.cpp +++ b/src/AggregateFunctions/AggregateFunctionGroupUniqArray.cpp @@ -323,12 +323,12 @@ AggregateFunctionPtr createAggregateFunctionGroupUniqArray( if (type != Field::Types::Int64 && type != Field::Types::UInt64) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Parameter for aggregate function {} should be positive number", name); - if ((type == Field::Types::Int64 && parameters[0].get() < 0) || - (type == Field::Types::UInt64 && parameters[0].get() == 0)) + if ((type == Field::Types::Int64 && parameters[0].safeGet() < 0) || + (type == Field::Types::UInt64 && parameters[0].safeGet() == 0)) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Parameter for aggregate function {} should be positive number", name); limit_size = true; - max_elems = parameters[0].get(); + max_elems = parameters[0].safeGet(); } else throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, diff --git a/src/AggregateFunctions/AggregateFunctionKolmogorovSmirnovTest.cpp b/src/AggregateFunctions/AggregateFunctionKolmogorovSmirnovTest.cpp index 04eebe9f485..28e8d37b8c8 100644 --- a/src/AggregateFunctions/AggregateFunctionKolmogorovSmirnovTest.cpp +++ b/src/AggregateFunctions/AggregateFunctionKolmogorovSmirnovTest.cpp @@ -238,7 +238,7 @@ public: if (params[0].getType() != Field::Types::String) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Aggregate function {} require first parameter to be a String", getName()); - const auto & param = params[0].get(); + const auto & param = params[0].safeGet(); if (param == "two-sided") alternative = Alternative::TwoSided; else if (param == "less") @@ -255,7 +255,7 @@ public: if (params[1].getType() != Field::Types::String) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Aggregate function {} require second parameter to be a String", getName()); - method = params[1].get(); + method = params[1].safeGet(); if (method != "auto" && method != "exact" && method != "asymp" && method != "asymptotic") throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unknown method in aggregate function {}. " "It must be one of: 'auto', 'exact', 'asymp' (or 'asymptotic')", getName()); diff --git a/src/AggregateFunctions/AggregateFunctionLargestTriangleThreeBuckets.cpp b/src/AggregateFunctions/AggregateFunctionLargestTriangleThreeBuckets.cpp index 6d1e3c0f64b..813b13b6f7b 100644 --- a/src/AggregateFunctions/AggregateFunctionLargestTriangleThreeBuckets.cpp +++ b/src/AggregateFunctions/AggregateFunctionLargestTriangleThreeBuckets.cpp @@ -181,7 +181,7 @@ public: throw Exception( ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Aggregate function {} require first parameter to be a UInt64", getName()); - total_buckets = params[0].get(); + total_buckets = params[0].safeGet(); this->x_type = WhichDataType(arguments[0]).idx; this->y_type = WhichDataType(arguments[1]).idx; diff --git a/src/AggregateFunctions/AggregateFunctionMannWhitney.cpp b/src/AggregateFunctions/AggregateFunctionMannWhitney.cpp index f088737c340..fa90846650d 100644 --- a/src/AggregateFunctions/AggregateFunctionMannWhitney.cpp +++ b/src/AggregateFunctions/AggregateFunctionMannWhitney.cpp @@ -152,7 +152,7 @@ public: if (params[0].getType() != Field::Types::String) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Aggregate function {} require first parameter to be a String", getName()); - const auto & param = params[0].get(); + const auto & param = params[0].safeGet(); if (param == "two-sided") alternative = Alternative::TwoSided; else if (param == "less") @@ -169,7 +169,7 @@ public: if (params[1].getType() != Field::Types::UInt64) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Aggregate function {} require second parameter to be a UInt64", getName()); - continuity_correction = static_cast(params[1].get()); + continuity_correction = static_cast(params[1].safeGet()); } String getName() const override diff --git a/src/AggregateFunctions/AggregateFunctionQuantile.h b/src/AggregateFunctions/AggregateFunctionQuantile.h index 127dc06b642..423fd4bc569 100644 --- a/src/AggregateFunctions/AggregateFunctionQuantile.h +++ b/src/AggregateFunctions/AggregateFunctionQuantile.h @@ -117,7 +117,7 @@ public: throw Exception( ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Aggregate function {} requires relative accuracy parameter with Float64 type", getName()); - relative_accuracy = relative_accuracy_field.get(); + relative_accuracy = relative_accuracy_field.safeGet(); if (relative_accuracy <= 0 || relative_accuracy >= 1 || isNaN(relative_accuracy)) throw Exception( @@ -147,9 +147,9 @@ public: ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Aggregate function {} requires accuracy parameter with integer type", getName()); if (accuracy_field.getType() == Field::Types::Int64) - accuracy = accuracy_field.get(); + accuracy = accuracy_field.safeGet(); else - accuracy = accuracy_field.get(); + accuracy = accuracy_field.safeGet(); if (accuracy <= 0) throw Exception( diff --git a/src/AggregateFunctions/AggregateFunctionSumMap.cpp b/src/AggregateFunctions/AggregateFunctionSumMap.cpp index 666136a91b7..9a94c3dfe1a 100644 --- a/src/AggregateFunctions/AggregateFunctionSumMap.cpp +++ b/src/AggregateFunctions/AggregateFunctionSumMap.cpp @@ -300,12 +300,12 @@ public: /// Compatibility with previous versions. if (value.getType() == Field::Types::Decimal32) { - auto source = value.get>(); + auto source = value.safeGet>(); value = DecimalField(source.getValue(), source.getScale()); } else if (value.getType() == Field::Types::Decimal64) { - auto source = value.get>(); + auto source = value.safeGet>(); value = DecimalField(source.getValue(), source.getScale()); } @@ -355,7 +355,7 @@ public: /// Compatibility with previous versions. if (value.getType() == Field::Types::Decimal128) { - auto source = value.get>(); + auto source = value.safeGet>(); WhichDataType value_type(values_types[col_idx]); if (value_type.isDecimal32()) { @@ -560,7 +560,7 @@ private: template bool compareImpl(FieldType & x) const { - auto val = rhs.get(); + auto val = rhs.safeGet(); if (val > x) { x = val; @@ -600,7 +600,7 @@ private: template bool compareImpl(FieldType & x) const { - auto val = rhs.get(); + auto val = rhs.safeGet(); if (val < x) { x = val; diff --git a/src/AggregateFunctions/WindowFunction.h b/src/AggregateFunctions/WindowFunction.h new file mode 100644 index 00000000000..f7fbd7389ea --- /dev/null +++ b/src/AggregateFunctions/WindowFunction.h @@ -0,0 +1,117 @@ +#pragma once +#include +#include +#include + + +namespace DB +{ +namespace ErrorCodes +{ +extern const int BAD_ARGUMENTS; +} +class WindowTransform; + + +// Interface for true window functions. It's not much of an interface, they just +// accept the guts of WindowTransform and do 'something'. Given a small number of +// true window functions, and the fact that the WindowTransform internals are +// pretty much well-defined in domain terms (e.g. frame boundaries), this is +// somewhat acceptable. +class IWindowFunction +{ +public: + virtual ~IWindowFunction() = default; + + // Must insert the result for current_row. + virtual void windowInsertResultInto(const WindowTransform * transform, size_t function_index) const = 0; + + virtual std::optional getDefaultFrame() const { return {}; } + + virtual ColumnPtr castColumn(const Columns &, const std::vector &) { return nullptr; } + + /// Is the frame type supported by this function. + virtual bool checkWindowFrameType(const WindowTransform * /*transform*/) const { return true; } +}; + +// Runtime data for computing one window function. +struct WindowFunctionWorkspace +{ + AggregateFunctionPtr aggregate_function; + + // Cached value of aggregate function isState virtual method + bool is_aggregate_function_state = false; + + // This field is set for pure window functions. When set, we ignore the + // window_function.aggregate_function, and work through this interface + // instead. + IWindowFunction * window_function_impl = nullptr; + + std::vector argument_column_indices; + + // Will not be initialized for a pure window function. + mutable AlignedBuffer aggregate_function_state; + + // Argument columns. Be careful, this is a per-block cache. + std::vector argument_columns; + UInt64 cached_block_number = std::numeric_limits::max(); +}; + +// A basic implementation for a true window function. It pretends to be an +// aggregate function, but refuses to work as such. +struct WindowFunction : public IAggregateFunctionHelper, public IWindowFunction +{ + std::string name; + + WindowFunction( + const std::string & name_, const DataTypes & argument_types_, const Array & parameters_, const DataTypePtr & result_type_) + : IAggregateFunctionHelper(argument_types_, parameters_, result_type_), name(name_) + { + } + + bool isOnlyWindowFunction() const override { return true; } + + [[noreturn]] void fail() const + { + throw Exception( + ErrorCodes::BAD_ARGUMENTS, "The function '{}' can only be used as a window function, not as an aggregate function", getName()); + } + + String getName() const override { return name; } + void create(AggregateDataPtr __restrict) const override { } + void destroy(AggregateDataPtr __restrict) const noexcept override { } + bool hasTrivialDestructor() const override { return true; } + size_t sizeOfData() const override { return 0; } + size_t alignOfData() const override { return 1; } + void add(AggregateDataPtr __restrict, const IColumn **, size_t, Arena *) const override { fail(); } + void merge(AggregateDataPtr __restrict, ConstAggregateDataPtr, Arena *) const override { fail(); } + void serialize(ConstAggregateDataPtr __restrict, WriteBuffer &, std::optional) const override { fail(); } + void deserialize(AggregateDataPtr __restrict, ReadBuffer &, std::optional, Arena *) const override { fail(); } + void insertResultInto(AggregateDataPtr __restrict, IColumn &, Arena *) const override { fail(); } +}; + +template +struct StatefulWindowFunction : public WindowFunction +{ + StatefulWindowFunction( + const std::string & name_, const DataTypes & argument_types_, const Array & parameters_, const DataTypePtr & result_type_) + : WindowFunction(name_, argument_types_, parameters_, result_type_) + { + } + + size_t sizeOfData() const override { return sizeof(State); } + size_t alignOfData() const override { return 1; } + + void create(AggregateDataPtr __restrict place) const override { new (place) State(); } + + void destroy(AggregateDataPtr __restrict place) const noexcept override { reinterpret_cast(place)->~State(); } + + bool hasTrivialDestructor() const override { return std::is_trivially_destructible_v; } + + State & getState(const WindowFunctionWorkspace & workspace) const + { + return *reinterpret_cast(workspace.aggregate_function_state.data()); + } +}; + +} diff --git a/src/AggregateFunctions/fuzzers/CMakeLists.txt b/src/AggregateFunctions/fuzzers/CMakeLists.txt index 3ff7c86a00c..1ce0c52feb7 100644 --- a/src/AggregateFunctions/fuzzers/CMakeLists.txt +++ b/src/AggregateFunctions/fuzzers/CMakeLists.txt @@ -1,2 +1,2 @@ clickhouse_add_executable(aggregate_function_state_deserialization_fuzzer aggregate_function_state_deserialization_fuzzer.cpp ${SRCS}) -target_link_libraries(aggregate_function_state_deserialization_fuzzer PRIVATE dbms clickhouse_aggregate_functions clickhouse_functions) +target_link_libraries(aggregate_function_state_deserialization_fuzzer PRIVATE clickhouse_functions clickhouse_aggregate_functions) diff --git a/src/Analyzer/ConstantNode.cpp b/src/Analyzer/ConstantNode.cpp index 46c1f7fb1ed..c65090f5b55 100644 --- a/src/Analyzer/ConstantNode.cpp +++ b/src/Analyzer/ConstantNode.cpp @@ -5,6 +5,7 @@ #include #include #include +#include #include #include @@ -162,6 +163,7 @@ QueryTreeNodePtr ConstantNode::cloneImpl() const ASTPtr ConstantNode::toASTImpl(const ConvertToASTOptions & options) const { const auto & constant_value_literal = constant_value->getValue(); + const auto & constant_value_type = constant_value->getType(); auto constant_value_ast = std::make_shared(constant_value_literal); if (!options.add_cast_for_constants) @@ -169,7 +171,25 @@ ASTPtr ConstantNode::toASTImpl(const ConvertToASTOptions & options) const if (requiresCastCall()) { - auto constant_type_name_ast = std::make_shared(constant_value->getType()->getName()); + /** Value for DateTime64 is Decimal64, which is serialized as a string literal. + * If we serialize it as is, DateTime64 would be parsed from that string literal, which can be incorrect. + * For example, DateTime64 cannot be parsed from the short value, like '1', while it's a valid Decimal64 value. + * It could also lead to ambiguous parsing because we don't know if the string literal represents a date or a Decimal64 literal. + * For this reason, we use a string literal representing a date instead of a Decimal64 literal. + */ + if (WhichDataType(constant_value_type->getTypeId()).isDateTime64()) + { + const auto * date_time_type = typeid_cast(constant_value_type.get()); + DecimalField decimal_value; + if (constant_value_literal.tryGet>(decimal_value)) + { + WriteBufferFromOwnString ostr; + writeDateTimeText(decimal_value.getValue(), date_time_type->getScale(), ostr, date_time_type->getTimeZone()); + constant_value_ast = std::make_shared(ostr.str()); + } + } + + auto constant_type_name_ast = std::make_shared(constant_value_type->getName()); return makeASTFunction("_CAST", std::move(constant_value_ast), std::move(constant_type_name_ast)); } diff --git a/src/Analyzer/FunctionNode.cpp b/src/Analyzer/FunctionNode.cpp index e98b04fe9a9..8e4e0725a2d 100644 --- a/src/Analyzer/FunctionNode.cpp +++ b/src/Analyzer/FunctionNode.cpp @@ -242,7 +242,8 @@ ASTPtr FunctionNode::toASTImpl(const ConvertToASTOptions & options) const /// Avoid cast for `IN tuple(...)` expression. /// Tuples could be quite big, and adding a type may significantly increase query size. /// It should be safe because set type for `column IN tuple` is deduced from `column` type. - if (isNameOfInFunction(function_name) && argument_nodes.size() > 1 && argument_nodes[1]->getNodeType() == QueryTreeNodeType::CONSTANT) + if (isNameOfInFunction(function_name) && argument_nodes.size() > 1 && argument_nodes[1]->getNodeType() == QueryTreeNodeType::CONSTANT + && !static_cast(argument_nodes[1].get())->hasSourceExpression()) new_options.add_cast_for_constants = false; const auto & parameters = getParameters(); diff --git a/src/Analyzer/Passes/ComparisonTupleEliminationPass.cpp b/src/Analyzer/Passes/ComparisonTupleEliminationPass.cpp index 76dc8ab94b4..f31920f8e33 100644 --- a/src/Analyzer/Passes/ComparisonTupleEliminationPass.cpp +++ b/src/Analyzer/Passes/ComparisonTupleEliminationPass.cpp @@ -137,7 +137,7 @@ private: if (constant_node_value.getType() != Field::Types::Which::Tuple) return {}; - const auto & constant_tuple = constant_node_value.get(); + const auto & constant_tuple = constant_node_value.safeGet(); const auto & function_arguments_nodes = function_node_typed.getArguments().getNodes(); size_t function_arguments_nodes_size = function_arguments_nodes.size(); diff --git a/src/Analyzer/Passes/ConvertOrLikeChainPass.cpp b/src/Analyzer/Passes/ConvertOrLikeChainPass.cpp index 2b2ac95d7b9..6c4ce789993 100644 --- a/src/Analyzer/Passes/ConvertOrLikeChainPass.cpp +++ b/src/Analyzer/Passes/ConvertOrLikeChainPass.cpp @@ -89,7 +89,7 @@ public: if (!pattern || !isString(pattern->getResultType())) continue; - auto regexp = likePatternToRegexp(pattern->getValue().get()); + auto regexp = likePatternToRegexp(pattern->getValue().safeGet()); /// Case insensitive. Works with UTF-8 as well. if (is_ilike) regexp = "(?i)" + regexp; diff --git a/src/Analyzer/Passes/FunctionToSubcolumnsPass.cpp b/src/Analyzer/Passes/FunctionToSubcolumnsPass.cpp index b968f43c6a6..1fc3eec6833 100644 --- a/src/Analyzer/Passes/FunctionToSubcolumnsPass.cpp +++ b/src/Analyzer/Passes/FunctionToSubcolumnsPass.cpp @@ -68,10 +68,10 @@ void optimizeFunctionEmpty(QueryTreeNodePtr &, FunctionNode & function_node, Col String getSubcolumnNameForElement(const Field & value, const DataTypeTuple & data_type_tuple) { if (value.getType() == Field::Types::String) - return value.get(); + return value.safeGet(); if (value.getType() == Field::Types::UInt64) - return data_type_tuple.getNameByPosition(value.get()); + return data_type_tuple.getNameByPosition(value.safeGet()); return ""; } @@ -79,7 +79,7 @@ String getSubcolumnNameForElement(const Field & value, const DataTypeTuple & dat String getSubcolumnNameForElement(const Field & value, const DataTypeVariant &) { if (value.getType() == Field::Types::String) - return value.get(); + return value.safeGet(); return ""; } diff --git a/src/Analyzer/Passes/FuseFunctionsPass.cpp b/src/Analyzer/Passes/FuseFunctionsPass.cpp index 0175e304a2b..f3b109a10ed 100644 --- a/src/Analyzer/Passes/FuseFunctionsPass.cpp +++ b/src/Analyzer/Passes/FuseFunctionsPass.cpp @@ -187,7 +187,7 @@ FunctionNodePtr createFusedQuantilesNode(std::vector & nodes /// Sort nodes and parameters in ascending order of quantile level std::vector permutation(nodes.size()); iota(permutation.data(), permutation.size(), size_t(0)); - std::sort(permutation.begin(), permutation.end(), [&](size_t i, size_t j) { return parameters[i].get() < parameters[j].get(); }); + std::sort(permutation.begin(), permutation.end(), [&](size_t i, size_t j) { return parameters[i].safeGet() < parameters[j].safeGet(); }); std::vector new_nodes; new_nodes.reserve(permutation.size()); diff --git a/src/Analyzer/Passes/IfTransformStringsToEnumPass.cpp b/src/Analyzer/Passes/IfTransformStringsToEnumPass.cpp index d966f129d08..f81327c5d55 100644 --- a/src/Analyzer/Passes/IfTransformStringsToEnumPass.cpp +++ b/src/Analyzer/Passes/IfTransformStringsToEnumPass.cpp @@ -134,8 +134,8 @@ public: return; std::set string_values; - string_values.insert(first_literal->getValue().get()); - string_values.insert(second_literal->getValue().get()); + string_values.insert(first_literal->getValue().safeGet()); + string_values.insert(second_literal->getValue().safeGet()); changeIfArguments(*function_if_node, string_values, context); wrapIntoToString(*function_node, std::move(modified_if_node), context); @@ -163,7 +163,7 @@ public: if (!isArray(literal_to->getResultType()) || !isString(literal_default->getResultType())) return; - auto array_to = literal_to->getValue().get(); + auto array_to = literal_to->getValue().safeGet(); if (array_to.empty()) return; @@ -178,9 +178,9 @@ public: std::set string_values; for (const auto & value : array_to) - string_values.insert(value.get()); + string_values.insert(value.safeGet()); - string_values.insert(literal_default->getValue().get()); + string_values.insert(literal_default->getValue().safeGet()); changeTransformArguments(*function_modified_transform_node, string_values, context); wrapIntoToString(*function_node, std::move(modified_transform_node), context); diff --git a/src/Analyzer/Passes/NormalizeCountVariantsPass.cpp b/src/Analyzer/Passes/NormalizeCountVariantsPass.cpp index 3a8b6e75d40..02f1c93ea7f 100644 --- a/src/Analyzer/Passes/NormalizeCountVariantsPass.cpp +++ b/src/Analyzer/Passes/NormalizeCountVariantsPass.cpp @@ -54,7 +54,7 @@ public: } else if (function_node->getFunctionName() == "sum" && first_argument_constant_literal.getType() == Field::Types::UInt64 && - first_argument_constant_literal.get() == 1) + first_argument_constant_literal.safeGet() == 1) { function_node->getArguments().getNodes().clear(); resolveAggregateFunctionNodeByName(*function_node, "count"); diff --git a/src/Analyzer/Passes/OptimizeDateOrDateTimeConverterWithPreimagePass.cpp b/src/Analyzer/Passes/OptimizeDateOrDateTimeConverterWithPreimagePass.cpp index feb8bcc792d..0f33c302265 100644 --- a/src/Analyzer/Passes/OptimizeDateOrDateTimeConverterWithPreimagePass.cpp +++ b/src/Analyzer/Passes/OptimizeDateOrDateTimeConverterWithPreimagePass.cpp @@ -143,13 +143,13 @@ private: const auto & column_type = column_node_typed.getColumnType().get(); if (isDateOrDate32(column_type)) { - start_date_or_date_time = date_lut.dateToString(range.first.get()); - end_date_or_date_time = date_lut.dateToString(range.second.get()); + start_date_or_date_time = date_lut.dateToString(range.first.safeGet()); + end_date_or_date_time = date_lut.dateToString(range.second.safeGet()); } else if (isDateTime(column_type) || isDateTime64(column_type)) { - start_date_or_date_time = date_lut.timeToString(range.first.get()); - end_date_or_date_time = date_lut.timeToString(range.second.get()); + start_date_or_date_time = date_lut.timeToString(range.first.safeGet()); + end_date_or_date_time = date_lut.timeToString(range.second.safeGet()); } else [[unlikely]] return {}; diff --git a/src/Analyzer/Passes/RewriteAggregateFunctionWithIfPass.cpp b/src/Analyzer/Passes/RewriteAggregateFunctionWithIfPass.cpp index a48e88132a6..091061ceb81 100644 --- a/src/Analyzer/Passes/RewriteAggregateFunctionWithIfPass.cpp +++ b/src/Analyzer/Passes/RewriteAggregateFunctionWithIfPass.cpp @@ -60,7 +60,7 @@ public: { const auto & second_const_value = second_const_node->getValue(); if (second_const_value.isNull() - || (lower_name == "sum" && isInt64OrUInt64FieldType(second_const_value.getType()) && second_const_value.get() == 0 + || (lower_name == "sum" && isInt64OrUInt64FieldType(second_const_value.getType()) && second_const_value.safeGet() == 0 && !if_node->getResultType()->isNullable())) { /// avg(if(cond, a, null)) -> avgIf(a::ResultTypeIf, cond) @@ -89,7 +89,7 @@ public: { const auto & first_const_value = first_const_node->getValue(); if (first_const_value.isNull() - || (lower_name == "sum" && isInt64OrUInt64FieldType(first_const_value.getType()) && first_const_value.get() == 0 + || (lower_name == "sum" && isInt64OrUInt64FieldType(first_const_value.getType()) && first_const_value.safeGet() == 0 && !if_node->getResultType()->isNullable())) { /// avg(if(cond, null, a) -> avgIf(a::ResultTypeIf, !cond)) diff --git a/src/Analyzer/Passes/ShardNumColumnToFunctionPass.cpp b/src/Analyzer/Passes/ShardNumColumnToFunctionPass.cpp index 82e3281121c..c58504064ce 100644 --- a/src/Analyzer/Passes/ShardNumColumnToFunctionPass.cpp +++ b/src/Analyzer/Passes/ShardNumColumnToFunctionPass.cpp @@ -46,7 +46,7 @@ public: return; const auto & storage_snapshot = table_node ? table_node->getStorageSnapshot() : table_function_node->getStorageSnapshot(); - if (!storage->isVirtualColumn(column.name, storage_snapshot->getMetadataForQuery())) + if (!storage->isVirtualColumn(column.name, storage_snapshot->metadata)) return; auto function_node = std::make_shared("shardNum"); diff --git a/src/Analyzer/Passes/SumIfToCountIfPass.cpp b/src/Analyzer/Passes/SumIfToCountIfPass.cpp index 1524629dc81..a987ced497a 100644 --- a/src/Analyzer/Passes/SumIfToCountIfPass.cpp +++ b/src/Analyzer/Passes/SumIfToCountIfPass.cpp @@ -66,7 +66,7 @@ public: resolveAggregateFunctionNodeByName(*function_node, "countIf"); - if (constant_value_literal.get() != 1) + if (constant_value_literal.safeGet() != 1) { /// Rewrite `sumIf(123, cond)` into `123 * countIf(cond)` node = getMultiplyFunction(std::move(multiplier_node), node); @@ -105,8 +105,8 @@ public: const auto & if_true_condition_constant_value_literal = if_true_condition_constant_node->getValue(); const auto & if_false_condition_constant_value_literal = if_false_condition_constant_node->getValue(); - auto if_true_condition_value = if_true_condition_constant_value_literal.get(); - auto if_false_condition_value = if_false_condition_constant_value_literal.get(); + auto if_true_condition_value = if_true_condition_constant_value_literal.safeGet(); + auto if_false_condition_value = if_false_condition_constant_value_literal.safeGet(); if (if_false_condition_value == 0) { diff --git a/src/Analyzer/QueryTreeBuilder.cpp b/src/Analyzer/QueryTreeBuilder.cpp index fb41826929f..9754897d54d 100644 --- a/src/Analyzer/QueryTreeBuilder.cpp +++ b/src/Analyzer/QueryTreeBuilder.cpp @@ -471,7 +471,7 @@ QueryTreeNodePtr QueryTreeBuilder::buildSortList(const ASTPtr & order_by_express std::shared_ptr collator; if (order_by_element.getCollation()) - collator = std::make_shared(order_by_element.getCollation()->as().value.get()); + collator = std::make_shared(order_by_element.getCollation()->as().value.safeGet()); const auto & sort_expression_ast = order_by_element.children.at(0); auto sort_expression = buildExpression(sort_expression_ast, context); diff --git a/src/Analyzer/Resolve/IdentifierResolver.cpp b/src/Analyzer/Resolve/IdentifierResolver.cpp index 447bf825836..a79433ac130 100644 --- a/src/Analyzer/Resolve/IdentifierResolver.cpp +++ b/src/Analyzer/Resolve/IdentifierResolver.cpp @@ -1273,7 +1273,7 @@ QueryTreeNodePtr IdentifierResolver::matchArrayJoinSubcolumns( const auto & constant_node_value = constant_node.getValue(); if (constant_node_value.getType() == Field::Types::String) { - array_join_subcolumn_prefix = constant_node_value.get() + "."; + array_join_subcolumn_prefix = constant_node_value.safeGet() + "."; array_join_parent_column = argument_nodes.at(0).get(); } } @@ -1287,7 +1287,7 @@ QueryTreeNodePtr IdentifierResolver::matchArrayJoinSubcolumns( if (!second_argument || second_argument->getValue().getType() != Field::Types::String) throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected constant string as second argument of getSubcolumn function {}", resolved_function->dumpTree()); - const auto & resolved_subcolumn_path = second_argument->getValue().get(); + const auto & resolved_subcolumn_path = second_argument->getValue().safeGet(); if (!startsWith(resolved_subcolumn_path, array_join_subcolumn_prefix)) return {}; @@ -1331,7 +1331,7 @@ QueryTreeNodePtr IdentifierResolver::tryResolveExpressionFromArrayJoinExpression size_t nested_function_arguments_size = nested_function_arguments.size(); const auto & nested_keys_names_constant_node = nested_function_arguments[0]->as(); - const auto & nested_keys_names = nested_keys_names_constant_node.getValue().get(); + const auto & nested_keys_names = nested_keys_names_constant_node.getValue().safeGet(); size_t nested_keys_names_size = nested_keys_names.size(); if (nested_keys_names_size == nested_function_arguments_size - 1) @@ -1344,7 +1344,7 @@ QueryTreeNodePtr IdentifierResolver::tryResolveExpressionFromArrayJoinExpression auto array_join_column = std::make_shared(array_join_column_expression_typed.getColumn(), array_join_column_expression_typed.getColumnSource()); - const auto & nested_key_name = nested_keys_names[i - 1].get(); + const auto & nested_key_name = nested_keys_names[i - 1].safeGet(); Identifier nested_identifier = Identifier(nested_key_name); array_join_resolved_expression = wrapExpressionNodeInTupleElement(array_join_column, nested_identifier, scope.context); break; diff --git a/src/Analyzer/Resolve/QueryAnalyzer.cpp b/src/Analyzer/Resolve/QueryAnalyzer.cpp index e04b93314c5..2e8acf2f201 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.cpp +++ b/src/Analyzer/Resolve/QueryAnalyzer.cpp @@ -748,11 +748,11 @@ void QueryAnalyzer::replaceNodesWithPositionalArguments(QueryTreeNodePtr & node_ UInt64 pos; if (constant_node->getValue().getType() == Field::Types::UInt64) { - pos = constant_node->getValue().get(); + pos = constant_node->getValue().safeGet(); } else // Int64 { - auto value = constant_node->getValue().get(); + auto value = constant_node->getValue().safeGet(); if (value > 0) pos = value; else diff --git a/src/Analyzer/SetUtils.cpp b/src/Analyzer/SetUtils.cpp index e7d1a22f45a..59a243b27f3 100644 --- a/src/Analyzer/SetUtils.cpp +++ b/src/Analyzer/SetUtils.cpp @@ -99,7 +99,7 @@ Block createBlockFromCollection(const Collection & collection, const DataTypes& "Invalid type in set. Expected tuple, got {}", value.getTypeName()); - const auto & tuple = value.template get(); + const auto & tuple = value.template safeGet(); const DataTypePtr & value_type = value_types[collection_index]; const DataTypes & tuple_value_type = typeid_cast(value_type.get())->getElements(); @@ -175,15 +175,15 @@ Block getSetElementsForConstantValue(const DataTypePtr & expression_type, const if (rhs_which_type.isArray()) { const DataTypeArray * value_array_type = assert_cast(value_type.get()); - size_t value_array_size = value.get().size(); + size_t value_array_size = value.safeGet().size(); DataTypes value_types(value_array_size, value_array_type->getNestedType()); - result_block = createBlockFromCollection(value.get(), value_types, set_element_types, transform_null_in); + result_block = createBlockFromCollection(value.safeGet(), value_types, set_element_types, transform_null_in); } else if (rhs_which_type.isTuple()) { const DataTypeTuple * value_tuple_type = assert_cast(value_type.get()); const DataTypes & value_types = value_tuple_type->getElements(); - result_block = createBlockFromCollection(value.get(), value_types, set_element_types, transform_null_in); + result_block = createBlockFromCollection(value.safeGet(), value_types, set_element_types, transform_null_in); } else throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, diff --git a/src/Backups/BackupSettings.cpp b/src/Backups/BackupSettings.cpp index 37ddd344001..e982a806b7c 100644 --- a/src/Backups/BackupSettings.cpp +++ b/src/Backups/BackupSettings.cpp @@ -126,7 +126,7 @@ std::vector BackupSettings::Util::clusterHostIDsFromAST(const IAST & as throw Exception( ErrorCodes::CANNOT_PARSE_BACKUP_SETTINGS, "Setting cluster_host_ids has wrong format, must be array of arrays of string literals"); - const auto & replicas = array_of_replicas->value.get(); + const auto & replicas = array_of_replicas->value.safeGet(); res[i].resize(replicas.size()); for (size_t j = 0; j != replicas.size(); ++j) { @@ -135,7 +135,7 @@ std::vector BackupSettings::Util::clusterHostIDsFromAST(const IAST & as throw Exception( ErrorCodes::CANNOT_PARSE_BACKUP_SETTINGS, "Setting cluster_host_ids has wrong format, must be array of arrays of string literals"); - res[i][j] = replica.get(); + res[i][j] = replica.safeGet(); } } } diff --git a/src/Backups/DDLAdjustingForBackupVisitor.cpp b/src/Backups/DDLAdjustingForBackupVisitor.cpp index 910831195a3..4dcbdcc1617 100644 --- a/src/Backups/DDLAdjustingForBackupVisitor.cpp +++ b/src/Backups/DDLAdjustingForBackupVisitor.cpp @@ -46,8 +46,8 @@ namespace if (zookeeper_path_ast && (zookeeper_path_ast->value.getType() == Field::Types::String) && replica_name_ast && (replica_name_ast->value.getType() == Field::Types::String)) { - String & zookeeper_path_arg = zookeeper_path_ast->value.get(); - String & replica_name_arg = replica_name_ast->value.get(); + String & zookeeper_path_arg = zookeeper_path_ast->value.safeGet(); + String & replica_name_arg = replica_name_ast->value.safeGet(); if (create.uuid != UUIDHelpers::Nil) { String table_uuid_str = toString(create.uuid); diff --git a/src/Backups/RestoreSettings.cpp b/src/Backups/RestoreSettings.cpp index a974fc11d00..8e60e8d129e 100644 --- a/src/Backups/RestoreSettings.cpp +++ b/src/Backups/RestoreSettings.cpp @@ -31,7 +31,7 @@ namespace { if (field.getType() == Field::Types::String) { - const String & str = field.get(); + const String & str = field.safeGet(); if (str == "1" || boost::iequals(str, "true") || boost::iequals(str, "create")) { value = RestoreTableCreationMode::kCreate; @@ -54,7 +54,7 @@ namespace if (field.getType() == Field::Types::UInt64) { - UInt64 number = field.get(); + UInt64 number = field.safeGet(); if (number == 1) { value = RestoreTableCreationMode::kCreate; @@ -95,7 +95,7 @@ namespace { if (field.getType() == Field::Types::String) { - const String & str = field.get(); + const String & str = field.safeGet(); if (str == "1" || boost::iequals(str, "true") || boost::iequals(str, "create")) { value = RestoreAccessCreationMode::kCreate; @@ -118,7 +118,7 @@ namespace if (field.getType() == Field::Types::UInt64) { - UInt64 number = field.get(); + UInt64 number = field.safeGet(); if (number == 1) { value = RestoreAccessCreationMode::kCreate; diff --git a/src/Backups/SettingsFieldOptionalString.cpp b/src/Backups/SettingsFieldOptionalString.cpp index 573fd1e052c..684407a533d 100644 --- a/src/Backups/SettingsFieldOptionalString.cpp +++ b/src/Backups/SettingsFieldOptionalString.cpp @@ -19,7 +19,7 @@ SettingFieldOptionalString::SettingFieldOptionalString(const Field & field) if (field.getType() == Field::Types::String) { - value = field.get(); + value = field.safeGet(); return; } diff --git a/src/Backups/SettingsFieldOptionalUUID.cpp b/src/Backups/SettingsFieldOptionalUUID.cpp index 3f14608b206..0011f7f1073 100644 --- a/src/Backups/SettingsFieldOptionalUUID.cpp +++ b/src/Backups/SettingsFieldOptionalUUID.cpp @@ -22,7 +22,7 @@ namespace ErrorCodes if (field.getType() == Field::Types::String) { - const String & str = field.get(); + const String & str = field.safeGet(); if (str.empty()) { value = std::nullopt; diff --git a/src/Backups/WithRetries.cpp b/src/Backups/WithRetries.cpp index 181e6331ac9..9f22085f5a9 100644 --- a/src/Backups/WithRetries.cpp +++ b/src/Backups/WithRetries.cpp @@ -68,13 +68,19 @@ const WithRetries::KeeperSettings & WithRetries::getKeeperSettings() const WithRetries::FaultyKeeper WithRetries::getFaultyZooKeeper() const { - /// We need to create new instance of ZooKeeperWithFaultInjection each time a copy a pointer to ZooKeeper client there + zkutil::ZooKeeperPtr current_zookeeper; + { + std::lock_guard lock(zookeeper_mutex); + current_zookeeper = zookeeper; + } + + /// We need to create new instance of ZooKeeperWithFaultInjection each time and copy a pointer to ZooKeeper client there /// The reason is that ZooKeeperWithFaultInjection may reset the underlying pointer and there could be a race condition /// when the same object is used from multiple threads. auto faulty_zookeeper = ZooKeeperWithFaultInjection::createInstance( settings.keeper_fault_injection_probability, settings.keeper_fault_injection_seed, - zookeeper, + current_zookeeper, log->name(), log); diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt index 8c133971785..95e431b54be 100644 --- a/src/CMakeLists.txt +++ b/src/CMakeLists.txt @@ -1,4 +1,4 @@ -add_compile_options($<$,$>:${COVERAGE_FLAGS}>) +add_compile_options("$<$,$>:${COVERAGE_FLAGS}>") if (USE_INCLUDE_WHAT_YOU_USE) set (CMAKE_CXX_INCLUDE_WHAT_YOU_USE ${IWYU_PATH}) @@ -78,6 +78,7 @@ add_headers_and_sources(clickhouse_common_io Common/Scheduler) add_headers_and_sources(clickhouse_common_io Common/Scheduler/Nodes) add_headers_and_sources(clickhouse_common_io IO) add_headers_and_sources(clickhouse_common_io IO/Archives) +add_headers_and_sources(clickhouse_common_io IO/Protobuf) add_headers_and_sources(clickhouse_common_io IO/S3) add_headers_and_sources(clickhouse_common_io IO/AzureBlobStorage) list (REMOVE_ITEM clickhouse_common_io_sources Common/malloc.cpp Common/new_delete.cpp) @@ -225,6 +226,7 @@ add_object_library(clickhouse_storages_liveview Storages/LiveView) add_object_library(clickhouse_storages_windowview Storages/WindowView) add_object_library(clickhouse_storages_s3queue Storages/ObjectStorageQueue) add_object_library(clickhouse_storages_materializedview Storages/MaterializedView) +add_object_library(clickhouse_storages_time_series Storages/TimeSeries) 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 @@ -351,8 +353,8 @@ target_link_libraries(clickhouse_common_io Poco::Foundation ) -if (TARGET ch_contrib::fiu) - target_link_libraries(clickhouse_common_io PUBLIC ch_contrib::fiu) +if (TARGET ch_contrib::libfiu) + target_link_libraries(clickhouse_common_io PUBLIC ch_contrib::libfiu) endif() if (TARGET ch_contrib::cpuid) @@ -469,6 +471,7 @@ dbms_target_link_libraries (PUBLIC ch_contrib::sparsehash) if (TARGET ch_contrib::protobuf) dbms_target_link_libraries (PRIVATE ch_contrib::protobuf) + target_link_libraries (clickhouse_common_io PUBLIC ch_contrib::protobuf) endif () if (TARGET clickhouse_grpc_protos) @@ -553,14 +556,13 @@ target_link_libraries (clickhouse_common_io PRIVATE ch_contrib::lz4) if (TARGET ch_contrib::qpl) dbms_target_link_libraries(PUBLIC ch_contrib::qpl) + target_link_libraries (clickhouse_compression PUBLIC ch_contrib::qpl) + target_link_libraries (clickhouse_compression PUBLIC ch_contrib::accel-config) endif () -if (TARGET ch_contrib::accel-config) - dbms_target_link_libraries(PUBLIC ch_contrib::accel-config) -endif () - -if (TARGET ch_contrib::qatzstd_plugin) +if (TARGET ch_contrib::accel-config AND TARGET ch_contrib::qatzstd_plugin) dbms_target_link_libraries(PUBLIC ch_contrib::qatzstd_plugin) + dbms_target_link_libraries(PUBLIC ch_contrib::accel-config) target_link_libraries(clickhouse_common_io PUBLIC ch_contrib::qatzstd_plugin) endif () @@ -599,10 +601,6 @@ endif() dbms_target_link_libraries(PUBLIC ch_contrib::consistent_hashing) -if (TARGET ch_contrib::annoy) - dbms_target_link_libraries(PUBLIC ch_contrib::annoy) -endif() - if (TARGET ch_contrib::usearch) dbms_target_link_libraries(PUBLIC ch_contrib::usearch) endif() diff --git a/src/Client/ClientApplicationBase.cpp b/src/Client/ClientApplicationBase.cpp index 9f133616d2e..0649aa5f5d7 100644 --- a/src/Client/ClientApplicationBase.cpp +++ b/src/Client/ClientApplicationBase.cpp @@ -158,6 +158,8 @@ void ClientApplicationBase::init(int argc, char ** argv) ("config-file,C", po::value(), "config-file path") + ("proto_caps", po::value(), "enable/disable chunked protocol: chunked_optional, notchunked, notchunked_optional, send_chunked, send_chunked_optional, send_notchunked, send_notchunked_optional, recv_chunked, recv_chunked_optional, recv_notchunked, recv_notchunked_optional") + ("query,q", po::value>()->multitoken(), R"(Query. Can be specified multiple times (--query "SELECT 1" --query "SELECT 2") or once with multiple comma-separated queries (--query "SELECT 1; SELECT 2;"). In the latter case, INSERT queries with non-VALUE format must be separated by empty lines.)") ("queries-file", po::value>()->multitoken(), "file path with queries to execute; multiple files can be specified (--queries-file file1 file2...)") ("multiquery,n", "Obsolete, does nothing") @@ -200,8 +202,6 @@ void ClientApplicationBase::init(int argc, char ** argv) ("pager", po::value(), "Pipe all output into this command (less or similar)") ("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)") ; @@ -339,6 +339,41 @@ void ClientApplicationBase::init(int argc, char ** argv) if (options.count("server_logs_file")) server_logs_file = options["server_logs_file"].as(); + if (options.count("proto_caps")) + { + std::string proto_caps_str = options["proto_caps"].as(); + + std::vector proto_caps; + splitInto<','>(proto_caps, proto_caps_str); + + for (auto cap_str : proto_caps) + { + std::string direction; + + if (cap_str.starts_with("send_")) + { + direction = "send"; + cap_str = cap_str.substr(std::string_view("send_").size()); + } + else if (cap_str.starts_with("recv_")) + { + direction = "recv"; + cap_str = cap_str.substr(std::string_view("recv_").size()); + } + + if (cap_str != "chunked" && cap_str != "notchunked" && cap_str != "chunked_optional" && cap_str != "notchunked_optional") + throw Exception(ErrorCodes::BAD_ARGUMENTS, "proto_caps option is incorrect ({})", proto_caps_str); + + if (direction.empty()) + { + config().setString("proto_caps.send", std::string(cap_str)); + config().setString("proto_caps.recv", std::string(cap_str)); + } + else + config().setString("proto_caps." + direction, std::string(cap_str)); + } + } + query_processing_stage = QueryProcessingStage::fromString(options["stage"].as()); query_kind = parseQueryKind(options["query_kind"].as()); profile_events.print = options.count("print-profile-events"); diff --git a/src/Client/ClientBase.cpp b/src/Client/ClientBase.cpp index 5642a72608c..187ef079eda 100644 --- a/src/Client/ClientBase.cpp +++ b/src/Client/ClientBase.cpp @@ -73,9 +73,11 @@ #include #include #include +#include #include #include +#include #include "config.h" #include #include @@ -477,6 +479,8 @@ void ClientBase::onProfileInfo(const ProfileInfo & profile_info) { if (profile_info.hasAppliedLimit() && output_format) output_format->setRowsBeforeLimit(profile_info.getRowsBeforeLimit()); + if (profile_info.hasAppliedAggregation() && output_format) + output_format->setRowsBeforeAggregation(profile_info.getRowsBeforeAggregation()); } @@ -912,6 +916,8 @@ void ClientBase::processTextAsSingleQuery(const String & full_query) } catch (Exception & e) { + if (server_exception) + server_exception->rethrow(); if (!is_interactive) e.addMessage("(in query: {})", full_query); throw; @@ -1030,19 +1036,28 @@ void ClientBase::processOrdinaryQuery(const String & query_to_execute, ASTPtr pa query_interrupt_handler.start(signals_before_stop); SCOPE_EXIT({ query_interrupt_handler.stop(); }); - connection->sendQuery( - connection_parameters.timeouts, - query, - query_parameters, - client_context->getCurrentQueryId(), - query_processing_stage, - &client_context->getSettingsRef(), - &client_context->getClientInfo(), - true, - [&](const Progress & progress) { onProgress(progress); }); + try { + connection->sendQuery( + connection_parameters.timeouts, + query, + query_parameters, + client_context->getCurrentQueryId(), + query_processing_stage, + &client_context->getSettingsRef(), + &client_context->getClientInfo(), + true, + [&](const Progress & progress) { onProgress(progress); }); + + if (send_external_tables) + sendExternalTables(parsed_query); + } + catch (const NetException &) + { + // We still want to attempt to process whatever we already received or can receive (socket receive buffer can be not empty) + receiveResult(parsed_query, signals_before_stop, settings.partial_result_on_first_cancel); + throw; + } - if (send_external_tables) - sendExternalTables(parsed_query); receiveResult(parsed_query, signals_before_stop, settings.partial_result_on_first_cancel); break; @@ -2749,7 +2764,7 @@ void ClientBase::runLibFuzzer() for (auto & arg : fuzzer_args_holder) fuzzer_args.emplace_back(arg.data()); - int fuzzer_argc = fuzzer_args.size(); + int fuzzer_argc = static_cast(fuzzer_args.size()); char ** fuzzer_argv = fuzzer_args.data(); LLVMFuzzerRunDriver(&fuzzer_argc, &fuzzer_argv, [](const uint8_t * data, size_t size) diff --git a/src/Client/ClientBase.h b/src/Client/ClientBase.h index 1a23b6b1363..45251aea28a 100644 --- a/src/Client/ClientBase.h +++ b/src/Client/ClientBase.h @@ -2,7 +2,7 @@ #include -#include +#include #include #include #include diff --git a/src/Client/Connection.cpp b/src/Client/Connection.cpp index 799c7511982..e89bd7a2bf5 100644 --- a/src/Client/Connection.cpp +++ b/src/Client/Connection.cpp @@ -5,8 +5,6 @@ #include #include #include -#include -#include #include #include #include @@ -85,6 +83,7 @@ Connection::~Connection() Connection::Connection(const String & host_, UInt16 port_, const String & default_database_, const String & user_, const String & password_, + const String & proto_send_chunked_, const String & proto_recv_chunked_, [[maybe_unused]] const SSHKey & ssh_private_key_, const String & jwt_, const String & quota_key_, @@ -95,6 +94,7 @@ Connection::Connection(const String & host_, UInt16 port_, Protocol::Secure secure_) : host(host_), port(port_), default_database(default_database_) , user(user_), password(password_) + , proto_send_chunked(proto_send_chunked_), proto_recv_chunked(proto_recv_chunked_) #if USE_SSH , ssh_private_key(ssh_private_key_) #endif @@ -211,10 +211,10 @@ void Connection::connect(const ConnectionTimeouts & timeouts) , tcp_keep_alive_timeout_in_sec); } - in = std::make_shared(*socket); + in = std::make_shared(*socket); in->setAsyncCallback(async_callback); - out = std::make_shared(*socket); + out = std::make_shared(*socket); out->setAsyncCallback(async_callback); connected = true; setDescription(); @@ -222,9 +222,61 @@ void Connection::connect(const ConnectionTimeouts & timeouts) sendHello(); receiveHello(timeouts.handshake_timeout); + if (server_revision >= DBMS_MIN_PROTOCOL_VERSION_WITH_CHUNKED_PACKETS) + { + /// Client side of chunked protocol negotiation. + /// Server advertises its protocol capabilities (separate for send and receive channels) by sending + /// in its 'Hello' response one of four types - chunked, notchunked, chunked_optional, notchunked_optional. + /// Not optional types are strict meaning that server only supports this type, optional means that + /// server prefer this type but capable to work in opposite. + /// Client selects which type it is going to communicate based on the settings from config or arguments, + /// and sends either "chunked" or "notchunked" protocol request in addendum section of handshake. + /// Client can detect if server's protocol capabilities are not compatible with client's settings (for example + /// server strictly requires chunked protocol but client's settings only allows notchunked protocol) - in such case + /// client should interrupt this connection. However if client continues with incompatible protocol type request, server + /// will send appropriate exception and disconnect client. + + auto is_chunked = [](const String & chunked_srv_str, const String & chunked_cl_str, const String & direction) + { + bool chunked_srv = chunked_srv_str.starts_with("chunked"); + bool optional_srv = chunked_srv_str.ends_with("_optional"); + bool chunked_cl = chunked_cl_str.starts_with("chunked"); + bool optional_cl = chunked_cl_str.ends_with("_optional"); + + if (optional_srv) + return chunked_cl; + if (optional_cl) + return chunked_srv; + if (chunked_cl != chunked_srv) + throw NetException( + ErrorCodes::NETWORK_ERROR, + "Incompatible protocol: {} set to {}, server requires {}", + direction, + chunked_cl ? "chunked" : "notchunked", + chunked_srv ? "chunked" : "notchunked"); + + return chunked_srv; + }; + + proto_send_chunked = is_chunked(proto_recv_chunked_srv, proto_send_chunked, "send") ? "chunked" : "notchunked"; + proto_recv_chunked = is_chunked(proto_send_chunked_srv, proto_recv_chunked, "recv") ? "chunked" : "notchunked"; + } + else + { + if (proto_send_chunked == "chunked" || proto_recv_chunked == "chunked") + throw NetException( + ErrorCodes::NETWORK_ERROR, + "Incompatible protocol: server's version is too old and doesn't support chunked protocol while client settings require it."); + } + if (server_revision >= DBMS_MIN_PROTOCOL_VERSION_WITH_ADDENDUM) sendAddendum(); + if (proto_send_chunked == "chunked") + out->enableChunked(); + if (proto_recv_chunked == "chunked") + in->enableChunked(); + LOG_TRACE(log_wrapper.get(), "Connected to {} server version {}.{}.{}.", server_name, server_version_major, server_version_minor, server_version_patch); } @@ -393,6 +445,13 @@ void Connection::sendAddendum() { if (server_revision >= DBMS_MIN_PROTOCOL_VERSION_WITH_QUOTA_KEY) writeStringBinary(quota_key, *out); + + if (server_revision >= DBMS_MIN_PROTOCOL_VERSION_WITH_CHUNKED_PACKETS) + { + writeStringBinary(proto_send_chunked, *out); + writeStringBinary(proto_recv_chunked, *out); + } + out->next(); } @@ -472,6 +531,12 @@ void Connection::receiveHello(const Poco::Timespan & handshake_timeout) else server_version_patch = server_revision; + if (server_revision >= DBMS_MIN_PROTOCOL_VERSION_WITH_CHUNKED_PACKETS) + { + readStringBinary(proto_send_chunked_srv, *in); + readStringBinary(proto_recv_chunked_srv, *in); + } + if (server_revision >= DBMS_MIN_PROTOCOL_VERSION_WITH_PASSWORD_COMPLEXITY_RULES) { UInt64 rules_size; @@ -611,6 +676,7 @@ bool Connection::ping(const ConnectionTimeouts & timeouts) UInt64 pong = 0; writeVarUInt(Protocol::Client::Ping, *out); + out->finishChunk(); out->next(); if (in->eof()) @@ -660,6 +726,7 @@ TablesStatusResponse Connection::getTablesStatus(const ConnectionTimeouts & time writeVarUInt(Protocol::Client::TablesStatusRequest, *out); request.write(*out, server_revision); + out->finishChunk(); out->next(); UInt64 response_type = 0; @@ -813,6 +880,8 @@ void Connection::sendQuery( block_profile_events_in.reset(); block_out.reset(); + out->finishChunk(); + /// Send empty block which means end of data. if (!with_pending_data) { @@ -829,6 +898,7 @@ void Connection::sendCancel() return; writeVarUInt(Protocol::Client::Cancel, *out); + out->finishChunk(); out->next(); } @@ -854,7 +924,10 @@ void Connection::sendData(const Block & block, const String & name, bool scalar) size_t prev_bytes = out->count(); block_out->write(block); - maybe_compressed_out->next(); + if (maybe_compressed_out != out) + maybe_compressed_out->next(); + if (!block) + out->finishChunk(); out->next(); if (throttler) @@ -865,6 +938,7 @@ void Connection::sendIgnoredPartUUIDs(const std::vector & uuids) { writeVarUInt(Protocol::Client::IgnoredPartUUIDs, *out); writeVectorBinary(uuids, *out); + out->finishChunk(); out->next(); } @@ -874,6 +948,7 @@ void Connection::sendReadTaskResponse(const String & response) writeVarUInt(Protocol::Client::ReadTaskResponse, *out); writeVarUInt(DBMS_CLUSTER_PROCESSING_PROTOCOL_VERSION, *out); writeStringBinary(response, *out); + out->finishChunk(); out->next(); } @@ -882,6 +957,7 @@ void Connection::sendMergeTreeReadTaskResponse(const ParallelReadResponse & resp { writeVarUInt(Protocol::Client::MergeTreeReadTaskResponse, *out); response.serialize(*out); + out->finishChunk(); out->next(); } @@ -899,6 +975,8 @@ void Connection::sendPreparedData(ReadBuffer & input, size_t size, const String copyData(input, *out); else copyData(input, *out, size); + + out->finishChunk(); out->next(); } @@ -927,6 +1005,8 @@ void Connection::sendScalarsData(Scalars & data) sendData(elem.second, elem.first, true /* scalar */); } + out->finishChunk(); + out_bytes = out->count() - out_bytes; maybe_compressed_out_bytes = maybe_compressed_out->count() - maybe_compressed_out_bytes; double elapsed = watch.elapsedSeconds(); @@ -1069,13 +1149,13 @@ std::optional Connection::getResolvedAddress() const bool Connection::poll(size_t timeout_microseconds) { - return static_cast(*in).poll(timeout_microseconds); + return in->poll(timeout_microseconds); } bool Connection::hasReadPendingData() const { - return last_input_packet_type.has_value() || static_cast(*in).hasPendingData(); + return last_input_packet_type.has_value() || in->hasBufferedData(); } @@ -1319,7 +1399,7 @@ Progress Connection::receiveProgress() const ProfileInfo Connection::receiveProfileInfo() const { ProfileInfo profile_info; - profile_info.read(*in); + profile_info.read(*in, server_revision); return profile_info; } @@ -1349,6 +1429,8 @@ ServerConnectionPtr Connection::createConnection(const ConnectionParameters & pa parameters.default_database, parameters.user, parameters.password, + parameters.proto_send_chunked, + parameters.proto_recv_chunked, parameters.ssh_private_key, parameters.jwt, parameters.quota_key, diff --git a/src/Client/Connection.h b/src/Client/Connection.h index 0f4b3e436df..ed84bc51318 100644 --- a/src/Client/Connection.h +++ b/src/Client/Connection.h @@ -8,8 +8,8 @@ #include -#include -#include +#include +#include #include #include @@ -52,6 +52,7 @@ public: Connection(const String & host_, UInt16 port_, const String & default_database_, const String & user_, const String & password_, + const String & proto_send_chunked_, const String & proto_recv_chunked_, const SSHKey & ssh_private_key_, const String & jwt_, const String & quota_key_, @@ -170,6 +171,10 @@ private: String default_database; String user; String password; + String proto_send_chunked; + String proto_recv_chunked; + String proto_send_chunked_srv; + String proto_recv_chunked_srv; #if USE_SSH SSHKey ssh_private_key; #endif @@ -209,8 +214,8 @@ private: String server_display_name; std::unique_ptr socket; - std::shared_ptr in; - std::shared_ptr out; + std::shared_ptr in; + std::shared_ptr out; std::optional last_input_packet_type; String query_id; diff --git a/src/Client/ConnectionEstablisher.cpp b/src/Client/ConnectionEstablisher.cpp index 4bab643ed31..f96546846c7 100644 --- a/src/Client/ConnectionEstablisher.cpp +++ b/src/Client/ConnectionEstablisher.cpp @@ -33,12 +33,12 @@ ConnectionEstablisher::ConnectionEstablisher( { } -void ConnectionEstablisher::run(ConnectionEstablisher::TryResult & result, std::string & fail_message) +void ConnectionEstablisher::run(ConnectionEstablisher::TryResult & result, std::string & fail_message, bool force_connected) { try { ProfileEvents::increment(ProfileEvents::DistributedConnectionTries); - result.entry = pool->get(*timeouts, settings, /* force_connected = */ false); + result.entry = pool->get(*timeouts, settings, force_connected); AsyncCallbackSetter async_setter(&*result.entry, std::move(async_callback)); UInt64 server_revision = 0; diff --git a/src/Client/ConnectionEstablisher.h b/src/Client/ConnectionEstablisher.h index a3a01e63246..ff071e59aea 100644 --- a/src/Client/ConnectionEstablisher.h +++ b/src/Client/ConnectionEstablisher.h @@ -24,7 +24,13 @@ public: const QualifiedTableName * table_to_check = nullptr); /// Establish connection and save it in result, write possible exception message in fail_message. - void run(TryResult & result, std::string & fail_message); + /// The connection is returned from connection pool and it can be stale. Use force_connected flag to ensure that connection is working one. + /// NOTE: force_connected is false by default due to the following consideration ... + /// When true, it implies sending a Ping packet to another peer and, if it fails - reestablishing the connection. + /// Ping-Pong round trip can be unnecessary in case of connection is still alive. + /// So, the optimistic approach is used by default. In this case, stale connections can be handled by retrying, + /// - see ConnectionPoolWithFailover, as example + void run(TryResult & result, std::string & fail_message, bool force_connected = false); /// Set async callback that will be called when reading from socket blocks. void setAsyncCallback(AsyncCallback async_callback_) { async_callback = std::move(async_callback_); } diff --git a/src/Client/ConnectionParameters.cpp b/src/Client/ConnectionParameters.cpp index 303bebc30d2..4d0a9ffa08c 100644 --- a/src/Client/ConnectionParameters.cpp +++ b/src/Client/ConnectionParameters.cpp @@ -107,6 +107,9 @@ ConnectionParameters::ConnectionParameters(const Poco::Util::AbstractConfigurati } } + proto_send_chunked = config.getString("proto_caps.send", "notchunked"); + proto_recv_chunked = config.getString("proto_caps.recv", "notchunked"); + quota_key = config.getString("quota_key", ""); /// By default compression is disabled if address looks like localhost. diff --git a/src/Client/ConnectionParameters.h b/src/Client/ConnectionParameters.h index c305c7813f2..382bfe34a3d 100644 --- a/src/Client/ConnectionParameters.h +++ b/src/Client/ConnectionParameters.h @@ -20,6 +20,8 @@ struct ConnectionParameters std::string default_database; std::string user; std::string password; + std::string proto_send_chunked = "notchunked"; + std::string proto_recv_chunked = "notchunked"; std::string quota_key; SSHKey ssh_private_key; std::string jwt; diff --git a/src/Client/ConnectionPool.cpp b/src/Client/ConnectionPool.cpp index ed2e7c3c725..ab8ad08826c 100644 --- a/src/Client/ConnectionPool.cpp +++ b/src/Client/ConnectionPool.cpp @@ -13,6 +13,8 @@ ConnectionPoolPtr ConnectionPoolFactory::get( String default_database, String user, String password, + String proto_send_chunked, + String proto_recv_chunked, String quota_key, String cluster, String cluster_secret, @@ -22,7 +24,7 @@ ConnectionPoolPtr ConnectionPoolFactory::get( Priority priority) { Key key{ - max_connections, host, port, default_database, user, password, quota_key, cluster, cluster_secret, client_name, compression, secure, priority}; + max_connections, host, port, default_database, user, password, proto_send_chunked, proto_recv_chunked, quota_key, cluster, cluster_secret, client_name, compression, secure, priority}; std::lock_guard lock(mutex); auto [it, inserted] = pools.emplace(key, ConnectionPoolPtr{}); @@ -39,6 +41,8 @@ ConnectionPoolPtr ConnectionPoolFactory::get( default_database, user, password, + proto_send_chunked, + proto_recv_chunked, quota_key, cluster, cluster_secret, diff --git a/src/Client/ConnectionPool.h b/src/Client/ConnectionPool.h index 0fcb3c4e7e1..219548b62a0 100644 --- a/src/Client/ConnectionPool.h +++ b/src/Client/ConnectionPool.h @@ -73,6 +73,8 @@ public: const String & default_database_, const String & user_, const String & password_, + const String & proto_send_chunked_, + const String & proto_recv_chunked_, const String & quota_key_, const String & cluster_, const String & cluster_secret_, @@ -85,6 +87,8 @@ public: , default_database(default_database_) , user(user_) , password(password_) + , proto_send_chunked(proto_send_chunked_) + , proto_recv_chunked(proto_recv_chunked_) , quota_key(quota_key_) , cluster(cluster_) , cluster_secret(cluster_secret_) @@ -116,7 +120,9 @@ protected: { return std::make_shared( host, port, - default_database, user, password, SSHKey(), /*jwt*/ "", quota_key, + default_database, user, password, + proto_send_chunked, proto_recv_chunked, + SSHKey(), /*jwt*/ "", quota_key, cluster, cluster_secret, client_name, compression, secure); } @@ -125,6 +131,8 @@ private: String default_database; String user; String password; + String proto_send_chunked; + String proto_recv_chunked; String quota_key; /// For inter-server authorization @@ -150,6 +158,8 @@ public: String default_database; String user; String password; + String proto_send_chunked; + String proto_recv_chunked; String quota_key; String cluster; String cluster_secret; @@ -173,6 +183,8 @@ public: String default_database, String user, String password, + String proto_send_chunked, + String proto_recv_chunked, String quota_key, String cluster, String cluster_secret, @@ -190,6 +202,7 @@ inline bool operator==(const ConnectionPoolFactory::Key & lhs, const ConnectionP { return lhs.max_connections == rhs.max_connections && lhs.host == rhs.host && lhs.port == rhs.port && lhs.default_database == rhs.default_database && lhs.user == rhs.user && lhs.password == rhs.password + && lhs.proto_send_chunked == rhs.proto_send_chunked && lhs.proto_recv_chunked == rhs.proto_recv_chunked && lhs.quota_key == rhs.quota_key && lhs.cluster == rhs.cluster && lhs.cluster_secret == rhs.cluster_secret && lhs.client_name == rhs.client_name && lhs.compression == rhs.compression && lhs.secure == rhs.secure && lhs.priority == rhs.priority; diff --git a/src/Client/LocalConnection.cpp b/src/Client/LocalConnection.cpp index 072184e0a66..7595a29912b 100644 --- a/src/Client/LocalConnection.cpp +++ b/src/Client/LocalConnection.cpp @@ -365,7 +365,7 @@ bool LocalConnection::poll(size_t) { while (pollImpl()) { - LOG_DEBUG(&Poco::Logger::get("LocalConnection"), "Executor timeout encountered, will retry"); + LOG_TEST(&Poco::Logger::get("LocalConnection"), "Executor timeout encountered, will retry"); if (needSendProgressOrMetrics()) return true; diff --git a/src/Client/Suggest.cpp b/src/Client/Suggest.cpp index 0188ebc8173..affd620f83a 100644 --- a/src/Client/Suggest.cpp +++ b/src/Client/Suggest.cpp @@ -214,7 +214,7 @@ void Suggest::fillWordsFromBlock(const Block & block) Words new_words; new_words.reserve(rows); for (size_t i = 0; i < rows; ++i) - new_words.emplace_back(column[i].get()); + new_words.emplace_back(column[i].safeGet()); addWords(std::move(new_words)); } diff --git a/src/Columns/ColumnAggregateFunction.cpp b/src/Columns/ColumnAggregateFunction.cpp index 4bc48c62eb4..d3363d91a46 100644 --- a/src/Columns/ColumnAggregateFunction.cpp +++ b/src/Columns/ColumnAggregateFunction.cpp @@ -457,9 +457,9 @@ MutableColumnPtr ColumnAggregateFunction::cloneEmpty() const Field ColumnAggregateFunction::operator[](size_t n) const { Field field = AggregateFunctionStateData(); - field.get().name = type_string; + field.safeGet().name = type_string; { - WriteBufferFromString buffer(field.get().data); + WriteBufferFromString buffer(field.safeGet().data); func->serialize(data[n], buffer, version); } return field; @@ -467,12 +467,7 @@ Field ColumnAggregateFunction::operator[](size_t n) const void ColumnAggregateFunction::get(size_t n, Field & res) const { - res = AggregateFunctionStateData(); - res.get().name = type_string; - { - WriteBufferFromString buffer(res.get().data); - func->serialize(data[n], buffer, version); - } + res = operator[](n); } StringRef ColumnAggregateFunction::getDataAt(size_t n) const @@ -552,7 +547,7 @@ void ColumnAggregateFunction::insert(const Field & x) "Inserting field of type {} into ColumnAggregateFunction. Expected {}", x.getTypeName(), Field::Types::AggregateFunctionState); - const auto & field_name = x.get().name; + const auto & field_name = x.safeGet().name; if (type_string != field_name) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Cannot insert filed with type {} into column with type {}", field_name, type_string); @@ -560,7 +555,7 @@ void ColumnAggregateFunction::insert(const Field & x) ensureOwnership(); Arena & arena = createOrGetArena(); pushBackAndCreateState(data, arena, func.get()); - ReadBufferFromString read_buffer(x.get().data); + ReadBufferFromString read_buffer(x.safeGet().data); func->deserialize(data.back(), read_buffer, version, &arena); } @@ -569,14 +564,14 @@ bool ColumnAggregateFunction::tryInsert(const DB::Field & x) if (x.getType() != Field::Types::AggregateFunctionState) return false; - const auto & field_name = x.get().name; + const auto & field_name = x.safeGet().name; if (type_string != field_name) return false; ensureOwnership(); Arena & arena = createOrGetArena(); pushBackAndCreateState(data, arena, func.get()); - ReadBufferFromString read_buffer(x.get().data); + ReadBufferFromString read_buffer(x.safeGet().data); func->deserialize(data.back(), read_buffer, version, &arena); return true; } diff --git a/src/Columns/ColumnArray.cpp b/src/Columns/ColumnArray.cpp index 19cce678cc7..26841879474 100644 --- a/src/Columns/ColumnArray.cpp +++ b/src/Columns/ColumnArray.cpp @@ -141,7 +141,7 @@ void ColumnArray::get(size_t n, Field & res) const size, max_array_size_as_field); res = Array(); - Array & res_arr = res.get(); + Array & res_arr = res.safeGet(); res_arr.reserve(size); for (size_t i = 0; i < size; ++i) @@ -309,7 +309,7 @@ void ColumnArray::updateHashFast(SipHash & hash) const void ColumnArray::insert(const Field & x) { - const Array & array = x.get(); + const Array & array = x.safeGet(); size_t size = array.size(); for (size_t i = 0; i < size; ++i) getData().insert(array[i]); @@ -321,7 +321,7 @@ bool ColumnArray::tryInsert(const Field & x) if (x.getType() != Field::Types::Which::Array) return false; - const Array & array = x.get(); + const Array & array = x.safeGet(); size_t size = array.size(); for (size_t i = 0; i < size; ++i) { @@ -452,6 +452,22 @@ void ColumnArray::reserve(size_t n) getData().reserve(n); /// The average size of arrays is not taken into account here. Or it is considered to be no more than 1. } +void ColumnArray::prepareForSquashing(const Columns & source_columns) +{ + size_t new_size = size(); + Columns source_data_columns; + source_data_columns.reserve(source_columns.size()); + for (const auto & source_column : source_columns) + { + const auto & source_array_column = assert_cast(*source_column); + new_size += source_array_column.size(); + source_data_columns.push_back(source_array_column.getDataPtr()); + } + + getOffsets().reserve_exact(new_size); + data->prepareForSquashing(source_data_columns); +} + void ColumnArray::shrinkToFit() { getOffsets().shrink_to_fit(); diff --git a/src/Columns/ColumnArray.h b/src/Columns/ColumnArray.h index 63affb86d9d..d6f71b72940 100644 --- a/src/Columns/ColumnArray.h +++ b/src/Columns/ColumnArray.h @@ -118,6 +118,7 @@ public: void updatePermutationWithCollation(const Collator & collator, PermutationSortDirection direction, PermutationSortStability stability, size_t limit, int nan_direction_hint, Permutation & res, EqualRanges& equal_ranges) const override; void reserve(size_t n) override; + void prepareForSquashing(const Columns & source_columns) override; void shrinkToFit() override; void ensureOwnership() override; size_t byteSize() const override; diff --git a/src/Columns/ColumnDecimal.h b/src/Columns/ColumnDecimal.h index 59bfbd2159c..07120f5f035 100644 --- a/src/Columns/ColumnDecimal.h +++ b/src/Columns/ColumnDecimal.h @@ -74,7 +74,7 @@ public: void insertData(const char * src, size_t /*length*/) override; void insertDefault() override { data.push_back(T()); } void insertManyDefaults(size_t length) override { data.resize_fill(data.size() + length); } - void insert(const Field & x) override { data.push_back(x.get()); } + void insert(const Field & x) override { data.push_back(x.safeGet()); } bool tryInsert(const Field & x) override; #if !defined(DEBUG_OR_SANITIZER_BUILD) void insertRangeFrom(const IColumn & src, size_t start, size_t length) override; diff --git a/src/Columns/ColumnDynamic.cpp b/src/Columns/ColumnDynamic.cpp index a92d54dd675..69b4c5dfc4e 100644 --- a/src/Columns/ColumnDynamic.cpp +++ b/src/Columns/ColumnDynamic.cpp @@ -1,16 +1,21 @@ #include #include +#include #include #include #include +#include #include #include #include #include #include #include -#include +#include +#include +#include +#include #include namespace DB @@ -22,31 +27,77 @@ namespace ErrorCodes extern const int PARAMETER_OUT_OF_BOUND; } - -ColumnDynamic::ColumnDynamic(size_t max_dynamic_types_) : max_dynamic_types(max_dynamic_types_) +namespace { - /// Create empty Variant. - variant_info.variant_type = std::make_shared(DataTypes{}); - variant_info.variant_name = variant_info.variant_type->getName(); - variant_column = variant_info.variant_type->createColumn(); + +/// Static default format settings to avoid creating it every time. +const FormatSettings & getFormatSettings() +{ + static const FormatSettings settings; + return settings; +} + +} + +/// Shared variant will contain String values but we cannot use usual String type +/// because we can have regular variant with type String. +/// To solve it, we use String type with custom name for shared variant. +DataTypePtr ColumnDynamic::getSharedVariantDataType() +{ + return DataTypeFactory::instance().getCustom("String", std::make_unique(std::make_unique(getSharedVariantTypeName()))); +} + +ColumnDynamic::ColumnDynamic(size_t max_dynamic_types_) : max_dynamic_types(max_dynamic_types_), global_max_dynamic_types(max_dynamic_types) +{ + /// Create Variant with shared variant. + setVariantType(std::make_shared(DataTypes{getSharedVariantDataType()})); } ColumnDynamic::ColumnDynamic( - MutableColumnPtr variant_column_, const VariantInfo & variant_info_, size_t max_dynamic_types_, const Statistics & statistics_) + MutableColumnPtr variant_column_, const DataTypePtr & variant_type_, size_t max_dynamic_types_, size_t global_max_dynamic_types_, const StatisticsPtr & statistics_) + : variant_column(std::move(variant_column_)) + , max_dynamic_types(max_dynamic_types_) + , global_max_dynamic_types(global_max_dynamic_types_) + , statistics(statistics_) +{ + createVariantInfo(variant_type_); +} + +ColumnDynamic::ColumnDynamic( + MutableColumnPtr variant_column_, const VariantInfo & variant_info_, size_t max_dynamic_types_, size_t global_max_dynamic_types_, const StatisticsPtr & statistics_) : variant_column(std::move(variant_column_)) , variant_info(variant_info_) , max_dynamic_types(max_dynamic_types_) + , global_max_dynamic_types(global_max_dynamic_types_) , statistics(statistics_) { } -ColumnDynamic::MutablePtr ColumnDynamic::create(MutableColumnPtr variant_column, const DataTypePtr & variant_type, size_t max_dynamic_types_, const Statistics & statistics_) +void ColumnDynamic::setVariantType(const DataTypePtr & variant_type) +{ + if (variant_column && !empty()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Setting specific variant type is allowed only for empty dynamic column"); + + variant_column = variant_type->createColumn(); + createVariantInfo(variant_type); +} + +void ColumnDynamic::setMaxDynamicPaths(size_t max_dynamic_type_) +{ + if (variant_column && !empty()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Setting specific max_dynamic_type parameter is allowed only for empty dynamic column"); + + max_dynamic_types = max_dynamic_type_; +} + +void ColumnDynamic::createVariantInfo(const DataTypePtr & variant_type) { - VariantInfo variant_info; variant_info.variant_type = variant_type; variant_info.variant_name = variant_type->getName(); const auto & variants = assert_cast(*variant_type).getVariants(); + variant_info.variant_names.clear(); variant_info.variant_names.reserve(variants.size()); + variant_info.variant_name_to_discriminator.clear(); variant_info.variant_name_to_discriminator.reserve(variants.size()); for (ColumnVariant::Discriminator discr = 0; discr != variants.size(); ++discr) { @@ -54,30 +105,26 @@ ColumnDynamic::MutablePtr ColumnDynamic::create(MutableColumnPtr variant_column, variant_info.variant_name_to_discriminator[variant_name] = discr; } - return create(std::move(variant_column), variant_info, max_dynamic_types_, statistics_); + if (!variant_info.variant_name_to_discriminator.contains(getSharedVariantTypeName())) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Variant in Dynamic column doesn't contain shared variant"); } -bool ColumnDynamic::addNewVariant(const DB::DataTypePtr & new_variant) +bool ColumnDynamic::addNewVariant(const DataTypePtr & new_variant, const String & new_variant_name) { /// Check if we already have such variant. - if (variant_info.variant_name_to_discriminator.contains(new_variant->getName())) + if (variant_info.variant_name_to_discriminator.contains(new_variant_name)) return true; /// Check if we reached maximum number of variants. - if (variant_info.variant_names.size() >= max_dynamic_types) + if (!canAddNewVariant()) { - /// ColumnDynamic can have max_dynamic_types number of variants only when it has String as a variant. - /// Otherwise we won't be able to cast new variants to Strings. - if (!variant_info.variant_name_to_discriminator.contains("String")) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Maximum number of variants reached, but no String variant exists"); + /// Dynamic column should always have shared variant. + if (!variant_info.variant_name_to_discriminator.contains(getSharedVariantTypeName())) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Maximum number of variants reached, but no shared variant exists"); return false; } - /// If we have (max_dynamic_types - 1) number of variants and don't have String variant, we can add only String variant. - if (variant_info.variant_names.size() == max_dynamic_types - 1 && new_variant->getName() != "String" && !variant_info.variant_name_to_discriminator.contains("String")) - return false; - const DataTypes & current_variants = assert_cast(*variant_info.variant_type).getVariants(); DataTypes all_variants = current_variants; all_variants.push_back(new_variant); @@ -86,21 +133,15 @@ bool ColumnDynamic::addNewVariant(const DB::DataTypePtr & new_variant) return true; } -void ColumnDynamic::addStringVariant() +void extendVariantColumn( + IColumn & variant_column, + const DataTypePtr & old_variant_type, + const DataTypePtr & new_variant_type, + std::unordered_map old_variant_name_to_discriminator) { - if (!addNewVariant(std::make_shared())) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot add String variant to Dynamic column, it's a bug"); -} - -void ColumnDynamic::updateVariantInfoAndExpandVariantColumn(const DB::DataTypePtr & new_variant_type) -{ - const DataTypes & current_variants = assert_cast(variant_info.variant_type.get())->getVariants(); + const DataTypes & current_variants = assert_cast(old_variant_type.get())->getVariants(); const DataTypes & new_variants = assert_cast(new_variant_type.get())->getVariants(); - Names new_variant_names; - new_variant_names.reserve(new_variants.size()); - std::unordered_map new_variant_name_to_discriminator; - new_variant_name_to_discriminator.reserve(new_variants.size()); std::vector> new_variant_columns_and_discriminators_to_add; new_variant_columns_and_discriminators_to_add.reserve(new_variants.size() - current_variants.size()); std::vector current_to_new_discriminators; @@ -108,26 +149,26 @@ void ColumnDynamic::updateVariantInfoAndExpandVariantColumn(const DB::DataTypePt for (ColumnVariant::Discriminator discr = 0; discr != new_variants.size(); ++discr) { - const auto & name = new_variant_names.emplace_back(new_variants[discr]->getName()); - new_variant_name_to_discriminator[name] = discr; - - auto current_it = variant_info.variant_name_to_discriminator.find(name); - if (current_it == variant_info.variant_name_to_discriminator.end()) + auto current_it = old_variant_name_to_discriminator.find(new_variants[discr]->getName()); + if (current_it == old_variant_name_to_discriminator.end()) new_variant_columns_and_discriminators_to_add.emplace_back(new_variants[discr]->createColumn(), discr); else current_to_new_discriminators[current_it->second] = discr; } - variant_info.variant_type = new_variant_type; - variant_info.variant_name = new_variant_type->getName(); - variant_info.variant_names = new_variant_names; - variant_info.variant_name_to_discriminator = new_variant_name_to_discriminator; - assert_cast(*variant_column).extend(current_to_new_discriminators, std::move(new_variant_columns_and_discriminators_to_add)); + assert_cast(variant_column).extend(current_to_new_discriminators, std::move(new_variant_columns_and_discriminators_to_add)); +} + +void ColumnDynamic::updateVariantInfoAndExpandVariantColumn(const DataTypePtr & new_variant_type) +{ + extendVariantColumn(*variant_column, variant_info.variant_type, new_variant_type, variant_info.variant_name_to_discriminator); + createVariantInfo(new_variant_type); + /// Clear mappings cache because now with new Variant we will have new mappings. variant_mappings_cache.clear(); } -std::vector * ColumnDynamic::combineVariants(const DB::ColumnDynamic::VariantInfo & other_variant_info) +std::vector * ColumnDynamic::combineVariants(const ColumnDynamic::VariantInfo & other_variant_info) { /// Check if we already have global discriminators mapping for other Variant in cache. /// It's used to not calculate the same mapping each call of insertFrom with the same columns. @@ -154,20 +195,13 @@ std::vector * ColumnDynamic::combineVariants(const const DataTypes & current_variants = assert_cast(*variant_info.variant_type).getVariants(); /// We cannot combine Variants if total number of variants exceeds max_dynamic_types. - if (current_variants.size() + num_new_variants > max_dynamic_types) + if (!canAddNewVariants(num_new_variants)) { /// Remember that we cannot combine our variant with this one, so we will not try to do it again. variants_with_failed_combination.insert(other_variant_info.variant_name); return nullptr; } - /// We cannot combine Variants if total number of variants reaches max_dynamic_types and we don't have String variant. - if (current_variants.size() + num_new_variants == max_dynamic_types && !variant_info.variant_name_to_discriminator.contains("String") && !other_variant_info.variant_name_to_discriminator.contains("String")) - { - variants_with_failed_combination.insert(other_variant_info.variant_name); - return nullptr; - } - DataTypes all_variants = current_variants; all_variants.insert(all_variants.end(), other_variants.begin(), other_variants.end()); auto new_variant_type = std::make_shared(all_variants); @@ -185,40 +219,93 @@ std::vector * ColumnDynamic::combineVariants(const return &it->second; } -void ColumnDynamic::insert(const DB::Field & x) +void ColumnDynamic::insert(const Field & x) { - /// Check if we can insert field without Variant extension. - if (variant_column->tryInsert(x)) + if (x.isNull()) + { + insertDefault(); return; + } + + auto & variant_col = getVariantColumn(); + auto shared_variant_discr = getSharedVariantDiscriminator(); + /// Check if we can insert field into existing variants and avoid Variant extension. + for (size_t i = 0; i != variant_col.getNumVariants(); ++i) + { + if (i != shared_variant_discr && variant_col.getVariantByGlobalDiscriminator(i).tryInsert(x)) + { + variant_col.getLocalDiscriminators().push_back(variant_col.localDiscriminatorByGlobal(i)); + variant_col.getOffsets().push_back(variant_col.getVariantByGlobalDiscriminator(i).size() - 1); + return; + } + } /// If we cannot insert field into current variant column, extend it with new variant for this field from its type. - if (addNewVariant(applyVisitor(FieldToDataType(), x))) + auto field_data_type = applyVisitor(FieldToDataType(), x); + auto field_data_type_name = field_data_type->getName(); + if (addNewVariant(field_data_type, field_data_type_name)) { - /// Now we should be able to insert this field into extended variant column. - variant_column->insert(x); + /// Insert this field into newly added variant. + auto discr = variant_info.variant_name_to_discriminator[field_data_type_name]; + variant_col.getVariantByGlobalDiscriminator(discr).insert(x); + variant_col.getLocalDiscriminators().push_back(variant_col.localDiscriminatorByGlobal(discr)); + variant_col.getOffsets().push_back(variant_col.getVariantByGlobalDiscriminator(discr).size() - 1); } else { /// We reached maximum number of variants and couldn't add new variant. - /// This case should be really rare in real use cases. - /// We should always be able to add String variant and cast inserted value to String. - addStringVariant(); - variant_column->insert(toString(x)); + /// In this case we add the value of this new variant into special shared variant. + /// We store values in shared variant in binary form with binary encoded type. + auto & shared_variant = getSharedVariant(); + auto & chars = shared_variant.getChars(); + WriteBufferFromVector value_buf(chars, AppendModeTag()); + encodeDataType(field_data_type, value_buf); + getVariantSerialization(field_data_type, field_data_type_name)->serializeBinary(x, value_buf, getFormatSettings()); + value_buf.finalize(); + chars.push_back(0); + shared_variant.getOffsets().push_back(chars.size()); + variant_col.getLocalDiscriminators().push_back(variant_col.localDiscriminatorByGlobal(shared_variant_discr)); + variant_col.getOffsets().push_back(shared_variant.size() - 1); } } -bool ColumnDynamic::tryInsert(const DB::Field & x) +bool ColumnDynamic::tryInsert(const Field & x) { /// We can insert any value into Dynamic column. insert(x); return true; } +Field ColumnDynamic::operator[](size_t n) const +{ + Field res; + get(n, res); + return res; +} + +void ColumnDynamic::get(size_t n, Field & res) const +{ + const auto & variant_col = getVariantColumn(); + /// Check if value is not in shared variant. + if (variant_col.globalDiscriminatorAt(n) != getSharedVariantDiscriminator()) + { + variant_col.get(n, res); + return; + } + + /// We should deeserialize value from shared variant. + const auto & shared_variant = getSharedVariant(); + auto value_data = shared_variant.getDataAt(variant_col.offsetAt(n)); + ReadBufferFromMemory buf(value_data.data, value_data.size); + auto type = decodeDataType(buf); + getVariantSerialization(type)->deserializeBinary(res, buf, getFormatSettings()); +} + #if !defined(DEBUG_OR_SANITIZER_BUILD) -void ColumnDynamic::insertFrom(const DB::IColumn & src_, size_t n) +void ColumnDynamic::insertFrom(const IColumn & src_, size_t n) #else -void ColumnDynamic::doInsertFrom(const DB::IColumn & src_, size_t n) +void ColumnDynamic::doInsertFrom(const IColumn & src_, size_t n) #endif { const auto & dynamic_src = assert_cast(src_); @@ -231,6 +318,28 @@ void ColumnDynamic::doInsertFrom(const DB::IColumn & src_, size_t n) } auto & variant_col = assert_cast(*variant_column); + const auto & src_variant_col = assert_cast(*dynamic_src.variant_column); + auto src_global_discr = src_variant_col.globalDiscriminatorAt(n); + auto src_offset = src_variant_col.offsetAt(n); + + /// Check if we insert from shared variant and process it separately. + if (src_global_discr == dynamic_src.getSharedVariantDiscriminator()) + { + const auto & src_shared_variant = dynamic_src.getSharedVariant(); + auto value = src_shared_variant.getDataAt(src_offset); + /// Decode data type of this value. + ReadBufferFromMemory buf(value.data, value.size); + auto type = decodeDataType(buf); + auto type_name = type->getName(); + /// Check if we have this variant and deserialize value into variant from shared variant data. + if (auto it = variant_info.variant_name_to_discriminator.find(type_name); it != variant_info.variant_name_to_discriminator.end()) + variant_col.deserializeBinaryIntoVariant(it->second, getVariantSerialization(type, type_name), buf, getFormatSettings()); + /// Otherwise just insert it into our shared variant. + else + variant_col.insertIntoVariantFrom(getSharedVariantDiscriminator(), src_shared_variant, src_offset); + + return; + } /// If variants are different, we need to extend our variant with new variants. if (auto * global_discriminators_mapping = combineVariants(dynamic_src.variant_info)) @@ -241,8 +350,6 @@ void ColumnDynamic::doInsertFrom(const DB::IColumn & src_, size_t n) /// We cannot combine 2 Variant types as total number of variants exceeds the limit. /// We need to insert single value, try to add only corresponding variant. - const auto & src_variant_col = assert_cast(*dynamic_src.variant_column); - auto src_global_discr = src_variant_col.globalDiscriminatorAt(n); /// NULL doesn't require Variant extension. if (src_global_discr == ColumnVariant::NULL_DISCRIMINATOR) @@ -260,19 +367,18 @@ void ColumnDynamic::doInsertFrom(const DB::IColumn & src_, size_t n) } /// We reached maximum number of variants and couldn't add new variant. - /// We should always be able to add String variant and cast inserted value to String. - addStringVariant(); - auto tmp_variant_column = src_variant_col.getVariantByGlobalDiscriminator(src_global_discr).cloneEmpty(); - tmp_variant_column->insertFrom(src_variant_col.getVariantByGlobalDiscriminator(src_global_discr), src_variant_col.offsetAt(n)); - auto tmp_string_column = castColumn(ColumnWithTypeAndName(tmp_variant_column->getPtr(), variant_type, ""), std::make_shared()); - auto string_variant_discr = variant_info.variant_name_to_discriminator["String"]; - variant_col.insertIntoVariantFrom(string_variant_discr, *tmp_string_column, 0); + /// Insert this value into shared variant. + insertValueIntoSharedVariant( + src_variant_col.getVariantByGlobalDiscriminator(src_global_discr), + variant_type, + dynamic_src.variant_info.variant_names[src_global_discr], + src_offset); } #if !defined(DEBUG_OR_SANITIZER_BUILD) -void ColumnDynamic::insertRangeFrom(const DB::IColumn & src_, size_t start, size_t length) +void ColumnDynamic::insertRangeFrom(const IColumn & src_, size_t start, size_t length) #else -void ColumnDynamic::doInsertRangeFrom(const DB::IColumn & src_, size_t start, size_t length) +void ColumnDynamic::doInsertRangeFrom(const IColumn & src_, size_t start, size_t length) #endif { if (start + length > src_.size()) @@ -293,156 +399,206 @@ void ColumnDynamic::doInsertRangeFrom(const DB::IColumn & src_, size_t start, si /// If variants are different, we need to extend our variant with new variants. if (auto * global_discriminators_mapping = combineVariants(dynamic_src.variant_info)) { - variant_col.insertRangeFrom(*dynamic_src.variant_column, start, length, *global_discriminators_mapping); + size_t prev_size = variant_col.size(); + auto shared_variant_discr = getSharedVariantDiscriminator(); + variant_col.insertRangeFrom(*dynamic_src.variant_column, start, length, *global_discriminators_mapping, shared_variant_discr); + + /// We should process insertion from src shared variant separately, because it can contain + /// values that should be extracted into our variants. insertRangeFrom above didn't insert + /// values into our shared variant (we specified shared_variant_discr as special skip discriminator). + + /// Check if src shared variant is empty, nothing to do in this case. + if (dynamic_src.getSharedVariant().empty()) + return; + + /// Iterate over src discriminators and process insertion from src shared variant. + const auto & src_variant_column = dynamic_src.getVariantColumn(); + const auto src_shared_variant_discr = dynamic_src.getSharedVariantDiscriminator(); + const auto src_shared_variant_local_discr = src_variant_column.localDiscriminatorByGlobal(src_shared_variant_discr); + const auto & src_local_discriminators = src_variant_column.getLocalDiscriminators(); + const auto & src_offsets = src_variant_column.getOffsets(); + const auto & src_shared_variant = assert_cast(src_variant_column.getVariantByLocalDiscriminator(src_shared_variant_local_discr)); + + auto & local_discriminators = variant_col.getLocalDiscriminators(); + auto & offsets = variant_col.getOffsets(); + const auto shared_variant_local_discr = variant_col.localDiscriminatorByGlobal(shared_variant_discr); + auto & shared_variant = assert_cast(variant_col.getVariantByLocalDiscriminator(shared_variant_local_discr)); + for (size_t i = 0; i != length; ++i) + { + if (src_local_discriminators[start + i] == src_shared_variant_local_discr) + { + chassert(local_discriminators[prev_size + i] == shared_variant_local_discr); + auto value = src_shared_variant.getDataAt(src_offsets[start + i]); + ReadBufferFromMemory buf(value.data, value.size); + auto type = decodeDataType(buf); + auto type_name = type->getName(); + /// Check if we have variant with this type. In this case we should extract + /// the value from src shared variant and insert it into this variant. + if (auto it = variant_info.variant_name_to_discriminator.find(type_name); it != variant_info.variant_name_to_discriminator.end()) + { + auto local_discr = variant_col.localDiscriminatorByGlobal(it->second); + auto & variant = variant_col.getVariantByLocalDiscriminator(local_discr); + getVariantSerialization(type, type_name)->deserializeBinary(variant, buf, getFormatSettings()); + /// Local discriminators were already filled in ColumnVariant::insertRangeFrom and this row should contain + /// shared_variant_local_discr. Change it to local discriminator of the found variant and update offsets. + local_discriminators[prev_size + i] = local_discr; + offsets[prev_size + i] = variant.size() - 1; + } + /// Otherwise, insert this value into shared variant. + else + { + shared_variant.insertData(value.data, value.size); + /// Update variant offset. + offsets[prev_size + i] = shared_variant.size() - 1; + } + } + } + return; } /// We cannot combine 2 Variant types as total number of variants exceeds the limit. - /// In this case we will add most frequent variants from this range and insert them as usual, - /// all other variants will be converted to String. - /// TODO: instead of keeping all current variants and just adding new most frequent variants - /// from source columns we can also try to replace rarest existing variants with frequent - /// variants from source column (so we will avoid casting new frequent variants to String - /// and keeping rare existing ones). It will require rewriting of existing data in Variant - /// column but will improve usability of Dynamic column for example during squashing blocks - /// during insert. - - const auto & src_variant_column = dynamic_src.getVariantColumn(); - - /// Calculate ranges for each variant in current range. - std::vector> variants_ranges(dynamic_src.variant_info.variant_names.size(), {0, 0}); - /// If we insert the whole column, no need to iterate through the range, we can just take variant sizes. - if (start == 0 && length == dynamic_src.size()) - { - for (size_t i = 0; i != dynamic_src.variant_info.variant_names.size(); ++i) - variants_ranges[i] = {0, src_variant_column.getVariantByGlobalDiscriminator(i).size()}; - } - /// Otherwise we need to iterate through discriminators and calculate the range for each variant. - else - { - const auto & local_discriminators = src_variant_column.getLocalDiscriminators(); - const auto & offsets = src_variant_column.getOffsets(); - size_t end = start + length; - for (size_t i = start; i != end; ++i) - { - auto discr = src_variant_column.globalDiscriminatorByLocal(local_discriminators[i]); - if (discr != ColumnVariant::NULL_DISCRIMINATOR) - { - if (!variants_ranges[discr].second) - variants_ranges[discr].first = offsets[i]; - ++variants_ranges[discr].second; - } - } - } - + /// In this case we will add most frequent variants and insert them as usual, + /// all other variants will be inserted into shared variant. const auto & src_variants = assert_cast(*dynamic_src.variant_info.variant_type).getVariants(); - /// List of variants that will be converted to String. - std::vector variants_to_convert_to_string; /// Mapping from global discriminators of src_variant to the new variant we will create. std::vector other_to_new_discriminators; other_to_new_discriminators.reserve(dynamic_src.variant_info.variant_names.size()); - /// Check if we cannot add any more new variants. In this case we will convert all new variants to String. - if (variant_info.variant_names.size() == max_dynamic_types || (variant_info.variant_names.size() == max_dynamic_types - 1 && !variant_info.variant_name_to_discriminator.contains("String"))) + /// Check if we cannot add any more new variants. In this case we will insert all new variants into shared variant. + if (!canAddNewVariant()) { - addStringVariant(); - for (size_t i = 0; i != dynamic_src.variant_info.variant_names.size(); ++i) - { - auto it = variant_info.variant_name_to_discriminator.find(dynamic_src.variant_info.variant_names[i]); - if (it == variant_info.variant_name_to_discriminator.end()) - { - variants_to_convert_to_string.push_back(i); - other_to_new_discriminators.push_back(variant_info.variant_name_to_discriminator["String"]); - } - else - { - other_to_new_discriminators.push_back(it->second); - } - } - } - /// We still can add some new variants, but not all of them. Let's choose the most frequent variants in specified range. - else - { - std::vector> new_variants_with_sizes; - new_variants_with_sizes.reserve(dynamic_src.variant_info.variant_names.size()); - for (size_t i = 0; i != dynamic_src.variant_info.variant_names.size(); ++i) - { - const auto & variant_name = dynamic_src.variant_info.variant_names[i]; - if (variant_name != "String" && !variant_info.variant_name_to_discriminator.contains(variant_name)) - new_variants_with_sizes.emplace_back(variants_ranges[i].second, i); - } - - std::sort(new_variants_with_sizes.begin(), new_variants_with_sizes.end(), std::greater()); - DataTypes new_variants = assert_cast(*variant_info.variant_type).getVariants(); - if (!variant_info.variant_name_to_discriminator.contains("String")) - new_variants.push_back(std::make_shared()); - - for (const auto & [_, discr] : new_variants_with_sizes) - { - if (new_variants.size() != max_dynamic_types) - new_variants.push_back(src_variants[discr]); - else - variants_to_convert_to_string.push_back(discr); - } - - auto new_variant_type = std::make_shared(new_variants); - updateVariantInfoAndExpandVariantColumn(new_variant_type); - auto string_variant_discriminator = variant_info.variant_name_to_discriminator.at("String"); + auto shared_variant_discr = getSharedVariantDiscriminator(); for (const auto & variant_name : dynamic_src.variant_info.variant_names) { auto it = variant_info.variant_name_to_discriminator.find(variant_name); if (it == variant_info.variant_name_to_discriminator.end()) - other_to_new_discriminators.push_back(string_variant_discriminator); + other_to_new_discriminators.push_back(shared_variant_discr); + else + other_to_new_discriminators.push_back(it->second); + } + } + /// We still can add some new variants, but not all of them. Let's choose the most frequent variants. + else + { + /// Create list of pairs and sort it. + std::vector> new_variants_with_sizes; + new_variants_with_sizes.reserve(dynamic_src.variant_info.variant_names.size()); + const auto & src_variant_column = dynamic_src.getVariantColumn(); + for (const auto & [name, discr] : dynamic_src.variant_info.variant_name_to_discriminator) + { + if (!variant_info.variant_name_to_discriminator.contains(name)) + new_variants_with_sizes.emplace_back(src_variant_column.getVariantByGlobalDiscriminator(discr).size(), discr); + } + + std::sort(new_variants_with_sizes.begin(), new_variants_with_sizes.end(), std::greater()); + DataTypes new_variants = assert_cast(*variant_info.variant_type).getVariants(); + /// Add new variants from sorted list until we reach max_dynamic_types. + for (const auto & [_, discr] : new_variants_with_sizes) + { + if (!canAddNewVariant(new_variants.size())) + break; + new_variants.push_back(src_variants[discr]); + } + + auto new_variant_type = std::make_shared(new_variants); + updateVariantInfoAndExpandVariantColumn(new_variant_type); + auto shared_variant_discr = getSharedVariantDiscriminator(); + for (const auto & variant_name : dynamic_src.variant_info.variant_names) + { + auto it = variant_info.variant_name_to_discriminator.find(variant_name); + if (it == variant_info.variant_name_to_discriminator.end()) + other_to_new_discriminators.push_back(shared_variant_discr); else other_to_new_discriminators.push_back(it->second); } } - /// Convert to String all variants that couldn't be added. - std::unordered_map variants_converted_to_string; - variants_converted_to_string.reserve(variants_to_convert_to_string.size()); - for (auto discr : variants_to_convert_to_string) - { - auto [variant_start, variant_length] = variants_ranges[discr]; - const auto & variant = src_variant_column.getVariantPtrByGlobalDiscriminator(discr); - if (variant_start == 0 && variant_length == variant->size()) - variants_converted_to_string[discr] = castColumn(ColumnWithTypeAndName(variant, src_variants[discr], ""), std::make_shared()); - else - variants_converted_to_string[discr] = castColumn(ColumnWithTypeAndName(variant->cut(variant_start, variant_length), src_variants[discr], ""), std::make_shared()); - } - + /// Iterate over the range and perform insertion. + const auto & src_variant_column = dynamic_src.getVariantColumn(); const auto & src_local_discriminators = src_variant_column.getLocalDiscriminators(); const auto & src_offsets = src_variant_column.getOffsets(); const auto & src_variant_columns = src_variant_column.getVariants(); + const auto src_shared_variant_discr = dynamic_src.getSharedVariantDiscriminator(); + const auto src_shared_variant_local_discr = src_variant_column.localDiscriminatorByGlobal(src_shared_variant_discr); + const auto & src_shared_variant = assert_cast(*src_variant_columns[src_shared_variant_local_discr]); + auto & local_discriminators = variant_col.getLocalDiscriminators(); + local_discriminators.reserve(local_discriminators.size() + length); + auto & offsets = variant_col.getOffsets(); + offsets.reserve(offsets.size() + length); + auto & variant_columns = variant_col.getVariants(); + const auto shared_variant_discr = getSharedVariantDiscriminator(); + const auto shared_variant_local_discr = variant_col.localDiscriminatorByGlobal(shared_variant_discr); + auto & shared_variant = assert_cast(*variant_columns[shared_variant_local_discr]); size_t end = start + length; for (size_t i = start; i != end; ++i) { - auto local_discr = src_local_discriminators[i]; - if (local_discr == ColumnVariant::NULL_DISCRIMINATOR) + auto src_local_discr = src_local_discriminators[i]; + auto src_offset = src_offsets[i]; + if (src_local_discr == ColumnVariant::NULL_DISCRIMINATOR) { - variant_col.insertDefault(); + local_discriminators.push_back(ColumnVariant::NULL_DISCRIMINATOR); + offsets.emplace_back(); } else { - auto global_discr = src_variant_column.globalDiscriminatorByLocal(local_discr); - auto to_global_discr = other_to_new_discriminators[global_discr]; - auto it = variants_converted_to_string.find(global_discr); - if (it == variants_converted_to_string.end()) + /// Process insertion from src shared variant separately. + if (src_local_discr == src_shared_variant_local_discr) { - variant_col.insertIntoVariantFrom(to_global_discr, *src_variant_columns[local_discr], src_offsets[i]); + auto value = src_shared_variant.getDataAt(src_offset); + ReadBufferFromMemory buf(value.data, value.size); + auto type = decodeDataType(buf); + auto type_name = type->getName(); + /// Check if we have variant with this type. In this case we should extract + /// the value from src shared variant and insert it into this variant. + if (auto it = variant_info.variant_name_to_discriminator.find(type_name); it != variant_info.variant_name_to_discriminator.end()) + { + auto local_discr = variant_col.localDiscriminatorByGlobal(it->second); + getVariantSerialization(type, type_name)->deserializeBinary(*variant_columns[local_discr], buf, getFormatSettings()); + local_discriminators.push_back(local_discr); + offsets.push_back(variant_columns[local_discr]->size() - 1); + } + /// Otherwise, insert this value into shared variant. + else + { + shared_variant.insertData(value.data, value.size); + local_discriminators.push_back(shared_variant_local_discr); + offsets.push_back(shared_variant.size() - 1); + } } + /// Insertion from usual variant. else { - variant_col.insertIntoVariantFrom(to_global_discr, *it->second, src_offsets[i] - variants_ranges[global_discr].first); + auto src_global_discr = src_variant_column.globalDiscriminatorByLocal(src_local_discr); + auto global_discr = other_to_new_discriminators[src_global_discr]; + /// Check if we need to insert this value into shared variant. + if (global_discr == shared_variant_discr) + { + serializeValueIntoSharedVariant( + shared_variant, + *src_variant_columns[src_local_discr], + src_variants[src_global_discr], + getVariantSerialization(src_variants[src_global_discr], dynamic_src.variant_info.variant_names[src_global_discr]), + src_offset); + local_discriminators.push_back(shared_variant_local_discr); + offsets.push_back(shared_variant.size() - 1); + } + else + { + auto local_discr = variant_col.localDiscriminatorByGlobal(global_discr); + variant_columns[local_discr]->insertFrom(*src_variant_columns[src_local_discr], src_offset); + local_discriminators.push_back(local_discr); + offsets.push_back(variant_columns[local_discr]->size() - 1); + } } } } } #if !defined(DEBUG_OR_SANITIZER_BUILD) -void ColumnDynamic::insertManyFrom(const DB::IColumn & src_, size_t position, size_t length) +void ColumnDynamic::insertManyFrom(const IColumn & src_, size_t position, size_t length) #else -void ColumnDynamic::doInsertManyFrom(const DB::IColumn & src_, size_t position, size_t length) +void ColumnDynamic::doInsertManyFrom(const IColumn & src_, size_t position, size_t length) #endif { const auto & dynamic_src = assert_cast(src_); @@ -455,6 +611,36 @@ void ColumnDynamic::doInsertManyFrom(const DB::IColumn & src_, size_t position, } auto & variant_col = assert_cast(*variant_column); + const auto & src_variant_col = assert_cast(*dynamic_src.variant_column); + auto src_global_discr = src_variant_col.globalDiscriminatorAt(position); + auto src_offset = src_variant_col.offsetAt(position); + + /// Check if we insert from shared variant and process it separately. + if (src_global_discr == dynamic_src.getSharedVariantDiscriminator()) + { + const auto & src_shared_variant = dynamic_src.getSharedVariant(); + auto value = src_shared_variant.getDataAt(src_offset); + /// Decode data type of this value. + ReadBufferFromMemory buf(value.data, value.size); + auto type = decodeDataType(buf); + auto type_name = type->getName(); + /// Check if we have this variant and deserialize value into variant from shared variant data. + if (auto it = variant_info.variant_name_to_discriminator.find(type_name); it != variant_info.variant_name_to_discriminator.end()) + { + /// Deserialize value into temporary column and use it in insertManyIntoVariantFrom. + auto tmp_column = type->createColumn(); + tmp_column->reserve(1); + getVariantSerialization(type, type_name)->deserializeBinary(*tmp_column, buf, getFormatSettings()); + variant_col.insertManyIntoVariantFrom(it->second, *tmp_column, 0, length); + } + /// Otherwise just insert it into our shared variant. + else + { + variant_col.insertManyIntoVariantFrom(getSharedVariantDiscriminator(), src_shared_variant, src_offset, length); + } + + return; + } /// If variants are different, we need to extend our variant with new variants. if (auto * global_discriminators_mapping = combineVariants(dynamic_src.variant_info)) @@ -465,8 +651,6 @@ void ColumnDynamic::doInsertManyFrom(const DB::IColumn & src_, size_t position, /// We cannot combine 2 Variant types as total number of variants exceeds the limit. /// We need to insert single value, try to add only corresponding variant. - const auto & src_variant_col = assert_cast(*dynamic_src.variant_column); - auto src_global_discr = src_variant_col.globalDiscriminatorAt(position); if (src_global_discr == ColumnVariant::NULL_DISCRIMINATOR) { insertDefault(); @@ -481,21 +665,51 @@ void ColumnDynamic::doInsertManyFrom(const DB::IColumn & src_, size_t position, return; } - addStringVariant(); - auto tmp_variant_column = src_variant_col.getVariantByGlobalDiscriminator(src_global_discr).cloneEmpty(); - tmp_variant_column->insertFrom(src_variant_col.getVariantByGlobalDiscriminator(src_global_discr), src_variant_col.offsetAt(position)); - auto tmp_string_column = castColumn(ColumnWithTypeAndName(tmp_variant_column->getPtr(), variant_type, ""), std::make_shared()); - auto string_variant_discr = variant_info.variant_name_to_discriminator["String"]; - variant_col.insertManyIntoVariantFrom(string_variant_discr, *tmp_string_column, 0, length); + /// We reached maximum number of variants and couldn't add new variant. + /// Insert this value into shared variant. + /// Create temporary string column, serialize value into it and use it in insertManyIntoVariantFrom. + auto tmp_shared_variant = ColumnString::create(); + serializeValueIntoSharedVariant( + *tmp_shared_variant, + src_variant_col.getVariantByGlobalDiscriminator(src_global_discr), + variant_type, + getVariantSerialization(variant_type, dynamic_src.variant_info.variant_names[src_global_discr]), + src_offset); + + variant_col.insertManyIntoVariantFrom(getSharedVariantDiscriminator(), *tmp_shared_variant, 0, length); } +void ColumnDynamic::insertValueIntoSharedVariant(const IColumn & src, const DataTypePtr & type, const String & type_name, size_t n) +{ + auto & variant_col = getVariantColumn(); + auto & shared_variant = getSharedVariant(); + serializeValueIntoSharedVariant(shared_variant, src, type, getVariantSerialization(type, type_name), n); + variant_col.getLocalDiscriminators().push_back(variant_col.localDiscriminatorByGlobal(getSharedVariantDiscriminator())); + variant_col.getOffsets().push_back(shared_variant.size() - 1); +} -StringRef ColumnDynamic::serializeValueIntoArena(size_t n, DB::Arena & arena, const char *& begin) const +void ColumnDynamic::serializeValueIntoSharedVariant( + ColumnString & shared_variant, + const IColumn & src, + const DataTypePtr & type, + const SerializationPtr & serialization, + size_t n) +{ + auto & chars = shared_variant.getChars(); + WriteBufferFromVector value_buf(chars, AppendModeTag()); + encodeDataType(type, value_buf); + serialization->serializeBinary(src, n, value_buf, getFormatSettings()); + value_buf.finalize(); + chars.push_back(0); + shared_variant.getOffsets().push_back(chars.size()); +} + +StringRef ColumnDynamic::serializeValueIntoArena(size_t n, Arena & arena, const char *& begin) const { /// 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 in binary format (size + bytes) + value. + /// Instead, we serialize null bit + variant type and value in binary format (size + data). const auto & variant_col = assert_cast(*variant_column); auto discr = variant_col.globalDiscriminatorAt(n); StringRef res; @@ -509,19 +723,29 @@ StringRef ColumnDynamic::serializeValueIntoArena(size_t n, DB::Arena & arena, co return res; } - 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_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_type_binary_data.size(); + WriteBufferFromOwnString buf; + StringRef type_and_value; + /// If we have value from shared variant, it's already stored in the desired format. + if (discr == getSharedVariantDiscriminator()) + { + type_and_value = getSharedVariant().getDataAt(variant_col.offsetAt(n)); + } + /// For regular variants serialize its type and value in binary format. + else + { + const auto & variant_type = assert_cast(*variant_info.variant_type).getVariant(discr); + encodeDataType(variant_type, buf); + getVariantSerialization(variant_type, variant_info.variant_names[discr]) + ->serializeBinary(variant_col.getVariantByGlobalDiscriminator(discr), variant_col.offsetAt(n), buf, getFormatSettings()); + type_and_value = buf.str(); + } - auto value_ref = variant_col.getVariantByGlobalDiscriminator(discr).serializeValueIntoArena(variant_col.offsetAt(n), arena, begin); - res.data = value_ref.data - res.size; - res.size += value_ref.size; + char * pos = arena.allocContinue(sizeof(UInt8) + sizeof(size_t) + type_and_value.size, begin); + memcpy(pos, &null_bit, sizeof(UInt8)); + memcpy(pos + sizeof(UInt8), &type_and_value.size, sizeof(size_t)); + memcpy(pos + sizeof(UInt8) + sizeof(size_t), type_and_value.data, type_and_value.size); + res.data = pos; + res.size = sizeof(UInt8) + sizeof(size_t) + type_and_value.size; return res; } @@ -536,39 +760,36 @@ const char * ColumnDynamic::deserializeAndInsertFromArena(const char * pos) return pos; } - /// 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); + /// Read variant type and value in binary format. + const size_t type_and_value_size = unalignedLoad(pos); + pos += sizeof(type_and_value_size); + std::string_view type_and_value(pos, type_and_value_size); + pos += type_and_value_size; + + ReadBufferFromMemory buf(type_and_value.data(), type_and_value.size()); + auto variant_type = decodeDataType(buf); 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()) { - auto discr = it->second; - return variant_col.deserializeVariantAndInsertFromArena(discr, pos); + variant_col.deserializeBinaryIntoVariant(it->second, getVariantSerialization(variant_type, variant_name), buf, getFormatSettings()); } - - /// If we don't have such variant, add it. - if (likely(addNewVariant(variant_type))) + /// If we don't have such variant, try to add it. + else if (likely(addNewVariant(variant_type))) { auto discr = variant_info.variant_name_to_discriminator[variant_name]; - return variant_col.deserializeVariantAndInsertFromArena(discr, pos); + variant_col.deserializeBinaryIntoVariant(discr, getVariantSerialization(variant_type, variant_name), buf, getFormatSettings()); + } + /// Otherwise insert this value into shared variant. + else + { + auto & shared_variant = getSharedVariant(); + shared_variant.insertData(type_and_value.data(), type_and_value.size()); + variant_col.getLocalDiscriminators().push_back(variant_col.localDiscriminatorByGlobal(getSharedVariantDiscriminator())); + variant_col.getOffsets().push_back(shared_variant.size() - 1); } - /// We reached maximum number of variants and couldn't add new variant. - /// We should always be able to add String variant and cast inserted value to String. - addStringVariant(); - /// Create temporary column of this variant type and deserialize value into it. - auto tmp_variant_column = variant_type->createColumn(); - pos = tmp_variant_column->deserializeAndInsertFromArena(pos); - /// Cast temporary column to String and insert this value into String variant. - auto str_column = castColumn(ColumnWithTypeAndName(tmp_variant_column->getPtr(), variant_type, ""), std::make_shared()); - variant_col.insertIntoVariantFrom(variant_info.variant_name_to_discriminator["String"], *str_column, 0); return pos; } @@ -579,14 +800,10 @@ const char * ColumnDynamic::skipSerializedInArena(const char * pos) const if (null_bit) return pos; - 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); + const size_t type_and_value_size = unalignedLoad(pos); + pos += sizeof(type_and_value_size); + pos += type_and_value_size; + return pos; } void ColumnDynamic::updateHashWithValue(size_t n, SipHash & hash) const @@ -604,9 +821,9 @@ void ColumnDynamic::updateHashWithValue(size_t n, SipHash & hash) const } #if !defined(DEBUG_OR_SANITIZER_BUILD) -int ColumnDynamic::compareAt(size_t n, size_t m, const DB::IColumn & rhs, int nan_direction_hint) const +int ColumnDynamic::compareAt(size_t n, size_t m, const 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 +int ColumnDynamic::doCompareAt(size_t n, size_t m, const IColumn & rhs, int nan_direction_hint) const #endif { const auto & left_variant = assert_cast(*variant_column); @@ -614,7 +831,9 @@ int ColumnDynamic::doCompareAt(size_t n, size_t m, const DB::IColumn & rhs, int const auto & right_variant = assert_cast(*right_dynamic.variant_column); auto left_discr = left_variant.globalDiscriminatorAt(n); + auto left_shared_variant_discr = getSharedVariantDiscriminator(); auto right_discr = right_variant.globalDiscriminatorAt(m); + auto right_shared_variant_discr = right_dynamic.getSharedVariantDiscriminator(); /// Check if we have NULLs and return result based on nan_direction_hint. if (left_discr == ColumnVariant::NULL_DISCRIMINATOR && right_discr == ColumnVariant::NULL_DISCRIMINATOR) @@ -624,12 +843,129 @@ int ColumnDynamic::doCompareAt(size_t n, size_t m, const DB::IColumn & rhs, int else if (right_discr == ColumnVariant::NULL_DISCRIMINATOR) return -nan_direction_hint; - /// If rows have different types, we compare type names. - if (variant_info.variant_names[left_discr] != right_dynamic.variant_info.variant_names[right_discr]) - return variant_info.variant_names[left_discr] < right_dynamic.variant_info.variant_names[right_discr] ? -1 : 1; + /// Check if both values are in shared variant. + if (left_discr == left_shared_variant_discr && right_discr == right_shared_variant_discr) + { + /// First check if both type and value are equal. + auto left_value = getSharedVariant().getDataAt(left_variant.offsetAt(n)); + auto right_value = right_dynamic.getSharedVariant().getDataAt(right_variant.offsetAt(m)); + if (left_value == right_value) + return 0; - /// If rows have the same types, compare actual values from corresponding variants. - return left_variant.getVariantByGlobalDiscriminator(left_discr).compareAt(left_variant.offsetAt(n), right_variant.offsetAt(m), right_variant.getVariantByGlobalDiscriminator(right_discr), nan_direction_hint); + /// Extract type names from both values. + ReadBufferFromMemory buf_left(left_value.data, left_value.size); + auto left_data_type = decodeDataType(buf_left); + auto left_data_type_name = left_data_type->getName(); + + ReadBufferFromMemory buf_right(right_value.data, right_value.size); + auto right_data_type = decodeDataType(buf_right); + auto right_data_type_name = right_data_type->getName(); + + /// If rows have different types, we compare type names. + if (left_data_type_name != right_data_type_name) + return left_data_type_name < right_data_type_name ? -1 : 1; + + /// If rows have the same type, we compare actual values. + /// We have both values serialized in binary format, so we need to + /// create temporary column, insert both values into it and compare. + auto tmp_column = left_data_type->createColumn(); + const auto & serialization = getVariantSerialization(left_data_type, left_data_type_name); + serialization->deserializeBinary(*tmp_column, buf_left, getFormatSettings()); + serialization->deserializeBinary(*tmp_column, buf_right, getFormatSettings()); + return tmp_column->compareAt(0, 1, *tmp_column, nan_direction_hint); + } + /// Check if only left value is in shared data. + else if (left_discr == left_shared_variant_discr) + { + /// Extract left type name from the value. + auto left_value = getSharedVariant().getDataAt(left_variant.offsetAt(n)); + ReadBufferFromMemory buf_left(left_value.data, left_value.size); + auto left_data_type = decodeDataType(buf_left); + auto left_data_type_name = left_data_type->getName(); + + /// If rows have different types, we compare type names. + if (left_data_type_name != right_dynamic.variant_info.variant_names[right_discr]) + return left_data_type_name < right_dynamic.variant_info.variant_names[right_discr] ? -1 : 1; + + /// If rows have the same type, we compare actual values. + /// We have left value serialized in binary format, we need to + /// create temporary column, insert the value into it and compare. + auto tmp_column = left_data_type->createColumn(); + getVariantSerialization(left_data_type, left_data_type_name)->deserializeBinary(*tmp_column, buf_left, getFormatSettings()); + return tmp_column->compareAt(0, right_variant.offsetAt(m), right_variant.getVariantByGlobalDiscriminator(right_discr), nan_direction_hint); + } + /// Check if only right value is in shared data. + else if (right_discr == right_shared_variant_discr) + { + /// Extract right type name from the value. + auto right_value = right_dynamic.getSharedVariant().getDataAt(right_variant.offsetAt(m)); + ReadBufferFromMemory buf_right(right_value.data, right_value.size); + auto right_data_type = decodeDataType(buf_right); + auto right_data_type_name = right_data_type->getName(); + + /// If rows have different types, we compare type names. + if (variant_info.variant_names[left_discr] != right_data_type_name) + return variant_info.variant_names[left_discr] < right_data_type_name ? -1 : 1; + + /// If rows have the same type, we compare actual values. + /// We have right value serialized in binary format, we need to + /// create temporary column, insert the value into it and compare. + auto tmp_column = right_data_type->createColumn(); + getVariantSerialization(right_data_type, right_data_type_name)->deserializeBinary(*tmp_column, buf_right, getFormatSettings()); + return left_variant.getVariantByGlobalDiscriminator(left_discr).compareAt(left_variant.offsetAt(n), 0, *tmp_column, nan_direction_hint); + } + /// Otherwise both values are regular variants. + else + { + /// If rows have different types, we compare type names. + if (variant_info.variant_names[left_discr] != right_dynamic.variant_info.variant_names[right_discr]) + return variant_info.variant_names[left_discr] < right_dynamic.variant_info.variant_names[right_discr] ? -1 : 1; + + /// If rows have the same types, compare actual values from corresponding variants. + return left_variant.getVariantByGlobalDiscriminator(left_discr).compareAt(left_variant.offsetAt(n), right_variant.offsetAt(m), right_variant.getVariantByGlobalDiscriminator(right_discr), nan_direction_hint); + } +} + +struct ColumnDynamic::ComparatorBase +{ + const ColumnDynamic & parent; + int nan_direction_hint; + + ComparatorBase(const ColumnDynamic & parent_, int nan_direction_hint_) + : parent(parent_), nan_direction_hint(nan_direction_hint_) + { + } + + ALWAYS_INLINE int compare(size_t lhs, size_t rhs) const + { + return parent.compareAt(lhs, rhs, parent, nan_direction_hint); + } +}; + +void ColumnDynamic::getPermutation(IColumn::PermutationSortDirection direction, IColumn::PermutationSortStability stability, size_t limit, int nan_direction_hint, IColumn::Permutation & res) const +{ + if (direction == IColumn::PermutationSortDirection::Ascending && stability == IColumn::PermutationSortStability::Unstable) + getPermutationImpl(limit, res, ComparatorAscendingUnstable(*this, nan_direction_hint), DefaultSort(), DefaultPartialSort()); + else if (direction == IColumn::PermutationSortDirection::Ascending && stability == IColumn::PermutationSortStability::Stable) + getPermutationImpl(limit, res, ComparatorAscendingStable(*this, nan_direction_hint), DefaultSort(), DefaultPartialSort()); + else if (direction == IColumn::PermutationSortDirection::Descending && stability == IColumn::PermutationSortStability::Unstable) + getPermutationImpl(limit, res, ComparatorDescendingUnstable(*this, nan_direction_hint), DefaultSort(), DefaultPartialSort()); + else if (direction == IColumn::PermutationSortDirection::Descending && stability == IColumn::PermutationSortStability::Stable) + getPermutationImpl(limit, res, ComparatorDescendingStable(*this, nan_direction_hint), DefaultSort(), DefaultPartialSort()); +} + +void ColumnDynamic::updatePermutation(IColumn::PermutationSortDirection direction, IColumn::PermutationSortStability stability, size_t limit, int nan_direction_hint, IColumn::Permutation & res, DB::EqualRanges & equal_ranges) const +{ + auto comparator_equal = ComparatorEqual(*this, nan_direction_hint); + + if (direction == IColumn::PermutationSortDirection::Ascending && stability == IColumn::PermutationSortStability::Unstable) + updatePermutationImpl(limit, res, equal_ranges, ComparatorAscendingUnstable(*this, nan_direction_hint), comparator_equal, DefaultSort(), DefaultPartialSort()); + else if (direction == IColumn::PermutationSortDirection::Ascending && stability == IColumn::PermutationSortStability::Stable) + updatePermutationImpl(limit, res, equal_ranges, ComparatorAscendingStable(*this, nan_direction_hint), comparator_equal, DefaultSort(), DefaultPartialSort()); + else if (direction == IColumn::PermutationSortDirection::Descending && stability == IColumn::PermutationSortStability::Unstable) + updatePermutationImpl(limit, res, equal_ranges, ComparatorDescendingUnstable(*this, nan_direction_hint), comparator_equal, DefaultSort(), DefaultPartialSort()); + else if (direction == IColumn::PermutationSortDirection::Descending && stability == IColumn::PermutationSortStability::Stable) + updatePermutationImpl(limit, res, equal_ranges, ComparatorDescendingStable(*this, nan_direction_hint), comparator_equal, DefaultSort(), DefaultPartialSort()); } ColumnPtr ColumnDynamic::compress() const @@ -637,12 +973,123 @@ ColumnPtr ColumnDynamic::compress() const ColumnPtr variant_compressed = variant_column->compress(); size_t byte_size = variant_compressed->byteSize(); return ColumnCompressed::create(size(), byte_size, - [my_variant_compressed = std::move(variant_compressed), my_variant_info = variant_info, my_max_dynamic_types = max_dynamic_types, my_statistics = statistics]() mutable + [my_variant_compressed = std::move(variant_compressed), my_variant_info = variant_info, my_max_dynamic_types = max_dynamic_types, my_global_max_dynamic_types = global_max_dynamic_types, my_statistics = statistics]() mutable { - return ColumnDynamic::create(my_variant_compressed->decompress(), my_variant_info, my_max_dynamic_types, my_statistics); + return ColumnDynamic::create(my_variant_compressed->decompress(), my_variant_info, my_max_dynamic_types, my_global_max_dynamic_types, my_statistics); }); } +void ColumnDynamic::prepareForSquashing(const Columns & source_columns) +{ + if (source_columns.empty()) + return; + + /// Internal variants of source dynamic columns may differ. + /// We want to preallocate memory for all variants we will have after squashing. + /// It may happen that the total number of variants in source columns will + /// exceed the limit, in this case we will choose the most frequent variants + /// and insert the rest types into the shared variant. + + /// First, preallocate memory for variant discriminators and offsets. + size_t new_size = size(); + for (const auto & source_column : source_columns) + new_size += source_column->size(); + auto & variant_col = getVariantColumn(); + variant_col.getLocalDiscriminators().reserve_exact(new_size); + variant_col.getOffsets().reserve_exact(new_size); + + /// Second, collect all variants and their total sizes. + std::unordered_map total_variant_sizes; + DataTypes all_variants; + + auto add_variants = [&](const ColumnDynamic & source_dynamic) + { + const auto & source_variant_column = source_dynamic.getVariantColumn(); + const auto & source_variant_info = source_dynamic.getVariantInfo(); + const auto & source_variants = assert_cast(*source_variant_info.variant_type).getVariants(); + + for (size_t i = 0; i != source_variants.size(); ++i) + { + const auto & variant_name = source_variant_info.variant_names[i]; + auto it = total_variant_sizes.find(variant_name); + /// Add this variant to the list of all variants if we didn't see it yet. + if (it == total_variant_sizes.end()) + { + all_variants.push_back(source_variants[i]); + it = total_variant_sizes.emplace(variant_name, 0).first; + } + + it->second += source_variant_column.getVariantByGlobalDiscriminator(i).size(); + } + }; + + for (const auto & source_column : source_columns) + add_variants(assert_cast(*source_column)); + + /// Add variants from this dynamic column. + add_variants(*this); + + DataTypePtr result_variant_type; + /// Check if the number of all variants exceeds the limit. + if (!canAddNewVariants(0, all_variants.size())) + { + /// We want to keep the most frequent variants in the resulting dynamic column. + DataTypes result_variants; + result_variants.reserve(max_dynamic_types + 1); /// +1 for shared variant. + /// Add variants from current variant column as we will not rewrite it. + for (const auto & variant : assert_cast(*variant_info.variant_type).getVariants()) + result_variants.push_back(variant); + + /// Create list of remaining variants with their sizes and sort it. + std::vector> variants_with_sizes; + variants_with_sizes.reserve(all_variants.size() - variant_info.variant_names.size()); + for (const auto & variant : all_variants) + { + /// Add variant to the list only of we didn't add it yet. + auto variant_name = variant->getName(); + if (!variant_info.variant_name_to_discriminator.contains(variant_name)) + variants_with_sizes.emplace_back(total_variant_sizes[variant_name], variant); + } + + std::sort(variants_with_sizes.begin(), variants_with_sizes.end(), std::greater()); + /// Add the most frequent variants until we reach max_dynamic_types. + for (const auto & [_, new_variant] : variants_with_sizes) + { + if (!canAddNewVariant(result_variants.size())) + break; + result_variants.push_back(new_variant); + } + + result_variant_type = std::make_shared(result_variants); + } + else + { + result_variant_type = std::make_shared(all_variants); + } + + if (!result_variant_type->equals(*variant_info.variant_type)) + updateVariantInfoAndExpandVariantColumn(result_variant_type); + + /// Now current dynamic column has all resulting variants and we can call + /// prepareForSquashing on them to preallocate the memory. + for (size_t i = 0; i != variant_info.variant_names.size(); ++i) + { + Columns source_variant_columns; + source_variant_columns.reserve(source_columns.size()); + for (const auto & source_column : source_columns) + { + const auto & source_dynamic_column = assert_cast(*source_column); + const auto & source_variant_info = source_dynamic_column.getVariantInfo(); + /// Try to find this variant in the current source column. + auto it = source_variant_info.variant_name_to_discriminator.find(variant_info.variant_names[i]); + if (it != source_variant_info.variant_name_to_discriminator.end()) + source_variant_columns.push_back(source_dynamic_column.getVariantColumn().getVariantPtrByGlobalDiscriminator(it->second)); + } + + variant_col.getVariantByGlobalDiscriminator(i).prepareForSquashing(source_variant_columns); + } +} + void ColumnDynamic::takeDynamicStructureFromSourceColumns(const Columns & source_columns) { if (!empty()) @@ -663,6 +1110,9 @@ void ColumnDynamic::takeDynamicStructureFromSourceColumns(const Columns & source /// First, collect all variants from all source columns and calculate total sizes. std::unordered_map total_sizes; DataTypes all_variants; + /// Add shared variant type in advance; + all_variants.push_back(getSharedVariantDataType()); + total_sizes[getSharedVariantTypeName()] = 0; for (const auto & source_column : source_columns) { @@ -671,7 +1121,7 @@ void ColumnDynamic::takeDynamicStructureFromSourceColumns(const Columns & source const auto & source_variant_info = source_dynamic.getVariantInfo(); const auto & source_variants = assert_cast(*source_variant_info.variant_type).getVariants(); /// During deserialization from MergeTree we will have variant sizes statistics from the whole data part. - const auto & source_statistics = source_dynamic.getStatistics(); + const auto & source_statistics = source_dynamic.getStatistics(); for (size_t i = 0; i != source_variants.size(); ++i) { const auto & variant_name = source_variant_info.variant_names[i]; @@ -682,35 +1132,67 @@ void ColumnDynamic::takeDynamicStructureFromSourceColumns(const Columns & source all_variants.push_back(source_variants[i]); it = total_sizes.emplace(variant_name, 0).first; } - auto statistics_it = source_statistics.data.find(variant_name); - size_t size = statistics_it == source_statistics.data.end() ? source_variant_column.getVariantByGlobalDiscriminator(i).size() : statistics_it->second; + size_t size = source_variant_column.getVariantByGlobalDiscriminator(i).size(); + if (source_statistics) + { + auto statistics_it = source_statistics->variants_statistics.find(variant_name); + if (statistics_it != source_statistics->variants_statistics.end()) + size = statistics_it->second; + } + it->second += size; } + + /// Use add variants from shared variant statistics. It can help extracting + /// frequent variants from shared variant to usual variants. + if (source_statistics) + { + for (const auto & [variant_name, size] : source_statistics->shared_variants_statistics) + { + auto it = total_sizes.find(variant_name); + /// Add this variant to the list of all variants if we didn't see it yet. + if (it == total_sizes.end()) + { + all_variants.push_back(DataTypeFactory::instance().get(variant_name)); + it = total_sizes.emplace(variant_name, 0).first; + } + it->second += size; + } + } } DataTypePtr result_variant_type; - /// Check if the number of all variants exceeds the limit. - if (all_variants.size() > max_dynamic_types || (all_variants.size() == max_dynamic_types && !total_sizes.contains("String"))) + Statistics new_statistics(Statistics::Source::MERGE); + /// Reset max_dynamic_types to global_max_dynamic_types. + max_dynamic_types = global_max_dynamic_types; + /// Check if the number of all dynamic types exceeds the limit. + if (!canAddNewVariants(0, all_variants.size())) { /// Create list of variants with their sizes and sort it. std::vector> variants_with_sizes; variants_with_sizes.reserve(all_variants.size()); for (const auto & variant : all_variants) - variants_with_sizes.emplace_back(total_sizes[variant->getName()], variant); + { + if (variant->getName() != getSharedVariantTypeName()) + variants_with_sizes.emplace_back(total_sizes[variant->getName()], variant); + } std::sort(variants_with_sizes.begin(), variants_with_sizes.end(), std::greater()); - /// Take first max_dynamic_types variants from sorted list. + /// Take first max_dynamic_types variants from sorted list and fill shared_variants_statistics with the rest. DataTypes result_variants; - result_variants.reserve(max_dynamic_types); - /// Add String variant in advance. - result_variants.push_back(std::make_shared()); - for (const auto & [_, variant] : variants_with_sizes) + result_variants.reserve(max_dynamic_types + 1); /// +1 for shared variant. + /// Add shared variant. + result_variants.push_back(getSharedVariantDataType()); + for (const auto & [size, variant] : variants_with_sizes) { - if (result_variants.size() == max_dynamic_types) - break; - - if (variant->getName() != "String") + /// Add variant to the resulting variants list until we reach max_dynamic_types. + if (canAddNewVariant(result_variants.size())) result_variants.push_back(variant); + /// Add all remaining variants into shared_variants_statistics until we reach its max size. + else if (new_statistics.shared_variants_statistics.size() < Statistics::MAX_SHARED_VARIANT_STATISTICS_SIZE) + new_statistics.shared_variants_statistics[variant->getName()] = size; + else + break; } result_variant_type = std::make_shared(result_variants); @@ -720,26 +1202,17 @@ void ColumnDynamic::takeDynamicStructureFromSourceColumns(const Columns & source result_variant_type = std::make_shared(all_variants); } - /// Now we have resulting Variant and can fill variant info. - variant_info.variant_type = result_variant_type; - variant_info.variant_name = result_variant_type->getName(); - const auto & result_variants = assert_cast(*result_variant_type).getVariants(); - variant_info.variant_names.clear(); - variant_info.variant_names.reserve(result_variants.size()); - variant_info.variant_name_to_discriminator.clear(); - variant_info.variant_name_to_discriminator.reserve(result_variants.size()); - statistics.data.clear(); - statistics.data.reserve(result_variants.size()); - statistics.source = Statistics::Source::MERGE; - for (size_t i = 0; i != result_variants.size(); ++i) - { - auto variant_name = result_variants[i]->getName(); - variant_info.variant_names.push_back(variant_name); - variant_info.variant_name_to_discriminator[variant_name] = i; - statistics.data[variant_name] = total_sizes[variant_name]; - } + /// Now we have resulting Variant and can fill variant info and create merge statistics. + setVariantType(result_variant_type); + new_statistics.variants_statistics.reserve(variant_info.variant_names.size()); + for (const auto & variant_name : variant_info.variant_names) + new_statistics.variants_statistics[variant_name] = total_sizes[variant_name]; + statistics = std::make_shared(std::move(new_statistics)); - variant_column = variant_info.variant_type->createColumn(); + /// Reduce max_dynamic_types to the number of selected variants, so there will be no possibility + /// to extend selected variants on inerts into this column during merges. + /// -1 because we don't count shared variant in the limit. + max_dynamic_types = variant_info.variant_names.size() - 1; /// Now we have the resulting Variant that will be used in all merged columns. /// Variants can also contain Dynamic columns inside, we should collect @@ -755,7 +1228,7 @@ void ColumnDynamic::takeDynamicStructureFromSourceColumns(const Columns & source { /// Try to find this variant in current source column. auto it = source_variant_info.variant_name_to_discriminator.find(variant_info.variant_names[i]); - if (it != source_variant_info.variant_name_to_discriminator.end()) + if (it != source_variant_info.variant_name_to_discriminator.end()) /// Add shared variant. variants_source_columns[i].push_back(source_dynamic_column.getVariantColumn().getVariantPtrByGlobalDiscriminator(it->second)); } } diff --git a/src/Columns/ColumnDynamic.h b/src/Columns/ColumnDynamic.h index e92cabd3db9..e6e720765f6 100644 --- a/src/Columns/ColumnDynamic.h +++ b/src/Columns/ColumnDynamic.h @@ -3,6 +3,7 @@ #include #include #include +#include #include #include @@ -19,11 +20,19 @@ namespace DB * * When new values are inserted into Dynamic column, the internal Variant * type and column are extended if the inserted value has new type. + * When the limit on number of dynamic types is exceeded, all values + * with new types are inserted into special shared variant with type String + * that contains values and their types in binary format. */ class ColumnDynamic final : public COWHelper, ColumnDynamic> { public: - /// + /// Maximum limit on dynamic types. We use ColumnVariant to store all the types, + /// so the limit cannot be greater then ColumnVariant::MAX_NESTED_COLUMNS. + /// We also always have reserved variant for shared variant. + static constexpr size_t MAX_DYNAMIC_TYPES_LIMIT = ColumnVariant::MAX_NESTED_COLUMNS - 1; + static constexpr const char * SHARED_VARIANT_TYPE_NAME = "SharedVariant"; + struct Statistics { enum class Source @@ -32,12 +41,27 @@ public: MERGE, /// Statistics were calculated during merge of several MergeTree parts. }; + explicit Statistics(Source source_) : source(source_) {} + /// Source of the statistics. Source source; - /// Statistics data: (variant name) -> (total variant size in data part). - std::unordered_map data; + /// Statistics data for usual variants: (variant name) -> (total variant size in data part). + std::unordered_map variants_statistics; + /// Statistics data for variants from shared variant: (variant name) -> (total variant size in data part). + /// For shared variant we store statistics only for first 256 variants (should cover almost all cases and it's not expensive). + static constexpr const size_t MAX_SHARED_VARIANT_STATISTICS_SIZE = 256; + std::unordered_map shared_variants_statistics; }; + using StatisticsPtr = std::shared_ptr; + + struct ComparatorBase; + using ComparatorAscendingUnstable = ComparatorAscendingUnstableImpl; + using ComparatorAscendingStable = ComparatorAscendingStableImpl; + using ComparatorDescendingUnstable = ComparatorDescendingUnstableImpl; + using ComparatorDescendingStable = ComparatorDescendingStableImpl; + using ComparatorEqual = ComparatorEqualImpl; + private: friend class COWHelper, ColumnDynamic>; @@ -54,28 +78,32 @@ private: }; explicit ColumnDynamic(size_t max_dynamic_types_); - ColumnDynamic(MutableColumnPtr variant_column_, const VariantInfo & variant_info_, size_t max_dynamic_types_, const Statistics & statistics_ = {}); + ColumnDynamic(MutableColumnPtr variant_column_, const DataTypePtr & variant_type_, size_t max_dynamic_types_, size_t global_max_dynamic_types_, const StatisticsPtr & statistics_ = {}); + ColumnDynamic(MutableColumnPtr variant_column_, const VariantInfo & variant_info_, size_t max_dynamic_types_, size_t global_max_dynamic_types_, const StatisticsPtr & statistics_ = {}); public: /** Create immutable column using immutable arguments. This arguments may be shared with other columns. * Use IColumn::mutate in order to make mutable column and mutate shared nested columns. */ using Base = COWHelper, ColumnDynamic>; - static Ptr create(const ColumnPtr & variant_column_, const VariantInfo & variant_info_, size_t max_dynamic_types_, const Statistics & statistics_ = {}) + static Ptr create(const ColumnPtr & variant_column_, const VariantInfo & variant_info_, size_t max_dynamic_types_, size_t global_max_dynamic_types_, const StatisticsPtr & statistics_ = {}) { - return ColumnDynamic::create(variant_column_->assumeMutable(), variant_info_, max_dynamic_types_, statistics_); + return ColumnDynamic::create(variant_column_->assumeMutable(), variant_info_, max_dynamic_types_, global_max_dynamic_types_, statistics_); } - static MutablePtr create(MutableColumnPtr variant_column_, const VariantInfo & variant_info_, size_t max_dynamic_types_, const Statistics & statistics_ = {}) + static MutablePtr create(MutableColumnPtr variant_column_, const VariantInfo & variant_info_, size_t max_dynamic_types_, size_t global_max_dynamic_types_, const StatisticsPtr & statistics_ = {}) { - return Base::create(std::move(variant_column_), variant_info_, max_dynamic_types_, statistics_); + return Base::create(std::move(variant_column_), variant_info_, max_dynamic_types_, global_max_dynamic_types_, statistics_); } - static MutablePtr create(MutableColumnPtr variant_column_, const DataTypePtr & variant_type, size_t max_dynamic_types_, const Statistics & statistics_ = {}); - - static ColumnPtr create(ColumnPtr variant_column_, const DataTypePtr & variant_type, size_t max_dynamic_types_, const Statistics & statistics_ = {}) + static MutablePtr create(MutableColumnPtr variant_column_, const DataTypePtr & variant_type_, size_t max_dynamic_types_, size_t global_max_dynamic_types_, const StatisticsPtr & statistics_ = {}) { - return create(variant_column_->assumeMutable(), variant_type, max_dynamic_types_, statistics_); + return Base::create(std::move(variant_column_), variant_type_, max_dynamic_types_, global_max_dynamic_types_, statistics_); + } + + static ColumnPtr create(ColumnPtr variant_column_, const DataTypePtr & variant_type, size_t max_dynamic_types_, size_t global_max_dynamic_types_, const StatisticsPtr & statistics_ = {}) + { + return create(variant_column_->assumeMutable(), variant_type, max_dynamic_types_, global_max_dynamic_types_, statistics_); } static MutablePtr create(size_t max_dynamic_types_) @@ -83,7 +111,7 @@ public: return Base::create(max_dynamic_types_); } - std::string getName() const override { return "Dynamic(max_types=" + std::to_string(max_dynamic_types) + ")"; } + std::string getName() const override { return "Dynamic(max_types=" + std::to_string(global_max_dynamic_types) + ")"; } const char * getFamilyName() const override { @@ -98,12 +126,12 @@ public: MutableColumnPtr cloneEmpty() const override { /// Keep current dynamic structure - return Base::create(variant_column->cloneEmpty(), variant_info, max_dynamic_types, statistics); + return Base::create(variant_column->cloneEmpty(), variant_info, max_dynamic_types, global_max_dynamic_types, statistics); } MutableColumnPtr cloneResized(size_t size) const override { - return Base::create(variant_column->cloneResized(size), variant_info, max_dynamic_types, statistics); + return Base::create(variant_column->cloneResized(size), variant_info, max_dynamic_types, global_max_dynamic_types, statistics); } size_t size() const override @@ -111,15 +139,9 @@ public: return variant_column->size(); } - Field operator[](size_t n) const override - { - return (*variant_column)[n]; - } + Field operator[](size_t n) const override; - void get(size_t n, Field & res) const override - { - variant_column->get(n, res); - } + void get(size_t n, Field & res) const override; bool isDefaultAt(size_t n) const override { @@ -187,7 +209,7 @@ public: ColumnPtr filter(const Filter & filt, ssize_t result_size_hint) const override { - return create(variant_column->filter(filt, result_size_hint), variant_info, max_dynamic_types); + return create(variant_column->filter(filt, result_size_hint), variant_info, max_dynamic_types, global_max_dynamic_types); } void expand(const Filter & mask, bool inverted) override @@ -197,17 +219,17 @@ public: ColumnPtr permute(const Permutation & perm, size_t limit) const override { - return create(variant_column->permute(perm, limit), variant_info, max_dynamic_types); + return create(variant_column->permute(perm, limit), variant_info, max_dynamic_types, global_max_dynamic_types); } ColumnPtr index(const IColumn & indexes, size_t limit) const override { - return create(variant_column->index(indexes, limit), variant_info, max_dynamic_types); + return create(variant_column->index(indexes, limit), variant_info, max_dynamic_types, global_max_dynamic_types); } ColumnPtr replicate(const Offsets & replicate_offsets) const override { - return create(variant_column->replicate(replicate_offsets), variant_info, max_dynamic_types); + return create(variant_column->replicate(replicate_offsets), variant_info, max_dynamic_types, global_max_dynamic_types); } MutableColumns scatter(ColumnIndex num_columns, const Selector & selector) const override @@ -216,7 +238,7 @@ public: MutableColumns scattered_columns; scattered_columns.reserve(num_columns); for (auto & scattered_variant_column : scattered_variant_columns) - scattered_columns.emplace_back(create(std::move(scattered_variant_column), variant_info, max_dynamic_types)); + scattered_columns.emplace_back(create(std::move(scattered_variant_column), variant_info, max_dynamic_types, global_max_dynamic_types)); return scattered_columns; } @@ -238,22 +260,18 @@ public: } void getPermutation(IColumn::PermutationSortDirection direction, IColumn::PermutationSortStability stability, - size_t limit, int nan_direction_hint, IColumn::Permutation & res) const override - { - variant_column->getPermutation(direction, stability, limit, nan_direction_hint, res); - } + size_t limit, int nan_direction_hint, IColumn::Permutation & res) const override; void updatePermutation(IColumn::PermutationSortDirection direction, IColumn::PermutationSortStability stability, - size_t limit, int nan_direction_hint, IColumn::Permutation & res, EqualRanges & equal_ranges) const override - { - variant_column->updatePermutation(direction, stability, limit, nan_direction_hint, res, equal_ranges); - } + size_t limit, int nan_direction_hint, IColumn::Permutation & res, EqualRanges & equal_ranges) const override; void reserve(size_t n) override { variant_column->reserve(n); } + void prepareForSquashing(const Columns & source_columns) override; + void ensureOwnership() override { variant_column->ensureOwnership(); @@ -293,7 +311,7 @@ public: bool structureEquals(const IColumn & rhs) const override { if (const auto * rhs_concrete = typeid_cast(&rhs)) - return max_dynamic_types == rhs_concrete->max_dynamic_types; + return global_max_dynamic_types == rhs_concrete->global_max_dynamic_types; return false; } @@ -336,17 +354,75 @@ public: const ColumnVariant & getVariantColumn() const { return assert_cast(*variant_column); } ColumnVariant & getVariantColumn() { return assert_cast(*variant_column); } - bool addNewVariant(const DataTypePtr & new_variant); - void addStringVariant(); + bool addNewVariant(const DataTypePtr & new_variant, const String & new_variant_name); + bool addNewVariant(const DataTypePtr & new_variant) { return addNewVariant(new_variant, new_variant->getName()); } bool hasDynamicStructure() const override { return true; } void takeDynamicStructureFromSourceColumns(const Columns & source_columns) override; - const Statistics & getStatistics() const { return statistics; } + const StatisticsPtr & getStatistics() const { return statistics; } + void setStatistics(const StatisticsPtr & statistics_) { statistics = statistics_; } size_t getMaxDynamicTypes() const { return max_dynamic_types; } + /// Check if we can add new variant types. + /// Shared variant doesn't count in the limit but always presents, + /// so we should subtract 1 from the total types count. + bool canAddNewVariants(size_t current_variants_count, size_t new_variants_count) const { return current_variants_count + new_variants_count - 1 <= max_dynamic_types; } + bool canAddNewVariant(size_t current_variants_count) const { return canAddNewVariants(current_variants_count, 1); } + bool canAddNewVariants(size_t new_variants_count) const { return canAddNewVariants(variant_info.variant_names.size(), new_variants_count); } + bool canAddNewVariant() const { return canAddNewVariants(variant_info.variant_names.size(), 1); } + + void setVariantType(const DataTypePtr & variant_type); + void setMaxDynamicPaths(size_t max_dynamic_type_); + + static const String & getSharedVariantTypeName() + { + static const String name = SHARED_VARIANT_TYPE_NAME; + return name; + } + + static DataTypePtr getSharedVariantDataType(); + + ColumnVariant::Discriminator getSharedVariantDiscriminator() const + { + return variant_info.variant_name_to_discriminator.at(getSharedVariantTypeName()); + } + + ColumnString & getSharedVariant() + { + return assert_cast(getVariantColumn().getVariantByGlobalDiscriminator(getSharedVariantDiscriminator())); + } + + const ColumnString & getSharedVariant() const + { + return assert_cast(getVariantColumn().getVariantByGlobalDiscriminator(getSharedVariantDiscriminator())); + } + + /// Serializes type and value in binary format into provided shared variant. Doesn't update Variant discriminators and offsets. + static void serializeValueIntoSharedVariant(ColumnString & shared_variant, const IColumn & src, const DataTypePtr & type, const SerializationPtr & serialization, size_t n); + + /// Insert value into shared variant. Also updates Variant discriminators and offsets. + void insertValueIntoSharedVariant(const IColumn & src, const DataTypePtr & type, const String & type_name, size_t n); + + const SerializationPtr & getVariantSerialization(const DataTypePtr & variant_type, const String & variant_name) const + { + /// Get serialization for provided data type. + /// To avoid calling type->getDefaultSerialization() every time we use simple cache with max size. + /// When max size is reached, just clear the cache. + if (serialization_cache.size() == SERIALIZATION_CACHE_MAX_SIZE) + serialization_cache.clear(); + + if (auto it = serialization_cache.find(variant_name); it != serialization_cache.end()) + return it->second; + + return serialization_cache.emplace(variant_name, variant_type->getDefaultSerialization()).first->second; + } + + const SerializationPtr & getVariantSerialization(const DataTypePtr & variant_type) const { return getVariantSerialization(variant_type, variant_type->getName()); } private: + void createVariantInfo(const DataTypePtr & variant_type); + /// Combine current variant with the other variant and return global discriminators mapping /// from other variant to the combined one. It's used for inserting from /// different variants. @@ -359,12 +435,19 @@ private: /// Store the type of current variant with some additional information. VariantInfo variant_info; /// The maximum number of different types that can be stored in this Dynamic column. - /// If exceeded, all new variants will be converted to String. + /// If exceeded, all new variants will be added to a special shared variant with type String + /// in binary format. This limit can be different for different instances of Dynamic column. + /// When max_dynamic_types = 0, we will have only shared variant and insert all values into it. size_t max_dynamic_types; + /// The types limit specified in the data type by the user Dynamic(max_types=N). + /// max_dynamic_types in all column instances of this Dynamic type can be only smaller + /// (for example, max_dynamic_types can be reduced in takeDynamicStructureFromSourceColumns + /// before merge of different Dynamic columns). + size_t global_max_dynamic_types; /// Size statistics of each variants from MergeTree data part. /// Used in takeDynamicStructureFromSourceColumns and set during deserialization. - Statistics statistics; + StatisticsPtr statistics; /// Cache (Variant name) -> (global discriminators mapping from this variant to current variant in Dynamic column). /// Used to avoid mappings recalculation in combineVariants for the same Variant types. @@ -372,6 +455,17 @@ private: /// Cache of Variant types that couldn't be combined with current variant in Dynamic column. /// Used to avoid checking if combination is possible for the same Variant types. std::unordered_set variants_with_failed_combination; + + /// We can use serializations of different data types to serialize values into shared variant. + /// To avoid creating the same serialization multiple times, use simple cache. + static const size_t SERIALIZATION_CACHE_MAX_SIZE = 256; + mutable std::unordered_map serialization_cache; }; +void extendVariantColumn( + IColumn & variant_column, + const DataTypePtr & old_variant_type, + const DataTypePtr & new_variant_type, + std::unordered_map old_variant_name_to_discriminator); + } diff --git a/src/Columns/ColumnFixedString.cpp b/src/Columns/ColumnFixedString.cpp index 0bb3f7edb14..04e894ee5ab 100644 --- a/src/Columns/ColumnFixedString.cpp +++ b/src/Columns/ColumnFixedString.cpp @@ -59,7 +59,7 @@ bool ColumnFixedString::isDefaultAt(size_t index) const void ColumnFixedString::insert(const Field & x) { - const String & s = x.get(); + const String & s = x.safeGet(); insertData(s.data(), s.size()); } @@ -67,7 +67,7 @@ bool ColumnFixedString::tryInsert(const Field & x) { if (x.getType() != Field::Types::Which::String) return false; - const String & s = x.get(); + const String & s = x.safeGet(); if (s.size() > n) return false; insertData(s.data(), s.size()); diff --git a/src/Columns/ColumnLowCardinality.h b/src/Columns/ColumnLowCardinality.h index 3766b247d60..d9c90e7dc21 100644 --- a/src/Columns/ColumnLowCardinality.h +++ b/src/Columns/ColumnLowCardinality.h @@ -46,8 +46,8 @@ public: return Base::create(std::move(column_unique), std::move(indexes), is_shared); } - std::string getName() const override { return "ColumnLowCardinality"; } - const char * getFamilyName() const override { return "ColumnLowCardinality"; } + std::string getName() const override { return "LowCardinality(" + getDictionary().getNestedColumn()->getName() + ")"; } + const char * getFamilyName() const override { return "LowCardinality"; } TypeIndex getDataType() const override { return TypeIndex::LowCardinality; } ColumnPtr convertToFullColumn() const { return getDictionary().getNestedColumn()->index(getIndexes(), 0); } diff --git a/src/Columns/ColumnMap.cpp b/src/Columns/ColumnMap.cpp index 1025b4e77b9..651b073fef7 100644 --- a/src/Columns/ColumnMap.cpp +++ b/src/Columns/ColumnMap.cpp @@ -72,7 +72,7 @@ void ColumnMap::get(size_t n, Field & res) const size_t size = offsets[n] - offsets[n - 1]; res = Map(); - auto & map = res.get(); + auto & map = res.safeGet(); map.reserve(size); for (size_t i = 0; i < size; ++i) @@ -96,7 +96,7 @@ void ColumnMap::insertData(const char *, size_t) void ColumnMap::insert(const Field & x) { - const auto & map = x.get(); + const auto & map = x.safeGet(); nested->insert(Array(map.begin(), map.end())); } @@ -105,7 +105,7 @@ bool ColumnMap::tryInsert(const Field & x) if (x.getType() != Field::Types::Which::Map) return false; - const auto & map = x.get(); + const auto & map = x.safeGet(); return nested->tryInsert(Array(map.begin(), map.end())); } @@ -249,6 +249,15 @@ void ColumnMap::reserve(size_t n) nested->reserve(n); } +void ColumnMap::prepareForSquashing(const Columns & source_columns) +{ + Columns nested_source_columns; + nested_source_columns.reserve(source_columns.size()); + for (const auto & source_column : source_columns) + nested_source_columns.push_back(assert_cast(*source_column).getNestedColumnPtr()); + nested->prepareForSquashing(nested_source_columns); +} + void ColumnMap::shrinkToFit() { nested->shrinkToFit(); @@ -288,8 +297,8 @@ void ColumnMap::getExtremes(Field & min, Field & max) const /// Convert result Array fields to Map fields because client expect min and max field to have type Map - Array nested_min_value = nested_min.get(); - Array nested_max_value = nested_max.get(); + Array nested_min_value = nested_min.safeGet(); + Array nested_max_value = nested_max.safeGet(); Map map_min_value(nested_min_value.begin(), nested_min_value.end()); Map map_max_value(nested_max_value.begin(), nested_max_value.end()); diff --git a/src/Columns/ColumnMap.h b/src/Columns/ColumnMap.h index 3eaaa0ad562..191476839f1 100644 --- a/src/Columns/ColumnMap.h +++ b/src/Columns/ColumnMap.h @@ -94,6 +94,7 @@ public: void updatePermutation(IColumn::PermutationSortDirection direction, IColumn::PermutationSortStability stability, size_t limit, int nan_direction_hint, IColumn::Permutation & res, EqualRanges & equal_ranges) const override; void reserve(size_t n) override; + void prepareForSquashing(const Columns & source_columns) override; void shrinkToFit() override; void ensureOwnership() override; size_t byteSize() const override; diff --git a/src/Columns/ColumnNullable.cpp b/src/Columns/ColumnNullable.cpp index 6529f0b78db..2a25cac6461 100644 --- a/src/Columns/ColumnNullable.cpp +++ b/src/Columns/ColumnNullable.cpp @@ -706,6 +706,22 @@ void ColumnNullable::reserve(size_t n) getNullMapData().reserve(n); } +void ColumnNullable::prepareForSquashing(const Columns & source_columns) +{ + size_t new_size = size(); + Columns nested_source_columns; + nested_source_columns.reserve(source_columns.size()); + for (const auto & source_column : source_columns) + { + const auto & source_nullable_column = assert_cast(*source_column); + new_size += source_nullable_column.size(); + nested_source_columns.push_back(source_nullable_column.getNestedColumnPtr()); + } + + nested_column->prepareForSquashing(nested_source_columns); + getNullMapData().reserve(new_size); +} + void ColumnNullable::shrinkToFit() { getNestedColumn().shrinkToFit(); diff --git a/src/Columns/ColumnNullable.h b/src/Columns/ColumnNullable.h index fe9f5b6dcc2..2c32e0fe5a0 100644 --- a/src/Columns/ColumnNullable.h +++ b/src/Columns/ColumnNullable.h @@ -125,6 +125,7 @@ public: size_t limit, int null_direction_hint, Permutation & res, EqualRanges& equal_ranges) const override; size_t estimateCardinalityInPermutedRange(const Permutation & permutation, const EqualRange & equal_range) const override; void reserve(size_t n) override; + void prepareForSquashing(const Columns & source_columns) override; void shrinkToFit() override; void ensureOwnership() override; size_t byteSize() const override; diff --git a/src/Columns/ColumnObject.cpp b/src/Columns/ColumnObject.cpp index a6431007cb6..eb99bb4081b 100644 --- a/src/Columns/ColumnObject.cpp +++ b/src/Columns/ColumnObject.cpp @@ -698,7 +698,7 @@ void ColumnObject::forEachSubcolumnRecursively(RecursiveMutableColumnCallback ca void ColumnObject::insert(const Field & field) { - const auto & object = field.get(); + const auto & object = field.safeGet(); HashSet inserted_paths; size_t old_size = size(); @@ -754,7 +754,7 @@ void ColumnObject::get(size_t n, Field & res) const { assert(n < size()); res = Object(); - auto & object = res.get(); + auto & object = res.safeGet(); for (const auto & entry : subcolumns) { diff --git a/src/Columns/ColumnString.cpp b/src/Columns/ColumnString.cpp index 7cfa2571f5a..9ed2c7e3d4d 100644 --- a/src/Columns/ColumnString.cpp +++ b/src/Columns/ColumnString.cpp @@ -557,6 +557,21 @@ void ColumnString::reserve(size_t n) offsets.reserve_exact(n); } +void ColumnString::prepareForSquashing(const Columns & source_columns) +{ + size_t new_size = size(); + size_t new_chars_size = chars.size(); + for (const auto & source_column : source_columns) + { + const auto & source_string_column = assert_cast(*source_column); + new_size += source_string_column.size(); + new_chars_size += source_string_column.chars.size(); + } + + offsets.reserve_exact(new_size); + chars.reserve_exact(new_chars_size); +} + void ColumnString::shrinkToFit() { chars.shrink_to_fit(); diff --git a/src/Columns/ColumnString.h b/src/Columns/ColumnString.h index c1012e1e55e..5bbb3ad5151 100644 --- a/src/Columns/ColumnString.h +++ b/src/Columns/ColumnString.h @@ -123,7 +123,7 @@ public: void insert(const Field & x) override { - const String & s = x.get(); + const String & s = x.safeGet(); const size_t old_size = chars.size(); const size_t size_to_append = s.size() + 1; const size_t new_size = old_size + size_to_append; @@ -283,6 +283,7 @@ public: ColumnPtr compress() const override; void reserve(size_t n) override; + void prepareForSquashing(const Columns & source_columns) override; void shrinkToFit() override; void getExtremes(Field & min, Field & max) const override; diff --git a/src/Columns/ColumnTuple.cpp b/src/Columns/ColumnTuple.cpp index 4fc3f88a87c..1539d7c8c2e 100644 --- a/src/Columns/ColumnTuple.cpp +++ b/src/Columns/ColumnTuple.cpp @@ -141,7 +141,7 @@ void ColumnTuple::get(size_t n, Field & res) const const size_t tuple_size = columns.size(); res = Tuple(); - Tuple & res_tuple = res.get(); + Tuple & res_tuple = res.safeGet(); res_tuple.reserve(tuple_size); for (size_t i = 0; i < tuple_size; ++i) @@ -169,7 +169,7 @@ void ColumnTuple::insertData(const char *, size_t) void ColumnTuple::insert(const Field & x) { - const auto & tuple = x.get(); + const auto & tuple = x.safeGet(); const size_t tuple_size = columns.size(); if (tuple.size() != tuple_size) @@ -185,7 +185,7 @@ bool ColumnTuple::tryInsert(const Field & x) if (x.getType() != Field::Types::Which::Tuple) return false; - const auto & tuple = x.get(); + const auto & tuple = x.safeGet(); const size_t tuple_size = columns.size(); if (tuple.size() != tuple_size) @@ -595,6 +595,19 @@ void ColumnTuple::reserve(size_t n) getColumn(i).reserve(n); } +void ColumnTuple::prepareForSquashing(const Columns & source_columns) +{ + const size_t tuple_size = columns.size(); + for (size_t i = 0; i < tuple_size; ++i) + { + Columns nested_columns; + nested_columns.reserve(source_columns.size()); + for (const auto & source_column : source_columns) + nested_columns.push_back(assert_cast(*source_column).getColumnPtr(i)); + getColumn(i).prepareForSquashing(nested_columns); + } +} + void ColumnTuple::shrinkToFit() { const size_t tuple_size = columns.size(); diff --git a/src/Columns/ColumnTuple.h b/src/Columns/ColumnTuple.h index 16b47a993f6..ef396d6a130 100644 --- a/src/Columns/ColumnTuple.h +++ b/src/Columns/ColumnTuple.h @@ -110,6 +110,7 @@ public: void updatePermutationWithCollation(const Collator & collator, IColumn::PermutationSortDirection direction, IColumn::PermutationSortStability stability, size_t limit, int nan_direction_hint, IColumn::Permutation & res, EqualRanges& equal_ranges) const override; void reserve(size_t n) override; + void prepareForSquashing(const Columns & source_columns) override; void shrinkToFit() override; void ensureOwnership() override; size_t byteSize() const override; diff --git a/src/Columns/ColumnUnique.h b/src/Columns/ColumnUnique.h index d6cb75679be..8a66f4e02ed 100644 --- a/src/Columns/ColumnUnique.h +++ b/src/Columns/ColumnUnique.h @@ -48,6 +48,8 @@ private: ColumnUnique(const ColumnUnique & other); public: + std::string getName() const override { return "Unique(" + getNestedColumn()->getName() + ")"; } + MutableColumnPtr cloneEmpty() const override; const ColumnPtr & getNestedColumn() const override; diff --git a/src/Columns/ColumnVariant.cpp b/src/Columns/ColumnVariant.cpp index de7efb41d19..d4294478633 100644 --- a/src/Columns/ColumnVariant.cpp +++ b/src/Columns/ColumnVariant.cpp @@ -476,7 +476,7 @@ void ColumnVariant::insertFromImpl(const DB::IColumn & src_, size_t n, const std } } -void ColumnVariant::insertRangeFromImpl(const DB::IColumn & src_, size_t start, size_t length, const std::vector * global_discriminators_mapping) +void ColumnVariant::insertRangeFromImpl(const DB::IColumn & src_, size_t start, size_t length, const std::vector * global_discriminators_mapping, const Discriminator * skip_discriminator) { const size_t num_variants = variants.size(); const auto & src = assert_cast(src_); @@ -557,9 +557,12 @@ void ColumnVariant::insertRangeFromImpl(const DB::IColumn & src_, size_t start, Discriminator global_discr = src_global_discr; if (global_discriminators_mapping && src_global_discr != NULL_DISCRIMINATOR) global_discr = (*global_discriminators_mapping)[src_global_discr]; - Discriminator local_discr = localDiscriminatorByGlobal(global_discr); - if (nested_length) - variants[local_discr]->insertRangeFrom(*src.variants[src_local_discr], nested_start, nested_length); + if (!skip_discriminator || global_discr != *skip_discriminator) + { + Discriminator local_discr = localDiscriminatorByGlobal(global_discr); + if (nested_length) + variants[local_discr]->insertRangeFrom(*src.variants[src_local_discr], nested_start, nested_length); + } } } @@ -610,7 +613,7 @@ void ColumnVariant::insertRangeFrom(const IColumn & src_, size_t start, size_t l void ColumnVariant::doInsertRangeFrom(const IColumn & src_, size_t start, size_t length) #endif { - insertRangeFromImpl(src_, start, length, nullptr); + insertRangeFromImpl(src_, start, length, nullptr, nullptr); } #if !defined(DEBUG_OR_SANITIZER_BUILD) @@ -627,9 +630,9 @@ void ColumnVariant::insertFrom(const DB::IColumn & src_, size_t n, const std::ve insertFromImpl(src_, n, &global_discriminators_mapping); } -void ColumnVariant::insertRangeFrom(const IColumn & src_, size_t start, size_t length, const std::vector & global_discriminators_mapping) +void ColumnVariant::insertRangeFrom(const IColumn & src_, size_t start, size_t length, const std::vector & global_discriminators_mapping, Discriminator skip_discriminator) { - insertRangeFromImpl(src_, start, length, &global_discriminators_mapping); + insertRangeFromImpl(src_, start, length, &global_discriminators_mapping, &skip_discriminator); } void ColumnVariant::insertManyFrom(const DB::IColumn & src_, size_t position, size_t length, const std::vector & global_discriminators_mapping) @@ -673,6 +676,14 @@ void ColumnVariant::insertManyIntoVariantFrom(DB::ColumnVariant::Discriminator g variants[local_discr]->insertManyFrom(src_, position, length); } +void ColumnVariant::deserializeBinaryIntoVariant(ColumnVariant::Discriminator global_discr, const SerializationPtr & serialization, ReadBuffer & buf, const FormatSettings & format_settings) +{ + auto local_discr = localDiscriminatorByGlobal(global_discr); + serialization->deserializeBinary(*variants[local_discr], buf, format_settings); + getLocalDiscriminators().push_back(local_discr); + getOffsets().push_back(variants[local_discr]->size() - 1); +} + void ColumnVariant::insertDefault() { getLocalDiscriminators().push_back(NULL_DISCRIMINATOR); @@ -1213,9 +1224,7 @@ struct ColumnVariant::ComparatorBase ALWAYS_INLINE int compare(size_t lhs, size_t rhs) const { - int res = parent.compareAt(lhs, rhs, parent, nan_direction_hint); - - return res; + return parent.compareAt(lhs, rhs, parent, nan_direction_hint); } }; @@ -1247,8 +1256,25 @@ void ColumnVariant::updatePermutation(IColumn::PermutationSortDirection directio void ColumnVariant::reserve(size_t n) { - local_discriminators->reserve(n); - offsets->reserve(n); + getLocalDiscriminators().reserve_exact(n); + getOffsets().reserve_exact(n); +} + +void ColumnVariant::prepareForSquashing(const Columns & source_columns) +{ + size_t new_size = size(); + for (const auto & source_column : source_columns) + new_size += source_column->size(); + reserve(new_size); + + for (size_t i = 0; i != variants.size(); ++i) + { + Columns source_variant_columns; + source_variant_columns.reserve(source_columns.size()); + for (const auto & source_column : source_columns) + source_variant_columns.push_back(assert_cast(*source_column).getVariantPtrByGlobalDiscriminator(i)); + getVariantByGlobalDiscriminator(i).prepareForSquashing(source_variant_columns); + } } void ColumnVariant::ensureOwnership() diff --git a/src/Columns/ColumnVariant.h b/src/Columns/ColumnVariant.h index 34c24b5428d..787b2f75247 100644 --- a/src/Columns/ColumnVariant.h +++ b/src/Columns/ColumnVariant.h @@ -2,6 +2,8 @@ #include #include +#include +#include namespace DB @@ -196,13 +198,15 @@ public: /// 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); - void insertRangeFrom(const IColumn & src_, size_t start, size_t length, const std::vector & global_discriminators_mapping); + /// Don't insert data into variant with skip_discriminator global discriminator, it will be processed separately. + void insertRangeFrom(const IColumn & src_, size_t start, size_t length, const std::vector & global_discriminators_mapping, Discriminator skip_discriminator); void insertManyFrom(const IColumn & src_, size_t position, size_t length, const std::vector & global_discriminators_mapping); /// Methods for insertion into a specific variant. void insertIntoVariantFrom(Discriminator global_discr, const IColumn & src_, size_t n); void insertRangeIntoVariantFrom(Discriminator global_discr, const IColumn & src_, size_t start, size_t length); void insertManyIntoVariantFrom(Discriminator global_discr, const IColumn & src_, size_t position, size_t length); + void deserializeBinaryIntoVariant(Discriminator global_discr, const SerializationPtr & serialization, ReadBuffer & buf, const FormatSettings & format_settings); void insertDefault() override; void insertManyDefaults(size_t length) override; @@ -237,6 +241,7 @@ public: size_t limit, int nan_direction_hint, IColumn::Permutation & res, EqualRanges & equal_ranges) const override; void reserve(size_t n) override; + void prepareForSquashing(const Columns & source_columns) override; void ensureOwnership() override; size_t byteSize() const override; size_t byteSizeAt(size_t n) const override; @@ -263,6 +268,7 @@ public: ColumnPtr & getVariantPtrByGlobalDiscriminator(size_t discr) { return variants[global_to_local_discriminators.at(discr)]; } const NestedColumns & getVariants() const { return variants; } + NestedColumns & getVariants() { return variants; } const IColumn & getLocalDiscriminatorsColumn() const { return *local_discriminators; } IColumn & getLocalDiscriminatorsColumn() { return *local_discriminators; } @@ -302,6 +308,8 @@ public: return true; } + std::vector getLocalToGlobalDiscriminatorsMapping() const { return local_to_global_discriminators; } + /// Check if we have only 1 non-empty variant and no NULL values, /// and if so, return the discriminator of this non-empty column. std::optional getLocalDiscriminatorOfOneNoneEmptyVariantNoNulls() const; @@ -322,7 +330,7 @@ public: private: void insertFromImpl(const IColumn & src_, size_t n, const std::vector * global_discriminators_mapping); - void insertRangeFromImpl(const IColumn & src_, size_t start, size_t length, const std::vector * global_discriminators_mapping); + void insertRangeFromImpl(const IColumn & src_, size_t start, size_t length, const std::vector * global_discriminators_mapping, const Discriminator * skip_discriminator); void insertManyFromImpl(const IColumn & src_, size_t position, size_t length, const std::vector * global_discriminators_mapping); void initIdentityGlobalToLocalDiscriminatorsMapping(); diff --git a/src/Columns/ColumnVector.h b/src/Columns/ColumnVector.h index 2fe5b635bd2..a5e1ee4b462 100644 --- a/src/Columns/ColumnVector.h +++ b/src/Columns/ColumnVector.h @@ -85,7 +85,7 @@ public: void insertMany(const Field & field, size_t length) override { - data.resize_fill(data.size() + length, static_cast(field.get())); + data.resize_fill(data.size() + length, static_cast(field.safeGet())); } void insertData(const char * pos, size_t) override @@ -235,7 +235,7 @@ public: void insert(const Field & x) override { - data.push_back(static_cast(x.get())); + data.push_back(static_cast(x.safeGet())); } bool tryInsert(const DB::Field & x) override; diff --git a/src/Columns/IColumn.h b/src/Columns/IColumn.h index f9c1a3e7034..edcb9f0bc30 100644 --- a/src/Columns/IColumn.h +++ b/src/Columns/IColumn.h @@ -475,6 +475,15 @@ public: /// It affects performance only (not correctness). virtual void reserve(size_t /*n*/) {} + /// Reserve memory before squashing all specified source columns into this column. + virtual void prepareForSquashing(const std::vector & source_columns) + { + size_t new_size = size(); + for (const auto & source_column : source_columns) + new_size += source_column->size(); + reserve(new_size); + } + /// Requests the removal of unused capacity. /// It is a non-binding request to reduce the capacity of the underlying container to its size. virtual void shrinkToFit() {} diff --git a/src/Columns/IColumnUnique.h b/src/Columns/IColumnUnique.h index a8e10e5e2b2..52b1bef3009 100644 --- a/src/Columns/IColumnUnique.h +++ b/src/Columns/IColumnUnique.h @@ -73,7 +73,7 @@ public: /// Returns dictionary hash which is SipHash is applied to each row of nested column. virtual UInt128 getHash() const = 0; - const char * getFamilyName() const override { return "ColumnUnique"; } + const char * getFamilyName() const override { return "Unique"; } TypeIndex getDataType() const override { return getNestedColumn()->getDataType(); } void insert(const Field &) override diff --git a/src/Columns/tests/gtest_column_dump_structure.cpp b/src/Columns/tests/gtest_column_dump_structure.cpp index e00c77798c8..d9647147157 100644 --- a/src/Columns/tests/gtest_column_dump_structure.cpp +++ b/src/Columns/tests/gtest_column_dump_structure.cpp @@ -10,7 +10,7 @@ TEST(IColumn, dumpStructure) { auto type_lc = std::make_shared(std::make_shared()); ColumnPtr column_lc = type_lc->createColumn(); - String expected_structure = "ColumnLowCardinality(size = 0, UInt8(size = 0), ColumnUnique(size = 1, String(size = 1)))"; + String expected_structure = "LowCardinality(size = 0, UInt8(size = 0), Unique(size = 1, String(size = 1)))"; std::vector threads; for (size_t i = 0; i < 6; ++i) diff --git a/src/Columns/tests/gtest_column_dynamic.cpp b/src/Columns/tests/gtest_column_dynamic.cpp index a2862b09de1..de76261229d 100644 --- a/src/Columns/tests/gtest_column_dynamic.cpp +++ b/src/Columns/tests/gtest_column_dynamic.cpp @@ -7,28 +7,34 @@ using namespace DB; TEST(ColumnDynamic, CreateEmpty) { - auto column = ColumnDynamic::create(255); + auto column = ColumnDynamic::create(254); ASSERT_TRUE(column->empty()); - ASSERT_EQ(column->getVariantInfo().variant_type->getName(), "Variant()"); - ASSERT_TRUE(column->getVariantInfo().variant_names.empty()); - ASSERT_TRUE(column->getVariantInfo().variant_name_to_discriminator.empty()); + ASSERT_EQ(column->getVariantInfo().variant_type->getName(), "Variant(SharedVariant)"); + ASSERT_EQ(column->getVariantInfo().variant_names.size(), 1); + ASSERT_EQ(column->getVariantInfo().variant_names[0], "SharedVariant"); + ASSERT_EQ(column->getVariantInfo().variant_name_to_discriminator.size(), 1); + ASSERT_EQ(column->getVariantInfo().variant_name_to_discriminator.at("SharedVariant"), 0); + ASSERT_TRUE(column->getVariantColumn().getVariantByGlobalDiscriminator(0).empty()); } TEST(ColumnDynamic, InsertDefault) { - auto column = ColumnDynamic::create(255); + auto column = ColumnDynamic::create(254); column->insertDefault(); ASSERT_TRUE(column->size() == 1); - ASSERT_EQ(column->getVariantInfo().variant_type->getName(), "Variant()"); - ASSERT_TRUE(column->getVariantInfo().variant_names.empty()); - ASSERT_TRUE(column->getVariantInfo().variant_name_to_discriminator.empty()); + ASSERT_EQ(column->getVariantInfo().variant_type->getName(), "Variant(SharedVariant)"); + ASSERT_EQ(column->getVariantInfo().variant_names.size(), 1); + ASSERT_EQ(column->getVariantInfo().variant_names[0], "SharedVariant"); + ASSERT_EQ(column->getVariantInfo().variant_name_to_discriminator.size(), 1); + ASSERT_EQ(column->getVariantInfo().variant_name_to_discriminator.at("SharedVariant"), 0); + ASSERT_TRUE(column->getVariantColumn().getVariantByGlobalDiscriminator(0).empty()); ASSERT_TRUE(column->isNullAt(0)); ASSERT_EQ((*column)[0], Field(Null())); } TEST(ColumnDynamic, InsertFields) { - auto column = ColumnDynamic::create(255); + auto column = ColumnDynamic::create(254); column->insert(Field(42)); column->insert(Field(-42)); column->insert(Field("str1")); @@ -41,16 +47,16 @@ TEST(ColumnDynamic, InsertFields) column->insert(Field(43.43)); ASSERT_TRUE(column->size() == 10); - ASSERT_EQ(column->getVariantInfo().variant_type->getName(), "Variant(Float64, Int8, String)"); - std::vector expected_names = {"Float64", "Int8", "String"}; + ASSERT_EQ(column->getVariantInfo().variant_type->getName(), "Variant(Float64, Int8, SharedVariant, String)"); + std::vector expected_names = {"Float64", "Int8", "SharedVariant", "String"}; ASSERT_EQ(column->getVariantInfo().variant_names, expected_names); - std::unordered_map expected_variant_name_to_discriminator = {{"Float64", 0}, {"Int8", 1}, {"String", 2}}; + std::unordered_map expected_variant_name_to_discriminator = {{"Float64", 0}, {"Int8", 1}, {"SharedVariant", 2}, {"String", 3}}; ASSERT_TRUE(column->getVariantInfo().variant_name_to_discriminator == expected_variant_name_to_discriminator); } ColumnDynamic::MutablePtr getDynamicWithManyVariants(size_t num_variants, Field tuple_element = Field(42)) { - auto column = ColumnDynamic::create(255); + auto column = ColumnDynamic::create(254); for (size_t i = 0; i != num_variants; ++i) { Tuple tuple; @@ -66,61 +72,71 @@ TEST(ColumnDynamic, InsertFieldsOverflow1) { auto column = getDynamicWithManyVariants(253); - ASSERT_EQ(column->getVariantInfo().variant_names.size(), 253); + ASSERT_EQ(column->getVariantInfo().variant_names.size(), 254); column->insert(Field(42.42)); - ASSERT_EQ(column->getVariantInfo().variant_names.size(), 254); + ASSERT_EQ(column->size(), 254); + ASSERT_EQ(column->getVariantInfo().variant_names.size(), 255); ASSERT_TRUE(column->getVariantInfo().variant_name_to_discriminator.contains("Float64")); column->insert(Field(42)); + ASSERT_EQ(column->size(), 255); ASSERT_EQ(column->getVariantInfo().variant_names.size(), 255); ASSERT_FALSE(column->getVariantInfo().variant_name_to_discriminator.contains("Int8")); - ASSERT_TRUE(column->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_EQ(column->getSharedVariant().size(), 1); Field field = (*column)[column->size() - 1]; - ASSERT_EQ(field, "42"); + ASSERT_EQ(field, 42); column->insert(Field(43)); + ASSERT_EQ(column->size(), 256); ASSERT_EQ(column->getVariantInfo().variant_names.size(), 255); ASSERT_FALSE(column->getVariantInfo().variant_name_to_discriminator.contains("Int8")); - ASSERT_TRUE(column->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_EQ(column->getSharedVariant().size(), 2); field = (*column)[column->size() - 1]; - ASSERT_EQ(field, "43"); + ASSERT_EQ(field, 43); column->insert(Field("str1")); + ASSERT_EQ(column->size(), 257); ASSERT_EQ(column->getVariantInfo().variant_names.size(), 255); ASSERT_FALSE(column->getVariantInfo().variant_name_to_discriminator.contains("Int8")); - ASSERT_TRUE(column->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_EQ(column->getSharedVariant().size(), 3); field = (*column)[column->size() - 1]; ASSERT_EQ(field, "str1"); column->insert(Field(Array({Field(42), Field(43)}))); ASSERT_EQ(column->getVariantInfo().variant_names.size(), 255); ASSERT_FALSE(column->getVariantInfo().variant_name_to_discriminator.contains("Array(Int8)")); - ASSERT_TRUE(column->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_EQ(column->getSharedVariant().size(), 4); field = (*column)[column->size() - 1]; - ASSERT_EQ(field, "[42, 43]"); + ASSERT_EQ(field, Field(Array({Field(42), Field(43)}))); } TEST(ColumnDynamic, InsertFieldsOverflow2) { auto column = getDynamicWithManyVariants(254); - ASSERT_EQ(column->getVariantInfo().variant_names.size(), 254); + ASSERT_EQ(column->getVariantInfo().variant_names.size(), 255); column->insert(Field("str1")); ASSERT_EQ(column->getVariantInfo().variant_names.size(), 255); - ASSERT_TRUE(column->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_EQ(column->getSharedVariant().size(), 1); + Field field = (*column)[column->size() - 1]; + ASSERT_EQ(field, "str1"); column->insert(Field(42)); ASSERT_EQ(column->getVariantInfo().variant_names.size(), 255); ASSERT_FALSE(column->getVariantInfo().variant_name_to_discriminator.contains("Int8")); - ASSERT_TRUE(column->getVariantInfo().variant_name_to_discriminator.contains("String")); - Field field = (*column)[column->size() - 1]; - ASSERT_EQ(field, "42"); + ASSERT_FALSE(column->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_EQ(column->getSharedVariant().size(), 2); + field = (*column)[column->size() - 1]; + ASSERT_EQ(field, 42); } ColumnDynamic::MutablePtr getInsertFromColumn(size_t num = 1) { - auto column_from = ColumnDynamic::create(255); + auto column_from = ColumnDynamic::create(254); for (size_t i = 0; i != num; ++i) { column_from->insert(Field(42)); @@ -154,41 +170,41 @@ void checkInsertFrom(const ColumnDynamic::MutablePtr & column_from, ColumnDynami TEST(ColumnDynamic, InsertFrom1) { - auto column_to = ColumnDynamic::create(255); - checkInsertFrom(getInsertFromColumn(), column_to, "Variant(Float64, Int8, String)", {"Float64", "Int8", "String"}, {{"Float64", 0}, {"Int8", 1}, {"String", 2}}); + auto column_to = ColumnDynamic::create(254); + checkInsertFrom(getInsertFromColumn(), column_to, "Variant(Float64, Int8, SharedVariant, String)", {"Float64", "Int8", "SharedVariant", "String"}, {{"Float64", 0}, {"Int8", 1}, {"SharedVariant", 2}, {"String", 3}}); } TEST(ColumnDynamic, InsertFrom2) { - auto column_to = ColumnDynamic::create(255); + auto column_to = ColumnDynamic::create(254); column_to->insert(Field(42)); column_to->insert(Field(42.42)); column_to->insert(Field("str")); - checkInsertFrom(getInsertFromColumn(), column_to, "Variant(Float64, Int8, String)", {"Float64", "Int8", "String"}, {{"Float64", 0}, {"Int8", 1}, {"String", 2}}); + checkInsertFrom(getInsertFromColumn(), column_to, "Variant(Float64, Int8, SharedVariant, String)", {"Float64", "Int8", "SharedVariant", "String"}, {{"Float64", 0}, {"Int8", 1}, {"SharedVariant", 2}, {"String", 3}}); } TEST(ColumnDynamic, InsertFrom3) { - auto column_to = ColumnDynamic::create(255); + auto column_to = ColumnDynamic::create(254); column_to->insert(Field(42)); column_to->insert(Field(42.42)); column_to->insert(Field("str")); column_to->insert(Array({Field(42)})); - checkInsertFrom(getInsertFromColumn(), column_to, "Variant(Array(Int8), Float64, Int8, String)", {"Array(Int8)", "Float64", "Int8", "String"}, {{"Array(Int8)", 0}, {"Float64", 1}, {"Int8", 2}, {"String", 3}}); + checkInsertFrom(getInsertFromColumn(), column_to, "Variant(Array(Int8), Float64, Int8, SharedVariant, String)", {"Array(Int8)", "Float64", "Int8", "SharedVariant", "String"}, {{"Array(Int8)", 0}, {"Float64", 1}, {"Int8", 2}, {"SharedVariant", 3}, {"String", 4}}); } TEST(ColumnDynamic, InsertFromOverflow1) { - auto column_from = ColumnDynamic::create(255); + auto column_from = ColumnDynamic::create(254); column_from->insert(Field(42)); column_from->insert(Field(42.42)); column_from->insert(Field("str")); auto column_to = getDynamicWithManyVariants(253); column_to->insertFrom(*column_from, 0); - ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 254); + ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 255); ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); auto field = (*column_to)[column_to->size() - 1]; ASSERT_EQ(field, 42); @@ -196,20 +212,22 @@ TEST(ColumnDynamic, InsertFromOverflow1) column_to->insertFrom(*column_from, 1); ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 255); ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); - ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_EQ(column_to->getSharedVariant().size(), 1); field = (*column_to)[column_to->size() - 1]; - ASSERT_EQ(field, "42.42"); + ASSERT_EQ(field, 42.42); column_to->insertFrom(*column_from, 2); ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 255); - ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_EQ(column_to->getSharedVariant().size(), 2); field = (*column_to)[column_to->size() - 1]; ASSERT_EQ(field, "str"); } TEST(ColumnDynamic, InsertFromOverflow2) { - auto column_from = ColumnDynamic::create(255); + auto column_from = ColumnDynamic::create(254); column_from->insert(Field(42)); column_from->insert(Field(42.42)); @@ -221,9 +239,32 @@ TEST(ColumnDynamic, InsertFromOverflow2) column_to->insertFrom(*column_from, 1); ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); - ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_EQ(column_to->getSharedVariant().size(), 1); field = (*column_to)[column_to->size() - 1]; - ASSERT_EQ(field, "42.42"); + ASSERT_EQ(field, 42.42); +} + +TEST(ColumnDynamic, InsertFromOverflow3) +{ + auto column_from = ColumnDynamic::create(1); + column_from->insert(Field(42)); + column_from->insert(Field(42.42)); + + auto column_to = ColumnDynamic::create(254); + column_to->insert(Field(41)); + + column_to->insertFrom(*column_from, 0); + ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); + ASSERT_EQ(column_to->getSharedVariant().size(), 0); + auto field = (*column_to)[column_to->size() - 1]; + ASSERT_EQ(field, 42); + + column_to->insertFrom(*column_from, 1); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); + ASSERT_EQ(column_to->getSharedVariant().size(), 1); + field = (*column_to)[column_to->size() - 1]; + ASSERT_EQ(field, 42.42); } void checkInsertManyFrom(const ColumnDynamic::MutablePtr & column_from, ColumnDynamic::MutablePtr & column_to, const std::string & expected_variant, const std::vector & expected_names, const std::unordered_map & expected_variant_name_to_discriminator) @@ -256,42 +297,43 @@ void checkInsertManyFrom(const ColumnDynamic::MutablePtr & column_from, ColumnDy TEST(ColumnDynamic, InsertManyFrom1) { - auto column_to = ColumnDynamic::create(255); - checkInsertManyFrom(getInsertFromColumn(), column_to, "Variant(Float64, Int8, String)", {"Float64", "Int8", "String"}, {{"Float64", 0}, {"Int8", 1}, {"String", 2}}); + auto column_to = ColumnDynamic::create(254); + checkInsertManyFrom(getInsertFromColumn(), column_to, "Variant(Float64, Int8, SharedVariant, String)", {"Float64", "Int8", "SharedVariant", "String"}, {{"Float64", 0}, {"Int8", 1}, {"SharedVariant", 2}, {"String", 3}}); } TEST(ColumnDynamic, InsertManyFrom2) { - auto column_to = ColumnDynamic::create(255); + auto column_to = ColumnDynamic::create(254); column_to->insert(Field(42)); column_to->insert(Field(42.42)); column_to->insert(Field("str")); - checkInsertManyFrom(getInsertFromColumn(), column_to, "Variant(Float64, Int8, String)", {"Float64", "Int8", "String"}, {{"Float64", 0}, {"Int8", 1}, {"String", 2}}); + checkInsertManyFrom(getInsertFromColumn(), column_to, "Variant(Float64, Int8, SharedVariant, String)", {"Float64", "Int8", "SharedVariant", "String"}, {{"Float64", 0}, {"Int8", 1}, {"SharedVariant", 2}, {"String", 3}}); } TEST(ColumnDynamic, InsertManyFrom3) { - auto column_to = ColumnDynamic::create(255); + auto column_to = ColumnDynamic::create(254); column_to->insert(Field(42)); column_to->insert(Field(42.42)); column_to->insert(Field("str")); column_to->insert(Array({Field(42)})); - checkInsertManyFrom(getInsertFromColumn(), column_to, "Variant(Array(Int8), Float64, Int8, String)", {"Array(Int8)", "Float64", "Int8", "String"}, {{"Array(Int8)", 0}, {"Float64", 1}, {"Int8", 2}, {"String", 3}}); + checkInsertManyFrom(getInsertFromColumn(), column_to, "Variant(Array(Int8), Float64, Int8, SharedVariant, String)", {"Array(Int8)", "Float64", "Int8", "SharedVariant", "String"}, {{"Array(Int8)", 0}, {"Float64", 1}, {"Int8", 2}, {"SharedVariant", 3}, {"String", 4}}); } TEST(ColumnDynamic, InsertManyFromOverflow1) { - auto column_from = ColumnDynamic::create(255); + auto column_from = ColumnDynamic::create(254); column_from->insert(Field(42)); column_from->insert(Field(42.42)); column_from->insert(Field("str")); auto column_to = getDynamicWithManyVariants(253); column_to->insertManyFrom(*column_from, 0, 2); - ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 254); + ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 255); ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); + ASSERT_EQ(column_to->getSharedVariant().size(), 0); auto field = (*column_to)[column_to->size() - 2]; ASSERT_EQ(field, 42); field = (*column_to)[column_to->size() - 1]; @@ -300,15 +342,17 @@ TEST(ColumnDynamic, InsertManyFromOverflow1) column_to->insertManyFrom(*column_from, 1, 2); ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 255); ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); - ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_EQ(column_to->getSharedVariant().size(), 2); field = (*column_to)[column_to->size() - 2]; - ASSERT_EQ(field, "42.42"); + ASSERT_EQ(field, 42.42); field = (*column_to)[column_to->size() - 1]; - ASSERT_EQ(field, "42.42"); + ASSERT_EQ(field, 42.42); column_to->insertManyFrom(*column_from, 2, 2); ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 255); - ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_EQ(column_to->getSharedVariant().size(), 4); field = (*column_to)[column_to->size() - 1]; ASSERT_EQ(field, "str"); field = (*column_to)[column_to->size() - 2]; @@ -317,14 +361,15 @@ TEST(ColumnDynamic, InsertManyFromOverflow1) TEST(ColumnDynamic, InsertManyFromOverflow2) { - auto column_from = ColumnDynamic::create(255); + auto column_from = ColumnDynamic::create(254); column_from->insert(Field(42)); column_from->insert(Field(42.42)); auto column_to = getDynamicWithManyVariants(253); column_to->insertManyFrom(*column_from, 0, 2); - ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 254); + ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 255); ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); + ASSERT_EQ(column_to->getSharedVariant().size(), 0); auto field = (*column_to)[column_to->size() - 2]; ASSERT_EQ(field, 42); field = (*column_to)[column_to->size() - 1]; @@ -333,11 +378,39 @@ TEST(ColumnDynamic, InsertManyFromOverflow2) column_to->insertManyFrom(*column_from, 1, 2); ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 255); ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); - ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_EQ(column_to->getSharedVariant().size(), 2); field = (*column_to)[column_to->size() - 2]; - ASSERT_EQ(field, "42.42"); + ASSERT_EQ(field, 42.42); field = (*column_to)[column_to->size() - 1]; - ASSERT_EQ(field, "42.42"); + ASSERT_EQ(field, 42.42); +} + + +TEST(ColumnDynamic, InsertManyFromOverflow3) +{ + auto column_from = ColumnDynamic::create(1); + column_from->insert(Field(42)); + column_from->insert(Field(42.42)); + + auto column_to = ColumnDynamic::create(254); + column_to->insert(Field(41)); + + column_to->insertManyFrom(*column_from, 0, 2); + ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); + ASSERT_EQ(column_to->getSharedVariant().size(), 0); + auto field = (*column_to)[column_to->size() - 2]; + ASSERT_EQ(field, 42); + field = (*column_to)[column_to->size() - 1]; + ASSERT_EQ(field, 42); + + column_to->insertManyFrom(*column_from, 1, 2); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); + ASSERT_EQ(column_to->getSharedVariant().size(), 2); + field = (*column_to)[column_to->size() - 2]; + ASSERT_EQ(field, 42.42); + field = (*column_to)[column_to->size() - 1]; + ASSERT_EQ(field, 42.42); } void checkInsertRangeFrom(const ColumnDynamic::MutablePtr & column_from, ColumnDynamic::MutablePtr & column_to, const std::string & expected_variant, const std::vector & expected_names, const std::unordered_map & expected_variant_name_to_discriminator) @@ -368,34 +441,34 @@ void checkInsertRangeFrom(const ColumnDynamic::MutablePtr & column_from, ColumnD TEST(ColumnDynamic, InsertRangeFrom1) { - auto column_to = ColumnDynamic::create(255); - checkInsertRangeFrom(getInsertFromColumn(2), column_to, "Variant(Float64, Int8, String)", {"Float64", "Int8", "String"}, {{"Float64", 0}, {"Int8", 1}, {"String", 2}}); + auto column_to = ColumnDynamic::create(254); + checkInsertRangeFrom(getInsertFromColumn(2), column_to, "Variant(Float64, Int8, SharedVariant, String)", {"Float64", "Int8", "SharedVariant", "String"}, {{"Float64", 0}, {"Int8", 1}, {"SharedVariant", 2}, {"String", 3}}); } TEST(ColumnDynamic, InsertRangeFrom2) { - auto column_to = ColumnDynamic::create(255); + auto column_to = ColumnDynamic::create(254); column_to->insert(Field(42)); column_to->insert(Field(42.42)); column_to->insert(Field("str1")); - checkInsertRangeFrom(getInsertFromColumn(2), column_to, "Variant(Float64, Int8, String)", {"Float64", "Int8", "String"}, {{"Float64", 0}, {"Int8", 1}, {"String", 2}}); + checkInsertRangeFrom(getInsertFromColumn(2), column_to, "Variant(Float64, Int8, SharedVariant, String)", {"Float64", "Int8", "SharedVariant", "String"}, {{"Float64", 0}, {"Int8", 1}, {"SharedVariant", 2}, {"String", 3}}); } TEST(ColumnDynamic, InsertRangeFrom3) { - auto column_to = ColumnDynamic::create(255); + auto column_to = ColumnDynamic::create(254); column_to->insert(Field(42)); column_to->insert(Field(42.42)); column_to->insert(Field("str1")); column_to->insert(Array({Field(42)})); - checkInsertRangeFrom(getInsertFromColumn(2), column_to, "Variant(Array(Int8), Float64, Int8, String)", {"Array(Int8)", "Float64", "Int8", "String"}, {{"Array(Int8)", 0}, {"Float64", 1}, {"Int8", 2}, {"String", 3}}); + checkInsertRangeFrom(getInsertFromColumn(2), column_to, "Variant(Array(Int8), Float64, Int8, SharedVariant, String)", {"Array(Int8)", "Float64", "Int8", "SharedVariant", "String"}, {{"Array(Int8)", 0}, {"Float64", 1}, {"Int8", 2}, {"SharedVariant", 3}, {"String", 4}}); } TEST(ColumnDynamic, InsertRangeFromOverflow1) { - auto column_from = ColumnDynamic::create(255); + auto column_from = ColumnDynamic::create(254); column_from->insert(Field(42)); column_from->insert(Field(43)); column_from->insert(Field(42.42)); @@ -403,23 +476,25 @@ TEST(ColumnDynamic, InsertRangeFromOverflow1) auto column_to = getDynamicWithManyVariants(253); column_to->insertRangeFrom(*column_from, 0, 4); + ASSERT_EQ(column_to->size(), 257); ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 255); ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); - ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); + ASSERT_EQ(column_to->getSharedVariant().size(), 2); auto field = (*column_to)[column_to->size() - 4]; ASSERT_EQ(field, Field(42)); field = (*column_to)[column_to->size() - 3]; ASSERT_EQ(field, Field(43)); field = (*column_to)[column_to->size() - 2]; - ASSERT_EQ(field, Field("42.42")); + ASSERT_EQ(field, Field(42.42)); field = (*column_to)[column_to->size() - 1]; ASSERT_EQ(field, Field("str")); } TEST(ColumnDynamic, InsertRangeFromOverflow2) { - auto column_from = ColumnDynamic::create(255); + auto column_from = ColumnDynamic::create(254); column_from->insert(Field(42)); column_from->insert(Field(43)); column_from->insert(Field(42.42)); @@ -428,19 +503,20 @@ TEST(ColumnDynamic, InsertRangeFromOverflow2) column_to->insertRangeFrom(*column_from, 0, 3); ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 255); ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); - ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); + ASSERT_EQ(column_to->getSharedVariant().size(), 1); auto field = (*column_to)[column_to->size() - 3]; ASSERT_EQ(field, Field(42)); field = (*column_to)[column_to->size() - 2]; ASSERT_EQ(field, Field(43)); field = (*column_to)[column_to->size() - 1]; - ASSERT_EQ(field, Field("42.42")); + ASSERT_EQ(field, Field(42.42)); } TEST(ColumnDynamic, InsertRangeFromOverflow3) { - auto column_from = ColumnDynamic::create(255); + auto column_from = ColumnDynamic::create(254); column_from->insert(Field(42)); column_from->insert(Field(43)); column_from->insert(Field(42.42)); @@ -449,20 +525,21 @@ TEST(ColumnDynamic, InsertRangeFromOverflow3) column_to->insert(Field("Str")); column_to->insertRangeFrom(*column_from, 0, 3); ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 255); - ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); + ASSERT_EQ(column_to->getSharedVariant().size(), 3); auto field = (*column_to)[column_to->size() - 3]; ASSERT_EQ(field, Field(42)); field = (*column_to)[column_to->size() - 2]; ASSERT_EQ(field, Field(43)); field = (*column_to)[column_to->size() - 1]; - ASSERT_EQ(field, Field("42.42")); + ASSERT_EQ(field, Field(42.42)); } TEST(ColumnDynamic, InsertRangeFromOverflow4) { - auto column_from = ColumnDynamic::create(255); + auto column_from = ColumnDynamic::create(254); column_from->insert(Field(42)); column_from->insert(Field(42.42)); column_from->insert(Field("str")); @@ -471,19 +548,20 @@ TEST(ColumnDynamic, InsertRangeFromOverflow4) column_to->insertRangeFrom(*column_from, 0, 3); ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 255); ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); - ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); + ASSERT_EQ(column_to->getSharedVariant().size(), 3); auto field = (*column_to)[column_to->size() - 3]; - ASSERT_EQ(field, Field("42")); + ASSERT_EQ(field, Field(42)); field = (*column_to)[column_to->size() - 2]; - ASSERT_EQ(field, Field("42.42")); + ASSERT_EQ(field, Field(42.42)); field = (*column_to)[column_to->size() - 1]; ASSERT_EQ(field, Field("str")); } TEST(ColumnDynamic, InsertRangeFromOverflow5) { - auto column_from = ColumnDynamic::create(255); + auto column_from = ColumnDynamic::create(254); column_from->insert(Field(42)); column_from->insert(Field(43)); column_from->insert(Field(42.42)); @@ -493,22 +571,23 @@ TEST(ColumnDynamic, InsertRangeFromOverflow5) column_to->insert(Field("str")); column_to->insertRangeFrom(*column_from, 0, 4); ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 255); - ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); + ASSERT_EQ(column_to->getSharedVariant().size(), 3); auto field = (*column_to)[column_to->size() - 4]; ASSERT_EQ(field, Field(42)); field = (*column_to)[column_to->size() - 3]; ASSERT_EQ(field, Field(43)); field = (*column_to)[column_to->size() - 2]; - ASSERT_EQ(field, Field("42.42")); + ASSERT_EQ(field, Field(42.42)); field = (*column_to)[column_to->size() - 1]; ASSERT_EQ(field, Field("str")); } TEST(ColumnDynamic, InsertRangeFromOverflow6) { - auto column_from = ColumnDynamic::create(255); + auto column_from = ColumnDynamic::create(254); column_from->insert(Field(42)); column_from->insert(Field(43)); column_from->insert(Field(44)); @@ -520,13 +599,14 @@ TEST(ColumnDynamic, InsertRangeFromOverflow6) auto column_to = getDynamicWithManyVariants(253); column_to->insertRangeFrom(*column_from, 2, 5); ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 255); - ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); - ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); - ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Array(Int8)")); + ASSERT_EQ(column_to->getSharedVariant().size(), 4); auto field = (*column_to)[column_to->size() - 5]; - ASSERT_EQ(field, Field("44")); + ASSERT_EQ(field, Field(44)); field = (*column_to)[column_to->size() - 4]; ASSERT_EQ(field, Field(42.42)); field = (*column_to)[column_to->size() - 3]; @@ -534,12 +614,136 @@ TEST(ColumnDynamic, InsertRangeFromOverflow6) field = (*column_to)[column_to->size() - 2]; ASSERT_EQ(field, Field("str")); field = (*column_to)[column_to->size() - 1]; - ASSERT_EQ(field, Field("[42]")); + ASSERT_EQ(field, Field(Array({Field(42)}))); +} + +TEST(ColumnDynamic, InsertRangeFromOverflow7) +{ + auto column_from = ColumnDynamic::create(2); + column_from->insert(Field(42.42)); + column_from->insert(Field("str1")); + column_from->insert(Field(42)); + column_from->insert(Field(43.43)); + column_from->insert(Field(Array({Field(41)}))); + column_from->insert(Field(43)); + column_from->insert(Field("str2")); + column_from->insert(Field(Array({Field(42)}))); + + auto column_to = ColumnDynamic::create(254); + column_to->insert(Field(42)); + + column_to->insertRangeFrom(*column_from, 0, 8); + ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 4); + ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); + ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Array(Int8)")); + ASSERT_EQ(column_to->getSharedVariant().size(), 2); + auto field = (*column_to)[column_to->size() - 8]; + ASSERT_EQ(field, Field(42.42)); + field = (*column_to)[column_to->size() - 7]; + ASSERT_EQ(field, Field("str1")); + field = (*column_to)[column_to->size() - 6]; + ASSERT_EQ(field, Field(42)); + field = (*column_to)[column_to->size() - 5]; + ASSERT_EQ(field, Field(43.43)); + field = (*column_to)[column_to->size() - 4]; + ASSERT_EQ(field, Field(Array({Field(41)}))); + field = (*column_to)[column_to->size() - 3]; + ASSERT_EQ(field, Field(43)); + field = (*column_to)[column_to->size() - 2]; + ASSERT_EQ(field, Field("str2")); + field = (*column_to)[column_to->size() - 1]; + ASSERT_EQ(field, Field(Array({Field(42)}))); +} + +TEST(ColumnDynamic, InsertRangeFromOverflow8) +{ + auto column_from = ColumnDynamic::create(2); + column_from->insert(Field(42.42)); + column_from->insert(Field("str1")); + column_from->insert(Field(42)); + column_from->insert(Field(43.43)); + column_from->insert(Field(Array({Field(41)}))); + column_from->insert(Field(43)); + column_from->insert(Field("str2")); + column_from->insert(Field(Array({Field(42)}))); + + auto column_to = ColumnDynamic::create(2); + column_to->insert(Field(42)); + column_from->insert(Field("str1")); + + column_to->insertRangeFrom(*column_from, 0, 8); + ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 3); + ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); + ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Array(Int8)")); + ASSERT_EQ(column_to->getSharedVariant().size(), 4); + auto field = (*column_to)[column_to->size() - 8]; + ASSERT_EQ(field, Field(42.42)); + field = (*column_to)[column_to->size() - 7]; + ASSERT_EQ(field, Field("str1")); + field = (*column_to)[column_to->size() - 6]; + ASSERT_EQ(field, Field(42)); + field = (*column_to)[column_to->size() - 5]; + ASSERT_EQ(field, Field(43.43)); + field = (*column_to)[column_to->size() - 4]; + ASSERT_EQ(field, Field(Array({Field(41)}))); + field = (*column_to)[column_to->size() - 3]; + ASSERT_EQ(field, Field(43)); + field = (*column_to)[column_to->size() - 2]; + ASSERT_EQ(field, Field("str2")); + field = (*column_to)[column_to->size() - 1]; + ASSERT_EQ(field, Field(Array({Field(42)}))); +} + +TEST(ColumnDynamic, InsertRangeFromOverflow9) +{ + auto column_from = ColumnDynamic::create(3); + column_from->insert(Field("str1")); + column_from->insert(Field(42.42)); + column_from->insert(Field("str2")); + column_from->insert(Field(42)); + column_from->insert(Field(43.43)); + column_from->insert(Field(Array({Field(41)}))); + column_from->insert(Field(43)); + column_from->insert(Field("str2")); + column_from->insert(Field(Array({Field(42)}))); + + auto column_to = ColumnDynamic::create(2); + column_to->insert(Field(42)); + + column_to->insertRangeFrom(*column_from, 0, 9); + ASSERT_EQ(column_to->getVariantInfo().variant_names.size(), 3); + ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); + ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Array(Int8)")); + ASSERT_EQ(column_to->getSharedVariant().size(), 4); + auto field = (*column_to)[column_to->size() - 9]; + ASSERT_EQ(field, Field("str1")); + field = (*column_to)[column_to->size() - 8]; + ASSERT_EQ(field, Field(42.42)); + field = (*column_to)[column_to->size() - 7]; + ASSERT_EQ(field, Field("str2")); + field = (*column_to)[column_to->size() - 6]; + ASSERT_EQ(field, Field(42)); + field = (*column_to)[column_to->size() - 5]; + ASSERT_EQ(field, Field(43.43)); + field = (*column_to)[column_to->size() - 4]; + ASSERT_EQ(field, Field(Array({Field(41)}))); + field = (*column_to)[column_to->size() - 3]; + ASSERT_EQ(field, Field(43)); + field = (*column_to)[column_to->size() - 2]; + ASSERT_EQ(field, Field("str2")); + field = (*column_to)[column_to->size() - 1]; + ASSERT_EQ(field, Field(Array({Field(42)}))); } TEST(ColumnDynamic, SerializeDeserializeFromArena1) { - auto column = ColumnDynamic::create(255); + auto column = ColumnDynamic::create(254); column->insert(Field(42)); column->insert(Field(42.42)); column->insert(Field("str")); @@ -564,7 +768,7 @@ TEST(ColumnDynamic, SerializeDeserializeFromArena1) TEST(ColumnDynamic, SerializeDeserializeFromArena2) { - auto column_from = ColumnDynamic::create(255); + auto column_from = ColumnDynamic::create(254); column_from->insert(Field(42)); column_from->insert(Field(42.42)); column_from->insert(Field("str")); @@ -577,26 +781,26 @@ TEST(ColumnDynamic, SerializeDeserializeFromArena2) column_from->serializeValueIntoArena(2, arena, pos); column_from->serializeValueIntoArena(3, arena, pos); - auto column_to = ColumnDynamic::create(255); + auto column_to = ColumnDynamic::create(254); pos = column_to->deserializeAndInsertFromArena(ref1.data); pos = column_to->deserializeAndInsertFromArena(pos); pos = column_to->deserializeAndInsertFromArena(pos); column_to->deserializeAndInsertFromArena(pos); - ASSERT_EQ((*column_from)[column_from->size() - 4], 42); - ASSERT_EQ((*column_from)[column_from->size() - 3], 42.42); - ASSERT_EQ((*column_from)[column_from->size() - 2], "str"); - ASSERT_EQ((*column_from)[column_from->size() - 1], Null()); - ASSERT_EQ(column_to->getVariantInfo().variant_type->getName(), "Variant(Float64, Int8, String)"); - std::vector expected_names = {"Float64", "Int8", "String"}; + ASSERT_EQ((*column_to)[column_to->size() - 4], 42); + ASSERT_EQ((*column_to)[column_to->size() - 3], 42.42); + ASSERT_EQ((*column_to)[column_to->size() - 2], "str"); + ASSERT_EQ((*column_to)[column_to->size() - 1], Null()); + ASSERT_EQ(column_to->getVariantInfo().variant_type->getName(), "Variant(Float64, Int8, SharedVariant, String)"); + std::vector expected_names = {"Float64", "Int8", "SharedVariant", "String"}; ASSERT_EQ(column_to->getVariantInfo().variant_names, expected_names); - std::unordered_map expected_variant_name_to_discriminator = {{"Float64", 0}, {"Int8", 1}, {"String", 2}}; + std::unordered_map expected_variant_name_to_discriminator = {{"Float64", 0}, {"Int8", 1}, {"SharedVariant", 2}, {"String", 3}}; ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator == expected_variant_name_to_discriminator); } -TEST(ColumnDynamic, SerializeDeserializeFromArenaOverflow) +TEST(ColumnDynamic, SerializeDeserializeFromArenaOverflow1) { - auto column_from = ColumnDynamic::create(255); + auto column_from = ColumnDynamic::create(254); column_from->insert(Field(42)); column_from->insert(Field(42.42)); column_from->insert(Field("str")); @@ -615,18 +819,56 @@ TEST(ColumnDynamic, SerializeDeserializeFromArenaOverflow) pos = column_to->deserializeAndInsertFromArena(pos); column_to->deserializeAndInsertFromArena(pos); - ASSERT_EQ((*column_from)[column_from->size() - 4], 42); - ASSERT_EQ((*column_from)[column_from->size() - 3], 42.42); - ASSERT_EQ((*column_from)[column_from->size() - 2], "str"); - ASSERT_EQ((*column_from)[column_from->size() - 1], Null()); + ASSERT_EQ((*column_to)[column_to->size() - 4], 42); + ASSERT_EQ((*column_to)[column_to->size() - 3], 42.42); + ASSERT_EQ((*column_to)[column_to->size() - 2], "str"); + ASSERT_EQ((*column_to)[column_to->size() - 1], Null()); ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); - ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_EQ(column_to->getSharedVariant().size(), 2); +} + +TEST(ColumnDynamic, SerializeDeserializeFromArenaOverflow2) +{ + auto column_from = ColumnDynamic::create(2); + column_from->insert(Field(42)); + column_from->insert(Field(42.42)); + column_from->insert(Field("str")); + column_from->insert(Field(Null())); + column_from->insert(Field(Array({Field(42)}))); + + Arena arena; + const char * pos = nullptr; + auto ref1 = column_from->serializeValueIntoArena(0, arena, pos); + column_from->serializeValueIntoArena(1, arena, pos); + column_from->serializeValueIntoArena(2, arena, pos); + column_from->serializeValueIntoArena(3, arena, pos); + column_from->serializeValueIntoArena(4, arena, pos); + + auto column_to = ColumnDynamic::create(2); + column_to->insert(Field(42.42)); + pos = column_to->deserializeAndInsertFromArena(ref1.data); + pos = column_to->deserializeAndInsertFromArena(pos); + pos = column_to->deserializeAndInsertFromArena(pos); + pos = column_to->deserializeAndInsertFromArena(pos); + column_to->deserializeAndInsertFromArena(pos); + + ASSERT_EQ((*column_to)[column_to->size() - 5], 42); + ASSERT_EQ((*column_to)[column_to->size() - 4], 42.42); + ASSERT_EQ((*column_to)[column_to->size() - 3], "str"); + ASSERT_EQ((*column_to)[column_to->size() - 2], Null()); + ASSERT_EQ((*column_to)[column_to->size() - 1], Field(Array({Field(42)}))); + ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Int8")); + ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Float64")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("String")); + ASSERT_FALSE(column_to->getVariantInfo().variant_name_to_discriminator.contains("Array(Int8)")); + ASSERT_EQ(column_to->getSharedVariant().size(), 2); } TEST(ColumnDynamic, skipSerializedInArena) { - auto column_from = ColumnDynamic::create(255); + auto column_from = ColumnDynamic::create(3); column_from->insert(Field(42)); column_from->insert(Field(42.42)); column_from->insert(Field("str")); @@ -640,13 +882,41 @@ TEST(ColumnDynamic, skipSerializedInArena) auto ref4 = column_from->serializeValueIntoArena(3, arena, pos); const char * end = ref4.data + ref4.size; - auto column_to = ColumnDynamic::create(255); + auto column_to = ColumnDynamic::create(254); pos = column_to->skipSerializedInArena(ref1.data); pos = column_to->skipSerializedInArena(pos); pos = column_to->skipSerializedInArena(pos); pos = column_to->skipSerializedInArena(pos); ASSERT_EQ(pos, end); - ASSERT_TRUE(column_to->getVariantInfo().variant_name_to_discriminator.empty()); - ASSERT_TRUE(column_to->getVariantInfo().variant_names.empty()); + ASSERT_EQ(column_to->getVariantInfo().variant_name_to_discriminator.at("SharedVariant"), 0); + ASSERT_EQ(column_to->getVariantInfo().variant_names, Names{"SharedVariant"}); +} + +TEST(ColumnDynamic, compare) +{ + auto column_from = ColumnDynamic::create(3); + column_from->insert(Field(42)); + column_from->insert(Field(42.42)); + column_from->insert(Field("str")); + column_from->insert(Field(Null())); + column_from->insert(Field(Array({Field(42)}))); + + ASSERT_EQ(column_from->compareAt(0, 0, *column_from, -1), 0); + ASSERT_EQ(column_from->compareAt(0, 1, *column_from, -1), 1); + ASSERT_EQ(column_from->compareAt(1, 1, *column_from, -1), 0); + ASSERT_EQ(column_from->compareAt(0, 2, *column_from, -1), -1); + ASSERT_EQ(column_from->compareAt(2, 0, *column_from, -1), 1); + ASSERT_EQ(column_from->compareAt(2, 4, *column_from, -1), 1); + ASSERT_EQ(column_from->compareAt(4, 2, *column_from, -1), -1); + ASSERT_EQ(column_from->compareAt(4, 4, *column_from, -1), 0); + ASSERT_EQ(column_from->compareAt(0, 3, *column_from, -1), 1); + ASSERT_EQ(column_from->compareAt(1, 3, *column_from, -1), 1); + ASSERT_EQ(column_from->compareAt(2, 3, *column_from, -1), 1); + ASSERT_EQ(column_from->compareAt(3, 3, *column_from, -1), 0); + ASSERT_EQ(column_from->compareAt(4, 3, *column_from, -1), 1); + ASSERT_EQ(column_from->compareAt(3, 0, *column_from, -1), -1); + ASSERT_EQ(column_from->compareAt(3, 1, *column_from, -1), -1); + ASSERT_EQ(column_from->compareAt(3, 2, *column_from, -1), -1); + ASSERT_EQ(column_from->compareAt(3, 4, *column_from, -1), -1); } diff --git a/src/Columns/tests/gtest_column_variant.cpp b/src/Columns/tests/gtest_column_variant.cpp index 25f276b9600..5e481b88409 100644 --- a/src/Columns/tests/gtest_column_variant.cpp +++ b/src/Columns/tests/gtest_column_variant.cpp @@ -108,10 +108,10 @@ void checkColumnVariant1(ColumnVariant * column) ASSERT_EQ(offsets[1], 0); ASSERT_EQ(offsets[3], 1); ASSERT_TRUE(column->isDefaultAt(2) && column->isDefaultAt(4)); - ASSERT_EQ((*column)[0].get(), 42); - ASSERT_EQ((*column)[1].get(), "Hello"); + ASSERT_EQ((*column)[0].safeGet(), 42); + ASSERT_EQ((*column)[1].safeGet(), "Hello"); ASSERT_TRUE((*column)[2].isNull()); - ASSERT_EQ((*column)[3].get(), "World"); + ASSERT_EQ((*column)[3].safeGet(), "World"); ASSERT_TRUE((*column)[4].isNull()); } @@ -209,9 +209,9 @@ TEST(ColumnVariant, CreateFromDiscriminatorsAndOneFullColumnNoNulls) ASSERT_EQ(offsets[0], 0); ASSERT_EQ(offsets[1], 1); ASSERT_EQ(offsets[2], 2); - ASSERT_EQ((*column)[0].get(), 0); - ASSERT_EQ((*column)[1].get(), 1); - ASSERT_EQ((*column)[2].get(), 2); + ASSERT_EQ((*column)[0].safeGet(), 0); + ASSERT_EQ((*column)[1].safeGet(), 1); + ASSERT_EQ((*column)[2].safeGet(), 2); } TEST(ColumnVariant, CreateFromDiscriminatorsAndOneFullColumnNoNullsWithLocalOrder) @@ -222,9 +222,9 @@ TEST(ColumnVariant, CreateFromDiscriminatorsAndOneFullColumnNoNullsWithLocalOrde ASSERT_EQ(offsets[0], 0); ASSERT_EQ(offsets[1], 1); ASSERT_EQ(offsets[2], 2); - ASSERT_EQ((*column)[0].get(), 0); - ASSERT_EQ((*column)[1].get(), 1); - ASSERT_EQ((*column)[2].get(), 2); + ASSERT_EQ((*column)[0].safeGet(), 0); + ASSERT_EQ((*column)[1].safeGet(), 1); + ASSERT_EQ((*column)[2].safeGet(), 2); ASSERT_EQ(column->localDiscriminatorAt(0), 2); ASSERT_EQ(column->localDiscriminatorAt(1), 2); ASSERT_EQ(column->localDiscriminatorAt(2), 2); @@ -331,9 +331,9 @@ TEST(ColumnVariant, CloneResizedGeneral1) ASSERT_EQ(offsets[0], 0); ASSERT_EQ(offsets[1], 0); ASSERT_EQ(offsets[3], 1); - ASSERT_EQ((*resized_column_variant)[0].get(), 42); - ASSERT_EQ((*resized_column_variant)[1].get(), "Hello"); - ASSERT_EQ((*resized_column_variant)[3].get(), 43); + ASSERT_EQ((*resized_column_variant)[0].safeGet(), 42); + ASSERT_EQ((*resized_column_variant)[1].safeGet(), "Hello"); + ASSERT_EQ((*resized_column_variant)[3].safeGet(), 43); } TEST(ColumnVariant, CloneResizedGeneral2) @@ -367,7 +367,7 @@ TEST(ColumnVariant, CloneResizedGeneral2) ASSERT_EQ(discriminators[2], ColumnVariant::NULL_DISCRIMINATOR); const auto & offsets = resized_column_variant->getOffsets(); ASSERT_EQ(offsets[0], 0); - ASSERT_EQ((*resized_column_variant)[0].get(), 42); + ASSERT_EQ((*resized_column_variant)[0].safeGet(), 42); } TEST(ColumnVariant, CloneResizedGeneral3) @@ -405,10 +405,10 @@ TEST(ColumnVariant, CloneResizedGeneral3) ASSERT_EQ(offsets[1], 0); ASSERT_EQ(offsets[2], 1); ASSERT_EQ(offsets[3], 1); - ASSERT_EQ((*resized_column_variant)[0].get(), 42); - ASSERT_EQ((*resized_column_variant)[1].get(), "Hello"); - ASSERT_EQ((*resized_column_variant)[2].get(), "World"); - ASSERT_EQ((*resized_column_variant)[3].get(), 43); + ASSERT_EQ((*resized_column_variant)[0].safeGet(), 42); + ASSERT_EQ((*resized_column_variant)[1].safeGet(), "Hello"); + ASSERT_EQ((*resized_column_variant)[2].safeGet(), "World"); + ASSERT_EQ((*resized_column_variant)[3].safeGet(), 43); } MutableColumnPtr createDiscriminators2() @@ -465,7 +465,7 @@ TEST(ColumnVariant, InsertFrom) auto column_from = createVariantColumn2(change_order); column_to->insertFrom(*column_from, 3); ASSERT_EQ(column_to->globalDiscriminatorAt(5), 0); - ASSERT_EQ((*column_to)[5].get(), 43); + ASSERT_EQ((*column_to)[5].safeGet(), 43); } } @@ -478,8 +478,8 @@ TEST(ColumnVariant, InsertRangeFromOneColumnNoNulls) column_to->insertRangeFrom(*column_from, 2, 2); ASSERT_EQ(column_to->globalDiscriminatorAt(7), 0); ASSERT_EQ(column_to->globalDiscriminatorAt(8), 0); - ASSERT_EQ((*column_to)[7].get(), 2); - ASSERT_EQ((*column_to)[8].get(), 3); + ASSERT_EQ((*column_to)[7].safeGet(), 2); + ASSERT_EQ((*column_to)[8].safeGet(), 3); } } @@ -494,9 +494,9 @@ TEST(ColumnVariant, InsertRangeFromGeneral) ASSERT_EQ(column_to->globalDiscriminatorAt(6), ColumnVariant::NULL_DISCRIMINATOR); ASSERT_EQ(column_to->globalDiscriminatorAt(7), 0); ASSERT_EQ(column_to->globalDiscriminatorAt(8), 1); - ASSERT_EQ((*column_to)[5].get(), "Hello"); - ASSERT_EQ((*column_to)[7].get(), 43); - ASSERT_EQ((*column_to)[8].get(), "World"); + ASSERT_EQ((*column_to)[5].safeGet(), "Hello"); + ASSERT_EQ((*column_to)[7].safeGet(), 43); + ASSERT_EQ((*column_to)[8].safeGet(), "World"); } } @@ -509,8 +509,8 @@ TEST(ColumnVariant, InsertManyFrom) column_to->insertManyFrom(*column_from, 3, 2); ASSERT_EQ(column_to->globalDiscriminatorAt(5), 0); ASSERT_EQ(column_to->globalDiscriminatorAt(6), 0); - ASSERT_EQ((*column_to)[5].get(), 43); - ASSERT_EQ((*column_to)[6].get(), 43); + ASSERT_EQ((*column_to)[5].safeGet(), 43); + ASSERT_EQ((*column_to)[6].safeGet(), 43); } } @@ -520,8 +520,8 @@ TEST(ColumnVariant, PopBackOneColumnNoNulls) column->popBack(3); ASSERT_EQ(column->size(), 2); ASSERT_EQ(column->getVariantByLocalDiscriminator(0).size(), 2); - ASSERT_EQ((*column)[0].get(), 0); - ASSERT_EQ((*column)[1].get(), 1); + ASSERT_EQ((*column)[0].safeGet(), 0); + ASSERT_EQ((*column)[1].safeGet(), 1); } TEST(ColumnVariant, PopBackGeneral) @@ -531,8 +531,8 @@ TEST(ColumnVariant, PopBackGeneral) ASSERT_EQ(column->size(), 3); ASSERT_EQ(column->getVariantByLocalDiscriminator(0).size(), 1); ASSERT_EQ(column->getVariantByLocalDiscriminator(1).size(), 1); - ASSERT_EQ((*column)[0].get(), 42); - ASSERT_EQ((*column)[1].get(), "Hello"); + ASSERT_EQ((*column)[0].safeGet(), 42); + ASSERT_EQ((*column)[1].safeGet(), "Hello"); ASSERT_TRUE((*column)[2].isNull()); } @@ -545,8 +545,8 @@ TEST(ColumnVariant, FilterOneColumnNoNulls) filter.push_back(1); auto filtered_column = column->filter(filter, -1); ASSERT_EQ(filtered_column->size(), 2); - ASSERT_EQ((*filtered_column)[0].get(), 0); - ASSERT_EQ((*filtered_column)[1].get(), 2); + ASSERT_EQ((*filtered_column)[0].safeGet(), 0); + ASSERT_EQ((*filtered_column)[1].safeGet(), 2); } TEST(ColumnVariant, FilterGeneral) @@ -562,7 +562,7 @@ TEST(ColumnVariant, FilterGeneral) filter.push_back(0); auto filtered_column = column->filter(filter, -1); ASSERT_EQ(filtered_column->size(), 3); - ASSERT_EQ((*filtered_column)[0].get(), "Hello"); + ASSERT_EQ((*filtered_column)[0].safeGet(), "Hello"); ASSERT_TRUE((*filtered_column)[1].isNull()); ASSERT_TRUE((*filtered_column)[2].isNull()); } @@ -577,9 +577,9 @@ TEST(ColumnVariant, PermuteAndIndexOneColumnNoNulls) permutation.push_back(0); auto permuted_column = column->permute(permutation, 3); ASSERT_EQ(permuted_column->size(), 3); - ASSERT_EQ((*permuted_column)[0].get(), 1); - ASSERT_EQ((*permuted_column)[1].get(), 3); - ASSERT_EQ((*permuted_column)[2].get(), 2); + ASSERT_EQ((*permuted_column)[0].safeGet(), 1); + ASSERT_EQ((*permuted_column)[1].safeGet(), 3); + ASSERT_EQ((*permuted_column)[2].safeGet(), 2); auto index = ColumnUInt64::create(); index->getData().push_back(1); @@ -588,9 +588,9 @@ TEST(ColumnVariant, PermuteAndIndexOneColumnNoNulls) index->getData().push_back(0); auto indexed_column = column->index(*index, 3); ASSERT_EQ(indexed_column->size(), 3); - ASSERT_EQ((*indexed_column)[0].get(), 1); - ASSERT_EQ((*indexed_column)[1].get(), 3); - ASSERT_EQ((*indexed_column)[2].get(), 2); + ASSERT_EQ((*indexed_column)[0].safeGet(), 1); + ASSERT_EQ((*indexed_column)[1].safeGet(), 3); + ASSERT_EQ((*indexed_column)[2].safeGet(), 2); } TEST(ColumnVariant, PermuteGeneral) @@ -603,9 +603,9 @@ TEST(ColumnVariant, PermuteGeneral) permutation.push_back(5); auto permuted_column = column->permute(permutation, 4); ASSERT_EQ(permuted_column->size(), 4); - ASSERT_EQ((*permuted_column)[0].get(), 43); - ASSERT_EQ((*permuted_column)[1].get(), "World"); - ASSERT_EQ((*permuted_column)[2].get(), "Hello"); + ASSERT_EQ((*permuted_column)[0].safeGet(), 43); + ASSERT_EQ((*permuted_column)[1].safeGet(), "World"); + ASSERT_EQ((*permuted_column)[2].safeGet(), "Hello"); ASSERT_TRUE((*permuted_column)[3].isNull()); } @@ -618,12 +618,12 @@ TEST(ColumnVariant, ReplicateOneColumnNoNull) offsets.push_back(6); auto replicated_column = column->replicate(offsets); ASSERT_EQ(replicated_column->size(), 6); - ASSERT_EQ((*replicated_column)[0].get(), 1); - ASSERT_EQ((*replicated_column)[1].get(), 1); - ASSERT_EQ((*replicated_column)[2].get(), 1); - ASSERT_EQ((*replicated_column)[3].get(), 2); - ASSERT_EQ((*replicated_column)[4].get(), 2); - ASSERT_EQ((*replicated_column)[5].get(), 2); + ASSERT_EQ((*replicated_column)[0].safeGet(), 1); + ASSERT_EQ((*replicated_column)[1].safeGet(), 1); + ASSERT_EQ((*replicated_column)[2].safeGet(), 1); + ASSERT_EQ((*replicated_column)[3].safeGet(), 2); + ASSERT_EQ((*replicated_column)[4].safeGet(), 2); + ASSERT_EQ((*replicated_column)[5].safeGet(), 2); } TEST(ColumnVariant, ReplicateGeneral) @@ -637,9 +637,9 @@ TEST(ColumnVariant, ReplicateGeneral) offsets.push_back(7); auto replicated_column = column->replicate(offsets); ASSERT_EQ(replicated_column->size(), 7); - ASSERT_EQ((*replicated_column)[0].get(), 42); - ASSERT_EQ((*replicated_column)[1].get(), "Hello"); - ASSERT_EQ((*replicated_column)[2].get(), "Hello"); + ASSERT_EQ((*replicated_column)[0].safeGet(), 42); + ASSERT_EQ((*replicated_column)[1].safeGet(), "Hello"); + ASSERT_EQ((*replicated_column)[2].safeGet(), "Hello"); ASSERT_TRUE((*replicated_column)[3].isNull()); ASSERT_TRUE((*replicated_column)[4].isNull()); ASSERT_TRUE((*replicated_column)[5].isNull()); @@ -657,13 +657,13 @@ TEST(ColumnVariant, ScatterOneColumnNoNulls) selector.push_back(1); auto columns = column->scatter(3, selector); ASSERT_EQ(columns[0]->size(), 2); - ASSERT_EQ((*columns[0])[0].get(), 0); - ASSERT_EQ((*columns[0])[1].get(), 3); + ASSERT_EQ((*columns[0])[0].safeGet(), 0); + ASSERT_EQ((*columns[0])[1].safeGet(), 3); ASSERT_EQ(columns[1]->size(), 2); - ASSERT_EQ((*columns[1])[0].get(), 1); - ASSERT_EQ((*columns[1])[1].get(), 4); + ASSERT_EQ((*columns[1])[0].safeGet(), 1); + ASSERT_EQ((*columns[1])[1].safeGet(), 4); ASSERT_EQ(columns[2]->size(), 1); - ASSERT_EQ((*columns[2])[0].get(), 2); + ASSERT_EQ((*columns[2])[0].safeGet(), 2); } TEST(ColumnVariant, ScatterGeneral) @@ -680,12 +680,12 @@ TEST(ColumnVariant, ScatterGeneral) auto columns = column->scatter(3, selector); ASSERT_EQ(columns[0]->size(), 3); - ASSERT_EQ((*columns[0])[0].get(), 42); - ASSERT_EQ((*columns[0])[1].get(), "Hello"); - ASSERT_EQ((*columns[0])[2].get(), 43); + ASSERT_EQ((*columns[0])[0].safeGet(), 42); + ASSERT_EQ((*columns[0])[1].safeGet(), "Hello"); + ASSERT_EQ((*columns[0])[2].safeGet(), 43); ASSERT_EQ(columns[1]->size(), 2); - ASSERT_EQ((*columns[1])[0].get(), "World"); - ASSERT_EQ((*columns[1])[1].get(), 44); + ASSERT_EQ((*columns[1])[0].safeGet(), "World"); + ASSERT_EQ((*columns[1])[1].safeGet(), 44); ASSERT_EQ(columns[2]->size(), 2); ASSERT_TRUE((*columns[2])[0].isNull()); ASSERT_TRUE((*columns[2])[1].isNull()); diff --git a/src/Columns/tests/gtest_low_cardinality.cpp b/src/Columns/tests/gtest_low_cardinality.cpp index 5e01279b7df..ce16d2cadb1 100644 --- a/src/Columns/tests/gtest_low_cardinality.cpp +++ b/src/Columns/tests/gtest_low_cardinality.cpp @@ -20,13 +20,13 @@ void testLowCardinalityNumberInsert(const DataTypePtr & data_type) Field value; column->get(0, value); - ASSERT_EQ(value.get(), 15); + ASSERT_EQ(value.safeGet(), 15); column->get(1, value); - ASSERT_EQ(value.get(), 20); + ASSERT_EQ(value.safeGet(), 20); column->get(2, value); - ASSERT_EQ(value.get(), 25); + ASSERT_EQ(value.safeGet(), 25); } TEST(ColumnLowCardinality, Insert) diff --git a/src/Common/AsynchronousMetrics.cpp b/src/Common/AsynchronousMetrics.cpp index 6309f6079f6..9b6a7428411 100644 --- a/src/Common/AsynchronousMetrics.cpp +++ b/src/Common/AsynchronousMetrics.cpp @@ -1,18 +1,24 @@ -#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" @@ -52,6 +58,12 @@ static std::unique_ptr openFileIfExists(const std::stri return {}; } +static void openCgroupv2MetricFile(const std::string & filename, std::optional & out) +{ + if (auto path = getCgroupsV2PathContainingFile(filename)) + openFileIfExists((path.value() + filename).c_str(), out); +}; + #endif @@ -63,21 +75,15 @@ AsynchronousMetrics::AsynchronousMetrics( , protocol_server_metrics_func(protocol_server_metrics_func_) { #if defined(OS_LINUX) - openFileIfExists("/proc/meminfo", meminfo); - openFileIfExists("/proc/loadavg", loadavg); - openFileIfExists("/proc/stat", proc_stat); openFileIfExists("/proc/cpuinfo", cpuinfo); openFileIfExists("/proc/sys/fs/file-nr", file_nr); - openFileIfExists("/proc/uptime", uptime); openFileIfExists("/proc/net/dev", net_dev); /// CGroups v2 - openFileIfExists("/sys/fs/cgroup/memory.max", cgroupmem_limit_in_bytes); - if (cgroupmem_limit_in_bytes) - { - openFileIfExists("/sys/fs/cgroup/memory.current", cgroupmem_usage_in_bytes); - } - openFileIfExists("/sys/fs/cgroup/cpu.max", cgroupcpu_max); + openCgroupv2MetricFile("memory.max", cgroupmem_limit_in_bytes); + openCgroupv2MetricFile("memory.current", cgroupmem_usage_in_bytes); + openCgroupv2MetricFile("cpu.max", cgroupcpu_max); + openCgroupv2MetricFile("cpu.stat", cgroupcpu_stat); /// CGroups v1 if (!cgroupmem_limit_in_bytes) @@ -90,6 +96,21 @@ AsynchronousMetrics::AsynchronousMetrics( openFileIfExists("/sys/fs/cgroup/cpu/cpu.cfs_period_us", cgroupcpu_cfs_period); openFileIfExists("/sys/fs/cgroup/cpu/cpu.cfs_quota_us", cgroupcpu_cfs_quota); } + if (!cgroupcpu_stat) + openFileIfExists("/sys/fs/cgroup/cpuacct/cpuacct.stat", cgroupcpuacct_stat); + + if (!cgroupcpu_stat && !cgroupcpuacct_stat) + { + /// The following metrics are not cgroup-aware and we've found cgroup-specific metric files for the similar metrics, + /// so we're better not reporting them at all to avoid confusion + openFileIfExists("/proc/loadavg", loadavg); + openFileIfExists("/proc/stat", proc_stat); + openFileIfExists("/proc/uptime", uptime); + } + + /// The same story for memory metrics + if (!cgroupmem_limit_in_bytes) + openFileIfExists("/proc/meminfo", meminfo); openFileIfExists("/proc/sys/vm/max_map_count", vm_max_map_count); openFileIfExists("/proc/self/maps", vm_maps); @@ -570,6 +591,151 @@ AsynchronousMetrics::NetworkInterfaceStatValues::operator-(const AsynchronousMet #endif +#if defined(OS_LINUX) +void AsynchronousMetrics::applyCPUMetricsUpdate( + AsynchronousMetricValues & new_values, const std::string & cpu_suffix, const ProcStatValuesCPU & delta_values, double multiplier) +{ + new_values["OSUserTime" + cpu_suffix] + = {delta_values.user * multiplier, + "The ratio of time the CPU core was running userspace code. This is a system-wide metric, it includes all the processes on the " + "host machine, not just clickhouse-server." + " This includes also the time when the CPU was under-utilized due to the reasons internal to the CPU (memory loads, pipeline " + "stalls, branch mispredictions, running another SMT core)." + " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across " + "them [0..num cores]."}; + new_values["OSNiceTime" + cpu_suffix] + = {delta_values.nice * multiplier, + "The ratio of time the CPU core was running userspace code with higher priority. This is a system-wide metric, it includes all " + "the processes on the host machine, not just clickhouse-server." + " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across " + "them [0..num cores]."}; + new_values["OSSystemTime" + cpu_suffix] + = {delta_values.system * multiplier, + "The ratio of time the CPU core was running OS kernel (system) code. This is a system-wide metric, it includes all the " + "processes on the host machine, not just clickhouse-server." + " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across " + "them [0..num cores]."}; + new_values["OSIdleTime" + cpu_suffix] + = {delta_values.idle * multiplier, + "The ratio of time the CPU core was idle (not even ready to run a process waiting for IO) from the OS kernel standpoint. This " + "is a system-wide metric, it includes all the processes on the host machine, not just clickhouse-server." + " This does not include the time when the CPU was under-utilized due to the reasons internal to the CPU (memory loads, pipeline " + "stalls, branch mispredictions, running another SMT core)." + " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across " + "them [0..num cores]."}; + new_values["OSIOWaitTime" + cpu_suffix] + = {delta_values.iowait * multiplier, + "The ratio of time the CPU core was not running the code but when the OS kernel did not run any other process on this CPU as " + "the processes were waiting for IO. This is a system-wide metric, it includes all the processes on the host machine, not just " + "clickhouse-server." + " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across " + "them [0..num cores]."}; + new_values["OSIrqTime" + cpu_suffix] + = {delta_values.irq * multiplier, + "The ratio of time spent for running hardware interrupt requests on the CPU. This is a system-wide metric, it includes all the " + "processes on the host machine, not just clickhouse-server." + " A high number of this metric may indicate hardware misconfiguration or a very high network load." + " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across " + "them [0..num cores]."}; + new_values["OSSoftIrqTime" + cpu_suffix] + = {delta_values.softirq * multiplier, + "The ratio of time spent for running software interrupt requests on the CPU. This is a system-wide metric, it includes all the " + "processes on the host machine, not just clickhouse-server." + " A high number of this metric may indicate inefficient software running on the system." + " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across " + "them [0..num cores]."}; + new_values["OSStealTime" + cpu_suffix] + = {delta_values.steal * multiplier, + "The ratio of time spent in other operating systems by the CPU when running in a virtualized environment. This is a system-wide " + "metric, it includes all the processes on the host machine, not just clickhouse-server." + " Not every virtualized environments present this metric, and most of them don't." + " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across " + "them [0..num cores]."}; + new_values["OSGuestTime" + cpu_suffix] + = {delta_values.guest * multiplier, + "The ratio of time spent running a virtual CPU for guest operating systems under the control of the Linux kernel (See `man " + "procfs`). This is a system-wide metric, it includes all the processes on the host machine, not just clickhouse-server." + " This metric is irrelevant for ClickHouse, but still exists for completeness." + " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across " + "them [0..num cores]."}; + new_values["OSGuestNiceTime" + cpu_suffix] + = {delta_values.guest_nice * multiplier, + "The ratio of time spent running a virtual CPU for guest operating systems under the control of the Linux kernel, when a guest " + "was set to a higher priority (See `man procfs`). This is a system-wide metric, it includes all the processes on the host " + "machine, not just clickhouse-server." + " This metric is irrelevant for ClickHouse, but still exists for completeness." + " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across " + "them [0..num cores]."}; +} + +void AsynchronousMetrics::applyNormalizedCPUMetricsUpdate( + AsynchronousMetricValues & new_values, double num_cpus_to_normalize, const ProcStatValuesCPU & delta_values_all_cpus, double multiplier) +{ + chassert(num_cpus_to_normalize); + + new_values["OSUserTimeNormalized"] + = {delta_values_all_cpus.user * multiplier / num_cpus_to_normalize, + "The value is similar to `OSUserTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless " + "of the number of cores." + " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is " + "non-uniform, and still get the average resource utilization metric."}; + new_values["OSNiceTimeNormalized"] + = {delta_values_all_cpus.nice * multiplier / num_cpus_to_normalize, + "The value is similar to `OSNiceTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless " + "of the number of cores." + " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is " + "non-uniform, and still get the average resource utilization metric."}; + new_values["OSSystemTimeNormalized"] + = {delta_values_all_cpus.system * multiplier / num_cpus_to_normalize, + "The value is similar to `OSSystemTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless " + "of the number of cores." + " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is " + "non-uniform, and still get the average resource utilization metric."}; + new_values["OSIdleTimeNormalized"] + = {delta_values_all_cpus.idle * multiplier / num_cpus_to_normalize, + "The value is similar to `OSIdleTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless " + "of the number of cores." + " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is " + "non-uniform, and still get the average resource utilization metric."}; + new_values["OSIOWaitTimeNormalized"] + = {delta_values_all_cpus.iowait * multiplier / num_cpus_to_normalize, + "The value is similar to `OSIOWaitTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless " + "of the number of cores." + " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is " + "non-uniform, and still get the average resource utilization metric."}; + new_values["OSIrqTimeNormalized"] + = {delta_values_all_cpus.irq * multiplier / num_cpus_to_normalize, + "The value is similar to `OSIrqTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless of " + "the number of cores." + " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is " + "non-uniform, and still get the average resource utilization metric."}; + new_values["OSSoftIrqTimeNormalized"] + = {delta_values_all_cpus.softirq * multiplier / num_cpus_to_normalize, + "The value is similar to `OSSoftIrqTime` but divided to the number of CPU cores to be measured in the [0..1] interval " + "regardless of the number of cores." + " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is " + "non-uniform, and still get the average resource utilization metric."}; + new_values["OSStealTimeNormalized"] + = {delta_values_all_cpus.steal * multiplier / num_cpus_to_normalize, + "The value is similar to `OSStealTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless " + "of the number of cores." + " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is " + "non-uniform, and still get the average resource utilization metric."}; + new_values["OSGuestTimeNormalized"] + = {delta_values_all_cpus.guest * multiplier / num_cpus_to_normalize, + "The value is similar to `OSGuestTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless " + "of the number of cores." + " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is " + "non-uniform, and still get the average resource utilization metric."}; + new_values["OSGuestNiceTimeNormalized"] + = {delta_values_all_cpus.guest_nice * multiplier / num_cpus_to_normalize, + "The value is similar to `OSGuestNiceTime` but divided to the number of CPU cores to be measured in the [0..1] interval " + "regardless of the number of cores." + " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is " + "non-uniform, and still get the average resource utilization metric."}; +} +#endif + void AsynchronousMetrics::update(TimePoint update_time, bool force_update) { Stopwatch watch; @@ -831,7 +997,68 @@ void AsynchronousMetrics::update(TimePoint update_time, bool force_update) new_values["CGroupMaxCPU"] = { max_cpu_cgroups, "The maximum number of CPU cores according to CGroups."}; } - if (proc_stat) + if (cgroupcpu_stat || cgroupcpuacct_stat) + { + try + { + ReadBufferFromFilePRead & in = cgroupcpu_stat ? *cgroupcpu_stat : *cgroupcpuacct_stat; + ProcStatValuesCPU current_values{}; + + /// We re-read the file from the beginning each time + in.rewind(); + + while (!in.eof()) + { + String name; + readStringUntilWhitespace(name, in); + skipWhitespaceIfAny(in); + + /// `user_usec` for cgroup v2 and `user` for cgroup v1 + if (name.starts_with("user")) + { + readText(current_values.user, in); + skipToNextLineOrEOF(in); + } + /// `system_usec` for cgroup v2 and `system` for cgroup v1 + else if (name.starts_with("system")) + { + readText(current_values.system, in); + skipToNextLineOrEOF(in); + } + else + skipToNextLineOrEOF(in); + } + + if (!first_run) + { + auto get_clock_ticks = [&]() + { + if (auto hz = sysconf(_SC_CLK_TCK); hz != -1) + return hz; + else + throw ErrnoException(ErrorCodes::CANNOT_SYSCONF, "Cannot call 'sysconf' to obtain system HZ"); + }; + const auto cgroup_version_specific_divisor = cgroupcpu_stat ? 1e6 : get_clock_ticks(); + const double multiplier = 1.0 / cgroup_version_specific_divisor + / (std::chrono::duration_cast(time_since_previous_update).count() / 1e9); + + const ProcStatValuesCPU delta_values = current_values - proc_stat_values_all_cpus; + applyCPUMetricsUpdate(new_values, /*cpu_suffix=*/"", delta_values, multiplier); + if (max_cpu_cgroups > 0) + applyNormalizedCPUMetricsUpdate(new_values, max_cpu_cgroups, delta_values, multiplier); + } + + proc_stat_values_all_cpus = current_values; + } + catch (...) + { + tryLogCurrentException(__PRETTY_FUNCTION__); + openCgroupv2MetricFile("cpu.stat", cgroupcpu_stat); + if (!cgroupcpu_stat) + openFileIfExists("/sys/fs/cgroup/cpuacct/cpuacct.stat", cgroupcpuacct_stat); + } + } + else if (proc_stat) { try { @@ -886,43 +1113,7 @@ void AsynchronousMetrics::update(TimePoint update_time, bool force_update) else delta_values_all_cpus = delta_values; - new_values["OSUserTime" + cpu_suffix] = { delta_values.user * multiplier, - "The ratio of time the CPU core was running userspace code. This is a system-wide metric, it includes all the processes on the host machine, not just clickhouse-server." - " This includes also the time when the CPU was under-utilized due to the reasons internal to the CPU (memory loads, pipeline stalls, branch mispredictions, running another SMT core)." - " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across them [0..num cores]."}; - new_values["OSNiceTime" + cpu_suffix] = { delta_values.nice * multiplier, - "The ratio of time the CPU core was running userspace code with higher priority. This is a system-wide metric, it includes all the processes on the host machine, not just clickhouse-server." - " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across them [0..num cores]."}; - new_values["OSSystemTime" + cpu_suffix] = { delta_values.system * multiplier, - "The ratio of time the CPU core was running OS kernel (system) code. This is a system-wide metric, it includes all the processes on the host machine, not just clickhouse-server." - " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across them [0..num cores]."}; - new_values["OSIdleTime" + cpu_suffix] = { delta_values.idle * multiplier, - "The ratio of time the CPU core was idle (not even ready to run a process waiting for IO) from the OS kernel standpoint. This is a system-wide metric, it includes all the processes on the host machine, not just clickhouse-server." - " This does not include the time when the CPU was under-utilized due to the reasons internal to the CPU (memory loads, pipeline stalls, branch mispredictions, running another SMT core)." - " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across them [0..num cores]."}; - new_values["OSIOWaitTime" + cpu_suffix] = { delta_values.iowait * multiplier, - "The ratio of time the CPU core was not running the code but when the OS kernel did not run any other process on this CPU as the processes were waiting for IO. This is a system-wide metric, it includes all the processes on the host machine, not just clickhouse-server." - " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across them [0..num cores]."}; - new_values["OSIrqTime" + cpu_suffix] = { delta_values.irq * multiplier, - "The ratio of time spent for running hardware interrupt requests on the CPU. This is a system-wide metric, it includes all the processes on the host machine, not just clickhouse-server." - " A high number of this metric may indicate hardware misconfiguration or a very high network load." - " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across them [0..num cores]."}; - new_values["OSSoftIrqTime" + cpu_suffix] = { delta_values.softirq * multiplier, - "The ratio of time spent for running software interrupt requests on the CPU. This is a system-wide metric, it includes all the processes on the host machine, not just clickhouse-server." - " A high number of this metric may indicate inefficient software running on the system." - " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across them [0..num cores]."}; - new_values["OSStealTime" + cpu_suffix] = { delta_values.steal * multiplier, - "The ratio of time spent in other operating systems by the CPU when running in a virtualized environment. This is a system-wide metric, it includes all the processes on the host machine, not just clickhouse-server." - " Not every virtualized environments present this metric, and most of them don't." - " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across them [0..num cores]."}; - new_values["OSGuestTime" + cpu_suffix] = { delta_values.guest * multiplier, - "The ratio of time spent running a virtual CPU for guest operating systems under the control of the Linux kernel (See `man procfs`). This is a system-wide metric, it includes all the processes on the host machine, not just clickhouse-server." - " This metric is irrelevant for ClickHouse, but still exists for completeness." - " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across them [0..num cores]."}; - new_values["OSGuestNiceTime" + cpu_suffix] = { delta_values.guest_nice * multiplier, - "The ratio of time spent running a virtual CPU for guest operating systems under the control of the Linux kernel, when a guest was set to a higher priority (See `man procfs`). This is a system-wide metric, it includes all the processes on the host machine, not just clickhouse-server." - " This metric is irrelevant for ClickHouse, but still exists for completeness." - " The value for a single CPU core will be in the interval [0..1]. The value for all CPU cores is calculated as a sum across them [0..num cores]."}; + applyCPUMetricsUpdate(new_values, cpu_suffix, delta_values, multiplier); } prev_values = current_values; @@ -978,38 +1169,7 @@ void AsynchronousMetrics::update(TimePoint update_time, bool force_update) Float64 num_cpus_to_normalize = max_cpu_cgroups > 0 ? max_cpu_cgroups : num_cpus; if (num_cpus_to_normalize > 0) - { - new_values["OSUserTimeNormalized"] = { delta_values_all_cpus.user * multiplier / num_cpus_to_normalize, - "The value is similar to `OSUserTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless of the number of cores." - " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is non-uniform, and still get the average resource utilization metric."}; - new_values["OSNiceTimeNormalized"] = { delta_values_all_cpus.nice * multiplier / num_cpus_to_normalize, - "The value is similar to `OSNiceTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless of the number of cores." - " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is non-uniform, and still get the average resource utilization metric."}; - new_values["OSSystemTimeNormalized"] = { delta_values_all_cpus.system * multiplier / num_cpus_to_normalize, - "The value is similar to `OSSystemTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless of the number of cores." - " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is non-uniform, and still get the average resource utilization metric."}; - new_values["OSIdleTimeNormalized"] = { delta_values_all_cpus.idle * multiplier / num_cpus_to_normalize, - "The value is similar to `OSIdleTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless of the number of cores." - " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is non-uniform, and still get the average resource utilization metric."}; - new_values["OSIOWaitTimeNormalized"] = { delta_values_all_cpus.iowait * multiplier / num_cpus_to_normalize, - "The value is similar to `OSIOWaitTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless of the number of cores." - " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is non-uniform, and still get the average resource utilization metric."}; - new_values["OSIrqTimeNormalized"] = { delta_values_all_cpus.irq * multiplier / num_cpus_to_normalize, - "The value is similar to `OSIrqTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless of the number of cores." - " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is non-uniform, and still get the average resource utilization metric."}; - new_values["OSSoftIrqTimeNormalized"] = { delta_values_all_cpus.softirq * multiplier / num_cpus_to_normalize, - "The value is similar to `OSSoftIrqTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless of the number of cores." - " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is non-uniform, and still get the average resource utilization metric."}; - new_values["OSStealTimeNormalized"] = { delta_values_all_cpus.steal * multiplier / num_cpus_to_normalize, - "The value is similar to `OSStealTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless of the number of cores." - " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is non-uniform, and still get the average resource utilization metric."}; - new_values["OSGuestTimeNormalized"] = { delta_values_all_cpus.guest * multiplier / num_cpus_to_normalize, - "The value is similar to `OSGuestTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless of the number of cores." - " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is non-uniform, and still get the average resource utilization metric."}; - new_values["OSGuestNiceTimeNormalized"] = { delta_values_all_cpus.guest_nice * multiplier / num_cpus_to_normalize, - "The value is similar to `OSGuestNiceTime` but divided to the number of CPU cores to be measured in the [0..1] interval regardless of the number of cores." - " This allows you to average the values of this metric across multiple servers in a cluster even if the number of cores is non-uniform, and still get the average resource utilization metric."}; - } + applyNormalizedCPUMetricsUpdate(new_values, num_cpus_to_normalize, delta_values_all_cpus, multiplier); } proc_stat_values_other = current_other_values; @@ -1042,8 +1202,7 @@ void AsynchronousMetrics::update(TimePoint update_time, bool force_update) tryLogCurrentException(__PRETTY_FUNCTION__); } } - - if (meminfo) + else if (meminfo) { try { @@ -1613,7 +1772,7 @@ void AsynchronousMetrics::update(TimePoint update_time, bool force_update) #endif { - auto get_metric_name_doc = [](const String & name) -> std::pair + auto threads_get_metric_name_doc = [](const String & name) -> std::pair { static std::map> metric_map = { @@ -1637,11 +1796,38 @@ void AsynchronousMetrics::update(TimePoint update_time, bool force_update) return it->second; }; + auto rejected_connections_get_metric_name_doc = [](const String & name) -> std::pair + { + static std::map> metric_map = + { + {"tcp_port", {"TCPRejectedConnections", "Number of rejected connections for the TCP protocol (without TLS)."}}, + {"tcp_port_secure", {"TCPSecureRejectedConnections", "Number of rejected connections for the TCP protocol (with TLS)."}}, + {"http_port", {"HTTPRejectedConnections", "Number of rejected connections for the HTTP interface (without TLS)."}}, + {"https_port", {"HTTPSecureRejectedConnections", "Number of rejected connections for the HTTPS interface."}}, + {"interserver_http_port", {"InterserverRejectedConnections", "Number of rejected connections for the replicas communication protocol (without TLS)."}}, + {"interserver_https_port", {"InterserverSecureRejectedConnections", "Number of rejected connections for the replicas communication protocol (with TLS)."}}, + {"mysql_port", {"MySQLRejectedConnections", "Number of rejected connections for the MySQL compatibility protocol."}}, + {"postgresql_port", {"PostgreSQLRejectedConnections", "Number of rejected connections for the PostgreSQL compatibility protocol."}}, + {"grpc_port", {"GRPCRejectedConnections", "Number of rejected connections for the GRPC protocol."}}, + {"prometheus.port", {"PrometheusRejectedConnections", "Number of rejected connections for the Prometheus endpoint. Note: prometheus endpoints can be also used via the usual HTTP/HTTPs ports."}}, + {"keeper_server.tcp_port", {"KeeperTCPRejectedConnections", "Number of rejected connections for the Keeper TCP protocol (without TLS)."}}, + {"keeper_server.tcp_port_secure", {"KeeperTCPSecureRejectedConnections", "Number of rejected connections for the Keeper TCP protocol (with TLS)."}} + }; + auto it = metric_map.find(name); + if (it == metric_map.end()) + return { nullptr, nullptr }; + else + return it->second; + }; + const auto server_metrics = protocol_server_metrics_func(); for (const auto & server_metric : server_metrics) { - if (auto name_doc = get_metric_name_doc(server_metric.port_name); name_doc.first != nullptr) + if (auto name_doc = threads_get_metric_name_doc(server_metric.port_name); name_doc.first != nullptr) new_values[name_doc.first] = { server_metric.current_threads, name_doc.second }; + + if (auto name_doc = rejected_connections_get_metric_name_doc(server_metric.port_name); name_doc.first != nullptr) + new_values[name_doc.first] = { server_metric.rejected_connections, name_doc.second }; } } diff --git a/src/Common/AsynchronousMetrics.h b/src/Common/AsynchronousMetrics.h index 10a972d2458..78d07ef4b6c 100644 --- a/src/Common/AsynchronousMetrics.h +++ b/src/Common/AsynchronousMetrics.h @@ -42,6 +42,7 @@ struct ProtocolServerMetrics { String port_name; size_t current_threads; + size_t rejected_connections; }; /** Periodically (by default, each second) @@ -125,6 +126,8 @@ private: std::optional cgroupcpu_cfs_period TSA_GUARDED_BY(data_mutex); std::optional cgroupcpu_cfs_quota TSA_GUARDED_BY(data_mutex); std::optional cgroupcpu_max TSA_GUARDED_BY(data_mutex); + std::optional cgroupcpu_stat TSA_GUARDED_BY(data_mutex); + std::optional cgroupcpuacct_stat TSA_GUARDED_BY(data_mutex); std::optional vm_max_map_count TSA_GUARDED_BY(data_mutex); std::optional vm_maps TSA_GUARDED_BY(data_mutex); @@ -220,6 +223,16 @@ private: void openBlockDevices(); void openSensorsChips(); void openEDAC(); + + void applyCPUMetricsUpdate( + AsynchronousMetricValues & new_values, const std::string & cpu_suffix, const ProcStatValuesCPU & delta_values, double multiplier); + + void applyNormalizedCPUMetricsUpdate( + AsynchronousMetricValues & new_values, + double num_cpus_to_normalize, + const ProcStatValuesCPU & delta_values_all_cpus, + double multiplier); + #endif void run(); diff --git a/src/Common/CgroupsMemoryUsageObserver.cpp b/src/Common/CgroupsMemoryUsageObserver.cpp index ef8bdfc1823..83b04360164 100644 --- a/src/Common/CgroupsMemoryUsageObserver.cpp +++ b/src/Common/CgroupsMemoryUsageObserver.cpp @@ -144,31 +144,6 @@ private: /// - 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"; @@ -179,7 +154,7 @@ std::optional getCgroupsV1Path() std::pair getCgroupsPath() { - auto v2_path = getCgroupsV2Path(); + auto v2_path = getCgroupsV2PathContainingFile("memory.current"); if (v2_path.has_value()) return {*v2_path, CgroupsMemoryUsageObserver::CgroupsVersion::V2}; diff --git a/src/Common/CollectionOfDerived.h b/src/Common/CollectionOfDerived.h index 9f80ff727b4..bcbcc36c67a 100644 --- a/src/Common/CollectionOfDerived.h +++ b/src/Common/CollectionOfDerived.h @@ -168,7 +168,7 @@ private: records.emplace(it, type_idx, item); } - Records::const_iterator getImpl(std::type_index type_idx) const + typename Records::const_iterator getImpl(std::type_index type_idx) const { auto it = std::lower_bound(records.cbegin(), records.cend(), type_idx); diff --git a/src/Common/ConcurrentBoundedQueue.h b/src/Common/ConcurrentBoundedQueue.h index 16b9488c98d..a830ae157a5 100644 --- a/src/Common/ConcurrentBoundedQueue.h +++ b/src/Common/ConcurrentBoundedQueue.h @@ -243,7 +243,7 @@ public: } /// Clear and finish queue - void clearAndFinish() + void clearAndFinish() noexcept { { std::lock_guard lock(queue_mutex); diff --git a/src/Common/CurrentMetrics.cpp b/src/Common/CurrentMetrics.cpp index b6dd14d292c..67890568941 100644 --- a/src/Common/CurrentMetrics.cpp +++ b/src/Common/CurrentMetrics.cpp @@ -307,7 +307,7 @@ M(FilteringMarksWithPrimaryKey, "Number of threads currently doing filtering of mark ranges by the primary key") \ M(FilteringMarksWithSecondaryKeys, "Number of threads currently doing filtering of mark ranges by secondary keys") \ \ - M(S3DiskNoKeyErrors, "The number of `NoSuchKey` errors that occur when reading data from S3 cloud storage through ClickHouse disks.") \ + M(DiskS3NoSuchKeyErrors, "The number of `NoSuchKey` errors that occur when reading data from S3 cloud storage through ClickHouse disks.") \ #ifdef APPLY_FOR_EXTERNAL_METRICS #define APPLY_FOR_METRICS(M) APPLY_FOR_BUILTIN_METRICS(M) APPLY_FOR_EXTERNAL_METRICS(M) diff --git a/src/Common/Dwarf.cpp b/src/Common/Dwarf.cpp index 1f22e3b05aa..37d0d51db91 100644 --- a/src/Common/Dwarf.cpp +++ b/src/Common/Dwarf.cpp @@ -1029,7 +1029,8 @@ bool Dwarf::findLocation( const LocationInfoMode mode, CompilationUnit & cu, LocationInfo & info, - std::vector & inline_frames) const + std::vector & inline_frames, + bool assume_in_cu_range) const { Die die = getDieAtOffset(cu, cu.first_die); // Partial compilation unit (DW_TAG_partial_unit) is not supported. @@ -1041,6 +1042,11 @@ bool Dwarf::findLocation( std::optional main_file_name; std::optional base_addr_cu; + std::optional low_pc; + std::optional high_pc; + std::optional is_high_pc_addr; + std::optional range_offset; + forEachAttribute(cu, die, [&](const Attribute & attr) { switch (attr.spec.name) // NOLINT(bugprone-switch-missing-default-case) @@ -1058,18 +1064,47 @@ bool Dwarf::findLocation( // File name of main file being compiled main_file_name = std::get(attr.attr_value); break; - case DW_AT_low_pc: case DW_AT_entry_pc: // 2.17.1: historically DW_AT_low_pc was used. DW_AT_entry_pc was // introduced in DWARF3. Support either to determine the base address of // the CU. base_addr_cu = std::get(attr.attr_value); break; + case DW_AT_ranges: + range_offset = std::get(attr.attr_value); + break; + case DW_AT_low_pc: + low_pc = std::get(attr.attr_value); + base_addr_cu = std::get(attr.attr_value); + break; + case DW_AT_high_pc: + // The value of the DW_AT_high_pc attribute can be + // an address (DW_FORM_addr*) or an offset (DW_FORM_data*). + is_high_pc_addr = attr.spec.form == DW_FORM_addr || // + attr.spec.form == DW_FORM_addrx || // + attr.spec.form == DW_FORM_addrx1 || // + attr.spec.form == DW_FORM_addrx2 || // + attr.spec.form == DW_FORM_addrx3 || // + attr.spec.form == DW_FORM_addrx4; + high_pc = std::get(attr.attr_value); + break; } // Iterate through all attributes until find all above. return true; }); + /// Check if the address falls inside this unit's address ranges. + if (!assume_in_cu_range && ((low_pc && high_pc) || range_offset)) + { + bool pc_match = low_pc && high_pc && is_high_pc_addr && address >= *low_pc + && (address < (*is_high_pc_addr ? *high_pc : *low_pc + *high_pc)); + bool range_match = range_offset && isAddrInRangeList(cu, address, base_addr_cu, range_offset.value(), cu.addr_size); + if (!pc_match && !range_match) + { + return false; + } + } + if (main_file_name) { info.has_main_file = true; @@ -1442,7 +1477,7 @@ bool Dwarf::findAddress( { return false; } - findLocation(address, mode, unit, locationInfo, inline_frames); + findLocation(address, mode, unit, locationInfo, inline_frames, /*assume_in_cu_range*/ true); return locationInfo.has_file_and_line; } else if (mode == LocationInfoMode::FAST) @@ -1471,7 +1506,7 @@ bool Dwarf::findAddress( { continue; } - findLocation(address, mode, unit, locationInfo, inline_frames); + findLocation(address, mode, unit, locationInfo, inline_frames, /*assume_in_cu_range*/ false); } return locationInfo.has_file_and_line; diff --git a/src/Common/Dwarf.h b/src/Common/Dwarf.h index da18b3affa0..d754191bfa9 100644 --- a/src/Common/Dwarf.h +++ b/src/Common/Dwarf.h @@ -283,7 +283,8 @@ private: LocationInfoMode mode, CompilationUnit & cu, LocationInfo & info, - std::vector & inline_frames) const; + std::vector & inline_frames, + bool assume_in_cu_range) const; /** * Finds a subprogram debugging info entry that contains a given address among diff --git a/src/Common/ErrorCodes.cpp b/src/Common/ErrorCodes.cpp index b1b8e2367a4..44a1cd071cb 100644 --- a/src/Common/ErrorCodes.cpp +++ b/src/Common/ErrorCodes.cpp @@ -604,6 +604,10 @@ M(723, PARQUET_EXCEPTION) \ M(724, TOO_MANY_TABLES) \ M(725, TOO_MANY_DATABASES) \ + M(726, UNEXPECTED_HTTP_HEADERS) \ + M(727, UNEXPECTED_TABLE_ENGINE) \ + M(728, UNEXPECTED_DATA_TYPE) \ + M(729, ILLEGAL_TIME_SERIES_TAGS) \ \ M(900, DISTRIBUTED_CACHE_ERROR) \ M(901, CANNOT_USE_DISTRIBUTED_CACHE) \ diff --git a/src/Common/ErrorHandlers.h b/src/Common/ErrorHandlers.h index a4a7c4683aa..4e7d391e66f 100644 --- a/src/Common/ErrorHandlers.h +++ b/src/Common/ErrorHandlers.h @@ -2,6 +2,7 @@ #include #include +#include /** ErrorHandler for Poco::Thread, @@ -26,8 +27,32 @@ public: void exception(const std::exception &) override { logException(); } void exception() override { logException(); } + void logMessageImpl(Poco::Message::Priority priority, const std::string & msg) override + { + switch (priority) + { + case Poco::Message::PRIO_FATAL: [[fallthrough]]; + case Poco::Message::PRIO_CRITICAL: + LOG_FATAL(trace_log, fmt::runtime(msg)); break; + case Poco::Message::PRIO_ERROR: + LOG_ERROR(trace_log, fmt::runtime(msg)); break; + case Poco::Message::PRIO_WARNING: + LOG_WARNING(trace_log, fmt::runtime(msg)); break; + case Poco::Message::PRIO_NOTICE: [[fallthrough]]; + case Poco::Message::PRIO_INFORMATION: + LOG_INFO(trace_log, fmt::runtime(msg)); break; + case Poco::Message::PRIO_DEBUG: + LOG_DEBUG(trace_log, fmt::runtime(msg)); break; + case Poco::Message::PRIO_TRACE: + LOG_TRACE(trace_log, fmt::runtime(msg)); break; + case Poco::Message::PRIO_TEST: + LOG_TEST(trace_log, fmt::runtime(msg)); break; + } + } + private: LoggerPtr log = getLogger("ServerErrorHandler"); + LoggerPtr trace_log = getLogger("Poco"); void logException() { diff --git a/src/Common/FailPoint.cpp b/src/Common/FailPoint.cpp index f5ec8cf0356..b2fcbc77c56 100644 --- a/src/Common/FailPoint.cpp +++ b/src/Common/FailPoint.cpp @@ -7,6 +7,8 @@ #include #include +#include "config.h" + namespace DB { @@ -15,7 +17,7 @@ namespace ErrorCodes extern const int LOGICAL_ERROR; }; -#if FIU_ENABLE +#if USE_LIBFIU static struct InitFiu { InitFiu() @@ -60,6 +62,7 @@ static struct InitFiu ONCE(receive_timeout_on_table_status_response) \ REGULAR(keepermap_fail_drop_data) \ REGULAR(lazy_pipe_fds_fail_close) \ + PAUSEABLE(infinite_sleep) \ namespace FailPoints @@ -134,7 +137,7 @@ void FailPointInjection::pauseFailPoint(const String & fail_point_name) void FailPointInjection::enableFailPoint(const String & fail_point_name) { -#if FIU_ENABLE +#if USE_LIBFIU #define SUB_M(NAME, flags, pause) \ if (fail_point_name == FailPoints::NAME) \ { \ diff --git a/src/Common/FailPoint.h b/src/Common/FailPoint.h index b3e1214d597..1af13d08553 100644 --- a/src/Common/FailPoint.h +++ b/src/Common/FailPoint.h @@ -1,17 +1,16 @@ #pragma once -#include "config.h" #include #include #include +#include "config.h" + #pragma clang diagnostic push #pragma clang diagnostic ignored "-Wdocumentation" #pragma clang diagnostic ignored "-Wreserved-macro-identifier" - -#include -#include - +# include +# include #pragma clang diagnostic pop #include diff --git a/src/Common/FieldBinaryEncoding.cpp b/src/Common/FieldBinaryEncoding.cpp index 6c1a8496fe6..23263c988c3 100644 --- a/src/Common/FieldBinaryEncoding.cpp +++ b/src/Common/FieldBinaryEncoding.cpp @@ -208,7 +208,7 @@ void FieldVisitorEncodeBinary::operator() (const Map & x, WriteBuffer & buf) con writeVarUInt(size, buf); for (size_t i = 0; i < size; ++i) { - const Tuple & key_and_value = x[i].get(); + const Tuple & key_and_value = x[i].safeGet(); 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]); } diff --git a/src/Common/FieldVisitorSum.cpp b/src/Common/FieldVisitorSum.cpp index b825f188586..af9503ac046 100644 --- a/src/Common/FieldVisitorSum.cpp +++ b/src/Common/FieldVisitorSum.cpp @@ -19,7 +19,7 @@ bool FieldVisitorSum::operator() (UInt64 & x) const return x != 0; } -bool FieldVisitorSum::operator() (Float64 & x) const { x += rhs.get(); return x != 0; } +bool FieldVisitorSum::operator() (Float64 & x) const { x += rhs.safeGet(); return x != 0; } bool FieldVisitorSum::operator() (Null &) const { diff --git a/src/Common/FieldVisitorSum.h b/src/Common/FieldVisitorSum.h index cbb4c4a1de3..d28676b5093 100644 --- a/src/Common/FieldVisitorSum.h +++ b/src/Common/FieldVisitorSum.h @@ -37,7 +37,7 @@ public: template bool operator() (DecimalField & x) const { - x += rhs.get>(); + x += rhs.safeGet>(); return x.getValue() != T(0); } diff --git a/src/Common/FieldVisitorToString.cpp b/src/Common/FieldVisitorToString.cpp index c4cb4266418..2148bac20d1 100644 --- a/src/Common/FieldVisitorToString.cpp +++ b/src/Common/FieldVisitorToString.cpp @@ -172,7 +172,7 @@ String FieldVisitorToString::operator() (const Object & x) const String convertFieldToString(const Field & field) { if (field.getType() == Field::Types::Which::String) - return field.get(); + return field.safeGet(); return applyVisitor(FieldVisitorToString(), field); } diff --git a/src/Common/HashTable/HashMap.h b/src/Common/HashTable/HashMap.h index a26797a687a..92621db5558 100644 --- a/src/Common/HashTable/HashMap.h +++ b/src/Common/HashTable/HashMap.h @@ -297,7 +297,7 @@ public: } /// Only inserts the value if key isn't already present - void ALWAYS_INLINE insertIfNotPresent(const Key & x, const Cell::Mapped & value) + void ALWAYS_INLINE insertIfNotPresent(const Key & x, const typename Cell::Mapped & value) { LookupResult it; bool inserted; diff --git a/src/Common/OptimizedRegularExpression.cpp b/src/Common/OptimizedRegularExpression.cpp index 712cab80aff..2cdb3409487 100644 --- a/src/Common/OptimizedRegularExpression.cpp +++ b/src/Common/OptimizedRegularExpression.cpp @@ -244,33 +244,43 @@ const char * analyzeImpl( is_trivial = false; if (!in_square_braces) { - /// Check for case-insensitive flag. - if (pos + 1 < end && pos[1] == '?') + /// it means flag negation + /// there are various possible flags + /// actually only imsU are supported by re2 + auto is_flag_char = [](char x) { - for (size_t offset = 2; pos + offset < end; ++offset) + return x == '-' || x == 'i' || x == 'm' || x == 's' || x == 'U' || x == 'u'; + }; + /// Check for case-insensitive flag. + if (pos + 2 < end && pos[1] == '?' && is_flag_char(pos[2])) + { + size_t offset = 2; + for (; pos + offset < end; ++offset) { - if (pos[offset] == '-' /// it means flag negation - /// various possible flags, actually only imsU are supported by re2 - || (pos[offset] >= 'a' && pos[offset] <= 'z') - || (pos[offset] >= 'A' && pos[offset] <= 'Z')) + if (pos[offset] == 'i') { - if (pos[offset] == 'i') - { - /// Actually it can be negated case-insensitive flag. But we don't care. - has_case_insensitive_flag = true; - break; - } + /// Actually it can be negated case-insensitive flag. But we don't care. + has_case_insensitive_flag = true; } - else + else if (!is_flag_char(pos[offset])) break; } + pos += offset; + if (pos == end) + return pos; + /// if this group only contains flags, we have nothing to do. + if (*pos == ')') + { + ++pos; + break; + } } /// (?:regex) means non-capturing parentheses group - if (pos + 2 < end && pos[1] == '?' && pos[2] == ':') + else if (pos + 2 < end && pos[1] == '?' && pos[2] == ':') { pos += 2; } - if (pos + 3 < end && pos[1] == '?' && (pos[2] == '<' || pos[2] == '\'' || (pos[2] == 'P' && pos[3] == '<'))) + else if (pos + 3 < end && pos[1] == '?' && (pos[2] == '<' || pos[2] == '\'' || (pos[2] == 'P' && pos[3] == '<'))) { pos = skipNameCapturingGroup(pos, pos[2] == 'P' ? 3: 2, end); } diff --git a/src/Common/ProfileEvents.cpp b/src/Common/ProfileEvents.cpp index f8bd84df131..d43d9fdcea8 100644 --- a/src/Common/ProfileEvents.cpp +++ b/src/Common/ProfileEvents.cpp @@ -209,8 +209,35 @@ \ M(Merge, "Number of launched background merges.") \ M(MergedRows, "Rows read for background merges. This is the number of rows before merge.") \ + M(MergedColumns, "Number of columns merged during the horizontal stage of merges.") \ + M(GatheredColumns, "Number of columns gathered during the vertical stage of merges.") \ M(MergedUncompressedBytes, "Uncompressed bytes (for columns as they stored in memory) that was read for background merges. This is the number before merge.") \ - M(MergesTimeMilliseconds, "Total time spent for background merges.")\ + M(MergeTotalMilliseconds, "Total time spent for background merges") \ + M(MergeExecuteMilliseconds, "Total busy time spent for execution of background merges") \ + M(MergeHorizontalStageTotalMilliseconds, "Total time spent for horizontal stage of background merges") \ + M(MergeHorizontalStageExecuteMilliseconds, "Total busy time spent for execution of horizontal stage of background merges") \ + M(MergeVerticalStageTotalMilliseconds, "Total time spent for vertical stage of background merges") \ + M(MergeVerticalStageExecuteMilliseconds, "Total busy time spent for execution of vertical stage of background merges") \ + M(MergeProjectionStageTotalMilliseconds, "Total time spent for projection stage of background merges") \ + M(MergeProjectionStageExecuteMilliseconds, "Total busy time spent for execution of projection stage of background merges") \ + \ + M(MergingSortedMilliseconds, "Total time spent while merging sorted columns") \ + M(AggregatingSortedMilliseconds, "Total time spent while aggregating sorted columns") \ + M(CollapsingSortedMilliseconds, "Total time spent while collapsing sorted columns") \ + M(ReplacingSortedMilliseconds, "Total time spent while replacing sorted columns") \ + M(SummingSortedMilliseconds, "Total time spent while summing sorted columns") \ + M(VersionedCollapsingSortedMilliseconds, "Total time spent while version collapsing sorted columns") \ + M(GatheringColumnMilliseconds, "Total time spent while gathering columns for vertical merge") \ + \ + M(MutationTotalParts, "Number of total parts for which mutations tried to be applied") \ + M(MutationUntouchedParts, "Number of total parts for which mutations tried to be applied but which was completely skipped according to predicate") \ + M(MutatedRows, "Rows read for mutations. This is the number of rows before mutation") \ + M(MutatedUncompressedBytes, "Uncompressed bytes (for columns as they stored in memory) that was read for mutations. This is the number before mutation.") \ + M(MutationTotalMilliseconds, "Total time spent for mutations.") \ + M(MutationExecuteMilliseconds, "Total busy time spent for execution of mutations.") \ + M(MutationAllPartColumns, "Number of times when task to mutate all columns in part was created") \ + M(MutationSomePartColumns, "Number of times when task to mutate some columns in part was created") \ + M(MutateTaskProjectionsCalculationMicroseconds, "Time spent calculating projections in mutations.") \ \ M(MergeTreeDataWriterRows, "Number of rows INSERTed to MergeTree tables.") \ M(MergeTreeDataWriterUncompressedBytes, "Uncompressed bytes (for columns as they stored in memory) INSERTed to MergeTree tables.") \ @@ -225,7 +252,6 @@ M(MergeTreeDataWriterProjectionsCalculationMicroseconds, "Time spent calculating projections") \ M(MergeTreeDataProjectionWriterSortingBlocksMicroseconds, "Time spent sorting blocks (for projection it might be a key different from table's sorting key)") \ M(MergeTreeDataProjectionWriterMergingBlocksMicroseconds, "Time spent merging blocks") \ - M(MutateTaskProjectionsCalculationMicroseconds, "Time spent calculating projections") \ \ M(InsertedWideParts, "Number of parts inserted in Wide format.") \ M(InsertedCompactParts, "Number of parts inserted in Compact format.") \ @@ -459,6 +485,7 @@ The server successfully detected this situation and will download merged part fr 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(AzureCreateContainer, "Number of Azure blob storage API CreateContainer calls.") \ \ M(DiskAzureGetObject, "Number of Disk Azure API GetObject calls.") \ M(DiskAzureUpload, "Number of Disk Azure blob storage API Upload calls") \ @@ -466,8 +493,9 @@ The server successfully detected this situation and will download merged part fr 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.") \ + M(DiskAzureDeleteObjects, "Number of Disk Azure blob storage API DeleteObject(s) calls.") \ M(DiskAzureGetProperties, "Number of Disk Azure blob storage API GetProperties calls.") \ + M(DiskAzureCreateContainer, "Number of Disk Azure blob storage API CreateContainer calls.") \ \ M(ReadBufferFromAzureMicroseconds, "Time spent on reading from Azure.") \ M(ReadBufferFromAzureInitMicroseconds, "Time spent initializing connection to Azure.") \ diff --git a/src/Client/QueryFuzzer.cpp b/src/Common/QueryFuzzer.cpp similarity index 95% rename from src/Client/QueryFuzzer.cpp rename to src/Common/QueryFuzzer.cpp index f5b700ea529..0b2f6c09b45 100644 --- a/src/Client/QueryFuzzer.cpp +++ b/src/Common/QueryFuzzer.cpp @@ -68,22 +68,21 @@ Field QueryFuzzer::getRandomField(int type) { case 0: { - return bad_int64_values[fuzz_rand() % (sizeof(bad_int64_values) - / sizeof(*bad_int64_values))]; + return bad_int64_values[fuzz_rand() % std::size(bad_int64_values)]; } case 1: { static constexpr double values[] = {NAN, INFINITY, -INFINITY, 0., -0., 0.0001, 0.5, 0.9999, 1., 1.0001, 2., 10.0001, 100.0001, 1000.0001, 1e10, 1e20, - FLT_MIN, FLT_MIN + FLT_EPSILON, FLT_MAX, FLT_MAX + FLT_EPSILON}; return values[fuzz_rand() % (sizeof(values) / sizeof(*values))]; + FLT_MIN, FLT_MIN + FLT_EPSILON, FLT_MAX, FLT_MAX + FLT_EPSILON}; return values[fuzz_rand() % std::size(values)]; } case 2: { static constexpr UInt64 scales[] = {0, 1, 2, 10}; return DecimalField( - bad_int64_values[fuzz_rand() % (sizeof(bad_int64_values) / sizeof(*bad_int64_values))], - static_cast(scales[fuzz_rand() % (sizeof(scales) / sizeof(*scales))]) + bad_int64_values[fuzz_rand() % std::size(bad_int64_values)], + static_cast(scales[fuzz_rand() % std::size(scales)]) ); } default: @@ -133,7 +132,7 @@ Field QueryFuzzer::fuzzField(Field field) if (type == Field::Types::String) { - auto & str = field.get(); + auto & str = field.safeGet(); UInt64 action = fuzz_rand() % 10; switch (action) { @@ -159,13 +158,14 @@ Field QueryFuzzer::fuzzField(Field field) } else if (type == Field::Types::Array) { - auto & arr = field.get(); + auto & arr = field.safeGet(); if (fuzz_rand() % 5 == 0 && !arr.empty()) { size_t pos = fuzz_rand() % arr.size(); arr.erase(arr.begin() + pos); - std::cerr << "erased\n"; + if (debug_stream) + *debug_stream << "erased\n"; } if (fuzz_rand() % 5 == 0) @@ -174,12 +174,14 @@ Field QueryFuzzer::fuzzField(Field field) { size_t pos = fuzz_rand() % arr.size(); arr.insert(arr.begin() + pos, fuzzField(arr[pos])); - std::cerr << fmt::format("inserted (pos {})\n", pos); + if (debug_stream) + *debug_stream << fmt::format("inserted (pos {})\n", pos); } else { arr.insert(arr.begin(), getRandomField(0)); - std::cerr << "inserted (0)\n"; + if (debug_stream) + *debug_stream << "inserted (0)\n"; } } @@ -191,13 +193,15 @@ Field QueryFuzzer::fuzzField(Field field) } else if (type == Field::Types::Tuple) { - auto & arr = field.get(); + auto & arr = field.safeGet(); if (fuzz_rand() % 5 == 0 && !arr.empty()) { size_t pos = fuzz_rand() % arr.size(); arr.erase(arr.begin() + pos); - std::cerr << "erased\n"; + + if (debug_stream) + *debug_stream << "erased\n"; } if (fuzz_rand() % 5 == 0) @@ -206,12 +210,16 @@ Field QueryFuzzer::fuzzField(Field field) { size_t pos = fuzz_rand() % arr.size(); arr.insert(arr.begin() + pos, fuzzField(arr[pos])); - std::cerr << fmt::format("inserted (pos {})\n", pos); + + if (debug_stream) + *debug_stream << fmt::format("inserted (pos {})\n", pos); } else { arr.insert(arr.begin(), getRandomField(0)); - std::cerr << "inserted (0)\n"; + + if (debug_stream) + *debug_stream << "inserted (0)\n"; } } @@ -344,7 +352,8 @@ void QueryFuzzer::fuzzOrderByList(IAST * ast) } else { - std::cerr << "No random column.\n"; + if (debug_stream) + *debug_stream << "No random column.\n"; } } @@ -378,7 +387,8 @@ void QueryFuzzer::fuzzColumnLikeExpressionList(IAST * ast) if (col) impl->children.insert(pos, col); else - std::cerr << "No random column.\n"; + if (debug_stream) + *debug_stream << "No random column.\n"; } // We don't have to recurse here to fuzz the children, this is handled by @@ -912,17 +922,17 @@ ASTPtr QueryFuzzer::fuzzLiteralUnderExpressionList(ASTPtr child) auto type = l->value.getType(); if (type == Field::Types::Which::String && fuzz_rand() % 7 == 0) { - String value = l->value.get(); + String value = l->value.safeGet(); child = makeASTFunction( "toFixedString", std::make_shared(value), std::make_shared(static_cast(value.size()))); } else if (type == Field::Types::Which::UInt64 && fuzz_rand() % 7 == 0) { - child = makeASTFunction(fuzz_rand() % 2 == 0 ? "toUInt128" : "toUInt256", std::make_shared(l->value.get())); + child = makeASTFunction(fuzz_rand() % 2 == 0 ? "toUInt128" : "toUInt256", std::make_shared(l->value.safeGet())); } else if (type == Field::Types::Which::Int64 && fuzz_rand() % 7 == 0) { - child = makeASTFunction(fuzz_rand() % 2 == 0 ? "toInt128" : "toInt256", std::make_shared(l->value.get())); + child = makeASTFunction(fuzz_rand() % 2 == 0 ? "toInt128" : "toInt256", std::make_shared(l->value.safeGet())); } else if (type == Field::Types::Which::Float64 && fuzz_rand() % 7 == 0) { @@ -930,22 +940,22 @@ ASTPtr QueryFuzzer::fuzzLiteralUnderExpressionList(ASTPtr child) if (decimal == 0) child = makeASTFunction( "toDecimal32", - std::make_shared(l->value.get()), + std::make_shared(l->value.safeGet()), std::make_shared(static_cast(fuzz_rand() % 9))); else if (decimal == 1) child = makeASTFunction( "toDecimal64", - std::make_shared(l->value.get()), + std::make_shared(l->value.safeGet()), std::make_shared(static_cast(fuzz_rand() % 18))); else if (decimal == 2) child = makeASTFunction( "toDecimal128", - std::make_shared(l->value.get()), + std::make_shared(l->value.safeGet()), std::make_shared(static_cast(fuzz_rand() % 38))); else child = makeASTFunction( "toDecimal256", - std::make_shared(l->value.get()), + std::make_shared(l->value.safeGet()), std::make_shared(static_cast(fuzz_rand() % 76))); } @@ -1361,11 +1371,15 @@ void QueryFuzzer::fuzzMain(ASTPtr & ast) collectFuzzInfoMain(ast); fuzz(ast); - std::cout << std::endl; - WriteBufferFromOStream ast_buf(std::cout, 4096); - formatAST(*ast, ast_buf, false /*highlight*/); - ast_buf.finalize(); - std::cout << std::endl << std::endl; + if (out_stream) + { + *out_stream << std::endl; + + WriteBufferFromOStream ast_buf(*out_stream, 4096); + formatAST(*ast, ast_buf, false /*highlight*/); + ast_buf.finalize(); + *out_stream << std::endl << std::endl; + } } } diff --git a/src/Client/QueryFuzzer.h b/src/Common/QueryFuzzer.h similarity index 91% rename from src/Client/QueryFuzzer.h rename to src/Common/QueryFuzzer.h index 6165e589cae..35d088809f2 100644 --- a/src/Client/QueryFuzzer.h +++ b/src/Common/QueryFuzzer.h @@ -35,9 +35,31 @@ struct ASTWindowDefinition; * queries, so you want to feed it a lot of queries to get some interesting mix * of them. Normally we feed SQL regression tests to it. */ -struct QueryFuzzer +class QueryFuzzer { - pcg64 fuzz_rand{randomSeed()}; +public: + explicit QueryFuzzer(pcg64 fuzz_rand_ = randomSeed(), std::ostream * out_stream_ = nullptr, std::ostream * debug_stream_ = nullptr) + : fuzz_rand(fuzz_rand_) + , out_stream(out_stream_) + , debug_stream(debug_stream_) + { + } + + // This is the only function you have to call -- it will modify the passed + // ASTPtr to point to new AST with some random changes. + void fuzzMain(ASTPtr & ast); + + ASTs getInsertQueriesForFuzzedTables(const String & full_query); + ASTs getDropQueriesForFuzzedTables(const ASTDropQuery & drop_query); + void notifyQueryFailed(ASTPtr ast); + + static bool isSuitableForFuzzing(const ASTCreateQuery & create); + +private: + pcg64 fuzz_rand; + + std::ostream * out_stream = nullptr; + std::ostream * debug_stream = nullptr; // We add elements to expression lists with fixed probability. Some elements // are so large, that the expected number of elements we add to them is @@ -66,10 +88,6 @@ struct QueryFuzzer std::unordered_map index_of_fuzzed_table; std::set created_tables_hashes; - // This is the only function you have to call -- it will modify the passed - // ASTPtr to point to new AST with some random changes. - void fuzzMain(ASTPtr & ast); - // Various helper functions follow, normally you shouldn't have to call them. Field getRandomField(int type); Field fuzzField(Field field); @@ -77,9 +95,6 @@ struct QueryFuzzer ASTPtr getRandomExpressionList(); DataTypePtr fuzzDataType(DataTypePtr type); DataTypePtr getRandomType(); - ASTs getInsertQueriesForFuzzedTables(const String & full_query); - ASTs getDropQueriesForFuzzedTables(const ASTDropQuery & drop_query); - void notifyQueryFailed(ASTPtr ast); void replaceWithColumnLike(ASTPtr & ast); void replaceWithTableLike(ASTPtr & ast); void fuzzOrderByElement(ASTOrderByElement * elem); @@ -102,8 +117,6 @@ struct QueryFuzzer void addTableLike(ASTPtr ast); void addColumnLike(ASTPtr ast); void collectFuzzInfoRecurse(ASTPtr ast); - - static bool isSuitableForFuzzing(const ASTCreateQuery & create); }; } diff --git a/src/Common/Scheduler/Nodes/DynamicResourceManager.cpp b/src/Common/Scheduler/Nodes/DynamicResourceManager.cpp index 01aa7df48d3..6b9f6318903 100644 --- a/src/Common/Scheduler/Nodes/DynamicResourceManager.cpp +++ b/src/Common/Scheduler/Nodes/DynamicResourceManager.cpp @@ -184,14 +184,20 @@ void DynamicResourceManager::updateConfiguration(const Poco::Util::AbstractConfi // Resource update leads to loss of runtime data of nodes and may lead to temporary violation of constraints (e.g. limits) // Try to minimise this by reusing "equal" resources (initialized with the same configuration). + std::vector resources_to_attach; for (auto & [name, new_resource] : new_state->resources) { if (auto iter = state->resources.find(name); iter != state->resources.end()) // Resource update { State::ResourcePtr old_resource = iter->second; if (old_resource->equals(*new_resource)) + { new_resource = old_resource; // Rewrite with older version to avoid loss of runtime data + continue; + } } + // It is new or updated resource + resources_to_attach.emplace_back(new_resource); } // Commit new state @@ -199,17 +205,14 @@ void DynamicResourceManager::updateConfiguration(const Poco::Util::AbstractConfi state = new_state; // Attach new and updated resources to the scheduler - for (auto & [name, resource] : new_state->resources) + for (auto & resource : resources_to_attach) { const SchedulerNodePtr & root = resource->nodes.find("/")->second.ptr; - if (root->parent == nullptr) + resource->attached_to = &scheduler; + scheduler.event_queue->enqueue([this, root] { - resource->attached_to = &scheduler; - scheduler.event_queue->enqueue([this, root] - { - scheduler.attachChild(root); - }); - } + scheduler.attachChild(root); + }); } // NOTE: after mutex unlock `state` became available for Classifier(s) and must be immutable diff --git a/src/Common/Scheduler/Nodes/tests/gtest_resource_class_fair.cpp b/src/Common/Scheduler/Nodes/tests/gtest_resource_class_fair.cpp index 4f0e8c80734..16cce309c2a 100644 --- a/src/Common/Scheduler/Nodes/tests/gtest_resource_class_fair.cpp +++ b/src/Common/Scheduler/Nodes/tests/gtest_resource_class_fair.cpp @@ -8,7 +8,9 @@ using namespace DB; using ResourceTest = ResourceTestClass; -TEST(SchedulerFairPolicy, Factory) +/// Tests disabled because of leaks in the test themselves: https://github.com/ClickHouse/ClickHouse/issues/67678 + +TEST(DISABLED_SchedulerFairPolicy, Factory) { ResourceTest t; @@ -17,7 +19,7 @@ TEST(SchedulerFairPolicy, Factory) EXPECT_TRUE(dynamic_cast(fair.get()) != nullptr); } -TEST(SchedulerFairPolicy, FairnessWeights) +TEST(DISABLED_SchedulerFairPolicy, FairnessWeights) { ResourceTest t; @@ -41,7 +43,7 @@ TEST(SchedulerFairPolicy, FairnessWeights) t.consumed("B", 20); } -TEST(SchedulerFairPolicy, Activation) +TEST(DISABLED_SchedulerFairPolicy, Activation) { ResourceTest t; @@ -77,7 +79,7 @@ TEST(SchedulerFairPolicy, Activation) t.consumed("B", 10); } -TEST(SchedulerFairPolicy, FairnessMaxMin) +TEST(DISABLED_SchedulerFairPolicy, FairnessMaxMin) { ResourceTest t; @@ -101,7 +103,7 @@ TEST(SchedulerFairPolicy, FairnessMaxMin) t.consumed("A", 20); } -TEST(SchedulerFairPolicy, HierarchicalFairness) +TEST(DISABLED_SchedulerFairPolicy, HierarchicalFairness) { ResourceTest t; diff --git a/src/Common/Scheduler/Nodes/tests/gtest_resource_class_priority.cpp b/src/Common/Scheduler/Nodes/tests/gtest_resource_class_priority.cpp index a447b7f6780..d3d38aae048 100644 --- a/src/Common/Scheduler/Nodes/tests/gtest_resource_class_priority.cpp +++ b/src/Common/Scheduler/Nodes/tests/gtest_resource_class_priority.cpp @@ -8,7 +8,9 @@ using namespace DB; using ResourceTest = ResourceTestClass; -TEST(SchedulerPriorityPolicy, Factory) +/// Tests disabled because of leaks in the test themselves: https://github.com/ClickHouse/ClickHouse/issues/67678 + +TEST(DISABLED_SchedulerPriorityPolicy, Factory) { ResourceTest t; @@ -17,7 +19,7 @@ TEST(SchedulerPriorityPolicy, Factory) EXPECT_TRUE(dynamic_cast(prio.get()) != nullptr); } -TEST(SchedulerPriorityPolicy, Priorities) +TEST(DISABLED_SchedulerPriorityPolicy, Priorities) { ResourceTest t; @@ -51,7 +53,7 @@ TEST(SchedulerPriorityPolicy, Priorities) t.consumed("C", 0); } -TEST(SchedulerPriorityPolicy, Activation) +TEST(DISABLED_SchedulerPriorityPolicy, Activation) { ResourceTest t; @@ -92,7 +94,7 @@ TEST(SchedulerPriorityPolicy, Activation) t.consumed("C", 0); } -TEST(SchedulerPriorityPolicy, SinglePriority) +TEST(DISABLED_SchedulerPriorityPolicy, SinglePriority) { ResourceTest t; diff --git a/src/Common/Scheduler/Nodes/tests/gtest_throttler_constraint.cpp b/src/Common/Scheduler/Nodes/tests/gtest_throttler_constraint.cpp index 6cfccb252fa..2bc24cdb292 100644 --- a/src/Common/Scheduler/Nodes/tests/gtest_throttler_constraint.cpp +++ b/src/Common/Scheduler/Nodes/tests/gtest_throttler_constraint.cpp @@ -10,7 +10,9 @@ using namespace DB; using ResourceTest = ResourceTestClass; -TEST(SchedulerThrottlerConstraint, LeakyBucketConstraint) +/// Tests disabled because of leaks in the test themselves: https://github.com/ClickHouse/ClickHouse/issues/67678 + +TEST(DISABLED_SchedulerThrottlerConstraint, LeakyBucketConstraint) { ResourceTest t; EventQueue::TimePoint start = std::chrono::system_clock::now(); @@ -40,7 +42,7 @@ TEST(SchedulerThrottlerConstraint, LeakyBucketConstraint) t.consumed("A", 10); } -TEST(SchedulerThrottlerConstraint, Unlimited) +TEST(DISABLED_SchedulerThrottlerConstraint, Unlimited) { ResourceTest t; EventQueue::TimePoint start = std::chrono::system_clock::now(); @@ -57,7 +59,7 @@ TEST(SchedulerThrottlerConstraint, Unlimited) } } -TEST(SchedulerThrottlerConstraint, Pacing) +TEST(DISABLED_SchedulerThrottlerConstraint, Pacing) { ResourceTest t; EventQueue::TimePoint start = std::chrono::system_clock::now(); @@ -77,7 +79,7 @@ TEST(SchedulerThrottlerConstraint, Pacing) } } -TEST(SchedulerThrottlerConstraint, BucketFilling) +TEST(DISABLED_SchedulerThrottlerConstraint, BucketFilling) { ResourceTest t; EventQueue::TimePoint start = std::chrono::system_clock::now(); @@ -111,7 +113,7 @@ TEST(SchedulerThrottlerConstraint, BucketFilling) t.consumed("A", 3); } -TEST(SchedulerThrottlerConstraint, PeekAndAvgLimits) +TEST(DISABLED_SchedulerThrottlerConstraint, PeekAndAvgLimits) { ResourceTest t; EventQueue::TimePoint start = std::chrono::system_clock::now(); @@ -139,7 +141,7 @@ TEST(SchedulerThrottlerConstraint, PeekAndAvgLimits) } } -TEST(SchedulerThrottlerConstraint, ThrottlerAndFairness) +TEST(DISABLED_SchedulerThrottlerConstraint, ThrottlerAndFairness) { ResourceTest t; EventQueue::TimePoint start = std::chrono::system_clock::now(); diff --git a/src/Common/ShellCommand.cpp b/src/Common/ShellCommand.cpp index 98a21b43d76..0d41669816c 100644 --- a/src/Common/ShellCommand.cpp +++ b/src/Common/ShellCommand.cpp @@ -237,7 +237,14 @@ std::unique_ptr ShellCommand::executeImpl( res->write_fds.emplace(fd, fds.fds_rw[1]); } - LOG_TRACE(getLogger(), "Started shell command '{}' with pid {}", filename, pid); + LOG_TRACE( + getLogger(), + "Started shell command '{}' with pid {} and file descriptors: out {}, err {}", + filename, + pid, + res->out.getFD(), + res->err.getFD()); + return res; } diff --git a/src/Common/SignalHandlers.cpp b/src/Common/SignalHandlers.cpp index e025e49e0a3..c4358da2453 100644 --- a/src/Common/SignalHandlers.cpp +++ b/src/Common/SignalHandlers.cpp @@ -629,6 +629,7 @@ void HandledSignals::setupTerminateHandler() void HandledSignals::setupCommonDeadlySignalHandlers() { /// SIGTSTP is added for debugging purposes. To output a stack trace of any running thread at anytime. + /// NOTE: that it is also used by clickhouse-test wrapper addSignalHandler({SIGABRT, SIGSEGV, SIGILL, SIGBUS, SIGSYS, SIGFPE, SIGPIPE, SIGTSTP, SIGTRAP}, signalHandler, true); #if defined(SANITIZER) diff --git a/src/Common/SystemLogBase.cpp b/src/Common/SystemLogBase.cpp index 7d2c15714e2..127c8862a35 100644 --- a/src/Common/SystemLogBase.cpp +++ b/src/Common/SystemLogBase.cpp @@ -27,12 +27,14 @@ #include #include + namespace DB { namespace ErrorCodes { extern const int TIMEOUT_EXCEEDED; + extern const int ABORTED; } ISystemLog::~ISystemLog() = default; @@ -86,32 +88,18 @@ void SystemLogQueue::push(LogElement&& element) // by one, under exclusive lock, so we will see each message count. // It is enough to only wake the flushing thread once, after the message // count increases past half available size. - const uint64_t queue_end = queue_front_index + queue.size(); - requested_flush_up_to = std::max(requested_flush_up_to, queue_end); - flush_event.notify_all(); + const auto last_log_index = queue_front_index + queue.size(); + notifyFlushUnlocked(last_log_index, /* should_prepare_tables_anyway */ false); } if (queue.size() >= settings.max_size_rows) { + chassert(queue.size() == settings.max_size_rows); + // Ignore all further entries until the queue is flushed. - // Log a message about that. Don't spam it -- this might be especially - // problematic in case of trace log. Remember what the front index of the - // queue was when we last logged the message. If it changed, it means the - // queue was flushed, and we can log again. - if (queue_front_index != logged_queue_full_at_index) - { - logged_queue_full_at_index = queue_front_index; - - // TextLog sets its logger level to 0, so this log is a noop and - // there is no recursive logging. - lock.unlock(); - LOG_ERROR(log, "Queue is full for system log '{}' at {}. max_size_rows {}", - demangle(typeid(*this).name()), - queue_front_index, - settings.max_size_rows); - } - + // To the next batch we add a log message about how much we have lost + ++ignored_logs; return; } @@ -127,20 +115,50 @@ template void SystemLogQueue::handleCrash() { if (settings.notify_flush_on_crash) - notifyFlush(/* force */ true); + { + notifyFlush(getLastLogIndex(), /* should_prepare_tables_anyway */ true); + } } template -void SystemLogQueue::waitFlush(uint64_t expected_flushed_up_to) +void SystemLogQueue::notifyFlushUnlocked(Index expected_flushed_index, bool should_prepare_tables_anyway) { + if (should_prepare_tables_anyway) + requested_prepare_tables = std::max(requested_prepare_tables, expected_flushed_index); + + requested_flush_index = std::max(requested_flush_index, expected_flushed_index); + + flush_event.notify_all(); +} + +template +void SystemLogQueue::notifyFlush(SystemLogQueue::Index expected_flushed_index, bool should_prepare_tables_anyway) +{ + std::lock_guard lock(mutex); + notifyFlushUnlocked(expected_flushed_index, should_prepare_tables_anyway); +} + +template +void SystemLogQueue::waitFlush(SystemLogQueue::Index expected_flushed_index, bool should_prepare_tables_anyway) +{ + LOG_DEBUG(log, "Requested flush up to offset {}", expected_flushed_index); + // Use an arbitrary timeout to avoid endless waiting. 60s proved to be // too fast for our parallel functional tests, probably because they // heavily load the disk. const int timeout_seconds = 180; + std::unique_lock lock(mutex); - bool result = flush_event.wait_for(lock, std::chrono::seconds(timeout_seconds), [&] + + // there is no obligation to call notifyFlush before waitFlush, than we have to be sure that flush_event has been triggered before we wait the result + notifyFlushUnlocked(expected_flushed_index, should_prepare_tables_anyway); + + auto result = confirm_event.wait_for(lock, std::chrono::seconds(timeout_seconds), [&] { - return flushed_up_to >= expected_flushed_up_to && !is_force_prepare_tables; + if (should_prepare_tables_anyway) + return (flushed_index >= expected_flushed_index && prepared_tables >= requested_prepare_tables) || is_shutdown; + else + return (flushed_index >= expected_flushed_index) || is_shutdown; }); if (!result) @@ -148,67 +166,63 @@ void SystemLogQueue::waitFlush(uint64_t expected_flushed_up_to) throw Exception(ErrorCodes::TIMEOUT_EXCEEDED, "Timeout exceeded ({} s) while flushing system log '{}'.", toString(timeout_seconds), demangle(typeid(*this).name())); } -} - -template -uint64_t SystemLogQueue::notifyFlush(bool should_prepare_tables_anyway) -{ - uint64_t this_thread_requested_offset; + if (is_shutdown) { - std::lock_guard lock(mutex); - if (is_shutdown) - return uint64_t(-1); - - this_thread_requested_offset = queue_front_index + queue.size(); - - // Publish our flush request, taking care not to overwrite the requests - // made by other threads. - is_force_prepare_tables |= should_prepare_tables_anyway; - requested_flush_up_to = std::max(requested_flush_up_to, this_thread_requested_offset); - - flush_event.notify_all(); + throw Exception(ErrorCodes::ABORTED, "Shutdown has been called while flushing system log '{}'. Aborting.", + demangle(typeid(*this).name())); } - - LOG_DEBUG(log, "Requested flush up to offset {}", this_thread_requested_offset); - return this_thread_requested_offset; } template -void SystemLogQueue::confirm(uint64_t to_flush_end) +SystemLogQueue::Index SystemLogQueue::getLastLogIndex() { std::lock_guard lock(mutex); - flushed_up_to = to_flush_end; - is_force_prepare_tables = false; - flush_event.notify_all(); + return queue_front_index + queue.size(); } template -typename SystemLogQueue::Index SystemLogQueue::pop(std::vector & output, - bool & should_prepare_tables_anyway, - bool & exit_this_thread) +void SystemLogQueue::confirm(SystemLogQueue::Index last_flashed_index) { - /// Call dtors and deallocate strings without holding the global lock - output.resize(0); + std::lock_guard lock(mutex); + prepared_tables = std::max(prepared_tables, last_flashed_index); + flushed_index = std::max(flushed_index, last_flashed_index); + confirm_event.notify_all(); +} - std::unique_lock lock(mutex); - flush_event.wait_for(lock, - std::chrono::milliseconds(settings.flush_interval_milliseconds), - [&] () +template +typename SystemLogQueue::PopResult SystemLogQueue::pop() +{ + PopResult result; + size_t prev_ignored_logs = 0; + + { + std::unique_lock lock(mutex); + + flush_event.wait_for(lock, std::chrono::milliseconds(settings.flush_interval_milliseconds), [&] () { - return requested_flush_up_to > flushed_up_to || is_shutdown || is_force_prepare_tables; - } - ); + return requested_flush_index > flushed_index || requested_prepare_tables > prepared_tables || is_shutdown; + }); - queue_front_index += queue.size(); - // Swap with existing array from previous flush, to save memory - // allocations. - queue.swap(output); + if (is_shutdown) + return PopResult{.is_shutdown = true}; - should_prepare_tables_anyway = is_force_prepare_tables; + queue_front_index += queue.size(); + prev_ignored_logs = ignored_logs; + ignored_logs = 0; - exit_this_thread = is_shutdown; - return queue_front_index; + result.last_log_index = queue_front_index; + result.logs.swap(queue); + result.create_table_force = requested_prepare_tables > prepared_tables; + } + + if (prev_ignored_logs) + LOG_ERROR(log, "Queue had been full at {}, accepted {} logs, ignored {} logs.", + result.last_log_index - result.logs.size(), + result.logs.size(), + prev_ignored_logs); + + return result; } template @@ -229,13 +243,21 @@ SystemLogBase::SystemLogBase( } template -void SystemLogBase::flush(bool force) +SystemLogBase::Index SystemLogBase::getLastLogIndex() { - uint64_t this_thread_requested_offset = queue->notifyFlush(force); - if (this_thread_requested_offset == uint64_t(-1)) - return; + return queue->getLastLogIndex(); +} - queue->waitFlush(this_thread_requested_offset); +template +void SystemLogBase::notifyFlush(Index expected_flushed_index, bool should_prepare_tables_anyway) +{ + queue->notifyFlush(expected_flushed_index, should_prepare_tables_anyway); +} + +template +void SystemLogBase::flush(Index expected_flushed_index, bool should_prepare_tables_anyway) +{ + queue->waitFlush(expected_flushed_index, should_prepare_tables_anyway); } template @@ -257,9 +279,6 @@ void SystemLogBase::add(LogElement element) queue->push(std::move(element)); } -template -void SystemLogBase::notifyFlush(bool force) { queue->notifyFlush(force); } - #define INSTANTIATE_SYSTEM_LOG_BASE(ELEMENT) template class SystemLogBase; SYSTEM_LOG_ELEMENTS(INSTANTIATE_SYSTEM_LOG_BASE) diff --git a/src/Common/SystemLogBase.h b/src/Common/SystemLogBase.h index b87fcf419d3..0d7b04d5c57 100644 --- a/src/Common/SystemLogBase.h +++ b/src/Common/SystemLogBase.h @@ -1,6 +1,7 @@ #pragma once #include +#include #include #include #include @@ -54,10 +55,19 @@ struct StorageID; class ISystemLog { public: + using Index = int64_t; + virtual String getName() const = 0; - //// force -- force table creation (used for SYSTEM FLUSH LOGS) - virtual void flush(bool force = false) = 0; /// NOLINT + /// Return the index of the latest added log element. That index no less than the flashed index. + /// The flashed index is the index of the last log element which has been flushed successfully. + /// Thereby all the records whose index is less than the flashed index are flushed already. + virtual Index getLastLogIndex() = 0; + /// Call this method to wake up the flush thread and flush the data in the background. It is non blocking call + virtual void notifyFlush(Index expected_flushed_index, bool should_prepare_tables_anyway) = 0; + /// Call this method to wait intill the logs are flushed up to expected_flushed_index. It is blocking call. + virtual void flush(Index expected_flushed_index, bool should_prepare_tables_anyway) = 0; + virtual void prepareTable() = 0; /// Start the background thread. @@ -97,26 +107,38 @@ struct SystemLogQueueSettings template class SystemLogQueue { - using Index = uint64_t; - public: + using Index = ISystemLog::Index; + explicit SystemLogQueue(const SystemLogQueueSettings & settings_); void shutdown(); // producer methods void push(LogElement && element); - Index notifyFlush(bool should_prepare_tables_anyway); - void waitFlush(Index expected_flushed_up_to); + + Index getLastLogIndex(); + void notifyFlush(Index expected_flushed_index, bool should_prepare_tables_anyway); + void waitFlush(Index expected_flushed_index, bool should_prepare_tables_anyway); /// Handles crash, flushes log without blocking if notify_flush_on_crash is set void handleCrash(); + struct PopResult + { + Index last_log_index = 0; + std::vector logs = {}; + bool create_table_force = false; + bool is_shutdown = false; + }; + // consumer methods - Index pop(std::vector& output, bool & should_prepare_tables_anyway, bool & exit_this_thread); - void confirm(Index to_flush_end); + PopResult pop(); + void confirm(Index last_flashed_index); private: + void notifyFlushUnlocked(Index expected_flushed_index, bool should_prepare_tables_anyway); + /// Data shared between callers of add()/flush()/shutdown(), and the saving thread std::mutex mutex; @@ -124,22 +146,32 @@ private: // Queue is bounded. But its size is quite large to not block in all normal cases. std::vector queue; + // An always-incrementing index of the first message currently in the queue. // We use it to give a global sequential index to every message, so that we // can wait until a particular message is flushed. This is used to implement // synchronous log flushing for SYSTEM FLUSH LOGS. Index queue_front_index = 0; - // A flag that says we must create the tables even if the queue is empty. - bool is_force_prepare_tables = false; + // Requested to flush logs up to this index, exclusive - Index requested_flush_up_to = 0; + Index requested_flush_index = std::numeric_limits::min(); // Flushed log up to this index, exclusive - Index flushed_up_to = 0; - // Logged overflow message at this queue front index - Index logged_queue_full_at_index = -1; + Index flushed_index = 0; + + // The same logic for the prepare tables: if requested_prepar_tables > prepared_tables we need to do prepare + // except that initial prepared_tables is -1 + // it is due to the difference: when no logs have been written and we call flush logs + // it becomes in the state: requested_flush_index = 0 and flushed_index = 0 -- we do not want to do anything + // but if we need to prepare tables it becomes requested_prepare_tables = 0 and prepared_tables = -1 + // we trigger background thread and do prepare + Index requested_prepare_tables = std::numeric_limits::min(); + Index prepared_tables = -1; + + size_t ignored_logs = 0; bool is_shutdown = false; + std::condition_variable confirm_event; std::condition_variable flush_event; const SystemLogQueueSettings settings; @@ -150,6 +182,7 @@ template class SystemLogBase : public ISystemLog { public: + using Index = ISystemLog::Index; using Self = SystemLogBase; explicit SystemLogBase( @@ -163,15 +196,16 @@ public: */ void add(LogElement element); + Index getLastLogIndex() override; + + void notifyFlush(Index expected_flushed_index, bool should_prepare_tables_anyway) override; + /// Flush data in the buffer to disk. Block the thread until the data is stored on disk. - void flush(bool force) override; + void flush(Index expected_flushed_index, bool should_prepare_tables_anyway) override; /// Handles crash, flushes log without blocking if notify_flush_on_crash is set void handleCrash() override; - /// Non-blocking flush data in the buffer to disk. - void notifyFlush(bool force); - String getName() const override { return LogElement::name(); } static const char * getDefaultOrderBy() { return "event_date, event_time"; } diff --git a/src/Common/ZooKeeper/ZooKeeper.h b/src/Common/ZooKeeper/ZooKeeper.h index 657c9cb2c03..7ccdc9d1b7f 100644 --- a/src/Common/ZooKeeper/ZooKeeper.h +++ b/src/Common/ZooKeeper/ZooKeeper.h @@ -44,7 +44,7 @@ namespace ErrorCodes namespace zkutil { -/// Preferred size of multi() command (in number of ops) +/// Preferred size of multi command (in the number of operations) constexpr size_t MULTI_BATCH_SIZE = 100; struct ShuffleHost diff --git a/src/Common/config.h.in b/src/Common/config.h.in index 6a0090130a3..2e3b8d84366 100644 --- a/src/Common/config.h.in +++ b/src/Common/config.h.in @@ -32,6 +32,8 @@ #cmakedefine01 USE_IDNA #cmakedefine01 USE_NLP #cmakedefine01 USE_VECTORSCAN +#cmakedefine01 USE_QPL +#cmakedefine01 USE_QATLIB #cmakedefine01 USE_LIBURING #cmakedefine01 USE_AVRO #cmakedefine01 USE_CAPNP @@ -56,10 +58,11 @@ #cmakedefine01 USE_FILELOG #cmakedefine01 USE_ODBC #cmakedefine01 USE_BLAKE3 +#cmakedefine01 USE_USEARCH #cmakedefine01 USE_SKIM #cmakedefine01 USE_PRQL #cmakedefine01 USE_ULID -#cmakedefine01 FIU_ENABLE +#cmakedefine01 USE_LIBFIU #cmakedefine01 USE_BCRYPT #cmakedefine01 USE_LIBARCHIVE #cmakedefine01 USE_POCKETFFT diff --git a/src/Common/examples/arena_with_free_lists.cpp b/src/Common/examples/arena_with_free_lists.cpp index 6793d567aca..3a1304e2d94 100644 --- a/src/Common/examples/arena_with_free_lists.cpp +++ b/src/Common/examples/arena_with_free_lists.cpp @@ -174,19 +174,19 @@ struct Dictionary { switch (attribute.type) { - case AttributeUnderlyingTypeTest::UInt8: std::get>(attribute.arrays)[idx] = value.get(); break; - case AttributeUnderlyingTypeTest::UInt16: std::get>(attribute.arrays)[idx] = value.get(); break; - case AttributeUnderlyingTypeTest::UInt32: std::get>(attribute.arrays)[idx] = static_cast(value.get()); break; - case AttributeUnderlyingTypeTest::UInt64: std::get>(attribute.arrays)[idx] = value.get(); break; - case AttributeUnderlyingTypeTest::Int8: std::get>(attribute.arrays)[idx] = value.get(); break; - case AttributeUnderlyingTypeTest::Int16: std::get>(attribute.arrays)[idx] = value.get(); break; - case AttributeUnderlyingTypeTest::Int32: std::get>(attribute.arrays)[idx] = static_cast(value.get()); break; - case AttributeUnderlyingTypeTest::Int64: std::get>(attribute.arrays)[idx] = value.get(); break; - case AttributeUnderlyingTypeTest::Float32: std::get>(attribute.arrays)[idx] = static_cast(value.get()); break; - case AttributeUnderlyingTypeTest::Float64: std::get>(attribute.arrays)[idx] = value.get(); break; + case AttributeUnderlyingTypeTest::UInt8: std::get>(attribute.arrays)[idx] = value.safeGet(); break; + case AttributeUnderlyingTypeTest::UInt16: std::get>(attribute.arrays)[idx] = value.safeGet(); break; + case AttributeUnderlyingTypeTest::UInt32: std::get>(attribute.arrays)[idx] = static_cast(value.safeGet()); break; + case AttributeUnderlyingTypeTest::UInt64: std::get>(attribute.arrays)[idx] = value.safeGet(); break; + case AttributeUnderlyingTypeTest::Int8: std::get>(attribute.arrays)[idx] = value.safeGet(); break; + case AttributeUnderlyingTypeTest::Int16: std::get>(attribute.arrays)[idx] = value.safeGet(); break; + case AttributeUnderlyingTypeTest::Int32: std::get>(attribute.arrays)[idx] = static_cast(value.safeGet()); break; + case AttributeUnderlyingTypeTest::Int64: std::get>(attribute.arrays)[idx] = value.safeGet(); break; + case AttributeUnderlyingTypeTest::Float32: std::get>(attribute.arrays)[idx] = static_cast(value.safeGet()); break; + case AttributeUnderlyingTypeTest::Float64: std::get>(attribute.arrays)[idx] = value.safeGet(); break; case AttributeUnderlyingTypeTest::String: { - const auto & string = value.get(); + const auto & string = value.safeGet(); auto & string_ref = std::get>(attribute.arrays)[idx]; const auto & null_value_ref = std::get(attribute.null_values); diff --git a/src/Common/getNumberOfPhysicalCPUCores.h b/src/Common/getNumberOfPhysicalCPUCores.h index 827e95e1bea..9e3412fdcba 100644 --- a/src/Common/getNumberOfPhysicalCPUCores.h +++ b/src/Common/getNumberOfPhysicalCPUCores.h @@ -1,4 +1,5 @@ #pragma once /// Get number of CPU cores without hyper-threading. +/// The calculation respects possible cgroups limits. unsigned getNumberOfPhysicalCPUCores(); diff --git a/src/Common/parseRemoteDescription.cpp b/src/Common/parseRemoteDescription.cpp index df3820b11f9..6a53098362d 100644 --- a/src/Common/parseRemoteDescription.cpp +++ b/src/Common/parseRemoteDescription.cpp @@ -79,11 +79,16 @@ std::vector parseRemoteDescription( /// Look for the corresponding closing bracket for (m = i + 1; m < r; ++m) { - if (description[m] == '{') ++cnt; - if (description[m] == '}') --cnt; - if (description[m] == '.' && description[m-1] == '.') last_dot = m; - if (description[m] == separator) have_splitter = true; - if (cnt == 0) break; + if (description[m] == '{') + ++cnt; + if (description[m] == '}') + --cnt; + if (description[m] == '.' && description[m-1] == '.') + last_dot = m; + if (description[m] == separator) + have_splitter = true; + if (cnt == 0) + break; } if (cnt != 0) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Table function '{}': incorrect brace sequence in first argument", func_name); diff --git a/src/Common/tests/gtest_lsan.cpp b/src/Common/tests/gtest_lsan.cpp index f6e1984ec58..7fc4ad2749e 100644 --- a/src/Common/tests/gtest_lsan.cpp +++ b/src/Common/tests/gtest_lsan.cpp @@ -14,20 +14,21 @@ /// because of broken getauxval() [1]. /// /// [1]: https://github.com/ClickHouse/ClickHouse/pull/33957 -TEST(Common, LSan) +TEST(SanitizerDeathTest, LSan) { - int sanitizers_exit_code = 1; - - ASSERT_EXIT({ - std::thread leak_in_thread([]() + EXPECT_DEATH( { - void * leak = malloc(4096); - ASSERT_NE(leak, nullptr); - }); - leak_in_thread.join(); + std::thread leak_in_thread( + []() + { + void * leak = malloc(4096); + ASSERT_NE(leak, nullptr); + }); + leak_in_thread.join(); - __lsan_do_leak_check(); - }, ::testing::ExitedWithCode(sanitizers_exit_code), ".*LeakSanitizer: detected memory leaks.*"); + __lsan_do_leak_check(); + }, + ".*LeakSanitizer: detected memory leaks.*"); } #endif diff --git a/src/Common/tests/gtest_optimize_re.cpp b/src/Common/tests/gtest_optimize_re.cpp index a9fcb918b24..d6735c3ccfe 100644 --- a/src/Common/tests/gtest_optimize_re.cpp +++ b/src/Common/tests/gtest_optimize_re.cpp @@ -19,6 +19,9 @@ TEST(OptimizeRE, analyze) }; test_f("abc", "abc", {}, true, true); test_f("c([^k]*)de", ""); + test_f("(?-s)bob", "bob", {}, false, true); + test_f("(?s)bob", "bob", {}, false, true); + test_f("(?ssss", ""); test_f("abc(de)fg", "abcdefg", {}, false, true); test_f("abc(de|xyz)fg", "abc", {"abcdefg", "abcxyzfg"}, false, true); test_f("abc(de?f|xyz)fg", "abc", {"abcd", "abcxyzfg"}, false, true); diff --git a/src/Compression/CompressionCodecDeflateQpl.cpp b/src/Compression/CompressionCodecDeflateQpl.cpp index f1b5b24e866..c82ee861a6f 100644 --- a/src/Compression/CompressionCodecDeflateQpl.cpp +++ b/src/Compression/CompressionCodecDeflateQpl.cpp @@ -1,7 +1,3 @@ -#ifdef ENABLE_QPL_COMPRESSION - -#include -#include #include #include #include @@ -11,6 +7,10 @@ #include #include #include +#include +#include + +#if USE_QPL #include "libaccel_config.h" diff --git a/src/Compression/CompressionCodecDeflateQpl.h b/src/Compression/CompressionCodecDeflateQpl.h index 86fd9051bd8..d9abc0fb7e0 100644 --- a/src/Compression/CompressionCodecDeflateQpl.h +++ b/src/Compression/CompressionCodecDeflateQpl.h @@ -4,6 +4,11 @@ #include #include #include + +#include "config.h" + +#if USE_QPL + #include namespace Poco @@ -117,3 +122,4 @@ private: }; } +#endif diff --git a/src/Compression/CompressionCodecZSTDQAT.cpp b/src/Compression/CompressionCodecZSTDQAT.cpp index 5a4ef70a30a..e19b7e4a001 100644 --- a/src/Compression/CompressionCodecZSTDQAT.cpp +++ b/src/Compression/CompressionCodecZSTDQAT.cpp @@ -1,4 +1,6 @@ -#ifdef ENABLE_ZSTD_QAT_CODEC +#include "config.h" + +#if USE_QATLIB #include #include @@ -6,6 +8,7 @@ #include #include #include + #include #include diff --git a/src/Compression/CompressionFactory.cpp b/src/Compression/CompressionFactory.cpp index 2e7aa0d086f..ac00f571568 100644 --- a/src/Compression/CompressionFactory.cpp +++ b/src/Compression/CompressionFactory.cpp @@ -1,20 +1,20 @@ -#include "config.h" - #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 { @@ -175,11 +175,11 @@ void registerCodecNone(CompressionCodecFactory & factory); void registerCodecLZ4(CompressionCodecFactory & factory); void registerCodecLZ4HC(CompressionCodecFactory & factory); void registerCodecZSTD(CompressionCodecFactory & factory); -#ifdef ENABLE_ZSTD_QAT_CODEC +#if USE_QATLIB void registerCodecZSTDQAT(CompressionCodecFactory & factory); #endif void registerCodecMultiple(CompressionCodecFactory & factory); -#ifdef ENABLE_QPL_COMPRESSION +#if USE_QPL void registerCodecDeflateQpl(CompressionCodecFactory & factory); #endif @@ -198,7 +198,7 @@ CompressionCodecFactory::CompressionCodecFactory() registerCodecNone(*this); registerCodecLZ4(*this); registerCodecZSTD(*this); -#ifdef ENABLE_ZSTD_QAT_CODEC +#if USE_QATLIB registerCodecZSTDQAT(*this); #endif registerCodecLZ4HC(*this); @@ -209,7 +209,7 @@ CompressionCodecFactory::CompressionCodecFactory() registerCodecGorilla(*this); registerCodecEncrypted(*this); registerCodecFPC(*this); -#ifdef ENABLE_QPL_COMPRESSION +#if USE_QPL registerCodecDeflateQpl(*this); #endif registerCodecGCD(*this); diff --git a/src/Compression/fuzzers/CMakeLists.txt b/src/Compression/fuzzers/CMakeLists.txt index a693faecc14..311f1eb3d35 100644 --- a/src/Compression/fuzzers/CMakeLists.txt +++ b/src/Compression/fuzzers/CMakeLists.txt @@ -5,19 +5,19 @@ # If you want really small size of the resulted binary, just link with fuzz_compression and clickhouse_common_io clickhouse_add_executable (compressed_buffer_fuzzer compressed_buffer_fuzzer.cpp) -target_link_libraries (compressed_buffer_fuzzer PRIVATE dbms) +target_link_libraries (compressed_buffer_fuzzer PRIVATE dbms clickhouse_functions) clickhouse_add_executable (lz4_decompress_fuzzer lz4_decompress_fuzzer.cpp) -target_link_libraries (lz4_decompress_fuzzer PUBLIC dbms ch_contrib::lz4) +target_link_libraries (lz4_decompress_fuzzer PUBLIC dbms ch_contrib::lz4 clickhouse_functions) clickhouse_add_executable (delta_decompress_fuzzer delta_decompress_fuzzer.cpp) -target_link_libraries (delta_decompress_fuzzer PRIVATE dbms) +target_link_libraries (delta_decompress_fuzzer PRIVATE dbms clickhouse_functions) clickhouse_add_executable (double_delta_decompress_fuzzer double_delta_decompress_fuzzer.cpp) -target_link_libraries (double_delta_decompress_fuzzer PRIVATE dbms) +target_link_libraries (double_delta_decompress_fuzzer PRIVATE dbms clickhouse_functions) clickhouse_add_executable (encrypted_decompress_fuzzer encrypted_decompress_fuzzer.cpp) -target_link_libraries (encrypted_decompress_fuzzer PRIVATE dbms) +target_link_libraries (encrypted_decompress_fuzzer PRIVATE dbms clickhouse_functions) clickhouse_add_executable (gcd_decompress_fuzzer gcd_decompress_fuzzer.cpp) -target_link_libraries (gcd_decompress_fuzzer PRIVATE dbms) +target_link_libraries (gcd_decompress_fuzzer PRIVATE dbms clickhouse_functions) diff --git a/src/Coordination/KeeperContext.cpp b/src/Coordination/KeeperContext.cpp index 1f66882ecad..dd2c1d59d56 100644 --- a/src/Coordination/KeeperContext.cpp +++ b/src/Coordination/KeeperContext.cpp @@ -23,6 +23,7 @@ #if USE_ROCKSDB #include #include +#include #include #endif @@ -88,7 +89,7 @@ static rocksdb::Options getRocksDBOptionsFromConfig(const Poco::Util::AbstractCo if (config.has("keeper_server.rocksdb.options")) { auto config_options = getOptionsFromConfig(config, "keeper_server.rocksdb.options"); - status = rocksdb::GetDBOptionsFromMap(merged, config_options, &merged); + status = rocksdb::GetDBOptionsFromMap({}, merged, config_options, &merged); if (!status.ok()) { throw Exception(ErrorCodes::ROCKSDB_ERROR, "Fail to merge rocksdb options from 'rocksdb.options' : {}", @@ -98,7 +99,7 @@ static rocksdb::Options getRocksDBOptionsFromConfig(const Poco::Util::AbstractCo 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); + 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()); @@ -107,7 +108,7 @@ static rocksdb::Options getRocksDBOptionsFromConfig(const Poco::Util::AbstractCo 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); + 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()); diff --git a/src/Coordination/KeeperSnapshotManager.cpp b/src/Coordination/KeeperSnapshotManager.cpp index 3f5ac055470..2ed89c414ff 100644 --- a/src/Coordination/KeeperSnapshotManager.cpp +++ b/src/Coordination/KeeperSnapshotManager.cpp @@ -54,7 +54,7 @@ namespace std::filesystem::path path(snapshot_path); std::string filename = path.stem(); Strings name_parts; - splitInto<'_'>(name_parts, filename); + splitInto<'_', '.'>(name_parts, filename); return parse(name_parts[1]); } diff --git a/src/Coordination/RaftServerConfig.cpp b/src/Coordination/RaftServerConfig.cpp index 929eeeb640e..bafc177b736 100644 --- a/src/Coordination/RaftServerConfig.cpp +++ b/src/Coordination/RaftServerConfig.cpp @@ -26,12 +26,16 @@ std::optional RaftServerConfig::parse(std::string_view server) if (!with_id_endpoint && !with_server_type && !with_priority) return std::nullopt; - const std::string_view id_str = parts[0]; + std::string_view id_str = parts[0]; if (!id_str.starts_with("server.")) return std::nullopt; + id_str = id_str.substr(7); + if (auto eq_pos = id_str.find('='); std::string_view::npos != eq_pos) + id_str = id_str.substr(0, eq_pos); + Int32 id; - if (!tryParse(id, std::next(id_str.begin(), 7))) + if (!tryParse(id, id_str)) return std::nullopt; if (id <= 0) return std::nullopt; diff --git a/src/Coordination/tests/gtest_coordination.cpp b/src/Coordination/tests/gtest_coordination.cpp index beae6254562..d39031773cd 100644 --- a/src/Coordination/tests/gtest_coordination.cpp +++ b/src/Coordination/tests/gtest_coordination.cpp @@ -111,10 +111,13 @@ public: } }; -using Implementation = testing::Types, - TestParam, - TestParam, - TestParam>; +using Implementation = testing::Types + ,TestParam +#if USE_ROCKSDB + ,TestParam + ,TestParam +#endif + >; TYPED_TEST_SUITE(CoordinationTest, Implementation); TYPED_TEST(CoordinationTest, RaftServerConfigParse) diff --git a/src/Core/ExternalTable.cpp b/src/Core/ExternalTable.cpp index af681cd5639..c2bcf6ec651 100644 --- a/src/Core/ExternalTable.cpp +++ b/src/Core/ExternalTable.cpp @@ -49,7 +49,7 @@ ExternalTableDataPtr BaseExternalTable::getData(ContextPtr context) { initReadBuffer(); initSampleBlock(); - auto input = context->getInputFormat(format, *read_buffer, sample_block, context->getSettingsRef().get("max_block_size").get()); + auto input = context->getInputFormat(format, *read_buffer, sample_block, context->getSettingsRef().get("max_block_size").safeGet()); auto data = std::make_unique(); data->pipe = std::make_unique(); diff --git a/src/Core/Field.h b/src/Core/Field.h index f1bb4a72b0d..ba8c66580ad 100644 --- a/src/Core/Field.h +++ b/src/Core/Field.h @@ -457,15 +457,6 @@ public: std::string_view getTypeName() const; bool isNull() const { return which == Types::Null; } - template - NearestFieldType> & get(); - - template - const auto & get() const - { - auto * mutable_this = const_cast *>(this); - return mutable_this->get(); - } bool isNegativeInfinity() const { return which == Types::Null && get().isNegativeInfinity(); } bool isPositiveInfinity() const { return which == Types::Null && get().isPositiveInfinity(); } @@ -681,6 +672,25 @@ private: Types::Which which; + /// This function is prone to type punning and should never be used outside of Field class, + /// whenever it is used within this class the stored type should be checked in advance. + template + NearestFieldType> & get() + { + // Before storing the value in the Field, we static_cast it to the field + // storage type, so here we return the value of storage type as well. + // Otherwise, it is easy to make a mistake of reinterpret_casting the stored + // value to a different and incompatible type. + // For example, a Float32 value is stored as Float64, and it is incorrect to + // return a reference to this value as Float32. + return *reinterpret_cast>*>(&storage); + } + + template + NearestFieldType> & get() const + { + return const_cast(this)->get(); + } /// Assuming there was no allocated state or it was deallocated (see destroy). template @@ -866,48 +876,21 @@ constexpr bool isInt64OrUInt64orBoolFieldType(Field::Types::Which t) || t == Field::Types::Bool; } -// Field value getter with type checking in debug builds. -template -NearestFieldType> & Field::get() -{ - // Before storing the value in the Field, we static_cast it to the field - // storage type, so here we return the value of storage type as well. - // Otherwise, it is easy to make a mistake of reinterpret_casting the stored - // value to a different and incompatible type. - // For example, a Float32 value is stored as Float64, and it is incorrect to - // return a reference to this value as Float32. - using StoredType = NearestFieldType>; - -#ifndef NDEBUG - // Disregard signedness when converting between int64 types. - constexpr Field::Types::Which target = TypeToEnum::value; - if (target != which - && (!isInt64OrUInt64orBoolFieldType(target) || !isInt64OrUInt64orBoolFieldType(which)) && target != Field::Types::IPv4) - throw Exception(ErrorCodes::LOGICAL_ERROR, - "Invalid Field get from type {} to type {}", which, target); -#endif - - StoredType * MAY_ALIAS ptr = reinterpret_cast(&storage); - - return *ptr; -} - - template auto & Field::safeGet() { const Types::Which target = TypeToEnum>>::value; - /// We allow converting int64 <-> uint64, int64 <-> bool, uint64 <-> bool in safeGet(). - if (target != which - && (!isInt64OrUInt64orBoolFieldType(target) || !isInt64OrUInt64orBoolFieldType(which))) - throw Exception(ErrorCodes::BAD_GET, - "Bad get: has {}, requested {}", getTypeName(), target); + /// bool is stored as uint64, will be returned as UInt64 when requested as bool or UInt64, as Int64 when requested as Int64 + /// also allow UInt64 <-> Int64 conversion + if (target != which && + !(which == Field::Types::Bool && (target == Field::Types::UInt64 || target == Field::Types::Int64)) && + !(isInt64OrUInt64FieldType(which) && isInt64OrUInt64FieldType(target))) + throw Exception(ErrorCodes::BAD_GET, "Bad get: has {}, requested {}", getTypeName(), target); return get(); } - template requires not_field_or_bool_or_stringlike Field::Field(T && rhs) diff --git a/src/Core/MySQL/MySQLGtid.cpp b/src/Core/MySQL/MySQLGtid.cpp index 7916f882979..28b583a0cfe 100644 --- a/src/Core/MySQL/MySQLGtid.cpp +++ b/src/Core/MySQL/MySQLGtid.cpp @@ -24,9 +24,7 @@ void GTIDSet::tryMerge(size_t i) void GTIDSets::parse(String gtid_format) { if (gtid_format.empty()) - { return; - } std::vector gtid_sets; boost::split(gtid_sets, gtid_format, [](char c) { return c == ','; }); diff --git a/src/Core/MySQL/tests/gtest_MySQLGtid.cpp b/src/Core/MySQL/tests/gtest_MySQLGtid.cpp index e31a87aaa39..e5a2fe44e5c 100644 --- a/src/Core/MySQL/tests/gtest_MySQLGtid.cpp +++ b/src/Core/MySQL/tests/gtest_MySQLGtid.cpp @@ -10,20 +10,19 @@ GTEST_TEST(GTIDSetsContains, Tests) contained1, contained2, contained3, contained4, contained5, not_contained1, not_contained2, not_contained3, not_contained4, not_contained5, not_contained6; - gtid_set.parse("2174B383-5441-11E8-B90A-C80AA9429562:1-3:11:47-49, 24DA167-0C0C-11E8-8442-00059A3C7B00:1-19:47-49:60"); - contained1.parse("2174B383-5441-11E8-B90A-C80AA9429562:1-3:11:47-49, 24DA167-0C0C-11E8-8442-00059A3C7B00:1-19:47-49:60"); + gtid_set.parse("2174B383-5441-11E8-B90A-C80AA9429562:1-3:11:47-49, FBC30C64-F8C9-4DDF-8CDD-066208EB433B:1-19:47-49:60"); + contained1.parse("2174B383-5441-11E8-B90A-C80AA9429562:1-3:11:47-49, FBC30C64-F8C9-4DDF-8CDD-066208EB433B:1-19:47-49:60"); contained2.parse("2174B383-5441-11E8-B90A-C80AA9429562:2-3:11:47-49"); contained3.parse("2174B383-5441-11E8-B90A-C80AA9429562:11"); - contained4.parse("24DA167-0C0C-11E8-8442-00059A3C7B00:2-16:47-49:60"); - contained5.parse("24DA167-0C0C-11E8-8442-00059A3C7B00:60"); + contained4.parse("FBC30C64-F8C9-4DDF-8CDD-066208EB433B:2-16:47-49:60"); + contained5.parse("FBC30C64-F8C9-4DDF-8CDD-066208EB433B:60"); - not_contained1.parse("2174B383-5441-11E8-B90A-C80AA9429562:1-3:11:47-50, 24DA167-0C0C-11E8-8442-00059A3C7B00:1-19:47-49:60"); + not_contained1.parse("2174B383-5441-11E8-B90A-C80AA9429562:1-3:11:47-50, FBC30C64-F8C9-4DDF-8CDD-066208EB433B:1-19:47-49:60"); not_contained2.parse("2174B383-5441-11E8-B90A-C80AA9429562:0-3:11:47-49"); not_contained3.parse("2174B383-5441-11E8-B90A-C80AA9429562:99"); - not_contained4.parse("24DA167-0C0C-11E8-8442-00059A3C7B00:2-16:46-49:60"); - not_contained5.parse("24DA167-0C0C-11E8-8442-00059A3C7B00:99"); - not_contained6.parse("2174B383-5441-11E8-B90A-C80AA9429562:1-3:11:47-49, 24DA167-0C0C-11E8-8442-00059A3C7B00:1-19:47-49:60, 00000000-0000-0000-0000-000000000000"); - + not_contained4.parse("FBC30C64-F8C9-4DDF-8CDD-066208EB433B:2-16:46-49:60"); + not_contained5.parse("FBC30C64-F8C9-4DDF-8CDD-066208EB433B:99"); + not_contained6.parse("2174B383-5441-11E8-B90A-C80AA9429562:1-3:11:47-49, FBC30C64-F8C9-4DDF-8CDD-066208EB433B:1-19:47-49:60, 00000000-0000-0000-0000-000000000000"); ASSERT_TRUE(gtid_set.contains(contained1)); ASSERT_TRUE(gtid_set.contains(contained2)); diff --git a/src/Core/ProtocolDefines.h b/src/Core/ProtocolDefines.h index 7e6893c6d85..790987272fa 100644 --- a/src/Core/ProtocolDefines.h +++ b/src/Core/ProtocolDefines.h @@ -81,6 +81,11 @@ static constexpr auto DBMS_MIN_REVISION_WITH_TABLE_READ_ONLY_CHECK = 54467; static constexpr auto DBMS_MIN_REVISION_WITH_SYSTEM_KEYWORDS_TABLE = 54468; +static constexpr auto DBMS_MIN_REVISION_WITH_ROWS_BEFORE_AGGREGATION = 54469; + +/// Packets size header +static constexpr auto DBMS_MIN_PROTOCOL_VERSION_WITH_CHUNKED_PACKETS = 54470; + /// Version of ClickHouse TCP protocol. /// /// Should be incremented manually on protocol changes. @@ -88,6 +93,6 @@ static constexpr auto DBMS_MIN_REVISION_WITH_SYSTEM_KEYWORDS_TABLE = 54468; /// NOTE: DBMS_TCP_PROTOCOL_VERSION has nothing common with VERSION_REVISION, /// later is just a number for server version (one number instead of commit SHA) /// for simplicity (sometimes it may be more convenient in some use cases). -static constexpr auto DBMS_TCP_PROTOCOL_VERSION = 54468; +static constexpr auto DBMS_TCP_PROTOCOL_VERSION = 54470; } diff --git a/src/Core/Range.cpp b/src/Core/Range.cpp index 956b96653a1..1a5ce1e012e 100644 --- a/src/Core/Range.cpp +++ b/src/Core/Range.cpp @@ -62,27 +62,27 @@ void Range::shrinkToIncludedIfPossible() { if (left.isExplicit() && !left_included) { - if (left.getType() == Field::Types::UInt64 && left.get() != std::numeric_limits::max()) + if (left.getType() == Field::Types::UInt64 && left.safeGet() != std::numeric_limits::max()) { - ++left.get(); + ++left.safeGet(); left_included = true; } - if (left.getType() == Field::Types::Int64 && left.get() != std::numeric_limits::max()) + if (left.getType() == Field::Types::Int64 && left.safeGet() != std::numeric_limits::max()) { - ++left.get(); + ++left.safeGet(); left_included = true; } } if (right.isExplicit() && !right_included) { - if (right.getType() == Field::Types::UInt64 && right.get() != std::numeric_limits::min()) + if (right.getType() == Field::Types::UInt64 && right.safeGet() != std::numeric_limits::min()) { - --right.get(); + --right.safeGet(); right_included = true; } - if (right.getType() == Field::Types::Int64 && right.get() != std::numeric_limits::min()) + if (right.getType() == Field::Types::Int64 && right.safeGet() != std::numeric_limits::min()) { - --right.get(); + --right.safeGet(); right_included = true; } } diff --git a/src/Core/ServerSettings.h b/src/Core/ServerSettings.h index d13e6251ca9..256f78ddea0 100644 --- a/src/Core/ServerSettings.h +++ b/src/Core/ServerSettings.h @@ -134,6 +134,7 @@ namespace DB M(Bool, async_load_databases, false, "Enable asynchronous loading of databases and tables to speedup server startup. Queries to not yet loaded entity will be blocked until load is finished.", 0) \ M(Bool, display_secrets_in_show_and_select, false, "Allow showing secrets in SHOW and SELECT queries via a format setting and a grant", 0) \ M(Seconds, keep_alive_timeout, DEFAULT_HTTP_KEEP_ALIVE_TIMEOUT, "The number of seconds that ClickHouse waits for incoming requests before closing the connection.", 0) \ + M(UInt64, max_keep_alive_requests, 10000, "The maximum number of requests handled via a single http keepalive connection before the server closes this connection.", 0) \ M(Seconds, replicated_fetches_http_connection_timeout, 0, "HTTP connection timeout for part fetch requests. Inherited from default profile `http_connection_timeout` if not set explicitly.", 0) \ M(Seconds, replicated_fetches_http_send_timeout, 0, "HTTP send timeout for part fetch requests. Inherited from default profile `http_send_timeout` if not set explicitly.", 0) \ M(Seconds, replicated_fetches_http_receive_timeout, 0, "HTTP receive timeout for fetch part requests. Inherited from default profile `http_receive_timeout` if not set explicitly.", 0) \ diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 9c9c9c1db00..45bd2b9eb42 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -118,7 +118,7 @@ void Settings::set(std::string_view name, const Field & value) { if (value.getType() != Field::Types::Which::String) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unexpected type of value for setting 'compatibility'. Expected String, got {}", value.getTypeName()); - applyCompatibilitySetting(value.get()); + applyCompatibilitySetting(value.safeGet()); } /// If we change setting that was changed by compatibility setting before /// we should remove it from settings_changed_by_compatibility_setting, diff --git a/src/Core/Settings.h b/src/Core/Settings.h index 0d498ce7699..33e7b3ae6dd 100644 --- a/src/Core/Settings.h +++ b/src/Core/Settings.h @@ -325,6 +325,7 @@ class IColumn; \ 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(Int32, join_output_by_rowlist_perkey_rows_threshold, 5, "The lower limit of per-key average rows in the right table to determine whether to output by row list in hash join.", 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) \ M(Bool, single_join_prefer_left_table, true, "For single JOIN in case of identifier ambiguity prefer left table", IMPORTANT) \ @@ -593,7 +594,6 @@ 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) \ @@ -605,7 +605,7 @@ class IColumn; M(Bool, optimize_if_chain_to_multiif, false, "Replace if(cond1, then1, if(cond2, ...)) chains to multiIf. Currently it's not beneficial for numeric types.", 0) \ M(Bool, optimize_multiif_to_if, true, "Replace 'multiIf' with only one condition to 'if'.", 0) \ M(Bool, optimize_if_transform_strings_to_enum, false, "Replaces string-type arguments in If and Transform to enum. Disabled by default cause it could make inconsistent change in distributed query that would lead to its fail.", 0) \ - M(Bool, optimize_functions_to_subcolumns, false, "Transform functions to subcolumns, if possible, to reduce amount of read data. E.g. 'length(arr)' -> 'arr.size0', 'col IS NULL' -> 'col.null' ", 0) \ + M(Bool, optimize_functions_to_subcolumns, true, "Transform functions to subcolumns, if possible, to reduce amount of read data. E.g. 'length(arr)' -> 'arr.size0', 'col IS NULL' -> 'col.null' ", 0) \ M(Bool, optimize_using_constraints, false, "Use constraints for query optimization", 0) \ M(Bool, optimize_substitute_columns, false, "Use constraints for column substitution", 0) \ M(Bool, optimize_append_index, false, "Use constraints in order to append index condition (indexHint)", 0) \ @@ -676,6 +676,7 @@ class IColumn; M(Bool, query_cache_squash_partial_results, true, "Squash partial result blocks to blocks of size 'max_block_size'. Reduces performance of inserts into the query cache but improves the compressability of cache entries.", 0) \ M(Seconds, query_cache_ttl, 60, "After this time in seconds entries in the query cache become stale", 0) \ M(Bool, query_cache_share_between_users, false, "Allow other users to read entry in the query cache", 0) \ + M(String, query_cache_tag, "", "A string which acts as a label for query cache entries. The same queries with different tags are considered different by the query cache.", 0) \ M(Bool, enable_sharing_sets_for_mutations, true, "Allow sharing set objects build for IN subqueries between different tasks of the same mutation. This reduces memory usage and CPU consumption", 0) \ \ M(Bool, optimize_rewrite_sum_if_to_count_if, true, "Rewrite sumIf() and sum(if()) function countIf() function when logically equivalent", 0) \ @@ -690,6 +691,7 @@ class IColumn; 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, allow_experimental_kafka_offsets_storage_in_keeper, false, "Allow experimental feature to store Kafka related offsets in ClickHouse Keeper. When enabled a ClickHouse Keeper path and replica name can be specified to the Kafka table engine. As a result instead of the regular Kafka engine, a new type of storage engine will be used that stores the committed offsets primarily in ClickHouse Keeper", 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) \ @@ -893,6 +895,9 @@ class IColumn; 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, restore_replace_external_engines_to_null, false, "Replace all the external table engines to Null on restore. Useful for testing purposes", 0) \ + M(Bool, restore_replace_external_table_functions_to_null, false, "Replace all table functions to Null on restore. Useful for testing purposes", 0) \ + M(Bool, create_if_not_exists, false, "Enable IF NOT EXISTS for CREATE statements by default", 0) \ \ \ /* ###################################### */ \ @@ -903,20 +908,19 @@ class IColumn; M(Bool, allow_experimental_nlp_functions, false, "Enable experimental functions for natural language processing.", 0) \ M(Bool, allow_experimental_hash_functions, false, "Enable experimental hash functions", 0) \ M(Bool, allow_experimental_object_type, false, "Allow Object and JSON data types", 0) \ + M(Bool, allow_experimental_time_series_table, false, "Allows experimental TimeSeries table engine", 0) \ + M(Bool, allow_experimental_vector_similarity_index, false, "Allow experimental vector similarity index", 0) \ M(Bool, allow_experimental_variant_type, false, "Allow Variant data type", 0) \ 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) \ M(Bool, throw_on_unsupported_query_inside_transaction, true, "Throw exception if unsupported query is used inside transaction", 0) \ M(TransactionsWaitCSNMode, wait_changes_become_visible_after_commit_mode, TransactionsWaitCSNMode::WAIT_UNKNOWN, "Wait for committed changes to become actually visible in the latest snapshot", 0) \ 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(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, use_hive_partitioning, false, "Allows to use hive partitioning for File, URL, S3, AzureBlobStorage and HDFS engines.", 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) \ @@ -943,8 +947,7 @@ class IColumn; \ 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, allow_experimental_analyzer, true, "Allow new query analyzer.", IMPORTANT) ALIAS(enable_analyzer) \ 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) \ @@ -1033,6 +1036,10 @@ class IColumn; MAKE_OBSOLETE(M, UInt64, parallel_replicas_min_number_of_granules_to_enable, 0) \ MAKE_OBSOLETE(M, Bool, query_plan_optimize_projection, true) \ MAKE_OBSOLETE(M, Bool, query_cache_store_results_of_queries_with_nondeterministic_functions, false) \ + MAKE_OBSOLETE(M, Bool, allow_experimental_annoy_index, false) \ + MAKE_OBSOLETE(M, UInt64, max_threads_for_annoy_index_creation, 4) \ + MAKE_OBSOLETE(M, Int64, annoy_index_search_k_nodes, -1) \ + MAKE_OBSOLETE(M, Bool, allow_experimental_usearch_index, false) \ MAKE_OBSOLETE(M, Bool, optimize_move_functions_out_of_any, false) \ MAKE_OBSOLETE(M, Bool, allow_experimental_undrop_table_query, true) \ MAKE_OBSOLETE(M, Bool, allow_experimental_s3queue, true) \ @@ -1126,11 +1133,11 @@ 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(UInt64, input_format_json_max_depth, 1000, "Maximum depth of a field in JSON. This is not a strict limit, it does not have to be applied precisely.", 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) \ + M(Bool, input_format_try_infer_datetimes_only_datetime64, false, "When input_format_try_infer_datetimes is enabled, infer only DateTime64 but not DateTime types", 0) \ M(Bool, input_format_try_infer_exponent_floats, false, "Try to infer floats in exponential notation while schema inference in text formats (except JSON, where exponent numbers are always inferred)", 0) \ M(Bool, output_format_markdown_escape_special_characters, false, "Escape special characters in Markdown", 0) \ M(Bool, input_format_protobuf_flatten_google_wrappers, false, "Enable Google wrappers for regular non-nested columns, e.g. google.protobuf.StringValue 'str' for String column 'str'. For Nullable columns empty wrappers are recognized as defaults, and missing as nulls", 0) \ @@ -1241,6 +1248,7 @@ class IColumn; M(Bool, insert_distributed_one_random_shard, false, "If setting is enabled, inserting into distributed table will choose a random shard to write when there is no sharding key", 0) \ \ M(Bool, exact_rows_before_limit, false, "When enabled, ClickHouse will provide exact value for rows_before_limit_at_least statistic, but with the cost that the data before limit will have to be read completely", 0) \ + M(Bool, rows_before_aggregation, false, "When enabled, ClickHouse will provide exact value for rows_before_aggregation statistic, represents the number of rows read before aggregation", 0) \ M(UInt64, cross_to_inner_join_rewrite, 1, "Use inner join instead of comma/cross join if there are joining expressions in the WHERE section. Values: 0 - no rewrite, 1 - apply if possible for comma/cross, 2 - force rewrite all comma joins, cross - if possible", 0) \ \ M(Bool, output_format_arrow_low_cardinality_as_dictionary, false, "Enable output LowCardinality type as Dictionary Arrow type", 0) \ diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 5b94391bade..b8639512b76 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -75,9 +75,22 @@ static std::initializer_list(f.get()); + return stringToNumber(f.safeGet()); } else if (f.getType() == Field::Types::UInt64) { T result; - if (!accurate::convertNumeric(f.get(), result)) + if (!accurate::convertNumeric(f.safeGet(), result)) throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Field value {} is out of range of {} type", f, demangle(typeid(T).name())); return result; } else if (f.getType() == Field::Types::Int64) { T result; - if (!accurate::convertNumeric(f.get(), result)) + if (!accurate::convertNumeric(f.safeGet(), result)) throw Exception(ErrorCodes::CANNOT_CONVERT_TYPE, "Field value {} is out of range of {} type", f, demangle(typeid(T).name())); return result; } else if (f.getType() == Field::Types::Bool) { - return T(f.get()); + return T(f.safeGet()); } else if (f.getType() == Field::Types::Float64) { - Float64 x = f.get(); + Float64 x = f.safeGet(); if constexpr (std::is_floating_point_v) { return T(x); @@ -120,7 +120,7 @@ namespace if (f.getType() == Field::Types::String) { /// Allow to parse Map from string field. For the convenience. - const auto & str = f.get(); + const auto & str = f.safeGet(); return stringToMap(str); } @@ -218,7 +218,7 @@ namespace UInt64 fieldToMaxThreads(const Field & f) { if (f.getType() == Field::Types::String) - return stringToMaxThreads(f.get()); + return stringToMaxThreads(f.safeGet()); else return fieldToNumber(f); } @@ -237,7 +237,7 @@ SettingFieldMaxThreads & SettingFieldMaxThreads::operator=(const Field & f) String SettingFieldMaxThreads::toString() const { if (is_auto) - return "'auto(" + ::DB::toString(value) + ")'"; + return "auto(" + ::DB::toString(value) + ")"; else return ::DB::toString(value); } diff --git a/src/Core/SettingsFields.h b/src/Core/SettingsFields.h index 266141815e3..533d69f3fbb 100644 --- a/src/Core/SettingsFields.h +++ b/src/Core/SettingsFields.h @@ -153,7 +153,7 @@ struct SettingFieldMaxThreads operator UInt64() const { return value; } /// NOLINT explicit operator Field() const { return value; } - /// Writes "auto()" instead of simple "" if `is_auto==true`. + /// Writes "auto()" instead of simple "" if `is_auto == true`. String toString() const; void parseFromString(const String & str); diff --git a/src/Core/SettingsQuirks.cpp b/src/Core/SettingsQuirks.cpp index 5541cc19653..3127a5ef36d 100644 --- a/src/Core/SettingsQuirks.cpp +++ b/src/Core/SettingsQuirks.cpp @@ -100,7 +100,7 @@ void doSettingsSanityCheckClamp(Settings & current_settings, LoggerPtr log) return current_value; }; - UInt64 max_threads = get_current_value("max_threads").get(); + UInt64 max_threads = get_current_value("max_threads").safeGet(); UInt64 max_threads_max_value = 256 * getNumberOfPhysicalCPUCores(); if (max_threads > max_threads_max_value) { @@ -120,7 +120,7 @@ void doSettingsSanityCheckClamp(Settings & current_settings, LoggerPtr log) "input_format_parquet_max_block_size"}; for (auto const & setting : block_rows_settings) { - if (auto block_size = get_current_value(setting).get(); + if (auto block_size = get_current_value(setting).safeGet(); block_size > max_sane_block_rows_size) { if (log) @@ -129,7 +129,7 @@ void doSettingsSanityCheckClamp(Settings & current_settings, LoggerPtr log) } } - if (auto max_block_size = get_current_value("max_block_size").get(); max_block_size == 0) + if (auto max_block_size = get_current_value("max_block_size").safeGet(); max_block_size == 0) { if (log) LOG_WARNING(log, "Sanity check: 'max_block_size' cannot be 0. Set to default value {}", DEFAULT_BLOCK_SIZE); diff --git a/src/Core/examples/field.cpp b/src/Core/examples/field.cpp index 110e11d0cb1..3064290e127 100644 --- a/src/Core/examples/field.cpp +++ b/src/Core/examples/field.cpp @@ -37,7 +37,7 @@ int main(int argc, char ** argv) std::cerr << applyVisitor(to_string, field) << std::endl; } - field.get().push_back(field); + field.safeGet().push_back(field); std::cerr << applyVisitor(to_string, field) << std::endl; std::cerr << (field < field2) << std::endl; diff --git a/src/Core/fuzzers/CMakeLists.txt b/src/Core/fuzzers/CMakeLists.txt index 3c5c0eed4e6..61d6b9629eb 100644 --- a/src/Core/fuzzers/CMakeLists.txt +++ b/src/Core/fuzzers/CMakeLists.txt @@ -1,2 +1,2 @@ clickhouse_add_executable (names_and_types_fuzzer names_and_types_fuzzer.cpp) -target_link_libraries (names_and_types_fuzzer PRIVATE dbms clickhouse_functions) +target_link_libraries (names_and_types_fuzzer PRIVATE clickhouse_functions) diff --git a/src/Core/tests/gtest_field.cpp b/src/Core/tests/gtest_field.cpp index 5585442d835..7e778be9575 100644 --- a/src/Core/tests/gtest_field.cpp +++ b/src/Core/tests/gtest_field.cpp @@ -8,31 +8,31 @@ GTEST_TEST(Field, FromBool) { Field f{false}; ASSERT_EQ(f.getType(), Field::Types::Bool); - ASSERT_EQ(f.get(), 0); - ASSERT_EQ(f.get(), false); + ASSERT_EQ(f.safeGet(), 0); + ASSERT_EQ(f.safeGet(), false); } { Field f{true}; ASSERT_EQ(f.getType(), Field::Types::Bool); - ASSERT_EQ(f.get(), 1); - ASSERT_EQ(f.get(), true); + ASSERT_EQ(f.safeGet(), 1); + ASSERT_EQ(f.safeGet(), true); } { Field f; f = false; ASSERT_EQ(f.getType(), Field::Types::Bool); - ASSERT_EQ(f.get(), 0); - ASSERT_EQ(f.get(), false); + ASSERT_EQ(f.safeGet(), 0); + ASSERT_EQ(f.safeGet(), false); } { Field f; f = true; ASSERT_EQ(f.getType(), Field::Types::Bool); - ASSERT_EQ(f.get(), 1); - ASSERT_EQ(f.get(), true); + ASSERT_EQ(f.safeGet(), 1); + ASSERT_EQ(f.safeGet(), true); } } @@ -42,15 +42,15 @@ GTEST_TEST(Field, Move) Field f; f = Field{String{"Hello, world (1)"}}; - ASSERT_EQ(f.get(), "Hello, world (1)"); + ASSERT_EQ(f.safeGet(), "Hello, world (1)"); f = Field{String{"Hello, world (2)"}}; - ASSERT_EQ(f.get(), "Hello, world (2)"); + ASSERT_EQ(f.safeGet(), "Hello, world (2)"); f = Field{Array{Field{String{"Hello, world (3)"}}}}; - ASSERT_EQ(f.get()[0].get(), "Hello, world (3)"); + ASSERT_EQ(f.safeGet()[0].safeGet(), "Hello, world (3)"); f = String{"Hello, world (4)"}; - ASSERT_EQ(f.get(), "Hello, world (4)"); + ASSERT_EQ(f.safeGet(), "Hello, world (4)"); f = Array{Field{String{"Hello, world (5)"}}}; - ASSERT_EQ(f.get()[0].get(), "Hello, world (5)"); + ASSERT_EQ(f.safeGet()[0].safeGet(), "Hello, world (5)"); f = Array{String{"Hello, world (6)"}}; - ASSERT_EQ(f.get()[0].get(), "Hello, world (6)"); + ASSERT_EQ(f.safeGet()[0].safeGet(), "Hello, world (6)"); } diff --git a/src/DataTypes/DataTypeAggregateFunction.cpp b/src/DataTypes/DataTypeAggregateFunction.cpp index 09175617bf1..1facaaab0d6 100644 --- a/src/DataTypes/DataTypeAggregateFunction.cpp +++ b/src/DataTypes/DataTypeAggregateFunction.cpp @@ -33,6 +33,16 @@ namespace ErrorCodes extern const int LOGICAL_ERROR; } + +DataTypeAggregateFunction::DataTypeAggregateFunction(AggregateFunctionPtr function_, const DataTypes & argument_types_, + const Array & parameters_, std::optional version_) + : function(std::move(function_)) + , argument_types(argument_types_) + , parameters(parameters_) + , version(version_) +{ +} + String DataTypeAggregateFunction::getFunctionName() const { return function->getName(); @@ -119,7 +129,7 @@ MutableColumnPtr DataTypeAggregateFunction::createColumn() const Field DataTypeAggregateFunction::getDefault() const { Field field = AggregateFunctionStateData(); - field.get().name = getName(); + field.safeGet().name = getName(); AlignedBuffer place_buffer(function->sizeOfData(), function->alignOfData()); AggregateDataPtr place = place_buffer.data(); @@ -128,7 +138,7 @@ Field DataTypeAggregateFunction::getDefault() const try { - WriteBufferFromString buffer_from_field(field.get().data); + WriteBufferFromString buffer_from_field(field.safeGet().data); function->serialize(place, buffer_from_field, version); } catch (...) diff --git a/src/DataTypes/DataTypeAggregateFunction.h b/src/DataTypes/DataTypeAggregateFunction.h index 52ed151107e..e3a4f9726d9 100644 --- a/src/DataTypes/DataTypeAggregateFunction.h +++ b/src/DataTypes/DataTypeAggregateFunction.h @@ -30,13 +30,7 @@ public: static constexpr bool is_parametric = true; DataTypeAggregateFunction(AggregateFunctionPtr function_, const DataTypes & argument_types_, - const Array & parameters_, std::optional version_ = std::nullopt) - : function(std::move(function_)) - , argument_types(argument_types_) - , parameters(parameters_) - , version(version_) - { - } + const Array & parameters_, std::optional version_ = std::nullopt); size_t getVersion() const; diff --git a/src/DataTypes/DataTypeCustomGeo.cpp b/src/DataTypes/DataTypeCustomGeo.cpp index 0736d837d46..f90788ec403 100644 --- a/src/DataTypes/DataTypeCustomGeo.cpp +++ b/src/DataTypes/DataTypeCustomGeo.cpp @@ -24,6 +24,13 @@ void registerDataTypeDomainGeo(DataTypeFactory & factory) std::make_unique(std::make_unique())); }); + // Custom type for multiple lines stored as Array(LineString) + factory.registerSimpleDataTypeCustom("MultiLineString", [] + { + return std::make_pair(DataTypeFactory::instance().get("Array(LineString)"), + std::make_unique(std::make_unique())); + }); + // Custom type for simple polygon without holes stored as Array(Point) factory.registerSimpleDataTypeCustom("Ring", [] { diff --git a/src/DataTypes/DataTypeCustomGeo.h b/src/DataTypes/DataTypeCustomGeo.h index 0a1c83e4638..6a632f0d05c 100644 --- a/src/DataTypes/DataTypeCustomGeo.h +++ b/src/DataTypes/DataTypeCustomGeo.h @@ -17,6 +17,12 @@ public: DataTypeLineStringName() : DataTypeCustomFixedName("LineString") {} }; +class DataTypeMultiLineStringName : public DataTypeCustomFixedName +{ +public: + DataTypeMultiLineStringName() : DataTypeCustomFixedName("MultiLineString") {} +}; + class DataTypeRingName : public DataTypeCustomFixedName { public: diff --git a/src/DataTypes/DataTypeDynamic.cpp b/src/DataTypes/DataTypeDynamic.cpp index a1b1f8325f0..5f713e9adc9 100644 --- a/src/DataTypes/DataTypeDynamic.cpp +++ b/src/DataTypes/DataTypeDynamic.cpp @@ -7,6 +7,7 @@ #include #include #include +#include #include #include #include @@ -14,6 +15,7 @@ #include #include #include +#include namespace DB { @@ -71,10 +73,10 @@ 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() > 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"); + if (!literal || literal->value.getType() != Field::Types::UInt64 || literal->value.safeGet() > ColumnDynamic::MAX_DYNAMIC_TYPES_LIMIT) + throw Exception(ErrorCodes::UNEXPECTED_AST_STRUCTURE, "'max_types' argument for Dynamic type should be a positive integer between 0 and {}", ColumnDynamic::MAX_DYNAMIC_TYPES_LIMIT); - return std::make_shared(literal->value.get()); + return std::make_shared(literal->value.safeGet()); } void registerDataTypeDynamic(DataTypeFactory & factory) @@ -84,30 +86,72 @@ void registerDataTypeDynamic(DataTypeFactory & factory) std::unique_ptr DataTypeDynamic::getDynamicSubcolumnData(std::string_view subcolumn_name, const DB::IDataType::SubstreamData & data, bool throw_if_null) const { - auto [subcolumn_type_name, subcolumn_nested_name] = Nested::splitName(subcolumn_name); + auto [type_subcolumn_name, subcolumn_nested_name] = Nested::splitName(subcolumn_name); /// Check if requested subcolumn is a valid data type. - auto subcolumn_type = DataTypeFactory::instance().tryGet(String(subcolumn_type_name)); + auto subcolumn_type = DataTypeFactory::instance().tryGet(String(type_subcolumn_name)); if (!subcolumn_type) { if (throw_if_null) - throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Dynamic type doesn't have subcolumn '{}'", subcolumn_type_name); + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Dynamic type doesn't have subcolumn '{}'", type_subcolumn_name); return nullptr; } std::unique_ptr res = std::make_unique(subcolumn_type->getDefaultSerialization()); res->type = subcolumn_type; std::optional discriminator; + ColumnPtr null_map_for_variant_from_shared_variant; if (data.column) { /// If column was provided, we should extract subcolumn from Dynamic column. const auto & dynamic_column = assert_cast(*data.column); const auto & variant_info = dynamic_column.getVariantInfo(); + const auto & variant_column = dynamic_column.getVariantColumn(); + const auto & shared_variant = dynamic_column.getSharedVariant(); /// Check if provided Dynamic column has subcolumn of this type. - auto it = variant_info.variant_name_to_discriminator.find(subcolumn_type->getName()); + String subcolumn_type_name = subcolumn_type->getName(); + auto it = variant_info.variant_name_to_discriminator.find(subcolumn_type_name); if (it != variant_info.variant_name_to_discriminator.end()) { discriminator = it->second; - res->column = dynamic_column.getVariantColumn().getVariantPtrByGlobalDiscriminator(*discriminator); + res->column = variant_column.getVariantPtrByGlobalDiscriminator(*discriminator); + } + /// Otherwise if there is data in shared variant try to find requested type there. + else if (!shared_variant.empty()) + { + /// Create null map for resulting subcolumn to make it Nullable. + auto null_map_column = ColumnUInt8::create(); + NullMap & null_map = assert_cast(*null_map_column).getData(); + null_map.reserve(variant_column.size()); + auto subcolumn = subcolumn_type->createColumn(); + auto shared_variant_local_discr = variant_column.localDiscriminatorByGlobal(dynamic_column.getSharedVariantDiscriminator()); + const auto & local_discriminators = variant_column.getLocalDiscriminators(); + const auto & offsets = variant_column.getOffsets(); + const FormatSettings format_settings; + for (size_t i = 0; i != local_discriminators.size(); ++i) + { + if (local_discriminators[i] == shared_variant_local_discr) + { + auto value = shared_variant.getDataAt(offsets[i]); + ReadBufferFromMemory buf(value.data, value.size); + auto type = decodeDataType(buf); + if (type->getName() == subcolumn_type_name) + { + dynamic_column.getVariantSerialization(subcolumn_type, subcolumn_type_name)->deserializeBinary(*subcolumn, buf, format_settings); + null_map.push_back(0); + } + else + { + null_map.push_back(1); + } + } + else + { + null_map.push_back(1); + } + } + + res->column = std::move(subcolumn); + null_map_for_variant_from_shared_variant = std::move(null_map_column); } } @@ -125,7 +169,7 @@ std::unique_ptr DataTypeDynamic::getDynamicSubcolumnDa return nullptr; } - res->serialization = std::make_shared(res->serialization, subcolumn_type->getName(), is_null_map_subcolumn); + res->serialization = std::make_shared(res->serialization, subcolumn_type->getName(), String(subcolumn_nested_name), 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) @@ -133,10 +177,10 @@ std::unique_ptr DataTypeDynamic::getDynamicSubcolumnDa if (data.column) { + /// Check if provided Dynamic column has subcolumn of this type. In this case we should use VariantSubcolumnCreator/VariantNullMapSubcolumnCreator to + /// create full subcolumn from variant according to discriminators. if (discriminator) { - /// 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(); std::unique_ptr creator; if (is_null_map_subcolumn) @@ -154,6 +198,21 @@ std::unique_ptr DataTypeDynamic::getDynamicSubcolumnDa make_subcolumn_nullable); res->column = creator->create(res->column); } + /// Check if requested type was extracted from shared variant. In this case we should use + /// VariantSubcolumnCreator to create full subcolumn from variant according to created null map. + else if (null_map_for_variant_from_shared_variant) + { + if (is_null_map_subcolumn) + { + res->column = null_map_for_variant_from_shared_variant; + } + else + { + SerializationVariantElement::VariantSubcolumnCreator creator( + null_map_for_variant_from_shared_variant, "", 0, 0, make_subcolumn_nullable, null_map_for_variant_from_shared_variant); + 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) { diff --git a/src/DataTypes/DataTypeEnum.cpp b/src/DataTypes/DataTypeEnum.cpp index 08e0c0d2045..b9a5a1a5a68 100644 --- a/src/DataTypes/DataTypeEnum.cpp +++ b/src/DataTypes/DataTypeEnum.cpp @@ -122,12 +122,12 @@ Field DataTypeEnum::castToName(const Field & value_or_name) const { if (value_or_name.getType() == Field::Types::String) { - this->getValue(value_or_name.get()); /// Check correctness - return value_or_name.get(); + this->getValue(value_or_name.safeGet()); /// Check correctness + return value_or_name.safeGet(); } else if (value_or_name.getType() == Field::Types::Int64) { - Int64 value = value_or_name.get(); + Int64 value = value_or_name.safeGet(); checkOverflow(value); return this->getNameForValue(static_cast(value)).toString(); } @@ -141,12 +141,12 @@ Field DataTypeEnum::castToValue(const Field & value_or_name) const { if (value_or_name.getType() == Field::Types::String) { - return this->getValue(value_or_name.get()); + return this->getValue(value_or_name.safeGet()); } else if (value_or_name.getType() == Field::Types::Int64 || value_or_name.getType() == Field::Types::UInt64) { - Int64 value = value_or_name.get(); + Int64 value = value_or_name.safeGet(); checkOverflow(value); this->getNameForValue(static_cast(value)); /// Check correctness return value; @@ -220,7 +220,7 @@ static void autoAssignNumberForEnum(const ASTPtr & arguments) "Elements of Enum data type must be of form: " "'name' = number or 'name', where name is string literal and number is an integer"); - literal_child_assign_num = value_literal->value.get(); + literal_child_assign_num = value_literal->value.safeGet(); } assign_number_child.emplace_back(child); } @@ -269,8 +269,8 @@ static DataTypePtr createExact(const ASTPtr & arguments) "Elements of Enum data type must be of form: " "'name' = number or 'name', where name is string literal and number is an integer"); - const String & field_name = name_literal->value.get(); - const auto value = value_literal->value.get(); + const String & field_name = name_literal->value.safeGet(); + const auto value = value_literal->value.safeGet(); if (value > std::numeric_limits::max() || value < std::numeric_limits::min()) throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "Value {} for element '{}' exceeds range of {}", @@ -302,7 +302,7 @@ static DataTypePtr create(const ASTPtr & arguments) "Elements of Enum data type must be of form: " "'name' = number or 'name', where name is string literal and number is an integer"); - Int64 value = value_literal->value.get(); + Int64 value = value_literal->value.safeGet(); if (value > std::numeric_limits::max() || value < std::numeric_limits::min()) return createExact(arguments); diff --git a/src/DataTypes/DataTypeFactory.cpp b/src/DataTypes/DataTypeFactory.cpp index 6f7dcd65b83..ca2ebdfbdbb 100644 --- a/src/DataTypes/DataTypeFactory.cpp +++ b/src/DataTypes/DataTypeFactory.cpp @@ -150,6 +150,12 @@ DataTypePtr DataTypeFactory::getCustom(DataTypeCustomDescPtr customization) cons return type; } +DataTypePtr DataTypeFactory::getCustom(const String & base_name, DataTypeCustomDescPtr customization) const +{ + auto type = get(base_name); + type->setCustomization(std::move(customization)); + return type; +} void DataTypeFactory::registerDataType(const String & family_name, Value creator, Case case_sensitiveness) { diff --git a/src/DataTypes/DataTypeFactory.h b/src/DataTypes/DataTypeFactory.h index edba9886d1c..a8324341691 100644 --- a/src/DataTypes/DataTypeFactory.h +++ b/src/DataTypes/DataTypeFactory.h @@ -34,6 +34,7 @@ public: DataTypePtr get(const String & family_name, const ASTPtr & parameters) const; DataTypePtr get(const ASTPtr & ast) const; DataTypePtr getCustom(DataTypeCustomDescPtr customization) const; + DataTypePtr getCustom(const String & base_name, DataTypeCustomDescPtr customization) const; /// Return nullptr in case of error. DataTypePtr tryGet(const String & full_name) const; diff --git a/src/DataTypes/DataTypeFixedString.cpp b/src/DataTypes/DataTypeFixedString.cpp index 080ff8826a5..63d5245287f 100644 --- a/src/DataTypes/DataTypeFixedString.cpp +++ b/src/DataTypes/DataTypeFixedString.cpp @@ -51,11 +51,11 @@ static DataTypePtr create(const ASTPtr & arguments) "FixedString data type family must have exactly one argument - size in bytes"); const auto * argument = arguments->children[0]->as(); - if (!argument || argument->value.getType() != Field::Types::UInt64 || argument->value.get() == 0) + if (!argument || argument->value.getType() != Field::Types::UInt64 || argument->value.safeGet() == 0) throw Exception(ErrorCodes::UNEXPECTED_AST_STRUCTURE, "FixedString data type family must have a number (positive integer) as its argument"); - return std::make_shared(argument->value.get()); + return std::make_shared(argument->value.safeGet()); } diff --git a/src/DataTypes/DataTypeObject.cpp b/src/DataTypes/DataTypeObject.cpp index 91b9bfcb2a5..5fd69688194 100644 --- a/src/DataTypes/DataTypeObject.cpp +++ b/src/DataTypes/DataTypeObject.cpp @@ -69,7 +69,7 @@ static DataTypePtr create(const ASTPtr & arguments) throw Exception(ErrorCodes::UNEXPECTED_AST_STRUCTURE, "Object data type family must have a const string as its schema name parameter"); - return std::make_shared(literal->value.get(), is_nullable); + return std::make_shared(literal->value.safeGet(), is_nullable); } void registerDataTypeObject(DataTypeFactory & factory) diff --git a/src/DataTypes/DataTypeTuple.cpp b/src/DataTypes/DataTypeTuple.cpp index e96937d522d..67b4a0a5e31 100644 --- a/src/DataTypes/DataTypeTuple.cpp +++ b/src/DataTypes/DataTypeTuple.cpp @@ -192,17 +192,12 @@ MutableColumnPtr DataTypeTuple::createColumn() const MutableColumnPtr DataTypeTuple::createColumn(const ISerialization & serialization) const { - /// If we read Tuple as Variant subcolumn, it may be wrapped to SerializationVariantElement. - /// Here we don't need it, so we drop this wrapper. - const auto * current_serialization = &serialization; - while (const auto * serialization_variant_element = typeid_cast(current_serialization)) - current_serialization = serialization_variant_element->getNested().get(); - - /// If we read subcolumn of nested Tuple, it may be wrapped to SerializationNamed + /// If we read subcolumn of nested Tuple or this Tuple is a subcolumn, it may be wrapped to SerializationWrapper /// several times to allow to reconstruct the substream path name. /// Here we don't need substream path name, so we drop first several wrapper serializations. - while (const auto * serialization_named = typeid_cast(current_serialization)) - current_serialization = serialization_named->getNested().get(); + const auto * current_serialization = &serialization; + while (const auto * serialization_wrapper = dynamic_cast(current_serialization)) + current_serialization = serialization_wrapper->getNested().get(); const auto * serialization_tuple = typeid_cast(current_serialization); if (!serialization_tuple) diff --git a/src/DataTypes/DataTypeVariant.cpp b/src/DataTypes/DataTypeVariant.cpp index 8a10ca7d06d..cc8d04e94da 100644 --- a/src/DataTypes/DataTypeVariant.cpp +++ b/src/DataTypes/DataTypeVariant.cpp @@ -117,7 +117,7 @@ bool DataTypeVariant::equals(const IDataType & rhs) const /// The same data types with different custom names considered different. /// For example, UInt8 and Bool. - if ((variants[i]->hasCustomName() || rhs_variant.variants[i]) && variants[i]->getName() != rhs_variant.variants[i]->getName()) + if ((variants[i]->hasCustomName() || rhs_variant.variants[i]->hasCustomName()) && variants[i]->getName() != rhs_variant.variants[i]->getName()) return false; } diff --git a/src/DataTypes/DataTypesBinaryEncoding.cpp b/src/DataTypes/DataTypesBinaryEncoding.cpp index bd994e313ba..610f246265e 100644 --- a/src/DataTypes/DataTypesBinaryEncoding.cpp +++ b/src/DataTypes/DataTypesBinaryEncoding.cpp @@ -444,7 +444,7 @@ void encodeDataType(const DataTypePtr & type, WriteBuffer & buf) case BinaryTypeIndex::Dynamic: { const auto & dynamic_type = assert_cast(*type); - /// Maximum number of dynamic types is 255, we can write it as 1 byte. + /// Maximum number of dynamic types is 254, we can write it as 1 byte. writeBinary(UInt8(dynamic_type.getMaxDynamicTypes()), buf); break; } diff --git a/src/DataTypes/DataTypesDecimal.cpp b/src/DataTypes/DataTypesDecimal.cpp index a427fd0717a..1d8f7711de1 100644 --- a/src/DataTypes/DataTypesDecimal.cpp +++ b/src/DataTypes/DataTypesDecimal.cpp @@ -80,14 +80,14 @@ static DataTypePtr create(const ASTPtr & arguments) const auto * precision_arg = arguments->children[0]->as(); if (!precision_arg || precision_arg->value.getType() != Field::Types::UInt64) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Decimal argument precision is invalid"); - precision = precision_arg->value.get(); + precision = precision_arg->value.safeGet(); if (arguments->children.size() == 2) { const auto * scale_arg = arguments->children[1]->as(); if (!scale_arg || !isInt64OrUInt64FieldType(scale_arg->value.getType())) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Decimal argument scale is invalid"); - scale = scale_arg->value.get(); + scale = scale_arg->value.safeGet(); } } @@ -107,7 +107,7 @@ static DataTypePtr createExact(const ASTPtr & arguments) "Decimal32 | Decimal64 | Decimal128 | Decimal256 data type family must have a one number as its argument"); UInt64 precision = DecimalUtils::max_precision; - UInt64 scale = scale_arg->value.get(); + UInt64 scale = scale_arg->value.safeGet(); return createDecimal(precision, scale); } diff --git a/src/DataTypes/IDataType.cpp b/src/DataTypes/IDataType.cpp index 1cb64b65d3a..824bc6e33b0 100644 --- a/src/DataTypes/IDataType.cpp +++ b/src/DataTypes/IDataType.cpp @@ -90,7 +90,9 @@ void IDataType::forEachSubcolumn( { auto name = ISerialization::getSubcolumnNameForStream(subpath, prefix_len); auto subdata = ISerialization::createFromPath(subpath, prefix_len); - callback(subpath, name, subdata); + auto path_copy = subpath; + path_copy.resize(prefix_len); + callback(path_copy, name, subdata); } subpath[i].visited = true; } diff --git a/src/DataTypes/ObjectUtils.cpp b/src/DataTypes/ObjectUtils.cpp index 1d525e5987f..fe8a5ede102 100644 --- a/src/DataTypes/ObjectUtils.cpp +++ b/src/DataTypes/ObjectUtils.cpp @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include @@ -66,6 +67,36 @@ DataTypePtr getBaseTypeOfArray(const DataTypePtr & type) return last_array ? last_array->getNestedType() : type; } +DataTypePtr getBaseTypeOfArray(DataTypePtr type, const Names & tuple_elements) +{ + auto it = tuple_elements.begin(); + while (true) + { + if (const auto * type_array = typeid_cast(type.get())) + { + type = type_array->getNestedType(); + } + else if (const auto * type_tuple = typeid_cast(type.get())) + { + if (it == tuple_elements.end()) + break; + + auto pos = type_tuple->tryGetPositionByName(*it); + if (!pos) + break; + + ++it; + type = type_tuple->getElement(*pos); + } + else + { + break; + } + } + + return type; +} + ColumnPtr getBaseColumnOfArray(const ColumnPtr & column) { /// Get raw pointers to avoid extra copying of column pointers. @@ -104,7 +135,7 @@ Array createEmptyArrayField(size_t num_dimensions) for (size_t i = 1; i < num_dimensions; ++i) { current_array->push_back(Array()); - current_array = ¤t_array->back().get(); + current_array = ¤t_array->back().safeGet(); } return array; diff --git a/src/DataTypes/ObjectUtils.h b/src/DataTypes/ObjectUtils.h index 6599d8adef1..21e5c3b2f59 100644 --- a/src/DataTypes/ObjectUtils.h +++ b/src/DataTypes/ObjectUtils.h @@ -27,6 +27,9 @@ size_t getNumberOfDimensions(const IColumn & column); /// Returns type of scalars of Array of arbitrary dimensions. DataTypePtr getBaseTypeOfArray(const DataTypePtr & type); +/// The same as above but takes into account Tuples of Nested. +DataTypePtr getBaseTypeOfArray(DataTypePtr type, const Names & tuple_elements); + /// Returns Array type with requested scalar type and number of dimensions. DataTypePtr createArrayOfType(DataTypePtr type, size_t num_dimensions); diff --git a/src/DataTypes/Serializations/ISerialization.h b/src/DataTypes/Serializations/ISerialization.h index 255dbbfadd2..5d0bf60c59f 100644 --- a/src/DataTypes/Serializations/ISerialization.h +++ b/src/DataTypes/Serializations/ISerialization.h @@ -195,7 +195,7 @@ public: /// Types of substreams that can have arbitrary name. static const std::set named_types; - Type type; + Type type = Type::Regular; /// The name of a variant element type. String variant_element_name; @@ -212,6 +212,7 @@ public: /// Flag, that may help to traverse substream paths. mutable bool visited = false; + Substream() = default; Substream(Type type_) : type(type_) {} /// NOLINT String toString() const; }; diff --git a/src/DataTypes/Serializations/JSONDataParser.cpp b/src/DataTypes/Serializations/JSONDataParser.cpp index 56641424396..0f74815f5b4 100644 --- a/src/DataTypes/Serializations/JSONDataParser.cpp +++ b/src/DataTypes/Serializations/JSONDataParser.cpp @@ -131,7 +131,7 @@ void JSONDataParser::traverseArrayElement(const Element & element, P auto nested_hash = getHashOfNestedPath(paths[i], values[i]); if (nested_hash) { - size_t array_size = values[i].template get().size(); + size_t array_size = values[i].template safeGet().size(); auto & current_nested_sizes = ctx.nested_sizes_by_path[*nested_hash]; if (current_nested_sizes.size() == ctx.current_size) @@ -154,7 +154,7 @@ void JSONDataParser::traverseArrayElement(const Element & element, P auto nested_hash = getHashOfNestedPath(paths[i], values[i]); if (nested_hash) { - size_t array_size = values[i].template get().size(); + size_t array_size = values[i].template safeGet().size(); auto & current_nested_sizes = ctx.nested_sizes_by_path[*nested_hash]; if (current_nested_sizes.empty()) diff --git a/src/DataTypes/Serializations/SerializationAggregateFunction.cpp b/src/DataTypes/Serializations/SerializationAggregateFunction.cpp index 55f7641e058..41b198890e4 100644 --- a/src/DataTypes/Serializations/SerializationAggregateFunction.cpp +++ b/src/DataTypes/Serializations/SerializationAggregateFunction.cpp @@ -16,14 +16,14 @@ namespace DB void SerializationAggregateFunction::serializeBinary(const Field & field, WriteBuffer & ostr, const FormatSettings &) const { - const AggregateFunctionStateData & state = field.get(); + const AggregateFunctionStateData & state = field.safeGet(); writeBinary(state.data, ostr); } void SerializationAggregateFunction::deserializeBinary(Field & field, ReadBuffer & istr, const FormatSettings &) const { field = AggregateFunctionStateData(); - AggregateFunctionStateData & s = field.get(); + AggregateFunctionStateData & s = field.safeGet(); readBinary(s.data, istr); s.name = type_name; } diff --git a/src/DataTypes/Serializations/SerializationArray.cpp b/src/DataTypes/Serializations/SerializationArray.cpp index b7d43332085..0a9c4529e23 100644 --- a/src/DataTypes/Serializations/SerializationArray.cpp +++ b/src/DataTypes/Serializations/SerializationArray.cpp @@ -29,7 +29,7 @@ static constexpr size_t MAX_ARRAYS_SIZE = 1ULL << 40; void SerializationArray::serializeBinary(const Field & field, WriteBuffer & ostr, const FormatSettings & settings) const { - const Array & a = field.get(); + const Array & a = field.safeGet(); writeVarUInt(a.size(), ostr); for (const auto & i : a) { @@ -51,7 +51,7 @@ void SerializationArray::deserializeBinary(Field & field, ReadBuffer & istr, con settings.binary.max_binary_string_size); field = Array(); - Array & arr = field.get(); + Array & arr = field.safeGet(); arr.reserve(size); for (size_t i = 0; i < size; ++i) nested->deserializeBinary(arr.emplace_back(), istr, settings); diff --git a/src/DataTypes/Serializations/SerializationDecimalBase.cpp b/src/DataTypes/Serializations/SerializationDecimalBase.cpp index 49dc042e872..8927f949368 100644 --- a/src/DataTypes/Serializations/SerializationDecimalBase.cpp +++ b/src/DataTypes/Serializations/SerializationDecimalBase.cpp @@ -13,7 +13,7 @@ namespace DB template void SerializationDecimalBase::serializeBinary(const Field & field, WriteBuffer & ostr, const FormatSettings &) const { - FieldType x = field.get>(); + FieldType x = field.safeGet>(); writeBinaryLittleEndian(x, ostr); } diff --git a/src/DataTypes/Serializations/SerializationDynamic.cpp b/src/DataTypes/Serializations/SerializationDynamic.cpp index 7609ffc91ca..9cd0adcc2ed 100644 --- a/src/DataTypes/Serializations/SerializationDynamic.cpp +++ b/src/DataTypes/Serializations/SerializationDynamic.cpp @@ -27,15 +27,21 @@ namespace ErrorCodes struct SerializeBinaryBulkStateDynamic : public ISerialization::SerializeBinaryBulkState { SerializationDynamic::DynamicStructureSerializationVersion structure_version; + size_t max_dynamic_types; DataTypePtr variant_type; Names variant_names; SerializationPtr variant_serialization; ISerialization::SerializeBinaryBulkStatePtr variant_state; - /// Variants statistics. Map (Variant name) -> (Variant size). - ColumnDynamic::Statistics statistics = { .source = ColumnDynamic::Statistics::Source::READ, .data = {} }; + /// Variants statistics. + ColumnDynamic::Statistics statistics; + /// If true, statistics will be recalculated during serialization. + bool recalculate_statistics = false; - explicit SerializeBinaryBulkStateDynamic(UInt64 structure_version_) : structure_version(structure_version_) {} + explicit SerializeBinaryBulkStateDynamic(UInt64 structure_version_) + : structure_version(structure_version_), statistics(ColumnDynamic::Statistics::Source::READ) + { + } }; struct DeserializeBinaryBulkStateDynamic : public ISerialization::DeserializeBinaryBulkState @@ -106,20 +112,41 @@ void SerializationDynamic::serializeBinaryBulkStatePrefix( writeBinaryLittleEndian(structure_version, *stream); auto dynamic_state = std::make_shared(structure_version); + dynamic_state->max_dynamic_types = column_dynamic.getMaxDynamicTypes(); + /// Write max_dynamic_types parameter, because it can differ from the max_dynamic_types + /// that is specified in the Dynamic type (we could decrease it before merge). + writeBinaryLittleEndian(dynamic_state->max_dynamic_types, *stream); + dynamic_state->variant_type = variant_info.variant_type; dynamic_state->variant_names = variant_info.variant_names; const auto & variant_column = column_dynamic.getVariantColumn(); - /// Write internal Variant type name. + /// Write information about variants. + size_t num_variants = dynamic_state->variant_names.size() - 1; /// Don't write shared variant, Dynamic column should always have it. + writeBinaryLittleEndian(num_variants, *stream); if (settings.data_types_binary_encoding) - encodeDataType(dynamic_state->variant_type, *stream); + { + const auto & variants = assert_cast(*dynamic_state->variant_type).getVariants(); + for (const auto & variant: variants) + { + if (variant->getName() != ColumnDynamic::getSharedVariantTypeName()) + encodeDataType(variant, *stream); + } + } else - writeStringBinary(dynamic_state->variant_type->getName(), *stream); + { + for (const auto & name : dynamic_state->variant_names) + { + if (name != ColumnDynamic::getSharedVariantTypeName()) + writeStringBinary(name, *stream); + } + } /// Write statistics in prefix if needed. if (settings.dynamic_write_statistics == SerializeBinaryBulkSettings::DynamicStatisticsMode::PREFIX) { const auto & statistics = column_dynamic.getStatistics(); + /// First, write statistics for usual variants. for (size_t i = 0; i != variant_info.variant_names.size(); ++i) { size_t size = 0; @@ -129,13 +156,55 @@ void SerializationDynamic::serializeBinaryBulkStatePrefix( /// - statistics read from the data part during deserialization of Dynamic column (Statistics::Source::READ). /// We can rely only on statistics calculated during the merge, because column with statistics that was read /// during deserialization from some data part could be filtered/limited/transformed/etc and so the statistics can be outdated. - if (!statistics.data.empty() && statistics.source == ColumnDynamic::Statistics::Source::MERGE) - size = statistics.data.at(variant_info.variant_names[i]); + if (statistics && statistics->source == ColumnDynamic::Statistics::Source::MERGE) + size = statistics->variants_statistics.at(variant_info.variant_names[i]); /// Otherwise we can use only variant sizes from current column. else size = variant_column.getVariantByGlobalDiscriminator(i).size(); writeVarUInt(size, *stream); } + + /// Second, write statistics for variants in shared variant. + /// Check if we have statistics calculated during merge of some data parts (Statistics::Source::MERGE). + if (statistics && statistics->source == ColumnDynamic::Statistics::Source::MERGE) + { + writeVarUInt(statistics->shared_variants_statistics.size(), *stream); + for (const auto & [variant_name, size] : statistics->shared_variants_statistics) + { + writeStringBinary(variant_name, *stream); + writeVarUInt(size, *stream); + } + } + /// If we don't have statistics for shared variants from merge, calculate it from the column. + else + { + std::unordered_map shared_variants_statistics; + const auto & shared_variant = column_dynamic.getSharedVariant(); + for (size_t i = 0; i != shared_variant.size(); ++i) + { + auto value = shared_variant.getDataAt(i); + ReadBufferFromMemory buf(value.data, value.size); + auto type = decodeDataType(buf); + auto type_name = type->getName(); + if (auto it = shared_variants_statistics.find(type_name); it != shared_variants_statistics.end()) + ++it->second; + else if (shared_variants_statistics.size() < ColumnDynamic::Statistics::MAX_SHARED_VARIANT_STATISTICS_SIZE) + shared_variants_statistics.emplace(type_name, 1); + } + + writeVarUInt(shared_variants_statistics.size(), *stream); + for (const auto & [variant_name, size] : shared_variants_statistics) + { + writeStringBinary(variant_name, *stream); + writeVarUInt(size, *stream); + } + } + } + /// Otherwise statistics will be written in the suffix, in this case we will recalculate + /// statistics during serialization to make it more precise. + else + { + dynamic_state->recalculate_statistics = true; } dynamic_state->variant_serialization = dynamic_state->variant_type->getDefaultSerialization(); @@ -182,33 +251,58 @@ ISerialization::DeserializeBinaryBulkStatePtr SerializationDynamic::deserializeD UInt64 structure_version; readBinaryLittleEndian(structure_version, *structure_stream); auto structure_state = std::make_shared(structure_version); - /// Read internal Variant type name. + /// Read max_dynamic_types parameter. + readBinaryLittleEndian(structure_state->max_dynamic_types, *structure_stream); + /// Read information about variants. + DataTypes variants; + size_t num_variants; + readBinaryLittleEndian(num_variants, *structure_stream); + variants.reserve(num_variants + 1); /// +1 for shared variant. if (settings.data_types_binary_encoding) { - structure_state->variant_type = decodeDataType(*structure_stream); + for (size_t i = 0; i != num_variants; ++i) + variants.push_back(decodeDataType(*structure_stream)); } else { String data_type_name; - readStringBinary(data_type_name, *structure_stream); - structure_state->variant_type = DataTypeFactory::instance().get(data_type_name); + for (size_t i = 0; i != num_variants; ++i) + { + readStringBinary(data_type_name, *structure_stream); + variants.push_back(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()); + /// Add shared variant, Dynamic column should always have it. + variants.push_back(ColumnDynamic::getSharedVariantDataType()); + auto variant_type = std::make_shared(variants); /// Read statistics. if (settings.dynamic_read_statistics) { - const auto & variants = variant_type->getVariants(); + ColumnDynamic::Statistics statistics(ColumnDynamic::Statistics::Source::READ); + /// First, read statistics for usual variants. size_t variant_size; - for (const auto & variant : variants) + for (const auto & variant : variant_type->getVariants()) { readVarUInt(variant_size, *structure_stream); - structure_state->statistics.data[variant->getName()] = variant_size; + statistics.variants_statistics[variant->getName()] = variant_size; } + + /// Second, read statistics for shared variants. + size_t statistics_size; + readVarUInt(statistics_size, *structure_stream); + String variant_name; + for (size_t i = 0; i != statistics_size; ++i) + { + readStringBinary(variant_name, *structure_stream); + readVarUInt(variant_size, *structure_stream); + statistics.shared_variants_statistics[variant_name] = variant_size; + } + + structure_state->statistics = std::make_shared(std::move(statistics)); } + structure_state->variant_type = std::move(variant_type); state = structure_state; addToSubstreamsDeserializeStatesCache(cache, settings.path, state); } @@ -231,8 +325,16 @@ void SerializationDynamic::serializeBinaryBulkStateSuffix( /// Write statistics in suffix if needed. if (settings.dynamic_write_statistics == SerializeBinaryBulkSettings::DynamicStatisticsMode::SUFFIX) { + /// First, write statistics for usual variants. for (const auto & variant_name : dynamic_state->variant_names) - writeVarUInt(dynamic_state->statistics.data[variant_name], *stream); + writeVarUInt(dynamic_state->statistics.variants_statistics[variant_name], *stream); + /// Second, write statistics for shared variants. + writeVarUInt(dynamic_state->statistics.shared_variants_statistics.size(), *stream); + for (const auto & [variant_name, size] : dynamic_state->statistics.shared_variants_statistics) + { + writeStringBinary(variant_name, *stream); + writeVarUInt(size, *stream); + } } settings.path.push_back(Substream::DynamicData); @@ -255,9 +357,42 @@ void SerializationDynamic::serializeBinaryBulkWithMultipleStreams( if (!variant_info.variant_type->equals(*dynamic_state->variant_type)) throw Exception(ErrorCodes::LOGICAL_ERROR, "Mismatch of internal columns of Dynamic. Expected: {}, Got: {}", dynamic_state->variant_type->getName(), variant_info.variant_type->getName()); + if (column_dynamic.getMaxDynamicTypes() != dynamic_state->max_dynamic_types) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Mismatch of max_dynamic_types parameter of Dynamic. Expected: {}, Got: {}", dynamic_state->max_dynamic_types, column_dynamic.getMaxDynamicTypes()); + settings.path.push_back(Substream::DynamicData); - assert_cast(*dynamic_state->variant_serialization) - .serializeBinaryBulkWithMultipleStreamsAndUpdateVariantStatistics(*variant_column, offset, limit, settings, dynamic_state->variant_state, dynamic_state->statistics.data); + if (dynamic_state->recalculate_statistics) + { + assert_cast(*dynamic_state->variant_serialization) + .serializeBinaryBulkWithMultipleStreamsAndUpdateVariantStatistics(*variant_column, offset, limit, settings, dynamic_state->variant_state, dynamic_state->statistics.variants_statistics); + /// Calculate statistics for shared variants. + const auto & shared_variant = column_dynamic.getSharedVariant(); + if (!shared_variant.empty()) + { + const auto & local_discriminators = variant_column->getLocalDiscriminators(); + const auto & offsets = variant_column->getOffsets(); + const auto shared_variant_discr = variant_column->localDiscriminatorByGlobal(column_dynamic.getSharedVariantDiscriminator()); + size_t end = limit == 0 || offset + limit > local_discriminators.size() ? local_discriminators.size() : offset + limit; + for (size_t i = offset; i != end; ++i) + { + if (local_discriminators[i] == shared_variant_discr) + { + auto value = shared_variant.getDataAt(offsets[i]); + ReadBufferFromMemory buf(value.data, value.size); + auto type = decodeDataType(buf); + auto type_name = type->getName(); + if (auto it = dynamic_state->statistics.shared_variants_statistics.find(type_name); it != dynamic_state->statistics.shared_variants_statistics.end()) + ++it->second; + else if (dynamic_state->statistics.shared_variants_statistics.size() < ColumnDynamic::Statistics::MAX_SHARED_VARIANT_STATISTICS_SIZE) + dynamic_state->statistics.shared_variants_statistics.emplace(type_name, 1); + } + } + } + } + else + { + assert_cast(*dynamic_state->variant_serialization).serializeBinaryBulkWithMultipleStreams(*variant_column, offset, limit, settings, dynamic_state->variant_state); + } settings.path.pop_back(); } @@ -272,13 +407,17 @@ void SerializationDynamic::deserializeBinaryBulkWithMultipleStreams( return; auto mutable_column = column->assumeMutable(); + auto & column_dynamic = assert_cast(*mutable_column); auto * dynamic_state = checkAndGetState(state); auto * structure_state = checkAndGetState(dynamic_state->structure_state); if (mutable_column->empty()) - mutable_column = ColumnDynamic::create(structure_state->variant_type->createColumn(), structure_state->variant_type, max_dynamic_types, structure_state->statistics); + { + column_dynamic.setMaxDynamicPaths(structure_state->max_dynamic_types); + column_dynamic.setVariantType(structure_state->variant_type); + column_dynamic.setStatistics(structure_state->statistics); + } - auto & column_dynamic = assert_cast(*mutable_column); const auto & variant_info = column_dynamic.getVariantInfo(); if (!variant_info.variant_type->equals(*structure_state->variant_type)) throw Exception(ErrorCodes::LOGICAL_ERROR, "Mismatch of internal columns of Dynamic. Expected: {}, Got: {}", structure_state->variant_type->getName(), variant_info.variant_type->getName()); @@ -329,24 +468,42 @@ void SerializationDynamic::serializeBinary(const IColumn & column, size_t row_nu encodeDataType(std::make_shared(), ostr); return; } + /// Check if this value is in shared variant. In this case it's already + /// in desired binary format. + else if (global_discr == dynamic_column.getSharedVariantDiscriminator()) + { + auto value = dynamic_column.getSharedVariant().getDataAt(variant_column.offsetAt(row_num)); + ostr.write(value.data, value.size); + 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]; encodeDataType(variant_type, ostr); - variant_type->getDefaultSerialization()->serializeBinary(variant_column.getVariantByGlobalDiscriminator(global_discr), variant_column.offsetAt(row_num), ostr, settings); + dynamic_column.getVariantSerialization(variant_type, variant_type_name)->serializeBinary(variant_column.getVariantByGlobalDiscriminator(global_discr), variant_column.offsetAt(row_num), ostr, settings); } -template -static void deserializeVariant( +template +static ReturnType deserializeVariant( ColumnVariant & variant_column, - const DataTypePtr & variant_type, + const SerializationPtr & variant_serialization, ColumnVariant::Discriminator global_discr, ReadBuffer & istr, DeserializeFunc deserialize) { auto & variant = variant_column.getVariantByGlobalDiscriminator(global_discr); - deserialize(*variant_type->getDefaultSerialization(), variant, istr); + if constexpr (std::is_same_v) + { + if (!deserialize(*variant_serialization, variant, istr)) + return ReturnType(false); + } + else + { + deserialize(*variant_serialization, variant, istr); + } variant_column.getLocalDiscriminators().push_back(variant_column.localDiscriminatorByGlobal(global_discr)); variant_column.getOffsets().push_back(variant.size() - 1); + return ReturnType(true); } void SerializationDynamic::deserializeBinary(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const @@ -360,11 +517,12 @@ void SerializationDynamic::deserializeBinary(IColumn & column, ReadBuffer & istr } auto variant_type_name = variant_type->getName(); + const auto & variant_serialization = dynamic_column.getVariantSerialization(variant_type, variant_type_name); 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()) { - deserializeVariant(dynamic_column.getVariantColumn(), variant_type, it->second, istr, [&settings](const ISerialization & serialization, IColumn & variant, ReadBuffer & buf){ serialization.deserializeBinary(variant, buf, settings); }); + deserializeVariant(dynamic_column.getVariantColumn(), variant_serialization, it->second, istr, [&settings](const ISerialization & serialization, IColumn & variant, ReadBuffer & buf){ serialization.deserializeBinary(variant, buf, settings); }); return; } @@ -372,25 +530,15 @@ void SerializationDynamic::deserializeBinary(IColumn & column, ReadBuffer & istr if (dynamic_column.addNewVariant(variant_type)) { auto discr = variant_info.variant_name_to_discriminator.at(variant_type_name); - deserializeVariant(dynamic_column.getVariantColumn(), variant_type, discr, istr, [&settings](const ISerialization & serialization, IColumn & variant, ReadBuffer & buf){ serialization.deserializeBinary(variant, buf, settings); }); + deserializeVariant(dynamic_column.getVariantColumn(), variant_serialization, discr, istr, [&settings](const ISerialization & serialization, IColumn & variant, ReadBuffer & buf){ serialization.deserializeBinary(variant, buf, settings); }); return; } /// We reached maximum number of variants and couldn't add new variant. - /// This case should be really rare in real use cases. - /// We should always be able to add String variant and insert value as String. - dynamic_column.addStringVariant(); + /// In this case we insert this value into shared variant in binary form. auto tmp_variant_column = variant_type->createColumn(); - variant_type->getDefaultSerialization()->deserializeBinary(*tmp_variant_column, istr, settings); - auto string_column = castColumn(ColumnWithTypeAndName(tmp_variant_column->getPtr(), variant_type, ""), std::make_shared()); - auto & variant_column = dynamic_column.getVariantColumn(); - variant_column.insertIntoVariantFrom(variant_info.variant_name_to_discriminator.at("String"), *string_column, 0); -} - -void SerializationDynamic::serializeTextCSV(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings & settings) const -{ - const auto & dynamic_column = assert_cast(column); - dynamic_column.getVariantInfo().variant_type->getDefaultSerialization()->serializeTextCSV(dynamic_column.getVariantColumn(), row_num, ostr, settings); + variant_serialization->deserializeBinary(*tmp_variant_column, istr, settings); + dynamic_column.insertValueIntoSharedVariant(*tmp_variant_column, variant_type, variant_type_name, 0); } template @@ -406,6 +554,7 @@ static void deserializeTextImpl( auto & dynamic_column = assert_cast(column); auto & variant_column = dynamic_column.getVariantColumn(); const auto & variant_info = dynamic_column.getVariantInfo(); + const auto & variant_types = assert_cast(*variant_info.variant_type).getVariants(); String field = read_field(istr); auto field_buf = std::make_unique(field); JSONInferenceInfo json_info; @@ -413,27 +562,81 @@ static void deserializeTextImpl( if (escaping_rule == FormatSettings::EscapingRule::JSON) transformFinalInferredJSONTypeIfNeeded(variant_type, settings, &json_info); - if (checkIfTypeIsComplete(variant_type) && dynamic_column.addNewVariant(variant_type)) + /// If inferred type is not complete, we cannot add it as a new variant. + /// Let's try to deserialize this field into existing variants. + /// If failed, insert this value as String. + if (!checkIfTypeIsComplete(variant_type)) + { + size_t shared_variant_discr = dynamic_column.getSharedVariantDiscriminator(); + for (size_t i = 0; i != variant_types.size(); ++i) + { + field_buf = std::make_unique(field); + if (i != shared_variant_discr + && deserializeVariant( + variant_column, + dynamic_column.getVariantSerialization(variant_types[i], variant_info.variant_names[i]), + i, + *field_buf, + try_deserialize_variant)) + return; + } + + variant_type = std::make_shared(); + /// To be able to deserialize field as String with Quoted escaping rule, it should be quoted. + if (escaping_rule == FormatSettings::EscapingRule::Quoted && (field.size() < 2 || field.front() != '\'' || field.back() != '\'')) + field = "'" + field + "'"; + } + else if (dynamic_column.addNewVariant(variant_type, variant_type->getName())) { auto discr = variant_info.variant_name_to_discriminator.at(variant_type->getName()); - deserializeVariant(dynamic_column.getVariantColumn(), variant_type, discr, *field_buf, deserialize_variant); + deserializeVariant(dynamic_column.getVariantColumn(), dynamic_column.getVariantSerialization(variant_type), discr, *field_buf, deserialize_variant); return; } - /// We couldn't infer type or add new variant. Try to insert field into current variants. + /// We couldn't infer type or add new variant. Insert it into shared variant. + auto tmp_variant_column = variant_type->createColumn(); field_buf = std::make_unique(field); - if (try_deserialize_variant(*variant_info.variant_type->getDefaultSerialization(), variant_column, *field_buf)) - return; + auto variant_type_name = variant_type->getName(); + deserialize_variant(*dynamic_column.getVariantSerialization(variant_type, variant_type_name), *tmp_variant_column, *field_buf); + dynamic_column.insertValueIntoSharedVariant(*tmp_variant_column, variant_type, variant_type_name, 0); +} - /// We couldn't insert field into any existing variant, add String variant and read value as String. - dynamic_column.addStringVariant(); +template +static void serializeTextImpl( + const IColumn & column, + size_t row_num, + WriteBuffer & ostr, + const FormatSettings & settings, + NestedSerialize nested_serialize) +{ + const auto & dynamic_column = assert_cast(column); + const auto & variant_column = dynamic_column.getVariantColumn(); + /// Check if this row has value in shared variant. In this case we should first deserialize it from binary format. + if (variant_column.globalDiscriminatorAt(row_num) == dynamic_column.getSharedVariantDiscriminator()) + { + auto value = dynamic_column.getSharedVariant().getDataAt(variant_column.offsetAt(row_num)); + ReadBufferFromMemory buf(value.data, value.size); + auto variant_type = decodeDataType(buf); + auto tmp_variant_column = variant_type->createColumn(); + auto variant_serialization = dynamic_column.getVariantSerialization(variant_type); + variant_serialization->deserializeBinary(*tmp_variant_column, buf, settings); + nested_serialize(*variant_serialization, *tmp_variant_column, 0, ostr); + } + /// Otherwise just use serialization for Variant. + else + { + nested_serialize(*dynamic_column.getVariantInfo().variant_type->getDefaultSerialization(), variant_column, row_num, ostr); + } +} - if (escaping_rule == FormatSettings::EscapingRule::Quoted && (field.size() < 2 || field.front() != '\'' || field.back() != '\'')) - field = "'" + field + "'"; +void SerializationDynamic::serializeTextCSV(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings & settings) const +{ + auto nested_serialize = [&settings](const ISerialization & serialization, const IColumn & col, size_t row, WriteBuffer & buf) + { + serialization.serializeTextCSV(col, row, buf, settings); + }; - field_buf = std::make_unique(field); - auto string_discr = variant_info.variant_name_to_discriminator.at("String"); - deserializeVariant(dynamic_column.getVariantColumn(), std::make_shared(), string_discr, *field_buf, deserialize_variant); + serializeTextImpl(column, row_num, ostr, settings, nested_serialize); } void SerializationDynamic::deserializeTextCSV(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const @@ -466,8 +669,12 @@ bool SerializationDynamic::tryDeserializeTextCSV(DB::IColumn & column, DB::ReadB void SerializationDynamic::serializeTextEscaped(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings & settings) const { - const auto & dynamic_column = assert_cast(column); - dynamic_column.getVariantInfo().variant_type->getDefaultSerialization()->serializeTextEscaped(dynamic_column.getVariantColumn(), row_num, ostr, settings); + auto nested_serialize = [&settings](const ISerialization & serialization, const IColumn & col, size_t row, WriteBuffer & buf) + { + serialization.serializeTextEscaped(col, row, buf, settings); + }; + + serializeTextImpl(column, row_num, ostr, settings, nested_serialize); } void SerializationDynamic::deserializeTextEscaped(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const @@ -500,8 +707,12 @@ bool SerializationDynamic::tryDeserializeTextEscaped(DB::IColumn & column, DB::R void SerializationDynamic::serializeTextQuoted(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings & settings) const { - const auto & dynamic_column = assert_cast(column); - dynamic_column.getVariantInfo().variant_type->getDefaultSerialization()->serializeTextQuoted(dynamic_column.getVariantColumn(), row_num, ostr, settings); + auto nested_serialize = [&settings](const ISerialization & serialization, const IColumn & col, size_t row, WriteBuffer & buf) + { + serialization.serializeTextQuoted(col, row, buf, settings); + }; + + serializeTextImpl(column, row_num, ostr, settings, nested_serialize); } void SerializationDynamic::deserializeTextQuoted(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const @@ -534,8 +745,12 @@ bool SerializationDynamic::tryDeserializeTextQuoted(DB::IColumn & column, DB::Re void SerializationDynamic::serializeTextJSON(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings & settings) const { - const auto & dynamic_column = assert_cast(column); - dynamic_column.getVariantInfo().variant_type->getDefaultSerialization()->serializeTextJSON(dynamic_column.getVariantColumn(), row_num, ostr, settings); + auto nested_serialize = [&settings](const ISerialization & serialization, const IColumn & col, size_t row, WriteBuffer & buf) + { + serialization.serializeTextJSON(col, row, buf, settings); + }; + + serializeTextImpl(column, row_num, ostr, settings, nested_serialize); } void SerializationDynamic::deserializeTextJSON(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const @@ -568,8 +783,12 @@ bool SerializationDynamic::tryDeserializeTextJSON(DB::IColumn & column, DB::Read void SerializationDynamic::serializeTextRaw(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings & settings) const { - const auto & dynamic_column = assert_cast(column); - dynamic_column.getVariantInfo().variant_type->getDefaultSerialization()->serializeTextRaw(dynamic_column.getVariantColumn(), row_num, ostr, settings); + auto nested_serialize = [&settings](const ISerialization & serialization, const IColumn & col, size_t row, WriteBuffer & buf) + { + serialization.serializeTextRaw(col, row, buf, settings); + }; + + serializeTextImpl(column, row_num, ostr, settings, nested_serialize); } void SerializationDynamic::deserializeTextRaw(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const @@ -602,8 +821,12 @@ bool SerializationDynamic::tryDeserializeTextRaw(DB::IColumn & column, DB::ReadB void SerializationDynamic::serializeText(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings & settings) const { - const auto & dynamic_column = assert_cast(column); - dynamic_column.getVariantInfo().variant_type->getDefaultSerialization()->serializeText(dynamic_column.getVariantColumn(), row_num, ostr, settings); + auto nested_serialize = [&settings](const ISerialization & serialization, const IColumn & col, size_t row, WriteBuffer & buf) + { + serialization.serializeText(col, row, buf, settings); + }; + + serializeTextImpl(column, row_num, ostr, settings, nested_serialize); } void SerializationDynamic::deserializeWholeText(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const @@ -636,8 +859,12 @@ bool SerializationDynamic::tryDeserializeWholeText(DB::IColumn & column, DB::Rea void SerializationDynamic::serializeTextXML(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings & settings) const { - const auto & dynamic_column = assert_cast(column); - dynamic_column.getVariantInfo().variant_type->getDefaultSerialization()->serializeTextXML(dynamic_column.getVariantColumn(), row_num, ostr, settings); + auto nested_serialize = [&settings](const ISerialization & serialization, const IColumn & col, size_t row, WriteBuffer & buf) + { + serialization.serializeTextXML(col, row, buf, settings); + }; + + serializeTextImpl(column, row_num, ostr, settings, nested_serialize); } } diff --git a/src/DataTypes/Serializations/SerializationDynamic.h b/src/DataTypes/Serializations/SerializationDynamic.h index 001a3cf87ce..3dbf311fb6c 100644 --- a/src/DataTypes/Serializations/SerializationDynamic.h +++ b/src/DataTypes/Serializations/SerializationDynamic.h @@ -105,9 +105,13 @@ private: { DynamicStructureSerializationVersion structure_version; DataTypePtr variant_type; - ColumnDynamic::Statistics statistics = {.source = ColumnDynamic::Statistics::Source::READ, .data = {}}; + size_t max_dynamic_types; + ColumnDynamic::StatisticsPtr statistics; - explicit DeserializeBinaryBulkStateDynamicStructure(UInt64 structure_version_) : structure_version(structure_version_) {} + explicit DeserializeBinaryBulkStateDynamicStructure(UInt64 structure_version_) + : structure_version(structure_version_) + { + } }; size_t max_dynamic_types; diff --git a/src/DataTypes/Serializations/SerializationDynamicElement.cpp b/src/DataTypes/Serializations/SerializationDynamicElement.cpp index 211f0ac9377..cffca14bca5 100644 --- a/src/DataTypes/Serializations/SerializationDynamicElement.cpp +++ b/src/DataTypes/Serializations/SerializationDynamicElement.cpp @@ -4,7 +4,10 @@ #include #include #include +#include #include +#include +#include #include namespace DB @@ -21,6 +24,8 @@ struct DeserializeBinaryBulkStateDynamicElement : public ISerialization::Deseria ISerialization::DeserializeBinaryBulkStatePtr structure_state; SerializationPtr variant_serialization; ISerialization::DeserializeBinaryBulkStatePtr variant_element_state; + bool read_from_shared_variant; + ColumnPtr shared_variant; }; void SerializationDynamicElement::enumerateStreams( @@ -73,9 +78,10 @@ void SerializationDynamicElement::deserializeBinaryBulkStatePrefix( auto dynamic_element_state = std::make_shared(); dynamic_element_state->structure_state = std::move(structure_state); - const auto & variant_type = checkAndGetState(dynamic_element_state->structure_state)->variant_type; + const auto & variant_type = assert_cast( + *checkAndGetState(dynamic_element_state->structure_state)->variant_type); /// Check if we actually have required element in the Variant. - if (auto global_discr = assert_cast(*variant_type).tryGetVariantDiscriminator(dynamic_element_name)) + if (auto global_discr = variant_type.tryGetVariantDiscriminator(dynamic_element_name)) { settings.path.push_back(Substream::DynamicData); if (is_null_map_subcolumn) @@ -83,6 +89,21 @@ void SerializationDynamicElement::deserializeBinaryBulkStatePrefix( 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); + dynamic_element_state->read_from_shared_variant = false; + settings.path.pop_back(); + } + /// If we don't have this element in the Variant, we will read shared variant and try to find it there. + else + { + auto shared_variant_global_discr = variant_type.tryGetVariantDiscriminator(ColumnDynamic::getSharedVariantTypeName()); + chassert(shared_variant_global_discr.has_value()); + settings.path.push_back(Substream::DynamicData); + dynamic_element_state->variant_serialization = std::make_shared( + ColumnDynamic::getSharedVariantDataType()->getDefaultSerialization(), + ColumnDynamic::getSharedVariantTypeName(), + *shared_variant_global_discr); + dynamic_element_state->variant_serialization->deserializeBinaryBulkStatePrefix(settings, dynamic_element_state->variant_element_state, cache); + dynamic_element_state->read_from_shared_variant = true; settings.path.pop_back(); } @@ -115,23 +136,103 @@ void SerializationDynamicElement::deserializeBinaryBulkWithMultipleStreams( auto * dynamic_element_state = checkAndGetState(state); - if (dynamic_element_state->variant_serialization) + /// Check if this subcolumn should not be read from shared variant. + /// In this case just read data from the corresponding variant. + if (!dynamic_element_state->read_from_shared_variant) { settings.path.push_back(Substream::DynamicData); - dynamic_element_state->variant_serialization->deserializeBinaryBulkWithMultipleStreams(result_column, limit, settings, dynamic_element_state->variant_element_state, cache); + 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); - } + /// Otherwise, read the shared variant column and extract requested type from it. else { - auto mutable_column = result_column->assumeMutable(); - mutable_column->insertManyDefaults(limit); - result_column = std::move(mutable_column); + settings.path.push_back(Substream::DynamicData); + /// Initialize shared_variant column if needed. + if (result_column->empty()) + dynamic_element_state->shared_variant = makeNullable(ColumnDynamic::getSharedVariantDataType()->createColumn()); + size_t prev_size = result_column->size(); + dynamic_element_state->variant_serialization->deserializeBinaryBulkWithMultipleStreams( + dynamic_element_state->shared_variant, limit, settings, dynamic_element_state->variant_element_state, cache); + settings.path.pop_back(); + + /// If we need to read a subcolumn from variant column, create an empty variant column, fill it and extract subcolumn. + auto variant_type = DataTypeFactory::instance().get(dynamic_element_name); + auto result_type = makeNullableOrLowCardinalityNullableSafe(variant_type); + MutableColumnPtr variant_column = nested_subcolumn.empty() || is_null_map_subcolumn ? result_column->assumeMutable() : result_type->createColumn(); + variant_column->reserve(variant_column->size() + limit); + MutableColumnPtr non_nullable_variant_column = variant_column->assumeMutable(); + NullMap * null_map = nullptr; + bool is_low_cardinality_nullable = isColumnLowCardinalityNullable(*variant_column); + /// Resulting subolumn can be Nullable, but value is serialized in shared variant as non-Nullable. + /// Extract non-nullable column and remember the null map to fill it during deserialization. + if (isColumnNullable(*variant_column)) + { + auto & nullable_variant_column = assert_cast(*variant_column); + non_nullable_variant_column = nullable_variant_column.getNestedColumnPtr()->assumeMutable(); + null_map = &nullable_variant_column.getNullMapData(); + } + else if (is_null_map_subcolumn) + { + null_map = &assert_cast(*variant_column).getData(); + } + + auto variant_serialization = variant_type->getDefaultSerialization(); + + const auto & nullable_shared_variant = assert_cast(*dynamic_element_state->shared_variant); + const auto & shared_null_map = nullable_shared_variant.getNullMapData(); + const auto & shared_variant = assert_cast(nullable_shared_variant.getNestedColumn()); + const FormatSettings format_settings; + for (size_t i = prev_size; i != shared_variant.size(); ++i) + { + if (!shared_null_map[i]) + { + auto value = shared_variant.getDataAt(i); + ReadBufferFromMemory buf(value.data, value.size); + auto type = decodeDataType(buf); + if (type->getName() == dynamic_element_name) + { + /// When requested type is LowCardinality the subcolumn type name will be LowCardinality(Nullable). + /// Value in shared variant is serialized as LowCardinality and we cannot simply deserialize it + /// inside LowCardinality(Nullable) column (it will try to deserialize null bit). In this case we + /// have to create temporary LowCardinality column, deserialize value into it and insert it into + /// resulting LowCardinality(Nullable) (insertion from LowCardinality column to LowCardinality(Nullable) + /// column is allowed). + if (is_low_cardinality_nullable) + { + auto tmp_column = variant_type->createColumn(); + variant_serialization->deserializeBinary(*tmp_column, buf, format_settings); + non_nullable_variant_column->insertFrom(*tmp_column, 0); + } + else if (is_null_map_subcolumn) + { + null_map->push_back(0); + } + else + { + variant_serialization->deserializeBinary(*non_nullable_variant_column, buf, format_settings); + if (null_map) + null_map->push_back(0); + } + } + else + { + variant_column->insertDefault(); + } + } + else + { + variant_column->insertDefault(); + } + } + + /// Extract nested subcolumn if needed. + if (!nested_subcolumn.empty() && !is_null_map_subcolumn) + { + auto subcolumn = result_type->getSubcolumn(nested_subcolumn, variant_column->getPtr()); + result_column->assumeMutable()->insertRangeFrom(*subcolumn, 0, subcolumn->size()); + } } } diff --git a/src/DataTypes/Serializations/SerializationDynamicElement.h b/src/DataTypes/Serializations/SerializationDynamicElement.h index 127d14a55e0..c674cf479ae 100644 --- a/src/DataTypes/Serializations/SerializationDynamicElement.h +++ b/src/DataTypes/Serializations/SerializationDynamicElement.h @@ -13,11 +13,15 @@ private: /// To be able to deserialize Dynamic element as a subcolumn /// we need its type name and global discriminator. String dynamic_element_name; + /// Nested subcolumn of a type dynamic type. For example, for `Tuple(a UInt32)`.a + /// subcolumn dynamic_element_name = 'Tuple(a UInt32)' and nested_subcolumn = 'a'. + /// Needed to extract nested subcolumn from values in shared variant. + String nested_subcolumn; bool is_null_map_subcolumn; public: - 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_) + SerializationDynamicElement(const SerializationPtr & nested_, const String & dynamic_element_name_, const String & nested_subcolumn_, bool is_null_map_subcolumn_ = false) + : SerializationWrapper(nested_), dynamic_element_name(dynamic_element_name_), nested_subcolumn(nested_subcolumn_), is_null_map_subcolumn(is_null_map_subcolumn_) { } diff --git a/src/DataTypes/Serializations/SerializationFixedString.cpp b/src/DataTypes/Serializations/SerializationFixedString.cpp index f919dc16d33..688c71792fa 100644 --- a/src/DataTypes/Serializations/SerializationFixedString.cpp +++ b/src/DataTypes/Serializations/SerializationFixedString.cpp @@ -28,7 +28,7 @@ static constexpr size_t MAX_STRINGS_SIZE = 1ULL << 30; void SerializationFixedString::serializeBinary(const Field & field, WriteBuffer & ostr, const FormatSettings &) const { - const String & s = field.get(); + const String & s = field.safeGet(); ostr.write(s.data(), std::min(s.size(), n)); if (s.size() < n) for (size_t i = s.size(); i < n; ++i) @@ -39,7 +39,7 @@ void SerializationFixedString::serializeBinary(const Field & field, WriteBuffer void SerializationFixedString::deserializeBinary(Field & field, ReadBuffer & istr, const FormatSettings &) const { field = String(); - String & s = field.get(); + String & s = field.safeGet(); s.resize(n); istr.readStrict(s.data(), n); } diff --git a/src/DataTypes/Serializations/SerializationIPv4andIPv6.cpp b/src/DataTypes/Serializations/SerializationIPv4andIPv6.cpp index dfcd24aff58..c1beceb4533 100644 --- a/src/DataTypes/Serializations/SerializationIPv4andIPv6.cpp +++ b/src/DataTypes/Serializations/SerializationIPv4andIPv6.cpp @@ -125,7 +125,7 @@ bool SerializationIP::tryDeserializeTextCSV(DB::IColumn & column, DB::ReadB template void SerializationIP::serializeBinary(const Field & field, WriteBuffer & ostr, const FormatSettings &) const { - IPv x = field.get(); + IPv x = field.safeGet(); if constexpr (std::is_same_v) writeBinary(x, ostr); else diff --git a/src/DataTypes/Serializations/SerializationMap.cpp b/src/DataTypes/Serializations/SerializationMap.cpp index 0bef3c7d79d..c722b3ac7a1 100644 --- a/src/DataTypes/Serializations/SerializationMap.cpp +++ b/src/DataTypes/Serializations/SerializationMap.cpp @@ -40,7 +40,7 @@ static IColumn & extractNestedColumn(IColumn & column) void SerializationMap::serializeBinary(const Field & field, WriteBuffer & ostr, const FormatSettings & settings) const { - const auto & map = field.get(); + const auto & map = field.safeGet(); writeVarUInt(map.size(), ostr); for (const auto & elem : map) { @@ -63,7 +63,7 @@ void SerializationMap::deserializeBinary(Field & field, ReadBuffer & istr, const size, settings.binary.max_binary_string_size); field = Map(); - Map & map = field.get(); + Map & map = field.safeGet(); map.reserve(size); for (size_t i = 0; i < size; ++i) { diff --git a/src/DataTypes/Serializations/SerializationNumber.cpp b/src/DataTypes/Serializations/SerializationNumber.cpp index bdb4dfc6735..bfc13af8ca3 100644 --- a/src/DataTypes/Serializations/SerializationNumber.cpp +++ b/src/DataTypes/Serializations/SerializationNumber.cpp @@ -169,7 +169,7 @@ template void SerializationNumber::serializeBinary(const Field & field, WriteBuffer & ostr, const FormatSettings &) const { /// ColumnVector::ValueType is a narrower type. For example, UInt8, when the Field type is UInt64 - typename ColumnVector::ValueType x = static_cast::ValueType>(field.get()); + typename ColumnVector::ValueType x = static_cast::ValueType>(field.safeGet()); writeBinaryLittleEndian(x, ostr); } diff --git a/src/DataTypes/Serializations/SerializationString.cpp b/src/DataTypes/Serializations/SerializationString.cpp index 9e523d0d745..ac5d4e3e128 100644 --- a/src/DataTypes/Serializations/SerializationString.cpp +++ b/src/DataTypes/Serializations/SerializationString.cpp @@ -32,7 +32,7 @@ namespace ErrorCodes void SerializationString::serializeBinary(const Field & field, WriteBuffer & ostr, const FormatSettings & settings) const { - const String & s = field.get(); + const String & s = field.safeGet(); if (settings.binary.max_binary_string_size && s.size() > settings.binary.max_binary_string_size) throw Exception( ErrorCodes::TOO_LARGE_STRING_SIZE, @@ -59,7 +59,7 @@ void SerializationString::deserializeBinary(Field & field, ReadBuffer & istr, co settings.binary.max_binary_string_size); field = String(); - String & s = field.get(); + String & s = field.safeGet(); s.resize(size); istr.readStrict(s.data(), size); } diff --git a/src/DataTypes/Serializations/SerializationTuple.cpp b/src/DataTypes/Serializations/SerializationTuple.cpp index 7a5227ca752..594a23ab507 100644 --- a/src/DataTypes/Serializations/SerializationTuple.cpp +++ b/src/DataTypes/Serializations/SerializationTuple.cpp @@ -34,7 +34,7 @@ static inline const IColumn & extractElementColumn(const IColumn & column, size_ void SerializationTuple::serializeBinary(const Field & field, WriteBuffer & ostr, const FormatSettings & settings) const { - const auto & tuple = field.get(); + const auto & tuple = field.safeGet(); for (size_t element_index = 0; element_index < elems.size(); ++element_index) { const auto & serialization = elems[element_index]; @@ -47,7 +47,7 @@ void SerializationTuple::deserializeBinary(Field & field, ReadBuffer & istr, con const size_t size = elems.size(); field = Tuple(); - Tuple & tuple = field.get(); + Tuple & tuple = field.safeGet(); tuple.reserve(size); for (size_t i = 0; i < size; ++i) elems[i]->deserializeBinary(tuple.emplace_back(), istr, settings); diff --git a/src/DataTypes/Serializations/SerializationUUID.cpp b/src/DataTypes/Serializations/SerializationUUID.cpp index 58178a896dc..f18466ad8ad 100644 --- a/src/DataTypes/Serializations/SerializationUUID.cpp +++ b/src/DataTypes/Serializations/SerializationUUID.cpp @@ -137,7 +137,7 @@ bool SerializationUUID::tryDeserializeTextCSV(IColumn & column, ReadBuffer & ist void SerializationUUID::serializeBinary(const Field & field, WriteBuffer & ostr, const FormatSettings &) const { - UUID x = field.get(); + UUID x = field.safeGet(); writeBinaryLittleEndian(x, ostr); } diff --git a/src/DataTypes/Serializations/SerializationVariantElement.cpp b/src/DataTypes/Serializations/SerializationVariantElement.cpp index 03b5d9584e0..36dc85f60ee 100644 --- a/src/DataTypes/Serializations/SerializationVariantElement.cpp +++ b/src/DataTypes/Serializations/SerializationVariantElement.cpp @@ -305,8 +305,10 @@ SerializationVariantElement::VariantSubcolumnCreator::VariantSubcolumnCreator( const String & variant_element_name_, ColumnVariant::Discriminator global_variant_discriminator_, ColumnVariant::Discriminator local_variant_discriminator_, - bool make_nullable_) + bool make_nullable_, + const ColumnPtr & null_map_) : local_discriminators(local_discriminators_) + , null_map(null_map_) , variant_element_name(variant_element_name_) , global_variant_discriminator(global_variant_discriminator_) , local_variant_discriminator(local_variant_discriminator_) @@ -314,12 +316,13 @@ SerializationVariantElement::VariantSubcolumnCreator::VariantSubcolumnCreator( { } -DataTypePtr SerializationVariantElement::VariantSubcolumnCreator::create(const DB::DataTypePtr & prev) const + +DataTypePtr SerializationVariantElement::VariantSubcolumnCreator::create(const DataTypePtr & prev) const { return make_nullable ? makeNullableOrLowCardinalityNullableSafe(prev) : prev; } -SerializationPtr SerializationVariantElement::VariantSubcolumnCreator::create(const DB::SerializationPtr & prev) const +SerializationPtr SerializationVariantElement::VariantSubcolumnCreator::create(const SerializationPtr & prev) const { return std::make_shared(prev, variant_element_name, global_variant_discriminator); } @@ -339,12 +342,16 @@ ColumnPtr SerializationVariantElement::VariantSubcolumnCreator::create(const DB: return res; } - /// In general case we should iterate through discriminators and create null-map for our variant. - NullMap null_map; - null_map.reserve(local_discriminators->size()); - const auto & local_discriminators_data = assert_cast(*local_discriminators).getData(); - for (auto local_discr : local_discriminators_data) - null_map.push_back(local_discr != local_variant_discriminator); + /// In general case we should iterate through discriminators and create null-map for our variant if we don't already have it. + std::optional null_map_from_discriminators; + if (!null_map) + { + null_map_from_discriminators = NullMap(); + null_map_from_discriminators->reserve(local_discriminators->size()); + const auto & local_discriminators_data = assert_cast(*local_discriminators).getData(); + for (auto local_discr : local_discriminators_data) + null_map_from_discriminators->push_back(local_discr != local_variant_discriminator); + } /// Now we can create new column from null-map and variant column using IColumn::expand. auto res_column = IColumn::mutate(prev); @@ -356,13 +363,21 @@ ColumnPtr SerializationVariantElement::VariantSubcolumnCreator::create(const DB: if (make_nullable && prev->lowCardinality()) res_column = assert_cast(*res_column).cloneNullable(); - res_column->expand(null_map, /*inverted = */ true); + if (null_map_from_discriminators) + res_column->expand(*null_map_from_discriminators, /*inverted = */ true); + else + res_column->expand(assert_cast(*null_map).getData(), /*inverted = */ true); if (make_nullable && prev->canBeInsideNullable()) { - auto null_map_col = ColumnUInt8::create(); - null_map_col->getData() = std::move(null_map); - return ColumnNullable::create(std::move(res_column), std::move(null_map_col)); + if (null_map_from_discriminators) + { + auto null_map_col = ColumnUInt8::create(); + null_map_col->getData() = std::move(*null_map_from_discriminators); + return ColumnNullable::create(std::move(res_column), std::move(null_map_col)); + } + + return ColumnNullable::create(std::move(res_column), null_map->assumeMutable()); } return res_column; diff --git a/src/DataTypes/Serializations/SerializationVariantElement.h b/src/DataTypes/Serializations/SerializationVariantElement.h index 69101aea0f5..64f86eb2190 100644 --- a/src/DataTypes/Serializations/SerializationVariantElement.h +++ b/src/DataTypes/Serializations/SerializationVariantElement.h @@ -63,18 +63,22 @@ public: struct VariantSubcolumnCreator : public ISubcolumnCreator { + private: const ColumnPtr local_discriminators; + const ColumnPtr null_map; /// optional const String variant_element_name; const ColumnVariant::Discriminator global_variant_discriminator; const ColumnVariant::Discriminator local_variant_discriminator; bool make_nullable; + public: VariantSubcolumnCreator( const ColumnPtr & local_discriminators_, const String & variant_element_name_, ColumnVariant::Discriminator global_variant_discriminator_, ColumnVariant::Discriminator local_variant_discriminator_, - bool make_nullable_); + bool make_nullable_, + const ColumnPtr & null_map_ = nullptr); DataTypePtr create(const DataTypePtr & prev) const override; ColumnPtr create(const ColumnPtr & prev) const override; diff --git a/src/DataTypes/fuzzers/CMakeLists.txt b/src/DataTypes/fuzzers/CMakeLists.txt index bc640358673..e54ef0a860c 100644 --- a/src/DataTypes/fuzzers/CMakeLists.txt +++ b/src/DataTypes/fuzzers/CMakeLists.txt @@ -1,2 +1,2 @@ clickhouse_add_executable(data_type_deserialization_fuzzer data_type_deserialization_fuzzer.cpp ${SRCS}) -target_link_libraries(data_type_deserialization_fuzzer PRIVATE dbms clickhouse_aggregate_functions clickhouse_functions) +target_link_libraries(data_type_deserialization_fuzzer PRIVATE clickhouse_functions clickhouse_aggregate_functions) diff --git a/src/DataTypes/registerDataTypeDateTime.cpp b/src/DataTypes/registerDataTypeDateTime.cpp index 84a52d4affb..9a632bd381b 100644 --- a/src/DataTypes/registerDataTypeDateTime.cpp +++ b/src/DataTypes/registerDataTypeDateTime.cpp @@ -55,7 +55,7 @@ getArgument(const ASTPtr & arguments, size_t argument_index, const char * argume } } - return argument->value.get(); + return argument->value.safeGet(); } static DataTypePtr create(const ASTPtr & arguments) diff --git a/src/Databases/DDLLoadingDependencyVisitor.cpp b/src/Databases/DDLLoadingDependencyVisitor.cpp index 67bce915168..b91aa84ecd3 100644 --- a/src/Databases/DDLLoadingDependencyVisitor.cpp +++ b/src/Databases/DDLLoadingDependencyVisitor.cpp @@ -183,7 +183,7 @@ void DDLLoadingDependencyVisitor::extractTableNameFromArgument(const ASTFunction if (name->value.getType() != Field::Types::String) return; - auto maybe_qualified_name = QualifiedTableName::tryParseFromString(name->value.get()); + auto maybe_qualified_name = QualifiedTableName::tryParseFromString(name->value.safeGet()); if (!maybe_qualified_name) return; @@ -194,7 +194,7 @@ void DDLLoadingDependencyVisitor::extractTableNameFromArgument(const ASTFunction if (literal->value.getType() != Field::Types::String) return; - auto maybe_qualified_name = QualifiedTableName::tryParseFromString(literal->value.get()); + auto maybe_qualified_name = QualifiedTableName::tryParseFromString(literal->value.safeGet()); /// Just return if name if invalid if (!maybe_qualified_name) return; diff --git a/src/Databases/DDLRenamingVisitor.cpp b/src/Databases/DDLRenamingVisitor.cpp index 38e100e2470..7556223b30e 100644 --- a/src/Databases/DDLRenamingVisitor.cpp +++ b/src/Databases/DDLRenamingVisitor.cpp @@ -180,7 +180,7 @@ namespace if (database_name_field && table_name_field) { - QualifiedTableName qualified_name{database_name_field->get(), table_name_field->get()}; + QualifiedTableName qualified_name{database_name_field->safeGet(), table_name_field->safeGet()}; if (!qualified_name.database.empty() && !qualified_name.table.empty()) { auto new_qualified_name = data.renaming_map.getNewTableName(qualified_name); @@ -207,7 +207,7 @@ namespace if (literal->value.getType() != Field::Types::String) return; - auto maybe_qualified_name = QualifiedTableName::tryParseFromString(literal->value.get()); + auto maybe_qualified_name = QualifiedTableName::tryParseFromString(literal->value.safeGet()); /// Just return if name if invalid if (!maybe_qualified_name || maybe_qualified_name->database.empty() || maybe_qualified_name->table.empty()) return; @@ -247,7 +247,7 @@ namespace if (!literal || (literal->value.getType() != Field::Types::String)) return; - auto database_name = literal->value.get(); + auto database_name = literal->value.safeGet(); if (database_name.empty()) return; diff --git a/src/Databases/DatabaseReplicated.cpp b/src/Databases/DatabaseReplicated.cpp index f127ccbc224..85217ba5c04 100644 --- a/src/Databases/DatabaseReplicated.cpp +++ b/src/Databases/DatabaseReplicated.cpp @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -338,9 +339,12 @@ ClusterPtr DatabaseReplicated::getClusterImpl(bool all_groups) const return std::make_shared(getContext()->getSettingsRef(), shards, params); } -std::vector DatabaseReplicated::tryGetAreReplicasActive(const ClusterPtr & cluster_) const +ReplicasInfo DatabaseReplicated::tryGetReplicasInfo(const ClusterPtr & cluster_) const { Strings paths; + + paths.emplace_back(fs::path(zookeeper_path) / "max_log_ptr"); + const auto & addresses_with_failover = cluster_->getShardsAddresses(); const auto & shards_info = cluster_->getShardsInfo(); for (size_t shard_index = 0; shard_index < shards_info.size(); ++shard_index) @@ -349,22 +353,50 @@ std::vector DatabaseReplicated::tryGetAreReplicasActive(const ClusterPtr { String full_name = getFullReplicaName(replica.database_shard_name, replica.database_replica_name); paths.emplace_back(fs::path(zookeeper_path) / "replicas" / full_name / "active"); + paths.emplace_back(fs::path(zookeeper_path) / "replicas" / full_name / "log_ptr"); } } try { auto current_zookeeper = getZooKeeper(); - auto res = current_zookeeper->exists(paths); + auto zk_res = current_zookeeper->tryGet(paths); - std::vector statuses; - statuses.resize(paths.size()); + auto max_log_ptr_zk = zk_res[0]; + if (max_log_ptr_zk.error != Coordination::Error::ZOK) + throw Coordination::Exception(max_log_ptr_zk.error); - for (size_t i = 0; i < res.size(); ++i) - if (res[i].error == Coordination::Error::ZOK) - statuses[i] = 1; + UInt32 max_log_ptr = parse(max_log_ptr_zk.data); - return statuses; + ReplicasInfo replicas_info; + replicas_info.resize((zk_res.size() - 1) / 2); + + size_t global_replica_index = 0; + for (size_t shard_index = 0; shard_index < shards_info.size(); ++shard_index) + { + for (const auto & replica : addresses_with_failover[shard_index]) + { + auto replica_active = zk_res[2 * global_replica_index + 1]; + auto replica_log_ptr = zk_res[2 * global_replica_index + 2]; + + UInt64 recovery_time = 0; + { + std::lock_guard lock(ddl_worker_mutex); + if (replica.is_local && ddl_worker) + recovery_time = ddl_worker->getCurrentInitializationDurationMs(); + } + + replicas_info[global_replica_index] = ReplicaInfo{ + .is_active = replica_active.error == Coordination::Error::ZOK, + .replication_lag = replica_log_ptr.error != Coordination::Error::ZNONODE ? std::optional(max_log_ptr - parse(replica_log_ptr.data)) : std::nullopt, + .recovery_time = recovery_time, + }; + + ++global_replica_index; + } + } + + return replicas_info; } catch (...) { @@ -373,7 +405,6 @@ std::vector DatabaseReplicated::tryGetAreReplicasActive(const ClusterPtr } } - void DatabaseReplicated::fillClusterAuthInfo(String collection_name, const Poco::Util::AbstractConfiguration & config_ref) { const auto & config_prefix = fmt::format("named_collections.{}", collection_name); @@ -802,8 +833,8 @@ void DatabaseReplicated::checkTableEngine(const ASTCreateQuery & query, ASTStora 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(); + String maybe_path = arg1->value.safeGet(); + String maybe_replica = arg2->value.safeGet(); /// Looks like it's ReplicatedMergeTree with explicit zookeeper_path and replica_name arguments. /// Let's ensure that some macros are used. @@ -1122,8 +1153,7 @@ void DatabaseReplicated::recoverLostReplica(const ZooKeeperPtr & current_zookeep query_context->setSetting("allow_experimental_object_type", 1); query_context->setSetting("allow_experimental_variant_type", 1); query_context->setSetting("allow_experimental_dynamic_type", 1); - query_context->setSetting("allow_experimental_annoy_index", 1); - query_context->setSetting("allow_experimental_usearch_index", 1); + query_context->setSetting("allow_experimental_vector_similarity_index", 1); query_context->setSetting("allow_experimental_bigint_types", 1); query_context->setSetting("allow_experimental_window_functions", 1); query_context->setSetting("allow_experimental_geo_types", 1); @@ -1553,6 +1583,8 @@ void DatabaseReplicated::dropTable(ContextPtr local_context, const String & tabl } auto table = tryGetTable(table_name, getContext()); + if (!table) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Table {} doesn't exist", table_name); if (table->getName() == "MaterializedView" || table->getName() == "WindowView") { /// Avoid recursive locking of metadata_mutex diff --git a/src/Databases/DatabaseReplicated.h b/src/Databases/DatabaseReplicated.h index 27ab262d1f1..db683be8f36 100644 --- a/src/Databases/DatabaseReplicated.h +++ b/src/Databases/DatabaseReplicated.h @@ -1,5 +1,7 @@ #pragma once +#include + #include #include #include @@ -17,6 +19,14 @@ using ZooKeeperPtr = std::shared_ptr; class Cluster; using ClusterPtr = std::shared_ptr; +struct ReplicaInfo +{ + bool is_active; + std::optional replication_lag; + UInt64 recovery_time; +}; +using ReplicasInfo = std::vector; + class DatabaseReplicated : public DatabaseAtomic { public: @@ -84,7 +94,7 @@ public: static void dropReplica(DatabaseReplicated * database, const String & database_zookeeper_path, const String & shard, const String & replica, bool throw_if_noop); - std::vector tryGetAreReplicasActive(const ClusterPtr & cluster_) const; + ReplicasInfo tryGetReplicasInfo(const ClusterPtr & cluster_) const; void renameDatabase(ContextPtr query_context, const String & new_name) override; diff --git a/src/Databases/DatabaseReplicatedWorker.cpp b/src/Databases/DatabaseReplicatedWorker.cpp index 1ef88dc03bc..4e7408aa96e 100644 --- a/src/Databases/DatabaseReplicatedWorker.cpp +++ b/src/Databases/DatabaseReplicatedWorker.cpp @@ -32,6 +32,12 @@ DatabaseReplicatedDDLWorker::DatabaseReplicatedDDLWorker(DatabaseReplicated * db bool DatabaseReplicatedDDLWorker::initializeMainThread() { + { + std::lock_guard lock(initialization_duration_timer_mutex); + initialization_duration_timer.emplace(); + initialization_duration_timer->start(); + } + while (!stop_flag) { try @@ -69,6 +75,10 @@ bool DatabaseReplicatedDDLWorker::initializeMainThread() initializeReplication(); initialized = true; + { + std::lock_guard lock(initialization_duration_timer_mutex); + initialization_duration_timer.reset(); + } return true; } catch (...) @@ -78,6 +88,11 @@ bool DatabaseReplicatedDDLWorker::initializeMainThread() } } + { + std::lock_guard lock(initialization_duration_timer_mutex); + initialization_duration_timer.reset(); + } + return false; } @@ -459,4 +474,10 @@ UInt32 DatabaseReplicatedDDLWorker::getLogPointer() const return max_id.load(); } +UInt64 DatabaseReplicatedDDLWorker::getCurrentInitializationDurationMs() const +{ + std::lock_guard lock(initialization_duration_timer_mutex); + return initialization_duration_timer ? initialization_duration_timer->elapsedMilliseconds() : 0; +} + } diff --git a/src/Databases/DatabaseReplicatedWorker.h b/src/Databases/DatabaseReplicatedWorker.h index 41edf2221b8..2309c831839 100644 --- a/src/Databases/DatabaseReplicatedWorker.h +++ b/src/Databases/DatabaseReplicatedWorker.h @@ -36,6 +36,8 @@ public: DatabaseReplicated * const database, bool committed = false); /// NOLINT UInt32 getLogPointer() const; + + UInt64 getCurrentInitializationDurationMs() const; private: bool initializeMainThread() override; void initializeReplication(); @@ -56,6 +58,9 @@ private: ZooKeeperPtr active_node_holder_zookeeper; /// It will remove "active" node when database is detached zkutil::EphemeralNodeHolderPtr active_node_holder; + + std::optional initialization_duration_timer; + mutable std::mutex initialization_duration_timer_mutex; }; } diff --git a/src/Databases/MySQL/MaterializedMySQLSyncThread.cpp b/src/Databases/MySQL/MaterializedMySQLSyncThread.cpp index 04b4070d5af..1364e9ae2b2 100644 --- a/src/Databases/MySQL/MaterializedMySQLSyncThread.cpp +++ b/src/Databases/MySQL/MaterializedMySQLSyncThread.cpp @@ -736,11 +736,11 @@ static void writeFieldsToColumn( { for (size_t index = 0; index < rows_data.size(); ++index) { - const Tuple & row_data = rows_data[index].get(); + const Tuple & row_data = rows_data[index].safeGet(); const Field & value = row_data[column_index]; if (write_data_to_null_map(value, index)) - casted_column->insertValue(static_cast(value.template get())); + casted_column->insertValue(static_cast(value.template safeGet())); } }; @@ -776,17 +776,17 @@ static void writeFieldsToColumn( { for (size_t index = 0; index < rows_data.size(); ++index) { - const Tuple & row_data = rows_data[index].get(); + const Tuple & row_data = rows_data[index].safeGet(); const Field & value = row_data[column_index]; if (write_data_to_null_map(value, index)) { if (value.getType() == Field::Types::UInt64) - casted_int32_column->insertValue(static_cast(value.get())); + casted_int32_column->insertValue(static_cast(value.safeGet())); else if (value.getType() == Field::Types::Int64) { /// For MYSQL_TYPE_INT24 - const Int32 & num = static_cast(value.get()); + const Int32 & num = static_cast(value.safeGet()); casted_int32_column->insertValue(num & 0x800000 ? num | 0xFF000000 : num); } else @@ -798,7 +798,7 @@ static void writeFieldsToColumn( { for (size_t index = 0; index < rows_data.size(); ++index) { - const Tuple & row_data = rows_data[index].get(); + const Tuple & row_data = rows_data[index].safeGet(); const Field & value = row_data[column_index]; if (write_data_to_null_map(value, index)) @@ -812,12 +812,12 @@ static void writeFieldsToColumn( { for (size_t index = 0; index < rows_data.size(); ++index) { - const Tuple & row_data = rows_data[index].get(); + const Tuple & row_data = rows_data[index].safeGet(); const Field & value = row_data[column_index]; if (write_data_to_null_map(value, index)) { - const String & data = value.get(); + const String & data = value.safeGet(); casted_fixed_string_column->insertData(data.data(), data.size()); } } @@ -864,7 +864,7 @@ static inline size_t onUpdateData(const Row & rows_data, Block & buffer, size_t { writeable_rows_mask[index + 1] = true; writeable_rows_mask[index] = differenceSortingKeys( - rows_data[index].get(), rows_data[index + 1].get(), sorting_columns_index); + rows_data[index].safeGet(), rows_data[index + 1].safeGet(), sorting_columns_index); } for (size_t column = 0; column < buffer.columns() - 2; ++column) diff --git a/src/Databases/MySQL/tests/gtest_mysql_binlog.cpp b/src/Databases/MySQL/tests/gtest_mysql_binlog.cpp index 11299c5b8b1..6f1ba26ee33 100644 --- a/src/Databases/MySQL/tests/gtest_mysql_binlog.cpp +++ b/src/Databases/MySQL/tests/gtest_mysql_binlog.cpp @@ -281,12 +281,12 @@ static void testFile1(IBinlog & binlog, UInt64 timeout, bool filtered = false) ASSERT_EQ(write_event->table, "a"); ASSERT_EQ(write_event->rows.size(), 1); ASSERT_EQ(write_event->rows[0].getType(), Field::Types::Tuple); - auto row_data = write_event->rows[0].get(); + auto row_data = write_event->rows[0].safeGet(); ASSERT_EQ(row_data.size(), 4u); - ASSERT_EQ(row_data[0].get(), 1u); - ASSERT_EQ(row_data[1].get(), 1u); - ASSERT_EQ(row_data[2].get(), 1u); - ASSERT_EQ(row_data[3].get(), 1u); + ASSERT_EQ(row_data[0].safeGet(), 1u); + ASSERT_EQ(row_data[1].safeGet(), 1u); + ASSERT_EQ(row_data[2].safeGet(), 1u); + ASSERT_EQ(row_data[3].safeGet(), 1u); ASSERT_TRUE(binlog.tryReadEvent(event, timeout)); ++count; @@ -342,18 +342,18 @@ static void testFile1(IBinlog & binlog, UInt64 timeout, bool filtered = false) ASSERT_EQ(update_event->table, "a"); ASSERT_EQ(update_event->rows.size(), 2); ASSERT_EQ(update_event->rows[0].getType(), Field::Types::Tuple); - row_data = update_event->rows[0].get(); + row_data = update_event->rows[0].safeGet(); ASSERT_EQ(row_data.size(), 4u); - ASSERT_EQ(row_data[0].get(), 1u); - ASSERT_EQ(row_data[1].get(), 1u); - ASSERT_EQ(row_data[2].get(), 1u); - ASSERT_EQ(row_data[3].get(), 1u); - row_data = update_event->rows[1].get(); + ASSERT_EQ(row_data[0].safeGet(), 1u); + ASSERT_EQ(row_data[1].safeGet(), 1u); + ASSERT_EQ(row_data[2].safeGet(), 1u); + ASSERT_EQ(row_data[3].safeGet(), 1u); + row_data = update_event->rows[1].safeGet(); ASSERT_EQ(row_data.size(), 4u); - ASSERT_EQ(row_data[0].get(), 1u); - ASSERT_EQ(row_data[1].get(), 2u); - ASSERT_EQ(row_data[2].get(), 1u); - ASSERT_EQ(row_data[3].get(), 1u); + ASSERT_EQ(row_data[0].safeGet(), 1u); + ASSERT_EQ(row_data[1].safeGet(), 2u); + ASSERT_EQ(row_data[2].safeGet(), 1u); + ASSERT_EQ(row_data[3].safeGet(), 1u); ASSERT_TRUE(binlog.tryReadEvent(event, timeout)); ++count; diff --git a/src/Databases/PostgreSQL/fetchPostgreSQLTableStructure.cpp b/src/Databases/PostgreSQL/fetchPostgreSQLTableStructure.cpp index 943f3ae502e..b9fd9c325f8 100644 --- a/src/Databases/PostgreSQL/fetchPostgreSQLTableStructure.cpp +++ b/src/Databases/PostgreSQL/fetchPostgreSQLTableStructure.cpp @@ -196,7 +196,7 @@ PostgreSQLTableStructure::ColumnsInfoPtr readNamesAndTypesList( } else { - std::tuple row; + std::tuple row; while (stream >> row) { const auto column_name = std::get<0>(row); @@ -206,13 +206,14 @@ PostgreSQLTableStructure::ColumnsInfoPtr readNamesAndTypesList( std::get<3>(row)); columns.push_back(NameAndTypePair(column_name, data_type)); - auto attgenerated = std::get<6>(row); + auto attgenerated = std::get<7>(row); attributes.emplace( column_name, PostgreSQLTableStructure::PGAttribute{ .atttypid = parse(std::get<4>(row)), .atttypmod = parse(std::get<5>(row)), + .attnum = parse(std::get<6>(row)), .atthasdef = false, .attgenerated = attgenerated.empty() ? char{} : char(attgenerated[0]), .attr_def = {} @@ -308,6 +309,7 @@ PostgreSQLTableStructure fetchPostgreSQLTableStructure( "attndims AS dims, " /// array dimensions "atttypid as type_id, " "atttypmod as type_modifier, " + "attnum as att_num, " "attgenerated as generated " /// if column has GENERATED "FROM pg_attribute " "WHERE attrelid = (SELECT oid FROM pg_class WHERE {}) " @@ -338,17 +340,29 @@ PostgreSQLTableStructure fetchPostgreSQLTableStructure( "WHERE adrelid = (SELECT oid FROM pg_class WHERE {});", where); pqxx::result result{tx.exec(attrdef_query)}; - for (const auto row : result) + if (static_cast(result.size()) > table.physical_columns->names.size()) { - size_t adnum = row[0].as(); - if (!adnum || adnum > table.physical_columns->names.size()) + throw Exception(ErrorCodes::LOGICAL_ERROR, + "Received {} attrdef, but currently fetched columns list has {} columns", + result.size(), table.physical_columns->attributes.size()); + } + + for (const auto & column_attrs : table.physical_columns->attributes) + { + if (column_attrs.second.attgenerated != 's') /// e.g. not a generated column { - throw Exception(ErrorCodes::LOGICAL_ERROR, - "Received adnum {}, but currently fetched columns list has {} columns", - adnum, table.physical_columns->attributes.size()); + continue; + } + + for (const auto row : result) + { + int adnum = row[0].as(); + if (column_attrs.second.attnum == adnum) + { + table.physical_columns->attributes.at(column_attrs.first).attr_def = row[1].as(); + break; + } } - const auto column_name = table.physical_columns->names[adnum - 1]; - table.physical_columns->attributes.at(column_name).attr_def = row[1].as(); } } diff --git a/src/Databases/PostgreSQL/fetchPostgreSQLTableStructure.h b/src/Databases/PostgreSQL/fetchPostgreSQLTableStructure.h index 81bf7b278fc..25ece6909fd 100644 --- a/src/Databases/PostgreSQL/fetchPostgreSQLTableStructure.h +++ b/src/Databases/PostgreSQL/fetchPostgreSQLTableStructure.h @@ -16,6 +16,7 @@ struct PostgreSQLTableStructure { Int32 atttypid; Int32 atttypmod; + Int32 attnum; bool atthasdef; char attgenerated; std::string attr_def; diff --git a/src/Databases/TablesLoader.h b/src/Databases/TablesLoader.h index 26b5777f1a9..bf469d83245 100644 --- a/src/Databases/TablesLoader.h +++ b/src/Databases/TablesLoader.h @@ -1,4 +1,5 @@ #pragma once + #include #include #include diff --git a/src/Dictionaries/CacheDictionaryStorage.h b/src/Dictionaries/CacheDictionaryStorage.h index 47f99bd1093..781822533e9 100644 --- a/src/Dictionaries/CacheDictionaryStorage.h +++ b/src/Dictionaries/CacheDictionaryStorage.h @@ -395,13 +395,13 @@ private: } else if constexpr (std::is_same_v) { - const String & string_value = column_value.get(); + const String & string_value = column_value.safeGet(); StringRef inserted_value = copyStringInArena(arena, string_value); container.back() = inserted_value; } else { - container.back() = static_cast(column_value.get()); + container.back() = static_cast(column_value.safeGet()); } }); } @@ -441,7 +441,7 @@ private: } else if constexpr (std::is_same_v) { - const String & string_value = column_value.get(); + const String & string_value = column_value.safeGet(); StringRef inserted_value = copyStringInArena(arena, string_value); if (!cell_was_default) @@ -454,7 +454,7 @@ private: } else { - container[index_to_use] = static_cast(column_value.get()); + container[index_to_use] = static_cast(column_value.safeGet()); } }); } @@ -651,12 +651,12 @@ private: } else if constexpr (std::is_same_v) { - auto & value = default_value.get(); + auto & value = default_value.safeGet(); value_setter(value); } else { - value_setter(default_value.get()); + value_setter(default_value.safeGet()); } } else diff --git a/src/Dictionaries/ClickHouseDictionarySource.cpp b/src/Dictionaries/ClickHouseDictionarySource.cpp index bf16f315ddf..b36d53a6159 100644 --- a/src/Dictionaries/ClickHouseDictionarySource.cpp +++ b/src/Dictionaries/ClickHouseDictionarySource.cpp @@ -51,6 +51,8 @@ namespace configuration.db, configuration.user, configuration.password, + configuration.proto_send_chunked, + configuration.proto_recv_chunked, configuration.quota_key, "", /* cluster */ "", /* cluster_secret */ @@ -222,7 +224,7 @@ void registerDictionarySourceClickHouse(DictionarySourceFactory & factory) { validateNamedCollection( *named_collection, {}, ValidateKeysMultiset{ - "secure", "host", "hostname", "port", "user", "username", "password", "quota_key", "name", + "secure", "host", "hostname", "port", "user", "username", "password", "proto_send_chunked", "proto_recv_chunked", "quota_key", "name", "db", "database", "table","query", "where", "invalidate_query", "update_field", "update_lag"}); const auto secure = named_collection->getOrDefault("secure", false); @@ -234,6 +236,8 @@ void registerDictionarySourceClickHouse(DictionarySourceFactory & factory) .host = host, .user = named_collection->getAnyOrDefault({"user", "username"}, "default"), .password = named_collection->getOrDefault("password", ""), + .proto_send_chunked = named_collection->getOrDefault("proto_send_chunked", "notchunked"), + .proto_recv_chunked = named_collection->getOrDefault("proto_recv_chunked", "notchunked"), .quota_key = named_collection->getOrDefault("quota_key", ""), .db = named_collection->getAnyOrDefault({"db", "database"}, default_database), .table = named_collection->getOrDefault("table", ""), @@ -258,6 +262,8 @@ void registerDictionarySourceClickHouse(DictionarySourceFactory & factory) .host = host, .user = config.getString(settings_config_prefix + ".user", "default"), .password = config.getString(settings_config_prefix + ".password", ""), + .proto_send_chunked = config.getString(settings_config_prefix + ".proto_caps.send", "notchunked"), + .proto_recv_chunked = config.getString(settings_config_prefix + ".proto_caps.recv", "notchunked"), .quota_key = config.getString(settings_config_prefix + ".quota_key", ""), .db = config.getString(settings_config_prefix + ".db", default_database), .table = config.getString(settings_config_prefix + ".table", ""), diff --git a/src/Dictionaries/ClickHouseDictionarySource.h b/src/Dictionaries/ClickHouseDictionarySource.h index 3357514eab2..faf9e5f8009 100644 --- a/src/Dictionaries/ClickHouseDictionarySource.h +++ b/src/Dictionaries/ClickHouseDictionarySource.h @@ -23,6 +23,8 @@ public: const std::string host; const std::string user; const std::string password; + const std::string proto_send_chunked; + const std::string proto_recv_chunked; const std::string quota_key; const std::string db; const std::string table; diff --git a/src/Dictionaries/DictionaryHelpers.h b/src/Dictionaries/DictionaryHelpers.h index 64fc05e99ab..43fd39640c3 100644 --- a/src/Dictionaries/DictionaryHelpers.h +++ b/src/Dictionaries/DictionaryHelpers.h @@ -345,7 +345,7 @@ public: if (attribute_default_value.isNull()) default_value_is_null = true; else - default_value = static_cast(attribute_default_value.get()); + default_value = static_cast(attribute_default_value.safeGet()); } else { @@ -377,7 +377,7 @@ public: if constexpr (std::is_same_v) { Field field = (*default_values_column)[row]; - return field.get(); + return field.safeGet(); } else if constexpr (std::is_same_v) return default_values_column->getDataAt(row); diff --git a/src/Dictionaries/FlatDictionary.cpp b/src/Dictionaries/FlatDictionary.cpp index 999160226d9..b0233766741 100644 --- a/src/Dictionaries/FlatDictionary.cpp +++ b/src/Dictionaries/FlatDictionary.cpp @@ -245,7 +245,7 @@ ColumnPtr FlatDictionary::getHierarchy(ColumnPtr key_column, const DataTypePtr & std::optional null_value; if (!dictionary_attribute.null_value.isNull()) - null_value = dictionary_attribute.null_value.get(); + null_value = dictionary_attribute.null_value.safeGet(); const ContainerType & parent_keys = std::get>(hierarchical_attribute.container); @@ -300,7 +300,7 @@ ColumnUInt8::Ptr FlatDictionary::isInHierarchy( std::optional null_value; if (!dictionary_attribute.null_value.isNull()) - null_value = dictionary_attribute.null_value.get(); + null_value = dictionary_attribute.null_value.safeGet(); const ContainerType & parent_keys = std::get>(hierarchical_attribute.container); @@ -701,7 +701,7 @@ void FlatDictionary::setAttributeValue(Attribute & attribute, const UInt64 key, return; } - auto & attribute_value = value.get(); + auto & attribute_value = value.safeGet(); auto & container = std::get>(attribute.container); loaded_keys[key] = true; diff --git a/src/Dictionaries/HashedArrayDictionary.cpp b/src/Dictionaries/HashedArrayDictionary.cpp index d7d50dfb0a6..8768be8e5ec 100644 --- a/src/Dictionaries/HashedArrayDictionary.cpp +++ b/src/Dictionaries/HashedArrayDictionary.cpp @@ -240,7 +240,7 @@ ColumnPtr HashedArrayDictionary::getHierarchy(Colu std::optional null_value; if (!dictionary_attribute.null_value.isNull()) - null_value = dictionary_attribute.null_value.get(); + null_value = dictionary_attribute.null_value.safeGet(); auto is_key_valid_func = [&, this](auto & key) @@ -313,7 +313,7 @@ ColumnUInt8::Ptr HashedArrayDictionary::isInHierar std::optional null_value; if (!dictionary_attribute.null_value.isNull()) - null_value = dictionary_attribute.null_value.get(); + null_value = dictionary_attribute.null_value.safeGet(); auto is_key_valid_func = [&](auto & key) @@ -581,13 +581,13 @@ void HashedArrayDictionary::blockToAttributes(cons if constexpr (std::is_same_v) { - String & value_to_insert = column_value_to_insert.get(); + String & value_to_insert = column_value_to_insert.safeGet(); StringRef string_in_arena_reference = copyStringInArena(*string_arenas[shard], value_to_insert); attribute_container.back() = string_in_arena_reference; } else { - auto value_to_insert = static_cast(column_value_to_insert.get()); + auto value_to_insert = static_cast(column_value_to_insert.safeGet()); attribute_container.back() = value_to_insert; } }; diff --git a/src/Dictionaries/HashedDictionary.h b/src/Dictionaries/HashedDictionary.h index 3a2b61e5149..7e935fe4855 100644 --- a/src/Dictionaries/HashedDictionary.h +++ b/src/Dictionaries/HashedDictionary.h @@ -636,7 +636,7 @@ ColumnPtr HashedDictionary::getHierarchy(C std::optional null_value; if (!dictionary_attribute.null_value.isNull()) - null_value = dictionary_attribute.null_value.get(); + null_value = dictionary_attribute.null_value.safeGet(); const CollectionsHolder & child_key_to_parent_key_maps = std::get>(hierarchical_attribute.containers); @@ -710,7 +710,7 @@ ColumnUInt8::Ptr HashedDictionary::isInHie std::optional null_value; if (!dictionary_attribute.null_value.isNull()) - null_value = dictionary_attribute.null_value.get(); + null_value = dictionary_attribute.null_value.safeGet(); const CollectionsHolder & child_key_to_parent_key_maps = std::get>(hierarchical_attribute.containers); @@ -1004,13 +1004,13 @@ void HashedDictionary::blockToAttributes(c if constexpr (std::is_same_v) { - String & value_to_insert = column_value_to_insert.get(); + String & value_to_insert = column_value_to_insert.safeGet(); StringRef arena_value = copyStringInArena(*string_arenas[shard], value_to_insert); container.insert({key, arena_value}); } else { - auto value_to_insert = static_cast(column_value_to_insert.get()); + auto value_to_insert = static_cast(column_value_to_insert.safeGet()); container.insert({key, value_to_insert}); } diff --git a/src/Dictionaries/HierarchyDictionariesUtils.cpp b/src/Dictionaries/HierarchyDictionariesUtils.cpp index e1119982a34..de532ade26d 100644 --- a/src/Dictionaries/HierarchyDictionariesUtils.cpp +++ b/src/Dictionaries/HierarchyDictionariesUtils.cpp @@ -50,7 +50,7 @@ namespace std::optional null_value; if (!hierarchical_attribute.null_value.isNull()) - null_value = hierarchical_attribute.null_value.get(); + null_value = hierarchical_attribute.null_value.safeGet(); ColumnPtr key_to_request_column = ColumnVector::create(); auto * key_to_request_column_typed = static_cast *>(key_to_request_column->assumeMutable().get()); @@ -190,7 +190,7 @@ ColumnPtr getKeysHierarchyDefaultImplementation( std::optional null_value; if (!hierarchical_attribute.null_value.isNull()) - null_value = hierarchical_attribute.null_value.get(); + null_value = hierarchical_attribute.null_value.safeGet(); auto get_parent_key_func = [&](auto & key) { @@ -252,7 +252,7 @@ ColumnUInt8::Ptr getKeysIsInHierarchyDefaultImplementation( std::optional null_value; if (!hierarchical_attribute.null_value.isNull()) - null_value = hierarchical_attribute.null_value.get(); + null_value = hierarchical_attribute.null_value.safeGet(); auto get_parent_key_func = [&](auto & key) { diff --git a/src/Dictionaries/IPAddressDictionary.cpp b/src/Dictionaries/IPAddressDictionary.cpp index 41fafcc162b..4f9e991752f 100644 --- a/src/Dictionaries/IPAddressDictionary.cpp +++ b/src/Dictionaries/IPAddressDictionary.cpp @@ -613,14 +613,14 @@ void IPAddressDictionary::calculateBytesAllocated() template void IPAddressDictionary::createAttributeImpl(Attribute & attribute, const Field & null_value) { - attribute.null_values = null_value.isNull() ? T{} : T(null_value.get()); + attribute.null_values = null_value.isNull() ? T{} : T(null_value.safeGet()); attribute.maps.emplace>(); } template <> void IPAddressDictionary::createAttributeImpl(Attribute & attribute, const Field & null_value) { - attribute.null_values = null_value.isNull() ? String() : null_value.get(); + attribute.null_values = null_value.isNull() ? String() : null_value.safeGet(); attribute.maps.emplace>(); attribute.string_arena = std::make_unique(); } @@ -976,13 +976,13 @@ void IPAddressDictionary::setAttributeValue(Attribute & attribute, const Field & if constexpr (std::is_same_v) { - const auto & string = value.get(); + const auto & string = value.safeGet(); const auto * string_in_arena = attribute.string_arena->insert(string.data(), string.size()); setAttributeValueImpl(attribute, StringRef{string_in_arena, string.size()}); } else { - setAttributeValueImpl(attribute, static_cast(value.get())); + setAttributeValueImpl(attribute, static_cast(value.safeGet())); } }; diff --git a/src/Dictionaries/MongoDBDictionarySource.cpp b/src/Dictionaries/MongoDBDictionarySource.cpp index 46910fa9f6a..c30a6f90e44 100644 --- a/src/Dictionaries/MongoDBDictionarySource.cpp +++ b/src/Dictionaries/MongoDBDictionarySource.cpp @@ -1,7 +1,6 @@ #include "MongoDBDictionarySource.h" #include "DictionarySourceFactory.h" #include "DictionaryStructure.h" -#include "registerDictionaries.h" #include #include @@ -233,7 +232,7 @@ QueryPipeline MongoDBDictionarySource::loadKeys(const Columns & key_columns, con } case AttributeUnderlyingType::String: { - String loaded_str((*key_columns[attribute_index])[row_idx].get()); + String loaded_str((*key_columns[attribute_index])[row_idx].safeGet()); /// Convert string to ObjectID if (key_attribute.is_object_id) { diff --git a/src/Dictionaries/PolygonDictionary.cpp b/src/Dictionaries/PolygonDictionary.cpp index dfc920623e3..ff29ca1f6b8 100644 --- a/src/Dictionaries/PolygonDictionary.cpp +++ b/src/Dictionaries/PolygonDictionary.cpp @@ -141,7 +141,7 @@ ColumnPtr IPolygonDictionary::getColumn( { getItemsShortCircuitImpl( requested_key_points, - [&](size_t row) { return (*attribute_values_column)[row].get(); }, + [&](size_t row) { return (*attribute_values_column)[row].safeGet(); }, [&](Array & value) { result_column_typed.insert(value); }, default_mask.value()); } @@ -149,7 +149,7 @@ ColumnPtr IPolygonDictionary::getColumn( { getItemsImpl( requested_key_points, - [&](size_t row) { return (*attribute_values_column)[row].get(); }, + [&](size_t row) { return (*attribute_values_column)[row].safeGet(); }, [&](Array & value) { result_column_typed.insert(value); }, default_value_provider.value()); } @@ -432,16 +432,16 @@ void IPolygonDictionary::getItemsImpl( } else if constexpr (std::is_same_v) { - set_value(default_value.get()); + set_value(default_value.safeGet()); } else if constexpr (std::is_same_v) { - auto default_value_string = default_value.get(); + auto default_value_string = default_value.safeGet(); set_value(default_value_string); } else { - set_value(default_value.get>()); + set_value(default_value.safeGet>()); } } } diff --git a/src/Dictionaries/RangeHashedDictionary.h b/src/Dictionaries/RangeHashedDictionary.h index fc6c98990d0..c264b480bcb 100644 --- a/src/Dictionaries/RangeHashedDictionary.h +++ b/src/Dictionaries/RangeHashedDictionary.h @@ -906,13 +906,13 @@ void RangeHashedDictionary::setAttributeValue(Attribute & a if constexpr (std::is_same_v) { - const auto & string = value.get(); + const auto & string = value.safeGet(); StringRef string_ref = copyStringInArena(string_arena, string); value_to_insert = string_ref; } else { - value_to_insert = static_cast(value.get()); + value_to_insert = static_cast(value.safeGet()); } container.back() = value_to_insert; diff --git a/src/Dictionaries/RedisDictionarySource.cpp b/src/Dictionaries/RedisDictionarySource.cpp index 1736cdff306..9db639a0ca4 100644 --- a/src/Dictionaries/RedisDictionarySource.cpp +++ b/src/Dictionaries/RedisDictionarySource.cpp @@ -1,7 +1,6 @@ #include "RedisDictionarySource.h" #include "DictionarySourceFactory.h" #include "DictionaryStructure.h" -#include "registerDictionaries.h" #include #include @@ -160,7 +159,7 @@ namespace DB if (isInteger(type)) key << DB::toString(key_columns[i]->get64(row)); else if (isString(type)) - key << (*key_columns[i])[row].get(); + key << (*key_columns[i])[row].safeGet(); else throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected type of key in Redis dictionary"); } diff --git a/src/Disks/DiskFomAST.cpp b/src/Disks/DiskFomAST.cpp new file mode 100644 index 00000000000..5329ff8748a --- /dev/null +++ b/src/Disks/DiskFomAST.cpp @@ -0,0 +1,150 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int BAD_ARGUMENTS; +} + +std::string getOrCreateCustomDisk(DiskConfigurationPtr config, const std::string & serialization, ContextPtr context, bool attach) +{ + Poco::Util::AbstractConfiguration::Keys disk_settings_keys; + config->keys(disk_settings_keys); + /// Check that no settings are defined when disk from the config is referred. + if (disk_settings_keys.empty()) + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "Disk function must have arguments. Invalid disk description."); + + if (disk_settings_keys.size() == 1 && disk_settings_keys.front() == "name" && !attach) + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "Disk function `{}` must have other arguments apart from `name`, which describe disk configuration. Invalid disk description.", + serialization); + + auto disk_settings_hash = sipHash128(serialization.data(), serialization.size()); + + std::string disk_name; + if (config->has("name")) + { + disk_name = config->getString("name"); + } + else + { + /// We need a unique name for a created custom disk, but it needs to be the same + /// after table is reattached or server is restarted, so take a hash of the disk + /// configuration serialized ast as a disk name suffix. + disk_name = DiskSelector::TMP_INTERNAL_DISK_PREFIX + toString(disk_settings_hash); + } + + + auto disk = context->getOrCreateDisk(disk_name, [&](const DisksMap & disks_map) -> DiskPtr { + auto result = DiskFactory::instance().create( + disk_name, *config, /* config_path */"", context, disks_map, /* attach */attach, /* custom_disk */true); + /// Mark that disk can be used without storage policy. + result->markDiskAsCustom(disk_settings_hash); + return result; + }); + + if (!disk->isCustomDisk()) + throw Exception(ErrorCodes::BAD_ARGUMENTS, + "Disk `{}` already exists and is described by the config." + " It is impossible to redefine it.", + disk_name); + + if (disk->getCustomDiskSettings() != disk_settings_hash && !attach) + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "The disk `{}` is already configured as a custom disk in another table. It can't be redefined with different settings.", + disk_name); + + if (!attach && !disk->isRemote()) + { + static constexpr auto custom_local_disks_base_dir_in_config = "custom_local_disks_base_directory"; + auto disk_path_expected_prefix = context->getConfigRef().getString(custom_local_disks_base_dir_in_config, ""); + + if (disk_path_expected_prefix.empty()) + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "Base path for custom local disks must be defined in config file by `{}`", + custom_local_disks_base_dir_in_config); + + if (!pathStartsWith(disk->getPath(), disk_path_expected_prefix)) + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "Path of the custom local disk must be inside `{}` directory", + disk_path_expected_prefix); + } + + return disk_name; +} + +class DiskConfigurationFlattener +{ +public: + struct Data + { + ContextPtr context; + bool attach; + }; + + static bool needChildVisit(const ASTPtr &, const ASTPtr &) { return true; } + + static void visit(ASTPtr & ast, Data & data) + { + if (isDiskFunction(ast)) + { + const auto * function = ast->as(); + const auto * function_args_expr = assert_cast(function->arguments.get()); + const auto & function_args = function_args_expr->children; + auto config = getDiskConfigurationFromAST(function_args, data.context); + auto disk_setting_string = serializeAST(*function); + auto disk_name = getOrCreateCustomDisk(config, disk_setting_string, data.context, data.attach); + ast = std::make_shared(disk_name); + } + } +}; + + +std::string DiskFomAST::createCustomDisk(const ASTPtr & disk_function_ast, ContextPtr context, bool attach) +{ + if (!isDiskFunction(disk_function_ast)) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected a disk function"); + + auto ast = disk_function_ast->clone(); + + using FlattenDiskConfigurationVisitor = InDepthNodeVisitor; + FlattenDiskConfigurationVisitor::Data data{context, attach}; + FlattenDiskConfigurationVisitor{data}.visit(ast); + + return assert_cast(*ast).value.safeGet(); +} + +void DiskFomAST::ensureDiskIsNotCustom(const std::string & disk_name, ContextPtr context) +{ + auto disk = context->getDisk(disk_name); + + if (disk->isCustomDisk()) + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "Disk name `{}` is a custom disk that is used in other table. " + "That disk could not be used by a reference by other tables. The custom disk should be fully specified with a disk function.", + disk_name); +} + +} diff --git a/src/Disks/DiskFomAST.h b/src/Disks/DiskFomAST.h new file mode 100644 index 00000000000..0a30834533e --- /dev/null +++ b/src/Disks/DiskFomAST.h @@ -0,0 +1,15 @@ +#pragma once +#include +#include +#include + +namespace DB +{ + +namespace DiskFomAST +{ + void ensureDiskIsNotCustom(const std::string & name, ContextPtr context); + std::string createCustomDisk(const ASTPtr & disk_function, ContextPtr context, bool attach); +} + +} diff --git a/src/Disks/DiskSelector.h b/src/Disks/DiskSelector.h index 49a1be5cf50..e6e2c257911 100644 --- a/src/Disks/DiskSelector.h +++ b/src/Disks/DiskSelector.h @@ -6,6 +6,8 @@ #include #include +#include +#include namespace DB { diff --git a/src/Disks/IDisk.h b/src/Disks/IDisk.h index 412ad27e94f..78d5f37e3a7 100644 --- a/src/Disks/IDisk.h +++ b/src/Disks/IDisk.h @@ -464,9 +464,9 @@ public: virtual void chmod(const String & /*path*/, mode_t /*mode*/) { throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Disk does not support chmod"); } /// Was disk created to be used without storage configuration? - bool isCustomDisk() const { return is_custom_disk; } - - void markDiskAsCustom() { is_custom_disk = true; } + bool isCustomDisk() const { return custom_disk_settings_hash != 0; } + UInt128 getCustomDiskSettings() const { return custom_disk_settings_hash; } + void markDiskAsCustom(UInt128 settings_hash) { custom_disk_settings_hash = settings_hash; } virtual DiskPtr getDelegateDiskIfExists() const { return nullptr; } @@ -504,7 +504,8 @@ protected: private: ThreadPool copying_thread_pool; - bool is_custom_disk = false; + // 0 means the disk is not custom, the disk is predefined in the config + UInt128 custom_disk_settings_hash = 0; /// Check access to the disk. void checkAccess(); diff --git a/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp b/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp index b471f3fc58f..56bfa019819 100644 --- a/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp +++ b/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp @@ -645,8 +645,9 @@ void CachedOnDiskReadBufferFromFile::predownload(FileSegment & file_segment) ProfileEvents::increment(ProfileEvents::CachedReadBufferReadFromSourceBytes, current_impl_buffer_size); + std::string failure_reason; bool continue_predownload = file_segment.reserve( - current_predownload_size, settings.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds); + current_predownload_size, settings.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds, failure_reason); if (continue_predownload) { LOG_TEST(log, "Left to predownload: {}, buffer size: {}", bytes_to_predownload, current_impl_buffer_size); @@ -1002,7 +1003,8 @@ bool CachedOnDiskReadBufferFromFile::nextImplStep() { chassert(file_offset_of_buffer_end + size - 1 <= file_segment.range().right); - bool success = file_segment.reserve(size, settings.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds); + std::string failure_reason; + bool success = file_segment.reserve(size, settings.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds, failure_reason); if (success) { chassert(file_segment.getCurrentWriteOffset() == static_cast(implementation_buffer->getPosition())); @@ -1028,7 +1030,8 @@ bool CachedOnDiskReadBufferFromFile::nextImplStep() LOG_TRACE(log, "Bypassing cache because writeCache method failed"); } else - LOG_TRACE(log, "No space left in cache to reserve {} bytes, will continue without cache download", size); + LOG_TRACE(log, "No space left in cache to reserve {} bytes, reason: {}, " + "will continue without cache download", size, failure_reason); if (!success) { diff --git a/src/Disks/IO/CachedOnDiskWriteBufferFromFile.cpp b/src/Disks/IO/CachedOnDiskWriteBufferFromFile.cpp index 382c4a80cc4..103ae0e1832 100644 --- a/src/Disks/IO/CachedOnDiskWriteBufferFromFile.cpp +++ b/src/Disks/IO/CachedOnDiskWriteBufferFromFile.cpp @@ -91,7 +91,8 @@ bool FileSegmentRangeWriter::write(char * data, size_t size, size_t offset, File size_t size_to_write = std::min(available_size, size); - bool reserved = file_segment->reserve(size_to_write, reserve_space_lock_wait_timeout_milliseconds); + std::string failure_reason; + bool reserved = file_segment->reserve(size_to_write, reserve_space_lock_wait_timeout_milliseconds, failure_reason); if (!reserved) { appendFilesystemCacheLog(*file_segment); diff --git a/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageCommon.cpp b/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageCommon.cpp index 0aa3b9c40b5..1a0b6157a86 100644 --- a/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageCommon.cpp +++ b/src/Disks/ObjectStorages/AzureBlobStorage/AzureBlobStorageCommon.cpp @@ -11,6 +11,14 @@ #include #include +namespace ProfileEvents +{ + extern const Event AzureGetProperties; + extern const Event DiskAzureGetProperties; + extern const Event AzureCreateContainer; + extern const Event DiskAzureCreateContainer; +} + namespace DB { @@ -214,20 +222,53 @@ void processURL(const String & url, const String & container_name, Endpoint & en } } +static bool containerExists(const ContainerClient & client) +{ + ProfileEvents::increment(ProfileEvents::AzureGetProperties); + if (client.GetClickhouseOptions().IsClientForDisk) + ProfileEvents::increment(ProfileEvents::DiskAzureGetProperties); + + try + { + client.GetProperties(); + return true; + } + catch (const Azure::Storage::StorageException & e) + { + if (e.StatusCode == Azure::Core::Http::HttpStatusCode::NotFound) + return false; + throw; + } +} + std::unique_ptr getContainerClient(const ConnectionParams & params, bool readonly) { if (params.endpoint.container_already_exists.value_or(false) || readonly) + { return params.createForContainer(); + } + + if (!params.endpoint.container_already_exists.has_value()) + { + auto container_client = params.createForContainer(); + if (containerExists(*container_client)) + return container_client; + } try { auto service_client = params.createForService(); + + ProfileEvents::increment(ProfileEvents::AzureCreateContainer); + if (params.client_options.ClickhouseOptions.IsClientForDisk) + ProfileEvents::increment(ProfileEvents::DiskAzureCreateContainer); + 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 container_already_exists is not set (in config), ignore already exists error. Conflict - The specified container already exists. + /// To avoid race with creation of container, handle this error despite that we have already checked the existence of container. if (!params.endpoint.container_already_exists.has_value() && e.StatusCode == Azure::Core::Http::HttpStatusCode::Conflict) return params.createForContainer(); throw; diff --git a/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp b/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp index 0d92561d142..f85b5f45b37 100644 --- a/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp +++ b/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp @@ -86,6 +86,7 @@ private: Poco::Timestamp::fromEpochTime( std::chrono::duration_cast( static_cast(blob.Details.LastModified).time_since_epoch()).count()), + blob.Details.ETag.ToString(), {}})); } @@ -186,6 +187,7 @@ void AzureObjectStorage::listObjects(const std::string & path, RelativePathsWith Poco::Timestamp::fromEpochTime( std::chrono::duration_cast( static_cast(blob.Details.LastModified).time_since_epoch()).count()), + blob.Details.ETag.ToString(), {}})); } diff --git a/src/Disks/ObjectStorages/DiskObjectStorageMetadata.cpp b/src/Disks/ObjectStorages/DiskObjectStorageMetadata.cpp index 56d5d11ef8a..233b13fb908 100644 --- a/src/Disks/ObjectStorages/DiskObjectStorageMetadata.cpp +++ b/src/Disks/ObjectStorages/DiskObjectStorageMetadata.cpp @@ -205,7 +205,7 @@ void DiskObjectStorageMetadata::addObject(ObjectStorageKey key, size_t size) } total_size += size; - keys_with_meta.emplace_back(std::move(key), ObjectMetadata{size, {}, {}}); + keys_with_meta.emplace_back(std::move(key), ObjectMetadata{size, {}, {}, {}}); } ObjectKeyWithMetadata DiskObjectStorageMetadata::popLastObject() diff --git a/src/Disks/ObjectStorages/HDFS/HDFSObjectStorage.cpp b/src/Disks/ObjectStorages/HDFS/HDFSObjectStorage.cpp index 00ef4b63e6f..512cc34ef44 100644 --- a/src/Disks/ObjectStorages/HDFS/HDFSObjectStorage.cpp +++ b/src/Disks/ObjectStorages/HDFS/HDFSObjectStorage.cpp @@ -222,6 +222,7 @@ void HDFSObjectStorage::listObjects(const std::string & path, RelativePathsWithM ObjectMetadata{ static_cast(ls.file_info[i].mSize), Poco::Timestamp::fromEpochTime(ls.file_info[i].mLastMod), + "", {}})); } diff --git a/src/Disks/ObjectStorages/IObjectStorage.h b/src/Disks/ObjectStorages/IObjectStorage.h index 529c79790fd..f3c587a1188 100644 --- a/src/Disks/ObjectStorages/IObjectStorage.h +++ b/src/Disks/ObjectStorages/IObjectStorage.h @@ -54,6 +54,7 @@ struct ObjectMetadata { uint64_t size_bytes = 0; Poco::Timestamp last_modified; + std::string etag; ObjectAttributes attributes; }; diff --git a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp index 3c4b4d76bf5..8de80971238 100644 --- a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp +++ b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp @@ -63,7 +63,7 @@ void throwIfError(const Aws::Utils::Outcome & response) { const auto & err = response.GetError(); throw S3Exception( - fmt::format("{} (Code: {}, s3 exception: {})", + fmt::format("{} (Code: {}, S3 exception: '{}')", err.GetMessage(), static_cast(err.GetErrorType()), err.GetExceptionName()), err.GetErrorType()); } @@ -146,7 +146,7 @@ private: auto objects = outcome.GetResult().GetContents(); for (const auto & object : objects) { - ObjectMetadata metadata{static_cast(object.GetSize()), Poco::Timestamp::fromEpochTime(object.GetLastModified().Seconds()), {}}; + ObjectMetadata metadata{static_cast(object.GetSize()), Poco::Timestamp::fromEpochTime(object.GetLastModified().Seconds()), object.GetETag(), {}}; batch.emplace_back(std::make_shared(object.GetKey(), std::move(metadata))); } @@ -305,7 +305,8 @@ void S3ObjectStorage::listObjects(const std::string & path, RelativePathsWithMet S3::ListObjectsV2Request request; request.SetBucket(uri.bucket); - request.SetPrefix(path); + if (path != "/") + request.SetPrefix(path); if (max_keys) request.SetMaxKeys(static_cast(max_keys)); else @@ -332,6 +333,7 @@ void S3ObjectStorage::listObjects(const std::string & path, RelativePathsWithMet ObjectMetadata{ static_cast(object.GetSize()), Poco::Timestamp::fromEpochTime(object.GetLastModified().Seconds()), + object.GetETag(), {}})); if (max_keys) @@ -479,6 +481,7 @@ ObjectMetadata S3ObjectStorage::getObjectMetadata(const std::string & path) cons ObjectMetadata result; result.size_bytes = object_info.size; result.last_modified = Poco::Timestamp::fromEpochTime(object_info.last_modification_time); + result.etag = object_info.etag; result.attributes = object_info.metadata; return result; diff --git a/src/Disks/StoragePolicy.h b/src/Disks/StoragePolicy.h index 501e033abc3..8e49ed910e3 100644 --- a/src/Disks/StoragePolicy.h +++ b/src/Disks/StoragePolicy.h @@ -12,7 +12,6 @@ #include #include -#include #include #include #include diff --git a/src/Disks/getOrCreateDiskFromAST.cpp b/src/Disks/getOrCreateDiskFromAST.cpp deleted file mode 100644 index fd43f31a009..00000000000 --- a/src/Disks/getOrCreateDiskFromAST.cpp +++ /dev/null @@ -1,121 +0,0 @@ -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -namespace DB -{ - -namespace ErrorCodes -{ - extern const int BAD_ARGUMENTS; -} - -namespace -{ - std::string getOrCreateDiskFromDiskAST(const ASTFunction & function, ContextPtr context, bool attach) - { - const auto * function_args_expr = assert_cast(function.arguments.get()); - const auto & function_args = function_args_expr->children; - auto config = getDiskConfigurationFromAST(function_args, context); - - std::string disk_name; - if (config->has("name")) - { - disk_name = config->getString("name"); - } - else - { - /// We need a unique name for a created custom disk, but it needs to be the same - /// after table is reattached or server is restarted, so take a hash of the disk - /// configuration serialized ast as a disk name suffix. - auto disk_setting_string = serializeAST(function); - disk_name = DiskSelector::TMP_INTERNAL_DISK_PREFIX - + toString(sipHash128(disk_setting_string.data(), disk_setting_string.size())); - } - - auto result_disk = context->getOrCreateDisk(disk_name, [&](const DisksMap & disks_map) -> DiskPtr { - auto disk = DiskFactory::instance().create( - disk_name, *config, /* config_path */"", context, disks_map, /* attach */attach, /* custom_disk */true); - /// Mark that disk can be used without storage policy. - disk->markDiskAsCustom(); - return disk; - }); - - if (!result_disk->isCustomDisk()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Disk with name `{}` already exist", disk_name); - - if (!attach && !result_disk->isRemote()) - { - static constexpr auto custom_local_disks_base_dir_in_config = "custom_local_disks_base_directory"; - auto disk_path_expected_prefix = context->getConfigRef().getString(custom_local_disks_base_dir_in_config, ""); - - if (disk_path_expected_prefix.empty()) - throw Exception( - ErrorCodes::BAD_ARGUMENTS, - "Base path for custom local disks must be defined in config file by `{}`", - custom_local_disks_base_dir_in_config); - - if (!pathStartsWith(result_disk->getPath(), disk_path_expected_prefix)) - throw Exception( - ErrorCodes::BAD_ARGUMENTS, - "Path of the custom local disk must be inside `{}` directory", - disk_path_expected_prefix); - } - - return disk_name; - } - - class DiskConfigurationFlattener - { - public: - struct Data - { - ContextPtr context; - bool attach; - }; - - static bool needChildVisit(const ASTPtr &, const ASTPtr &) { return true; } - - static void visit(ASTPtr & ast, Data & data) - { - if (isDiskFunction(ast)) - { - auto disk_name = getOrCreateDiskFromDiskAST(*ast->as(), data.context, data.attach); - ast = std::make_shared(disk_name); - } - } - }; - - /// Visits children first. - using FlattenDiskConfigurationVisitor = InDepthNodeVisitor; -} - - -std::string getOrCreateDiskFromDiskAST(const ASTPtr & disk_function, ContextPtr context, bool attach) -{ - if (!isDiskFunction(disk_function)) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected a disk function"); - - auto ast = disk_function->clone(); - - FlattenDiskConfigurationVisitor::Data data{context, attach}; - FlattenDiskConfigurationVisitor{data}.visit(ast); - - auto disk_name = assert_cast(*ast).value.get(); - LOG_TRACE(getLogger("getOrCreateDiskFromDiskAST"), "Result disk name: {}", disk_name); - return disk_name; -} - -} diff --git a/src/Disks/getOrCreateDiskFromAST.h b/src/Disks/getOrCreateDiskFromAST.h deleted file mode 100644 index 61e1decbee9..00000000000 --- a/src/Disks/getOrCreateDiskFromAST.h +++ /dev/null @@ -1,18 +0,0 @@ -#pragma once -#include -#include -#include - -namespace DB -{ - -class ASTFunction; - -/** - * Create a DiskPtr from disk AST function like disk(), - * add it to DiskSelector by a unique (but always the same for given configuration) disk name - * and return this name. - */ -std::string getOrCreateDiskFromDiskAST(const ASTPtr & disk_function, ContextPtr context, bool attach); - -} diff --git a/src/Formats/EscapingRuleUtils.cpp b/src/Formats/EscapingRuleUtils.cpp index 58407a810c5..e7d9be39ec9 100644 --- a/src/Formats/EscapingRuleUtils.cpp +++ b/src/Formats/EscapingRuleUtils.cpp @@ -419,10 +419,11 @@ String getAdditionalFormatInfoByEscapingRule(const FormatSettings & settings, Fo String result = getAdditionalFormatInfoForAllRowBasedFormats(settings); /// First, settings that are common for all text formats: result += fmt::format( - ", try_infer_integers={}, try_infer_dates={}, try_infer_datetimes={}", + ", try_infer_integers={}, try_infer_dates={}, try_infer_datetimes={}, try_infer_datetimes_only_datetime64={}", settings.try_infer_integers, settings.try_infer_dates, - settings.try_infer_datetimes); + settings.try_infer_datetimes, + settings.try_infer_datetimes_only_datetime64); /// Second, format-specific settings: switch (escaping_rule) diff --git a/src/Formats/FormatFactory.cpp b/src/Formats/FormatFactory.cpp index 68946c895bb..da57c59bdfc 100644 --- a/src/Formats/FormatFactory.cpp +++ b/src/Formats/FormatFactory.cpp @@ -151,7 +151,6 @@ 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; @@ -267,6 +266,7 @@ FormatSettings getFormatSettings(const ContextPtr & context, const Settings & se format_settings.try_infer_integers = settings.input_format_try_infer_integers; format_settings.try_infer_dates = settings.input_format_try_infer_dates; format_settings.try_infer_datetimes = settings.input_format_try_infer_datetimes; + format_settings.try_infer_datetimes_only_datetime64 = settings.input_format_try_infer_datetimes_only_datetime64; format_settings.try_infer_exponent_floats = settings.input_format_try_infer_exponent_floats; 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; diff --git a/src/Formats/FormatSettings.h b/src/Formats/FormatSettings.h index 72104518ecc..3970c776ad2 100644 --- a/src/Formats/FormatSettings.h +++ b/src/Formats/FormatSettings.h @@ -46,6 +46,7 @@ struct FormatSettings bool try_infer_integers = true; bool try_infer_dates = true; bool try_infer_datetimes = true; + bool try_infer_datetimes_only_datetime64 = false; bool try_infer_exponent_floats = false; enum class DateTimeInputFormat : uint8_t @@ -234,7 +235,6 @@ 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 diff --git a/src/Formats/JSONExtractTree.cpp b/src/Formats/JSONExtractTree.cpp index 242d2dc9f80..86fde3852b8 100644 --- a/src/Formats/JSONExtractTree.cpp +++ b/src/Formats/JSONExtractTree.cpp @@ -1362,13 +1362,14 @@ public: } auto & variant_column = column_dynamic.getVariantColumn(); - auto variant_info = column_dynamic.getVariantInfo(); + const 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 element_type_name = element_type->getName(); + if (column_dynamic.addNewVariant(element_type, element_type_name)) { auto node = buildJSONExtractTree(element_type, "Dynamic inference"); - auto global_discriminator = variant_info.variant_name_to_discriminator[element_type->getName()]; + auto global_discriminator = variant_info.variant_name_to_discriminator.at(element_type_name); auto & variant = variant_column.getVariantByGlobalDiscriminator(global_discriminator); if (!node->insertResultToColumn(variant, element, insert_settings, format_settings, error)) return false; @@ -1377,29 +1378,15 @@ public: 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)) + /// We couldn't add this variant, insert it into shared variant. + auto tmp_variant_column = element_type->createColumn(); + auto node = buildJSONExtractTree(element_type, "Dynamic inference"); + if (!node->insertResultToColumn(*tmp_variant_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); + column_dynamic.insertValueIntoSharedVariant(*tmp_variant_column, element_type, element_type_name, 0); 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; diff --git a/src/Formats/JSONUtils.cpp b/src/Formats/JSONUtils.cpp index f0985f4a6b7..017befe5b0e 100644 --- a/src/Formats/JSONUtils.cpp +++ b/src/Formats/JSONUtils.cpp @@ -487,6 +487,8 @@ namespace JSONUtils size_t rows, size_t rows_before_limit, bool applied_limit, + size_t rows_before_aggregation, + bool applied_aggregation, const Stopwatch & watch, const Progress & progress, bool write_statistics, @@ -502,7 +504,12 @@ namespace JSONUtils writeTitle("rows_before_limit_at_least", out, 1, " "); writeIntText(rows_before_limit, out); } - + if (applied_aggregation) + { + writeFieldDelimiter(out, 2); + writeTitle("rows_before_aggregation", out, 1, " "); + writeIntText(rows_before_aggregation, out); + } if (write_statistics) { writeFieldDelimiter(out, 2); diff --git a/src/Formats/JSONUtils.h b/src/Formats/JSONUtils.h index 7ee111c1285..e2ac3467971 100644 --- a/src/Formats/JSONUtils.h +++ b/src/Formats/JSONUtils.h @@ -104,6 +104,8 @@ namespace JSONUtils size_t rows, size_t rows_before_limit, bool applied_limit, + size_t rows_before_aggregation, + bool applied_aggregation, const Stopwatch & watch, const Progress & progress, bool write_statistics, diff --git a/src/Formats/ReadSchemaUtils.cpp b/src/Formats/ReadSchemaUtils.cpp index 617595c19a2..dbbd728ed72 100644 --- a/src/Formats/ReadSchemaUtils.cpp +++ b/src/Formats/ReadSchemaUtils.cpp @@ -164,7 +164,7 @@ try return {*iterator_data.cached_columns, *format_name}; } - schemas_for_union_mode.emplace_back(iterator_data.cached_columns->getAll(), read_buffer_iterator.getLastFileName()); + schemas_for_union_mode.emplace_back(iterator_data.cached_columns->getAll(), read_buffer_iterator.getLastFilePath()); continue; } @@ -250,7 +250,7 @@ try if (!names_and_types.empty()) read_buffer_iterator.setSchemaToLastFile(ColumnsDescription(names_and_types)); - schemas_for_union_mode.emplace_back(names_and_types, read_buffer_iterator.getLastFileName()); + schemas_for_union_mode.emplace_back(names_and_types, read_buffer_iterator.getLastFilePath()); } catch (...) { @@ -411,7 +411,7 @@ try throw Exception(ErrorCodes::CANNOT_DETECT_FORMAT, "The data format cannot be detected by the contents of the files. You can specify the format manually"); read_buffer_iterator.setSchemaToLastFile(ColumnsDescription(names_and_types)); - schemas_for_union_mode.emplace_back(names_and_types, read_buffer_iterator.getLastFileName()); + schemas_for_union_mode.emplace_back(names_and_types, read_buffer_iterator.getLastFilePath()); } if (format_name && mode == SchemaInferenceMode::DEFAULT) @@ -527,9 +527,9 @@ try } catch (Exception & e) { - auto file_name = read_buffer_iterator.getLastFileName(); - if (!file_name.empty()) - e.addMessage(fmt::format("(in file/uri {})", file_name)); + auto file_path = read_buffer_iterator.getLastFilePath(); + if (!file_path.empty()) + e.addMessage(fmt::format("(in file/uri {})", file_path)); throw; } diff --git a/src/Formats/ReadSchemaUtils.h b/src/Formats/ReadSchemaUtils.h index bb5e068f696..7168e7f0817 100644 --- a/src/Formats/ReadSchemaUtils.h +++ b/src/Formats/ReadSchemaUtils.h @@ -56,8 +56,8 @@ struct IReadBufferIterator /// Set auto detected format name. virtual void setFormatName(const String & /*format_name*/) {} - /// Get last processed file name for better exception messages. - virtual String getLastFileName() const { return ""; } + /// Get last processed file path for better exception messages. + virtual String getLastFilePath() const { return ""; } /// Return true if method recreateLastReadBuffer is implemented. virtual bool supportsLastReadBufferRecreation() const { return false; } diff --git a/src/Formats/SchemaInferenceUtils.cpp b/src/Formats/SchemaInferenceUtils.cpp index 3c374ada9e6..5bd41e33f58 100644 --- a/src/Formats/SchemaInferenceUtils.cpp +++ b/src/Formats/SchemaInferenceUtils.cpp @@ -306,37 +306,45 @@ namespace type_indexes.erase(TypeIndex::UInt64); } - /// If we have only Date and DateTime types, convert Date to DateTime, - /// otherwise, convert all Date and DateTime to String. + /// If we have only date/datetimes types (Date/DateTime/DateTime64), convert all of them to the common type, + /// otherwise, convert all Date, DateTime and DateTime64 to String. void transformDatesAndDateTimes(DataTypes & data_types, TypeIndexesSet & type_indexes) { bool have_dates = type_indexes.contains(TypeIndex::Date); - bool have_datetimes = type_indexes.contains(TypeIndex::DateTime64); - bool all_dates_or_datetimes = (type_indexes.size() == (static_cast(have_dates) + static_cast(have_datetimes))); + bool have_datetimes = type_indexes.contains(TypeIndex::DateTime); + bool have_datetimes64 = type_indexes.contains(TypeIndex::DateTime64); + bool all_dates_or_datetimes = (type_indexes.size() == (static_cast(have_dates) + static_cast(have_datetimes) + static_cast(have_datetimes64))); - if (!all_dates_or_datetimes && (have_dates || have_datetimes)) + if (!all_dates_or_datetimes && (have_dates || have_datetimes || have_datetimes64)) { for (auto & type : data_types) { - if (isDate(type) || isDateTime64(type)) + if (isDate(type) || isDateTime(type) || isDateTime64(type)) type = std::make_shared(); } type_indexes.erase(TypeIndex::Date); type_indexes.erase(TypeIndex::DateTime); + type_indexes.erase(TypeIndex::DateTime64); type_indexes.insert(TypeIndex::String); return; } - if (have_dates && have_datetimes) + for (auto & type : data_types) { - for (auto & type : data_types) + if (isDate(type) && (have_datetimes || have_datetimes64)) { - if (isDate(type)) + if (have_datetimes64) type = std::make_shared(9); + else + type = std::make_shared(); + type_indexes.erase(TypeIndex::Date); + } + else if (isDateTime(type) && have_datetimes64) + { + type = std::make_shared(9); + type_indexes.erase(TypeIndex::DateTime); } - - type_indexes.erase(TypeIndex::Date); } } @@ -697,55 +705,87 @@ namespace bool tryInferDate(std::string_view field) { - if (field.empty()) + /// Minimum length of Date text representation is 8 (YYYY-M-D) and maximum is 10 (YYYY-MM-DD) + if (field.size() < 8 || field.size() > 10) return false; - ReadBufferFromString buf(field); - Float64 tmp_float; /// Check if it's just a number, and if so, don't try to infer Date from it, /// because we can interpret this number as a Date (for example 20000101 will be 2000-01-01) /// and it will lead to inferring Date instead of simple Int64/UInt64 in some cases. - if (tryReadFloatText(tmp_float, buf) && buf.eof()) - return false; - - buf.seek(0, SEEK_SET); /// Return position to the beginning - - DayNum tmp; - return tryReadDateText(tmp, buf) && buf.eof(); - } - - bool tryInferDateTime(std::string_view field, const FormatSettings & settings) - { - if (field.empty()) + if (std::all_of(field.begin(), field.end(), isNumericASCII)) return false; ReadBufferFromString buf(field); - Float64 tmp_float; + DayNum tmp; + return tryReadDateText(tmp, buf, DateLUT::instance(), /*allowed_delimiters=*/"-/:") && buf.eof(); + } + + DataTypePtr tryInferDateTimeOrDateTime64(std::string_view field, const FormatSettings & settings) + { + /// Don't try to infer DateTime if string is too long. + /// It's difficult to say what is the real maximum length of + /// DateTime we can parse using BestEffort approach. + /// 50 symbols is more or less valid limit for date times that makes sense. + if (field.empty() || field.size() > 50) + return nullptr; + + /// Check that we have at least one digit, don't infer datetime form strings like "Apr"/"May"/etc. + if (!std::any_of(field.begin(), field.end(), isNumericASCII)) + return nullptr; + /// Check if it's just a number, and if so, don't try to infer DateTime from it, /// because we can interpret this number as a timestamp and it will lead to - /// inferring DateTime instead of simple Int64/Float64 in some cases. + /// inferring DateTime instead of simple Int64 in some cases. + if (std::all_of(field.begin(), field.end(), isNumericASCII)) + return nullptr; + + ReadBufferFromString buf(field); + Float64 tmp_float; + /// Check if it's a float value, and if so, don't try to infer DateTime from it, + /// because it will lead to inferring DateTime instead of simple Float64 in some cases. if (tryReadFloatText(tmp_float, buf) && buf.eof()) - return false; + return nullptr; + + buf.seek(0, SEEK_SET); /// Return position to the beginning + if (!settings.try_infer_datetimes_only_datetime64) + { + time_t tmp; + switch (settings.date_time_input_format) + { + case FormatSettings::DateTimeInputFormat::Basic: + if (tryReadDateTimeText(tmp, buf, DateLUT::instance(), /*allowed_date_delimiters=*/"-/:", /*allowed_time_delimiters=*/":") && buf.eof()) + return std::make_shared(); + break; + case FormatSettings::DateTimeInputFormat::BestEffort: + if (tryParseDateTimeBestEffortStrict(tmp, buf, DateLUT::instance(), DateLUT::instance("UTC"), /*allowed_date_delimiters=*/"-/:") && buf.eof()) + return std::make_shared(); + break; + case FormatSettings::DateTimeInputFormat::BestEffortUS: + if (tryParseDateTimeBestEffortUSStrict(tmp, buf, DateLUT::instance(), DateLUT::instance("UTC"), /*allowed_date_delimiters=*/"-/:") && buf.eof()) + return std::make_shared(); + break; + } + } buf.seek(0, SEEK_SET); /// Return position to the beginning DateTime64 tmp; switch (settings.date_time_input_format) { case FormatSettings::DateTimeInputFormat::Basic: - if (tryReadDateTime64Text(tmp, 9, buf) && buf.eof()) - return true; + if (tryReadDateTime64Text(tmp, 9, buf, DateLUT::instance(), /*allowed_date_delimiters=*/"-/:", /*allowed_time_delimiters=*/":") && buf.eof()) + return std::make_shared(9); break; case FormatSettings::DateTimeInputFormat::BestEffort: - if (tryParseDateTime64BestEffort(tmp, 9, buf, DateLUT::instance(), DateLUT::instance("UTC")) && buf.eof()) - return true; + if (tryParseDateTime64BestEffortStrict(tmp, 9, buf, DateLUT::instance(), DateLUT::instance("UTC"), /*allowed_date_delimiters=*/"-/:") && buf.eof()) + return std::make_shared(9); break; case FormatSettings::DateTimeInputFormat::BestEffortUS: - if (tryParseDateTime64BestEffortUS(tmp, 9, buf, DateLUT::instance(), DateLUT::instance("UTC")) && buf.eof()) - return true; + if (tryParseDateTime64BestEffortUSStrict(tmp, 9, buf, DateLUT::instance(), DateLUT::instance("UTC"), /*allowed_date_delimiters=*/"-/:") && buf.eof()) + return std::make_shared(9); break; } - return false; + return nullptr; } template @@ -1439,8 +1479,11 @@ DataTypePtr tryInferDateOrDateTimeFromString(std::string_view field, const Forma if (settings.try_infer_dates && tryInferDate(field)) return std::make_shared(); - if (settings.try_infer_datetimes && tryInferDateTime(field, settings)) - return std::make_shared(9); + if (settings.try_infer_datetimes) + { + if (auto type = tryInferDateTimeOrDateTime64(field, settings)) + return type; + } return nullptr; } diff --git a/src/Formats/fuzzers/CMakeLists.txt b/src/Formats/fuzzers/CMakeLists.txt index 99e57bfbca3..b8a7e78b6e2 100644 --- a/src/Formats/fuzzers/CMakeLists.txt +++ b/src/Formats/fuzzers/CMakeLists.txt @@ -1,2 +1,2 @@ clickhouse_add_executable(format_fuzzer format_fuzzer.cpp ${SRCS}) -target_link_libraries(format_fuzzer PRIVATE dbms clickhouse_aggregate_functions clickhouse_functions) +target_link_libraries(format_fuzzer PRIVATE clickhouse_functions clickhouse_aggregate_functions) diff --git a/src/Formats/fuzzers/format_fuzzer.cpp b/src/Formats/fuzzers/format_fuzzer.cpp index 408e7218221..12cd40f9442 100644 --- a/src/Formats/fuzzers/format_fuzzer.cpp +++ b/src/Formats/fuzzers/format_fuzzer.cpp @@ -3,7 +3,6 @@ #include #include -#include #include #include diff --git a/src/Functions/CastOverloadResolver.cpp b/src/Functions/CastOverloadResolver.cpp index 49f63073aaf..6cb4d492fd8 100644 --- a/src/Functions/CastOverloadResolver.cpp +++ b/src/Functions/CastOverloadResolver.cpp @@ -3,6 +3,7 @@ #include #include #include +#include #include #include #include @@ -35,7 +36,7 @@ FunctionBasePtr createFunctionBaseCast( class CastOverloadResolverImpl : public IFunctionOverloadResolver { public: - const char * getNameImpl() const + static const char * getNameImpl(CastType cast_type, bool internal) { if (cast_type == CastType::accurate) return "accurateCast"; @@ -49,7 +50,7 @@ public: String getName() const override { - return getNameImpl(); + return getNameImpl(cast_type, internal); } size_t getNumberOfArguments() const override { return 2; } @@ -79,10 +80,22 @@ public: } } + static FunctionBasePtr createInternalCast(ColumnWithTypeAndName from, DataTypePtr to, CastType cast_type, std::optional diagnostic) + { + if (cast_type == CastType::accurateOrNull && !isVariant(to)) + to = makeNullable(to); + + ColumnsWithTypeAndName arguments; + arguments.emplace_back(std::move(from)); + arguments.emplace_back().type = std::make_unique(); + + return createFunctionBaseCast(nullptr, getNameImpl(cast_type, true), arguments, to, diagnostic, cast_type); + } + protected: FunctionBasePtr buildImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & return_type) const override { - return createFunctionBaseCast(context, getNameImpl(), arguments, return_type, diagnostic, cast_type); + return createFunctionBaseCast(context, getNameImpl(cast_type, internal), arguments, return_type, diagnostic, cast_type); } DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override @@ -130,9 +143,9 @@ private: }; -FunctionOverloadResolverPtr createInternalCastOverloadResolver(CastType type, std::optional diagnostic) +FunctionBasePtr createInternalCast(ColumnWithTypeAndName from, DataTypePtr to, CastType cast_type, std::optional diagnostic) { - return CastOverloadResolverImpl::create(ContextPtr{}, type, true, diagnostic); + return CastOverloadResolverImpl::createInternalCast(std::move(from), std::move(to), cast_type, std::move(diagnostic)); } REGISTER_FUNCTION(CastOverloadResolvers) diff --git a/src/Functions/CastOverloadResolver.h b/src/Functions/CastOverloadResolver.h index 7d98f774812..66f9d6cfcaf 100644 --- a/src/Functions/CastOverloadResolver.h +++ b/src/Functions/CastOverloadResolver.h @@ -3,6 +3,7 @@ #include #include #include +#include namespace DB @@ -11,6 +12,9 @@ namespace DB class IFunctionOverloadResolver; using FunctionOverloadResolverPtr = std::shared_ptr; +class IFunctionBase; +using FunctionBasePtr = std::shared_ptr; + enum class CastType : uint8_t { nonAccurate, @@ -24,6 +28,6 @@ struct CastDiagnostic std::string column_to; }; -FunctionOverloadResolverPtr createInternalCastOverloadResolver(CastType type, std::optional diagnostic); +FunctionBasePtr createInternalCast(ColumnWithTypeAndName from, DataTypePtr to, CastType cast_type, std::optional diagnostic); } diff --git a/src/Functions/DateTimeTransforms.h b/src/Functions/DateTimeTransforms.h index 6f0bdf406db..dfb4b76e5e2 100644 --- a/src/Functions/DateTimeTransforms.h +++ b/src/Functions/DateTimeTransforms.h @@ -1198,7 +1198,7 @@ struct ToYearImpl { if (point.getType() != Field::Types::UInt64) return std::nullopt; - auto year = point.get(); + auto year = point.safeGet(); if (year < DATE_LUT_MIN_YEAR || year >= DATE_LUT_MAX_YEAR) return std::nullopt; const DateLUTImpl & date_lut = DateLUT::instance("UTC"); @@ -2003,7 +2003,7 @@ struct ToYYYYMMImpl { if (point.getType() != Field::Types::UInt64) return std::nullopt; - auto year_month = point.get(); + auto year_month = point.safeGet(); auto year = year_month / 100; auto month = year_month % 100; diff --git a/src/Functions/FunctionsConsistentHashing.h b/src/Functions/FunctionsConsistentHashing.h index 306b6395dc5..210bb69e16d 100644 --- a/src/Functions/FunctionsConsistentHashing.h +++ b/src/Functions/FunctionsConsistentHashing.h @@ -101,9 +101,9 @@ private: BucketsType num_buckets; if (buckets_field.getType() == Field::Types::Int64) - num_buckets = checkBucketsRange(buckets_field.get()); + num_buckets = checkBucketsRange(buckets_field.safeGet()); else if (buckets_field.getType() == Field::Types::UInt64) - num_buckets = checkBucketsRange(buckets_field.get()); + num_buckets = checkBucketsRange(buckets_field.safeGet()); else throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Illegal type {} of the second argument of function {}", diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp index 675283d011e..0f54898177b 100644 --- a/src/Functions/FunctionsConversion.cpp +++ b/src/Functions/FunctionsConversion.cpp @@ -43,6 +43,7 @@ #include #include #include +#include #include #include #include @@ -117,7 +118,7 @@ UInt32 extractToDecimalScale(const ColumnWithTypeAndName & named_column) Field field; named_column.column->get(0, field); - return static_cast(field.get()); + return static_cast(field.safeGet()); } @@ -2604,8 +2605,8 @@ struct ToNumberMonotonicity if (left.isNull() || right.isNull()) return {}; - Float64 left_float = left.get(); - Float64 right_float = right.get(); + Float64 left_float = left.safeGet(); + Float64 right_float = right.safeGet(); if (left_float >= static_cast(std::numeric_limits::min()) && left_float <= static_cast(std::numeric_limits::max()) @@ -2633,11 +2634,11 @@ struct ToNumberMonotonicity const bool left_in_first_half = left.isNull() ? from_is_unsigned - : (left.get() >= 0); + : (left.safeGet() >= 0); const bool right_in_first_half = right.isNull() ? !from_is_unsigned - : (right.get() >= 0); + : (right.safeGet() >= 0); /// Size of type is the same. if (size_of_from == size_of_to) @@ -2675,7 +2676,7 @@ struct ToNumberMonotonicity return {}; /// Function cannot be monotonic when left and right are not on the same ranges. - if (divideByRangeOfType(left.get()) != divideByRangeOfType(right.get())) + if (divideByRangeOfType(left.safeGet()) != divideByRangeOfType(right.safeGet())) return {}; if (to_is_unsigned) @@ -2683,7 +2684,7 @@ struct ToNumberMonotonicity else { // If To is signed, it's possible that the signedness is different after conversion. So we check it explicitly. - const bool is_monotonic = (T(left.get()) >= 0) == (T(right.get()) >= 0); + const bool is_monotonic = (T(left.safeGet()) >= 0) == (T(right.safeGet()) >= 0); return { .is_monotonic = is_monotonic }; } @@ -2707,13 +2708,13 @@ struct ToDateMonotonicity } else if ( ((left.getType() == Field::Types::UInt64 || left.isNull()) && (right.getType() == Field::Types::UInt64 || right.isNull()) - && ((left.isNull() || left.get() < 0xFFFF) && (right.isNull() || right.get() >= 0xFFFF))) + && ((left.isNull() || left.safeGet() < 0xFFFF) && (right.isNull() || right.safeGet() >= 0xFFFF))) || ((left.getType() == Field::Types::Int64 || left.isNull()) && (right.getType() == Field::Types::Int64 || right.isNull()) - && ((left.isNull() || left.get() < 0xFFFF) && (right.isNull() || right.get() >= 0xFFFF))) + && ((left.isNull() || left.safeGet() < 0xFFFF) && (right.isNull() || right.safeGet() >= 0xFFFF))) || (( (left.getType() == Field::Types::Float64 || left.isNull()) && (right.getType() == Field::Types::Float64 || right.isNull()) - && ((left.isNull() || left.get() < 0xFFFF) && (right.isNull() || right.get() >= 0xFFFF)))) + && ((left.isNull() || left.safeGet() < 0xFFFF) && (right.isNull() || right.safeGet() >= 0xFFFF)))) || !isNativeNumber(type)) { return {}; @@ -2768,16 +2769,16 @@ struct ToStringMonotonicity if (left.getType() == Field::Types::UInt64 && right.getType() == Field::Types::UInt64) { - return (left.get() == 0 && right.get() == 0) - || (floor(log10(left.get())) == floor(log10(right.get()))) + return (left.safeGet() == 0 && right.safeGet() == 0) + || (floor(log10(left.safeGet())) == floor(log10(right.safeGet()))) ? positive : not_monotonic; } if (left.getType() == Field::Types::Int64 && right.getType() == Field::Types::Int64) { - return (left.get() == 0 && right.get() == 0) - || (left.get() > 0 && right.get() > 0 && floor(log10(left.get())) == floor(log10(right.get()))) + return (left.safeGet() == 0 && right.safeGet() == 0) + || (left.safeGet() > 0 && right.safeGet() > 0 && floor(log10(left.safeGet())) == floor(log10(right.safeGet()))) ? positive : not_monotonic; } @@ -4287,13 +4288,98 @@ private: WrapperType createDynamicToColumnWrapper(const DataTypePtr &) const { return [this] - (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * col_nullable, size_t input_rows_count) -> ColumnPtr + (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable *, size_t input_rows_count) -> ColumnPtr { + /// When casting Dynamic to regular column we should cast all variants from current Dynamic column + /// and construct the result based on discriminators. const auto & column_dynamic = assert_cast(*arguments.front().column.get()); + const auto & variant_column = column_dynamic.getVariantColumn(); const auto & variant_info = column_dynamic.getVariantInfo(); - auto variant_wrapper = createVariantToColumnWrapper(assert_cast(*variant_info.variant_type), result_type); - ColumnsWithTypeAndName args = {ColumnWithTypeAndName(column_dynamic.getVariantColumnPtr(), variant_info.variant_type, "")}; - return variant_wrapper(args, result_type, col_nullable, input_rows_count); + + /// First, cast usual variants to result type. + const auto & variant_types = assert_cast(*variant_info.variant_type).getVariants(); + std::vector casted_variant_columns; + casted_variant_columns.reserve(variant_types.size()); + for (size_t i = 0; i != variant_types.size(); ++i) + { + const auto & variant_col = variant_column.getVariantPtrByGlobalDiscriminator(i); + ColumnsWithTypeAndName variant = {{variant_col, variant_types[i], ""}}; + auto variant_wrapper = prepareUnpackDictionaries(variant_types[i], result_type); + casted_variant_columns.push_back(variant_wrapper(variant, result_type, nullptr, variant_col->size())); + } + + /// Second, collect all variants stored in shared variant and cast them to result type. + std::vector variant_columns_from_shared_variant; + DataTypes variant_types_from_shared_variant; + /// We will need to know what variant to use when we see discriminator of a shared variant. + /// To do it, we remember what variant was extracted from each row and what was it's offset. + PaddedPODArray shared_variant_indexes; + PaddedPODArray shared_variant_offsets; + std::unordered_map shared_variant_to_index; + const auto & shared_variant = column_dynamic.getSharedVariant(); + const auto shared_variant_discr = column_dynamic.getSharedVariantDiscriminator(); + const auto & local_discriminators = variant_column.getLocalDiscriminators(); + const auto & offsets = variant_column.getOffsets(); + if (!shared_variant.empty()) + { + shared_variant_indexes.reserve(input_rows_count); + shared_variant_offsets.reserve(input_rows_count); + FormatSettings format_settings; + const auto shared_variant_local_discr = variant_column.localDiscriminatorByGlobal(shared_variant_discr); + for (size_t i = 0; i != input_rows_count; ++i) + { + if (local_discriminators[i] == shared_variant_local_discr) + { + auto value = shared_variant.getDataAt(offsets[i]); + ReadBufferFromMemory buf(value.data, value.size); + auto type = decodeDataType(buf); + auto type_name = type->getName(); + auto it = shared_variant_to_index.find(type_name); + /// Check if we didn't create column for this variant yet. + if (it == shared_variant_to_index.end()) + { + it = shared_variant_to_index.emplace(type_name, variant_columns_from_shared_variant.size()).first; + variant_columns_from_shared_variant.push_back(type->createColumn()); + variant_types_from_shared_variant.push_back(type); + } + + shared_variant_indexes.push_back(it->second); + shared_variant_offsets.push_back(variant_columns_from_shared_variant[it->second]->size()); + type->getDefaultSerialization()->deserializeBinary(*variant_columns_from_shared_variant[it->second], buf, format_settings); + } + else + { + shared_variant_indexes.emplace_back(); + shared_variant_offsets.emplace_back(); + } + } + } + + /// Cast all extracted variants into result type. + std::vector casted_shared_variant_columns; + casted_shared_variant_columns.reserve(variant_types_from_shared_variant.size()); + for (size_t i = 0; i != variant_types_from_shared_variant.size(); ++i) + { + ColumnsWithTypeAndName variant = {{variant_columns_from_shared_variant[i]->getPtr(), variant_types_from_shared_variant[i], ""}}; + auto variant_wrapper = prepareUnpackDictionaries(variant_types_from_shared_variant[i], result_type); + casted_shared_variant_columns.push_back(variant_wrapper(variant, result_type, nullptr, variant_columns_from_shared_variant[i]->size())); + } + + /// Construct result column from all casted variants. + auto res = result_type->createColumn(); + res->reserve(input_rows_count); + for (size_t i = 0; i != input_rows_count; ++i) + { + auto global_discr = variant_column.globalDiscriminatorByLocal(local_discriminators[i]); + if (global_discr == ColumnVariant::NULL_DISCRIMINATOR) + res->insertDefault(); + else if (global_discr == shared_variant_discr) + res->insertFrom(*casted_shared_variant_columns[shared_variant_indexes[i]], shared_variant_offsets[i]); + else + res->insertFrom(*casted_variant_columns[global_discr], offsets[i]); + } + + return res; }; } @@ -4320,200 +4406,51 @@ private: }; } - std::pair getReducedVariant( - const ColumnVariant & variant_column, - const DataTypePtr & variant_type, - const std::unordered_map & variant_name_to_discriminator, - size_t max_result_num_variants, - const ColumnDynamic::Statistics & statistics = {}) const + WrapperType createVariantToDynamicWrapper(const DataTypeVariant & from_variant_type, const DataTypeDynamic & dynamic_type) const { - const auto & variant_types = assert_cast(*variant_type).getVariants(); - /// First check if we don't exceed the limit in current Variant column. - if (variant_types.size() < max_result_num_variants || (variant_types.size() == max_result_num_variants && variant_name_to_discriminator.contains("String"))) - return {variant_column.getPtr(), variant_type}; - - /// We want to keep the most frequent variants and convert to string the rarest. - std::vector> variant_sizes; - variant_sizes.reserve(variant_types.size()); - std::optional old_string_discriminator; - /// List of variants that should be converted to a single String variant. - std::vector variants_to_convert_to_string; - for (size_t i = 0; i != variant_types.size(); ++i) + /// First create extended Variant with shared variant type and cast this Variant to it. + auto variants_for_dynamic = from_variant_type.getVariants(); + size_t number_of_variants = variants_for_dynamic.size(); + variants_for_dynamic.push_back(ColumnDynamic::getSharedVariantDataType()); + const auto & variant_type_for_dynamic = std::make_shared(variants_for_dynamic); + auto old_to_new_variant_wrapper = createVariantToVariantWrapper(from_variant_type, *variant_type_for_dynamic); + auto max_dynamic_types = dynamic_type.getMaxDynamicTypes(); + return [old_to_new_variant_wrapper, variant_type_for_dynamic, number_of_variants, max_dynamic_types] + (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * col_nullable, size_t input_rows_count) -> ColumnPtr { - /// String variant won't be removed. - String variant_name = variant_types[i]->getName(); + auto variant_column_for_dynamic = old_to_new_variant_wrapper(arguments, result_type, col_nullable, input_rows_count); + /// If resulting Dynamic column can contain all variants from this Variant column, just create Dynamic column from it. + if (max_dynamic_types >= number_of_variants) + return ColumnDynamic::create(variant_column_for_dynamic, variant_type_for_dynamic, max_dynamic_types, max_dynamic_types); - if (variant_name == "String") - { - old_string_discriminator = i; - /// For simplicity, add this variant to the list that will be converted to string, - /// so we will process it with other variants when constructing the new String variant. - variants_to_convert_to_string.push_back(i); - } - else - { - size_t size = 0; - if (statistics.data.empty()) - size = variant_column.getVariantByGlobalDiscriminator(i).size(); - else - size = statistics.data.at(variant_name); - variant_sizes.emplace_back(size, i); - } - } - - /// Sort variants by sizes, so we will keep the most frequent. - std::sort(variant_sizes.begin(), variant_sizes.end(), std::greater()); - - DataTypes remaining_variants; - remaining_variants.reserve(max_result_num_variants); - /// Add String variant in advance. - remaining_variants.push_back(std::make_shared()); - for (auto [_, discr] : variant_sizes) - { - if (remaining_variants.size() != max_result_num_variants) - remaining_variants.push_back(variant_types[discr]); - else - variants_to_convert_to_string.push_back(discr); - } - - auto reduced_variant = std::make_shared(remaining_variants); - const auto & new_variants = reduced_variant->getVariants(); - /// To construct reduced variant column we will need mapping from old to new discriminators. - std::vector old_to_new_discriminators_mapping; - old_to_new_discriminators_mapping.resize(variant_types.size()); - ColumnVariant::Discriminator string_variant_discriminator = 0; - for (size_t i = 0; i != new_variants.size(); ++i) - { - String variant_name = new_variants[i]->getName(); - if (variant_name == "String") - { - string_variant_discriminator = i; - for (auto discr : variants_to_convert_to_string) - old_to_new_discriminators_mapping[discr] = i; - } - else - { - auto old_discr = variant_name_to_discriminator.at(variant_name); - old_to_new_discriminators_mapping[old_discr] = i; - } - } - - /// Convert all reduced variants to String. - std::unordered_map variants_converted_to_string; - variants_converted_to_string.reserve(variants_to_convert_to_string.size()); - size_t string_variant_size = 0; - for (auto discr : variants_to_convert_to_string) - { - auto string_type = std::make_shared(); - auto string_wrapper = prepareUnpackDictionaries(variant_types[discr], string_type); - auto column_to_convert = ColumnWithTypeAndName(variant_column.getVariantPtrByGlobalDiscriminator(discr), variant_types[discr], ""); - ColumnsWithTypeAndName args = {column_to_convert}; - auto variant_string_column = string_wrapper(args, string_type, nullptr, column_to_convert.column->size()); - string_variant_size += variant_string_column->size(); - variants_converted_to_string[discr] = variant_string_column; - } - - /// Create new discriminators and offsets and fill new String variant according to old discriminators. - auto string_variant = ColumnString::create(); - string_variant->reserve(string_variant_size); - auto new_discriminators_column = variant_column.getLocalDiscriminatorsPtr()->cloneEmpty(); - auto & new_discriminators_data = assert_cast(*new_discriminators_column).getData(); - new_discriminators_data.reserve(variant_column.size()); - auto new_offsets = variant_column.getOffsetsPtr()->cloneEmpty(); - auto & new_offsets_data = assert_cast(*new_offsets).getData(); - new_offsets_data.reserve(variant_column.size()); - const auto & old_local_discriminators = variant_column.getLocalDiscriminators(); - const auto & old_offsets = variant_column.getOffsets(); - for (size_t i = 0; i != old_local_discriminators.size(); ++i) - { - auto old_discr = variant_column.globalDiscriminatorByLocal(old_local_discriminators[i]); - - if (old_discr == ColumnVariant::NULL_DISCRIMINATOR) - { - new_discriminators_data.push_back(ColumnVariant::NULL_DISCRIMINATOR); - new_offsets_data.push_back(0); - continue; - } - - auto new_discr = old_to_new_discriminators_mapping[old_discr]; - new_discriminators_data.push_back(new_discr); - if (new_discr != string_variant_discriminator) - { - new_offsets_data.push_back(old_offsets[i]); - } - else - { - new_offsets_data.push_back(string_variant->size()); - string_variant->insertFrom(*variants_converted_to_string[old_discr], old_offsets[i]); - } - } - - /// Create new list of variant columns. - Columns new_variant_columns; - new_variant_columns.resize(new_variants.size()); - for (size_t i = 0; i != variant_types.size(); ++i) - { - auto new_discr = old_to_new_discriminators_mapping[i]; - if (new_discr != string_variant_discriminator) - new_variant_columns[new_discr] = variant_column.getVariantPtrByGlobalDiscriminator(i); - } - new_variant_columns[string_variant_discriminator] = std::move(string_variant); - return {ColumnVariant::create(std::move(new_discriminators_column), std::move(new_offsets), new_variant_columns), reduced_variant}; - } - - WrapperType createVariantToDynamicWrapper(const DataTypePtr & from_type, const DataTypeDynamic & dynamic_type) const - { - const auto & from_variant_type = assert_cast(*from_type); - size_t max_dynamic_types = dynamic_type.getMaxDynamicTypes(); - const auto & variants = from_variant_type.getVariants(); - std::unordered_map variant_name_to_discriminator; - variant_name_to_discriminator.reserve(variants.size()); - for (size_t i = 0; i != variants.size(); ++i) - variant_name_to_discriminator[variants[i]->getName()] = i; - - return [from_type, max_dynamic_types, variant_name_to_discriminator, this] - (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable *, size_t) -> ColumnPtr - { - const auto & variant_column = assert_cast(*arguments.front().column); - auto [reduced_variant_column, reduced_variant_type] = getReducedVariant(variant_column, from_type, variant_name_to_discriminator, max_dynamic_types); - return ColumnDynamic::create(reduced_variant_column, reduced_variant_type, max_dynamic_types); + /// Otherwise some variants should go to the shared variant. Create temporary Dynamic column from this Variant and insert + /// all data to the resulting Dynamic column, this insertion will do all the logic with shared variant. + auto tmp_dynamic_column = ColumnDynamic::create(variant_column_for_dynamic, variant_type_for_dynamic, number_of_variants, number_of_variants); + auto result_dynamic_column = ColumnDynamic::create(max_dynamic_types); + result_dynamic_column->insertRangeFrom(*tmp_dynamic_column, 0, tmp_dynamic_column->size()); + return result_dynamic_column; }; } WrapperType createColumnToDynamicWrapper(const DataTypePtr & from_type, const DataTypeDynamic & dynamic_type) const { if (const auto * variant_type = typeid_cast(from_type.get())) - return createVariantToDynamicWrapper(from_type, dynamic_type); - - if (dynamic_type.getMaxDynamicTypes() == 1) - { - DataTypePtr string_type = std::make_shared(); - if (from_type->isNullable()) - string_type = makeNullable(string_type); - auto string_wrapper = prepareUnpackDictionaries(from_type, string_type); - auto variant_type = std::make_shared(DataTypes{removeNullable(string_type)}); - auto variant_wrapper = createColumnToVariantWrapper(string_type, *variant_type); - return [string_wrapper, variant_wrapper, string_type, variant_type, max_dynamic_types=dynamic_type.getMaxDynamicTypes()] - (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * col_nullable, size_t input_rows_count) -> ColumnPtr - { - auto string_column = string_wrapper(arguments, string_type, col_nullable, input_rows_count); - auto column = ColumnWithTypeAndName(string_column, string_type, ""); - ColumnsWithTypeAndName args = {column}; - auto variant_column = variant_wrapper(args, variant_type, nullptr, string_column->size()); - return ColumnDynamic::create(variant_column, variant_type, max_dynamic_types); - }; - } + return createVariantToDynamicWrapper(*variant_type, dynamic_type); if (context && context->getSettingsRef().cast_string_to_dynamic_use_inference && isStringOrFixedString(removeNullable(removeLowCardinality(from_type)))) return createStringToDynamicThroughParsingWrapper(); + /// First, cast column to Variant with 2 variants - the type of the column we cast and shared variant type. auto variant_type = std::make_shared(DataTypes{removeNullableOrLowCardinalityNullable(from_type)}); - auto variant_wrapper = createColumnToVariantWrapper(from_type, *variant_type); - return [variant_wrapper, variant_type, max_dynamic_types=dynamic_type.getMaxDynamicTypes()] - (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable * col_nullable, size_t input_rows_count) -> ColumnPtr + auto column_to_variant_wrapper = createColumnToVariantWrapper(from_type, *variant_type); + /// Second, cast this Variant to Dynamic. + auto variant_to_dynamic_wrapper = createVariantToDynamicWrapper(*variant_type, dynamic_type); + return [column_to_variant_wrapper, variant_to_dynamic_wrapper, variant_type] + (ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * col_nullable, size_t input_rows_count) -> ColumnPtr { - auto variant_res = variant_wrapper(arguments, variant_type, col_nullable, input_rows_count); - return ColumnDynamic::create(variant_res, variant_type, max_dynamic_types); + auto variant_res = column_to_variant_wrapper(arguments, variant_type, col_nullable, input_rows_count); + ColumnsWithTypeAndName args = {{variant_res, variant_type, ""}}; + return variant_to_dynamic_wrapper(args, result_type, nullptr, input_rows_count); }; } @@ -4530,21 +4467,26 @@ private: (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable *, size_t) -> ColumnPtr { const auto & column_dynamic = assert_cast(*arguments[0].column); - return ColumnDynamic::create(column_dynamic.getVariantColumnPtr(), column_dynamic.getVariantInfo(), to_max_types); + /// We should use the same limit as already used in column and change only global limit. + /// It's needed because shared variant should contain values only when limit is exceeded, + /// so if there are already some data, we cannot increase the limit. + return ColumnDynamic::create(column_dynamic.getVariantColumnPtr(), column_dynamic.getVariantInfo(), column_dynamic.getMaxDynamicTypes(), to_max_types); }; } - return [to_max_types, this] + return [to_max_types] (ColumnsWithTypeAndName & arguments, const DataTypePtr &, const ColumnNullable *, size_t) -> ColumnPtr { const auto & column_dynamic = assert_cast(*arguments[0].column); - auto [reduced_variant_column, reduced_variant_type] = getReducedVariant( - column_dynamic.getVariantColumn(), - column_dynamic.getVariantInfo().variant_type, - column_dynamic.getVariantInfo().variant_name_to_discriminator, - to_max_types, - column_dynamic.getStatistics()); - return ColumnDynamic::create(reduced_variant_column, reduced_variant_type, to_max_types); + /// If real limit in the column is not greater than desired, just use the same variant column. + if (column_dynamic.getMaxDynamicTypes() <= to_max_types) + return ColumnDynamic::create(column_dynamic.getVariantColumnPtr(), column_dynamic.getVariantInfo(), column_dynamic.getMaxDynamicTypes(), to_max_types); + + /// Otherwise some variants should go to the shared variant. In this case we can just insert all + /// the data into resulting column and it will do all the logic with shared variant. + auto result_dynamic_column = ColumnDynamic::create(to_max_types); + result_dynamic_column->insertRangeFrom(column_dynamic, 0, column_dynamic.size()); + return result_dynamic_column; }; } @@ -4673,7 +4615,7 @@ private: return [function_name] ( ColumnsWithTypeAndName & arguments, const DataTypePtr & res_type, const ColumnNullable * nullable_col, size_t /*input_rows_count*/) { - using ColumnEnumType = EnumType::ColumnType; + using ColumnEnumType = typename EnumType::ColumnType; const auto & first_col = arguments.front().column.get(); const auto & first_type = arguments.front().type.get(); diff --git a/src/Functions/FunctionsJSON.cpp b/src/Functions/FunctionsJSON.cpp index c35df8ba72d..1eaf0d1609a 100644 --- a/src/Functions/FunctionsJSON.cpp +++ b/src/Functions/FunctionsJSON.cpp @@ -243,7 +243,7 @@ private: } case MoveType::Index: { - Int64 index = (*arguments[j + 1].column)[row].get(); + Int64 index = (*arguments[j + 1].column)[row].safeGet(); if (!moveToElementByIndex(res_element, static_cast(index), key)) return false; break; diff --git a/src/Functions/FunctionsLogical.cpp b/src/Functions/FunctionsLogical.cpp index 65d7473b945..ff0cff09c9e 100644 --- a/src/Functions/FunctionsLogical.cpp +++ b/src/Functions/FunctionsLogical.cpp @@ -701,11 +701,11 @@ ColumnPtr FunctionAnyArityLogical::getConstantResultForNonConstArgum bool constant_value_bool = false; if (field_type == Field::Types::Float64) - constant_value_bool = static_cast(constant_field_value.get()); + constant_value_bool = static_cast(constant_field_value.safeGet()); else if (field_type == Field::Types::Int64) - constant_value_bool = static_cast(constant_field_value.get()); + constant_value_bool = static_cast(constant_field_value.safeGet()); else if (field_type == Field::Types::UInt64) - constant_value_bool = static_cast(constant_field_value.get()); + constant_value_bool = static_cast(constant_field_value.safeGet()); has_true_constant = has_true_constant || constant_value_bool; has_false_constant = has_false_constant || !constant_value_bool; diff --git a/src/Functions/FunctionsRound.h b/src/Functions/FunctionsRound.h index 6b65a5feaec..ed7fe1a5de1 100644 --- a/src/Functions/FunctionsRound.h +++ b/src/Functions/FunctionsRound.h @@ -500,7 +500,7 @@ inline Scale getScaleArg(const ColumnConst* scale_col) { const auto & scale_field = scale_col->getField(); - Int64 scale64 = scale_field.get(); + Int64 scale64 = scale_field.safeGet(); validateScale(scale64); return scale64; @@ -632,7 +632,7 @@ public: Scale raw_scale = scale64; DecimalRoundingImpl::applyOne(value_col_typed->getElement(i), value_col_typed->getScale(), - reinterpret_cast::NativeT&>(col_res->getElement(i)), raw_scale); + reinterpret_cast::NativeT&>(col_res->getElement(i)), raw_scale); } } } @@ -854,7 +854,7 @@ private: using ValueType = typename Container::value_type; std::vector boundary_values(boundaries.size()); for (size_t i = 0; i < boundaries.size(); ++i) - boundary_values[i] = static_cast(boundaries[i].get()); + boundary_values[i] = static_cast(boundaries[i].safeGet()); ::sort(boundary_values.begin(), boundary_values.end()); boundary_values.erase(std::unique(boundary_values.begin(), boundary_values.end()), boundary_values.end()); diff --git a/src/Functions/FunctionsTimeWindow.cpp b/src/Functions/FunctionsTimeWindow.cpp index 77d740803be..88b85c48326 100644 --- a/src/Functions/FunctionsTimeWindow.cpp +++ b/src/Functions/FunctionsTimeWindow.cpp @@ -267,7 +267,12 @@ struct TimeWindowImpl { auto type = WhichDataType(arguments[0].type); if (type.isTuple()) - return std::static_pointer_cast(arguments[0].type)->getElement(0); + { + const auto & tuple_elems = std::static_pointer_cast(arguments[0].type)->getElements(); + if (tuple_elems.empty()) + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Tuple passed to {} should not be empty", function_name); + return tuple_elems[0]; + } else if (type.isUInt32()) return std::make_shared(); else @@ -622,7 +627,12 @@ struct TimeWindowImpl { auto type = WhichDataType(arguments[0].type); if (type.isTuple()) - return std::static_pointer_cast(arguments[0].type)->getElement(0); + { + const auto & tuple_elems = std::static_pointer_cast(arguments[0].type)->getElements(); + if (tuple_elems.empty()) + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Tuple passed to {} should not be empty", function_name); + return tuple_elems[0]; + } else if (type.isUInt32()) return std::make_shared(); else diff --git a/src/Functions/IFunctionCustomWeek.h b/src/Functions/IFunctionCustomWeek.h index 51542c9cab1..ba0baa35819 100644 --- a/src/Functions/IFunctionCustomWeek.h +++ b/src/Functions/IFunctionCustomWeek.h @@ -50,15 +50,15 @@ public: if (checkAndGetDataType(&type)) { - return Transform::FactorTransform::execute(UInt16(left.get()), date_lut) - == Transform::FactorTransform::execute(UInt16(right.get()), date_lut) + return Transform::FactorTransform::execute(UInt16(left.safeGet()), date_lut) + == Transform::FactorTransform::execute(UInt16(right.safeGet()), date_lut) ? is_monotonic : is_not_monotonic; } else { - return Transform::FactorTransform::execute(UInt32(left.get()), date_lut) - == Transform::FactorTransform::execute(UInt32(right.get()), date_lut) + return Transform::FactorTransform::execute(UInt32(left.safeGet()), date_lut) + == Transform::FactorTransform::execute(UInt32(right.safeGet()), date_lut) ? is_monotonic : is_not_monotonic; } diff --git a/src/Functions/IFunctionDateOrDateTime.h b/src/Functions/IFunctionDateOrDateTime.h index 762b79bfafc..899aa2c305d 100644 --- a/src/Functions/IFunctionDateOrDateTime.h +++ b/src/Functions/IFunctionDateOrDateTime.h @@ -72,30 +72,30 @@ public: if (checkAndGetDataType(type_ptr)) { - return Transform::FactorTransform::execute(UInt16(left.get()), *date_lut) - == Transform::FactorTransform::execute(UInt16(right.get()), *date_lut) + return Transform::FactorTransform::execute(UInt16(left.safeGet()), *date_lut) + == Transform::FactorTransform::execute(UInt16(right.safeGet()), *date_lut) ? is_monotonic : is_not_monotonic; } else if (checkAndGetDataType(type_ptr)) { - return Transform::FactorTransform::execute(Int32(left.get()), *date_lut) - == Transform::FactorTransform::execute(Int32(right.get()), *date_lut) + return Transform::FactorTransform::execute(Int32(left.safeGet()), *date_lut) + == Transform::FactorTransform::execute(Int32(right.safeGet()), *date_lut) ? is_monotonic : is_not_monotonic; } else if (checkAndGetDataType(type_ptr)) { - return Transform::FactorTransform::execute(UInt32(left.get()), *date_lut) - == Transform::FactorTransform::execute(UInt32(right.get()), *date_lut) + return Transform::FactorTransform::execute(UInt32(left.safeGet()), *date_lut) + == Transform::FactorTransform::execute(UInt32(right.safeGet()), *date_lut) ? is_monotonic : is_not_monotonic; } else { assert(checkAndGetDataType(type_ptr)); - const auto & left_date_time = left.get(); + const auto & left_date_time = left.safeGet(); TransformDateTime64 transformer_left(left_date_time.getScale()); - const auto & right_date_time = right.get(); + const auto & right_date_time = right.safeGet(); TransformDateTime64 transformer_right(right_date_time.getScale()); return transformer_left.execute(left_date_time.getValue(), *date_lut) diff --git a/src/Functions/JSONPath/Parsers/ParserJSONPathRange.cpp b/src/Functions/JSONPath/Parsers/ParserJSONPathRange.cpp index fb74018b330..84ac0ff08f3 100644 --- a/src/Functions/JSONPath/Parsers/ParserJSONPathRange.cpp +++ b/src/Functions/JSONPath/Parsers/ParserJSONPathRange.cpp @@ -46,7 +46,7 @@ bool ParserJSONPathRange::parseImpl(Pos & pos, ASTPtr & node, Expected & expecte { return false; } - range_indices.first = static_cast(number_ptr->as()->value.get()); + range_indices.first = static_cast(number_ptr->as()->value.safeGet()); if (pos->type == TokenType::Comma || pos->type == TokenType::ClosingSquareBracket) { @@ -63,7 +63,7 @@ bool ParserJSONPathRange::parseImpl(Pos & pos, ASTPtr & node, Expected & expecte { return false; } - range_indices.second = static_cast(number_ptr->as()->value.get()); + range_indices.second = static_cast(number_ptr->as()->value.safeGet()); } else { diff --git a/src/Functions/MultiMatchAllIndicesImpl.h b/src/Functions/MultiMatchAllIndicesImpl.h index 3aeac808880..e7c3aebf794 100644 --- a/src/Functions/MultiMatchAllIndicesImpl.h +++ b/src/Functions/MultiMatchAllIndicesImpl.h @@ -77,7 +77,7 @@ struct MultiMatchAllIndicesImpl std::vector needles; needles.reserve(needles_arr.size()); for (const auto & needle : needles_arr) - needles.emplace_back(needle.get()); + needles.emplace_back(needle.safeGet()); checkHyperscanRegexp(needles, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length); diff --git a/src/Functions/MultiMatchAnyImpl.h b/src/Functions/MultiMatchAnyImpl.h index fda752cbacc..54413cbc1cd 100644 --- a/src/Functions/MultiMatchAnyImpl.h +++ b/src/Functions/MultiMatchAnyImpl.h @@ -91,7 +91,7 @@ struct MultiMatchAnyImpl std::vector needles; needles.reserve(needles_arr.size()); for (const auto & needle : needles_arr) - needles.emplace_back(needle.get()); + needles.emplace_back(needle.safeGet()); checkHyperscanRegexp(needles, max_hyperscan_regexp_length, max_hyperscan_regexp_total_length); diff --git a/src/Functions/MultiSearchAllPositionsImpl.h b/src/Functions/MultiSearchAllPositionsImpl.h index cfe60e51bcd..6c2cd215638 100644 --- a/src/Functions/MultiSearchAllPositionsImpl.h +++ b/src/Functions/MultiSearchAllPositionsImpl.h @@ -33,7 +33,7 @@ struct MultiSearchAllPositionsImpl std::vector needles; needles.reserve(needles_arr.size()); for (const auto & needle : needles_arr) - needles.emplace_back(needle.get()); + needles.emplace_back(needle.safeGet()); auto res_callback = [](const UInt8 * start, const UInt8 * end) -> UInt64 { diff --git a/src/Functions/MultiSearchFirstIndexImpl.h b/src/Functions/MultiSearchFirstIndexImpl.h index 5b34dbfe601..f1dc9ab9e11 100644 --- a/src/Functions/MultiSearchFirstIndexImpl.h +++ b/src/Functions/MultiSearchFirstIndexImpl.h @@ -45,7 +45,7 @@ struct MultiSearchFirstIndexImpl std::vector needles; needles.reserve(needles_arr.size()); for (const auto & needle : needles_arr) - needles.emplace_back(needle.get()); + needles.emplace_back(needle.safeGet()); auto searcher = Impl::createMultiSearcherInBigHaystack(needles); diff --git a/src/Functions/MultiSearchFirstPositionImpl.h b/src/Functions/MultiSearchFirstPositionImpl.h index 06bf7aa94d8..4380eeb1b29 100644 --- a/src/Functions/MultiSearchFirstPositionImpl.h +++ b/src/Functions/MultiSearchFirstPositionImpl.h @@ -45,7 +45,7 @@ struct MultiSearchFirstPositionImpl std::vector needles; needles.reserve(needles_arr.size()); for (const auto & needle : needles_arr) - needles.emplace_back(needle.get()); + needles.emplace_back(needle.safeGet()); auto res_callback = [](const UInt8 * start, const UInt8 * end) -> UInt64 { diff --git a/src/Functions/MultiSearchImpl.h b/src/Functions/MultiSearchImpl.h index 909425f5a93..5c652ddcb74 100644 --- a/src/Functions/MultiSearchImpl.h +++ b/src/Functions/MultiSearchImpl.h @@ -45,7 +45,7 @@ struct MultiSearchImpl std::vector needles; needles.reserve(needles_arr.size()); for (const auto & needle : needles_arr) - needles.emplace_back(needle.get()); + needles.emplace_back(needle.safeGet()); auto searcher = Impl::createMultiSearcherInBigHaystack(needles); diff --git a/src/Functions/URL/cutURLParameter.cpp b/src/Functions/URL/cutURLParameter.cpp index 3ab9cad1ea7..4439e79e962 100644 --- a/src/Functions/URL/cutURLParameter.cpp +++ b/src/Functions/URL/cutURLParameter.cpp @@ -156,7 +156,7 @@ public: for (size_t j = 0; j < num_needles; ++j) { auto field = col_needle_const_array->getData()[j]; - cutURL(res_data, field.get(), res_offset, cur_res_offset); + cutURL(res_data, field.safeGet(), res_offset, cur_res_offset); } } else diff --git a/src/Functions/array/arrayElement.cpp b/src/Functions/array/arrayElement.cpp index 227b29d5d9f..81f3f97979b 100644 --- a/src/Functions/array/arrayElement.cpp +++ b/src/Functions/array/arrayElement.cpp @@ -904,10 +904,10 @@ ColumnPtr FunctionArrayElement::executeNumberConst( return nullptr; if (index.getType() == Field::Types::UInt64 - || (index.getType() == Field::Types::Int64 && index.get() >= 0)) + || (index.getType() == Field::Types::Int64 && index.safeGet() >= 0)) { ArrayElementNumImpl::template vectorConst( - col_nested->getData(), col_array->getOffsets(), index.get() - 1, col_res_vec->getData(), builder); + col_nested->getData(), col_array->getOffsets(), index.safeGet() - 1, col_res_vec->getData(), builder); } else if (index.getType() == Field::Types::Int64) { @@ -972,14 +972,14 @@ FunctionArrayElement::executeStringConst(const ColumnsWithTypeAndName & argument auto col_res = ColumnString::create(); if (index.getType() == Field::Types::UInt64 - || (index.getType() == Field::Types::Int64 && index.get() >= 0)) + || (index.getType() == Field::Types::Int64 && index.safeGet() >= 0)) { if (builder) ArrayElementStringImpl::vectorConst( col_nested->getChars(), col_array->getOffsets(), col_nested->getOffsets(), - index.get() - 1, + index.safeGet() - 1, col_res->getChars(), col_res->getOffsets(), builder); @@ -988,7 +988,7 @@ FunctionArrayElement::executeStringConst(const ColumnsWithTypeAndName & argument col_nested->getChars(), col_array->getOffsets(), col_nested->getOffsets(), - index.get() - 1, + index.safeGet() - 1, col_res->getChars(), col_res->getOffsets(), builder); @@ -1000,7 +1000,7 @@ FunctionArrayElement::executeStringConst(const ColumnsWithTypeAndName & argument col_nested->getChars(), col_array->getOffsets(), col_nested->getOffsets(), - -(UInt64(index.get()) + 1), + -(UInt64(index.safeGet()) + 1), col_res->getChars(), col_res->getOffsets(), builder); @@ -1009,7 +1009,7 @@ FunctionArrayElement::executeStringConst(const ColumnsWithTypeAndName & argument col_nested->getChars(), col_array->getOffsets(), col_nested->getOffsets(), - -(UInt64(index.get()) + 1), + -(UInt64(index.safeGet()) + 1), col_res->getChars(), col_res->getOffsets(), builder); @@ -1046,7 +1046,7 @@ ColumnPtr FunctionArrayElement::executeArrayStringConst( auto res_offsets = ColumnArray::ColumnOffsets::create(); auto res_string_null_map = col_nullable ? ColumnUInt8::create() : nullptr; if (index.getType() == Field::Types::UInt64 - || (index.getType() == Field::Types::Int64 && index.get() >= 0)) + || (index.getType() == Field::Types::Int64 && index.safeGet() >= 0)) { if (col_nullable) ArrayElementArrayStringImpl::vectorConst( @@ -1055,7 +1055,7 @@ ColumnPtr FunctionArrayElement::executeArrayStringConst( col_nested_array->getOffsets(), col_nested_elem->getOffsets(), &string_null_map->getData(), - index.get() - 1, + index.safeGet() - 1, res_string->getChars(), res_offsets->getData(), res_string->getOffsets(), @@ -1068,7 +1068,7 @@ ColumnPtr FunctionArrayElement::executeArrayStringConst( col_nested_array->getOffsets(), col_nested_elem->getOffsets(), nullptr, - index.get() - 1, + index.safeGet() - 1, res_string->getChars(), res_offsets->getData(), res_string->getOffsets(), @@ -1084,7 +1084,7 @@ ColumnPtr FunctionArrayElement::executeArrayStringConst( col_nested_array->getOffsets(), col_nested_elem->getOffsets(), &string_null_map->getData(), - -(UInt64(index.get()) + 1), + -(UInt64(index.safeGet()) + 1), res_string->getChars(), res_offsets->getData(), res_string->getOffsets(), @@ -1097,7 +1097,7 @@ ColumnPtr FunctionArrayElement::executeArrayStringConst( col_nested_array->getOffsets(), col_nested_elem->getOffsets(), nullptr, - -(UInt64(index.get()) + 1), + -(UInt64(index.safeGet()) + 1), res_string->getChars(), res_offsets->getData(), res_string->getOffsets(), @@ -1153,7 +1153,7 @@ ColumnPtr FunctionArrayElement::executeArrayNumberConst( auto & res_offsets = res_array->getOffsets(); NullMap * res_null_map = res_nullable ? &res_nullable->getNullMapData() : nullptr; - if (index.getType() == Field::Types::UInt64 || (index.getType() == Field::Types::Int64 && index.get() >= 0)) + if (index.getType() == Field::Types::UInt64 || (index.getType() == Field::Types::Int64 && index.safeGet() >= 0)) { if (col_nullable) ArrayElementArrayNumImpl::template vectorConst( @@ -1161,7 +1161,7 @@ ColumnPtr FunctionArrayElement::executeArrayNumberConst( col_array->getOffsets(), col_nested_array->getOffsets(), null_map, - index.get() - 1, + index.safeGet() - 1, res_data->getData(), res_offsets, res_null_map, @@ -1172,7 +1172,7 @@ ColumnPtr FunctionArrayElement::executeArrayNumberConst( col_array->getOffsets(), col_nested_array->getOffsets(), null_map, - index.get() - 1, + index.safeGet() - 1, res_data->getData(), res_offsets, res_null_map, @@ -1392,12 +1392,12 @@ ColumnPtr FunctionArrayElement::executeGenericConst( auto col_res = col_nested.cloneEmpty(); if (index.getType() == Field::Types::UInt64 - || (index.getType() == Field::Types::Int64 && index.get() >= 0)) + || (index.getType() == Field::Types::Int64 && index.safeGet() >= 0)) ArrayElementGenericImpl::vectorConst( - col_nested, col_array->getOffsets(), index.get() - 1, *col_res, builder); + col_nested, col_array->getOffsets(), index.safeGet() - 1, *col_res, builder); else if (index.getType() == Field::Types::Int64) ArrayElementGenericImpl::vectorConst( - col_nested, col_array->getOffsets(), -(static_cast(index.get() + 1)), *col_res, builder); + col_nested, col_array->getOffsets(), -(static_cast(index.safeGet() + 1)), *col_res, builder); else throw Exception(ErrorCodes::LOGICAL_ERROR, "Illegal type of array index"); @@ -1789,7 +1789,7 @@ bool FunctionArrayElement::matchKeyToIndexStringConst( using DataColumn = std::decay_t; if (index.getType() != Field::Types::String) return false; - MatcherStringConst matcher{data_column, index.get()}; + MatcherStringConst matcher{data_column, index.safeGet()}; executeMatchKeyToIndex(offsets, matched_idxs, matcher); return true; }); diff --git a/src/Functions/array/mapOp.cpp b/src/Functions/array/mapOp.cpp index 86797cb5db0..614b01c2ac8 100644 --- a/src/Functions/array/mapOp.cpp +++ b/src/Functions/array/mapOp.cpp @@ -237,7 +237,7 @@ private: } arg.val_column->get(offset + j, temp_val); - ValType value = temp_val.get(); + ValType value = temp_val.safeGet(); if constexpr (op_type == OpTypes::ADD) { diff --git a/src/Functions/bitSlice.cpp b/src/Functions/bitSlice.cpp index f24473351ae..908c534b228 100644 --- a/src/Functions/bitSlice.cpp +++ b/src/Functions/bitSlice.cpp @@ -42,11 +42,11 @@ public: { FunctionArgumentDescriptors mandatory_args{ {"s", static_cast(&isStringOrFixedString), nullptr, "String"}, - {"offset", static_cast(&isNativeNumber), nullptr, "(U)Int8, (U)Int16, (U)Int32, (U)Int64 or Float"}, + {"offset", static_cast(&isNativeNumber), nullptr, "(U)Int8/16/32/64 or Float"}, }; FunctionArgumentDescriptors optional_args{ - {"length", static_cast(&isNativeNumber), nullptr, "(U)Int8, (U)Int16, (U)Int32, (U)Int64 or Float"}, + {"length", static_cast(&isNativeNumber), nullptr, "(U)Int8/16/32/64 or Float"}, }; validateFunctionArguments(*this, arguments, mandatory_args, optional_args); diff --git a/src/Functions/dynamicType.cpp b/src/Functions/dynamicType.cpp index e8ca73597d6..327cdfe1616 100644 --- a/src/Functions/dynamicType.cpp +++ b/src/Functions/dynamicType.cpp @@ -2,10 +2,14 @@ #include #include #include +#include +#include #include #include #include #include +#include +#include #include @@ -65,11 +69,15 @@ public: const auto & variant_column = dynamic_column->getVariantColumn(); auto res = result_type->createColumn(); String element_type; + auto shared_variant_discr = dynamic_column->getSharedVariantDiscriminator(); + const auto & shared_variant = dynamic_column->getSharedVariant(); for (size_t i = 0; i != input_rows_count; ++i) { auto global_discr = variant_column.globalDiscriminatorAt(i); if (global_discr == ColumnVariant::NULL_DISCRIMINATOR) element_type = name_for_null; + else if (global_discr == shared_variant_discr) + element_type = getTypeNameFromSharedVariantValue(shared_variant.getDataAt(variant_column.offsetAt(i))); else element_type = variant_info.variant_names[global_discr]; @@ -78,6 +86,63 @@ public: return res; } + + String getTypeNameFromSharedVariantValue(StringRef value) const + { + ReadBufferFromMemory buf(value.data, value.size); + return decodeDataType(buf)->getName(); + } +}; + +class FunctionIsDynamicElementInSharedData : public IFunction +{ +public: + static constexpr auto name = "isDynamicElementInSharedData"; + + static FunctionPtr create(ContextPtr) { return std::make_shared(); } + String getName() const override { return name; } + size_t getNumberOfArguments() const override { return 1; } + bool useDefaultImplementationForConstants() const override { return true; } + bool useDefaultImplementationForNulls() const override { return false; } + bool useDefaultImplementationForLowCardinalityColumns() const override { return false; } + bool isSuitableForShortCircuitArgumentsExecution(const DataTypesWithConstInfo & /*arguments*/) const override { return true; } + + DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override + { + if (arguments.empty() || arguments.size() > 1) + throw Exception( + ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, + "Number of arguments for function {} doesn't match: passed {}, should be 1", + getName(), arguments.empty()); + + if (!isDynamic(arguments[0].type.get())) + throw Exception( + ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + "First argument for function {} must be Dynamic, got {} instead", + getName(), arguments[0].type->getName()); + + return DataTypeFactory::instance().get("Bool"); + } + + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, size_t input_rows_count) const override + { + const ColumnDynamic * dynamic_column = checkAndGetColumn(arguments[0].column.get()); + if (!dynamic_column) + throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, + "First argument for function {} must be Dynamic, got {} instead", + getName(), arguments[0].type->getName()); + + const auto & variant_column = dynamic_column->getVariantColumn(); + const auto & local_discriminators = variant_column.getLocalDiscriminators(); + auto res = result_type->createColumn(); + auto & res_data = assert_cast(*res).getData(); + res_data.reserve(dynamic_column->size()); + auto shared_variant_local_discr = variant_column.localDiscriminatorByGlobal(dynamic_column->getSharedVariantDiscriminator()); + for (size_t i = 0; i != input_rows_count; ++i) + res_data.push_back(local_discriminators[i] == shared_variant_local_discr); + + return res; + } }; } @@ -88,7 +153,7 @@ REGISTER_FUNCTION(DynamicType) .description = R"( Returns the variant type name for each row of `Dynamic` column. If row contains NULL, it returns 'None' for it. )", - .syntax = {"dynamicType(variant)"}, + .syntax = {"dynamicType(dynamic)"}, .arguments = {{"dynamic", "Dynamic column"}}, .examples = {{{ "Example", @@ -104,6 +169,30 @@ SELECT d, dynamicType(d) FROM test; ā”‚ Hello, World! ā”‚ String ā”‚ ā”‚ [1,2,3] ā”‚ Array(Int64) ā”‚ ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ +)"}}}, + .categories{"Variant"}, + }); + + factory.registerFunction(FunctionDocumentation{ + .description = R"( +Returns true for rows in Dynamic column that are not separated into subcolumns and stored inside shared variant in binary form. +)", + .syntax = {"isDynamicElementInSharedData(dynamic)"}, + .arguments = {{"dynamic", "Dynamic column"}}, + .examples = {{{ + "Example", + R"( +CREATE TABLE test (d Dynamic(max_types=2)) ENGINE = Memory; +INSERT INTO test VALUES (NULL), (42), ('Hello, World!'), ([1, 2, 3]); +SELECT d, isDynamicElementInSharedData(d) FROM test; +)", + R"( +ā”Œā”€dā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€isDynamicElementInSharedData(d)ā”€ā” +ā”‚ į“ŗįµį“øį“ø ā”‚ false ā”‚ +ā”‚ 42 ā”‚ false ā”‚ +ā”‚ Hello, World! ā”‚ true ā”‚ +ā”‚ [1,2,3] ā”‚ true ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”“ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ )"}}}, .categories{"Variant"}, }); diff --git a/src/Functions/geometryConverters.h b/src/Functions/geometryConverters.h index 03831d37e0c..f1156d81f01 100644 --- a/src/Functions/geometryConverters.h +++ b/src/Functions/geometryConverters.h @@ -31,6 +31,9 @@ namespace ErrorCodes template using LineString = boost::geometry::model::linestring; +template +using MultiLineString = boost::geometry::model::multi_linestring>; + template using Ring = boost::geometry::model::ring; @@ -42,12 +45,14 @@ using MultiPolygon = boost::geometry::model::multi_polygon>; using CartesianPoint = boost::geometry::model::d2::point_xy; using CartesianLineString = LineString; +using CartesianMultiLineString = MultiLineString; using CartesianRing = Ring; using CartesianPolygon = Polygon; using CartesianMultiPolygon = MultiPolygon; using SphericalPoint = boost::geometry::model::point>; using SphericalLineString = LineString; +using SphericalMultiLineString = MultiLineString; using SphericalRing = Ring; using SphericalPolygon = Polygon; using SphericalMultiPolygon = MultiPolygon; @@ -113,6 +118,28 @@ struct ColumnToLineStringsConverter } }; +/** + * Class which converts Column with type Array(Array(Tuple(Float64, Float64))) to a vector of boost multi_linestring type. +*/ +template +struct ColumnToMultiLineStringsConverter +{ + static std::vector> convert(ColumnPtr col) + { + const IColumn::Offsets & offsets = typeid_cast(*col).getOffsets(); + size_t prev_offset = 0; + std::vector> answer(offsets.size()); + auto all_linestrings = ColumnToLineStringsConverter::convert(typeid_cast(*col).getDataPtr()); + for (size_t iter = 0; iter < offsets.size() && iter < all_linestrings.size(); ++iter) + { + for (size_t linestring_iter = prev_offset; linestring_iter < offsets[iter]; ++linestring_iter) + answer[iter].emplace_back(std::move(all_linestrings[linestring_iter])); + prev_offset = offsets[iter]; + } + return answer; + } +}; + /** * Class which converts Column with type Array(Tuple(Float64, Float64)) to a vector of boost ring type. */ @@ -268,6 +295,38 @@ private: ColumnUInt64::MutablePtr offsets; }; +/// Serialize Point, MultiLineString as MultiLineString +template +class MultiLineStringSerializer +{ +public: + MultiLineStringSerializer() + : offsets(ColumnUInt64::create()) + {} + + explicit MultiLineStringSerializer(size_t n) + : offsets(ColumnUInt64::create(n)) + {} + + void add(const MultiLineString & multilinestring) + { + size += multilinestring.size(); + offsets->insertValue(size); + for (const auto & linestring : multilinestring) + linestring_serializer.add(linestring); + } + + ColumnPtr finalize() + { + return ColumnArray::create(linestring_serializer.finalize(), std::move(offsets)); + } + +private: + size_t size = 0; + LineStringSerializer linestring_serializer; + ColumnUInt64::MutablePtr offsets; +}; + /// Almost the same as LineStringSerializer /// Serialize Point, Ring as Ring template @@ -411,6 +470,11 @@ static void callOnGeometryDataType(DataTypePtr type, F && f) else if (factory.get("LineString")->equals(*type) && type->getCustomName() && type->getCustomName()->getName() == "LineString") return f(ConverterType>()); + /// We should take the name into consideration to avoid ambiguity. + /// Because for example both MultiLineString and Polygon are resolved to Array(Array(Point)). + else if (factory.get("MultiLineString")->equals(*type) && type->getCustomName() && type->getCustomName()->getName() == "MultiLineString") + return f(ConverterType>()); + /// For backward compatibility if we call this function not on a custom type, we will consider Array(Tuple(Point)) as type Ring. else if (factory.get("Ring")->equals(*type)) return f(ConverterType>()); diff --git a/src/Functions/getClientHTTPHeader.cpp b/src/Functions/getClientHTTPHeader.cpp index 140f39d03b8..50a6275fc82 100644 --- a/src/Functions/getClientHTTPHeader.cpp +++ b/src/Functions/getClientHTTPHeader.cpp @@ -58,7 +58,7 @@ public: { Field header; source->get(row, header); - if (auto it = client_info.http_headers.find(header.get()); it != client_info.http_headers.end()) + if (auto it = client_info.http_headers.find(header.safeGet()); it != client_info.http_headers.end()) result->insert(it->second); else result->insertDefault(); diff --git a/src/Functions/multiIf.cpp b/src/Functions/multiIf.cpp index c4b675fcf6c..14b8b70b22c 100644 --- a/src/Functions/multiIf.cpp +++ b/src/Functions/multiIf.cpp @@ -200,7 +200,7 @@ public: if (value.isNull()) continue; - if (value.get() == 0) + if (value.safeGet() == 0) continue; instruction.condition_always_true = true; diff --git a/src/Functions/nested.cpp b/src/Functions/nested.cpp index bdaf57d65c9..85c342b5e7c 100644 --- a/src/Functions/nested.cpp +++ b/src/Functions/nested.cpp @@ -145,7 +145,7 @@ private: if (nested_names_field.getType() != Field::Types::Array) return {}; - const auto & nested_names_array = nested_names_field.get(); + const auto & nested_names_array = nested_names_field.safeGet(); Names nested_names; nested_names.reserve(nested_names_array.size()); @@ -155,7 +155,7 @@ private: if (nested_name_field.getType() != Field::Types::String) return {}; - nested_names.push_back(nested_name_field.get()); + nested_names.push_back(nested_name_field.safeGet()); } return nested_names; diff --git a/src/Functions/polygonsIntersection.cpp b/src/Functions/polygonsIntersection.cpp index 329242e762e..43ab03f8c1f 100644 --- a/src/Functions/polygonsIntersection.cpp +++ b/src/Functions/polygonsIntersection.cpp @@ -75,6 +75,8 @@ public: throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be Point", getName()); else if constexpr (std::is_same_v, LeftConverter> || std::is_same_v, RightConverter>) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be LineString", getName()); + else if constexpr (std::is_same_v, LeftConverter> || std::is_same_v, RightConverter>) + throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be MultiLineString", getName()); else { auto first = LeftConverter::convert(arguments[0].column->convertToFullColumnIfConst()); diff --git a/src/Functions/polygonsSymDifference.cpp b/src/Functions/polygonsSymDifference.cpp index 3c219d0facb..6faec95bb7b 100644 --- a/src/Functions/polygonsSymDifference.cpp +++ b/src/Functions/polygonsSymDifference.cpp @@ -73,6 +73,8 @@ public: throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be Point", getName()); else if constexpr (std::is_same_v, LeftConverter> || std::is_same_v, RightConverter>) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be LineString", getName()); + else if constexpr (std::is_same_v, LeftConverter> || std::is_same_v, RightConverter>) + throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be MultiLineString", getName()); else { auto first = LeftConverter::convert(arguments[0].column->convertToFullColumnIfConst()); diff --git a/src/Functions/polygonsUnion.cpp b/src/Functions/polygonsUnion.cpp index 969eb2f78fb..5378ff636f8 100644 --- a/src/Functions/polygonsUnion.cpp +++ b/src/Functions/polygonsUnion.cpp @@ -73,6 +73,8 @@ public: throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be Point", getName()); else if constexpr (std::is_same_v, LeftConverter> || std::is_same_v, RightConverter>) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be LineString", getName()); + else if constexpr (std::is_same_v, LeftConverter> || std::is_same_v, RightConverter>) + throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be MultiLineString", getName()); else { auto first = LeftConverter::convert(arguments[0].column->convertToFullColumnIfConst()); diff --git a/src/Functions/polygonsWithin.cpp b/src/Functions/polygonsWithin.cpp index c63ad5ef868..dacd1c0e18f 100644 --- a/src/Functions/polygonsWithin.cpp +++ b/src/Functions/polygonsWithin.cpp @@ -77,6 +77,8 @@ public: throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be Point", getName()); else if constexpr (std::is_same_v, LeftConverter> || std::is_same_v, RightConverter>) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be LineString", getName()); + else if constexpr (std::is_same_v, LeftConverter> || std::is_same_v, RightConverter>) + throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Any argument of function {} must not be MultiLineString", getName()); else { auto first = LeftConverter::convert(arguments[0].column->convertToFullColumnIfConst()); diff --git a/src/Functions/readWkt.cpp b/src/Functions/readWkt.cpp index eb262777b0d..2010b5167e7 100644 --- a/src/Functions/readWkt.cpp +++ b/src/Functions/readWkt.cpp @@ -87,6 +87,11 @@ struct ReadWKTLineStringNameHolder static constexpr const char * name = "readWKTLineString"; }; +struct ReadWKTMultiLineStringNameHolder +{ + static constexpr const char * name = "readWKTMultiLineString"; +}; + struct ReadWKTRingNameHolder { static constexpr const char * name = "readWKTRing"; @@ -131,6 +136,31 @@ Parses a Well-Known Text (WKT) representation of a LineString geometry and retur }, .categories{"Unique identifiers"} }); + factory.registerFunction, ReadWKTMultiLineStringNameHolder>>(FunctionDocumentation + { + .description=R"( +Parses a Well-Known Text (WKT) representation of a MultiLineString geometry and returns it in the internal ClickHouse format. +)", + .syntax = "readWKTMultiLineString(wkt_string)", + .arguments{ + {"wkt_string", "The input WKT string representing a MultiLineString geometry."} + }, + .returned_value = "The function returns a ClickHouse internal representation of the multilinestring geometry.", + .examples{ + {"first call", "SELECT readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3), (4 4, 5 5, 6 6))');", R"( +ā”Œā”€readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3), (4 4, 5 5, 6 6))')ā”€ā” +ā”‚ [[(1,1),(2,2),(3,3)],[(4,4),(5,5),(6,6)]] ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ + + )"}, + {"second call", "SELECT toTypeName(readWKTLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))'));", R"( +ā”Œā”€toTypeName(readWKTLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))'))ā”€ā” +ā”‚ MultiLineString ā”‚ +ā””ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”˜ + )"}, + }, + .categories{"Unique identifiers"} + }); factory.registerFunction, ReadWKTRingNameHolder>>(); factory.registerFunction, ReadWKTPolygonNameHolder>>(); factory.registerFunction, ReadWKTMultiPolygonNameHolder>>(); diff --git a/src/Functions/seriesPeriodDetectFFT.cpp b/src/Functions/seriesPeriodDetectFFT.cpp index 471354235d5..ecf8398bbd5 100644 --- a/src/Functions/seriesPeriodDetectFFT.cpp +++ b/src/Functions/seriesPeriodDetectFFT.cpp @@ -153,12 +153,8 @@ public: return true; } - std::vector xfreq(spec_len); double step = 0.5 / (spec_len - 1); - for (size_t i = 0; i < spec_len; ++i) - xfreq[i] = i * step; - - auto freq = xfreq[idx]; + auto freq = idx * step; period = std::round(1 / freq); return true; diff --git a/src/Functions/sleep.h b/src/Functions/sleep.h index 62ee19fa904..b6e4b36ee64 100644 --- a/src/Functions/sleep.h +++ b/src/Functions/sleep.h @@ -7,6 +7,7 @@ #include #include #include +#include #include #include #include @@ -32,6 +33,11 @@ namespace ErrorCodes extern const int BAD_ARGUMENTS; } +namespace FailPoints +{ + extern const char infinite_sleep[]; +} + /** sleep(seconds) - the specified number of seconds sleeps each columns. */ @@ -107,6 +113,8 @@ public: { /// When sleeping, the query cannot be cancelled. For ability to cancel query, we limit sleep time. UInt64 microseconds = static_cast(seconds * 1e6); + FailPointInjection::pauseFailPoint(FailPoints::infinite_sleep); + if (max_microseconds && microseconds > max_microseconds) throw Exception(ErrorCodes::TOO_SLOW, "The maximum sleep time is {} microseconds. Requested: {} microseconds", max_microseconds, microseconds); diff --git a/src/Functions/toBool.cpp b/src/Functions/toBool.cpp index 6f2c436c1ea..ac595d313e3 100644 --- a/src/Functions/toBool.cpp +++ b/src/Functions/toBool.cpp @@ -54,8 +54,7 @@ namespace } }; - FunctionOverloadResolverPtr func_builder_cast = createInternalCastOverloadResolver(CastType::nonAccurate, {}); - auto func_cast = func_builder_cast->build(cast_args); + auto func_cast = createInternalCast(arguments[0], result_type, CastType::nonAccurate, {}); return func_cast->execute(cast_args, result_type, arguments[0].column->size()); } }; diff --git a/src/IO/HTTPCommon.cpp b/src/IO/HTTPCommon.cpp index 9704d034b2a..fcd96e97b4e 100644 --- a/src/IO/HTTPCommon.cpp +++ b/src/IO/HTTPCommon.cpp @@ -34,14 +34,20 @@ namespace ErrorCodes extern const int RECEIVED_ERROR_TOO_MANY_REQUESTS; } -void setResponseDefaultHeaders(HTTPServerResponse & response, size_t keep_alive_timeout) +void setResponseDefaultHeaders(HTTPServerResponse & response) { if (!response.getKeepAlive()) return; - Poco::Timespan timeout(keep_alive_timeout, 0); - if (timeout.totalSeconds()) - response.set("Keep-Alive", "timeout=" + std::to_string(timeout.totalSeconds())); + const size_t keep_alive_timeout = response.getSession().getKeepAliveTimeout(); + const size_t keep_alive_max_requests = response.getSession().getMaxKeepAliveRequests(); + if (keep_alive_timeout) + { + if (keep_alive_max_requests) + response.set("Keep-Alive", fmt::format("timeout={}, max={}", keep_alive_timeout, keep_alive_max_requests)); + else + response.set("Keep-Alive", fmt::format("timeout={}", keep_alive_timeout)); + } } HTTPSessionPtr makeHTTPSession( diff --git a/src/IO/HTTPCommon.h b/src/IO/HTTPCommon.h index 3a1fa5bebee..4d0580acaba 100644 --- a/src/IO/HTTPCommon.h +++ b/src/IO/HTTPCommon.h @@ -54,7 +54,7 @@ private: using HTTPSessionPtr = std::shared_ptr; -void setResponseDefaultHeaders(HTTPServerResponse & response, size_t keep_alive_timeout); +void setResponseDefaultHeaders(HTTPServerResponse & response); /// Create session object to perform requests and set required parameters. HTTPSessionPtr makeHTTPSession( diff --git a/src/IO/NetUtils.h b/src/IO/NetUtils.h new file mode 100644 index 00000000000..12f09524ae7 --- /dev/null +++ b/src/IO/NetUtils.h @@ -0,0 +1,58 @@ +#pragma once + +#include +#include + + +namespace DB +{ + +template +constexpr T netToHost(T value) noexcept +{ + if constexpr (std::endian::native != std::endian::big) + return std::byteswap(value); + return value; +} + +template +constexpr T hostToNet(T value) noexcept +{ + if constexpr (std::endian::native != std::endian::big) + return std::byteswap(value); + return value; +} + +template +constexpr T toLittleEndian(T value) noexcept +{ + if constexpr (std::endian::native == std::endian::big) + return std::byteswap(value); + return value; +} + +template +constexpr T toBigEndian(T value) noexcept +{ + if constexpr (std::endian::native != std::endian::big) + return std::byteswap(value); + return value; +} + +template +constexpr T fromLittleEndian(T value) noexcept +{ + if constexpr (std::endian::native == std::endian::big) + return std::byteswap(value); + return value; +} + +template +constexpr T fromBigEndian(T value) noexcept +{ + if constexpr (std::endian::native != std::endian::big) + return std::byteswap(value); + return value; +} + +} diff --git a/src/IO/Protobuf/ProtobufZeroCopyInputStreamFromReadBuffer.cpp b/src/IO/Protobuf/ProtobufZeroCopyInputStreamFromReadBuffer.cpp new file mode 100644 index 00000000000..86b7eb4d7f7 --- /dev/null +++ b/src/IO/Protobuf/ProtobufZeroCopyInputStreamFromReadBuffer.cpp @@ -0,0 +1,56 @@ +#include "config.h" + +#if USE_PROTOBUF +#include +#include + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +} + +ProtobufZeroCopyInputStreamFromReadBuffer::ProtobufZeroCopyInputStreamFromReadBuffer(std::unique_ptr in_) : in(std::move(in_)) +{ +} + +ProtobufZeroCopyInputStreamFromReadBuffer::~ProtobufZeroCopyInputStreamFromReadBuffer() = default; + +bool ProtobufZeroCopyInputStreamFromReadBuffer::Next(const void ** data, int * size) +{ + if (in->eof()) + return false; + *data = in->position(); + *size = static_cast(in->available()); + in->position() += *size; + return true; +} + +void ProtobufZeroCopyInputStreamFromReadBuffer::BackUp(int count) +{ + if (static_cast(in->offset()) < count) + throw Exception( + ErrorCodes::LOGICAL_ERROR, + "ProtobufZeroCopyInputStreamFromReadBuffer::BackUp() cannot back up {} bytes (max = {} bytes)", + count, + in->offset()); + + in->position() -= count; +} + +bool ProtobufZeroCopyInputStreamFromReadBuffer::Skip(int count) +{ + return static_cast(in->tryIgnore(count)) == count; +} + +int64_t ProtobufZeroCopyInputStreamFromReadBuffer::ByteCount() const +{ + return in->count(); +} + +} + +#endif diff --git a/src/IO/Protobuf/ProtobufZeroCopyInputStreamFromReadBuffer.h b/src/IO/Protobuf/ProtobufZeroCopyInputStreamFromReadBuffer.h new file mode 100644 index 00000000000..3f86815ef3f --- /dev/null +++ b/src/IO/Protobuf/ProtobufZeroCopyInputStreamFromReadBuffer.h @@ -0,0 +1,38 @@ +#pragma once + +#include "config.h" +#if USE_PROTOBUF + +#include + + +namespace DB +{ +class ReadBuffer; + +class ProtobufZeroCopyInputStreamFromReadBuffer : public google::protobuf::io::ZeroCopyInputStream +{ +public: + explicit ProtobufZeroCopyInputStreamFromReadBuffer(std::unique_ptr in_); + ~ProtobufZeroCopyInputStreamFromReadBuffer() override; + + // Obtains a chunk of data from the stream. + bool Next(const void ** data, int * size) override; + + // Backs up a number of bytes, so that the next call to Next() returns + // data again that was already returned by the last call to Next(). + void BackUp(int count) override; + + // Skips a number of bytes. + bool Skip(int count) override; + + // Returns the total number of bytes read since this object was created. + int64_t ByteCount() const override; + +private: + std::unique_ptr in; +}; + +} + +#endif diff --git a/src/IO/Protobuf/ProtobufZeroCopyOutputStreamFromWriteBuffer.cpp b/src/IO/Protobuf/ProtobufZeroCopyOutputStreamFromWriteBuffer.cpp new file mode 100644 index 00000000000..d1e02b436f3 --- /dev/null +++ b/src/IO/Protobuf/ProtobufZeroCopyOutputStreamFromWriteBuffer.cpp @@ -0,0 +1,60 @@ +#include "config.h" + +#if USE_PROTOBUF +#include +#include + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +} + +ProtobufZeroCopyOutputStreamFromWriteBuffer::ProtobufZeroCopyOutputStreamFromWriteBuffer(WriteBuffer & out_) : out(&out_) +{ +} + +ProtobufZeroCopyOutputStreamFromWriteBuffer::ProtobufZeroCopyOutputStreamFromWriteBuffer(std::unique_ptr out_) + : ProtobufZeroCopyOutputStreamFromWriteBuffer(*out_) +{ + out_holder = std::move(out_); +} + +ProtobufZeroCopyOutputStreamFromWriteBuffer::~ProtobufZeroCopyOutputStreamFromWriteBuffer() = default; + +bool ProtobufZeroCopyOutputStreamFromWriteBuffer::Next(void ** data, int * size) +{ + *data = out->position(); + *size = static_cast(out->available()); + out->position() += *size; + return true; +} + +void ProtobufZeroCopyOutputStreamFromWriteBuffer::BackUp(int count) +{ + if (static_cast(out->offset()) < count) + throw Exception( + ErrorCodes::LOGICAL_ERROR, + "ProtobufZeroCopyOutputStreamFromWriteBuffer::BackUp() cannot back up {} bytes (max = {} bytes)", + count, + out->offset()); + + out->position() -= count; +} + +int64_t ProtobufZeroCopyOutputStreamFromWriteBuffer::ByteCount() const +{ + return out->count(); +} + +void ProtobufZeroCopyOutputStreamFromWriteBuffer::finalize() +{ + out->finalize(); +} + +} + +#endif diff --git a/src/IO/Protobuf/ProtobufZeroCopyOutputStreamFromWriteBuffer.h b/src/IO/Protobuf/ProtobufZeroCopyOutputStreamFromWriteBuffer.h new file mode 100644 index 00000000000..c47cef9ff4d --- /dev/null +++ b/src/IO/Protobuf/ProtobufZeroCopyOutputStreamFromWriteBuffer.h @@ -0,0 +1,40 @@ +#pragma once + +#include "config.h" +#if USE_PROTOBUF + +#include + + +namespace DB +{ +class WriteBuffer; + +class ProtobufZeroCopyOutputStreamFromWriteBuffer : public google::protobuf::io::ZeroCopyOutputStream +{ +public: + explicit ProtobufZeroCopyOutputStreamFromWriteBuffer(WriteBuffer & out_); + explicit ProtobufZeroCopyOutputStreamFromWriteBuffer(std::unique_ptr out_); + + ~ProtobufZeroCopyOutputStreamFromWriteBuffer() override; + + // Obtains a buffer into which data can be written. + bool Next(void ** data, int * size) override; + + // Backs up a number of bytes, so that the end of the last buffer returned + // by Next() is not actually written. + void BackUp(int count) override; + + // Returns the total number of bytes written since this object was created. + int64_t ByteCount() const override; + + void finalize(); + +private: + WriteBuffer * out; + std::unique_ptr out_holder; +}; + +} + +#endif diff --git a/src/IO/ReadBufferFromPocoSocket.cpp b/src/IO/ReadBufferFromPocoSocket.cpp index 26cdee4140c..bbf9f96404f 100644 --- a/src/IO/ReadBufferFromPocoSocket.cpp +++ b/src/IO/ReadBufferFromPocoSocket.cpp @@ -32,7 +32,7 @@ namespace ErrorCodes extern const int LOGICAL_ERROR; } -bool ReadBufferFromPocoSocket::nextImpl() +ssize_t ReadBufferFromPocoSocketBase::socketReceiveBytesImpl(char * ptr, size_t size) { ssize_t bytes_read = 0; Stopwatch watch; @@ -43,14 +43,11 @@ bool ReadBufferFromPocoSocket::nextImpl() ProfileEvents::increment(ProfileEvents::NetworkReceiveBytes, bytes_read); }); + CurrentMetrics::Increment metric_increment(CurrentMetrics::NetworkReceive); + /// Add more details to exceptions. try { - CurrentMetrics::Increment metric_increment(CurrentMetrics::NetworkReceive); - - if (internal_buffer.size() > INT_MAX) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Buffer overflow"); - /// If async_callback is specified, set socket to non-blocking mode /// and try to read data from it, if socket is not ready for reading, /// run async_callback and try again later. @@ -61,7 +58,7 @@ bool ReadBufferFromPocoSocket::nextImpl() socket.setBlocking(false); SCOPE_EXIT(socket.setBlocking(true)); bool secure = socket.secure(); - bytes_read = socket.impl()->receiveBytes(internal_buffer.begin(), static_cast(internal_buffer.size())); + bytes_read = socket.impl()->receiveBytes(ptr, static_cast(size)); /// Check EAGAIN and ERR_SSL_WANT_READ/ERR_SSL_WANT_WRITE for secure socket (reading from secure socket can write too). while (bytes_read < 0 && (errno == EAGAIN || (secure && (checkSSLWantRead(bytes_read) || checkSSLWantWrite(bytes_read))))) @@ -73,12 +70,12 @@ bool ReadBufferFromPocoSocket::nextImpl() async_callback(socket.impl()->sockfd(), socket.getReceiveTimeout(), AsyncEventTimeoutType::RECEIVE, socket_description, AsyncTaskExecutor::Event::READ | AsyncTaskExecutor::Event::ERROR); /// Try to read again. - bytes_read = socket.impl()->receiveBytes(internal_buffer.begin(), static_cast(internal_buffer.size())); + bytes_read = socket.impl()->receiveBytes(ptr, static_cast(size)); } } else { - bytes_read = socket.impl()->receiveBytes(internal_buffer.begin(), static_cast(internal_buffer.size())); + bytes_read = socket.impl()->receiveBytes(ptr, static_cast(size)); } } catch (const Poco::Net::NetException & e) @@ -99,6 +96,16 @@ bool ReadBufferFromPocoSocket::nextImpl() if (bytes_read < 0) throw NetException(ErrorCodes::CANNOT_READ_FROM_SOCKET, "Cannot read from socket (peer: {}, local: {})", peer_address.toString(), socket.address().toString()); + return bytes_read; +} + +bool ReadBufferFromPocoSocketBase::nextImpl() +{ + if (internal_buffer.size() > INT_MAX) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Buffer overflow"); + + ssize_t bytes_read = socketReceiveBytesImpl(internal_buffer.begin(), internal_buffer.size()); + if (read_event != ProfileEvents::end()) ProfileEvents::increment(read_event, bytes_read); @@ -110,7 +117,7 @@ bool ReadBufferFromPocoSocket::nextImpl() return true; } -ReadBufferFromPocoSocket::ReadBufferFromPocoSocket(Poco::Net::Socket & socket_, size_t buf_size) +ReadBufferFromPocoSocketBase::ReadBufferFromPocoSocketBase(Poco::Net::Socket & socket_, size_t buf_size) : BufferWithOwnMemory(buf_size) , socket(socket_) , peer_address(socket.peerAddress()) @@ -119,19 +126,22 @@ ReadBufferFromPocoSocket::ReadBufferFromPocoSocket(Poco::Net::Socket & socket_, { } -ReadBufferFromPocoSocket::ReadBufferFromPocoSocket(Poco::Net::Socket & socket_, const ProfileEvents::Event & read_event_, size_t buf_size) - : ReadBufferFromPocoSocket(socket_, buf_size) +ReadBufferFromPocoSocketBase::ReadBufferFromPocoSocketBase(Poco::Net::Socket & socket_, const ProfileEvents::Event & read_event_, size_t buf_size) + : ReadBufferFromPocoSocketBase(socket_, buf_size) { read_event = read_event_; } -bool ReadBufferFromPocoSocket::poll(size_t timeout_microseconds) const +bool ReadBufferFromPocoSocketBase::poll(size_t timeout_microseconds) const { - if (available()) + /// For secure socket it is important to check if any remaining data available in underlying decryption buffer - + /// read always retrieves the whole encrypted frame from the wire and puts it into underlying buffer while returning only requested size - + /// further poll() can block though there is still data to read in the underlying decryption buffer. + if (available() || socket.impl()->available()) return true; Stopwatch watch; - bool res = socket.poll(timeout_microseconds, Poco::Net::Socket::SELECT_READ | Poco::Net::Socket::SELECT_ERROR); + bool res = socket.impl()->poll(timeout_microseconds, Poco::Net::Socket::SELECT_READ | Poco::Net::Socket::SELECT_ERROR); ProfileEvents::increment(ProfileEvents::NetworkReceiveElapsedMicroseconds, watch.elapsedMicroseconds()); return res; } diff --git a/src/IO/ReadBufferFromPocoSocket.h b/src/IO/ReadBufferFromPocoSocket.h index 76156612764..912388adaac 100644 --- a/src/IO/ReadBufferFromPocoSocket.h +++ b/src/IO/ReadBufferFromPocoSocket.h @@ -9,7 +9,7 @@ namespace DB { /// Works with the ready Poco::Net::Socket. Blocking operations. -class ReadBufferFromPocoSocket : public BufferWithOwnMemory +class ReadBufferFromPocoSocketBase : public BufferWithOwnMemory { protected: Poco::Net::Socket & socket; @@ -25,16 +25,29 @@ protected: bool nextImpl() override; public: - explicit ReadBufferFromPocoSocket(Poco::Net::Socket & socket_, size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE); - explicit ReadBufferFromPocoSocket(Poco::Net::Socket & socket_, const ProfileEvents::Event & read_event_, size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE); + explicit ReadBufferFromPocoSocketBase(Poco::Net::Socket & socket_, size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE); + explicit ReadBufferFromPocoSocketBase(Poco::Net::Socket & socket_, const ProfileEvents::Event & read_event_, size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE); bool poll(size_t timeout_microseconds) const; void setAsyncCallback(AsyncCallback async_callback_) { async_callback = std::move(async_callback_); } + ssize_t socketReceiveBytesImpl(char * ptr, size_t size); + private: AsyncCallback async_callback; std::string socket_description; }; +class ReadBufferFromPocoSocket : public ReadBufferFromPocoSocketBase +{ +public: + explicit ReadBufferFromPocoSocket(Poco::Net::Socket & socket_, size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE) + : ReadBufferFromPocoSocketBase(socket_, buf_size) + {} + explicit ReadBufferFromPocoSocket(Poco::Net::Socket & socket_, const ProfileEvents::Event & read_event_, size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE) + : ReadBufferFromPocoSocketBase(socket_, read_event_, buf_size) + {} +}; + } diff --git a/src/IO/ReadBufferFromPocoSocketChunked.cpp b/src/IO/ReadBufferFromPocoSocketChunked.cpp new file mode 100644 index 00000000000..4a1e3732a55 --- /dev/null +++ b/src/IO/ReadBufferFromPocoSocketChunked.cpp @@ -0,0 +1,166 @@ +#include +#include +#include + + +namespace DB::ErrorCodes +{ + extern const int LOGICAL_ERROR; +} + +namespace DB +{ + +ReadBufferFromPocoSocketChunked::ReadBufferFromPocoSocketChunked(Poco::Net::Socket & socket_, size_t buf_size) + : ReadBufferFromPocoSocketChunked(socket_, ProfileEvents::end(), buf_size) +{} + +ReadBufferFromPocoSocketChunked::ReadBufferFromPocoSocketChunked(Poco::Net::Socket & socket_, const ProfileEvents::Event & read_event_, size_t buf_size) + : ReadBufferFromPocoSocketBase( + socket_, read_event_, + std::min(buf_size, static_cast(std::numeric_limits::max()))), + our_address(socket_.address()), log(getLogger("Protocol")) +{} + +void ReadBufferFromPocoSocketChunked::enableChunked() +{ + if (chunked) + return; + chunked = 1; + data_end = buffer().end(); + /// Resize working buffer so any next read will call nextImpl + working_buffer.resize(offset()); + chunk_left = 0; + next_chunk = 0; +} + +bool ReadBufferFromPocoSocketChunked::hasBufferedData() const +{ + if (available()) + return true; + + return chunked && (static_cast(data_end - working_buffer.end()) > sizeof(next_chunk)); +} + +bool ReadBufferFromPocoSocketChunked::poll(size_t timeout_microseconds) const +{ + if (chunked) + if (available() || static_cast(data_end - working_buffer.end()) > sizeof(next_chunk)) + return true; + + return ReadBufferFromPocoSocketBase::poll(timeout_microseconds); +} + + +bool ReadBufferFromPocoSocketChunked::loadNextChunk(Position c_pos, bool cont) +{ + auto buffered = std::min(static_cast(data_end - c_pos), sizeof(next_chunk)); + + if (buffered) + std::memcpy(&next_chunk, c_pos, buffered); + if (buffered < sizeof(next_chunk)) + if (socketReceiveBytesImpl(reinterpret_cast(&next_chunk) + buffered, sizeof(next_chunk) - buffered) < static_cast(sizeof(next_chunk) - buffered)) + return false; + next_chunk = fromLittleEndian(next_chunk); + + if (next_chunk) + { + if (cont) + LOG_TEST(log, "{} <- {} Chunk receive continued. Size {}", ourAddress().toString(), peerAddress().toString(), next_chunk); + } + else + LOG_TEST(log, "{} <- {} Chunk receive ended.", ourAddress().toString(), peerAddress().toString()); + + return true; +} + +bool ReadBufferFromPocoSocketChunked::processChunkLeft(Position c_pos) +{ + if (data_end - c_pos < chunk_left) + { + working_buffer.resize(data_end - buffer().begin()); + nextimpl_working_buffer_offset = c_pos - buffer().begin(); + chunk_left -= (data_end - c_pos); + return true; + } + + nextimpl_working_buffer_offset = c_pos - buffer().begin(); + working_buffer.resize(nextimpl_working_buffer_offset + chunk_left); + + c_pos += chunk_left; + + if (!loadNextChunk(c_pos, true)) + return false; + + chunk_left = 0; + return true; +} + + +bool ReadBufferFromPocoSocketChunked::nextImpl() +{ + if (!chunked) + return ReadBufferFromPocoSocketBase::nextImpl(); + + auto * c_pos = pos; + + if (chunk_left == 0) + { + if (next_chunk == 0) + { + if (chunked == 1) + chunked = 2; // first chunked block - no end marker + else + c_pos = pos + sizeof(next_chunk); // bypass chunk end marker + + if (c_pos > data_end) + c_pos = data_end; + + if (!loadNextChunk(c_pos)) + return false; + + chunk_left = next_chunk; + next_chunk = 0; + + if (chunk_left == 0) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Native protocol: empty chunk received"); + + c_pos += sizeof(next_chunk); + + if (c_pos >= data_end) + { + if (!ReadBufferFromPocoSocketBase::nextImpl()) + return false; + data_end = buffer().end(); + c_pos = buffer().begin(); + } + + LOG_TEST(log, "{} <- {} Chunk receive started. Message {}, size {}", ourAddress().toString(), peerAddress().toString(), static_cast(*c_pos), chunk_left); + } + else + { + c_pos += sizeof(next_chunk); + if (c_pos >= data_end) + { + if (!ReadBufferFromPocoSocketBase::nextImpl()) + return false; + data_end = buffer().end(); + c_pos = buffer().begin(); + } + + chunk_left = next_chunk; + next_chunk = 0; + } + } + else + { + if (!ReadBufferFromPocoSocketBase::nextImpl()) + return false; + data_end = buffer().end(); + c_pos = buffer().begin(); + } + + return processChunkLeft(c_pos); +} + +} diff --git a/src/IO/ReadBufferFromPocoSocketChunked.h b/src/IO/ReadBufferFromPocoSocketChunked.h new file mode 100644 index 00000000000..8bc4024b978 --- /dev/null +++ b/src/IO/ReadBufferFromPocoSocketChunked.h @@ -0,0 +1,109 @@ +#pragma once + +#include +#include + +/* + +Handshake +============= + | 'Hello' type + | handshake exchange + | chunked protocol negotiation + +============= + + +Basic chunk: + +============= +Chunk begins | 0x12345678 chunk size, 4 bytes little endian + +------------- + | Packet type always follows beginning of the chunk + | packet data + +------------- +Chunk ends | 0x00000000 4 zero bytes + +============= + + +Datastream chunk: + +============= +Chunk begins | 0x12345678 + +------------- + | Packet type + | packet data + +------------- + | Packet type + | packet data + +------------- +...arbitrary number ..... +of packets... ..... + +------------- + | Packet type + | packet data + +------------- +Chunk ends | 0x00000000 + +============= + + +Multipart chunk: + +============= +Chunk begins | 0x12345678 chunk part size, 4 bytes little endian + +------------- + | Packet type + | packet data + +------------- + | Packet type + | (partial) packet data + +============= +Chunk continues | 0x12345678 chunk next part size, 4 bytes little endian + +============= + | possibly previous packet's data + +------------- + | Packet type + | packet data + +------------- +...arbitrary number ..... +of chunk parts... ..... + +------------- + | Packet type + | packet data + +------------- +Chunk ends | 0x00000000 + +============= + +*/ + +namespace DB +{ + +class ReadBufferFromPocoSocketChunked: public ReadBufferFromPocoSocketBase +{ +public: + using ReadBufferFromPocoSocketBase::setAsyncCallback; + + explicit ReadBufferFromPocoSocketChunked(Poco::Net::Socket & socket_, size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE); + explicit ReadBufferFromPocoSocketChunked(Poco::Net::Socket & socket_, const ProfileEvents::Event & read_event_, size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE); + + void enableChunked(); + + bool hasBufferedData() const; + + bool poll(size_t timeout_microseconds) const; + + Poco::Net::SocketAddress peerAddress() { return peer_address; } + Poco::Net::SocketAddress ourAddress() { return our_address; } + +protected: + bool loadNextChunk(Position c_pos, bool cont = false); + bool processChunkLeft(Position c_pos); + bool nextImpl() override; + + Poco::Net::SocketAddress our_address; + +private: + LoggerPtr log; + Position data_end = nullptr; // end position of data in the internal_buffer + UInt32 chunk_left = 0; // chunk left to read from socket + UInt32 next_chunk = 0; // size of the next cnunk + UInt8 chunked = 0; // 0 - disabled; 1 - started; 2 - enabled; +}; + +} diff --git a/src/IO/ReadHelpers.cpp b/src/IO/ReadHelpers.cpp index 9559462e62b..48d788512e4 100644 --- a/src/IO/ReadHelpers.cpp +++ b/src/IO/ReadHelpers.cpp @@ -1271,7 +1271,7 @@ template void readJSONArrayInto, void>(PaddedPODArray, bool>(PaddedPODArray & s, ReadBuffer & buf); template -ReturnType readDateTextFallback(LocalDate & date, ReadBuffer & buf) +ReturnType readDateTextFallback(LocalDate & date, ReadBuffer & buf, const char * allowed_delimiters) { static constexpr bool throw_exception = std::is_same_v; @@ -1318,6 +1318,9 @@ ReturnType readDateTextFallback(LocalDate & date, ReadBuffer & buf) } else { + if (!isSymbolIn(*buf.position(), allowed_delimiters)) + return error(); + ++buf.position(); if (!append_digit(month)) @@ -1325,7 +1328,11 @@ ReturnType readDateTextFallback(LocalDate & date, ReadBuffer & buf) append_digit(month); if (!buf.eof() && !isNumericASCII(*buf.position())) + { + if (!isSymbolIn(*buf.position(), allowed_delimiters)) + return error(); ++buf.position(); + } else return error(); @@ -1338,12 +1345,12 @@ ReturnType readDateTextFallback(LocalDate & date, ReadBuffer & buf) return ReturnType(true); } -template void readDateTextFallback(LocalDate &, ReadBuffer &); -template bool readDateTextFallback(LocalDate &, ReadBuffer &); +template void readDateTextFallback(LocalDate &, ReadBuffer &, const char * allowed_delimiters); +template bool readDateTextFallback(LocalDate &, ReadBuffer &, const char * allowed_delimiters); template -ReturnType readDateTimeTextFallback(time_t & datetime, ReadBuffer & buf, const DateLUTImpl & date_lut) +ReturnType readDateTimeTextFallback(time_t & datetime, ReadBuffer & buf, const DateLUTImpl & date_lut, const char * allowed_date_delimiters, const char * allowed_time_delimiters) { static constexpr bool throw_exception = std::is_same_v; @@ -1413,6 +1420,9 @@ ReturnType readDateTimeTextFallback(time_t & datetime, ReadBuffer & buf, const D if (!isNumericASCII(s[0]) || !isNumericASCII(s[1]) || !isNumericASCII(s[2]) || !isNumericASCII(s[3]) || !isNumericASCII(s[5]) || !isNumericASCII(s[6]) || !isNumericASCII(s[8]) || !isNumericASCII(s[9])) return false; + + if (!isSymbolIn(s[4], allowed_date_delimiters) || !isSymbolIn(s[7], allowed_date_delimiters)) + return false; } UInt16 year = (s[0] - '0') * 1000 + (s[1] - '0') * 100 + (s[2] - '0') * 10 + (s[3] - '0'); @@ -1443,6 +1453,9 @@ ReturnType readDateTimeTextFallback(time_t & datetime, ReadBuffer & buf, const D if (!isNumericASCII(s[0]) || !isNumericASCII(s[1]) || !isNumericASCII(s[3]) || !isNumericASCII(s[4]) || !isNumericASCII(s[6]) || !isNumericASCII(s[7])) return false; + + if (!isSymbolIn(s[2], allowed_time_delimiters) || !isSymbolIn(s[5], allowed_time_delimiters)) + return false; } hour = (s[0] - '0') * 10 + (s[1] - '0'); @@ -1488,10 +1501,10 @@ ReturnType readDateTimeTextFallback(time_t & datetime, ReadBuffer & buf, const D return ReturnType(true); } -template void readDateTimeTextFallback(time_t &, ReadBuffer &, const DateLUTImpl &); -template void readDateTimeTextFallback(time_t &, ReadBuffer &, const DateLUTImpl &); -template bool readDateTimeTextFallback(time_t &, ReadBuffer &, const DateLUTImpl &); -template bool readDateTimeTextFallback(time_t &, ReadBuffer &, const DateLUTImpl &); +template void readDateTimeTextFallback(time_t &, ReadBuffer &, const DateLUTImpl &, const char *, const char *); +template void readDateTimeTextFallback(time_t &, ReadBuffer &, const DateLUTImpl &, const char *, const char *); +template bool readDateTimeTextFallback(time_t &, ReadBuffer &, const DateLUTImpl &, const char *, const char *); +template bool readDateTimeTextFallback(time_t &, ReadBuffer &, const DateLUTImpl &, const char *, const char *); template diff --git a/src/IO/ReadHelpers.h b/src/IO/ReadHelpers.h index ffba4fafb5c..8e3e9a2fbe1 100644 --- a/src/IO/ReadHelpers.h +++ b/src/IO/ReadHelpers.h @@ -258,6 +258,20 @@ inline void readBoolText(bool & x, ReadBuffer & buf) char tmp = '0'; readChar(tmp, buf); x = tmp != '0'; + + if (!buf.eof() && isAlphaASCII(tmp)) + { + if (tmp == 't' || tmp == 'T') + { + assertStringCaseInsensitive("rue", buf); + x = true; + } + else if (tmp == 'f' || tmp == 'F') + { + assertStringCaseInsensitive("alse", buf); + x = false; + } + } } template @@ -703,13 +717,28 @@ struct NullOutput }; template -ReturnType readDateTextFallback(LocalDate & date, ReadBuffer & buf); +ReturnType readDateTextFallback(LocalDate & date, ReadBuffer & buf, const char * allowed_delimiters); + +inline bool isSymbolIn(char symbol, const char * symbols) +{ + if (symbols == nullptr) + return true; + + const char * pos = symbols; + while (*pos) + { + if (*pos == symbol) + return true; + ++pos; + } + return false; +} /// In YYYY-MM-DD format. /// For convenience, Month and Day parts can have single digit instead of two digits. /// Any separators other than '-' are supported. template -inline ReturnType readDateTextImpl(LocalDate & date, ReadBuffer & buf) +inline ReturnType readDateTextImpl(LocalDate & date, ReadBuffer & buf, const char * allowed_delimiters = nullptr) { static constexpr bool throw_exception = std::is_same_v; @@ -753,6 +782,9 @@ inline ReturnType readDateTextImpl(LocalDate & date, ReadBuffer & buf) } else { + if (!isSymbolIn(pos[-1], allowed_delimiters)) + return error(); + if (!isNumericASCII(pos[0])) return error(); @@ -768,6 +800,9 @@ inline ReturnType readDateTextImpl(LocalDate & date, ReadBuffer & buf) if (isNumericASCII(pos[-1]) || !isNumericASCII(pos[0])) return error(); + if (!isSymbolIn(pos[-1], allowed_delimiters)) + return error(); + day = pos[0] - '0'; if (isNumericASCII(pos[1])) { @@ -783,7 +818,7 @@ inline ReturnType readDateTextImpl(LocalDate & date, ReadBuffer & buf) return ReturnType(true); } else - return readDateTextFallback(date, buf); + return readDateTextFallback(date, buf, allowed_delimiters); } inline void convertToDayNum(DayNum & date, ExtendedDayNum & from) @@ -797,15 +832,15 @@ inline void convertToDayNum(DayNum & date, ExtendedDayNum & from) } template -inline ReturnType readDateTextImpl(DayNum & date, ReadBuffer & buf, const DateLUTImpl & date_lut) +inline ReturnType readDateTextImpl(DayNum & date, ReadBuffer & buf, const DateLUTImpl & date_lut, const char * allowed_delimiters = nullptr) { static constexpr bool throw_exception = std::is_same_v; LocalDate local_date; if constexpr (throw_exception) - readDateTextImpl(local_date, buf); - else if (!readDateTextImpl(local_date, buf)) + readDateTextImpl(local_date, buf, allowed_delimiters); + else if (!readDateTextImpl(local_date, buf, allowed_delimiters)) return false; ExtendedDayNum ret = date_lut.makeDayNum(local_date.year(), local_date.month(), local_date.day()); @@ -814,15 +849,15 @@ inline ReturnType readDateTextImpl(DayNum & date, ReadBuffer & buf, const DateLU } template -inline ReturnType readDateTextImpl(ExtendedDayNum & date, ReadBuffer & buf, const DateLUTImpl & date_lut) +inline ReturnType readDateTextImpl(ExtendedDayNum & date, ReadBuffer & buf, const DateLUTImpl & date_lut, const char * allowed_delimiters = nullptr) { static constexpr bool throw_exception = std::is_same_v; LocalDate local_date; if constexpr (throw_exception) - readDateTextImpl(local_date, buf); - else if (!readDateTextImpl(local_date, buf)) + readDateTextImpl(local_date, buf, allowed_delimiters); + else if (!readDateTextImpl(local_date, buf, allowed_delimiters)) return false; /// When the parameter is out of rule or out of range, Date32 uses 1925-01-01 as the default value (-DateLUT::instance().getDayNumOffsetEpoch(), -16436) and Date uses 1970-01-01. @@ -846,19 +881,19 @@ inline void readDateText(ExtendedDayNum & date, ReadBuffer & buf, const DateLUTI readDateTextImpl(date, buf, date_lut); } -inline bool tryReadDateText(LocalDate & date, ReadBuffer & buf) +inline bool tryReadDateText(LocalDate & date, ReadBuffer & buf, const char * allowed_delimiters = nullptr) { - return readDateTextImpl(date, buf); + return readDateTextImpl(date, buf, allowed_delimiters); } -inline bool tryReadDateText(DayNum & date, ReadBuffer & buf, const DateLUTImpl & time_zone = DateLUT::instance()) +inline bool tryReadDateText(DayNum & date, ReadBuffer & buf, const DateLUTImpl & time_zone = DateLUT::instance(), const char * allowed_delimiters = nullptr) { - return readDateTextImpl(date, buf, time_zone); + return readDateTextImpl(date, buf, time_zone, allowed_delimiters); } -inline bool tryReadDateText(ExtendedDayNum & date, ReadBuffer & buf, const DateLUTImpl & time_zone = DateLUT::instance()) +inline bool tryReadDateText(ExtendedDayNum & date, ReadBuffer & buf, const DateLUTImpl & time_zone = DateLUT::instance(), const char * allowed_delimiters = nullptr) { - return readDateTextImpl(date, buf, time_zone); + return readDateTextImpl(date, buf, time_zone, allowed_delimiters); } UUID parseUUID(std::span src); @@ -975,13 +1010,13 @@ inline T parseFromString(std::string_view str) template -ReturnType readDateTimeTextFallback(time_t & datetime, ReadBuffer & buf, const DateLUTImpl & date_lut); +ReturnType readDateTimeTextFallback(time_t & datetime, ReadBuffer & buf, const DateLUTImpl & date_lut, const char * allowed_date_delimiters = nullptr, const char * allowed_time_delimiters = nullptr); /** In YYYY-MM-DD hh:mm:ss or YYYY-MM-DD format, according to specified time zone. * As an exception, also supported parsing of unix timestamp in form of decimal number. */ template -inline ReturnType readDateTimeTextImpl(time_t & datetime, ReadBuffer & buf, const DateLUTImpl & date_lut) +inline ReturnType readDateTimeTextImpl(time_t & datetime, ReadBuffer & buf, const DateLUTImpl & date_lut, const char * allowed_date_delimiters = nullptr, const char * allowed_time_delimiters = nullptr) { static constexpr bool throw_exception = std::is_same_v; @@ -1014,6 +1049,9 @@ inline ReturnType readDateTimeTextImpl(time_t & datetime, ReadBuffer & buf, cons if (!isNumericASCII(s[0]) || !isNumericASCII(s[1]) || !isNumericASCII(s[2]) || !isNumericASCII(s[3]) || !isNumericASCII(s[5]) || !isNumericASCII(s[6]) || !isNumericASCII(s[8]) || !isNumericASCII(s[9])) return ReturnType(false); + + if (!isSymbolIn(s[4], allowed_date_delimiters) || !isSymbolIn(s[7], allowed_date_delimiters)) + return ReturnType(false); } UInt16 year = (s[0] - '0') * 1000 + (s[1] - '0') * 100 + (s[2] - '0') * 10 + (s[3] - '0'); @@ -1033,6 +1071,9 @@ inline ReturnType readDateTimeTextImpl(time_t & datetime, ReadBuffer & buf, cons if (!isNumericASCII(s[11]) || !isNumericASCII(s[12]) || !isNumericASCII(s[14]) || !isNumericASCII(s[15]) || !isNumericASCII(s[17]) || !isNumericASCII(s[18])) return ReturnType(false); + + if (!isSymbolIn(s[13], allowed_time_delimiters) || !isSymbolIn(s[16], allowed_time_delimiters)) + return ReturnType(false); } hour = (s[11] - '0') * 10 + (s[12] - '0'); @@ -1057,11 +1098,11 @@ inline ReturnType readDateTimeTextImpl(time_t & datetime, ReadBuffer & buf, cons return readIntTextImpl(datetime, buf); } else - return readDateTimeTextFallback(datetime, buf, date_lut); + return readDateTimeTextFallback(datetime, buf, date_lut, allowed_date_delimiters, allowed_time_delimiters); } template -inline ReturnType readDateTimeTextImpl(DateTime64 & datetime64, UInt32 scale, ReadBuffer & buf, const DateLUTImpl & date_lut) +inline ReturnType readDateTimeTextImpl(DateTime64 & datetime64, UInt32 scale, ReadBuffer & buf, const DateLUTImpl & date_lut, const char * allowed_date_delimiters = nullptr, const char * allowed_time_delimiters = nullptr) { static constexpr bool throw_exception = std::is_same_v; @@ -1075,7 +1116,7 @@ inline ReturnType readDateTimeTextImpl(DateTime64 & datetime64, UInt32 scale, Re { try { - readDateTimeTextImpl(whole, buf, date_lut); + readDateTimeTextImpl(whole, buf, date_lut, allowed_date_delimiters, allowed_time_delimiters); } catch (const DB::Exception &) { @@ -1085,7 +1126,7 @@ inline ReturnType readDateTimeTextImpl(DateTime64 & datetime64, UInt32 scale, Re } else { - auto ok = readDateTimeTextImpl(whole, buf, date_lut); + auto ok = readDateTimeTextImpl(whole, buf, date_lut, allowed_date_delimiters, allowed_time_delimiters); if (!ok && (buf.eof() || *buf.position() != '.')) return ReturnType(false); } @@ -1168,14 +1209,14 @@ inline void readDateTime64Text(DateTime64 & datetime64, UInt32 scale, ReadBuffer readDateTimeTextImpl(datetime64, scale, buf, date_lut); } -inline bool tryReadDateTimeText(time_t & datetime, ReadBuffer & buf, const DateLUTImpl & time_zone = DateLUT::instance()) +inline bool tryReadDateTimeText(time_t & datetime, ReadBuffer & buf, const DateLUTImpl & time_zone = DateLUT::instance(), const char * allowed_date_delimiters = nullptr, const char * allowed_time_delimiters = nullptr) { - return readDateTimeTextImpl(datetime, buf, time_zone); + return readDateTimeTextImpl(datetime, buf, time_zone, allowed_date_delimiters, allowed_time_delimiters); } -inline bool tryReadDateTime64Text(DateTime64 & datetime64, UInt32 scale, ReadBuffer & buf, const DateLUTImpl & date_lut = DateLUT::instance()) +inline bool tryReadDateTime64Text(DateTime64 & datetime64, UInt32 scale, ReadBuffer & buf, const DateLUTImpl & date_lut = DateLUT::instance(), const char * allowed_date_delimiters = nullptr, const char * allowed_time_delimiters = nullptr) { - return readDateTimeTextImpl(datetime64, scale, buf, date_lut); + return readDateTimeTextImpl(datetime64, scale, buf, date_lut, allowed_date_delimiters, allowed_time_delimiters); } inline void readDateTimeText(LocalDateTime & datetime, ReadBuffer & buf) @@ -1708,6 +1749,7 @@ inline T parse(const char * data, size_t size) T res; ReadBufferFromMemory buf(data, size); readText(res, buf); + assertEOF(buf); return res; } @@ -1715,7 +1757,9 @@ template inline bool tryParse(T & res, const char * data, size_t size) { ReadBufferFromMemory buf(data, size); - return tryReadText(res, buf); + if (!tryReadText(res, buf)) + return false; + return buf.eof(); } template diff --git a/src/IO/S3/Client.cpp b/src/IO/S3/Client.cpp index a966e370ca1..8338a235387 100644 --- a/src/IO/S3/Client.cpp +++ b/src/IO/S3/Client.cpp @@ -46,7 +46,7 @@ namespace ProfileEvents namespace CurrentMetrics { - extern const Metric S3DiskNoKeyErrors; + extern const Metric DiskS3NoSuchKeyErrors; } namespace DB @@ -701,7 +701,7 @@ RequestResult Client::processRequestResult(RequestResult && outcome) const return std::forward(outcome); if (outcome.GetError().GetErrorType() == Aws::S3::S3Errors::NO_SUCH_KEY) - CurrentMetrics::add(CurrentMetrics::S3DiskNoKeyErrors); + CurrentMetrics::add(CurrentMetrics::DiskS3NoSuchKeyErrors); String enriched_message = fmt::format( "{} {}", diff --git a/src/IO/S3/Credentials.cpp b/src/IO/S3/Credentials.cpp index dfb7727fca4..d6f7542da6b 100644 --- a/src/IO/S3/Credentials.cpp +++ b/src/IO/S3/Credentials.cpp @@ -145,12 +145,16 @@ Aws::String AWSEC2MetadataClient::getDefaultCredentialsSecurely() const { String user_agent_string = awsComputeUserAgentString(); auto [new_token, response_code] = getEC2MetadataToken(user_agent_string); - if (response_code == Aws::Http::HttpResponseCode::BAD_REQUEST) + if (response_code == Aws::Http::HttpResponseCode::BAD_REQUEST + || response_code == Aws::Http::HttpResponseCode::REQUEST_NOT_MADE) + { + /// At least the host should be available and reply, otherwise neither IMDSv2 nor IMDSv1 are usable. return {}; + } else if (response_code != Aws::Http::HttpResponseCode::OK || new_token.empty()) { LOG_TRACE(logger, "Calling EC2MetadataService to get token failed, " - "falling back to less secure way. HTTP response code: {}", response_code); + "falling back to a less secure way. HTTP response code: {}", response_code); return getDefaultCredentials(); } @@ -247,7 +251,7 @@ static Aws::String getAWSMetadataEndpoint() return ec2_metadata_service_endpoint; } -std::shared_ptr InitEC2MetadataClient(const Aws::Client::ClientConfiguration & client_configuration) +std::shared_ptr createEC2MetadataClient(const Aws::Client::ClientConfiguration & client_configuration) { auto endpoint = getAWSMetadataEndpoint(); return std::make_shared(client_configuration, endpoint.c_str()); @@ -781,11 +785,13 @@ S3CredentialsProviderChain::S3CredentialsProviderChain( /// EC2MetadataService throttles by delaying the response so the service client should set a large read timeout. /// EC2MetadataService delay is in order of seconds so it only make sense to retry after a couple of seconds. - aws_client_configuration.connectTimeoutMs = 1000; + /// But the connection timeout should be small because there is the case when there is no IMDS at all, + /// like outside of the cloud, on your own machines. + aws_client_configuration.connectTimeoutMs = 10; aws_client_configuration.requestTimeoutMs = 1000; aws_client_configuration.retryStrategy = std::make_shared(1, 1000); - auto ec2_metadata_client = InitEC2MetadataClient(aws_client_configuration); + auto ec2_metadata_client = createEC2MetadataClient(aws_client_configuration); auto config_loader = std::make_shared(ec2_metadata_client, !credentials_configuration.use_insecure_imds_request); AddProvider(std::make_shared(config_loader)); diff --git a/src/IO/S3/Credentials.h b/src/IO/S3/Credentials.h index 95297ab0538..042c48ec15a 100644 --- a/src/IO/S3/Credentials.h +++ b/src/IO/S3/Credentials.h @@ -70,7 +70,7 @@ private: LoggerPtr logger; }; -std::shared_ptr InitEC2MetadataClient(const Aws::Client::ClientConfiguration & client_configuration); +std::shared_ptr createEC2MetadataClient(const Aws::Client::ClientConfiguration & client_configuration); class AWSEC2InstanceProfileConfigLoader : public Aws::Config::AWSProfileConfigLoader { diff --git a/src/IO/S3/PocoHTTPClient.cpp b/src/IO/S3/PocoHTTPClient.cpp index aab7a39534d..de43f34d838 100644 --- a/src/IO/S3/PocoHTTPClient.cpp +++ b/src/IO/S3/PocoHTTPClient.cpp @@ -128,7 +128,7 @@ void PocoHTTPClientConfiguration::updateSchemeAndRegion() } else { - /// In global mode AWS C++ SDK send `us-east-1` but accept switching to another one if being suggested. + /// In global mode AWS C++ SDK sends `us-east-1` but accepts switching to another one if being suggested. region = Aws::Region::AWS_GLOBAL; } } diff --git a/src/IO/S3/URI.cpp b/src/IO/S3/URI.cpp index fead18315d8..9c80b377661 100644 --- a/src/IO/S3/URI.cpp +++ b/src/IO/S3/URI.cpp @@ -1,8 +1,8 @@ #include -#include -#include -#include "Common/Macros.h" + #if USE_AWS_S3 +#include +#include #include #include #include @@ -10,6 +10,7 @@ #include + namespace DB { @@ -40,21 +41,13 @@ URI::URI(const std::string & uri_, bool allow_archive_path_syntax) /// Case when AWS Private Link Interface is being used /// E.g. (bucket.vpce-07a1cd78f1bd55c5f-j3a3vg6w.s3.us-east-1.vpce.amazonaws.com/bucket-name/key) /// https://docs.aws.amazon.com/AmazonS3/latest/userguide/privatelink-interface-endpoints.html - static const RE2 aws_private_link_style_pattern(R"(bucket\.vpce\-([a-z0-9\-.]+)\.vpce.amazonaws.com(:\d{1,5})?)"); + static const RE2 aws_private_link_style_pattern(R"(bucket\.vpce\-([a-z0-9\-.]+)\.vpce\.amazonaws\.com(:\d{1,5})?)"); - /// Case when bucket name and key represented in path of S3 URL. + /// Case when bucket name and key represented in the path of S3 URL. /// E.g. (https://s3.region.amazonaws.com/bucket-name/key) /// https://docs.aws.amazon.com/AmazonS3/latest/dev/VirtualHosting.html#path-style-access static const RE2 path_style_pattern("^/([^/]*)/(.*)"); - static constexpr auto S3 = "S3"; - static constexpr auto S3EXPRESS = "S3EXPRESS"; - static constexpr auto COSN = "COSN"; - static constexpr auto COS = "COS"; - static constexpr auto OBS = "OBS"; - static constexpr auto OSS = "OSS"; - static constexpr auto EOS = "EOS"; - if (allow_archive_path_syntax) std::tie(uri_str, archive_pattern) = getURIAndArchivePattern(uri_); else @@ -85,7 +78,7 @@ URI::URI(const std::string & uri_, bool allow_archive_path_syntax) URIConverter::modifyURI(uri, mapper); } - storage_name = S3; + storage_name = "S3"; if (uri.getHost().empty()) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Host is empty in S3 URI."); @@ -93,11 +86,13 @@ URI::URI(const std::string & uri_, bool allow_archive_path_syntax) /// Extract object version ID from query string. bool has_version_id = false; for (const auto & [query_key, query_value] : uri.getQueryParameters()) + { if (query_key == "versionId") { version_id = query_value; has_version_id = true; } + } /// Poco::URI will ignore '?' when parsing the path, but if there is a versionId in the http parameter, /// '?' can not be used as a wildcard, otherwise it will be ambiguous. @@ -129,15 +124,8 @@ URI::URI(const std::string & uri_, bool allow_archive_path_syntax) } boost::to_upper(name); - /// For S3Express it will look like s3express-eun1-az1, i.e. contain region and AZ info - if (name != S3 && !name.starts_with(S3EXPRESS) && name != COS && name != OBS && name != OSS && name != EOS) - throw Exception( - ErrorCodes::BAD_ARGUMENTS, - "Object storage system name is unrecognized in virtual hosted style S3 URI: {}", - quoteString(name)); - - if (name == COS) - storage_name = COSN; + if (name == "COS") + storage_name = "COSN"; else storage_name = name; } @@ -148,13 +136,22 @@ URI::URI(const std::string & uri_, bool allow_archive_path_syntax) validateBucket(bucket, uri); } else - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Bucket or key name are invalid in S3 URI."); + { + /// Custom endpoint, e.g. a public domain of Cloudflare R2, + /// which could be served by a custom server-side code. + storage_name = "S3"; + bucket = "default"; + is_virtual_hosted_style = false; + endpoint = uri.getScheme() + "://" + uri.getAuthority(); + if (!uri.getPath().empty()) + key = uri.getPath().substr(1); + } } void URI::addRegionToURI(const std::string ®ion) { - if (auto pos = endpoint.find("amazonaws.com"); pos != std::string::npos) - endpoint = endpoint.substr(0, pos) + region + "." + endpoint.substr(pos); + if (auto pos = endpoint.find(".amazonaws.com"); pos != std::string::npos) + endpoint = endpoint.substr(0, pos) + "." + region + endpoint.substr(pos); } void URI::validateBucket(const String & bucket, const Poco::URI & uri) diff --git a/src/IO/S3/URI.h b/src/IO/S3/URI.h index 80e2da96cd4..c8d0b28cd15 100644 --- a/src/IO/S3/URI.h +++ b/src/IO/S3/URI.h @@ -1,14 +1,14 @@ #pragma once -#include -#include - #include "config.h" #if USE_AWS_S3 +#include +#include #include + namespace DB::S3 { @@ -23,7 +23,7 @@ namespace DB::S3 struct URI { Poco::URI uri; - // Custom endpoint if URI scheme is not S3. + // Custom endpoint if URI scheme, if not S3. std::string endpoint; std::string bucket; std::string key; diff --git a/src/IO/S3/getObjectInfo.cpp b/src/IO/S3/getObjectInfo.cpp index 9271ad820e4..a21fb9fce54 100644 --- a/src/IO/S3/getObjectInfo.cpp +++ b/src/IO/S3/getObjectInfo.cpp @@ -54,6 +54,7 @@ namespace ObjectInfo object_info; object_info.size = static_cast(result.GetContentLength()); object_info.last_modification_time = result.GetLastModified().Seconds(); + object_info.etag = result.GetETag(); if (with_metadata) object_info.metadata = result.GetMetadata(); diff --git a/src/IO/S3/getObjectInfo.h b/src/IO/S3/getObjectInfo.h index 32f34f74069..30d4c627d37 100644 --- a/src/IO/S3/getObjectInfo.h +++ b/src/IO/S3/getObjectInfo.h @@ -15,6 +15,7 @@ struct ObjectInfo { size_t size = 0; time_t last_modification_time = 0; + String etag; std::map metadata = {}; /// Set only if getObjectInfo() is called with `with_metadata = true`. }; diff --git a/src/IO/S3Common.cpp b/src/IO/S3Common.cpp index a794cdbcf05..59040bf1fea 100644 --- a/src/IO/S3Common.cpp +++ b/src/IO/S3Common.cpp @@ -294,21 +294,21 @@ void RequestSettings::finishInit(const DB::Settings & settings, bool validate_se /// to avoid losing token bucket state on every config reload, /// which could lead to exceeding limit for short time. /// But it is good enough unless very high `burst` values are used. - if (UInt64 max_get_rps = isChanged("max_get_rps") ? get("max_get_rps").get() : settings.s3_max_get_rps) + if (UInt64 max_get_rps = isChanged("max_get_rps") ? get("max_get_rps").safeGet() : settings.s3_max_get_rps) { size_t default_max_get_burst = settings.s3_max_get_burst ? settings.s3_max_get_burst : (Throttler::default_burst_seconds * max_get_rps); - size_t max_get_burst = isChanged("max_get_burts") ? get("max_get_burst").get() : default_max_get_burst; + size_t max_get_burst = isChanged("max_get_burts") ? get("max_get_burst").safeGet() : default_max_get_burst; get_request_throttler = std::make_shared(max_get_rps, max_get_burst); } - if (UInt64 max_put_rps = isChanged("max_put_rps") ? get("max_put_rps").get() : settings.s3_max_put_rps) + if (UInt64 max_put_rps = isChanged("max_put_rps") ? get("max_put_rps").safeGet() : settings.s3_max_put_rps) { size_t default_max_put_burst = settings.s3_max_put_burst ? settings.s3_max_put_burst : (Throttler::default_burst_seconds * max_put_rps); - size_t max_put_burst = isChanged("max_put_burts") ? get("max_put_burst").get() : default_max_put_burst; + size_t max_put_burst = isChanged("max_put_burts") ? get("max_put_burst").safeGet() : default_max_put_burst; put_request_throttler = std::make_shared(max_put_rps, max_put_burst); } } diff --git a/src/IO/SnappyWriteBuffer.cpp b/src/IO/SnappyWriteBuffer.cpp index ca40d0656d1..0e02b48e1e0 100644 --- a/src/IO/SnappyWriteBuffer.cpp +++ b/src/IO/SnappyWriteBuffer.cpp @@ -16,7 +16,13 @@ namespace ErrorCodes } SnappyWriteBuffer::SnappyWriteBuffer(std::unique_ptr out_, size_t buf_size, char * existing_memory, size_t alignment) - : BufferWithOwnMemory(buf_size, existing_memory, alignment), out(std::move(out_)) + : SnappyWriteBuffer(*out_, buf_size, existing_memory, alignment) +{ + out_holder = std::move(out_); +} + +SnappyWriteBuffer::SnappyWriteBuffer(WriteBuffer & out_, size_t buf_size, char * existing_memory, size_t alignment) + : BufferWithOwnMemory(buf_size, existing_memory, alignment), out(&out_) { } diff --git a/src/IO/SnappyWriteBuffer.h b/src/IO/SnappyWriteBuffer.h index 2ff86fb64ef..b7a084d0f80 100644 --- a/src/IO/SnappyWriteBuffer.h +++ b/src/IO/SnappyWriteBuffer.h @@ -18,6 +18,12 @@ public: char * existing_memory = nullptr, size_t alignment = 0); + explicit SnappyWriteBuffer( + WriteBuffer & out_, + size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE, + char * existing_memory = nullptr, + size_t alignment = 0); + ~SnappyWriteBuffer() override; void finalizeImpl() override { finish(); } @@ -28,7 +34,9 @@ private: void finishImpl(); void finish(); - std::unique_ptr out; + WriteBuffer * out; + std::unique_ptr out_holder; + bool finished = false; String uncompress_buffer; diff --git a/src/IO/WriteBuffer.h b/src/IO/WriteBuffer.h index 4759f96a235..84b1079b824 100644 --- a/src/IO/WriteBuffer.h +++ b/src/IO/WriteBuffer.h @@ -64,7 +64,8 @@ public: } bytes += bytes_in_buffer; - pos = working_buffer.begin(); + pos = working_buffer.begin() + nextimpl_working_buffer_offset; + nextimpl_working_buffer_offset = 0; } /// Calling finalize() in the destructor of derived classes is a bad practice. @@ -164,6 +165,11 @@ protected: bool finalized = false; bool canceled = false; + /// The number of bytes to preserve from the initial position of `working_buffer` + /// buffer. Apparently this is an additional out-parameter for nextImpl(), + /// not a real field. + size_t nextimpl_working_buffer_offset = 0; + private: /** Write the data in the buffer (from the beginning of the buffer to the current position). * Throw an exception if something is wrong. diff --git a/src/IO/WriteBufferFromPocoSocket.cpp b/src/IO/WriteBufferFromPocoSocket.cpp index 5ed4dbdc787..ffb38a384a0 100644 --- a/src/IO/WriteBufferFromPocoSocket.cpp +++ b/src/IO/WriteBufferFromPocoSocket.cpp @@ -183,6 +183,7 @@ WriteBufferFromPocoSocket::WriteBufferFromPocoSocket(Poco::Net::Socket & socket_ , socket(socket_) , peer_address(socket.peerAddress()) , our_address(socket.address()) + , write_event(ProfileEvents::end()) , socket_description("socket (" + peer_address.toString() + ")") { } diff --git a/src/IO/WriteBufferFromPocoSocketChunked.cpp b/src/IO/WriteBufferFromPocoSocketChunked.cpp new file mode 100644 index 00000000000..9da46ee2d10 --- /dev/null +++ b/src/IO/WriteBufferFromPocoSocketChunked.cpp @@ -0,0 +1,210 @@ +#include +#include +#include + + +namespace +{ + +template +void setValue(T * typed_ptr, std::type_identity_t val) +{ + memcpy(static_cast(typed_ptr), &val, sizeof(T)); +} + +} + +namespace DB +{ + +WriteBufferFromPocoSocketChunked::WriteBufferFromPocoSocketChunked(Poco::Net::Socket & socket_, size_t buf_size) + : WriteBufferFromPocoSocketChunked(socket_, ProfileEvents::end(), buf_size) +{} + +WriteBufferFromPocoSocketChunked::WriteBufferFromPocoSocketChunked(Poco::Net::Socket & socket_, const ProfileEvents::Event & write_event_, size_t buf_size) + : WriteBufferFromPocoSocket( + socket_, write_event_, + std::clamp(buf_size, sizeof(*chunk_size_ptr) + 1, static_cast(std::numeric_limits>::max()))), + log(getLogger("Protocol")) +{} + +void WriteBufferFromPocoSocketChunked::enableChunked() +{ + chunked = true; + /// Initialize next chunk + chunk_size_ptr = reinterpret_cast(pos); + pos += std::min(available(), sizeof(*chunk_size_ptr)); + /// Pretend finishChunk() was just called to prevent sending empty chunk if finishChunk() called immediately + last_finish_chunk = chunk_size_ptr; +} + +void WriteBufferFromPocoSocketChunked::finishChunk() +{ + if (!chunked) + return; + + if (pos <= reinterpret_cast(chunk_size_ptr) + sizeof(*chunk_size_ptr)) + { + /// Prevent duplicate finish chunk (and finish chunk right after enableChunked()) + if (chunk_size_ptr == last_finish_chunk) + return; + + /// If current chunk is empty it means we are finishing a chunk previously sent by next(), + /// we want to convert current chunk header into end-of-chunk marker and initialize next chunk. + /// We don't need to worry about if it's the end of the buffer because next() always sends the whole buffer + /// so it should be a beginning of the buffer. + + chassert(reinterpret_cast(chunk_size_ptr) == working_buffer.begin()); + + setValue(chunk_size_ptr, 0); + /// Initialize next chunk + chunk_size_ptr = reinterpret_cast(pos); + pos += std::min(available(), sizeof(*chunk_size_ptr)); + + last_finish_chunk = chunk_size_ptr; + + return; + } + + /// Previously finished chunk wasn't sent yet + if (last_finish_chunk == chunk_size_ptr) + { + chunk_started = false; + LOG_TEST(log, "{} -> {} Chunk send ended.", ourAddress().toString(), peerAddress().toString()); + } + + /// Fill up current chunk size + setValue(chunk_size_ptr, toLittleEndian(static_cast(pos - reinterpret_cast(chunk_size_ptr) - sizeof(*chunk_size_ptr)))); + + if (!chunk_started) + LOG_TEST(log, "{} -> {} Chunk send started. Message {}, size {}", + ourAddress().toString(), peerAddress().toString(), + static_cast(*(reinterpret_cast(chunk_size_ptr) + sizeof(*chunk_size_ptr))), + *chunk_size_ptr); + else + { + chunk_started = false; + LOG_TEST(log, "{} -> {} Chunk send continued. Size {}", ourAddress().toString(), peerAddress().toString(), *chunk_size_ptr); + } + + LOG_TEST(log, "{} -> {} Chunk send ended.", ourAddress().toString(), peerAddress().toString()); + + if (available() < sizeof(*chunk_size_ptr)) + { + finishing = available(); + pos += available(); + chunk_size_ptr = reinterpret_cast(pos); + last_finish_chunk = chunk_size_ptr; + return; + } + + /// Buffer end-of-chunk + setValue(reinterpret_cast(pos), 0); + pos += sizeof(*chunk_size_ptr); + /// Initialize next chunk + chunk_size_ptr = reinterpret_cast(pos); + pos += std::min(available(), sizeof(*chunk_size_ptr)); + + last_finish_chunk = chunk_size_ptr; +} + +WriteBufferFromPocoSocketChunked::~WriteBufferFromPocoSocketChunked() +{ + try + { + finalize(); + } + catch (...) + { + tryLogCurrentException(__PRETTY_FUNCTION__); + } +} + +void WriteBufferFromPocoSocketChunked::nextImpl() +{ + if (!chunked) + { + WriteBufferFromPocoSocket::nextImpl(); + return; + } + + /// next() after finishChunk at the end of the buffer + if (finishing < sizeof(*chunk_size_ptr)) + { + pos -= finishing; + /// Send current chunk + WriteBufferFromPocoSocket::nextImpl(); + /// Send end-of-chunk directly + UInt32 s = 0; + socketSendBytes(reinterpret_cast(&s), sizeof(s)); + + finishing = sizeof(*chunk_size_ptr); + + /// Initialize next chunk + chunk_size_ptr = reinterpret_cast(working_buffer.begin()); + nextimpl_working_buffer_offset = sizeof(*chunk_size_ptr); + + last_finish_chunk = chunk_size_ptr; + + return; + } + + /// Prevent sending empty chunk + if (offset() == sizeof(*chunk_size_ptr)) + { + nextimpl_working_buffer_offset = sizeof(*chunk_size_ptr); + return; + } + + /// Finish chunk at the end of the buffer + if (working_buffer.end() - reinterpret_cast(chunk_size_ptr) <= static_cast(sizeof(*chunk_size_ptr))) + { + pos = reinterpret_cast(chunk_size_ptr); + /// Send current chunk + WriteBufferFromPocoSocket::nextImpl(); + /// Initialize next chunk + chunk_size_ptr = reinterpret_cast(working_buffer.begin()); + nextimpl_working_buffer_offset = sizeof(*chunk_size_ptr); + + last_finish_chunk = nullptr; + + return; + } + + bool initialize_last_finish_chunk = false; + if (pos - reinterpret_cast(chunk_size_ptr) == sizeof(*chunk_size_ptr)) // next() after finishChunk + { + pos -= sizeof(*chunk_size_ptr); + initialize_last_finish_chunk = true; + } + else // fill up current chunk size + { + setValue(chunk_size_ptr, toLittleEndian(static_cast(pos - reinterpret_cast(chunk_size_ptr) - sizeof(*chunk_size_ptr)))); + if (!chunk_started) + { + chunk_started = true; + LOG_TEST(log, "{} -> {} Chunk send started. Message {}, size {}", + ourAddress().toString(), peerAddress().toString(), + static_cast(*(reinterpret_cast(chunk_size_ptr) + sizeof(*chunk_size_ptr))), + *chunk_size_ptr); + } + else + LOG_TEST(log, "{} -> {} Chunk send continued. Size {}", ourAddress().toString(), peerAddress().toString(), *chunk_size_ptr); + } + /// Send current chunk + WriteBufferFromPocoSocket::nextImpl(); + /// Initialize next chunk + chunk_size_ptr = reinterpret_cast(working_buffer.begin()); + nextimpl_working_buffer_offset = sizeof(*chunk_size_ptr); + + last_finish_chunk = initialize_last_finish_chunk ? chunk_size_ptr : nullptr; +} + +void WriteBufferFromPocoSocketChunked::finalizeImpl() +{ + if (chunked && offset() == sizeof(*chunk_size_ptr)) + pos -= sizeof(*chunk_size_ptr); + WriteBufferFromPocoSocket::finalizeImpl(); +} + +} diff --git a/src/IO/WriteBufferFromPocoSocketChunked.h b/src/IO/WriteBufferFromPocoSocketChunked.h new file mode 100644 index 00000000000..13a277e3bfb --- /dev/null +++ b/src/IO/WriteBufferFromPocoSocketChunked.h @@ -0,0 +1,36 @@ +#pragma once + +#include +#include +#include + + +namespace DB +{ + +class WriteBufferFromPocoSocketChunked: public WriteBufferFromPocoSocket +{ +public: + explicit WriteBufferFromPocoSocketChunked(Poco::Net::Socket & socket_, size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE); + explicit WriteBufferFromPocoSocketChunked(Poco::Net::Socket & socket_, const ProfileEvents::Event & write_event_, size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE); + + void enableChunked(); + void finishChunk(); + ~WriteBufferFromPocoSocketChunked() override; + +protected: + void nextImpl() override; + void finalizeImpl() override; + Poco::Net::SocketAddress peerAddress() const { return peer_address; } + Poco::Net::SocketAddress ourAddress() const { return our_address; } + +private: + LoggerPtr log; + bool chunked = false; + UInt32 * last_finish_chunk = nullptr; // pointer to the last chunk header created by finishChunk + bool chunk_started = false; // chunk started flag + UInt32 * chunk_size_ptr = nullptr; // pointer to the chunk size holder in the buffer + size_t finishing = sizeof(*chunk_size_ptr); // indicates not enough buffer for end-of-chunk marker +}; + +} diff --git a/src/IO/parseDateTimeBestEffort.cpp b/src/IO/parseDateTimeBestEffort.cpp index e046e837689..f220577f2cb 100644 --- a/src/IO/parseDateTimeBestEffort.cpp +++ b/src/IO/parseDateTimeBestEffort.cpp @@ -82,13 +82,14 @@ struct DateTimeSubsecondPart UInt8 digits; }; -template +template ReturnType parseDateTimeBestEffortImpl( time_t & res, ReadBuffer & in, const DateLUTImpl & local_time_zone, const DateLUTImpl & utc_time_zone, - DateTimeSubsecondPart * fractional) + DateTimeSubsecondPart * fractional, + const char * allowed_date_delimiters = nullptr) { auto on_error = [&](int error_code [[maybe_unused]], FormatStringHelper fmt_string [[maybe_unused]], @@ -170,22 +171,36 @@ ReturnType parseDateTimeBestEffortImpl( fractional->digits = 3; readDecimalNumber<3>(fractional->value, digits + 10); } + else if constexpr (strict) + { + /// Fractional part is not allowed. + return on_error(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot read DateTime: unexpected fractional part"); + } return ReturnType(true); } else if (num_digits == 10 && !year && !has_time) { + if (strict && month) + return on_error(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot read DateTime: month component is duplicated"); + /// This is unix timestamp. readDecimalNumber<10>(res, digits); return ReturnType(true); } else if (num_digits == 9 && !year && !has_time) { + if (strict && month) + return on_error(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot read DateTime: month component is duplicated"); + /// This is unix timestamp. readDecimalNumber<9>(res, digits); return ReturnType(true); } else if (num_digits == 14 && !year && !has_time) { + if (strict && month) + return on_error(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot read DateTime: month component is duplicated"); + /// This is YYYYMMDDhhmmss readDecimalNumber<4>(year, digits); readDecimalNumber<2>(month, digits + 4); @@ -197,6 +212,9 @@ ReturnType parseDateTimeBestEffortImpl( } else if (num_digits == 8 && !year) { + if (strict && month) + return on_error(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot read DateTime: month component is duplicated"); + /// This is YYYYMMDD readDecimalNumber<4>(year, digits); readDecimalNumber<2>(month, digits + 4); @@ -272,6 +290,9 @@ ReturnType parseDateTimeBestEffortImpl( else return on_error(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot read DateTime: unexpected number of decimal digits after year and month: {}", num_digits); } + + if (!isSymbolIn(delimiter_after_year, allowed_date_delimiters)) + return on_error(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot read DateTime: '{}' delimiter between date parts is not allowed", delimiter_after_year); } } else if (num_digits == 2 || num_digits == 1) @@ -403,9 +424,16 @@ ReturnType parseDateTimeBestEffortImpl( else { if (day_of_month) + { + if (strict && hour) + return on_error(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot read DateTime: hour component is duplicated"); + hour = hour_or_day_of_month_or_month; + } else + { day_of_month = hour_or_day_of_month_or_month; + } } } else if (num_digits != 0) @@ -446,6 +474,11 @@ ReturnType parseDateTimeBestEffortImpl( fractional->digits = num_digits; readDecimalNumber(fractional->value, num_digits, digits); } + else if (strict) + { + /// Fractional part is not allowed. + return on_error(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot read DateTime: unexpected fractional part"); + } } else if (c == '+' || c == '-') { @@ -582,12 +615,24 @@ ReturnType parseDateTimeBestEffortImpl( return on_error(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot read DateTime: neither Date nor Time was parsed successfully"); if (!day_of_month) + { + if constexpr (strict) + return on_error(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot read DateTime: day of month is required"); day_of_month = 1; + } + if (!month) + { + if constexpr (strict) + return on_error(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot read DateTime: month is required"); month = 1; + } if (!year) { + if constexpr (strict) + return on_error(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot read DateTime: year is required"); + /// If year is not specified, it will be the current year if the date is unknown or not greater than today, /// otherwise it will be the previous year. /// This convoluted logic is needed to parse the syslog format, which looks as follows: "Mar 3 01:33:48". @@ -641,6 +686,20 @@ ReturnType parseDateTimeBestEffortImpl( } }; + if constexpr (strict) + { + if constexpr (is_64) + { + if (year < 1900) + return on_error(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot read DateTime64: year {} is less than minimum supported year 1900", year); + } + else + { + if (year < 1970) + return on_error(ErrorCodes::CANNOT_PARSE_DATETIME, "Cannot read DateTime: year {} is less than minimum supported year 1970", year); + } + } + if (has_time_zone_offset) { res = utc_time_zone.makeDateTime(year, month, day_of_month, hour, minute, second); @@ -654,20 +713,20 @@ ReturnType parseDateTimeBestEffortImpl( return ReturnType(true); } -template -ReturnType parseDateTime64BestEffortImpl(DateTime64 & res, UInt32 scale, ReadBuffer & in, const DateLUTImpl & local_time_zone, const DateLUTImpl & utc_time_zone) +template +ReturnType parseDateTime64BestEffortImpl(DateTime64 & res, UInt32 scale, ReadBuffer & in, const DateLUTImpl & local_time_zone, const DateLUTImpl & utc_time_zone, const char * allowed_date_delimiters = nullptr) { time_t whole; DateTimeSubsecondPart subsecond = {0, 0}; // needs to be explicitly initialized sine it could be missing from input string if constexpr (std::is_same_v) { - if (!parseDateTimeBestEffortImpl(whole, in, local_time_zone, utc_time_zone, &subsecond)) + if (!parseDateTimeBestEffortImpl(whole, in, local_time_zone, utc_time_zone, &subsecond, allowed_date_delimiters)) return false; } else { - parseDateTimeBestEffortImpl(whole, in, local_time_zone, utc_time_zone, &subsecond); + parseDateTimeBestEffortImpl(whole, in, local_time_zone, utc_time_zone, &subsecond, allowed_date_delimiters); } @@ -730,4 +789,24 @@ bool tryParseDateTime64BestEffortUS(DateTime64 & res, UInt32 scale, ReadBuffer & return parseDateTime64BestEffortImpl(res, scale, in, local_time_zone, utc_time_zone); } +bool tryParseDateTimeBestEffortStrict(time_t & res, ReadBuffer & in, const DateLUTImpl & local_time_zone, const DateLUTImpl & utc_time_zone, const char * allowed_date_delimiters) +{ + return parseDateTimeBestEffortImpl(res, in, local_time_zone, utc_time_zone, nullptr, allowed_date_delimiters); +} + +bool tryParseDateTimeBestEffortUSStrict(time_t & res, ReadBuffer & in, const DateLUTImpl & local_time_zone, const DateLUTImpl & utc_time_zone, const char * allowed_date_delimiters) +{ + return parseDateTimeBestEffortImpl(res, in, local_time_zone, utc_time_zone, nullptr, allowed_date_delimiters); +} + +bool tryParseDateTime64BestEffortStrict(DateTime64 & res, UInt32 scale, ReadBuffer & in, const DateLUTImpl & local_time_zone, const DateLUTImpl & utc_time_zone, const char * allowed_date_delimiters) +{ + return parseDateTime64BestEffortImpl(res, scale, in, local_time_zone, utc_time_zone, allowed_date_delimiters); +} + +bool tryParseDateTime64BestEffortUSStrict(DateTime64 & res, UInt32 scale, ReadBuffer & in, const DateLUTImpl & local_time_zone, const DateLUTImpl & utc_time_zone, const char * allowed_date_delimiters) +{ + return parseDateTime64BestEffortImpl(res, scale, in, local_time_zone, utc_time_zone, allowed_date_delimiters); +} + } diff --git a/src/IO/parseDateTimeBestEffort.h b/src/IO/parseDateTimeBestEffort.h index 22af44f9e76..6dd052b67a3 100644 --- a/src/IO/parseDateTimeBestEffort.h +++ b/src/IO/parseDateTimeBestEffort.h @@ -63,4 +63,12 @@ void parseDateTime64BestEffort(DateTime64 & res, UInt32 scale, ReadBuffer & in, bool tryParseDateTime64BestEffort(DateTime64 & res, UInt32 scale, ReadBuffer & in, const DateLUTImpl & local_time_zone, const DateLUTImpl & utc_time_zone); void parseDateTime64BestEffortUS(DateTime64 & res, UInt32 scale, ReadBuffer & in, const DateLUTImpl & local_time_zone, const DateLUTImpl & utc_time_zone); bool tryParseDateTime64BestEffortUS(DateTime64 & res, UInt32 scale, ReadBuffer & in, const DateLUTImpl & local_time_zone, const DateLUTImpl & utc_time_zone); + +/// More strict version of best effort parsing. Requires day, month and year to be present, checks for allowed +/// delimiters between date components, makes additional correctness checks. Used in schema inference if date times. +bool tryParseDateTimeBestEffortStrict(time_t & res, ReadBuffer & in, const DateLUTImpl & local_time_zone, const DateLUTImpl & utc_time_zone, const char * allowed_date_delimiters); +bool tryParseDateTimeBestEffortUSStrict(time_t & res, ReadBuffer & in, const DateLUTImpl & local_time_zone, const DateLUTImpl & utc_time_zone, const char * allowed_date_delimiters); +bool tryParseDateTime64BestEffortStrict(DateTime64 & res, UInt32 scale, ReadBuffer & in, const DateLUTImpl & local_time_zone, const DateLUTImpl & utc_time_zone, const char * allowed_date_delimiters); +bool tryParseDateTime64BestEffortUSStrict(DateTime64 & res, UInt32 scale, ReadBuffer & in, const DateLUTImpl & local_time_zone, const DateLUTImpl & utc_time_zone, const char * allowed_date_delimiters); + } diff --git a/src/IO/tests/gtest_s3_uri.cpp b/src/IO/tests/gtest_s3_uri.cpp index 0ec28f80072..c0bf7fcb28a 100644 --- a/src/IO/tests/gtest_s3_uri.cpp +++ b/src/IO/tests/gtest_s3_uri.cpp @@ -206,11 +206,6 @@ TEST(S3UriTest, validPatterns) } } -TEST_P(S3UriTest, invalidPatterns) -{ - ASSERT_ANY_THROW(S3::URI new_uri(GetParam())); -} - TEST(S3UriTest, versionIdChecks) { for (const auto& test_case : TestCases) @@ -223,19 +218,5 @@ TEST(S3UriTest, versionIdChecks) } } -INSTANTIATE_TEST_SUITE_P( - S3, - S3UriTest, - testing::Values( - "https:///", - "https://.s3.amazonaws.com/key", - "https://s3.amazonaws.com/key", - "https://jokserfn.s3amazonaws.com/key", - "https://s3.amazonaws.com//", - "https://amazonaws.com/", - "https://amazonaws.com//", - "https://amazonaws.com//key")); - } - #endif diff --git a/src/Interpreters/ActionsDAG.cpp b/src/Interpreters/ActionsDAG.cpp index 4aaecc491e0..2a594839c6a 100644 --- a/src/Interpreters/ActionsDAG.cpp +++ b/src/Interpreters/ActionsDAG.cpp @@ -100,6 +100,13 @@ bool isConstantFromScalarSubquery(const ActionsDAG::Node * node) } +bool ActionsDAG::Node::isDeterministic() const +{ + bool deterministic_if_func = type != ActionType::FUNCTION || function_base->isDeterministic(); + bool deterministic_if_const = type != ActionType::COLUMN || is_deterministic_constant; + return deterministic_if_func && deterministic_if_const; +} + void ActionsDAG::Node::toTree(JSONBuilder::JSONMap & map) const { map.add("Node Type", magic_enum::enum_name(type)); @@ -294,11 +301,11 @@ const ActionsDAG::Node & ActionsDAG::addCast(const Node & node_to_cast, const Da column.column = DataTypeString().createColumnConst(0, cast_type_constant_value); column.type = std::make_shared(); - const auto * cast_type_constant_node = &addColumn(std::move(column)); + const auto * cast_type_constant_node = &addColumn(column); ActionsDAG::NodeRawConstPtrs children = {&node_to_cast, cast_type_constant_node}; - FunctionOverloadResolverPtr func_builder_cast = createInternalCastOverloadResolver(CastType::nonAccurate, {}); + auto func_base_cast = createInternalCast(ColumnWithTypeAndName{node_to_cast.result_type, node_to_cast.result_name}, cast_type, CastType::nonAccurate, {}); - return addFunction(func_builder_cast, std::move(children), result_name); + return addFunction(func_base_cast, std::move(children), result_name); } const ActionsDAG::Node & ActionsDAG::addFunctionImpl( @@ -318,7 +325,6 @@ const ActionsDAG::Node & ActionsDAG::addFunctionImpl( node.function_base = function_base; node.result_type = result_type; node.function = node.function_base->prepare(arguments); - node.is_deterministic = node.function_base->isDeterministic(); /// If all arguments are constants, and function is suitable to be executed in 'prepare' stage - execute function. if (node.function_base->isSuitableForConstantFolding()) @@ -536,64 +542,99 @@ void ActionsDAG::removeUnusedActions(bool allow_remove_inputs, bool allow_consta void ActionsDAG::removeUnusedActions(const std::unordered_set & used_inputs, bool allow_constant_folding) { - std::unordered_set visited_nodes; - std::stack stack; - - for (const auto * node : outputs) - { - visited_nodes.insert(node); - stack.push(const_cast(node)); - } + NodeRawConstPtrs roots; + roots.reserve(outputs.size() + used_inputs.size()); + roots = outputs; for (auto & node : nodes) { /// We cannot remove arrayJoin because it changes the number of rows. - bool is_array_join = node.type == ActionType::ARRAY_JOIN; - - if (is_array_join && !visited_nodes.contains(&node)) - { - visited_nodes.insert(&node); - stack.push(&node); - } + if (node.type == ActionType::ARRAY_JOIN) + roots.push_back(&node); if (node.type == ActionType::INPUT && used_inputs.contains(&node)) - visited_nodes.insert(&node); + roots.push_back(&node); } - while (!stack.empty()) + std::unordered_set required_nodes; + std::unordered_set non_deterministic_nodes; + + struct Frame { - auto * node = stack.top(); - stack.pop(); + const ActionsDAG::Node * node; + size_t next_child_to_visit = 0; + }; - /// Constant folding. - if (allow_constant_folding && !node->children.empty() && node->column && isColumnConst(*node->column)) + std::stack stack; + + enum class VisitStage { NonDeterministic, Required }; + + for (auto stage : {VisitStage::NonDeterministic, VisitStage::Required}) + { + required_nodes.clear(); + + for (const auto * root : roots) { - node->type = ActionsDAG::ActionType::COLUMN; - - for (const auto & child : node->children) + if (!required_nodes.contains(root)) { - if (!child->is_deterministic) + required_nodes.insert(root); + stack.push({.node = root}); + } + + while (!stack.empty()) + { + auto & frame = stack.top(); + auto * node = const_cast(frame.node); + + while (frame.next_child_to_visit < node->children.size()) { - node->is_deterministic = false; - break; + const auto * child = node->children[frame.next_child_to_visit]; + ++frame.next_child_to_visit; + + if (!required_nodes.contains(child)) + { + required_nodes.insert(child); + stack.push({.node = child}); + break; + } + } + + if (stack.top().node != node) + continue; + + stack.pop(); + + if (stage == VisitStage::Required) + continue; + + if (!node->isDeterministic()) + non_deterministic_nodes.insert(node); + else + { + for (const auto * child : node->children) + { + if (non_deterministic_nodes.contains(child)) + { + non_deterministic_nodes.insert(node); + break; + } + } + } + + /// Constant folding. + if (allow_constant_folding && !node->children.empty() + && node->column && isColumnConst(*node->column)) + { + node->type = ActionsDAG::ActionType::COLUMN; + node->children.clear(); + node->is_deterministic_constant = !non_deterministic_nodes.contains(node); } } - - node->children.clear(); - } - - for (const auto * child : node->children) - { - if (!visited_nodes.contains(child)) - { - stack.push(const_cast(child)); - visited_nodes.insert(child); - } } } - std::erase_if(nodes, [&](const Node & node) { return !visited_nodes.contains(&node); }); - std::erase_if(inputs, [&](const Node * node) { return !visited_nodes.contains(node); }); + std::erase_if(nodes, [&](const Node & node) { return !required_nodes.contains(&node); }); + std::erase_if(inputs, [&](const Node * node) { return !required_nodes.contains(node); }); } @@ -1379,7 +1420,7 @@ bool ActionsDAG::trivial() const void ActionsDAG::assertDeterministic() const { for (const auto & node : nodes) - if (!node.is_deterministic) + if (!node.isDeterministic()) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expression must be deterministic but it contains non-deterministic part `{}`", node.result_name); } @@ -1387,7 +1428,7 @@ void ActionsDAG::assertDeterministic() const bool ActionsDAG::hasNonDeterministic() const { for (const auto & node : nodes) - if (!node.is_deterministic) + if (!node.isDeterministic()) return true; return false; } @@ -1506,11 +1547,11 @@ ActionsDAG ActionsDAG::makeConvertingActions( const auto * left_arg = dst_node; CastDiagnostic diagnostic = {dst_node->result_name, res_elem.name}; - FunctionOverloadResolverPtr func_builder_cast - = createInternalCastOverloadResolver(CastType::nonAccurate, std::move(diagnostic)); + ColumnWithTypeAndName left_column{nullptr, dst_node->result_type, {}}; + auto func_base_cast = createInternalCast(std::move(left_column), res_elem.type, CastType::nonAccurate, std::move(diagnostic)); NodeRawConstPtrs children = { left_arg, right_arg }; - dst_node = &actions_dag.addFunction(func_builder_cast, std::move(children), {}); + dst_node = &actions_dag.addFunction(func_base_cast, std::move(children), {}); } if (dst_node->column && isColumnConst(*dst_node->column) && !(res_elem.column && isColumnConst(*res_elem.column))) diff --git a/src/Interpreters/ActionsDAG.h b/src/Interpreters/ActionsDAG.h index 43c1b41a240..ee2b3fbf4f2 100644 --- a/src/Interpreters/ActionsDAG.h +++ b/src/Interpreters/ActionsDAG.h @@ -80,13 +80,15 @@ public: ExecutableFunctionPtr function; /// If function is a compiled statement. bool is_function_compiled = false; - /// It is deterministic (See IFunction::isDeterministic). - /// This property is kept after constant folding of non-deterministic functions like 'now', 'today'. - bool is_deterministic = true; + /// It is a constant calculated from deterministic functions (See IFunction::isDeterministic). + /// This property is kept after constant folding of non-deterministic functions like 'now', 'today'. + bool is_deterministic_constant = true; /// For COLUMN node and propagated constants. ColumnPtr column; + /// If result of this not is deterministic. Checks only this node, not a subtree. + bool isDeterministic() const; void toTree(JSONBuilder::JSONMap & map) const; }; diff --git a/src/Interpreters/ActionsVisitor.cpp b/src/Interpreters/ActionsVisitor.cpp index e1b7e92ee5d..368eb8174f0 100644 --- a/src/Interpreters/ActionsVisitor.cpp +++ b/src/Interpreters/ActionsVisitor.cpp @@ -131,7 +131,7 @@ static Block createBlockFromCollection(const Collection & collection, const Data throw Exception(ErrorCodes::INCORRECT_ELEMENT_OF_SET, "Invalid type in set. Expected tuple, got {}", String(value.getTypeName())); - const auto & tuple = value.template get(); + const auto & tuple = value.template safeGet(); size_t tuple_size = tuple.size(); if (tuple_size != columns_num) throw Exception(ErrorCodes::INCORRECT_ELEMENT_OF_SET, "Incorrect size of tuple in set: {} instead of {}", @@ -233,7 +233,7 @@ static Block createBlockFromAST(const ASTPtr & node, const DataTypes & types, Co "Invalid type of set. Expected tuple, got {}", function_result.getTypeName()); - tuple = &function_result.get(); + tuple = &function_result.safeGet(); } /// Tuple can be represented as a literal in AST. @@ -246,7 +246,7 @@ static Block createBlockFromAST(const ASTPtr & node, const DataTypes & types, Co "Invalid type in set. Expected tuple, got {}", literal->value.getTypeName()); - tuple = &literal->value.get(); + tuple = &literal->value.safeGet(); } assert(tuple || func); @@ -332,14 +332,14 @@ Block createBlockForSet( if (type_index == TypeIndex::Tuple) { const DataTypes & value_types = assert_cast(right_arg_type.get())->getElements(); - block = createBlockFromCollection(right_arg_value.get(), value_types, set_element_types, tranform_null_in); + block = createBlockFromCollection(right_arg_value.safeGet(), value_types, set_element_types, tranform_null_in); } else if (type_index == TypeIndex::Array) { const auto* right_arg_array_type = assert_cast(right_arg_type.get()); - size_t right_arg_array_size = right_arg_value.get().size(); + size_t right_arg_array_size = right_arg_value.safeGet().size(); DataTypes value_types(right_arg_array_size, right_arg_array_type->getNestedType()); - block = createBlockFromCollection(right_arg_value.get(), value_types, set_element_types, tranform_null_in); + block = createBlockFromCollection(right_arg_value.safeGet(), value_types, set_element_types, tranform_null_in); } else throw_unsupported_type(right_arg_type); diff --git a/src/Interpreters/AddDefaultDatabaseVisitor.h b/src/Interpreters/AddDefaultDatabaseVisitor.h index 356bffa75e9..d59fd35df77 100644 --- a/src/Interpreters/AddDefaultDatabaseVisitor.h +++ b/src/Interpreters/AddDefaultDatabaseVisitor.h @@ -1,6 +1,7 @@ #pragma once #include +#include #include #include #include @@ -100,6 +101,7 @@ private: const String database_name; std::set external_tables; + mutable std::unordered_set with_aliases; bool only_replace_current_database_function = false; bool only_replace_in_join = false; @@ -117,6 +119,10 @@ private: void visit(ASTSelectQuery & select, ASTPtr &) const { + if (select.recursive_with) + for (const auto & child : select.with()->children) + with_aliases.insert(child->as()->name); + if (select.tables()) tryVisit(select.refTables()); @@ -165,6 +171,9 @@ private: /// There is temporary table with such name, should not be rewritten. if (external_tables.contains(identifier.shortName())) return; + /// This is WITH RECURSIVE alias. + if (with_aliases.contains(identifier.name())) + return; auto qualified_identifier = std::make_shared(database_name, identifier.name()); if (!identifier.alias.empty()) @@ -201,7 +210,7 @@ private: if (literal_value.getType() != Field::Types::String) continue; - auto dictionary_name = literal_value.get(); + auto dictionary_name = literal_value.safeGet(); auto qualified_dictionary_name = context->getExternalDictionariesLoader().qualifyDictionaryNameWithDatabase(dictionary_name, context); literal_value = qualified_dictionary_name.getFullName(); } diff --git a/src/Interpreters/Aggregator.cpp b/src/Interpreters/Aggregator.cpp index 90bbde8be35..d1aa8a0fff0 100644 --- a/src/Interpreters/Aggregator.cpp +++ b/src/Interpreters/Aggregator.cpp @@ -3307,6 +3307,17 @@ void NO_INLINE Aggregator::destroyImpl(Table & table) const data = nullptr; }); + + if constexpr (Method::low_cardinality_optimization || Method::one_key_nullable_optimization) + { + if (table.getNullKeyData() != nullptr) + { + for (size_t i = 0; i < params.aggregates_size; ++i) + aggregate_functions[i]->destroy(table.getNullKeyData() + offsets_of_aggregate_states[i]); + + table.getNullKeyData() = nullptr; + } + } } diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp index 13c70b38543..b1a1f629b00 100644 --- a/src/Interpreters/Cache/FileCache.cpp +++ b/src/Interpreters/Cache/FileCache.cpp @@ -804,7 +804,8 @@ bool FileCache::tryReserve( const size_t size, FileCacheReserveStat & reserve_stat, const UserInfo & user, - size_t lock_wait_timeout_milliseconds) + size_t lock_wait_timeout_milliseconds, + std::string & failure_reason) { ProfileEventTimeIncrement watch(ProfileEvents::FilesystemCacheReserveMicroseconds); @@ -817,6 +818,7 @@ bool FileCache::tryReserve( if (cache_is_being_resized.load(std::memory_order_relaxed)) { ProfileEvents::increment(ProfileEvents::FilesystemCacheFailToReserveSpaceBecauseOfCacheResize); + failure_reason = "cache is being resized"; return false; } @@ -824,6 +826,7 @@ bool FileCache::tryReserve( if (!cache_lock) { ProfileEvents::increment(ProfileEvents::FilesystemCacheFailToReserveSpaceBecauseOfLockContention); + failure_reason = "cache contention"; return false; } @@ -847,6 +850,7 @@ bool FileCache::tryReserve( LOG_TEST(log, "Query limit exceeded, space reservation failed, " "recache_on_query_limit_exceeded is disabled (while reserving for {}:{})", file_segment.key(), file_segment.offset()); + failure_reason = "query limit exceeded"; return false; } @@ -877,6 +881,7 @@ bool FileCache::tryReserve( if (!query_priority->collectCandidatesForEviction( size, required_elements_num, reserve_stat, eviction_candidates, {}, user.user_id, cache_lock)) { + failure_reason = "cannot evict enough space for query limit"; return false; } @@ -891,11 +896,15 @@ bool FileCache::tryReserve( if (!main_priority->collectCandidatesForEviction( size, required_elements_num, reserve_stat, eviction_candidates, queue_iterator, user.user_id, cache_lock)) { + failure_reason = "cannot evict enough space"; return false; } if (!file_segment.getKeyMetadata()->createBaseDirectory()) + { + failure_reason = "not enough space on device"; return false; + } if (eviction_candidates.size() > 0) { diff --git a/src/Interpreters/Cache/FileCache.h b/src/Interpreters/Cache/FileCache.h index 07be802a940..efa504689eb 100644 --- a/src/Interpreters/Cache/FileCache.h +++ b/src/Interpreters/Cache/FileCache.h @@ -165,7 +165,8 @@ public: size_t size, FileCacheReserveStat & stat, const UserInfo & user, - size_t lock_wait_timeout_milliseconds); + size_t lock_wait_timeout_milliseconds, + std::string & failure_reason); std::vector getFileSegmentInfos(const UserID & user_id); diff --git a/src/Interpreters/Cache/FileSegment.cpp b/src/Interpreters/Cache/FileSegment.cpp index c46fb978ae4..cfbdfbaa257 100644 --- a/src/Interpreters/Cache/FileSegment.cpp +++ b/src/Interpreters/Cache/FileSegment.cpp @@ -502,7 +502,11 @@ LockedKeyPtr FileSegment::lockKeyMetadata(bool assert_exists) const return metadata->tryLock(); } -bool FileSegment::reserve(size_t size_to_reserve, size_t lock_wait_timeout_milliseconds, FileCacheReserveStat * reserve_stat) +bool FileSegment::reserve( + size_t size_to_reserve, + size_t lock_wait_timeout_milliseconds, + std::string & failure_reason, + FileCacheReserveStat * reserve_stat) { if (!size_to_reserve) throw Exception(ErrorCodes::LOGICAL_ERROR, "Zero space reservation is not allowed"); @@ -554,7 +558,7 @@ bool FileSegment::reserve(size_t size_to_reserve, size_t lock_wait_timeout_milli if (!reserve_stat) reserve_stat = &dummy_stat; - bool reserved = cache->tryReserve(*this, size_to_reserve, *reserve_stat, getKeyMetadata()->user, lock_wait_timeout_milliseconds); + bool reserved = cache->tryReserve(*this, size_to_reserve, *reserve_stat, getKeyMetadata()->user, lock_wait_timeout_milliseconds, failure_reason); if (!reserved) setDownloadFailedUnlocked(lock()); diff --git a/src/Interpreters/Cache/FileSegment.h b/src/Interpreters/Cache/FileSegment.h index 25ffb880b45..e90ebdbf8fe 100644 --- a/src/Interpreters/Cache/FileSegment.h +++ b/src/Interpreters/Cache/FileSegment.h @@ -201,7 +201,11 @@ public: /// Try to reserve exactly `size` bytes (in addition to the getDownloadedSize() bytes already downloaded). /// Returns true if reservation was successful, false otherwise. - bool reserve(size_t size_to_reserve, size_t lock_wait_timeout_milliseconds, FileCacheReserveStat * reserve_stat = nullptr); + bool reserve( + size_t size_to_reserve, + size_t lock_wait_timeout_milliseconds, + std::string & failure_reason, + FileCacheReserveStat * reserve_stat = nullptr); /// Write data into reserved space. void write(char * from, size_t size, size_t offset_in_file); diff --git a/src/Interpreters/Cache/Metadata.cpp b/src/Interpreters/Cache/Metadata.cpp index 7e4b76d3cc6..6399691bcf6 100644 --- a/src/Interpreters/Cache/Metadata.cpp +++ b/src/Interpreters/Cache/Metadata.cpp @@ -705,7 +705,8 @@ void CacheMetadata::downloadImpl(FileSegment & file_segment, std::optionalavailable(); - if (!file_segment.reserve(size, reserve_space_lock_wait_timeout_milliseconds)) + std::string failure_reason; + if (!file_segment.reserve(size, reserve_space_lock_wait_timeout_milliseconds, failure_reason)) { LOG_TEST( log, "Failed to reserve space during background download " diff --git a/src/Interpreters/Cache/QueryCache.cpp b/src/Interpreters/Cache/QueryCache.cpp index a3fe8c2e779..ab926037c67 100644 --- a/src/Interpreters/Cache/QueryCache.cpp +++ b/src/Interpreters/Cache/QueryCache.cpp @@ -128,7 +128,7 @@ namespace bool isQueryCacheRelatedSetting(const String & setting_name) { - return setting_name.starts_with("query_cache_") || setting_name.ends_with("_query_cache"); + return (setting_name.starts_with("query_cache_") || setting_name.ends_with("_query_cache")) && setting_name != "query_cache_tag"; } class RemoveQueryCacheSettingsMatcher @@ -242,11 +242,18 @@ QueryCache::Key::Key( , expires_at(expires_at_) , is_compressed(is_compressed_) , query_string(queryStringFromAST(ast_)) + , tag(settings.query_cache_tag) { } -QueryCache::Key::Key(ASTPtr ast_, const String & current_database, const Settings & settings, std::optional user_id_, const std::vector & current_user_roles_) - : QueryCache::Key(ast_, current_database, settings, {}, user_id_, current_user_roles_, false, std::chrono::system_clock::from_time_t(1), false) /// dummy values for everything != AST, current database, user name/roles +QueryCache::Key::Key( + ASTPtr ast_, + const String & current_database, + const Settings & settings, + std::optional user_id_, + const std::vector & current_user_roles_) + : QueryCache::Key(ast_, current_database, settings, {}, user_id_, current_user_roles_, false, std::chrono::system_clock::from_time_t(1), false) + /// ^^ dummy values for everything != AST, current database, user name/roles { } diff --git a/src/Interpreters/Cache/QueryCache.h b/src/Interpreters/Cache/QueryCache.h index 461197cac32..c7ebaf4d26a 100644 --- a/src/Interpreters/Cache/QueryCache.h +++ b/src/Interpreters/Cache/QueryCache.h @@ -88,6 +88,11 @@ public: /// SYSTEM.QUERY_CACHE. const String query_string; + /// A tag (namespace) for distinguish multiple entries of the same query. + /// This member has currently no use besides that SYSTEM.QUERY_CACHE can populate the 'tag' column conveniently without having to + /// compute the tag from the query AST. + const String tag; + /// Ctor to construct a Key for writing into query cache. Key(ASTPtr ast_, const String & current_database, @@ -99,7 +104,10 @@ public: bool is_compressed); /// Ctor to construct a Key for reading from query cache (this operation only needs the AST + user name). - Key(ASTPtr ast_, const String & current_database, const Settings & settings, std::optional user_id_, const std::vector & current_user_roles_); + Key(ASTPtr ast_, + const String & current_database, + const Settings & settings, + std::optional user_id_, const std::vector & current_user_roles_); bool operator==(const Key & other) const; }; diff --git a/src/Interpreters/Cache/WriteBufferToFileSegment.cpp b/src/Interpreters/Cache/WriteBufferToFileSegment.cpp index e6ebf6ad50c..e43bbacdbc5 100644 --- a/src/Interpreters/Cache/WriteBufferToFileSegment.cpp +++ b/src/Interpreters/Cache/WriteBufferToFileSegment.cpp @@ -75,7 +75,8 @@ void WriteBufferToFileSegment::nextImpl() FileCacheReserveStat reserve_stat; /// In case of an error, we don't need to finalize the file segment /// because it will be deleted soon and completed in the holder's destructor. - bool ok = file_segment->reserve(bytes_to_write, reserve_space_lock_wait_timeout_milliseconds, &reserve_stat); + std::string failure_reason; + bool ok = file_segment->reserve(bytes_to_write, reserve_space_lock_wait_timeout_milliseconds, failure_reason, &reserve_stat); if (!ok) { @@ -84,9 +85,10 @@ void WriteBufferToFileSegment::nextImpl() reserve_stat_msg += fmt::format("{} hold {}, can release {}; ", toString(kind), ReadableSize(stat.non_releasable_size), ReadableSize(stat.releasable_size)); - throw Exception(ErrorCodes::NOT_ENOUGH_SPACE, "Failed to reserve {} bytes for {}: {}(segment info: {})", + throw Exception(ErrorCodes::NOT_ENOUGH_SPACE, "Failed to reserve {} bytes for {}: reason {}, {}(segment info: {})", bytes_to_write, file_segment->getKind() == FileSegmentKind::Temporary ? "temporary file" : "the file in cache", + failure_reason, reserve_stat_msg, file_segment->getInfoForLog() ); diff --git a/src/Interpreters/ClientInfo.cpp b/src/Interpreters/ClientInfo.cpp index 39fdef23baa..daf1e300046 100644 --- a/src/Interpreters/ClientInfo.cpp +++ b/src/Interpreters/ClientInfo.cpp @@ -254,6 +254,8 @@ String toString(ClientInfo::Interface interface) return "LOCAL"; case ClientInfo::Interface::TCP_INTERSERVER: return "TCP_INTERSERVER"; + case ClientInfo::Interface::PROMETHEUS: + return "PROMETHEUS"; } return std::format("Unknown server interface ({}).", static_cast(interface)); diff --git a/src/Interpreters/ClientInfo.h b/src/Interpreters/ClientInfo.h index ca32b4c5cfa..48dea3cc3ea 100644 --- a/src/Interpreters/ClientInfo.h +++ b/src/Interpreters/ClientInfo.h @@ -38,6 +38,7 @@ public: POSTGRESQL = 5, LOCAL = 6, TCP_INTERSERVER = 7, + PROMETHEUS = 8, }; enum class HTTPMethod : uint8_t diff --git a/src/Interpreters/Cluster.cpp b/src/Interpreters/Cluster.cpp index 59c98491c14..7b7bedc850d 100644 --- a/src/Interpreters/Cluster.cpp +++ b/src/Interpreters/Cluster.cpp @@ -113,6 +113,9 @@ Cluster::Address::Address( secure = ConfigHelper::getBool(config, config_prefix + ".secure", false, /* empty_as */true) ? Protocol::Secure::Enable : Protocol::Secure::Disable; priority = Priority{config.getInt(config_prefix + ".priority", 1)}; + proto_send_chunked = config.getString(config_prefix + ".proto_caps.send", "notchunked"); + proto_recv_chunked = config.getString(config_prefix + ".proto_caps.recv", "notchunked"); + const char * port_type = secure == Protocol::Secure::Enable ? "tcp_port_secure" : "tcp_port"; auto default_port = config.getInt(port_type, 0); @@ -227,21 +230,37 @@ String Cluster::Address::toFullString(bool use_compact_format) const } } -Cluster::Address Cluster::Address::fromFullString(const String & full_string) +Cluster::Address Cluster::Address::fromFullString(std::string_view full_string) { - const char * address_begin = full_string.data(); - const char * address_end = address_begin + full_string.size(); - - const char * user_pw_end = strchr(full_string.data(), '@'); + std::string_view user_password; + if (auto pos = full_string.find('@'); pos != std::string_view::npos) + user_password = full_string.substr(pos + 1); /// parsing with the new shard{shard_index}[_replica{replica_index}] format - if (!user_pw_end && startsWith(full_string, "shard")) + if (user_password.empty() && full_string.starts_with("shard")) { - const char * underscore = strchr(full_string.data(), '_'); - Address address; - address.shard_index = parse(address_begin + strlen("shard")); - address.replica_index = underscore ? parse(underscore + strlen("_replica")) : 0; + + if (auto underscore_pos = full_string.find('_'); underscore_pos != std::string_view::npos) + { + address.shard_index = parse(full_string.substr(0, underscore_pos).substr(strlen("shard"))); + + if (full_string.substr(underscore_pos + 1).starts_with("replica")) + { + address.replica_index = parse(full_string.substr(underscore_pos + 1 + strlen("replica"))); + } + else if (full_string.substr(underscore_pos + 1).starts_with("all_replicas")) + { + address.replica_index = 0; + } + else + throw Exception(ErrorCodes::SYNTAX_ERROR, "Incorrect address '{}', should be in a form of `shardN_all_replicas` or `shardN_replicaM`", full_string); + } + else + { + address.shard_index = parse(full_string.substr(strlen("shard"))); + address.replica_index = 0; + } return address; } @@ -252,9 +271,13 @@ Cluster::Address Cluster::Address::fromFullString(const String & full_string) /// - credentials are exposed in file name; /// - the file name can be too long. + const char * address_begin = full_string.data(); + const char * address_end = address_begin + full_string.size(); + const char * user_pw_end = strchr(address_begin, '@'); + Protocol::Secure secure = Protocol::Secure::Disable; const char * secure_tag = "+secure"; - if (endsWith(full_string, secure_tag)) + if (full_string.ends_with(secure_tag)) { address_end -= strlen(secure_tag); secure = Protocol::Secure::Enable; @@ -425,7 +448,9 @@ Cluster::Cluster(const Poco::Util::AbstractConfiguration & config, auto pool = ConnectionPoolFactory::instance().get( static_cast(settings.distributed_connections_pool_size), address.host_name, address.port, - address.default_database, address.user, address.password, address.quota_key, + address.default_database, address.user, address.password, + address.proto_send_chunked, address.proto_recv_chunked, + address.quota_key, address.cluster, address.cluster_secret, "server", address.compression, address.secure, address.priority); @@ -589,6 +614,8 @@ void Cluster::addShard( replica.default_database, replica.user, replica.password, + replica.proto_send_chunked, + replica.proto_recv_chunked, replica.quota_key, replica.cluster, replica.cluster_secret, @@ -744,6 +771,8 @@ Cluster::Cluster(Cluster::ReplicasAsShardsTag, const Cluster & from, const Setti address.default_database, address.user, address.password, + address.proto_send_chunked, + address.proto_recv_chunked, address.quota_key, address.cluster, address.cluster_secret, diff --git a/src/Interpreters/Cluster.h b/src/Interpreters/Cluster.h index dc5790ac339..82d77941b76 100644 --- a/src/Interpreters/Cluster.h +++ b/src/Interpreters/Cluster.h @@ -114,6 +114,8 @@ public: UInt16 port{0}; String user; String password; + String proto_send_chunked = "notchunked"; + String proto_recv_chunked = "notchunked"; String quota_key; /// For inter-server authorization @@ -166,7 +168,7 @@ public: String toFullString(bool use_compact_format) const; /// Returns address with only shard index and replica index or full address without shard index and replica index - static Address fromFullString(const String & address_full_string); + static Address fromFullString(std::string_view full_string); /// Returns resolved address if it does resolve. std::optional getResolvedAddress() const; diff --git a/src/Interpreters/ComparisonTupleEliminationVisitor.cpp b/src/Interpreters/ComparisonTupleEliminationVisitor.cpp index 4f06f345b96..b9f7f37b338 100644 --- a/src/Interpreters/ComparisonTupleEliminationVisitor.cpp +++ b/src/Interpreters/ComparisonTupleEliminationVisitor.cpp @@ -22,7 +22,7 @@ ASTs splitTuple(const ASTPtr & node) if (const auto * literal = node->as(); literal && literal->value.getType() == Field::Types::Tuple) { ASTs result; - const auto & tuple = literal->value.get(); + const auto & tuple = literal->value.safeGet(); for (const auto & child : tuple) result.emplace_back(std::make_shared(child)); return result; diff --git a/src/Interpreters/Context.cpp b/src/Interpreters/Context.cpp index 5413b568068..4a08fd5fe5b 100644 --- a/src/Interpreters/Context.cpp +++ b/src/Interpreters/Context.cpp @@ -99,6 +99,7 @@ #include #include #include +#include #include #include #include @@ -618,7 +619,7 @@ struct ContextSharedPart : boost::noncopyable /** After system_logs have been shut down it is guaranteed that no system table gets created or written to. * Note that part changes at shutdown won't be logged to part log. */ - SHUTDOWN(log, "system logs", system_logs, shutdown()); + SHUTDOWN(log, "system logs", system_logs, flushAndShutdown()); LOG_TRACE(log, "Shutting down database catalog"); DatabaseCatalog::shutdown(); @@ -4255,7 +4256,7 @@ std::shared_ptr Context::getS3QueueLog() const if (!shared->system_logs) return {}; - return shared->system_logs->s3_queue_log; + return shared->system_logs->s3queue_log; } std::shared_ptr Context::getAzureQueueLog() const @@ -4312,13 +4313,13 @@ std::shared_ptr Context::getBlobStorageLog() const return shared->system_logs->blob_storage_log; } -std::vector Context::getSystemLogs() const +SystemLogs Context::getSystemLogs() const { SharedLockGuard lock(shared->mutex); if (!shared->system_logs) return {}; - return shared->system_logs->logs; + return *shared->system_logs; } std::optional Context::getDashboards() const diff --git a/src/Interpreters/Context.h b/src/Interpreters/Context.h index d5e35c3e4b3..3da4f124553 100644 --- a/src/Interpreters/Context.h +++ b/src/Interpreters/Context.h @@ -48,6 +48,8 @@ namespace DB class ASTSelectQuery; +class SystemLogs; + struct ContextSharedPart; class ContextAccess; class ContextAccessWrapper; @@ -1150,7 +1152,7 @@ public: std::shared_ptr getBackupLog() const; std::shared_ptr getBlobStorageLog() const; - std::vector getSystemLogs() const; + SystemLogs getSystemLogs() const; using Dashboards = std::vector>; std::optional getDashboards() const; diff --git a/src/Interpreters/ConvertFunctionOrLikeVisitor.cpp b/src/Interpreters/ConvertFunctionOrLikeVisitor.cpp index 084bb0a1bb9..220355e0741 100644 --- a/src/Interpreters/ConvertFunctionOrLikeVisitor.cpp +++ b/src/Interpreters/ConvertFunctionOrLikeVisitor.cpp @@ -45,7 +45,7 @@ void ConvertFunctionOrLikeData::visit(ASTFunction & function, ASTPtr &) if (!identifier || !literal || literal->value.getType() != Field::Types::String) continue; - String regexp = likePatternToRegexp(literal->value.get()); + String regexp = likePatternToRegexp(literal->value.safeGet()); /// Case insensitive. Works with UTF-8 as well. if (is_ilike) regexp = "(?i)" + regexp; @@ -61,7 +61,7 @@ void ConvertFunctionOrLikeData::visit(ASTFunction & function, ASTPtr &) match->arguments->children.push_back(it->second); unique_elems.push_back(std::move(match)); } - it->second->value.get().push_back(regexp); + it->second->value.safeGet().push_back(regexp); } } diff --git a/src/Interpreters/ConvertStringsToEnumVisitor.cpp b/src/Interpreters/ConvertStringsToEnumVisitor.cpp index 7cc95dc521b..d35baa92900 100644 --- a/src/Interpreters/ConvertStringsToEnumVisitor.cpp +++ b/src/Interpreters/ConvertStringsToEnumVisitor.cpp @@ -33,8 +33,8 @@ String makeStringsEnum(const std::set & values) void changeIfArguments(ASTPtr & first, ASTPtr & second) { - String first_value = first->as()->value.get(); - String second_value = second->as()->value.get(); + String first_value = first->as()->value.safeGet(); + String second_value = second->as()->value.safeGet(); std::set values; values.insert(first_value); @@ -59,9 +59,9 @@ void changeTransformArguments(ASTPtr & array_to, ASTPtr & other) { std::set values; - for (const auto & item : array_to->as()->value.get()) - values.insert(item.get()); - values.insert(other->as()->value.get()); + for (const auto & item : array_to->as()->value.safeGet()) + values.insert(item.safeGet()); + values.insert(other->as()->value.safeGet()); String enum_string = makeStringsEnum(values); @@ -168,7 +168,7 @@ void ConvertStringsToEnumMatcher::visit(ASTFunction & function_node, Data & data if (literal_to->value.getTypeName() != "Array" || literal_other->value.getTypeName() != "String") return; - Array array_to = literal_to->value.get(); + Array array_to = literal_to->value.safeGet(); if (array_to.empty()) return; diff --git a/src/Interpreters/DDLTask.cpp b/src/Interpreters/DDLTask.cpp index fa197d59c13..6e08dd5e2cc 100644 --- a/src/Interpreters/DDLTask.cpp +++ b/src/Interpreters/DDLTask.cpp @@ -538,7 +538,7 @@ void DatabaseReplicatedTask::createSyncedNodeIfNeed(const ZooKeeperPtr & zookeep /// Bool type is really weird, sometimes it's Bool and sometimes it's UInt64... assert(value.getType() == Field::Types::Bool || value.getType() == Field::Types::UInt64); - if (!value.get()) + if (!value.safeGet()) return; zookeeper->createIfNotExists(getSyncedNodePath(), ""); diff --git a/src/Interpreters/ExpressionActions.h b/src/Interpreters/ExpressionActions.h index 6ff39ee07f7..7652fe49eab 100644 --- a/src/Interpreters/ExpressionActions.h +++ b/src/Interpreters/ExpressionActions.h @@ -102,7 +102,7 @@ public: /// /// @param allow_duplicates_in_input - actions are allowed to have /// duplicated input (that will refer into the block). This is needed for - /// preliminary query filtering (filterBlockWithDAG()), because they just + /// preliminary query filtering (filterBlockWithExpression()), because they just /// pass available virtual columns, which cannot be moved in case they are /// used multiple times. void execute(Block & block, size_t & num_rows, bool dry_run = false, bool allow_duplicates_in_input = false) const; diff --git a/src/Interpreters/ExpressionAnalyzer.cpp b/src/Interpreters/ExpressionAnalyzer.cpp index c767aeb2ec2..7063b2162a0 100644 --- a/src/Interpreters/ExpressionAnalyzer.cpp +++ b/src/Interpreters/ExpressionAnalyzer.cpp @@ -38,6 +38,7 @@ #include #include +#include #include #include @@ -590,6 +591,7 @@ void ExpressionAnalyzer::makeAggregateDescriptions(ActionsDAG & actions, Aggrega void ExpressionAnalyzer::makeWindowDescriptionFromAST(const Context & context_, const WindowDescriptions & existing_descriptions, + AggregateFunctionPtr aggregate_function, WindowDescription & desc, const IAST * ast) { const auto & definition = ast->as(); @@ -698,7 +700,21 @@ void ExpressionAnalyzer::makeWindowDescriptionFromAST(const Context & context_, ast->formatForErrorMessage()); } + const auto * window_function = aggregate_function ? dynamic_cast(aggregate_function.get()) : nullptr; desc.frame.is_default = definition.frame_is_default; + if (desc.frame.is_default && window_function) + { + auto default_window_frame_opt = window_function->getDefaultFrame(); + if (default_window_frame_opt) + { + desc.frame = *default_window_frame_opt; + /// Append the default frame description to window_name, make sure it will be put into + /// a proper window description. + desc.window_name += " " + desc.frame.toString(); + return; + } + } + desc.frame.type = definition.frame_type; desc.frame.begin_type = definition.frame_begin_type; desc.frame.begin_preceding = definition.frame_begin_preceding; @@ -734,7 +750,7 @@ void ExpressionAnalyzer::makeWindowDescriptions(ActionsDAG & actions) WindowDescription desc; desc.window_name = elem.name; makeWindowDescriptionFromAST(*current_context, window_descriptions, - desc, elem.definition.get()); + nullptr, desc, elem.definition.get()); auto [it, inserted] = window_descriptions.insert( {elem.name, std::move(desc)}); @@ -821,12 +837,12 @@ void ExpressionAnalyzer::makeWindowDescriptions(ActionsDAG & actions) WindowDescription desc; desc.window_name = default_window_name; makeWindowDescriptionFromAST(*current_context, window_descriptions, - desc, &definition); + window_function.aggregate_function, desc, &definition); auto full_sort_description = desc.full_sort_description; auto [it, inserted] = window_descriptions.insert( - {default_window_name, std::move(desc)}); + {desc.window_name, std::move(desc)}); if (!inserted) { diff --git a/src/Interpreters/ExpressionAnalyzer.h b/src/Interpreters/ExpressionAnalyzer.h index 0c00247df85..dc038e10594 100644 --- a/src/Interpreters/ExpressionAnalyzer.h +++ b/src/Interpreters/ExpressionAnalyzer.h @@ -135,7 +135,12 @@ public: /// A list of windows for window functions. const WindowDescriptions & windowDescriptions() const { return window_descriptions; } - void makeWindowDescriptionFromAST(const Context & context, const WindowDescriptions & existing_descriptions, WindowDescription & desc, const IAST * ast); + void makeWindowDescriptionFromAST( + const Context & context, + const WindowDescriptions & existing_descriptions, + AggregateFunctionPtr aggregate_function, + WindowDescription & desc, + const IAST * ast); void makeWindowDescriptions(ActionsDAG & actions); /** Checks if subquery is not a plain StorageSet. diff --git a/src/Interpreters/HashJoin/AddedColumns.cpp b/src/Interpreters/HashJoin/AddedColumns.cpp index 930a352744d..21cb6e401ed 100644 --- a/src/Interpreters/HashJoin/AddedColumns.cpp +++ b/src/Interpreters/HashJoin/AddedColumns.cpp @@ -15,48 +15,115 @@ JoinOnKeyColumns::JoinOnKeyColumns(const Block & block, const Names & key_names_ { } -template<> void AddedColumns::buildOutput() -{ -} +template<> +void AddedColumns::buildOutput() {} + +template<> +void AddedColumns::buildJoinGetOutput() {} + +template<> +template +void AddedColumns::buildOutputFromBlocks() {} template<> void AddedColumns::buildOutput() { - for (size_t i = 0; i < this->size(); ++i) + if (!output_by_row_list) + buildOutputFromBlocks(); + else { - auto& col = columns[i]; - size_t default_count = 0; - auto apply_default = [&]() + if (join_data_avg_perkey_rows < output_by_row_list_threshold) + buildOutputFromBlocks(); + else { - if (default_count > 0) + for (size_t i = 0; i < this->size(); ++i) { - 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()) + auto & col = columns[i]; + for (auto row_ref_i : lazy_output.row_refs) { - nullable_col->insertFromNotNullable(*column_from_block.column, lazy_output.row_nums[j]); - continue; + if (row_ref_i) + { + const RowRefList * row_ref_list = reinterpret_cast(row_ref_i); + for (auto it = row_ref_list->begin(); it.ok(); ++it) + col->insertFrom(*it->block->getByPosition(right_indexes[i]).column, it->row_num); + } + else + type_name[i].type->insertDefaultInto(*col); } } - col->insertFrom(*column_from_block.column, lazy_output.row_nums[j]); } - apply_default(); + } +} + +template<> +void AddedColumns::buildJoinGetOutput() +{ + for (size_t i = 0; i < this->size(); ++i) + { + auto & col = columns[i]; + for (auto row_ref_i : lazy_output.row_refs) + { + if (!row_ref_i) + { + type_name[i].type->insertDefaultInto(*col); + continue; + } + const auto * row_ref = reinterpret_cast(row_ref_i); + const auto & column_from_block = row_ref->block->getByPosition(right_indexes[i]); + if (auto * nullable_col = typeid_cast(col.get()); nullable_col && !column_from_block.column->isNullable()) + nullable_col->insertFromNotNullable(*column_from_block.column, row_ref->row_num); + else + col->insertFrom(*column_from_block.column, row_ref->row_num); + } + } +} + +template<> +template +void AddedColumns::buildOutputFromBlocks() +{ + if (this->size() == 0) + return; + std::vector blocks; + std::vector row_nums; + blocks.reserve(lazy_output.row_refs.size()); + row_nums.reserve(lazy_output.row_refs.size()); + for (auto row_ref_i : lazy_output.row_refs) + { + if (row_ref_i) + { + if constexpr (from_row_list) + { + const RowRefList * row_ref_list = reinterpret_cast(row_ref_i); + for (auto it = row_ref_list->begin(); it.ok(); ++it) + { + blocks.emplace_back(it->block); + row_nums.emplace_back(it->row_num); + } + } + else + { + const RowRef * row_ref = reinterpret_cast(row_ref_i); + blocks.emplace_back(row_ref->block); + row_nums.emplace_back(row_ref->row_num); + } + } + else + { + blocks.emplace_back(nullptr); + row_nums.emplace_back(0); + } + } + for (size_t i = 0; i < this->size(); ++i) + { + auto & col = columns[i]; + for (size_t j = 0; j < blocks.size(); ++j) + { + if (blocks[j]) + col->insertFrom(*blocks[j]->getByPosition(right_indexes[i]).column, row_nums[j]); + else + type_name[i].type->insertDefaultInto(*col); + } } } @@ -72,29 +139,27 @@ void AddedColumns::applyLazyDefaults() } template<> -void AddedColumns::applyLazyDefaults() -{ -} +void AddedColumns::applyLazyDefaults() {} template <> -void AddedColumns::appendFromBlock(const Block & block, size_t row_num,const bool has_defaults) +void AddedColumns::appendFromBlock(const RowRef * row_ref, const bool has_defaults) { if (has_defaults) applyLazyDefaults(); #ifndef NDEBUG - checkBlock(block); + checkBlock(*row_ref->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]); + const auto & column_from_block = row_ref->block->getByPosition(right_indexes[j]); if (auto * nullable_col = nullable_column_ptrs[j]) - nullable_col->insertFromNotNullable(*column_from_block.column, row_num); + nullable_col->insertFromNotNullable(*column_from_block.column, row_ref->row_num); else - columns[j]->insertFrom(*column_from_block.column, row_num); + columns[j]->insertFrom(*column_from_block.column, row_ref->row_num); } } else @@ -102,22 +167,21 @@ void AddedColumns::appendFromBlock(const Block & block, size_t row_num,co 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); + const auto & column_from_block = row_ref->block->getByPosition(right_indexes[j]); + columns[j]->insertFrom(*column_from_block.column, row_ref->row_num); } } } template <> -void AddedColumns::appendFromBlock(const Block & block, size_t row_num, bool) +void AddedColumns::appendFromBlock(const RowRef * row_ref, bool) { #ifndef NDEBUG - checkBlock(block); + checkBlock(*row_ref->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)); + lazy_output.row_refs.emplace_back(reinterpret_cast(row_ref)); } } template<> @@ -131,8 +195,7 @@ void AddedColumns::appendDefaultRow() { if (has_columns_to_add) { - lazy_output.blocks.emplace_back(0); - lazy_output.row_nums.emplace_back(0); + lazy_output.row_refs.emplace_back(0); } } } diff --git a/src/Interpreters/HashJoin/AddedColumns.h b/src/Interpreters/HashJoin/AddedColumns.h index 13a7df6f498..f1b95a63be6 100644 --- a/src/Interpreters/HashJoin/AddedColumns.h +++ b/src/Interpreters/HashJoin/AddedColumns.h @@ -50,8 +50,7 @@ public: struct LazyOutput { - PaddedPODArray blocks; - PaddedPODArray row_nums; + PaddedPODArray row_refs; }; AddedColumns( @@ -76,8 +75,7 @@ public: 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); + lazy_output.row_refs.reserve(rows_to_add); } columns.reserve(num_columns_to_add); @@ -115,18 +113,22 @@ public: if (columns[j]->isNullable() && !saved_column->isNullable()) nullable_column_ptrs[j] = typeid_cast(columns[j].get()); } + join_data_avg_perkey_rows = join.getJoinedData()->avgPerKeyRows(); + output_by_row_list_threshold = join.getTableJoin().outputByRowListPerkeyRowsThreshold(); } size_t size() const { return columns.size(); } void buildOutput(); + void buildJoinGetOutput(); + 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 appendFromBlock(const RowRef * row_ref, bool has_default); void appendDefaultRow(); @@ -134,6 +136,8 @@ public: const IColumn & leftAsofKey() const { return *left_asof_key; } + static constexpr bool isLazy() { return lazy; } + Block left_block; std::vector join_on_keys; ExpressionActionsPtr additional_filter_expression; @@ -142,6 +146,9 @@ public: size_t rows_to_add; std::unique_ptr offsets_to_replicate; bool need_filter = false; + bool output_by_row_list = false; + size_t join_data_avg_perkey_rows = 0; + size_t output_by_row_list_threshold = 0; IColumn::Filter filter; void reserve(bool need_replicate) @@ -212,15 +219,22 @@ private: columns.back()->reserve(src_column.column->size()); type_name.emplace_back(src_column.type, src_column.name, qualified_name); } + + /** Build output from the blocks that extract from `RowRef` or `RowRefList`, to avoid block cache miss which may cause performance slow down. + * And This problem would happen it we directly build output from `RowRef` or `RowRefList`. + */ + template + void buildOutputFromBlocks(); }; /// 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 +class PreSelectedRows : public std::vector { public: - void appendFromBlock(const Block & block, size_t row_num, bool /* has_default */) { this->emplace_back(&block, row_num); } + void appendFromBlock(const RowRef * row_ref, bool /* has_default */) { this->emplace_back(row_ref); } + static constexpr bool isLazy() { return false; } }; } diff --git a/src/Interpreters/HashJoin/FullHashJoin.cpp b/src/Interpreters/HashJoin/FullHashJoin.cpp index 5d058d10fc2..4cdb2e757a4 100644 --- a/src/Interpreters/HashJoin/FullHashJoin.cpp +++ b/src/Interpreters/HashJoin/FullHashJoin.cpp @@ -1,4 +1,4 @@ -#include +#include namespace DB { diff --git a/src/Interpreters/HashJoin/HashJoin.cpp b/src/Interpreters/HashJoin/HashJoin.cpp index 0c7cad4360d..9c07a71e614 100644 --- a/src/Interpreters/HashJoin/HashJoin.cpp +++ b/src/Interpreters/HashJoin/HashJoin.cpp @@ -291,12 +291,13 @@ 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); }); + auto prefer_use_maps_all = table_join->getMixedJoinExpression() != nullptr; + joinDispatchInit(kind, strictness, map, prefer_use_maps_all); + joinDispatch(kind, strictness, map, prefer_use_maps_all, [&](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); }); + joinDispatch(kind, strictness, map, prefer_use_maps_all, [&](auto, auto, auto & map_) { map_.reserve(data->type, reserve_num); }); } if (!data) @@ -327,9 +328,10 @@ size_t HashJoin::getTotalRowCount() const } else { + auto prefer_use_maps_all = table_join->getMixedJoinExpression() != nullptr; for (const auto & map : data->maps) { - joinDispatch(kind, strictness, map, [&](auto, auto, auto & map_) { res += map_.getTotalRowCount(data->type); }); + joinDispatch(kind, strictness, map, prefer_use_maps_all, [&](auto, auto, auto & map_) { res += map_.getTotalRowCount(data->type); }); } } @@ -367,9 +369,10 @@ size_t HashJoin::getTotalByteCount() const if (data->type != Type::CROSS) { + auto prefer_use_maps_all = table_join->getMixedJoinExpression() != nullptr; for (const auto & map : data->maps) { - joinDispatch(kind, strictness, map, [&](auto, auto, auto & map_) { res += map_.getTotalByteCountImpl(data->type); }); + joinDispatch(kind, strictness, map, prefer_use_maps_all, [&](auto, auto, auto & map_) { res += map_.getTotalByteCountImpl(data->type); }); } } return res; @@ -492,7 +495,7 @@ bool HashJoin::addBlockToJoin(const Block & source_block_, bool check_limits) } size_t rows = source_block.rows(); - + data->rows_to_join += 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) @@ -520,6 +523,8 @@ bool HashJoin::addBlockToJoin(const Block & source_block_, bool check_limits) return true; } + bool prefer_use_maps_all = table_join->getMixedJoinExpression() != nullptr; + size_t total_rows = 0; size_t total_bytes = 0; { @@ -592,7 +597,7 @@ bool HashJoin::addBlockToJoin(const Block & source_block_, bool check_limits) bool is_inserted = false; if (kind != JoinKind::Cross) { - joinDispatch(kind, strictness, data->maps[onexpr_idx], [&](auto kind_, auto strictness_, auto & map) + joinDispatch(kind, strictness, data->maps[onexpr_idx], prefer_use_maps_all, [&](auto kind_, auto strictness_, auto & map) { size_t size = HashJoinMethods>::insertFromBlockImpl( *this, @@ -608,10 +613,10 @@ bool HashJoin::addBlockToJoin(const Block & source_block_, bool check_limits) is_inserted); if (flag_per_row) - used_flags->reinit(stored_block); + used_flags->reinit, MapsAll>>(stored_block); else if (is_inserted) /// Number of buckets + 1 value from zero storage - used_flags->reinit(size + 1); + used_flags->reinit, MapsAll>>(size + 1); }); } @@ -642,16 +647,14 @@ bool HashJoin::addBlockToJoin(const Block & source_block_, bool check_limits) total_bytes = getTotalByteCount(); } } - + data->keys_to_join = total_rows; 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) +void HashJoin::shrinkStoredBlocksToFit(size_t & total_bytes_in_join, bool force_optimize) { - if (shrink_blocks) - return; /// Already shrunk Int64 current_memory_usage = getCurrentQueryMemoryUsage(); Int64 query_memory_usage_delta = current_memory_usage - memory_usage_before_adding_blocks; @@ -659,15 +662,21 @@ void HashJoin::shrinkStoredBlocksToFit(size_t & total_bytes_in_join) 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; + if (!force_optimize) + { + if (shrink_blocks) + return; /// Already shrunk + + /** 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)) : "", @@ -869,7 +878,7 @@ ColumnWithTypeAndName HashJoin::joinGet(const Block & block, const Block & block keys.insert(std::move(key)); } - static_assert(!MapGetter::flagged, + static_assert(!MapGetter::flagged, "joinGet are not protected from hash table changes between block processing"); std::vector maps_vector; @@ -910,16 +919,34 @@ void HashJoin::joinBlock(Block & block, ExtraBlockPtr & not_processed) materializeBlockInplace(block); } + bool prefer_use_maps_all = table_join->getMixedJoinExpression() != nullptr; { 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_) + if (joinDispatch(kind, strictness, maps_vector, prefer_use_maps_all, [&](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_); + Block remaining_block; + if constexpr (std::is_same_v, std::vector>) + { + remaining_block = HashJoinMethods::joinBlockImpl( + *this, block, sample_block_with_columns_to_add, maps_vector_); + } + else if constexpr (std::is_same_v, std::vector>) + { + remaining_block = HashJoinMethods::joinBlockImpl( + *this, block, sample_block_with_columns_to_add, maps_vector_); + } + else if constexpr (std::is_same_v, std::vector>) + { + remaining_block = HashJoinMethods::joinBlockImpl( + *this, block, sample_block_with_columns_to_add, maps_vector_); + } + else + { + throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown maps type"); + } if (remaining_block.rows()) not_processed = std::make_shared(ExtraBlock{std::move(remaining_block)}); else @@ -1019,7 +1046,8 @@ public: rows_added = fillColumnsFromMap(map, columns_right); }; - if (!joinDispatch(parent.kind, parent.strictness, parent.data->maps.front(), fill_callback)) + bool prefer_use_maps_all = parent.table_join->getMixedJoinExpression() != nullptr; + if (!joinDispatch(parent.kind, parent.strictness, parent.data->maps.front(), prefer_use_maps_all, fill_callback)) throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown JOIN strictness '{}' (must be on of: ANY, ALL, ASOF)", parent.strictness); } @@ -1216,11 +1244,12 @@ void HashJoin::reuseJoinedData(const HashJoin & join) if (flag_per_row) throw Exception(ErrorCodes::NOT_IMPLEMENTED, "StorageJoin with ORs is not supported"); + bool prefer_use_maps_all = join.table_join->getMixedJoinExpression() != nullptr; for (auto & map : data->maps) { - joinDispatch(kind, strictness, map, [this](auto kind_, auto strictness_, auto & map_) + joinDispatch(kind, strictness, map, prefer_use_maps_all, [this](auto kind_, auto strictness_, auto & map_) { - used_flags->reinit(map_.getBufferSizeInCells(data->type) + 1); + used_flags->reinit, MapsAll>>(map_.getBufferSizeInCells(data->type) + 1); }); } } @@ -1300,7 +1329,9 @@ void HashJoin::validateAdditionalFilterExpression(ExpressionActionsPtr additiona additional_filter_expression->dumpActions()); } - bool is_supported = (strictness == JoinStrictness::All) && (isInnerOrLeft(kind) || isRightOrFull(kind)); + bool is_supported = ((strictness == JoinStrictness::All) && (isInnerOrLeft(kind) || isRightOrFull(kind))) + || ((strictness == JoinStrictness::Semi || strictness == JoinStrictness::Any || strictness == JoinStrictness::Anti) + && (isLeft(kind) || isRight(kind))) || (strictness == JoinStrictness::Any && (isInner(kind))); if (!is_supported) { throw Exception(ErrorCodes::INVALID_JOIN_ON_EXPRESSION, diff --git a/src/Interpreters/HashJoin/HashJoin.h b/src/Interpreters/HashJoin/HashJoin.h index 0b115b9fdbb..d645b8e9273 100644 --- a/src/Interpreters/HashJoin/HashJoin.h +++ b/src/Interpreters/HashJoin/HashJoin.h @@ -345,6 +345,18 @@ public: size_t blocks_allocated_size = 0; size_t blocks_nullmaps_allocated_size = 0; + + /// Number of rows of right table to join + size_t rows_to_join = 0; + /// Number of keys of right table to join + size_t keys_to_join = 0; + + size_t avgPerKeyRows() const + { + if (keys_to_join == 0) + return 0; + return rows_to_join / keys_to_join; + } }; using RightTableDataPtr = std::shared_ptr; @@ -372,7 +384,7 @@ public: void debugKeys() const; - void shrinkStoredBlocksToFit(size_t & total_bytes_in_join); + void shrinkStoredBlocksToFit(size_t & total_bytes_in_join, bool force_optimize = false); void setMaxJoinedBlockRows(size_t value) { max_joined_block_rows = value; } diff --git a/src/Interpreters/HashJoin/HashJoinMethods.h b/src/Interpreters/HashJoin/HashJoinMethods.h index 0dfafa94efc..97ad57d26ea 100644 --- a/src/Interpreters/HashJoin/HashJoinMethods.h +++ b/src/Interpreters/HashJoin/HashJoinMethods.h @@ -12,15 +12,8 @@ #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 @@ -64,7 +57,6 @@ struct Inserter } }; - /// MapsTemplate is one of MapsOne, MapsAll and MapsAsof template class HashJoinMethods @@ -81,27 +73,7 @@ public: 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 - } - } + bool & is_inserted); using MapsTemplateVector = std::vector; @@ -110,280 +82,37 @@ public: 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; - } + bool is_join_get = false); 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); - } + static KeyGetter createKeyGetter(const ColumnRawPtrs & key_columns, const Sizes & key_sizes); template - static size_t NO_INLINE insertFromBlockImplTypeCase( + static size_t 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(); - } + const Sizes & key_sizes, Block * stored_block, ConstNullMapPtr null_map, UInt8ColumnDataPtr join_mask, Arena & pool, bool & is_inserted); 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); - } - } + JoinStuff::JoinUsedFlags & used_flags); 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); - } - } + JoinStuff::JoinUsedFlags & 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); - } + JoinStuff::JoinUsedFlags & 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). @@ -392,464 +121,30 @@ private: 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; - } + JoinStuff::JoinUsedFlags & used_flags); template - static void setUsed(IColumn::Filter & filter [[maybe_unused]], size_t pos [[maybe_unused]]) - { - if constexpr (need_filter) - filter[pos] = 1; - } + static void setUsed(IColumn::Filter & filter [[maybe_unused]], size_t pos [[maybe_unused]]); template static ColumnPtr buildAdditionalFilter( size_t left_start_row, - const std::vector & selected_rows, + 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; - } + AddedColumns & added_columns); /// First to collect all matched rows refs by join keys, then filter out rows which are not true in additional filter expression. - template + 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; - } + bool flag_per_row [[maybe_unused]]); /// 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; - } + static Block sliceBlock(Block & block, size_t num_rows); /** Since we do not store right key columns, * this function is used to copy left key columns to right key columns. @@ -864,70 +159,22 @@ private: 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; + const IColumn::Filter * null_map_filter = nullptr); - if (null_map_filter) - right_column.column = JoinCommon::filterWithBlanks(right_column.column, *null_map_filter); + static void correctNullabilityInplace(ColumnWithTypeAndName & column, bool nullable); - 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); - } + static void correctNullabilityInplace(ColumnWithTypeAndName & column, bool nullable, const IColumn::Filter & negative_null_map); }; /// 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; @@ -939,6 +186,7 @@ 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/HashJoinMethodsImpl.h b/src/Interpreters/HashJoin/HashJoinMethodsImpl.h new file mode 100644 index 00000000000..320c8851ce4 --- /dev/null +++ b/src/Interpreters/HashJoin/HashJoinMethodsImpl.h @@ -0,0 +1,935 @@ +#pragma once +#include + +namespace DB +{ +namespace ErrorCodes +{ +extern const int UNSUPPORTED_JOIN_KEYS; +extern const int LOGICAL_ERROR; +} +template +size_t HashJoinMethods::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< \ + typename KeyGetterForType>::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 + } +} + +template +Block HashJoinMethods::joinBlockImpl( + const HashJoin & join, Block & block, const Block & block_with_columns_to_add, const MapsTemplateVector & maps_, bool is_join_get) +{ + 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); + + if (is_join_get) + added_columns.buildJoinGetOutput(); + else + 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; +} + +template +template +KeyGetter HashJoinMethods::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 +template +size_t HashJoinMethods::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 +template +size_t HashJoinMethods::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 +template +size_t HashJoinMethods::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 +template +size_t HashJoinMethods::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_maps_all) + { + 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, 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 +template +size_t HashJoinMethods::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); + if constexpr (!flag_per_row && (STRICTNESS == JoinStrictness::All || (STRICTNESS == JoinStrictness::Semi && KIND == JoinKind::Right))) + added_columns.output_by_row_list = true; + + 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(i); + added_columns.filter.resize(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 && 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, 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 && join_features.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, 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, 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 +template +void HashJoinMethods::setUsed(IColumn::Filter & filter [[maybe_unused]], size_t pos [[maybe_unused]]) +{ + if constexpr (need_filter) + filter[pos] = 1; +} + +template +template +ColumnPtr HashJoinMethods::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; +} + +template +template +size_t HashJoinMethods::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 flag_per_row [[maybe_unused]]) +{ + constexpr JoinFeatures join_features; + 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 (join_features.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 (join_features.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); + /// We don't add missing in addFoundRowAll here. we will add it after filter is applied. + /// it's different from `joinRightColumns`. + 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; + /// right/full join or multiple disjuncts, we need to 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]) + { + if constexpr (join_features.is_semi_join || join_features.is_any_join) + { + /// For LEFT/INNER SEMI/ANY JOIN, we need to add only first appeared row from left, + if constexpr (join_features.left || join_features.inner) + { + if (!any_matched) + { + // For inner join, we need mark each right row'flag, because we only use each right row once. + auto used_once = used_flags.template setUsedOnce( + (*selected_right_row_it)->block, (*selected_right_row_it)->row_num, 0); + if (used_once) + { + any_matched = true; + total_added_rows += 1; + added_columns.appendFromBlock(*selected_right_row_it, join_features.add_missing); + } + } + } + else + { + auto used_once = used_flags.template setUsedOnce( + (*selected_right_row_it)->block, (*selected_right_row_it)->row_num, 0); + if (used_once) + { + any_matched = true; + total_added_rows += 1; + added_columns.appendFromBlock(*selected_right_row_it, join_features.add_missing); + } + } + } + else if constexpr (join_features.is_anti_join) + { + any_matched = true; + if constexpr (join_features.right && join_features.need_flags) + used_flags.template setUsed((*selected_right_row_it)->block, (*selected_right_row_it)->row_num, 0); + } + else + { + any_matched = true; + total_added_rows += 1; + added_columns.appendFromBlock(*selected_right_row_it, join_features.add_missing); + 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 constexpr (join_features.is_anti_join) + { + any_matched |= filter_flags[replicated_row]; + } + else if constexpr (join_features.need_replication) + { + if (filter_flags[replicated_row]) + { + any_matched = true; + added_columns.appendFromBlock(*selected_right_row_it, join_features.add_missing); + total_added_rows += 1; + } + ++selected_right_row_it; + } + else + { + if (filter_flags[replicated_row]) + { + any_matched = true; + added_columns.appendFromBlock(*selected_right_row_it, join_features.add_missing); + total_added_rows += 1; + selected_right_row_it = selected_right_row_it + row_replicate_offset[i] - replicated_row; + break; + } + else + ++selected_right_row_it; + } + } + } + + + if constexpr (join_features.is_anti_join) + { + if (!any_matched) + { + if constexpr (join_features.left) + if (need_filter) + setUsed(added_columns.filter, left_start_row + i - 1); + addNotFoundRow(added_columns, total_added_rows); + } + } + else + { + if (!any_matched) + { + addNotFoundRow(added_columns, total_added_rows); + } + else + { + if (!flag_per_row) + used_flags.template setUsed(find_results[find_result_index]); + if (need_filter) + setUsed(added_columns.filter, left_start_row + i - 1); + if constexpr (join_features.add_missing) + added_columns.applyLazyDefaults(); + } + } + find_result_index += (prev_replicated_row != row_replicate_offset[i]); + + if constexpr (join_features.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 (join_features.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 (join_features.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; +} + +template +Block HashJoinMethods::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; +} + +template +ColumnWithTypeAndName HashJoinMethods::copyLeftKeyColumnToRight( + const DataTypePtr & right_key_type, + const String & renamed_right_column, + const ColumnWithTypeAndName & left_column, + const IColumn::Filter * null_map_filter) +{ + 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; +} + +template +void HashJoinMethods::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); + } +} + +template +void HashJoinMethods::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); +} +} diff --git a/src/Interpreters/HashJoin/InnerHashJoin.cpp b/src/Interpreters/HashJoin/InnerHashJoin.cpp index 85aedf3a8e5..69f4c620cb8 100644 --- a/src/Interpreters/HashJoin/InnerHashJoin.cpp +++ b/src/Interpreters/HashJoin/InnerHashJoin.cpp @@ -1,10 +1,11 @@ -#include +#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 index 2f2bd1e29a2..b39593e7cac 100644 --- a/src/Interpreters/HashJoin/JoinFeatures.h +++ b/src/Interpreters/HashJoin/JoinFeatures.h @@ -3,26 +3,41 @@ #include namespace DB { -template +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 is_any_or_semi_join = is_any_join || STRICTNESS == JoinStrictness::RightAny || (is_semi_join && KIND == JoinKind::Left); 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; + /** Whether we may need duplicate rows from the left table. + * For example, when we have row (key1, attr1) in left table + * and rows (key1, attr2), (key1, attr3) in right table, + * then we need to duplicate row (key1, attr1) for each of joined rows from right table, so result will be + * (key1, attr1, key1, attr2) + * (key1, attr1, key1, attr3) + */ static constexpr bool need_replication = is_all_join || (is_any_join && right) || (is_semi_join && right); + + /// Whether we need to filter rows from the left table that do not have matches in the right table. static constexpr bool need_filter = !need_replication && (inner || right || (is_semi_join && left) || (is_anti_join && left)); + + /// Whether we need to add default values for columns from the left table. static constexpr bool add_missing = (left || full) && !is_semi_join; - static constexpr bool need_flags = MapGetter::flagged; + /// Whether we need to store flags for rows from the right table table + /// that indicates if they have matches in the left table. + static constexpr bool need_flags = MapGetter, HashJoin::MapsAll>>::flagged; + + static constexpr bool is_maps_all = std::is_same_v, HashJoin::MapsAll>; }; } diff --git a/src/Interpreters/HashJoin/JoinUsedFlags.h b/src/Interpreters/HashJoin/JoinUsedFlags.h index bd41ba2073f..c84c6ec3fea 100644 --- a/src/Interpreters/HashJoin/JoinUsedFlags.h +++ b/src/Interpreters/HashJoin/JoinUsedFlags.h @@ -26,10 +26,10 @@ 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 + template void reinit(size_t size) { - if constexpr (MapGetter::flagged) + if constexpr (MapGetter::flagged) { assert(flags[nullptr].size() <= size); need_flags = true; @@ -43,10 +43,10 @@ public: } } - template + template void reinit(const Block * block_ptr) { - if constexpr (MapGetter::flagged) + if constexpr (MapGetter::flagged) { assert(flags[block_ptr].size() <= block_ptr->rows()); need_flags = true; @@ -148,6 +148,31 @@ public: } } + template + bool setUsedOnce(const Block * block, size_t row_num, size_t offset) + { + if constexpr (!use_flags) + return true; + + if constexpr (flag_per_row) + { + /// fast check to prevent heavy CAS with seq_cst order + if (flags[block][row_num].load(std::memory_order_relaxed)) + return false; + + bool expected = false; + return flags[block][row_num].compare_exchange_strong(expected, true); + } + else + { + /// fast check to prevent heavy CAS with seq_cst order + if (flags[nullptr][offset].load(std::memory_order_relaxed)) + return false; + + bool expected = false; + return flags[nullptr][offset].compare_exchange_strong(expected, true); + } + } }; } diff --git a/src/Interpreters/HashJoin/KnowRowsHolder.h b/src/Interpreters/HashJoin/KnowRowsHolder.h index d51c96893c5..9223e98d13c 100644 --- a/src/Interpreters/HashJoin/KnowRowsHolder.h +++ b/src/Interpreters/HashJoin/KnowRowsHolder.h @@ -104,7 +104,7 @@ void addFoundRowAll( { if (!known_rows.isKnown(std::make_pair(it->block, it->row_num))) { - added.appendFromBlock(*it->block, it->row_num, false); + added.appendFromBlock(*it, false); ++current_offset; if (!new_known_rows_ptr) { @@ -124,11 +124,16 @@ void addFoundRowAll( known_rows.add(std::cbegin(*new_known_rows_ptr), std::cend(*new_known_rows_ptr)); } } + else if constexpr (AddedColumns::isLazy()) + { + added.appendFromBlock(&mapped, false); + current_offset += mapped.rows; + } else { for (auto it = mapped.begin(); it.ok(); ++it) { - added.appendFromBlock(*it->block, it->row_num, false); + added.appendFromBlock(*it, false); ++current_offset; } } diff --git a/src/Interpreters/HashJoin/LeftHashJoin.cpp b/src/Interpreters/HashJoin/LeftHashJoin.cpp index 69e17ff70bd..4e06789570e 100644 --- a/src/Interpreters/HashJoin/LeftHashJoin.cpp +++ b/src/Interpreters/HashJoin/LeftHashJoin.cpp @@ -1,11 +1,14 @@ -#include +#include namespace DB { template class HashJoinMethods; template class HashJoinMethods; +template class HashJoinMethods; 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 index 8e304754f5c..d9d41d7d63c 100644 --- a/src/Interpreters/HashJoin/RightHashJoin.cpp +++ b/src/Interpreters/HashJoin/RightHashJoin.cpp @@ -1,4 +1,4 @@ -#include +#include namespace DB { diff --git a/src/Interpreters/InterpreterCreateQuery.cpp b/src/Interpreters/InterpreterCreateQuery.cpp index 971f90bd3cd..f30c0585663 100644 --- a/src/Interpreters/InterpreterCreateQuery.cpp +++ b/src/Interpreters/InterpreterCreateQuery.cpp @@ -37,6 +37,7 @@ #include #include #include +#include #include #include @@ -689,7 +690,7 @@ ColumnsDescription InterpreterCreateQuery::getColumnsDescription( throw Exception(ErrorCodes::LOGICAL_ERROR, "Neither default value expression nor type is provided for a column"); if (col_decl.comment) - column.comment = col_decl.comment->as().value.get(); + column.comment = col_decl.comment->as().value.safeGet(); if (col_decl.codec) { @@ -751,6 +752,10 @@ InterpreterCreateQuery::TableProperties InterpreterCreateQuery::getTableProperti if (create.storage && create.storage->engine) getContext()->checkAccess(AccessType::TABLE_ENGINE, create.storage->engine->name); + /// If this is a TimeSeries table then we need to normalize list of columns (add missing columns and reorder), and also set inner table engines. + if (create.is_time_series_table && (mode < LoadingStrictnessLevel::ATTACH)) + StorageTimeSeries::normalizeTableDefinition(create, getContext()); + TableProperties properties; TableLockHolder as_storage_lock; @@ -782,10 +787,8 @@ InterpreterCreateQuery::TableProperties InterpreterCreateQuery::getTableProperti if (index_desc.type == INVERTED_INDEX_NAME && !settings.allow_experimental_inverted_index) throw Exception(ErrorCodes::ILLEGAL_INDEX, "Please use index type 'full_text' instead of 'inverted'"); /// ---- - if (index_desc.type == "annoy" && !settings.allow_experimental_annoy_index) - throw Exception(ErrorCodes::INCORRECT_QUERY, "Annoy index is disabled. Turn on allow_experimental_annoy_index"); - if (index_desc.type == "usearch" && !settings.allow_experimental_usearch_index) - throw Exception(ErrorCodes::INCORRECT_QUERY, "USearch index is disabled. Turn on allow_experimental_usearch_index"); + if (index_desc.type == "vector_similarity" && !settings.allow_experimental_vector_similarity_index) + throw Exception(ErrorCodes::INCORRECT_QUERY, "Vector similarity index is disabled. Turn on allow_experimental_vector_similarity_index"); properties.indices.push_back(index_desc); } @@ -951,12 +954,40 @@ namespace engine_ast->no_empty_args = true; storage.set(storage.engine, engine_ast); } + + void setNullTableEngine(ASTStorage & storage) + { + auto engine_ast = std::make_shared(); + engine_ast->name = "Null"; + engine_ast->no_empty_args = true; + storage.set(storage.engine, engine_ast); + } + } void InterpreterCreateQuery::setEngine(ASTCreateQuery & create) const { if (create.as_table_function) + { + if (getContext()->getSettingsRef().restore_replace_external_table_functions_to_null) + { + const auto & factory = TableFunctionFactory::instance(); + + auto properties = factory.tryGetProperties(create.as_table_function->as()->name); + if (properties && properties->allow_readonly) + return; + if (!create.storage) + { + auto storage_ast = std::make_shared(); + create.set(create.storage, storage_ast); + } + else + throw Exception(ErrorCodes::LOGICAL_ERROR, "Storage should not be created yet, it's a bug."); + create.as_table_function = nullptr; + setNullTableEngine(*create.storage); + } return; + } if (create.is_dictionary || create.is_ordinary_view || create.is_live_view || create.is_window_view) return; @@ -1007,6 +1038,13 @@ void InterpreterCreateQuery::setEngine(ASTCreateQuery & create) const /// 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); } + /// For external tables with restore_replace_external_engine_to_null setting we replace external engines to + /// Null table engine. + else if (getContext()->getSettingsRef().restore_replace_external_engines_to_null) + { + if (StorageFactory::instance().getStorageFeatures(create.storage->engine->name).source_access_type != AccessType::NONE) + setNullTableEngine(*create.storage); + } return; } @@ -1058,6 +1096,7 @@ void InterpreterCreateQuery::setEngine(ASTCreateQuery & create) const else if (as_create.storage) { storage_def = typeid_cast>(as_create.storage->ptr()); + create.is_time_series_table = as_create.is_time_series_table; } else { @@ -1875,7 +1914,7 @@ void InterpreterCreateQuery::prepareOnClusterQuery(ASTCreateQuery & create, Cont if (has_explicit_zk_path_arg) { - String zk_path = create.storage->engine->arguments->children[0]->as()->value.get(); + String zk_path = create.storage->engine->arguments->children[0]->as()->value.safeGet(); Macros::MacroExpansionInfo info; info.table_id.uuid = create.uuid; info.ignore_unknown = true; @@ -1905,6 +1944,8 @@ BlockIO InterpreterCreateQuery::execute() FunctionNameNormalizer::visit(query_ptr.get()); auto & create = query_ptr->as(); + create.if_not_exists |= getContext()->getSettingsRef().create_if_not_exists; + bool is_create_database = create.database && !create.table; if (!create.cluster.empty() && !maybeRemoveOnCluster(query_ptr, getContext())) { diff --git a/src/Interpreters/InterpreterDeleteQuery.cpp b/src/Interpreters/InterpreterDeleteQuery.cpp index 291c8e19db0..4827edc6c2a 100644 --- a/src/Interpreters/InterpreterDeleteQuery.cpp +++ b/src/Interpreters/InterpreterDeleteQuery.cpp @@ -17,6 +17,7 @@ #include #include #include +#include namespace DB @@ -27,7 +28,6 @@ namespace ErrorCodes extern const int TABLE_IS_READ_ONLY; extern const int SUPPORT_IS_DISABLED; extern const int BAD_ARGUMENTS; - extern const int NOT_IMPLEMENTED; extern const int QUERY_IS_PROHIBITED; } @@ -67,13 +67,42 @@ BlockIO InterpreterDeleteQuery::execute() auto table_lock = table->lockForShare(getContext()->getCurrentQueryId(), getContext()->getSettingsRef().lock_acquire_timeout); auto metadata_snapshot = table->getInMemoryMetadataPtr(); - auto lightweightDelete = [&]() + 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()) { if (!getContext()->getSettingsRef().enable_lightweight_delete) throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, "Lightweight delete mutate is disabled. " "Set `enable_lightweight_delete` setting to enable it"); + if (metadata_snapshot->hasProjections()) + { + if (const auto * merge_tree_data = dynamic_cast(table.get())) + if (merge_tree_data->getSettings()->lightweight_mutation_projection_mode == LightweightMutationProjectionMode::THROW) + throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, + "DELETE query is not allowed for table {} because as it has projections and setting " + "lightweight_mutation_projection_mode is set to THROW. " + "User should change lightweight_mutation_projection_mode OR " + "drop all the projections manually before running the query", + table_id.getFullTableName()); + } + /// Build "ALTER ... UPDATE _row_exists = 0 WHERE predicate" query String alter_query = "ALTER TABLE " + table->getStorageID().getFullTableName() @@ -94,79 +123,9 @@ 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 { - if (table->hasProjection()) - { - 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, "DELETE query is not supported for table {}", table->getStorageID().getFullTableName()); diff --git a/src/Interpreters/InterpreterExplainQuery.cpp b/src/Interpreters/InterpreterExplainQuery.cpp index bedd9cb4a80..c820f999e0c 100644 --- a/src/Interpreters/InterpreterExplainQuery.cpp +++ b/src/Interpreters/InterpreterExplainQuery.cpp @@ -332,7 +332,7 @@ ExplainSettings checkAndGetSettings(const ASTPtr & ast_settings) if (settings.hasBooleanSetting(change.name)) { - auto value = change.value.get(); + auto value = change.value.safeGet(); if (value > 1) throw Exception(ErrorCodes::INVALID_SETTING_VALUE, "Invalid value {} for setting \"{}\". " "Expected boolean type", value, change.name); @@ -341,7 +341,7 @@ ExplainSettings checkAndGetSettings(const ASTPtr & ast_settings) } else { - auto value = change.value.get(); + auto value = change.value.safeGet(); settings.setIntegerSetting(change.name, value); } } diff --git a/src/Interpreters/InterpreterKillQueryQuery.cpp b/src/Interpreters/InterpreterKillQueryQuery.cpp index 7eb487ba7b3..2c579f3b468 100644 --- a/src/Interpreters/InterpreterKillQueryQuery.cpp +++ b/src/Interpreters/InterpreterKillQueryQuery.cpp @@ -334,7 +334,7 @@ BlockIO InterpreterKillQueryQuery::execute() for (size_t i = 0; i < moves_block.rows(); ++i) { table_id = StorageID{database_col.getDataAt(i).toString(), table_col.getDataAt(i).toString()}; - auto task_uuid = task_uuid_col[i].get(); + auto task_uuid = task_uuid_col[i].safeGet(); CancellationCode code = CancellationCode::Unknown; diff --git a/src/Interpreters/InterpreterSelectQuery.cpp b/src/Interpreters/InterpreterSelectQuery.cpp index cb42a8abf9c..0c79f4310ce 100644 --- a/src/Interpreters/InterpreterSelectQuery.cpp +++ b/src/Interpreters/InterpreterSelectQuery.cpp @@ -1232,7 +1232,7 @@ SortDescription InterpreterSelectQuery::getSortDescription(const ASTSelectQuery std::shared_ptr collator; if (order_by_elem.getCollation()) - collator = std::make_shared(order_by_elem.getCollation()->as().value.get()); + collator = std::make_shared(order_by_elem.getCollation()->as().value.safeGet()); if (order_by_elem.with_fill) { diff --git a/src/Interpreters/InterpreterSystemQuery.cpp b/src/Interpreters/InterpreterSystemQuery.cpp index c284acfa308..1cd55a0020c 100644 --- a/src/Interpreters/InterpreterSystemQuery.cpp +++ b/src/Interpreters/InterpreterSystemQuery.cpp @@ -663,13 +663,16 @@ BlockIO InterpreterSystemQuery::execute() startStopAction(ActionLocks::ViewRefresh, false); break; case Type::REFRESH_VIEW: - getRefreshTask()->run(); + for (const auto & task : getRefreshTasks()) + task->run(); break; case Type::CANCEL_VIEW: - getRefreshTask()->cancel(); + for (const auto & task : getRefreshTasks()) + task->cancel(); break; case Type::TEST_VIEW: - getRefreshTask()->setFakeTime(query.fake_time_for_view); + for (const auto & task : getRefreshTasks()) + task->setFakeTime(query.fake_time_for_view); break; case Type::DROP_REPLICA: dropReplica(query); @@ -710,14 +713,8 @@ BlockIO InterpreterSystemQuery::execute() case Type::FLUSH_LOGS: { getContext()->checkAccess(AccessType::SYSTEM_FLUSH_LOGS); - - auto logs = getContext()->getSystemLogs(); - std::vector> commands; - commands.reserve(logs.size()); - for (auto * system_log : logs) - commands.emplace_back([system_log] { system_log->flush(true); }); - - executeCommandsAndThrowIfError(commands); + auto system_logs = getContext()->getSystemLogs(); + system_logs.flush(true); break; } case Type::STOP_LISTEN: @@ -1248,15 +1245,15 @@ void InterpreterSystemQuery::flushDistributed(ASTSystemQuery & query) throw Exception(ErrorCodes::NOT_IMPLEMENTED, "SYSTEM RESTART DISK is not supported"); } -RefreshTaskHolder InterpreterSystemQuery::getRefreshTask() +RefreshTaskList InterpreterSystemQuery::getRefreshTasks() { auto ctx = getContext(); ctx->checkAccess(AccessType::SYSTEM_VIEWS); - auto task = ctx->getRefreshSet().getTask(table_id); - if (!task) + auto tasks = ctx->getRefreshSet().findTasks(table_id); + if (tasks.empty()) throw Exception( ErrorCodes::BAD_ARGUMENTS, "Refreshable view {} doesn't exist", table_id.getNameForLogs()); - return task; + return tasks; } diff --git a/src/Interpreters/InterpreterSystemQuery.h b/src/Interpreters/InterpreterSystemQuery.h index 776dd7915f0..f44fe930b04 100644 --- a/src/Interpreters/InterpreterSystemQuery.h +++ b/src/Interpreters/InterpreterSystemQuery.h @@ -74,7 +74,7 @@ private: void flushDistributed(ASTSystemQuery & query); [[noreturn]] void restartDisk(String & name); - RefreshTaskHolder getRefreshTask(); + RefreshTaskList getRefreshTasks(); AccessRightsElements getRequiredAccessForDDLOnCluster() const; void startStopAction(StorageActionBlockType action_type, bool start); diff --git a/src/Interpreters/MutationsInterpreter.cpp b/src/Interpreters/MutationsInterpreter.cpp index 57ad5caa4c7..24635870e62 100644 --- a/src/Interpreters/MutationsInterpreter.cpp +++ b/src/Interpreters/MutationsInterpreter.cpp @@ -146,7 +146,6 @@ ColumnDependencies getAllColumnDependencies( bool isStorageTouchedByMutations( - MergeTreeData & storage, MergeTreeData::DataPartPtr source_part, const StorageMetadataPtr & metadata_snapshot, const std::vector & commands, @@ -155,7 +154,9 @@ bool isStorageTouchedByMutations( if (commands.empty()) return false; + auto storage_from_part = std::make_shared(source_part); bool all_commands_can_be_skipped = true; + for (const auto & command : commands) { if (command.type == MutationCommand::APPLY_DELETED_MASK) @@ -170,7 +171,7 @@ bool isStorageTouchedByMutations( if (command.partition) { - const String partition_id = storage.getPartitionIDFromQuery(command.partition, context); + const String partition_id = storage_from_part->getPartitionIDFromQuery(command.partition, context); if (partition_id == source_part->info.partition_id) all_commands_can_be_skipped = false; } @@ -184,20 +185,18 @@ bool isStorageTouchedByMutations( if (all_commands_can_be_skipped) return false; - auto storage_from_part = std::make_shared(source_part); - std::optional interpreter_select_query; BlockIO io; if (context->getSettingsRef().allow_experimental_analyzer) { - auto select_query_tree = prepareQueryAffectedQueryTree(commands, storage.shared_from_this(), context); + auto select_query_tree = prepareQueryAffectedQueryTree(commands, storage_from_part, context); InterpreterSelectQueryAnalyzer interpreter(select_query_tree, context, SelectQueryOptions().ignoreLimits()); io = interpreter.execute(); } else { - ASTPtr select_query = prepareQueryAffectedAST(commands, storage.shared_from_this(), context); + ASTPtr select_query = prepareQueryAffectedAST(commands, storage_from_part, context); /// Interpreter must be alive, when we use result of execute() method. /// For some reason it may copy context and give it into ExpressionTransform /// after that we will use context from destroyed stack frame in our stream. @@ -220,7 +219,7 @@ bool isStorageTouchedByMutations( Block tmp_block; while (executor.pull(tmp_block)); - auto count = (*block.getByName("count()").column)[0].get(); + auto count = (*block.getByName("count()").column)[0].safeGet(); return count != 0; } diff --git a/src/Interpreters/MutationsInterpreter.h b/src/Interpreters/MutationsInterpreter.h index 6aaa233cda3..57863e9ae73 100644 --- a/src/Interpreters/MutationsInterpreter.h +++ b/src/Interpreters/MutationsInterpreter.h @@ -19,7 +19,6 @@ using QueryPipelineBuilderPtr = std::unique_ptr; /// Return false if the data isn't going to be changed by mutations. bool isStorageTouchedByMutations( - MergeTreeData & storage, MergeTreeData::DataPartPtr source_part, const StorageMetadataPtr & metadata_snapshot, const std::vector & commands, diff --git a/src/Interpreters/OptimizeDateOrDateTimeConverterWithPreimageVisitor.cpp b/src/Interpreters/OptimizeDateOrDateTimeConverterWithPreimageVisitor.cpp index dd205ae6508..913f9900b77 100644 --- a/src/Interpreters/OptimizeDateOrDateTimeConverterWithPreimageVisitor.cpp +++ b/src/Interpreters/OptimizeDateOrDateTimeConverterWithPreimageVisitor.cpp @@ -42,13 +42,13 @@ ASTPtr generateOptimizedDateFilterAST(const String & comparator, const NameAndTy if (isDateOrDate32(column.type.get())) { - start_date_or_date_time = date_lut.dateToString(range.first.get()); - end_date_or_date_time = date_lut.dateToString(range.second.get()); + start_date_or_date_time = date_lut.dateToString(range.first.safeGet()); + end_date_or_date_time = date_lut.dateToString(range.second.safeGet()); } else if (isDateTime(column.type.get()) || isDateTime64(column.type.get())) { - start_date_or_date_time = date_lut.timeToString(range.first.get()); - end_date_or_date_time = date_lut.timeToString(range.second.get()); + start_date_or_date_time = date_lut.timeToString(range.first.safeGet()); + end_date_or_date_time = date_lut.timeToString(range.second.safeGet()); } else [[unlikely]] return {}; diff --git a/src/Interpreters/OptimizeIfWithConstantConditionVisitor.cpp b/src/Interpreters/OptimizeIfWithConstantConditionVisitor.cpp index 48c9988b6fc..e9a663d53b0 100644 --- a/src/Interpreters/OptimizeIfWithConstantConditionVisitor.cpp +++ b/src/Interpreters/OptimizeIfWithConstantConditionVisitor.cpp @@ -24,7 +24,7 @@ static bool tryExtractConstValueFromCondition(const ASTPtr & condition, bool & v if (literal->value.getType() == Field::Types::Int64 || literal->value.getType() == Field::Types::UInt64) { - value = literal->value.get(); + value = literal->value.safeGet(); return true; } if (literal->value.getType() == Field::Types::Null) @@ -51,7 +51,7 @@ static bool tryExtractConstValueFromCondition(const ASTPtr & condition, bool & v { if (type_literal->value.getType() == Field::Types::String) { - const auto & type_str = type_literal->value.get(); + const auto & type_str = type_literal->value.safeGet(); if (type_str == "UInt8" || type_str == "Nullable(UInt8)") return tryExtractConstValueFromCondition(expr_list->children.at(0), value); } diff --git a/src/Interpreters/OptimizeShardingKeyRewriteInVisitor.cpp b/src/Interpreters/OptimizeShardingKeyRewriteInVisitor.cpp index 54515ea072a..86cec8659f5 100644 --- a/src/Interpreters/OptimizeShardingKeyRewriteInVisitor.cpp +++ b/src/Interpreters/OptimizeShardingKeyRewriteInVisitor.cpp @@ -72,7 +72,7 @@ bool shardContains( if (sharding_value.isNull()) return false; - UInt64 value = sharding_value.get(); + UInt64 value = sharding_value.safeGet(); const auto shard_num = data.slots[value % data.slots.size()] + 1; return data.shard_info.shard_num == shard_num; } @@ -120,11 +120,20 @@ void OptimizeShardingKeyRewriteInMatcher::visit(ASTFunction & function, Data & d else if (auto * tuple_literal = right->as(); tuple_literal && tuple_literal->value.getType() == Field::Types::Tuple) { - auto & tuple = tuple_literal->value.get(); - std::erase_if(tuple, [&](auto & child) + auto & tuple = tuple_literal->value.safeGet(); + if (tuple.size() > 1) { - return tuple.size() > 1 && !shardContains(child, name, data); - }); + Tuple new_tuple; + + for (auto & child : tuple) + if (shardContains(child, name, data)) + new_tuple.emplace_back(std::move(child)); + + if (new_tuple.empty()) + new_tuple.emplace_back(std::move(tuple.back())); + + tuple_literal->value = std::move(new_tuple); + } } } @@ -159,7 +168,7 @@ public: { if (isTuple(constant->getResultType())) { - const auto & tuple = constant->getValue().get(); + const auto & tuple = constant->getValue().safeGet(); Tuple new_tuple; new_tuple.reserve(tuple.size()); diff --git a/src/Interpreters/PreparedSets.h b/src/Interpreters/PreparedSets.h index bf99a8ece3c..a6aee974d0e 100644 --- a/src/Interpreters/PreparedSets.h +++ b/src/Interpreters/PreparedSets.h @@ -90,9 +90,18 @@ private: using FutureSetFromTuplePtr = std::shared_ptr; -/// Set from subquery can be built inplace for PK or in CreatingSet step. -/// If use_index_for_in_with_subqueries_max_values is reached, set for PK won't be created, -/// but ordinary set would be created instead. +/// Set from subquery can be filled (by running the subquery) in one of two ways: +/// 1. During query analysis. Specifically, inside `SourceStepWithFilter::applyFilters()`. +/// Useful if the query plan depends on the set contents, e.g. to determine which files to read. +/// 2. During query execution. This is the preferred way. +/// Sets are created by CreatingSetStep, which runs before other steps. +/// Be careful: to build the set during query analysis, the `buildSetInplace()` call must happen +/// inside `SourceStepWithFilter::applyFilters()`. Calling it later, e.g. from `initializePipeline()` +/// will result in LOGICAL_ERROR "Not-ready Set is passed" (because a CreatingSetStep was already +/// added to pipeline but hasn't executed yet). +/// +/// If use_index_for_in_with_subqueries_max_values is reached, the built set won't be suitable for +/// key analysis, but will work with function IN (the set will contain only hashes of elements). class FutureSetFromSubquery final : public FutureSet { public: diff --git a/src/Interpreters/RewriteCountVariantsVisitor.cpp b/src/Interpreters/RewriteCountVariantsVisitor.cpp index 4a541c3765a..272e1ac735f 100644 --- a/src/Interpreters/RewriteCountVariantsVisitor.cpp +++ b/src/Interpreters/RewriteCountVariantsVisitor.cpp @@ -53,7 +53,7 @@ void RewriteCountVariantsVisitor::visit(ASTFunction & func) { if (first_arg_literal->value.getType() == Field::Types::UInt64) { - auto constant = first_arg_literal->value.get(); + auto constant = first_arg_literal->value.safeGet(); if (constant == 1 && !context->getSettingsRef().aggregate_functions_null_for_empty) transform = true; } diff --git a/src/Interpreters/RowRefs.cpp b/src/Interpreters/RowRefs.cpp index 9785ba46dab..1b397ab56ef 100644 --- a/src/Interpreters/RowRefs.cpp +++ b/src/Interpreters/RowRefs.cpp @@ -144,7 +144,7 @@ public: return low; } - RowRef findAsof(const IColumn & asof_column, size_t row_num) override + RowRef * findAsof(const IColumn & asof_column, size_t row_num) override { sort(); @@ -156,10 +156,10 @@ public: if (pos != entries.size()) { size_t row_ref_index = entries[pos].row_ref_index; - return row_refs[row_ref_index]; + return &row_refs[row_ref_index]; } - return {nullptr, 0}; + return nullptr; } private: diff --git a/src/Interpreters/RowRefs.h b/src/Interpreters/RowRefs.h index 650b2311ba7..7c98c47dd11 100644 --- a/src/Interpreters/RowRefs.h +++ b/src/Interpreters/RowRefs.h @@ -122,7 +122,7 @@ struct RowRefList : RowRef }; RowRefList() {} /// NOLINT - RowRefList(const Block * block_, size_t row_num_) : RowRef(block_, row_num_) {} + RowRefList(const Block * block_, size_t row_num_) : RowRef(block_, row_num_), rows(1) {} ForwardIterator begin() const { return ForwardIterator(this); } @@ -135,8 +135,11 @@ struct RowRefList : RowRef *next = Batch(nullptr); } next = next->insert(std::move(row_ref), pool); + ++rows; } +public: + SizeT rows = 0; private: Batch * next = nullptr; }; @@ -158,7 +161,7 @@ struct SortedLookupVectorBase virtual void insert(const IColumn &, const Block *, size_t) = 0; // This needs to be synchronized internally - virtual RowRef findAsof(const IColumn &, size_t) = 0; + virtual RowRef * findAsof(const IColumn &, size_t) = 0; }; diff --git a/src/Interpreters/SessionLog.cpp b/src/Interpreters/SessionLog.cpp index 0615a2a1d62..866f5ba8c0a 100644 --- a/src/Interpreters/SessionLog.cpp +++ b/src/Interpreters/SessionLog.cpp @@ -105,9 +105,10 @@ ColumnsDescription SessionLogElement::getColumnsDescription() {"MySQL", static_cast(Interface::MYSQL)}, {"PostgreSQL", static_cast(Interface::POSTGRESQL)}, {"Local", static_cast(Interface::LOCAL)}, - {"TCP_Interserver", static_cast(Interface::TCP_INTERSERVER)} + {"TCP_Interserver", static_cast(Interface::TCP_INTERSERVER)}, + {"Prometheus", static_cast(Interface::PROMETHEUS)}, }); - static_assert(magic_enum::enum_count() == 7); + static_assert(magic_enum::enum_count() == 8); auto lc_string_datatype = std::make_shared(std::make_shared()); diff --git a/src/Interpreters/Squashing.cpp b/src/Interpreters/Squashing.cpp index 488177c3b4f..95b76c60063 100644 --- a/src/Interpreters/Squashing.cpp +++ b/src/Interpreters/Squashing.cpp @@ -5,7 +5,6 @@ #include #include - namespace DB { @@ -106,6 +105,17 @@ Chunk Squashing::convertToChunk(CurrentData && data) const Chunk Squashing::squash(std::vector && input_chunks, Chunk::ChunkInfoCollection && infos) { + if (input_chunks.size() == 1) + { + /// this is just optimization, no logic changes + Chunk result = std::move(input_chunks.front()); + infos.appendIfUniq(std::move(result.getChunkInfos())); + result.setChunkInfos(infos); + + chassert(result); + return result; + } + std::vector mutable_columns = {}; size_t rows = 0; for (const Chunk & chunk : input_chunks) @@ -114,20 +124,32 @@ Chunk Squashing::squash(std::vector && input_chunks, Chunk::ChunkInfoColl { auto & first_chunk = input_chunks[0]; Columns columns = first_chunk.detachColumns(); + mutable_columns.reserve(columns.size()); for (auto & column : columns) - { mutable_columns.push_back(IColumn::mutate(std::move(column))); - mutable_columns.back()->reserve(rows); - } } + size_t num_columns = mutable_columns.size(); + /// Collect the list of source columns for each column. + std::vector source_columns_list(num_columns, Columns{}); + for (size_t i = 0; i != num_columns; ++i) + source_columns_list[i].reserve(input_chunks.size() - 1); + for (size_t i = 1; i < input_chunks.size(); ++i) // We've already processed the first chunk above { - Columns columns = input_chunks[i].detachColumns(); - for (size_t j = 0, size = mutable_columns.size(); j < size; ++j) + auto columns = input_chunks[i].detachColumns(); + for (size_t j = 0; j != num_columns; ++j) + source_columns_list[j].emplace_back(std::move(columns[j])); + } + + for (size_t i = 0; i != num_columns; ++i) + { + /// We know all the data we will insert in advance and can make all necessary pre-allocations. + mutable_columns[i]->prepareForSquashing(source_columns_list[i]); + for (auto & source_column : source_columns_list[i]) { - const auto source_column = columns[j]; - mutable_columns[j]->insertRangeFrom(*source_column, 0, source_column->size()); + auto column = std::move(source_column); + mutable_columns[i]->insertRangeFrom(*column, 0, column->size()); } } diff --git a/src/Interpreters/SystemLog.cpp b/src/Interpreters/SystemLog.cpp index 572481e6b12..832c39bfaf8 100644 --- a/src/Interpreters/SystemLog.cpp +++ b/src/Interpreters/SystemLog.cpp @@ -1,6 +1,7 @@ #include #include +#include #include #include #include @@ -49,6 +50,7 @@ #include + namespace DB { @@ -282,85 +284,21 @@ ASTPtr getCreateTableQueryClean(const StorageID & table_id, ContextPtr context) SystemLogs::SystemLogs(ContextPtr global_context, const Poco::Util::AbstractConfiguration & config) { - query_log = createSystemLog(global_context, "system", "query_log", config, "query_log", "Contains information about executed queries, for example, start time, duration of processing, error messages."); - query_thread_log = createSystemLog(global_context, "system", "query_thread_log", config, "query_thread_log", "Contains information about threads that execute queries, for example, thread name, thread start time, duration of query processing."); - part_log = createSystemLog(global_context, "system", "part_log", config, "part_log", "This table contains information about events that occurred with data parts in the MergeTree family tables, such as adding or merging data."); - trace_log = createSystemLog(global_context, "system", "trace_log", config, "trace_log", "Contains stack traces collected by the sampling query profiler."); - crash_log = createSystemLog(global_context, "system", "crash_log", config, "crash_log", "Contains information about stack traces for fatal errors. The table does not exist in the database by default, it is created only when fatal errors occur."); - text_log = createSystemLog(global_context, "system", "text_log", config, "text_log", "Contains logging entries which are normally written to a log file or to stdout."); - metric_log = createSystemLog(global_context, "system", "metric_log", config, "metric_log", "Contains history of metrics values from tables system.metrics and system.events, periodically flushed to disk."); - error_log = createSystemLog(global_context, "system", "error_log", config, "error_log", "Contains history of error values from table system.errors, periodically flushed to disk."); - filesystem_cache_log = createSystemLog(global_context, "system", "filesystem_cache_log", config, "filesystem_cache_log", "Contains a history of all events occurred with filesystem cache for objects on a remote filesystem."); - filesystem_read_prefetches_log = createSystemLog( - global_context, "system", "filesystem_read_prefetches_log", config, "filesystem_read_prefetches_log", "Contains a history of all prefetches done during reading from MergeTables backed by a remote filesystem."); - asynchronous_metric_log = createSystemLog( - global_context, "system", "asynchronous_metric_log", config, - "asynchronous_metric_log", "Contains the historical values for system.asynchronous_metrics, once per time interval (one second by default)."); - opentelemetry_span_log = createSystemLog( - global_context, "system", "opentelemetry_span_log", config, - "opentelemetry_span_log", "Contains information about trace spans for executed queries."); - query_views_log = createSystemLog(global_context, "system", "query_views_log", config, "query_views_log", "Contains information about the dependent views executed when running a query, for example, the view type or the execution time."); - zookeeper_log = createSystemLog(global_context, "system", "zookeeper_log", config, "zookeeper_log", "This table contains information about the parameters of the request to the ZooKeeper server and the response from it."); - session_log = createSystemLog(global_context, "system", "session_log", config, "session_log", "Contains information about all successful and failed login and logout events."); - transactions_info_log = createSystemLog( - global_context, "system", "transactions_info_log", config, "transactions_info_log", "Contains information about all transactions executed on a current server."); - 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."); - 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."); +/// NOLINTBEGIN(bugprone-macro-parentheses) +#define CREATE_PUBLIC_MEMBERS(log_type, member, descr) \ + member = createSystemLog(global_context, "system", #member, config, #member, descr); \ + + LIST_OF_ALL_SYSTEM_LOGS(CREATE_PUBLIC_MEMBERS) +#undef CREATE_PUBLIC_MEMBERS +/// NOLINTEND(bugprone-macro-parentheses) - if (query_log) - logs.emplace_back(query_log.get()); - if (query_thread_log) - logs.emplace_back(query_thread_log.get()); - if (part_log) - logs.emplace_back(part_log.get()); - if (trace_log) - logs.emplace_back(trace_log.get()); - if (crash_log) - logs.emplace_back(crash_log.get()); - if (text_log) - logs.emplace_back(text_log.get()); - if (metric_log) - logs.emplace_back(metric_log.get()); - if (error_log) - logs.emplace_back(error_log.get()); - if (asynchronous_metric_log) - logs.emplace_back(asynchronous_metric_log.get()); - if (opentelemetry_span_log) - logs.emplace_back(opentelemetry_span_log.get()); - if (query_views_log) - logs.emplace_back(query_views_log.get()); - if (zookeeper_log) - logs.emplace_back(zookeeper_log.get()); if (session_log) - { - logs.emplace_back(session_log.get()); global_context->addWarningMessage("Table system.session_log is enabled. It's unreliable and may contain garbage. Do not use it for any kind of security monitoring."); - } - if (transactions_info_log) - logs.emplace_back(transactions_info_log.get()); - if (processors_profile_log) - logs.emplace_back(processors_profile_log.get()); - if (filesystem_cache_log) - logs.emplace_back(filesystem_cache_log.get()); - if (filesystem_read_prefetches_log) - logs.emplace_back(filesystem_read_prefetches_log.get()); - if (asynchronous_insert_log) - logs.emplace_back(asynchronous_insert_log.get()); - if (backup_log) - logs.emplace_back(backup_log.get()); - if (s3_queue_log) - logs.emplace_back(s3_queue_log.get()); - 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) + for (auto & log : getAllLogs()) { log->startup(); if (should_prepare) @@ -394,20 +332,54 @@ SystemLogs::SystemLogs(ContextPtr global_context, const Poco::Util::AbstractConf } } - -SystemLogs::~SystemLogs() +std::vector SystemLogs::getAllLogs() const { +#define GET_RAW_POINTERS(log_type, member, descr) \ + (member).get(), \ + + std::vector result = { + LIST_OF_ALL_SYSTEM_LOGS(GET_RAW_POINTERS) + }; +#undef GET_RAW_POINTERS + + auto last_it = std::remove(result.begin(), result.end(), nullptr); + result.erase(last_it, result.end()); + + return result; +} + +void SystemLogs::flush(bool should_prepare_tables_anyway) +{ + auto logs = getAllLogs(); + std::vector logs_indexes(logs.size(), 0); + + for (size_t i = 0; i < logs.size(); ++i) + { + auto last_log_index = logs[i]->getLastLogIndex(); + logs_indexes[i] = last_log_index; + logs[i]->notifyFlush(last_log_index, should_prepare_tables_anyway); + } + + for (size_t i = 0; i < logs.size(); ++i) + logs[i]->flush(logs_indexes[i], should_prepare_tables_anyway); +} + +void SystemLogs::flushAndShutdown() +{ + flush(/* should_prepare_tables_anyway */ false); shutdown(); } void SystemLogs::shutdown() { + auto logs = getAllLogs(); for (auto & log : logs) log->shutdown(); } void SystemLogs::handleCrash() { + auto logs = getAllLogs(); for (auto & log : logs) log->handleCrash(); } @@ -462,33 +434,26 @@ void SystemLog::savingThreadFunction() { setThreadName("SystemLogFlush"); - std::vector to_flush; - bool exit_this_thread = false; - while (!exit_this_thread) + while (true) { try { - // The end index (exclusive, like std end()) of the messages we are - // going to flush. - uint64_t to_flush_end = 0; - // Should we prepare table even if there are no new messages. - bool should_prepare_tables_anyway = false; + auto result = queue->pop(); - to_flush_end = queue->pop(to_flush, should_prepare_tables_anyway, exit_this_thread); - - if (to_flush.empty()) + if (result.is_shutdown) { - if (should_prepare_tables_anyway) - { - prepareTable(); - LOG_TRACE(log, "Table created (force)"); - - queue->confirm(to_flush_end); - } + LOG_TRACE(log, "Terminating"); + return; } - else + + if (!result.logs.empty()) { - flushImpl(to_flush, to_flush_end); + flushImpl(result.logs, result.last_log_index); + } + else if (result.create_table_force) + { + prepareTable(); + queue->confirm(result.last_log_index); } } catch (...) @@ -496,7 +461,6 @@ void SystemLog::savingThreadFunction() tryLogCurrentException(__PRETTY_FUNCTION__); } } - LOG_TRACE(log, "Terminating"); } diff --git a/src/Interpreters/SystemLog.h b/src/Interpreters/SystemLog.h index 0ac468b15ec..9e1af3578bd 100644 --- a/src/Interpreters/SystemLog.h +++ b/src/Interpreters/SystemLog.h @@ -5,6 +5,32 @@ #include #include +#include + +#define LIST_OF_ALL_SYSTEM_LOGS(M) \ + M(QueryLog, query_log, "Contains information about executed queries, for example, start time, duration of processing, error messages.") \ + M(QueryThreadLog, query_thread_log, "Contains information about threads that execute queries, for example, thread name, thread start time, duration of query processing.") \ + M(PartLog, part_log, "This table contains information about events that occurred with data parts in the MergeTree family tables, such as adding or merging data.") \ + M(TraceLog, trace_log, "Contains stack traces collected by the sampling query profiler.") \ + M(CrashLog, crash_log, "Contains information about stack traces for fatal errors. The table does not exist in the database by default, it is created only when fatal errors occur.") \ + M(TextLog, text_log, "Contains logging entries which are normally written to a log file or to stdout.") \ + M(MetricLog, metric_log, "Contains history of metrics values from tables system.metrics and system.events, periodically flushed to disk.") \ + M(ErrorLog, error_log, "Contains history of error values from table system.errors, periodically flushed to disk.") \ + M(FilesystemCacheLog, filesystem_cache_log, "Contains a history of all events occurred with filesystem cache for objects on a remote filesystem.") \ + M(FilesystemReadPrefetchesLog, filesystem_read_prefetches_log, "Contains a history of all prefetches done during reading from MergeTables backed by a remote filesystem.") \ + M(ObjectStorageQueueLog, s3queue_log, "Contains logging entries with the information files processes by S3Queue engine.") \ + M(ObjectStorageQueueLog, azure_queue_log, "Contains logging entries with the information files processes by S3Queue engine.") \ + M(AsynchronousMetricLog, asynchronous_metric_log, "Contains the historical values for system.asynchronous_metrics, once per time interval (one second by default).") \ + M(OpenTelemetrySpanLog, opentelemetry_span_log, "Contains information about trace spans for executed queries.") \ + M(QueryViewsLog, query_views_log, "Contains information about the dependent views executed when running a query, for example, the view type or the execution time.") \ + M(ZooKeeperLog, zookeeper_log, "This table contains information about the parameters of the request to the ZooKeeper server and the response from it.") \ + M(SessionLog, session_log, "Contains information about all successful and failed login and logout events.") \ + M(TransactionsInfoLog, transactions_info_log, "Contains information about all transactions executed on a current server.") \ + M(ProcessorsProfileLog, processors_profile_log, "Contains profiling information on processors level (building blocks for a pipeline for query execution.") \ + M(AsynchronousInsertLog, asynchronous_insert_log, "Contains a history for all asynchronous inserts executed on current server.") \ + M(BackupLog, backup_log, "Contains logging entries with the information about BACKUP and RESTORE operations.") \ + M(BlobStorageLog, blob_storage_log, "Contains logging entries with information about various blob storage operations such as uploads and deletes.") \ + namespace DB { @@ -34,71 +60,37 @@ namespace DB }; */ -class QueryLog; -class QueryThreadLog; -class PartLog; -class TextLog; -class TraceLog; -class CrashLog; -class ErrorLog; -class MetricLog; -class AsynchronousMetricLog; -class OpenTelemetrySpanLog; -class QueryViewsLog; -class ZooKeeperLog; -class SessionLog; -class TransactionsInfoLog; -class ProcessorsProfileLog; -class FilesystemCacheLog; -class FilesystemReadPrefetchesLog; -class AsynchronousInsertLog; -class BackupLog; -class ObjectStorageQueueLog; -class BlobStorageLog; +/// NOLINTBEGIN(bugprone-macro-parentheses) +#define FORWARD_DECLARATION(log_type, member, descr) \ + class log_type; \ + +LIST_OF_ALL_SYSTEM_LOGS(FORWARD_DECLARATION) +#undef FORWARD_DECLARATION +/// NOLINTEND(bugprone-macro-parentheses) + /// System logs should be destroyed in destructor of the last Context and before tables, /// because SystemLog destruction makes insert query while flushing data into underlying tables -struct SystemLogs +class SystemLogs { +public: + SystemLogs() = default; SystemLogs(ContextPtr global_context, const Poco::Util::AbstractConfiguration & config); - ~SystemLogs(); + SystemLogs(const SystemLogs & other) = default; + void flush(bool should_prepare_tables_anyway); + void flushAndShutdown(); void shutdown(); void handleCrash(); - std::shared_ptr query_log; /// Used to log queries. - std::shared_ptr query_thread_log; /// Used to log query threads. - std::shared_ptr part_log; /// Used to log operations with parts - std::shared_ptr trace_log; /// Used to log traces from query profiler - std::shared_ptr crash_log; /// Used to log server crashes. - std::shared_ptr text_log; /// Used to log all text messages. - std::shared_ptr metric_log; /// Used to log all metrics. - 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 azure_queue_log; - /// Metrics from system.asynchronous_metrics. - std::shared_ptr asynchronous_metric_log; - /// OpenTelemetry trace spans. - std::shared_ptr opentelemetry_span_log; - /// Used to log queries of materialized and live views - std::shared_ptr query_views_log; - /// Used to log all actions of ZooKeeper client - std::shared_ptr zookeeper_log; - /// Login, LogOut and Login failure events - std::shared_ptr session_log; - /// Events related to transactions - std::shared_ptr transactions_info_log; - /// Used to log processors profiling - std::shared_ptr processors_profile_log; - std::shared_ptr asynchronous_insert_log; - /// Backup and restore events - std::shared_ptr backup_log; - /// Log blob storage operations - std::shared_ptr blob_storage_log; +#define DECLARE_PUBLIC_MEMBERS(log_type, member, descr) \ + std::shared_ptr member; \ - std::vector logs; + LIST_OF_ALL_SYSTEM_LOGS(DECLARE_PUBLIC_MEMBERS) +#undef DECLARE_PUBLIC_MEMBERS + +private: + std::vector getAllLogs() const; }; struct SystemLogSettings diff --git a/src/Interpreters/TableJoin.cpp b/src/Interpreters/TableJoin.cpp index c8c926db13c..138085f0710 100644 --- a/src/Interpreters/TableJoin.cpp +++ b/src/Interpreters/TableJoin.cpp @@ -115,6 +115,7 @@ TableJoin::TableJoin(const Settings & settings, VolumePtr tmp_volume_, Temporary , partial_merge_join_left_table_buffer_bytes(settings.partial_merge_join_left_table_buffer_bytes) , max_files_to_merge(settings.join_on_disk_max_files_to_merge) , temporary_files_codec(settings.temporary_files_codec) + , output_by_rowlist_perkey_rows_threshold(settings.join_output_by_rowlist_perkey_rows_threshold) , max_memory_usage(settings.max_memory_usage) , tmp_volume(tmp_volume_) , tmp_data(tmp_data_) diff --git a/src/Interpreters/TableJoin.h b/src/Interpreters/TableJoin.h index 3f2bebb5816..4d626084d81 100644 --- a/src/Interpreters/TableJoin.h +++ b/src/Interpreters/TableJoin.h @@ -148,6 +148,7 @@ private: const size_t partial_merge_join_left_table_buffer_bytes = 0; const size_t max_files_to_merge = 0; const String temporary_files_codec = "LZ4"; + const size_t output_by_rowlist_perkey_rows_threshold = 0; /// Value if setting max_memory_usage for query, can be used when max_bytes_in_join is not specified. size_t max_memory_usage = 0; @@ -295,6 +296,7 @@ public: return join_use_nulls && isRightOrFull(kind()); } + size_t outputByRowListPerkeyRowsThreshold() const { return output_by_rowlist_perkey_rows_threshold; } size_t defaultMaxBytes() const { return default_max_bytes; } size_t maxJoinedBlockRows() const { return max_joined_block_rows; } size_t maxRowsInRightBlock() const { return partial_merge_join_rows_in_right_blocks; } diff --git a/src/Interpreters/TreeOptimizer.cpp b/src/Interpreters/TreeOptimizer.cpp index b872eb94fde..6483dd3be48 100644 --- a/src/Interpreters/TreeOptimizer.cpp +++ b/src/Interpreters/TreeOptimizer.cpp @@ -184,7 +184,7 @@ void optimizeGroupBy(ASTSelectQuery * select_query, ContextPtr context) const auto & value = group_exprs[i]->as()->value; if (value.getType() == Field::Types::UInt64) { - auto pos = value.get(); + auto pos = value.safeGet(); if (pos > 0 && pos <= select_query->select()->children.size()) keep_position = true; } diff --git a/src/Interpreters/TreeRewriter.cpp b/src/Interpreters/TreeRewriter.cpp index 6ce6f5e454e..14094c3cccf 100644 --- a/src/Interpreters/TreeRewriter.cpp +++ b/src/Interpreters/TreeRewriter.cpp @@ -1158,7 +1158,8 @@ bool TreeRewriterResult::collectUsedColumns(const ASTPtr & query, bool is_select } } - has_virtual_shard_num = is_remote_storage && storage->isVirtualColumn("_shard_num", storage_snapshot->getMetadataForQuery()) && virtuals->has("_shard_num"); + has_virtual_shard_num + = is_remote_storage && storage->isVirtualColumn("_shard_num", storage_snapshot->metadata) && virtuals->has("_shard_num"); } /// Collect missed object subcolumns diff --git a/src/Interpreters/WindowDescription.cpp b/src/Interpreters/WindowDescription.cpp index 31a881001e3..b1e12ff8048 100644 --- a/src/Interpreters/WindowDescription.cpp +++ b/src/Interpreters/WindowDescription.cpp @@ -94,8 +94,8 @@ void WindowFrame::checkValid() const if (begin_type == BoundaryType::Offset && !((begin_offset.getType() == Field::Types::UInt64 || begin_offset.getType() == Field::Types::Int64) - && begin_offset.get() >= 0 - && begin_offset.get() < INT_MAX)) + && begin_offset.safeGet() >= 0 + && begin_offset.safeGet() < INT_MAX)) { throw Exception(ErrorCodes::BAD_ARGUMENTS, "Frame start offset for '{}' frame must be a nonnegative 32-bit integer, '{}' of type '{}' given", @@ -107,8 +107,8 @@ void WindowFrame::checkValid() const if (end_type == BoundaryType::Offset && !((end_offset.getType() == Field::Types::UInt64 || end_offset.getType() == Field::Types::Int64) - && end_offset.get() >= 0 - && end_offset.get() < INT_MAX)) + && end_offset.safeGet() >= 0 + && end_offset.safeGet() < INT_MAX)) { throw Exception(ErrorCodes::BAD_ARGUMENTS, "Frame end offset for '{}' frame must be a nonnegative 32-bit integer, '{}' of type '{}' given", diff --git a/src/Interpreters/castColumn.cpp b/src/Interpreters/castColumn.cpp index 906dfb84b14..a779c9bc34d 100644 --- a/src/Interpreters/castColumn.cpp +++ b/src/Interpreters/castColumn.cpp @@ -26,11 +26,9 @@ static ColumnPtr castColumn(CastType cast_type, const ColumnWithTypeAndName & ar "" } }; - auto get_cast_func = [cast_type, &arguments] + auto get_cast_func = [from = arg, to = type, cast_type] { - - FunctionOverloadResolverPtr func_builder_cast = createInternalCastOverloadResolver(cast_type, {}); - return func_builder_cast->build(arguments); + return createInternalCast(from, to, cast_type, {}); }; FunctionBasePtr func_cast = cache ? cache->getOrSet(cast_type, from_name, to_name, std::move(get_cast_func)) : get_cast_func(); diff --git a/src/Interpreters/convertFieldToType.cpp b/src/Interpreters/convertFieldToType.cpp index 1a7c166c6a5..738c51baa64 100644 --- a/src/Interpreters/convertFieldToType.cpp +++ b/src/Interpreters/convertFieldToType.cpp @@ -57,7 +57,7 @@ template Field convertNumericTypeImpl(const Field & from) { To result; - if (!accurate::convertNumeric(from.get(), result)) + if (!accurate::convertNumeric(from.safeGet(), result)) return {}; return result; } @@ -88,7 +88,7 @@ Field convertNumericType(const Field & from, const IDataType & type) template Field convertIntToDecimalType(const Field & from, const DataTypeDecimal & type) { - From value = from.get(); + From value = from.safeGet(); if (!type.canStoreWhole(value)) throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "Number is too big to place in {}", type.getName()); @@ -100,7 +100,7 @@ Field convertIntToDecimalType(const Field & from, const DataTypeDecimal & typ template Field convertStringToDecimalType(const Field & from, const DataTypeDecimal & type) { - const String & str_value = from.get(); + const String & str_value = from.safeGet(); T value = type.parseFromString(str_value); return DecimalField(value, type.getScale()); } @@ -108,7 +108,7 @@ Field convertStringToDecimalType(const Field & from, const DataTypeDecimal & template Field convertDecimalToDecimalType(const Field & from, const DataTypeDecimal & type) { - auto field = from.get>(); + auto field = from.safeGet>(); T value = convertDecimals, DataTypeDecimal>(field.getValue(), field.getScale(), type.getScale()); return DecimalField(value, type.getScale()); } @@ -116,7 +116,7 @@ Field convertDecimalToDecimalType(const Field & from, const DataTypeDecimal & template Field convertFloatToDecimalType(const Field & from, const DataTypeDecimal & type) { - From value = from.get(); + From value = from.safeGet(); if (!type.canStoreWhole(value)) throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "Number is too big to place in {}", type.getName()); @@ -182,24 +182,24 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID /// Conversion between Date and DateTime and vice versa. if (which_type.isDate() && which_from_type.isDateTime()) { - return static_cast(static_cast(*from_type_hint).getTimeZone().toDayNum(src.get()).toUnderType()); + return static_cast(static_cast(*from_type_hint).getTimeZone().toDayNum(src.safeGet()).toUnderType()); } else if (which_type.isDate32() && which_from_type.isDateTime()) { - return static_cast(static_cast(*from_type_hint).getTimeZone().toDayNum(src.get()).toUnderType()); + return static_cast(static_cast(*from_type_hint).getTimeZone().toDayNum(src.safeGet()).toUnderType()); } else if (which_type.isDateTime() && which_from_type.isDate()) { - return static_cast(type).getTimeZone().fromDayNum(DayNum(src.get())); + return static_cast(type).getTimeZone().fromDayNum(DayNum(src.safeGet())); } else if (which_type.isDateTime() && which_from_type.isDate32()) { - return static_cast(type).getTimeZone().fromDayNum(DayNum(src.get())); + return static_cast(type).getTimeZone().fromDayNum(DayNum(src.safeGet())); } else if (which_type.isDateTime64() && which_from_type.isDate()) { const auto & date_time64_type = static_cast(type); - const auto value = date_time64_type.getTimeZone().fromDayNum(DayNum(src.get())); + const auto value = date_time64_type.getTimeZone().fromDayNum(DayNum(src.safeGet())); return DecimalField( DecimalUtils::decimalFromComponentsWithMultiplier(value, 0, date_time64_type.getScaleMultiplier()), date_time64_type.getScale()); @@ -207,13 +207,17 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID else if (which_type.isDateTime64() && which_from_type.isDate32()) { const auto & date_time64_type = static_cast(type); - const auto value = date_time64_type.getTimeZone().fromDayNum(ExtendedDayNum(static_cast(src.get()))); + const auto value = date_time64_type.getTimeZone().fromDayNum(ExtendedDayNum(static_cast(src.safeGet()))); return DecimalField( DecimalUtils::decimalFromComponentsWithMultiplier(value, 0, date_time64_type.getScaleMultiplier()), date_time64_type.getScale()); } else if (type.isValueRepresentedByNumber() && src.getType() != Field::Types::String) { + /// Bool is not represented in which_type, so we need to type it separately + if (isInt64OrUInt64orBoolFieldType(src.getType()) && type.getName() == "Bool") + return bool(src.safeGet()); + if (which_type.isUInt8()) return convertNumericType(src, type); if (which_type.isUInt16()) return convertNumericType(src, type); if (which_type.isUInt32()) return convertNumericType(src, type); @@ -253,7 +257,7 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID if (which_type.isDateTime64() && src.getType() == Field::Types::Decimal64) { - const auto & from_type = src.get(); + const auto & from_type = src.safeGet(); const auto & to_type = static_cast(type); const auto scale_from = from_type.getScale(); @@ -300,8 +304,8 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID } if (which_type.isIPv4() && src.getType() == Field::Types::UInt64) { - /// convert to UInt32 which is the underlying type for native IPv4 - return convertNumericType(src, type); + /// convert through UInt32 which is the underlying type for native IPv4 + return static_cast(convertNumericType(src, type).safeGet()); } } else if (which_type.isUUID() && src.getType() == Field::Types::UUID) @@ -318,7 +322,7 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID if (which_from_type.isFixedString() && assert_cast(from_type_hint)->getN() == IPV6_BINARY_LENGTH) { const auto col = type.createColumn(); - ReadBufferFromString in_buffer(src.get()); + ReadBufferFromString in_buffer(src.safeGet()); type.getDefaultSerialization()->deserializeBinary(*col, in_buffer, {}); return (*col)[0]; } @@ -330,7 +334,7 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID if (which_type.isFixedString()) { size_t n = assert_cast(type).getN(); - const auto & src_str = src.get(); + const auto & src_str = src.safeGet(); if (src_str.size() < n) { String src_str_extended = src_str; @@ -347,7 +351,7 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID { if (src.getType() == Field::Types::Array) { - const Array & src_arr = src.get(); + const Array & src_arr = src.safeGet(); size_t src_arr_size = src_arr.size(); const auto & element_type = *(type_array->getNestedType()); @@ -370,7 +374,7 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID { if (src.getType() == Field::Types::Tuple) { - const auto & src_tuple = src.get(); + const auto & src_tuple = src.safeGet(); size_t src_tuple_size = src_tuple.size(); size_t dst_tuple_size = type_tuple->getElements().size(); @@ -415,7 +419,7 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID const auto & key_type = *type_map->getKeyType(); const auto & value_type = *type_map->getValueType(); - const auto & map = src.get(); + const auto & map = src.safeGet(); size_t map_size = map.size(); Map res(map_size); @@ -424,7 +428,7 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID for (size_t i = 0; i < map_size; ++i) { - const auto & map_entry = map[i].get(); + const auto & map_entry = map[i].safeGet(); const auto & key = map_entry[0]; const auto & value = map_entry[1]; @@ -453,7 +457,7 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID "Cannot convert {} to {}", src.getTypeName(), agg_func_type->getName()); - const auto & name = src.get().name; + const auto & name = src.safeGet().name; if (agg_func_type->getName() != name) throw Exception(ErrorCodes::TYPE_MISMATCH, "Cannot convert {} to {}", name, agg_func_type->getName()); @@ -468,7 +472,7 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID if (src.getType() == Field::Types::Tuple && from_type_tuple && from_type_tuple->haveExplicitNames()) { const auto & names = from_type_tuple->getElementNames(); - const auto & tuple = src.get(); + const auto & tuple = src.safeGet(); if (names.size() != tuple.size()) throw Exception(ErrorCodes::TYPE_MISMATCH, @@ -485,10 +489,10 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID if (src.getType() == Field::Types::Map) { Object object; - const auto & map = src.get(); + const auto & map = src.safeGet(); for (const auto & element : map) { - const auto & map_entry = element.get(); + const auto & map_entry = element.safeGet(); const auto & key = map_entry[0]; const auto & value = map_entry[1]; @@ -496,7 +500,7 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID throw Exception(ErrorCodes::TYPE_MISMATCH, "Cannot convert from Map with key of type {} to Object", key.getTypeName()); - object[key.get()] = value; + object[key.safeGet()] = value; } return object; @@ -537,7 +541,7 @@ Field convertFieldToTypeImpl(const Field & src, const IDataType & type, const ID } const auto col = type_to_parse->createColumn(); - ReadBufferFromString in_buffer(src.get()); + ReadBufferFromString in_buffer(src.safeGet()); try { type_to_parse->getDefaultSerialization()->deserializeWholeText(*col, in_buffer, FormatSettings{}); @@ -545,9 +549,9 @@ 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.safeGet(), type.getName()); - e.addMessage(fmt::format("while converting '{}' to {}", src.get(), type.getName())); + e.addMessage(fmt::format("while converting '{}' to {}", src.safeGet(), type.getName())); throw; } @@ -610,7 +614,7 @@ Field convertFieldToTypeOrThrow(const Field & from_value, const IDataType & to_t template static bool decimalEqualsFloat(Field field, Float64 float_value) { - auto decimal_field = field.get>(); + auto decimal_field = field.safeGet>(); auto decimal_to_float = DecimalUtils::convertTo(decimal_field.getValue(), decimal_field.getScale()); return decimal_to_float == float_value; } @@ -629,13 +633,13 @@ std::optional convertFieldToTypeStrict(const Field & from_value, const ID { /// Convert back to Float64 and compare if (result_value.getType() == Field::Types::Decimal32) - return decimalEqualsFloat(result_value, from_value.get()) ? result_value : std::optional{}; + return decimalEqualsFloat(result_value, from_value.safeGet()) ? result_value : std::optional{}; if (result_value.getType() == Field::Types::Decimal64) - return decimalEqualsFloat(result_value, from_value.get()) ? result_value : std::optional{}; + return decimalEqualsFloat(result_value, from_value.safeGet()) ? result_value : std::optional{}; if (result_value.getType() == Field::Types::Decimal128) - return decimalEqualsFloat(result_value, from_value.get()) ? result_value : std::optional{}; + return decimalEqualsFloat(result_value, from_value.safeGet()) ? result_value : std::optional{}; if (result_value.getType() == Field::Types::Decimal256) - return decimalEqualsFloat(result_value, from_value.get()) ? result_value : std::optional{}; + return decimalEqualsFloat(result_value, from_value.safeGet()) ? result_value : std::optional{}; throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown decimal type {}", result_value.getTypeName()); } diff --git a/src/Interpreters/evaluateConstantExpression.cpp b/src/Interpreters/evaluateConstantExpression.cpp index 4bfc80af1fe..d4bb0cc2f8a 100644 --- a/src/Interpreters/evaluateConstantExpression.cpp +++ b/src/Interpreters/evaluateConstantExpression.cpp @@ -297,7 +297,7 @@ namespace { if (tuple_literal->value.getType() == Field::Types::Tuple) { - const auto & tuple = tuple_literal->value.get(); + const auto & tuple = tuple_literal->value.safeGet(); for (const auto & child : tuple) { const auto dnf = analyzeEquals(identifier, child, expr); @@ -792,7 +792,7 @@ std::optional evaluateExpressionOverConstantCondition(const ASTPtr & nod else if (const auto * literal = node->as()) { // Check if it's always true or false. - if (literal->value.getType() == Field::Types::UInt64 && literal->value.get() == 0) + if (literal->value.getType() == Field::Types::UInt64 && literal->value.safeGet() == 0) return {result}; else return {}; diff --git a/src/Interpreters/executeQuery.cpp b/src/Interpreters/executeQuery.cpp index ce58f7f922c..fe87eed5570 100644 --- a/src/Interpreters/executeQuery.cpp +++ b/src/Interpreters/executeQuery.cpp @@ -690,6 +690,12 @@ void validateAnalyzerSettings(ASTPtr ast, bool context_value) if (top_level != value->safeGet()) throw Exception(ErrorCodes::INCORRECT_QUERY, "Setting 'allow_experimental_analyzer' is changed in the subquery. Top level value: {}", top_level); } + + if (auto * value = set_query->changes.tryGet("enable_analyzer")) + { + if (top_level != value->safeGet()) + throw Exception(ErrorCodes::INCORRECT_QUERY, "Setting 'enable_analyzer' is changed in the subquery. Top level value: {}", top_level); + } } for (auto child : node->children) diff --git a/src/Interpreters/inplaceBlockConversions.cpp b/src/Interpreters/inplaceBlockConversions.cpp index 62f8aea86d1..68254768a7d 100644 --- a/src/Interpreters/inplaceBlockConversions.cpp +++ b/src/Interpreters/inplaceBlockConversions.cpp @@ -18,6 +18,7 @@ #include #include #include +#include #include #include @@ -35,8 +36,13 @@ namespace /// Add all required expressions for missing columns calculation void addDefaultRequiredExpressionsRecursively( - const Block & block, const String & required_column_name, DataTypePtr required_column_type, - const ColumnsDescription & columns, ASTPtr default_expr_list_accum, NameSet & added_columns, bool null_as_default) + const Block & block, + const String & required_column_name, + DataTypePtr required_column_type, + const ColumnsDescription & columns, + ASTPtr default_expr_list_accum, + NameSet & added_columns, + bool null_as_default) { checkStackSize(); @@ -271,6 +277,53 @@ static std::unordered_map collectOffsetsColumns( return offsets_columns; } +static ColumnPtr createColumnWithDefaultValue(const IDataType & data_type, const String & subcolumn_name, size_t num_rows) +{ + auto column = data_type.createColumnConstWithDefaultValue(num_rows); + + /// We must turn a constant column into a full column because the interpreter could infer + /// that it is constant everywhere but in some blocks (from other parts) it can be a full column. + + if (subcolumn_name.empty()) + return column->convertToFullColumnIfConst(); + + /// Firstly get subcolumn from const column and then replicate. + column = assert_cast(*column).getDataColumnPtr(); + column = data_type.getSubcolumn(subcolumn_name, column); + + return ColumnConst::create(std::move(column), num_rows)->convertToFullColumnIfConst(); +} + +static bool hasDefault(const StorageMetadataPtr & metadata_snapshot, const NameAndTypePair & column) +{ + if (!metadata_snapshot) + return false; + + const auto & columns = metadata_snapshot->getColumns(); + if (columns.has(column.name)) + return columns.hasDefault(column.name); + + auto name_in_storage = column.getNameInStorage(); + return columns.hasDefault(name_in_storage); +} + +static String removeTupleElementsFromSubcolumn(String subcolumn_name, const Names & tuple_elements) +{ + /// Add a dot to the end of name for convenience. + subcolumn_name += "."; + for (const auto & elem : tuple_elements) + { + auto pos = subcolumn_name.find(elem + "."); + if (pos != std::string::npos) + subcolumn_name.erase(pos, elem.size() + 1); + } + + if (subcolumn_name.ends_with(".")) + subcolumn_name.pop_back(); + + return subcolumn_name; +} + void fillMissingColumns( Columns & res_columns, size_t num_rows, @@ -296,21 +349,17 @@ void fillMissingColumns( auto requested_column = requested_columns.begin(); for (size_t i = 0; i < num_columns; ++i, ++requested_column) { - const auto & [name, type] = *requested_column; - - if (res_columns[i] && partially_read_columns.contains(name)) + if (res_columns[i] && partially_read_columns.contains(requested_column->name)) res_columns[i] = nullptr; - if (res_columns[i]) - continue; - - if (metadata_snapshot && metadata_snapshot->getColumns().hasDefault(name)) + /// Nothing to fill or default should be filled in evaluateMissingDefaults + if (res_columns[i] || hasDefault(metadata_snapshot, *requested_column)) continue; std::vector current_offsets; size_t num_dimensions = 0; - const auto * array_type = typeid_cast(type.get()); + const auto * array_type = typeid_cast(requested_column->type.get()); if (array_type && !offsets_columns.empty()) { num_dimensions = getNumberOfDimensions(*array_type); @@ -345,20 +394,34 @@ void fillMissingColumns( if (!current_offsets.empty()) { - size_t num_empty_dimensions = num_dimensions - current_offsets.size(); - auto scalar_type = createArrayOfType(getBaseTypeOfArray(type), num_empty_dimensions); + Names tuple_elements; + auto serialization = IDataType::getSerialization(*requested_column); + /// For Nested columns collect names of tuple elements and skip them while getting the base type of array. + IDataType::forEachSubcolumn([&](const auto & path, const auto &, const auto &) + { + if (path.back().type == ISerialization::Substream::TupleElement) + tuple_elements.push_back(path.back().name_of_substream); + }, ISerialization::SubstreamData(serialization)); + + /// The number of dimensions that belongs to the array itself but not shared in Nested column. + /// For example for column "n Nested(a UInt64, b Array(UInt64))" this value is 0 for `n.a` and 1 for `n.b`. + size_t num_empty_dimensions = num_dimensions - current_offsets.size(); + + auto base_type = getBaseTypeOfArray(requested_column->getTypeInStorage(), tuple_elements); + auto scalar_type = createArrayOfType(base_type, num_empty_dimensions); size_t data_size = assert_cast(*current_offsets.back()).getData().back(); - res_columns[i] = scalar_type->createColumnConstWithDefaultValue(data_size)->convertToFullColumnIfConst(); + + /// Remove names of tuple elements because they are already processed by 'getBaseTypeOfArray'. + auto subcolumn_name = removeTupleElementsFromSubcolumn(requested_column->getSubcolumnName(), tuple_elements); + res_columns[i] = createColumnWithDefaultValue(*scalar_type, subcolumn_name, data_size); for (auto it = current_offsets.rbegin(); it != current_offsets.rend(); ++it) res_columns[i] = ColumnArray::create(res_columns[i], *it); } else { - /// We must turn a constant column into a full column because the interpreter could infer - /// that it is constant everywhere but in some blocks (from other parts) it can be a full column. - res_columns[i] = type->createColumnConstWithDefaultValue(num_rows)->convertToFullColumnIfConst(); + res_columns[i] = createColumnWithDefaultValue(*requested_column->getTypeInStorage(), requested_column->getSubcolumnName(), num_rows); } } } diff --git a/src/Interpreters/joinDispatch.h b/src/Interpreters/joinDispatch.h index 54c5c7dc83a..4aabc49c29b 100644 --- a/src/Interpreters/joinDispatch.h +++ b/src/Interpreters/joinDispatch.h @@ -12,38 +12,53 @@ namespace DB { -template +/// HashJoin::MapsOne is more efficient, it only store one row for each key in the map. It is recommended to use it whenever possible. +/// When only need to match only one row from right table, use HashJoin::MapsOne. For example, LEFT ANY/SEMI/ANTI. +/// +/// HashJoin::MapsAll will store all rows for each key in the map. It is used when need to match multiple rows from right table. +/// For example, LEFT ALL, INNER ALL, RIGHT ALL/ANY. +/// +/// prefer_use_maps_all is true when there is mixed inequal condition in the join condition. For example, `t1.a = t2.a AND t1.b > t2.b`. +/// In this case, we need to use HashJoin::MapsAll to store all rows for each key in the map. We will select all matched rows from the map +/// and filter them by `t1.b > t2.b`. +/// +/// flagged indicates whether we need to store flags for each row whether it has been used in the join. See JoinUsedFlags.h. +template struct MapGetter; -template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; -template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; -template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = true; }; -template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = true; }; +template struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; +template struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; +template struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = true; }; +template struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = true; }; -template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; -template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = true; }; -template <> struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = true; }; -template <> struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = true; }; +template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; +template <> struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = false; }; +template <> struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = true; }; +template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = true; }; +template struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = true; }; +template struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = true; }; -template <> struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = false; }; -template <> struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = false; }; -template <> struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = true; }; -template <> struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = true; }; +template struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = false; }; +template struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = false; }; +template struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = true; }; +template struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = true; }; /// Only SEMI LEFT and SEMI RIGHT are valid. INNER and FULL are here for templates instantiation. -template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; -template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; -template <> struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = true; }; -template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; +template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; +template <> struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = false; }; +template struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; +template struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = true; }; +template struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; -/// Only SEMI LEFT and SEMI RIGHT are valid. INNER and FULL are here for templates instantiation. -template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; -template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; -template <> struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = true; }; -template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; +/// Only ANTI LEFT and ANTI RIGHT are valid. INNER and FULL are here for templates instantiation. +template <> struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; +template <> struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = true; }; +template struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; +template struct MapGetter { using Map = HashJoin::MapsAll; static constexpr bool flagged = true; }; +template struct MapGetter { using Map = HashJoin::MapsOne; static constexpr bool flagged = false; }; -template -struct MapGetter { using Map = HashJoin::MapsAsof; static constexpr bool flagged = false; }; +template +struct MapGetter { using Map = HashJoin::MapsAsof; static constexpr bool flagged = false; }; static constexpr std::array STRICTNESSES = { JoinStrictness::RightAny, @@ -62,7 +77,7 @@ static constexpr std::array KINDS = { }; /// Init specified join map -inline bool joinDispatchInit(JoinKind kind, JoinStrictness strictness, HashJoin::MapsVariant & maps) +inline bool joinDispatchInit(JoinKind kind, JoinStrictness strictness, HashJoin::MapsVariant & maps, bool prefer_use_maps_all = false) { return static_for<0, KINDS.size() * STRICTNESSES.size()>([&](auto ij) { @@ -70,7 +85,10 @@ inline bool joinDispatchInit(JoinKind kind, JoinStrictness strictness, HashJoin: constexpr auto j = ij % STRICTNESSES.size(); if (kind == KINDS[i] && strictness == STRICTNESSES[j]) { - maps = typename MapGetter::Map(); + if (prefer_use_maps_all) + maps = typename MapGetter::Map(); + else + maps = typename MapGetter::Map(); return true; } return false; @@ -79,7 +97,7 @@ inline bool joinDispatchInit(JoinKind kind, JoinStrictness strictness, HashJoin: /// Call function on specified join map template -inline bool joinDispatch(JoinKind kind, JoinStrictness strictness, MapsVariant & maps, Func && func) +inline bool joinDispatch(JoinKind kind, JoinStrictness strictness, MapsVariant & maps, bool prefer_use_maps_all, Func && func) { return static_for<0, KINDS.size() * STRICTNESSES.size()>([&](auto ij) { @@ -89,10 +107,16 @@ inline bool joinDispatch(JoinKind kind, JoinStrictness strictness, MapsVariant & constexpr auto j = ij % STRICTNESSES.size(); if (kind == KINDS[i] && strictness == STRICTNESSES[j]) { - func( - std::integral_constant(), - std::integral_constant(), - std::get::Map>(maps)); + if (prefer_use_maps_all) + func( + std::integral_constant(), + std::integral_constant(), + std::get::Map>(maps)); + else + func( + std::integral_constant(), + std::integral_constant(), + std::get::Map>(maps)); return true; } return false; @@ -101,7 +125,7 @@ inline bool joinDispatch(JoinKind kind, JoinStrictness strictness, MapsVariant & /// Call function on specified join map template -inline bool joinDispatch(JoinKind kind, JoinStrictness strictness, std::vector & mapsv, Func && func) +inline bool joinDispatch(JoinKind kind, JoinStrictness strictness, std::vector & mapsv, bool prefer_use_maps_all, Func && func) { return static_for<0, KINDS.size() * STRICTNESSES.size()>([&](auto ij) { @@ -111,17 +135,31 @@ inline bool joinDispatch(JoinKind kind, JoinStrictness strictness, std::vector::Map; - std::vector v; - v.reserve(mapsv.size()); - for (const auto & el : mapsv) - v.push_back(&std::get(*el)); + if (prefer_use_maps_all) + { + using MapType = typename MapGetter::Map; + std::vector v; + v.reserve(mapsv.size()); + for (const auto & el : mapsv) + v.push_back(&std::get(*el)); - func( - std::integral_constant(), - std::integral_constant(), - v - /*std::get::Map>(maps)*/); + func( + std::integral_constant(), std::integral_constant(), v + /*std::get::Map>(maps)*/); + } + else + { + using MapType = typename MapGetter::Map; + std::vector v; + v.reserve(mapsv.size()); + for (const auto & el : mapsv) + v.push_back(&std::get(*el)); + + func( + std::integral_constant(), std::integral_constant(), v + /*std::get::Map>(maps)*/); + + } return true; } return false; diff --git a/src/Interpreters/replaceForPositionalArguments.cpp b/src/Interpreters/replaceForPositionalArguments.cpp index 3d60723a167..ee967f45c74 100644 --- a/src/Interpreters/replaceForPositionalArguments.cpp +++ b/src/Interpreters/replaceForPositionalArguments.cpp @@ -35,11 +35,11 @@ bool replaceForPositionalArguments(ASTPtr & argument, const ASTSelectQuery * sel if (which == Field::Types::UInt64) { - pos = ast_literal->value.get(); + pos = ast_literal->value.safeGet(); } else if (which == Field::Types::Int64) { - auto value = ast_literal->value.get(); + auto value = ast_literal->value.safeGet(); if (value > 0) pos = value; else diff --git a/src/Interpreters/tests/gtest_comparison_graph.cpp b/src/Interpreters/tests/gtest_comparison_graph.cpp index ac24a8de368..5f93bb983c1 100644 --- a/src/Interpreters/tests/gtest_comparison_graph.cpp +++ b/src/Interpreters/tests/gtest_comparison_graph.cpp @@ -29,7 +29,7 @@ TEST(ComparisonGraph, Bounds) const auto & [lower, strict] = *res; - ASSERT_EQ(lower.get(), 3); + ASSERT_EQ(lower.safeGet(), 3); ASSERT_TRUE(strict); } @@ -39,7 +39,7 @@ TEST(ComparisonGraph, Bounds) const auto & [upper, strict] = *res; - ASSERT_EQ(upper.get(), 7); + ASSERT_EQ(upper.safeGet(), 7); ASSERT_TRUE(strict); } diff --git a/src/Interpreters/tests/gtest_filecache.cpp b/src/Interpreters/tests/gtest_filecache.cpp index 36acc319f4e..fd602ab5918 100644 --- a/src/Interpreters/tests/gtest_filecache.cpp +++ b/src/Interpreters/tests/gtest_filecache.cpp @@ -246,7 +246,8 @@ void download(FileSegment & file_segment) ASSERT_EQ(file_segment.state(), State::DOWNLOADING); ASSERT_EQ(file_segment.getDownloadedSize(), 0); - ASSERT_TRUE(file_segment.reserve(file_segment.range().size(), 1000)); + std::string failure_reason; + ASSERT_TRUE(file_segment.reserve(file_segment.range().size(), 1000, failure_reason)); download(cache_base_path, file_segment); ASSERT_EQ(file_segment.state(), State::DOWNLOADING); @@ -258,7 +259,8 @@ void assertDownloadFails(FileSegment & file_segment) { ASSERT_EQ(file_segment.getOrSetDownloader(), FileSegment::getCallerId()); ASSERT_EQ(file_segment.getDownloadedSize(), 0); - ASSERT_FALSE(file_segment.reserve(file_segment.range().size(), 1000)); + std::string failure_reason; + ASSERT_FALSE(file_segment.reserve(file_segment.range().size(), 1000, failure_reason)); file_segment.complete(); } @@ -957,10 +959,11 @@ TEST_F(FileCacheTest, temporaryData) { ASSERT_EQ(some_data_holder->size(), 5); + std::string failure_reason; for (auto & segment : *some_data_holder) { ASSERT_TRUE(segment->getOrSetDownloader() == DB::FileSegment::getCallerId()); - ASSERT_TRUE(segment->reserve(segment->range().size(), 1000)); + ASSERT_TRUE(segment->reserve(segment->range().size(), 1000, failure_reason)); download(*segment); segment->complete(); } diff --git a/src/Parsers/ASTCreateQuery.cpp b/src/Parsers/ASTCreateQuery.cpp index f0f782c0a63..359e93ab269 100644 --- a/src/Parsers/ASTCreateQuery.cpp +++ b/src/Parsers/ASTCreateQuery.cpp @@ -483,6 +483,13 @@ void ASTCreateQuery::formatQueryImpl(const FormatSettings & settings, FormatStat if (auto to_storage = getTargetInnerEngine(ViewTarget::To)) to_storage->formatImpl(settings, state, frame); + if (targets) + { + targets->formatTarget(ViewTarget::Data, settings, state, frame); + targets->formatTarget(ViewTarget::Tags, settings, state, frame); + targets->formatTarget(ViewTarget::Metrics, settings, state, frame); + } + if (dictionary) dictionary->formatImpl(settings, state, frame); diff --git a/src/Parsers/ASTCreateQuery.h b/src/Parsers/ASTCreateQuery.h index a95010aea31..6be0fa78903 100644 --- a/src/Parsers/ASTCreateQuery.h +++ b/src/Parsers/ASTCreateQuery.h @@ -97,6 +97,7 @@ public: bool is_materialized_view{false}; bool is_live_view{false}; bool is_window_view{false}; + bool is_time_series_table{false}; /// CREATE TABLE ... ENGINE=TimeSeries() ... bool is_populate{false}; bool is_create_empty{false}; /// CREATE TABLE ... EMPTY AS SELECT ... bool replace_view{false}; /// CREATE OR REPLACE VIEW diff --git a/src/Parsers/ASTFunction.cpp b/src/Parsers/ASTFunction.cpp index cd9e910d45a..d42728addb7 100644 --- a/src/Parsers/ASTFunction.cpp +++ b/src/Parsers/ASTFunction.cpp @@ -522,7 +522,7 @@ void ASTFunction::formatImplWithoutAlias(const FormatSettings & settings, Format if (tuple_arguments_valid && lit_right) { if (isInt64OrUInt64FieldType(lit_right->value.getType()) - && lit_right->value.get() >= 0) + && lit_right->value.safeGet() >= 0) { if (frame.need_parens) settings.ostr << '('; diff --git a/src/Parsers/ASTIndexDeclaration.h b/src/Parsers/ASTIndexDeclaration.h index dd05ad08184..72f3f017a99 100644 --- a/src/Parsers/ASTIndexDeclaration.h +++ b/src/Parsers/ASTIndexDeclaration.h @@ -13,8 +13,7 @@ class ASTIndexDeclaration : public IAST { public: static const auto DEFAULT_INDEX_GRANULARITY = 1uz; - static const auto DEFAULT_ANNOY_INDEX_GRANULARITY = 100'000'000uz; - static const auto DEFAULT_USEARCH_INDEX_GRANULARITY = 100'000'000uz; + static const auto DEFAULT_VECTOR_SIMILARITY_INDEX_GRANULARITY = 100'000'000uz; ASTIndexDeclaration(ASTPtr expression, ASTPtr type, const String & name_); diff --git a/src/Parsers/ASTLiteral.cpp b/src/Parsers/ASTLiteral.cpp index 8dedc5dc95d..515f4f0cb9f 100644 --- a/src/Parsers/ASTLiteral.cpp +++ b/src/Parsers/ASTLiteral.cpp @@ -73,8 +73,8 @@ void ASTLiteral::appendColumnNameImpl(WriteBuffer & ostr) const /// Special case for very large arrays and tuples. Instead of listing all elements, will use hash of them. /// (Otherwise column name will be too long, that will lead to significant slowdown of expression analysis.) auto type = value.getType(); - if ((type == Field::Types::Array && value.get().size() > min_elements_for_hashing) - || (type == Field::Types::Tuple && value.get().size() > min_elements_for_hashing)) + if ((type == Field::Types::Array && value.safeGet().size() > min_elements_for_hashing) + || (type == Field::Types::Tuple && value.safeGet().size() > min_elements_for_hashing)) { SipHash hash; applyVisitor(FieldVisitorHash(hash), value); @@ -92,7 +92,7 @@ void ASTLiteral::appendColumnNameImpl(WriteBuffer & ostr) const /// for tons of literals as it creates temporary String. if (value.getType() == Field::Types::String) { - writeQuoted(value.get(), ostr); + writeQuoted(value.safeGet(), ostr); } else { @@ -110,7 +110,7 @@ void ASTLiteral::appendColumnNameImplLegacy(WriteBuffer & ostr) const /// Special case for very large arrays. Instead of listing all elements, will use hash of them. /// (Otherwise column name will be too long, that will lead to significant slowdown of expression analysis.) auto type = value.getType(); - if ((type == Field::Types::Array && value.get().size() > min_elements_for_hashing)) + if ((type == Field::Types::Array && value.safeGet().size() > min_elements_for_hashing)) { SipHash hash; applyVisitor(FieldVisitorHash(hash), value); diff --git a/src/Parsers/ASTSystemQuery.cpp b/src/Parsers/ASTSystemQuery.cpp index a730ea0ba3d..7780544d5c2 100644 --- a/src/Parsers/ASTSystemQuery.cpp +++ b/src/Parsers/ASTSystemQuery.cpp @@ -198,6 +198,29 @@ void ASTSystemQuery::formatImpl(const FormatSettings & settings, FormatState & s print_database_table(); } + if (sync_replica_mode != SyncReplicaMode::DEFAULT) + { + settings.ostr << ' '; + print_keyword(magic_enum::enum_name(sync_replica_mode)); + + // If the mode is LIGHTWEIGHT and specific source replicas are specified + if (sync_replica_mode == SyncReplicaMode::LIGHTWEIGHT && !src_replicas.empty()) + { + settings.ostr << ' '; + print_keyword("FROM"); + settings.ostr << ' '; + + bool first = true; + for (const auto & src : src_replicas) + { + if (!first) + settings.ostr << ", "; + first = false; + settings.ostr << quoteString(src); + } + } + } + if (query_settings) { settings.ostr << (settings.hilite ? hilite_keyword : "") << settings.nl_or_ws << "SETTINGS " << (settings.hilite ? hilite_none : ""); @@ -233,28 +256,6 @@ void ASTSystemQuery::formatImpl(const FormatSettings & settings, FormatState & s print_identifier(disk); } - if (sync_replica_mode != SyncReplicaMode::DEFAULT) - { - settings.ostr << ' '; - print_keyword(magic_enum::enum_name(sync_replica_mode)); - - // If the mode is LIGHTWEIGHT and specific source replicas are specified - if (sync_replica_mode == SyncReplicaMode::LIGHTWEIGHT && !src_replicas.empty()) - { - settings.ostr << ' '; - print_keyword("FROM"); - settings.ostr << ' '; - - bool first = true; - for (const auto & src : src_replicas) - { - if (!first) - settings.ostr << ", "; - first = false; - settings.ostr << quoteString(src); - } - } - } break; } case Type::SYNC_DATABASE_REPLICA: diff --git a/src/Parsers/ASTViewTargets.cpp b/src/Parsers/ASTViewTargets.cpp index 8ee98e704df..ffd746cc38a 100644 --- a/src/Parsers/ASTViewTargets.cpp +++ b/src/Parsers/ASTViewTargets.cpp @@ -21,6 +21,9 @@ std::string_view toString(ViewTarget::Kind kind) { case ViewTarget::To: return "to"; case ViewTarget::Inner: return "inner"; + case ViewTarget::Data: return "data"; + case ViewTarget::Tags: return "tags"; + case ViewTarget::Metrics: return "metrics"; } throw Exception(ErrorCodes::LOGICAL_ERROR, "{} doesn't support kind {}", __FUNCTION__, kind); } @@ -254,6 +257,9 @@ std::optional ASTViewTargets::getKeywordForTableID(ViewTarget::Kind kin { case ViewTarget::To: return Keyword::TO; /// TO mydb.mydata case ViewTarget::Inner: return std::nullopt; + case ViewTarget::Data: return Keyword::DATA; /// DATA mydb.mydata + case ViewTarget::Tags: return Keyword::TAGS; /// TAGS mydb.mytags + case ViewTarget::Metrics: return Keyword::METRICS; /// METRICS mydb.mymetrics } UNREACHABLE(); } @@ -264,6 +270,9 @@ std::optional ASTViewTargets::getKeywordForInnerStorage(ViewTarget::Kin { case ViewTarget::To: return std::nullopt; /// ENGINE = MergeTree() case ViewTarget::Inner: return Keyword::INNER; /// INNER ENGINE = MergeTree() + case ViewTarget::Data: return Keyword::DATA; /// DATA ENGINE = MergeTree() + case ViewTarget::Tags: return Keyword::TAGS; /// TAGS ENGINE = MergeTree() + case ViewTarget::Metrics: return Keyword::METRICS; /// METRICS ENGINE = MergeTree() } UNREACHABLE(); } @@ -274,6 +283,9 @@ std::optional ASTViewTargets::getKeywordForInnerUUID(ViewTarget::Kind k { case ViewTarget::To: return Keyword::TO_INNER_UUID; /// TO INNER UUID 'XXX' case ViewTarget::Inner: return std::nullopt; + case ViewTarget::Data: return Keyword::DATA_INNER_UUID; /// DATA INNER UUID 'XXX' + case ViewTarget::Tags: return Keyword::TAGS_INNER_UUID; /// TAGS INNER UUID 'XXX' + case ViewTarget::Metrics: return Keyword::METRICS_INNER_UUID; /// METRICS INNER UUID 'XXX' } UNREACHABLE(); } diff --git a/src/Parsers/ASTViewTargets.h b/src/Parsers/ASTViewTargets.h index 12182919f0e..7814dd5249c 100644 --- a/src/Parsers/ASTViewTargets.h +++ b/src/Parsers/ASTViewTargets.h @@ -9,7 +9,7 @@ namespace DB class ASTStorage; enum class Keyword : size_t; -/// Information about target tables (external or inner) of a materialized view or a window view. +/// Information about target tables (external or inner) of a materialized view or a window view or a TimeSeries table. /// See ASTViewTargets for more details. struct ViewTarget { @@ -24,6 +24,15 @@ struct ViewTarget /// 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, + + /// The "data" table for a TimeSeries table, contains time series. + Data, + + /// The "tags" table for a TimeSeries table, contains identifiers for each combination of a metric name and tags (labels). + Tags, + + /// The "metrics" table for a TimeSeries table, contains general information (metadata) about metrics. + Metrics, }; Kind kind = To; diff --git a/src/Parsers/Access/ASTGrantQuery.cpp b/src/Parsers/Access/ASTGrantQuery.cpp index f60fa7e4a23..eac88c75513 100644 --- a/src/Parsers/Access/ASTGrantQuery.cpp +++ b/src/Parsers/Access/ASTGrantQuery.cpp @@ -97,24 +97,9 @@ namespace void formatCurrentGrantsElements(const AccessRightsElements & elements, const IAST::FormatSettings & settings) { - for (size_t i = 0; i != elements.size(); ++i) - { - const auto & element = elements[i]; - - bool next_element_on_same_db_and_table = false; - if (i != elements.size() - 1) - { - const auto & next_element = elements[i + 1]; - if (element.sameDatabaseAndTableAndParameter(next_element)) - next_element_on_same_db_and_table = true; - } - - if (!next_element_on_same_db_and_table) - { - settings.ostr << " "; - formatONClause(element, settings); - } - } + settings.ostr << "("; + formatElementsWithoutOptions(elements, settings); + settings.ostr << ")"; } } diff --git a/src/Parsers/Access/ParserCreateQuotaQuery.cpp b/src/Parsers/Access/ParserCreateQuotaQuery.cpp index ddfdbe38903..ddf4e9ecda5 100644 --- a/src/Parsers/Access/ParserCreateQuotaQuery.cpp +++ b/src/Parsers/Access/ParserCreateQuotaQuery.cpp @@ -114,7 +114,7 @@ namespace T fieldToNumber(const Field & f) { if (f.getType() == Field::Types::String) - return parseWithSizeSuffix(boost::algorithm::trim_copy(f.get())); + return parseWithSizeSuffix(boost::algorithm::trim_copy(f.safeGet())); else return applyVisitor(FieldVisitorConvertToNumber(), f); } diff --git a/src/Parsers/CommonParsers.h b/src/Parsers/CommonParsers.h index 0ae9ee4833c..34df4b135bb 100644 --- a/src/Parsers/CommonParsers.h +++ b/src/Parsers/CommonParsers.h @@ -116,6 +116,8 @@ namespace DB MR_MACROS(CURRENT_TRANSACTION, "CURRENT TRANSACTION") \ MR_MACROS(CURRENTUSER, "CURRENTUSER") \ MR_MACROS(D, "D") \ + MR_MACROS(DATA, "DATA") \ + MR_MACROS(DATA_INNER_UUID, "DATA INNER UUID") \ MR_MACROS(DATABASE, "DATABASE") \ MR_MACROS(DATABASES, "DATABASES") \ MR_MACROS(DATE, "DATE") \ @@ -288,6 +290,8 @@ namespace DB MR_MACROS(MCS, "MCS") \ MR_MACROS(MEMORY, "MEMORY") \ MR_MACROS(MERGES, "MERGES") \ + MR_MACROS(METRICS, "METRICS") \ + MR_MACROS(METRICS_INNER_UUID, "METRICS INNER UUID") \ MR_MACROS(MI, "MI") \ MR_MACROS(MICROSECOND, "MICROSECOND") \ MR_MACROS(MICROSECONDS, "MICROSECONDS") \ @@ -464,6 +468,8 @@ namespace DB MR_MACROS(TABLE_OVERRIDE, "TABLE OVERRIDE") \ MR_MACROS(TABLE, "TABLE") \ MR_MACROS(TABLES, "TABLES") \ + MR_MACROS(TAGS, "TAGS") \ + MR_MACROS(TAGS_INNER_UUID, "TAGS INNER UUID") \ MR_MACROS(TEMPORARY_TABLE, "TEMPORARY TABLE") \ MR_MACROS(TEMPORARY, "TEMPORARY") \ MR_MACROS(TEST, "TEST") \ diff --git a/src/Parsers/CreateQueryUUIDs.cpp b/src/Parsers/CreateQueryUUIDs.cpp index 4dfee67b537..fbdc6161408 100644 --- a/src/Parsers/CreateQueryUUIDs.cpp +++ b/src/Parsers/CreateQueryUUIDs.cpp @@ -45,6 +45,13 @@ CreateQueryUUIDs::CreateQueryUUIDs(const ASTCreateQuery & query, bool generate_r /// then MV will create inner table. We should generate UUID of inner table here. if (query.is_materialized_view) generate_target_uuid(ViewTarget::To); + + if (query.is_time_series_table) + { + generate_target_uuid(ViewTarget::Data); + generate_target_uuid(ViewTarget::Tags); + generate_target_uuid(ViewTarget::Metrics); + } } } } diff --git a/src/Parsers/ExpressionElementParsers.cpp b/src/Parsers/ExpressionElementParsers.cpp index 9927acdcf17..de395d120d7 100644 --- a/src/Parsers/ExpressionElementParsers.cpp +++ b/src/Parsers/ExpressionElementParsers.cpp @@ -314,7 +314,7 @@ bool ParserCompoundIdentifier::parseImpl(Pos & pos, ASTPtr & node, Expected & ex ASTPtr ast_uuid; if (!uuid_p.parse(pos, ast_uuid, expected)) return false; - uuid = parseFromString(ast_uuid->as()->value.get()); + uuid = parseFromString(ast_uuid->as()->value.safeGet()); } if (parts.size() == 1) node = std::make_shared(parts[0], std::move(params)); @@ -1626,7 +1626,7 @@ bool ParserColumnsTransformers::parseImpl(Pos & pos, ASTPtr & node, Expected & e if (!parser_string_literal.parse(pos, ast_prefix_name, expected)) return false; - column_name_prefix = ast_prefix_name->as().value.get(); + column_name_prefix = ast_prefix_name->as().value.safeGet(); } if (with_open_round_bracket) @@ -1689,7 +1689,7 @@ bool ParserColumnsTransformers::parseImpl(Pos & pos, ASTPtr & node, Expected & e auto res = std::make_shared(); if (regexp_node) - res->setPattern(regexp_node->as().value.get()); + res->setPattern(regexp_node->as().value.safeGet()); else res->children = std::move(identifiers); res->is_strict = is_strict; @@ -1861,7 +1861,7 @@ static bool parseColumnsMatcherBody(IParser::Pos & pos, ASTPtr & node, Expected else { auto regexp_matcher = std::make_shared(); - regexp_matcher->setPattern(regexp_node->as().value.get()); + regexp_matcher->setPattern(regexp_node->as().value.safeGet()); if (!transformers->children.empty()) { @@ -2310,7 +2310,7 @@ bool ParserTTLElement::parseImpl(Pos & pos, ASTPtr & node, Expected & expected) if (!parser_string_literal.parse(pos, ast_space_name, expected)) return false; - destination_name = ast_space_name->as().value.get(); + destination_name = ast_space_name->as().value.safeGet(); } else if (mode == TTLMode::GROUP_BY) { diff --git a/src/Parsers/FieldFromAST.cpp b/src/Parsers/FieldFromAST.cpp index ad1eab49eeb..64aeae1b570 100644 --- a/src/Parsers/FieldFromAST.cpp +++ b/src/Parsers/FieldFromAST.cpp @@ -1,5 +1,4 @@ #include -#include #include #include #include diff --git a/src/Parsers/ParserAlterQuery.cpp b/src/Parsers/ParserAlterQuery.cpp index dbefb0cb966..73fd563faf6 100644 --- a/src/Parsers/ParserAlterQuery.cpp +++ b/src/Parsers/ParserAlterQuery.cpp @@ -517,7 +517,7 @@ bool ParserAlterCommand::parseImpl(Pos & pos, ASTPtr & node, Expected & expected if (!parser_string_literal.parse(pos, ast_space_name, expected)) return false; - command->move_destination_name = ast_space_name->as().value.get(); + command->move_destination_name = ast_space_name->as().value.safeGet(); } else if (s_move_partition.ignore(pos, expected)) { @@ -545,7 +545,7 @@ bool ParserAlterCommand::parseImpl(Pos & pos, ASTPtr & node, Expected & expected if (!parser_string_literal.parse(pos, ast_space_name, expected)) return false; - command->move_destination_name = ast_space_name->as().value.get(); + command->move_destination_name = ast_space_name->as().value.safeGet(); } } else if (s_add_constraint.ignore(pos, expected)) @@ -638,7 +638,7 @@ bool ParserAlterCommand::parseImpl(Pos & pos, ASTPtr & node, Expected & expected if (!parser_string_literal.parse(pos, ast_from, expected)) return false; - command->from = ast_from->as().value.get(); + command->from = ast_from->as().value.safeGet(); command->type = ASTAlterCommand::FETCH_PARTITION; } else if (s_fetch_part.ignore(pos, expected)) @@ -652,7 +652,7 @@ bool ParserAlterCommand::parseImpl(Pos & pos, ASTPtr & node, Expected & expected ASTPtr ast_from; if (!parser_string_literal.parse(pos, ast_from, expected)) return false; - command->from = ast_from->as().value.get(); + command->from = ast_from->as().value.safeGet(); command->part = true; command->type = ASTAlterCommand::FETCH_PARTITION; } @@ -680,7 +680,7 @@ bool ParserAlterCommand::parseImpl(Pos & pos, ASTPtr & node, Expected & expected if (!parser_string_literal.parse(pos, ast_with_name, expected)) return false; - command->with_name = ast_with_name->as().value.get(); + command->with_name = ast_with_name->as().value.safeGet(); } } else if (s_unfreeze.ignore(pos, expected)) @@ -707,7 +707,7 @@ bool ParserAlterCommand::parseImpl(Pos & pos, ASTPtr & node, Expected & expected if (!parser_string_literal.parse(pos, ast_with_name, expected)) return false; - command->with_name = ast_with_name->as().value.get(); + command->with_name = ast_with_name->as().value.safeGet(); } else { diff --git a/src/Parsers/ParserCheckQuery.cpp b/src/Parsers/ParserCheckQuery.cpp index 42716ba7f2c..33b6a5a1ac2 100644 --- a/src/Parsers/ParserCheckQuery.cpp +++ b/src/Parsers/ParserCheckQuery.cpp @@ -55,7 +55,7 @@ bool ParserCheckQuery::parseCheckTable(Pos & pos, ASTPtr & node, Expected & expe const auto * ast_literal = ast_part_name->as(); if (!ast_literal || ast_literal->value.getType() != Field::Types::String) return false; - query->part_name = ast_literal->value.get(); + query->part_name = ast_literal->value.safeGet(); } if (query->database) diff --git a/src/Parsers/ParserCreateIndexQuery.cpp b/src/Parsers/ParserCreateIndexQuery.cpp index 9ebee4cc852..ed89b80edca 100644 --- a/src/Parsers/ParserCreateIndexQuery.cpp +++ b/src/Parsers/ParserCreateIndexQuery.cpp @@ -89,10 +89,8 @@ bool ParserCreateIndexDeclaration::parseImpl(Pos & pos, ASTPtr & node, Expected else { auto index_type = index->getType(); - if (index_type && index_type->name == "annoy") - index->granularity = ASTIndexDeclaration::DEFAULT_ANNOY_INDEX_GRANULARITY; - else if (index_type && index_type->name == "usearch") - index->granularity = ASTIndexDeclaration::DEFAULT_USEARCH_INDEX_GRANULARITY; + if (index_type && index_type->name == "vector_similarity") + index->granularity = ASTIndexDeclaration::DEFAULT_VECTOR_SIMILARITY_INDEX_GRANULARITY; else index->granularity = ASTIndexDeclaration::DEFAULT_INDEX_GRANULARITY; } diff --git a/src/Parsers/ParserCreateQuery.cpp b/src/Parsers/ParserCreateQuery.cpp index a592975613b..cc4e02f46a3 100644 --- a/src/Parsers/ParserCreateQuery.cpp +++ b/src/Parsers/ParserCreateQuery.cpp @@ -214,10 +214,8 @@ bool ParserIndexDeclaration::parseImpl(Pos & pos, ASTPtr & node, Expected & expe else { auto index_type = index->getType(); - if (index_type->name == "annoy") - index->granularity = ASTIndexDeclaration::DEFAULT_ANNOY_INDEX_GRANULARITY; - else if (index_type->name == "usearch") - index->granularity = ASTIndexDeclaration::DEFAULT_USEARCH_INDEX_GRANULARITY; + if (index_type->name == "vector_similarity") + index->granularity = ASTIndexDeclaration::DEFAULT_VECTOR_SIMILARITY_INDEX_GRANULARITY; else index->granularity = ASTIndexDeclaration::DEFAULT_INDEX_GRANULARITY; } @@ -696,6 +694,7 @@ bool ParserCreateTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe ASTPtr table; ASTPtr columns_list; std::shared_ptr storage; + bool is_time_series_table = false; ASTPtr targets; ASTPtr as_database; ASTPtr as_table; @@ -784,6 +783,13 @@ bool ParserCreateTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe return false; storage = typeid_cast>(ast); + + if (storage && storage->engine && (storage->engine->name == "TimeSeries")) + { + is_time_series_table = true; + ParserViewTargets({ViewTarget::Data, ViewTarget::Tags, ViewTarget::Metrics}).parse(pos, targets, expected); + } + return true; }; @@ -873,6 +879,7 @@ bool ParserCreateTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe query->create_or_replace = or_replace; query->if_not_exists = if_not_exists; query->temporary = is_temporary; + query->is_time_series_table = is_time_series_table; query->database = table_id->getDatabase(); query->table = table_id->getTable(); @@ -922,7 +929,7 @@ bool ParserCreateTableQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe query->is_create_empty = is_create_empty; if (from_path) - query->attach_from_path = from_path->as().value.get(); + query->attach_from_path = from_path->as().value.safeGet(); return true; } @@ -1431,7 +1438,7 @@ bool ParserCreateDatabaseQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & e ASTPtr ast_uuid; if (!uuid_p.parse(pos, ast_uuid, expected)) return false; - uuid = parseFromString(ast_uuid->as()->value.get()); + uuid = parseFromString(ast_uuid->as()->value.safeGet()); } if (s_on.ignore(pos, expected)) diff --git a/src/Parsers/ParserDictionary.cpp b/src/Parsers/ParserDictionary.cpp index 83a006231d9..ce38d1b54d1 100644 --- a/src/Parsers/ParserDictionary.cpp +++ b/src/Parsers/ParserDictionary.cpp @@ -33,7 +33,7 @@ bool ParserDictionaryLifetime::parseImpl(Pos & pos, ASTPtr & node, Expected & ex if (literal.value.getType() != Field::Types::UInt64) return false; - res->max_sec = literal.value.get(); + res->max_sec = literal.value.safeGet(); node = res; return true; } @@ -58,10 +58,10 @@ bool ParserDictionaryLifetime::parseImpl(Pos & pos, ASTPtr & node, Expected & ex return false; if (pair.first == "min") - res->min_sec = literal->value.get(); + res->min_sec = literal->value.safeGet(); else if (pair.first == "max") { - res->max_sec = literal->value.get(); + res->max_sec = literal->value.safeGet(); initialized_max = true; } else diff --git a/src/Parsers/ParserPartition.cpp b/src/Parsers/ParserPartition.cpp index 80a28f4803e..ab97b3d0e3b 100644 --- a/src/Parsers/ParserPartition.cpp +++ b/src/Parsers/ParserPartition.cpp @@ -65,7 +65,7 @@ bool ParserPartition::parseImpl(Pos & pos, ASTPtr & node, Expected & expected) { if (literal_ast->value.getType() == Field::Types::Tuple) { - fields_count = literal_ast->value.get().size(); + fields_count = literal_ast->value.safeGet().size(); } else { diff --git a/src/Parsers/ParserSystemQuery.cpp b/src/Parsers/ParserSystemQuery.cpp index 0545c3e5568..81b64ab47c6 100644 --- a/src/Parsers/ParserSystemQuery.cpp +++ b/src/Parsers/ParserSystemQuery.cpp @@ -445,7 +445,7 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & ASTPtr ast; if (!ParserStringLiteral{}.parse(pos, ast, expected)) return false; - String time_str = ast->as().value.get(); + String time_str = ast->as().value.safeGet(); ReadBufferFromString buf(time_str); time_t time; readDateTimeText(time, buf); @@ -467,7 +467,7 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & return false; } - res->seconds = seconds->as()->value.get(); + res->seconds = seconds->as()->value.safeGet(); break; } case Type::DROP_FILESYSTEM_CACHE: @@ -538,7 +538,7 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & ASTPtr ast; if (ParserKeyword{Keyword::WITH_NAME}.ignore(pos, expected) && ParserStringLiteral{}.parse(pos, ast, expected)) { - res->backup_name = ast->as().value.get(); + res->backup_name = ast->as().value.safeGet(); } else { @@ -577,7 +577,7 @@ bool ParserSystemQuery::parseImpl(IParser::Pos & pos, ASTPtr & node, Expected & if (!ParserStringLiteral{}.parse(pos, ast, expected)) return false; - custom_name = ast->as().value.get(); + custom_name = ast->as().value.safeGet(); } return true; diff --git a/src/Parsers/ParserUndropQuery.cpp b/src/Parsers/ParserUndropQuery.cpp index 07ca8a3b5fd..57da47df70d 100644 --- a/src/Parsers/ParserUndropQuery.cpp +++ b/src/Parsers/ParserUndropQuery.cpp @@ -41,7 +41,7 @@ bool parseUndropQuery(IParser::Pos & pos, ASTPtr & node, Expected & expected) ASTPtr ast_uuid; if (!uuid_p.parse(pos, ast_uuid, expected)) return false; - uuid = parseFromString(ast_uuid->as()->value.get()); + uuid = parseFromString(ast_uuid->as()->value.safeGet()); } if (ParserKeyword{Keyword::ON}.ignore(pos, expected)) { diff --git a/src/Parsers/fuzzers/codegen_fuzzer/CMakeLists.txt b/src/Parsers/fuzzers/codegen_fuzzer/CMakeLists.txt index 20fd951d390..74fdcff79f7 100644 --- a/src/Parsers/fuzzers/codegen_fuzzer/CMakeLists.txt +++ b/src/Parsers/fuzzers/codegen_fuzzer/CMakeLists.txt @@ -39,7 +39,7 @@ set(CMAKE_INCLUDE_CURRENT_DIR TRUE) clickhouse_add_executable(codegen_select_fuzzer ${FUZZER_SRCS}) -set_source_files_properties("${PROTO_SRCS}" "out.cpp" PROPERTIES COMPILE_FLAGS "-Wno-reserved-identifier") +set_source_files_properties("${PROTO_SRCS}" "out.cpp" PROPERTIES COMPILE_FLAGS "-Wno-reserved-identifier -Wno-extra-semi-stmt -Wno-used-but-marked-unused") # contrib/libprotobuf-mutator/src/libfuzzer/libfuzzer_macro.h:143:44: error: no newline at end of file [-Werror,-Wnewline-eof] target_compile_options (codegen_select_fuzzer PRIVATE -Wno-newline-eof) diff --git a/src/Parsers/tests/gtest_dictionary_parser.cpp b/src/Parsers/tests/gtest_dictionary_parser.cpp index a1ba46125a7..af3591750a1 100644 --- a/src/Parsers/tests/gtest_dictionary_parser.cpp +++ b/src/Parsers/tests/gtest_dictionary_parser.cpp @@ -56,21 +56,21 @@ TEST(ParserDictionaryDDL, SimpleDictionary) EXPECT_EQ(create->dictionary->source->name, "clickhouse"); auto children = create->dictionary->source->elements->children; EXPECT_EQ(children[0]->as() -> first, "host"); - EXPECT_EQ(children[0]->as()->second->as()->value.get(), "localhost"); + EXPECT_EQ(children[0]->as()->second->as()->value.safeGet(), "localhost"); EXPECT_EQ(children[1]->as()->first, "port"); - EXPECT_EQ(children[1]->as()->second->as()->value.get(), 9000); + EXPECT_EQ(children[1]->as()->second->as()->value.safeGet(), 9000); EXPECT_EQ(children[2]->as()->first, "user"); - EXPECT_EQ(children[2]->as()->second->as()->value.get(), "default"); + EXPECT_EQ(children[2]->as()->second->as()->value.safeGet(), "default"); EXPECT_EQ(children[3]->as()->first, "password"); - EXPECT_EQ(children[3]->as()->second->as()->value.get(), ""); + EXPECT_EQ(children[3]->as()->second->as()->value.safeGet(), ""); EXPECT_EQ(children[4]->as()->first, "db"); - EXPECT_EQ(children[4]->as()->second->as()->value.get(), "test"); + EXPECT_EQ(children[4]->as()->second->as()->value.safeGet(), "test"); EXPECT_EQ(children[5]->as()->first, "table"); - EXPECT_EQ(children[5]->as()->second->as()->value.get(), "table_for_dict"); + EXPECT_EQ(children[5]->as()->second->as()->value.safeGet(), "table_for_dict"); /// layout test auto * layout = create->dictionary->layout; @@ -102,9 +102,9 @@ TEST(ParserDictionaryDDL, SimpleDictionary) EXPECT_EQ(attributes_children[1]->as()->name, "second_column"); EXPECT_EQ(attributes_children[2]->as()->name, "third_column"); - EXPECT_EQ(attributes_children[0]->as()->default_value->as()->value.get(), 0); - EXPECT_EQ(attributes_children[1]->as()->default_value->as()->value.get(), 1); - EXPECT_EQ(attributes_children[2]->as()->default_value->as()->value.get(), 2); + EXPECT_EQ(attributes_children[0]->as()->default_value->as()->value.safeGet(), 0); + EXPECT_EQ(attributes_children[1]->as()->default_value->as()->value.safeGet(), 1); + EXPECT_EQ(attributes_children[2]->as()->default_value->as()->value.safeGet(), 2); EXPECT_EQ(attributes_children[0]->as()->expression, nullptr); EXPECT_EQ(attributes_children[1]->as()->expression, nullptr); @@ -150,8 +150,8 @@ TEST(ParserDictionaryDDL, AttributesWithMultipleProperties) EXPECT_EQ(attributes_children[2]->as()->name, "third_column"); EXPECT_EQ(attributes_children[0]->as()->default_value, nullptr); - EXPECT_EQ(attributes_children[1]->as()->default_value->as()->value.get(), 1); - EXPECT_EQ(attributes_children[2]->as()->default_value->as()->value.get(), 2); + EXPECT_EQ(attributes_children[1]->as()->default_value->as()->value.safeGet(), 1); + EXPECT_EQ(attributes_children[2]->as()->default_value->as()->value.safeGet(), 2); EXPECT_EQ(attributes_children[0]->as()->expression, nullptr); EXPECT_EQ(attributes_children[1]->as()->expression, nullptr); @@ -195,9 +195,9 @@ TEST(ParserDictionaryDDL, CustomAttributePropertiesOrder) EXPECT_EQ(attributes_children[1]->as()->name, "second_column"); EXPECT_EQ(attributes_children[2]->as()->name, "third_column"); - EXPECT_EQ(attributes_children[0]->as()->default_value->as()->value.get(), 100); - EXPECT_EQ(attributes_children[1]->as()->default_value->as()->value.get(), 1); - EXPECT_EQ(attributes_children[2]->as()->default_value->as()->value.get(), 2); + EXPECT_EQ(attributes_children[0]->as()->default_value->as()->value.safeGet(), 100); + EXPECT_EQ(attributes_children[1]->as()->default_value->as()->value.safeGet(), 1); + EXPECT_EQ(attributes_children[2]->as()->default_value->as()->value.safeGet(), 2); EXPECT_EQ(attributes_children[0]->as()->expression, nullptr); EXPECT_EQ(attributes_children[1]->as()->expression, nullptr); @@ -248,25 +248,25 @@ TEST(ParserDictionaryDDL, NestedSource) auto children = create->dictionary->source->elements->children; EXPECT_EQ(children[0]->as()->first, "host"); - EXPECT_EQ(children[0]->as()->second->as()->value.get(), "localhost"); + EXPECT_EQ(children[0]->as()->second->as()->value.safeGet(), "localhost"); EXPECT_EQ(children[1]->as()->first, "port"); - EXPECT_EQ(children[1]->as()->second->as()->value.get(), 9000); + EXPECT_EQ(children[1]->as()->second->as()->value.safeGet(), 9000); EXPECT_EQ(children[2]->as()->first, "user"); - EXPECT_EQ(children[2]->as()->second->as()->value.get(), "default"); + EXPECT_EQ(children[2]->as()->second->as()->value.safeGet(), "default"); EXPECT_EQ(children[3]->as()->first, "replica"); auto replica = children[3]->as()->second->children; EXPECT_EQ(replica[0]->as()->first, "host"); - EXPECT_EQ(replica[0]->as()->second->as()->value.get(), "127.0.0.1"); + EXPECT_EQ(replica[0]->as()->second->as()->value.safeGet(), "127.0.0.1"); EXPECT_EQ(replica[1]->as()->first, "priority"); - EXPECT_EQ(replica[1]->as()->second->as()->value.get(), 1); + EXPECT_EQ(replica[1]->as()->second->as()->value.safeGet(), 1); EXPECT_EQ(children[4]->as()->first, "password"); - EXPECT_EQ(children[4]->as()->second->as()->value.get(), ""); + EXPECT_EQ(children[4]->as()->second->as()->value.safeGet(), ""); } diff --git a/src/Planner/Planner.cpp b/src/Planner/Planner.cpp index b837d9428a1..d3d20c6fba0 100644 --- a/src/Planner/Planner.cpp +++ b/src/Planner/Planner.cpp @@ -1099,7 +1099,7 @@ void addBuildSubqueriesForSetsStepIfNeeded( auto query_tree = subquery->detachQueryTree(); auto subquery_options = select_query_options.subquery(); /// I don't know if this is a good decision, - /// But for now it is done in the same way as in old analyzer. + /// but for now it is done in the same way as in old analyzer. /// This would not ignore limits for subqueries (affects mutations only). /// See test_build_sets_from_multiple_threads-analyzer. subquery_options.ignore_limits = false; diff --git a/src/Planner/PlannerActionsVisitor.cpp b/src/Planner/PlannerActionsVisitor.cpp index 57457493844..43177fc73c0 100644 --- a/src/Planner/PlannerActionsVisitor.cpp +++ b/src/Planner/PlannerActionsVisitor.cpp @@ -1,5 +1,6 @@ #include +#include #include #include #include @@ -237,7 +238,7 @@ public: if (function_node.isWindowFunction()) { buffer << " OVER ("; - buffer << calculateWindowNodeActionName(function_node.getWindowNode()); + buffer << calculateWindowNodeActionName(node, function_node.getWindowNode()); buffer << ')'; } @@ -298,21 +299,22 @@ public: return calculateConstantActionNodeName(constant_literal, applyVisitor(FieldToDataType(), constant_literal)); } - String calculateWindowNodeActionName(const QueryTreeNodePtr & node) + String calculateWindowNodeActionName(const QueryTreeNodePtr & function_nodew_node_, const QueryTreeNodePtr & window_node_) { - auto & window_node = node->as(); + const auto & function_node = function_nodew_node_->as(); + const auto & window_node = window_node_->as(); WriteBufferFromOwnString buffer; if (window_node.hasPartitionBy()) { buffer << "PARTITION BY "; - auto & partition_by_nodes = window_node.getPartitionBy().getNodes(); + const auto & partition_by_nodes = window_node.getPartitionBy().getNodes(); size_t partition_by_nodes_size = partition_by_nodes.size(); for (size_t i = 0; i < partition_by_nodes_size; ++i) { - auto & partition_by_node = partition_by_nodes[i]; + const auto & partition_by_node = partition_by_nodes[i]; buffer << calculateActionNodeName(partition_by_node); if (i + 1 != partition_by_nodes_size) buffer << ", "; @@ -326,7 +328,7 @@ public: buffer << "ORDER BY "; - auto & order_by_nodes = window_node.getOrderBy().getNodes(); + const auto & order_by_nodes = window_node.getOrderBy().getNodes(); size_t order_by_nodes_size = order_by_nodes.size(); for (size_t i = 0; i < order_by_nodes_size; ++i) @@ -364,44 +366,14 @@ public: } } - auto & window_frame = window_node.getWindowFrame(); - if (!window_frame.is_default) + auto window_frame_opt = extractWindowFrame(function_node); + if (window_frame_opt) { + auto & window_frame = *window_frame_opt; if (window_node.hasPartitionBy() || window_node.hasOrderBy()) buffer << ' '; - buffer << window_frame.type << " BETWEEN "; - if (window_frame.begin_type == WindowFrame::BoundaryType::Current) - { - buffer << "CURRENT ROW"; - } - else if (window_frame.begin_type == WindowFrame::BoundaryType::Unbounded) - { - buffer << "UNBOUNDED"; - buffer << " " << (window_frame.begin_preceding ? "PRECEDING" : "FOLLOWING"); - } - else - { - buffer << calculateActionNodeName(window_node.getFrameBeginOffsetNode()); - buffer << " " << (window_frame.begin_preceding ? "PRECEDING" : "FOLLOWING"); - } - - buffer << " AND "; - - if (window_frame.end_type == WindowFrame::BoundaryType::Current) - { - buffer << "CURRENT ROW"; - } - else if (window_frame.end_type == WindowFrame::BoundaryType::Unbounded) - { - buffer << "UNBOUNDED"; - buffer << " " << (window_frame.end_preceding ? "PRECEDING" : "FOLLOWING"); - } - else - { - buffer << calculateActionNodeName(window_node.getFrameEndOffsetNode()); - buffer << " " << (window_frame.end_preceding ? "PRECEDING" : "FOLLOWING"); - } + window_frame.toString(buffer); } return buffer.str(); @@ -1056,20 +1028,11 @@ String calculateConstantActionNodeName(const Field & constant_literal) return ActionNodeNameHelper::calculateConstantActionNodeName(constant_literal); } -String calculateWindowNodeActionName(const QueryTreeNodePtr & node, - const PlannerContext & planner_context, - QueryTreeNodeToName & node_to_name, - bool use_column_identifier_as_action_node_name) -{ - ActionNodeNameHelper helper(node_to_name, planner_context, use_column_identifier_as_action_node_name); - return helper.calculateWindowNodeActionName(node); -} - -String calculateWindowNodeActionName(const QueryTreeNodePtr & node, const PlannerContext & planner_context, bool use_column_identifier_as_action_node_name) +String calculateWindowNodeActionName(const QueryTreeNodePtr & function_node, const QueryTreeNodePtr & window_node, const PlannerContext & planner_context, bool use_column_identifier_as_action_node_name) { QueryTreeNodeToName empty_map; ActionNodeNameHelper helper(empty_map, planner_context, use_column_identifier_as_action_node_name); - return helper.calculateWindowNodeActionName(node); + return helper.calculateWindowNodeActionName(function_node, window_node); } } diff --git a/src/Planner/PlannerActionsVisitor.h b/src/Planner/PlannerActionsVisitor.h index 6bb32047327..4f608ad3f7b 100644 --- a/src/Planner/PlannerActionsVisitor.h +++ b/src/Planner/PlannerActionsVisitor.h @@ -1,5 +1,6 @@ #pragma once +#include #include #include @@ -8,6 +9,7 @@ #include #include +#include namespace DB { @@ -73,16 +75,8 @@ String calculateConstantActionNodeName(const Field & constant_literal); * Window node action name can only be part of window function action name. * For column node column node identifier from planner context is used, if use_column_identifier_as_action_node_name = true. */ -String calculateWindowNodeActionName(const QueryTreeNodePtr & node, - const PlannerContext & planner_context, - QueryTreeNodeToName & node_to_name, - bool use_column_identifier_as_action_node_name = true); - -/** Calculate action node name for window node. - * Window node action name can only be part of window function action name. - * For column node column node identifier from planner context is used, if use_column_identifier_as_action_node_name = true. - */ -String calculateWindowNodeActionName(const QueryTreeNodePtr & node, +String calculateWindowNodeActionName(const QueryTreeNodePtr & function_node, + const QueryTreeNodePtr & window_node, const PlannerContext & planner_context, bool use_column_identifier_as_action_node_name = true); diff --git a/src/Planner/PlannerJoinTree.cpp b/src/Planner/PlannerJoinTree.cpp index a3db0395ccc..bc31af32a20 100644 --- a/src/Planner/PlannerJoinTree.cpp +++ b/src/Planner/PlannerJoinTree.cpp @@ -417,20 +417,20 @@ void updatePrewhereOutputsIfNeeded(SelectQueryInfo & table_expression_query_info /// We evaluate sampling for Merge lazily so we need to get all the columns if (storage_snapshot->storage.getName() == "Merge") { - const auto columns = storage_snapshot->getMetadataForQuery()->getColumns().getAll(); + const auto columns = storage_snapshot->metadata->getColumns().getAll(); for (const auto & column : columns) required_columns.insert(column.name); } else { - auto columns_required_for_sampling = storage_snapshot->getMetadataForQuery()->getColumnsRequiredForSampling(); + auto columns_required_for_sampling = storage_snapshot->metadata->getColumnsRequiredForSampling(); required_columns.insert(columns_required_for_sampling.begin(), columns_required_for_sampling.end()); } } if (table_expression_modifiers->hasFinal()) { - auto columns_required_for_final = storage_snapshot->getMetadataForQuery()->getColumnsRequiredForFinal(); + auto columns_required_for_final = storage_snapshot->metadata->getColumnsRequiredForFinal(); required_columns.insert(columns_required_for_final.begin(), columns_required_for_final.end()); } } diff --git a/src/Planner/PlannerWindowFunctions.cpp b/src/Planner/PlannerWindowFunctions.cpp index 225852de5a7..f91cf644cf0 100644 --- a/src/Planner/PlannerWindowFunctions.cpp +++ b/src/Planner/PlannerWindowFunctions.cpp @@ -1,5 +1,7 @@ +#include #include +#include #include #include #include @@ -8,8 +10,9 @@ #include -#include #include +#include +#include namespace DB { @@ -22,27 +25,33 @@ namespace ErrorCodes namespace { -WindowDescription extractWindowDescriptionFromWindowNode(const QueryTreeNodePtr & node, const PlannerContext & planner_context) +WindowDescription extractWindowDescriptionFromWindowNode(const QueryTreeNodePtr & func_node_, const PlannerContext & planner_context) { + const auto & func_node = func_node_->as(); + auto node = func_node.getWindowNode(); auto & window_node = node->as(); WindowDescription window_description; - window_description.window_name = calculateWindowNodeActionName(node, planner_context); + window_description.window_name = calculateWindowNodeActionName(func_node_, node, planner_context); for (const auto & partition_by_node : window_node.getPartitionBy().getNodes()) { auto partition_by_node_action_name = calculateActionNodeName(partition_by_node, planner_context); - auto partition_by_sort_column_description = SortColumnDescription(partition_by_node_action_name, 1 /* direction */, 1 /* nulls_direction */); + auto partition_by_sort_column_description + = SortColumnDescription(partition_by_node_action_name, 1 /* direction */, 1 /* nulls_direction */); window_description.partition_by.push_back(std::move(partition_by_sort_column_description)); } window_description.order_by = extractSortDescription(window_node.getOrderByNode(), planner_context); window_description.full_sort_description = window_description.partition_by; - window_description.full_sort_description.insert(window_description.full_sort_description.end(), window_description.order_by.begin(), window_description.order_by.end()); + window_description.full_sort_description.insert( + window_description.full_sort_description.end(), window_description.order_by.begin(), window_description.order_by.end()); /// WINDOW frame is validated during query analysis stage - window_description.frame = window_node.getWindowFrame(); + auto window_frame = extractWindowFrame(func_node); + window_description.frame = window_frame ? *window_frame : window_node.getWindowFrame(); + auto node_frame = window_node.getWindowFrame(); const auto & query_context = planner_context.getQueryContext(); const auto & query_context_settings = query_context->getSettingsRef(); @@ -64,7 +73,8 @@ WindowDescription extractWindowDescriptionFromWindowNode(const QueryTreeNodePtr } -std::vector extractWindowDescriptions(const QueryTreeNodes & window_function_nodes, const PlannerContext & planner_context) +std::vector +extractWindowDescriptions(const QueryTreeNodes & window_function_nodes, const PlannerContext & planner_context) { std::unordered_map window_name_to_description; @@ -72,7 +82,7 @@ std::vector extractWindowDescriptions(const QueryTreeNodes & { auto & window_function_node_typed = window_function_node->as(); - auto function_window_description = extractWindowDescriptionFromWindowNode(window_function_node_typed.getWindowNode(), planner_context); + auto function_window_description = extractWindowDescriptionFromWindowNode(window_function_node, planner_context); auto frame_type = function_window_description.frame.type; if (frame_type != WindowFrame::FrameType::ROWS && frame_type != WindowFrame::FrameType::RANGE) diff --git a/src/Planner/Utils.cpp b/src/Planner/Utils.cpp index a6e94a124e6..822a3e9465e 100644 --- a/src/Planner/Utils.cpp +++ b/src/Planner/Utils.cpp @@ -22,6 +22,8 @@ #include +#include + #include #include #include @@ -34,6 +36,7 @@ #include #include #include +#include #include @@ -507,4 +510,20 @@ void appendSetsFromActionsDAG(const ActionsDAG & dag, UsefulSets & useful_sets) } } +std::optional extractWindowFrame(const FunctionNode & node) +{ + if (!node.isWindowFunction()) + return {}; + auto & window_node = node.getWindowNode()->as(); + const auto & window_frame = window_node.getWindowFrame(); + if (!window_frame.is_default) + return window_frame; + auto aggregate_function = node.getAggregateFunction(); + if (const auto * win_func = dynamic_cast(aggregate_function.get())) + { + return win_func->getDefaultFrame(); + } + return {}; +} + } diff --git a/src/Planner/Utils.h b/src/Planner/Utils.h index ae60976a8d6..254b8f4eae1 100644 --- a/src/Planner/Utils.h +++ b/src/Planner/Utils.h @@ -19,6 +19,8 @@ #include +#include + namespace DB { @@ -91,4 +93,9 @@ ASTPtr parseAdditionalResultFilter(const Settings & settings); using UsefulSets = std::unordered_set; void appendSetsFromActionsDAG(const ActionsDAG & dag, UsefulSets & useful_sets); +/// If the window frame is not set in sql, try to use the default frame from window function +/// if it have any one. Otherwise return empty. +/// If the window frame is set in sql, use it anyway. +std::optional extractWindowFrame(const FunctionNode & node); + } diff --git a/src/Processors/Executors/CompletedPipelineExecutor.cpp b/src/Processors/Executors/CompletedPipelineExecutor.cpp index 598a51bf0c7..888835c9beb 100644 --- a/src/Processors/Executors/CompletedPipelineExecutor.cpp +++ b/src/Processors/Executors/CompletedPipelineExecutor.cpp @@ -97,7 +97,9 @@ void CompletedPipelineExecutor::execute() break; if (is_cancelled_callback()) + { data->executor->cancel(); + } } if (data->has_exception) @@ -116,7 +118,9 @@ CompletedPipelineExecutor::~CompletedPipelineExecutor() try { if (data && data->executor) + { data->executor->cancel(); + } } catch (...) { diff --git a/src/Processors/Executors/ExecutingGraph.cpp b/src/Processors/Executors/ExecutingGraph.cpp index 6d5b60d8159..10470325bb8 100644 --- a/src/Processors/Executors/ExecutingGraph.cpp +++ b/src/Processors/Executors/ExecutingGraph.cpp @@ -96,7 +96,7 @@ bool ExecutingGraph::addEdges(uint64_t node) return was_edge_added; } -bool ExecutingGraph::expandPipeline(std::stack & stack, uint64_t pid) +ExecutingGraph::UpdateNodeStatus ExecutingGraph::expandPipeline(std::stack & stack, uint64_t pid) { auto & cur_node = *nodes[pid]; Processors new_processors; @@ -108,7 +108,7 @@ bool ExecutingGraph::expandPipeline(std::stack & stack, uint64_t pid) catch (...) { cur_node.exception = std::current_exception(); - return false; + return UpdateNodeStatus::Exception; } { @@ -118,7 +118,7 @@ bool ExecutingGraph::expandPipeline(std::stack & stack, uint64_t pid) { for (auto & processor : new_processors) processor->cancel(); - return false; + return UpdateNodeStatus::Cancelled; } processors->insert(processors->end(), new_processors.begin(), new_processors.end()); @@ -178,7 +178,7 @@ bool ExecutingGraph::expandPipeline(std::stack & stack, uint64_t pid) } } - return true; + return UpdateNodeStatus::Done; } void ExecutingGraph::initializeExecution(Queue & queue) @@ -213,7 +213,7 @@ void ExecutingGraph::initializeExecution(Queue & queue) } -bool ExecutingGraph::updateNode(uint64_t pid, Queue & queue, Queue & async_queue) +ExecutingGraph::UpdateNodeStatus ExecutingGraph::updateNode(uint64_t pid, Queue & queue, Queue & async_queue) { std::stack updated_edges; std::stack updated_processors; @@ -309,7 +309,7 @@ bool ExecutingGraph::updateNode(uint64_t pid, Queue & queue, Queue & async_queue catch (...) { node.exception = std::current_exception(); - return false; + return UpdateNodeStatus::Exception; } #ifndef NDEBUG @@ -386,8 +386,9 @@ bool ExecutingGraph::updateNode(uint64_t pid, Queue & queue, Queue & async_queue read_lock.unlock(); { std::unique_lock lock(nodes_mutex); - if (!expandPipeline(updated_processors, pid)) - return false; + auto status = expandPipeline(updated_processors, pid); + if (status != UpdateNodeStatus::Done) + return status; } read_lock.lock(); @@ -397,7 +398,7 @@ bool ExecutingGraph::updateNode(uint64_t pid, Queue & queue, Queue & async_queue } } - return true; + return UpdateNodeStatus::Done; } void ExecutingGraph::cancel(bool cancel_all_processors) diff --git a/src/Processors/Executors/ExecutingGraph.h b/src/Processors/Executors/ExecutingGraph.h index 71dcd360a2c..e1a6ac96203 100644 --- a/src/Processors/Executors/ExecutingGraph.h +++ b/src/Processors/Executors/ExecutingGraph.h @@ -138,10 +138,17 @@ public: /// Traverse graph the first time to update all the childless nodes. void initializeExecution(Queue & queue); + enum class UpdateNodeStatus + { + Done, + Exception, + Cancelled, + }; + /// Update processor with pid number (call IProcessor::prepare). /// Check parents and children of current processor and push them to stacks if they also need to be updated. /// If processor wants to be expanded, lock will be upgraded to get write access to pipeline. - bool updateNode(uint64_t pid, Queue & queue, Queue & async_queue); + UpdateNodeStatus updateNode(uint64_t pid, Queue & queue, Queue & async_queue); void cancel(bool cancel_all_processors = true); @@ -155,7 +162,7 @@ private: /// Update graph after processor (pid) returned ExpandPipeline status. /// All new nodes and nodes with updated ports are pushed into stack. - bool expandPipeline(std::stack & stack, uint64_t pid); + UpdateNodeStatus expandPipeline(std::stack & stack, uint64_t pid); std::shared_ptr processors; std::vector source_processors; diff --git a/src/Processors/Executors/PipelineExecutor.cpp b/src/Processors/Executors/PipelineExecutor.cpp index 82cad471a29..23b3a6d9f5f 100644 --- a/src/Processors/Executors/PipelineExecutor.cpp +++ b/src/Processors/Executors/PipelineExecutor.cpp @@ -77,9 +77,9 @@ const Processors & PipelineExecutor::getProcessors() const return graph->getProcessors(); } -void PipelineExecutor::cancel() +void PipelineExecutor::cancel(ExecutionStatus reason) { - cancelled = true; + tryUpdateExecutionStatus(ExecutionStatus::Executing, reason); finish(); graph->cancel(); } @@ -98,6 +98,11 @@ void PipelineExecutor::finish() tasks.finish(); } +bool PipelineExecutor::tryUpdateExecutionStatus(ExecutionStatus expected, ExecutionStatus desired) +{ + return execution_status.compare_exchange_strong(expected, desired); +} + void PipelineExecutor::execute(size_t num_threads, bool concurrency_control) { checkTimeLimit(); @@ -120,7 +125,7 @@ void PipelineExecutor::execute(size_t num_threads, bool concurrency_control) } catch (...) { - span.addAttribute(ExecutionStatus::fromCurrentException()); + span.addAttribute(DB::ExecutionStatus::fromCurrentException()); #ifndef NDEBUG LOG_TRACE(log, "Exception while executing query. Current state:\n{}", dumpPipeline()); @@ -169,7 +174,7 @@ bool PipelineExecutor::checkTimeLimitSoft() // We call cancel here so that all processors are notified and tasks waken up // so that the "break" is faster and doesn't wait for long events if (!continuing) - cancel(); + cancel(ExecutionStatus::CancelledByTimeout); return continuing; } @@ -195,7 +200,8 @@ void PipelineExecutor::finalizeExecution() { checkTimeLimit(); - if (cancelled) + auto status = execution_status.load(); + if (status == ExecutionStatus::CancelledByTimeout || status == ExecutionStatus::CancelledByUser) return; bool all_processors_finished = true; @@ -271,7 +277,7 @@ void PipelineExecutor::executeStepImpl(size_t thread_num, std::atomic_bool * yie break; if (!context.executeTask()) - cancel(); + cancel(ExecutionStatus::Exception); if (tasks.isFinished()) break; @@ -289,11 +295,13 @@ void PipelineExecutor::executeStepImpl(size_t thread_num, std::atomic_bool * yie Queue async_queue; /// Prepare processor after execution. - if (!graph->updateNode(context.getProcessorID(), queue, async_queue)) - cancel(); + auto status = graph->updateNode(context.getProcessorID(), queue, async_queue); + if (status == ExecutingGraph::UpdateNodeStatus::Exception) + cancel(ExecutionStatus::Exception); /// Push other tasks to global queue. - tasks.pushTasks(queue, async_queue, context); + if (status == ExecutingGraph::UpdateNodeStatus::Done) + tasks.pushTasks(queue, async_queue, context); } #ifndef NDEBUG @@ -309,7 +317,7 @@ void PipelineExecutor::executeStepImpl(size_t thread_num, std::atomic_bool * yie { /// spawnThreads can throw an exception, for example CANNOT_SCHEDULE_TASK. /// We should cancel execution properly before rethrow. - cancel(); + cancel(ExecutionStatus::Exception); throw; } @@ -328,6 +336,7 @@ void PipelineExecutor::executeStepImpl(size_t thread_num, std::atomic_bool * yie void PipelineExecutor::initializeExecution(size_t num_threads, bool concurrency_control) { is_execution_initialized = true; + tryUpdateExecutionStatus(ExecutionStatus::NotStarted, ExecutionStatus::Executing); size_t use_threads = num_threads; @@ -393,7 +402,7 @@ void PipelineExecutor::executeImpl(size_t num_threads, bool concurrency_control) { /// If finished_flag is not set, there was an exception. /// Cancel execution in this case. - cancel(); + cancel(ExecutionStatus::Exception); if (pool) pool->wait(); } diff --git a/src/Processors/Executors/PipelineExecutor.h b/src/Processors/Executors/PipelineExecutor.h index ae119355cb5..79d0a29d4e1 100644 --- a/src/Processors/Executors/PipelineExecutor.h +++ b/src/Processors/Executors/PipelineExecutor.h @@ -48,8 +48,20 @@ public: const Processors & getProcessors() const; + enum class ExecutionStatus + { + NotStarted, + Executing, + Finished, + Exception, + CancelledByUser, + CancelledByTimeout, + }; + /// Cancel execution. May be called from another thread. - void cancel(); + void cancel() { cancel(ExecutionStatus::CancelledByUser); } + + ExecutionStatus getExecutionStatus() const { return execution_status.load(); } /// Cancel processors which only read data from source. May be called from another thread. void cancelReading(); @@ -81,7 +93,7 @@ private: /// system.opentelemetry_span_log bool trace_processors = false; - std::atomic_bool cancelled = false; + std::atomic execution_status = ExecutionStatus::NotStarted; std::atomic_bool cancelled_reading = false; LoggerPtr log = getLogger("PipelineExecutor"); @@ -105,6 +117,10 @@ private: void executeStepImpl(size_t thread_num, std::atomic_bool * yield_flag = nullptr); void executeSingleThread(size_t thread_num); void finish(); + void cancel(ExecutionStatus reason); + + /// If execution_status == from, change it to desired. + bool tryUpdateExecutionStatus(ExecutionStatus expected, ExecutionStatus desired); String dumpPipeline() const; }; diff --git a/src/Processors/Executors/PushingAsyncPipelineExecutor.cpp b/src/Processors/Executors/PushingAsyncPipelineExecutor.cpp index 830a96533ed..866d224a08d 100644 --- a/src/Processors/Executors/PushingAsyncPipelineExecutor.cpp +++ b/src/Processors/Executors/PushingAsyncPipelineExecutor.cpp @@ -15,6 +15,7 @@ namespace DB namespace ErrorCodes { extern const int LOGICAL_ERROR; + extern const int QUERY_WAS_CANCELLED; } class PushingAsyncSource : public ISource @@ -176,6 +177,16 @@ void PushingAsyncPipelineExecutor::start() data->thread = ThreadFromGlobalPool(std::move(func)); } +[[noreturn]] static void throwOnExecutionStatus(PipelineExecutor::ExecutionStatus status) +{ + if (status == PipelineExecutor::ExecutionStatus::CancelledByTimeout + || status == PipelineExecutor::ExecutionStatus::CancelledByUser) + throw Exception(ErrorCodes::QUERY_WAS_CANCELLED, "Query was cancelled"); + + throw Exception(ErrorCodes::LOGICAL_ERROR, + "Pipeline for PushingPipelineExecutor was finished before all data was inserted"); +} + void PushingAsyncPipelineExecutor::push(Chunk chunk) { if (!started) @@ -185,8 +196,7 @@ void PushingAsyncPipelineExecutor::push(Chunk chunk) data->rethrowExceptionIfHas(); if (!is_pushed) - throw Exception(ErrorCodes::LOGICAL_ERROR, - "Pipeline for PushingAsyncPipelineExecutor was finished before all data was inserted"); + throwOnExecutionStatus(data->executor->getExecutionStatus()); } void PushingAsyncPipelineExecutor::push(Block block) diff --git a/src/Processors/Executors/PushingPipelineExecutor.cpp b/src/Processors/Executors/PushingPipelineExecutor.cpp index 696932932df..7a1c0111a3a 100644 --- a/src/Processors/Executors/PushingPipelineExecutor.cpp +++ b/src/Processors/Executors/PushingPipelineExecutor.cpp @@ -11,6 +11,7 @@ namespace DB namespace ErrorCodes { extern const int LOGICAL_ERROR; + extern const int QUERY_WAS_CANCELLED; } class PushingSource : public ISource @@ -80,6 +81,15 @@ const Block & PushingPipelineExecutor::getHeader() const return pushing_source->getPort().getHeader(); } +[[noreturn]] static void throwOnExecutionStatus(PipelineExecutor::ExecutionStatus status) +{ + if (status == PipelineExecutor::ExecutionStatus::CancelledByTimeout + || status == PipelineExecutor::ExecutionStatus::CancelledByUser) + throw Exception(ErrorCodes::QUERY_WAS_CANCELLED, "Query was cancelled"); + + throw Exception(ErrorCodes::LOGICAL_ERROR, + "Pipeline for PushingPipelineExecutor was finished before all data was inserted"); +} void PushingPipelineExecutor::start() { @@ -91,8 +101,7 @@ void PushingPipelineExecutor::start() executor->setReadProgressCallback(pipeline.getReadProgressCallback()); if (!executor->executeStep(&input_wait_flag)) - throw Exception(ErrorCodes::LOGICAL_ERROR, - "Pipeline for PushingPipelineExecutor was finished before all data was inserted"); + throwOnExecutionStatus(executor->getExecutionStatus()); } void PushingPipelineExecutor::push(Chunk chunk) @@ -103,8 +112,7 @@ void PushingPipelineExecutor::push(Chunk chunk) pushing_source->setData(std::move(chunk)); if (!executor->executeStep(&input_wait_flag)) - throw Exception(ErrorCodes::LOGICAL_ERROR, - "Pipeline for PushingPipelineExecutor was finished before all data was inserted"); + throwOnExecutionStatus(executor->getExecutionStatus()); } void PushingPipelineExecutor::push(Block block) diff --git a/src/Processors/Formats/IOutputFormat.cpp b/src/Processors/Formats/IOutputFormat.cpp index 88a6fb1e92f..97628778adb 100644 --- a/src/Processors/Formats/IOutputFormat.cpp +++ b/src/Processors/Formats/IOutputFormat.cpp @@ -69,9 +69,10 @@ void IOutputFormat::work() if (finished && !finalized) { - if (rows_before_limit_counter && rows_before_limit_counter->hasAppliedLimit()) + if (rows_before_limit_counter && rows_before_limit_counter->hasAppliedStep()) setRowsBeforeLimit(rows_before_limit_counter->get()); - + if (rows_before_aggregation_counter && rows_before_aggregation_counter->hasAppliedStep()) + setRowsBeforeAggregation(rows_before_aggregation_counter->get()); finalize(); if (auto_flush) flush(); diff --git a/src/Processors/Formats/IOutputFormat.h b/src/Processors/Formats/IOutputFormat.h index cae2ab7691e..e9af4ca7cf5 100644 --- a/src/Processors/Formats/IOutputFormat.h +++ b/src/Processors/Formats/IOutputFormat.h @@ -1,9 +1,9 @@ #pragma once #include -#include -#include #include +#include +#include #include namespace DB @@ -36,14 +36,20 @@ public: void setAutoFlush() { auto_flush = true; } /// Value for rows_before_limit_at_least field. - virtual void setRowsBeforeLimit(size_t /*rows_before_limit*/) {} + virtual void setRowsBeforeLimit(size_t /*rows_before_limit*/) { } /// Counter to calculate rows_before_limit_at_least in processors pipeline. - void setRowsBeforeLimitCounter(RowsBeforeLimitCounterPtr counter) override { rows_before_limit_counter.swap(counter); } + void setRowsBeforeLimitCounter(RowsBeforeStepCounterPtr counter) override { rows_before_limit_counter.swap(counter); } + + /// Value for rows_before_aggregation field. + virtual void setRowsBeforeAggregation(size_t /*rows_before_aggregation*/) { } + + /// Counter to calculate rows_before_aggregation in processors pipeline. + void setRowsBeforeAggregationCounter(RowsBeforeStepCounterPtr counter) override { rows_before_aggregation_counter.swap(counter); } /// Notify about progress. Method could be called from different threads. /// Passed value are delta, that must be summarized. - virtual void onProgress(const Progress & /*progress*/) {} + virtual void onProgress(const Progress & /*progress*/) { } /// Content-Type to set when sending HTTP response. virtual std::string getContentType() const { return "text/plain; charset=UTF-8"; } @@ -151,6 +157,8 @@ protected: Progress progress; bool applied_limit = false; size_t rows_before_limit = 0; + bool applied_aggregation = false; + size_t rows_before_aggregation = 0; Chunk totals; Chunk extremes; }; @@ -184,7 +192,8 @@ protected: bool need_write_prefix = true; bool need_write_suffix = true; - RowsBeforeLimitCounterPtr rows_before_limit_counter; + RowsBeforeStepCounterPtr rows_before_limit_counter; + RowsBeforeStepCounterPtr rows_before_aggregation_counter; Statistics statistics; private: diff --git a/src/Processors/Formats/Impl/ArrowBlockInputFormat.h b/src/Processors/Formats/Impl/ArrowBlockInputFormat.h index cdbc5e57e4e..cb74a9dd93e 100644 --- a/src/Processors/Formats/Impl/ArrowBlockInputFormat.h +++ b/src/Processors/Formats/Impl/ArrowBlockInputFormat.h @@ -32,7 +32,7 @@ public: private: Chunk read() override; - void onCancel() override + void onCancel() noexcept override { is_stopped = 1; } diff --git a/src/Processors/Formats/Impl/ArrowColumnToCHColumn.cpp b/src/Processors/Formats/Impl/ArrowColumnToCHColumn.cpp index ed91913de4d..fb56fdd4fe0 100644 --- a/src/Processors/Formats/Impl/ArrowColumnToCHColumn.cpp +++ b/src/Processors/Formats/Impl/ArrowColumnToCHColumn.cpp @@ -133,16 +133,31 @@ static ColumnWithTypeAndName readColumnWithStringData(const std::shared_ptr buffer = chunk.value_data(); const size_t chunk_length = chunk.length(); - for (size_t offset_i = 0; offset_i != chunk_length; ++offset_i) + const size_t null_count = chunk.null_count(); + if (null_count == 0) { - if (!chunk.IsNull(offset_i) && buffer) + for (size_t offset_i = 0; offset_i != chunk_length; ++offset_i) { const auto * raw_data = buffer->data() + chunk.value_offset(offset_i); column_chars_t.insert_assume_reserved(raw_data, raw_data + chunk.value_length(offset_i)); - } - column_chars_t.emplace_back('\0'); + column_chars_t.emplace_back('\0'); - column_offsets.emplace_back(column_chars_t.size()); + column_offsets.emplace_back(column_chars_t.size()); + } + } + else + { + for (size_t offset_i = 0; offset_i != chunk_length; ++offset_i) + { + if (!chunk.IsNull(offset_i) && buffer) + { + const auto * raw_data = buffer->data() + chunk.value_offset(offset_i); + column_chars_t.insert_assume_reserved(raw_data, raw_data + chunk.value_length(offset_i)); + } + column_chars_t.emplace_back('\0'); + + column_offsets.emplace_back(column_chars_t.size()); + } } } return {std::move(internal_column), std::move(internal_type), column_name}; diff --git a/src/Processors/Formats/Impl/CHColumnToArrowColumn.cpp b/src/Processors/Formats/Impl/CHColumnToArrowColumn.cpp index 58bf4c1a2fc..30301b242db 100644 --- a/src/Processors/Formats/Impl/CHColumnToArrowColumn.cpp +++ b/src/Processors/Formats/Impl/CHColumnToArrowColumn.cpp @@ -185,7 +185,7 @@ namespace DB } else { - auto value = static_cast(column[value_i].get>().getValue()); + auto value = static_cast(column[value_i].safeGet>().getValue()); if (need_rescale) { if (common::mulOverflow(value, rescale_multiplier, value)) diff --git a/src/Processors/Formats/Impl/ConstantExpressionTemplate.cpp b/src/Processors/Formats/Impl/ConstantExpressionTemplate.cpp index 06e8668cd7c..566a036d79c 100644 --- a/src/Processors/Formats/Impl/ConstantExpressionTemplate.cpp +++ b/src/Processors/Formats/Impl/ConstantExpressionTemplate.cpp @@ -208,20 +208,20 @@ private: /// Do not replace empty array and array of NULLs if (literal->value.getType() == Field::Types::Array) { - const Array & array = literal->value.get(); + const Array & array = literal->value.safeGet(); auto not_null = std::find_if_not(array.begin(), array.end(), [](const auto & elem) { return elem.isNull(); }); if (not_null == array.end()) return true; } else if (literal->value.getType() == Field::Types::Map) { - const Map & map = literal->value.get(); + const Map & map = literal->value.safeGet(); if (map.size() % 2) return false; } else if (literal->value.getType() == Field::Types::Tuple) { - const Tuple & tuple = literal->value.get(); + const Tuple & tuple = literal->value.safeGet(); for (const auto & value : tuple) if (value.isNull()) diff --git a/src/Processors/Formats/Impl/DWARFBlockInputFormat.h b/src/Processors/Formats/Impl/DWARFBlockInputFormat.h index d8f5fc3d896..2d94d166708 100644 --- a/src/Processors/Formats/Impl/DWARFBlockInputFormat.h +++ b/src/Processors/Formats/Impl/DWARFBlockInputFormat.h @@ -32,7 +32,7 @@ public: protected: Chunk read() override; - void onCancel() override + void onCancel() noexcept override { is_stopped = 1; } diff --git a/src/Processors/Formats/Impl/JSONColumnsWithMetadataBlockOutputFormat.cpp b/src/Processors/Formats/Impl/JSONColumnsWithMetadataBlockOutputFormat.cpp index 1e8f57aa9a6..2f285e3d202 100644 --- a/src/Processors/Formats/Impl/JSONColumnsWithMetadataBlockOutputFormat.cpp +++ b/src/Processors/Formats/Impl/JSONColumnsWithMetadataBlockOutputFormat.cpp @@ -81,6 +81,8 @@ void JSONColumnsWithMetadataBlockOutputFormat::finalizeImpl() rows, statistics.rows_before_limit, statistics.applied_limit, + statistics.rows_before_aggregation, + statistics.applied_aggregation, statistics.watch, statistics.progress, format_settings.write_statistics, diff --git a/src/Processors/Formats/Impl/JSONColumnsWithMetadataBlockOutputFormat.h b/src/Processors/Formats/Impl/JSONColumnsWithMetadataBlockOutputFormat.h index c72b4d87234..e5208440483 100644 --- a/src/Processors/Formats/Impl/JSONColumnsWithMetadataBlockOutputFormat.h +++ b/src/Processors/Formats/Impl/JSONColumnsWithMetadataBlockOutputFormat.h @@ -44,6 +44,11 @@ public: String getName() const override { return "JSONCompactColumnsBlockOutputFormat"; } void setRowsBeforeLimit(size_t rows_before_limit_) override { statistics.rows_before_limit = rows_before_limit_; statistics.applied_limit = true; } + void setRowsBeforeAggregation(size_t rows_before_aggregation_) override + { + statistics.rows_before_aggregation = rows_before_aggregation_; + statistics.applied_aggregation = true; + } void onProgress(const Progress & progress_) override { statistics.progress.incrementPiecewiseAtomically(progress_); } protected: diff --git a/src/Processors/Formats/Impl/JSONEachRowRowInputFormat.cpp b/src/Processors/Formats/Impl/JSONEachRowRowInputFormat.cpp index 0e3972e8c45..8855a1bc28d 100644 --- a/src/Processors/Formats/Impl/JSONEachRowRowInputFormat.cpp +++ b/src/Processors/Formats/Impl/JSONEachRowRowInputFormat.cpp @@ -46,15 +46,6 @@ 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) @@ -180,15 +171,7 @@ void JSONEachRowRowInputFormat::readJSONObject(MutableColumns & columns) skipUnknownField(name_ref); continue; } - 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); + const size_t 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 8a1cef8fa9f..d97aa2dad8d 100644 --- a/src/Processors/Formats/Impl/JSONEachRowRowInputFormat.h +++ b/src/Processors/Formats/Impl/JSONEachRowRowInputFormat.h @@ -55,13 +55,7 @@ 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 @@ -80,8 +74,7 @@ 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/JSONRowOutputFormat.cpp b/src/Processors/Formats/Impl/JSONRowOutputFormat.cpp index 20182d84917..fec24b10c11 100644 --- a/src/Processors/Formats/Impl/JSONRowOutputFormat.cpp +++ b/src/Processors/Formats/Impl/JSONRowOutputFormat.cpp @@ -116,6 +116,8 @@ void JSONRowOutputFormat::finalizeImpl() row_count, statistics.rows_before_limit, statistics.applied_limit, + statistics.rows_before_aggregation, + statistics.applied_aggregation, statistics.watch, statistics.progress, settings.write_statistics && exception_message.empty(), diff --git a/src/Processors/Formats/Impl/JSONRowOutputFormat.h b/src/Processors/Formats/Impl/JSONRowOutputFormat.h index a38cd0e8db9..c36adb5ee3e 100644 --- a/src/Processors/Formats/Impl/JSONRowOutputFormat.h +++ b/src/Processors/Formats/Impl/JSONRowOutputFormat.h @@ -35,6 +35,11 @@ public: statistics.applied_limit = true; statistics.rows_before_limit = rows_before_limit_; } + void setRowsBeforeAggregation(size_t rows_before_aggregation_) override + { + statistics.applied_aggregation = true; + statistics.rows_before_aggregation = rows_before_aggregation_; + } protected: void writeField(const IColumn & column, const ISerialization & serialization, size_t row_num) override; diff --git a/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp b/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp index 649721f28bf..e68286bfcc5 100644 --- a/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp +++ b/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp @@ -262,7 +262,7 @@ convertFieldToORCLiteral(const orc::Type & orc_type, const Field & field, DataTy { case orc::BOOLEAN: { /// May throw exception - auto val = field.get(); + auto val = field.safeGet(); return orc::Literal(val != 0); } case orc::BYTE: @@ -275,7 +275,7 @@ convertFieldToORCLiteral(const orc::Type & orc_type, const Field & field, DataTy /// SELECT * FROM file('t.orc', ORC, 'x UInt8') WHERE x > 10 /// We have to reject this, otherwise it would miss values > 127 (because /// they're treated as negative by ORC). - auto val = field.get(); + auto val = field.safeGet(); return orc::Literal(val); } case orc::FLOAT: @@ -1143,24 +1143,42 @@ readColumnWithStringData(const orc::ColumnVectorBatch * orc_column, const orc::T reserver_size += 1; } - column_chars_t.reserve(reserver_size); - column_offsets.reserve(orc_str_column->numElements); + column_chars_t.resize_exact(reserver_size); + column_offsets.resize_exact(orc_str_column->numElements); size_t curr_offset = 0; - for (size_t i = 0; i < orc_str_column->numElements; ++i) + if (!orc_str_column->hasNulls) { - if (!orc_str_column->hasNulls || orc_str_column->notNull[i]) + for (size_t i = 0; i < orc_str_column->numElements; ++i) { const auto * buf = orc_str_column->data[i]; size_t buf_size = orc_str_column->length[i]; - column_chars_t.insert_assume_reserved(buf, buf + buf_size); + memcpy(&column_chars_t[curr_offset], buf, buf_size); curr_offset += buf_size; + + column_chars_t[curr_offset] = 0; + ++curr_offset; + + column_offsets[i] = curr_offset; } + } + else + { + for (size_t i = 0; i < orc_str_column->numElements; ++i) + { + if (orc_str_column->notNull[i]) + { + const auto * buf = orc_str_column->data[i]; + size_t buf_size = orc_str_column->length[i]; + memcpy(&column_chars_t[curr_offset], buf, buf_size); + curr_offset += buf_size; + } - column_chars_t.push_back(0); - ++curr_offset; + column_chars_t[curr_offset] = 0; + ++curr_offset; - column_offsets.push_back(curr_offset); + column_offsets[i] = curr_offset; + } } return {std::move(internal_column), std::move(internal_type), column_name}; } diff --git a/src/Processors/Formats/Impl/NativeORCBlockInputFormat.h b/src/Processors/Formats/Impl/NativeORCBlockInputFormat.h index a3ef9ed4b8f..e4f2ef9ebe3 100644 --- a/src/Processors/Formats/Impl/NativeORCBlockInputFormat.h +++ b/src/Processors/Formats/Impl/NativeORCBlockInputFormat.h @@ -64,7 +64,7 @@ public: protected: Chunk read() override; - void onCancel() override { is_stopped = 1; } + void onCancel() noexcept override { is_stopped = 1; } private: void prepareFileReader(); diff --git a/src/Processors/Formats/Impl/ORCBlockInputFormat.h b/src/Processors/Formats/Impl/ORCBlockInputFormat.h index 34630345849..85f1636d3dc 100644 --- a/src/Processors/Formats/Impl/ORCBlockInputFormat.h +++ b/src/Processors/Formats/Impl/ORCBlockInputFormat.h @@ -34,7 +34,7 @@ public: protected: Chunk read() override; - void onCancel() override + void onCancel() noexcept override { is_stopped = 1; } diff --git a/src/Processors/Formats/Impl/ORCBlockOutputFormat.cpp b/src/Processors/Formats/Impl/ORCBlockOutputFormat.cpp index 6f543a05fba..4a7a23158ff 100644 --- a/src/Processors/Formats/Impl/ORCBlockOutputFormat.cpp +++ b/src/Processors/Formats/Impl/ORCBlockOutputFormat.cpp @@ -12,6 +12,7 @@ #include #include #include +#include #include #include @@ -203,25 +204,15 @@ template void ORCBlockOutputFormat::writeNumbers( orc::ColumnVectorBatch & orc_column, const IColumn & column, - const PaddedPODArray * null_bytemap, + const PaddedPODArray * /*null_bytemap*/, ConvertFunc convert) { NumberVectorBatch & number_orc_column = dynamic_cast(orc_column); const auto & number_column = assert_cast &>(column); - number_orc_column.resize(number_column.size()); + number_orc_column.data.resize(number_column.size()); for (size_t i = 0; i != number_column.size(); ++i) - { - if (null_bytemap && (*null_bytemap)[i]) - { - number_orc_column.notNull[i] = 0; - continue; - } - - number_orc_column.notNull[i] = 1; number_orc_column.data[i] = convert(number_column.getElement(i)); - } - number_orc_column.numElements = number_column.size(); } template @@ -229,7 +220,7 @@ void ORCBlockOutputFormat::writeDecimals( orc::ColumnVectorBatch & orc_column, const IColumn & column, DataTypePtr & type, - const PaddedPODArray * null_bytemap, + const PaddedPODArray * /*null_bytemap*/, ConvertFunc convert) { DecimalVectorBatch & decimal_orc_column = dynamic_cast(orc_column); @@ -238,71 +229,49 @@ void ORCBlockOutputFormat::writeDecimals( decimal_orc_column.precision = decimal_type->getPrecision(); decimal_orc_column.scale = decimal_type->getScale(); decimal_orc_column.resize(decimal_column.size()); - for (size_t i = 0; i != decimal_column.size(); ++i) - { - if (null_bytemap && (*null_bytemap)[i]) - { - decimal_orc_column.notNull[i] = 0; - continue; - } - decimal_orc_column.notNull[i] = 1; + decimal_orc_column.values.resize(decimal_column.size()); + for (size_t i = 0; i != decimal_column.size(); ++i) decimal_orc_column.values[i] = convert(decimal_column.getElement(i).value); - } - decimal_orc_column.numElements = decimal_column.size(); } template void ORCBlockOutputFormat::writeStrings( orc::ColumnVectorBatch & orc_column, const IColumn & column, - const PaddedPODArray * null_bytemap) + const PaddedPODArray * /*null_bytemap*/) { orc::StringVectorBatch & string_orc_column = dynamic_cast(orc_column); const auto & string_column = assert_cast(column); - string_orc_column.resize(string_column.size()); + string_orc_column.data.resize(string_column.size()); + string_orc_column.length.resize(string_column.size()); for (size_t i = 0; i != string_column.size(); ++i) { - if (null_bytemap && (*null_bytemap)[i]) - { - string_orc_column.notNull[i] = 0; - continue; - } - - string_orc_column.notNull[i] = 1; const std::string_view & string = string_column.getDataAt(i).toView(); string_orc_column.data[i] = const_cast(string.data()); string_orc_column.length[i] = string.size(); } - string_orc_column.numElements = string_column.size(); } template void ORCBlockOutputFormat::writeDateTimes( orc::ColumnVectorBatch & orc_column, const IColumn & column, - const PaddedPODArray * null_bytemap, + const PaddedPODArray * /*null_bytemap*/, GetSecondsFunc get_seconds, GetNanosecondsFunc get_nanoseconds) { orc::TimestampVectorBatch & timestamp_orc_column = dynamic_cast(orc_column); const auto & timestamp_column = assert_cast(column); - timestamp_orc_column.resize(timestamp_column.size()); + timestamp_orc_column.data.resize(timestamp_column.size()); + timestamp_orc_column.nanoseconds.resize(timestamp_column.size()); for (size_t i = 0; i != timestamp_column.size(); ++i) { - if (null_bytemap && (*null_bytemap)[i]) - { - timestamp_orc_column.notNull[i] = 0; - continue; - } - - timestamp_orc_column.notNull[i] = 1; timestamp_orc_column.data[i] = static_cast(get_seconds(timestamp_column.getElement(i))); timestamp_orc_column.nanoseconds[i] = static_cast(get_nanoseconds(timestamp_column.getElement(i))); } - timestamp_orc_column.numElements = timestamp_column.size(); } void ORCBlockOutputFormat::writeColumn( @@ -311,9 +280,27 @@ void ORCBlockOutputFormat::writeColumn( DataTypePtr & type, const PaddedPODArray * null_bytemap) { - orc_column.notNull.resize(column.size()); + size_t rows = column.size(); + orc_column.resize(rows); + orc_column.numElements = rows; + + /// Calculate orc_column.hasNulls if (null_bytemap) - orc_column.hasNulls = true; + orc_column.hasNulls = !memoryIsZero(null_bytemap->data(), 0, null_bytemap->size()); + else + orc_column.hasNulls = false; + + /// Fill orc_column.notNull + if (orc_column.hasNulls) + { + for (size_t i = 0; i < rows; ++i) + orc_column.notNull[i] = !(*null_bytemap)[i]; + } + else + { + for (size_t i = 0; i < rows; ++i) + orc_column.notNull[i] = 1; + } /// ORC doesn't have unsigned types, so cast everything to signed and sign-extend to Int64 to /// make the ORC library calculate min and max correctly. @@ -471,6 +458,7 @@ void ORCBlockOutputFormat::writeColumn( } case TypeIndex::Nullable: { + chassert(!null_bytemap); const auto & nullable_column = assert_cast(column); const PaddedPODArray & new_null_bytemap = assert_cast &>(*nullable_column.getNullMapColumnPtr()).getData(); auto nested_type = removeNullable(type); @@ -485,19 +473,15 @@ void ORCBlockOutputFormat::writeColumn( const ColumnArray::Offsets & offsets = list_column.getOffsets(); size_t column_size = list_column.size(); - list_orc_column.resize(column_size); + list_orc_column.offsets.resize(column_size + 1); /// The length of list i in ListVectorBatch is offsets[i+1] - offsets[i]. list_orc_column.offsets[0] = 0; for (size_t i = 0; i != column_size; ++i) - { list_orc_column.offsets[i + 1] = offsets[i]; - list_orc_column.notNull[i] = 1; - } orc::ColumnVectorBatch & nested_orc_column = *list_orc_column.elements; - writeColumn(nested_orc_column, list_column.getData(), nested_type, null_bytemap); - list_orc_column.numElements = column_size; + writeColumn(nested_orc_column, list_column.getData(), nested_type, nullptr); break; } case TypeIndex::Tuple: @@ -505,10 +489,8 @@ void ORCBlockOutputFormat::writeColumn( orc::StructVectorBatch & struct_orc_column = dynamic_cast(orc_column); const auto & tuple_column = assert_cast(column); auto nested_types = assert_cast(type.get())->getElements(); - for (size_t i = 0; i != tuple_column.size(); ++i) - struct_orc_column.notNull[i] = 1; for (size_t i = 0; i != tuple_column.tupleSize(); ++i) - writeColumn(*struct_orc_column.fields[i], tuple_column.getColumn(i), nested_types[i], null_bytemap); + writeColumn(*struct_orc_column.fields[i], tuple_column.getColumn(i), nested_types[i], nullptr); break; } case TypeIndex::Map: @@ -520,25 +502,21 @@ void ORCBlockOutputFormat::writeColumn( size_t column_size = list_column.size(); - map_orc_column.resize(list_column.size()); + map_orc_column.offsets.resize(column_size + 1); /// The length of list i in ListVectorBatch is offsets[i+1] - offsets[i]. map_orc_column.offsets[0] = 0; for (size_t i = 0; i != column_size; ++i) - { map_orc_column.offsets[i + 1] = offsets[i]; - map_orc_column.notNull[i] = 1; - } + const auto nested_columns = assert_cast(list_column.getDataPtr().get())->getColumns(); orc::ColumnVectorBatch & keys_orc_column = *map_orc_column.keys; auto key_type = map_type.getKeyType(); - writeColumn(keys_orc_column, *nested_columns[0], key_type, null_bytemap); + writeColumn(keys_orc_column, *nested_columns[0], key_type, nullptr); orc::ColumnVectorBatch & values_orc_column = *map_orc_column.elements; auto value_type = map_type.getValueType(); - writeColumn(values_orc_column, *nested_columns[1], value_type, null_bytemap); - - map_orc_column.numElements = column_size; + writeColumn(values_orc_column, *nested_columns[1], value_type, nullptr); break; } default: @@ -546,27 +524,6 @@ void ORCBlockOutputFormat::writeColumn( } } -size_t ORCBlockOutputFormat::getColumnSize(const IColumn & column, DataTypePtr & type) -{ - if (type->getTypeId() == TypeIndex::Array) - { - auto nested_type = assert_cast(*type).getNestedType(); - const IColumn & nested_column = assert_cast(column).getData(); - return std::max(column.size(), getColumnSize(nested_column, nested_type)); - } - - return column.size(); -} - -size_t ORCBlockOutputFormat::getMaxColumnSize(Chunk & chunk) -{ - size_t columns_num = chunk.getNumColumns(); - size_t max_column_size = 0; - for (size_t i = 0; i != columns_num; ++i) - max_column_size = std::max(max_column_size, getColumnSize(*chunk.getColumns()[i], data_types[i])); - return max_column_size; -} - void ORCBlockOutputFormat::consume(Chunk chunk) { if (!writer) @@ -575,10 +532,7 @@ void ORCBlockOutputFormat::consume(Chunk chunk) size_t columns_num = chunk.getNumColumns(); size_t rows_num = chunk.getNumRows(); - /// getMaxColumnSize is needed to write arrays. - /// The size of the batch must be no less than total amount of array elements - /// and no less than the number of rows (ORC writes a null bit for every row). - std::unique_ptr batch = writer->createRowBatch(getMaxColumnSize(chunk)); + std::unique_ptr batch = writer->createRowBatch(chunk.getNumRows()); orc::StructVectorBatch & root = dynamic_cast(*batch); auto columns = chunk.detachColumns(); diff --git a/src/Processors/Formats/Impl/ORCBlockOutputFormat.h b/src/Processors/Formats/Impl/ORCBlockOutputFormat.h index 28837193d1a..06ecac9b820 100644 --- a/src/Processors/Formats/Impl/ORCBlockOutputFormat.h +++ b/src/Processors/Formats/Impl/ORCBlockOutputFormat.h @@ -69,11 +69,6 @@ private: void writeColumn(orc::ColumnVectorBatch & orc_column, const IColumn & column, DataTypePtr & type, const PaddedPODArray * null_bytemap); - /// These two functions are needed to know maximum nested size of arrays to - /// create an ORC Batch with the appropriate size - size_t getColumnSize(const IColumn & column, DataTypePtr & type); - size_t getMaxColumnSize(Chunk & chunk); - void prepareWriter(); const FormatSettings format_settings; diff --git a/src/Processors/Formats/Impl/ParallelFormattingOutputFormat.cpp b/src/Processors/Formats/Impl/ParallelFormattingOutputFormat.cpp index b2871310be5..5d404d493a6 100644 --- a/src/Processors/Formats/Impl/ParallelFormattingOutputFormat.cpp +++ b/src/Processors/Formats/Impl/ParallelFormattingOutputFormat.cpp @@ -96,7 +96,7 @@ namespace DB } - void ParallelFormattingOutputFormat::finishAndWait() + void ParallelFormattingOutputFormat::finishAndWait() noexcept { emergency_stop = true; diff --git a/src/Processors/Formats/Impl/ParallelFormattingOutputFormat.h b/src/Processors/Formats/Impl/ParallelFormattingOutputFormat.h index 341141dd633..02c74742226 100644 --- a/src/Processors/Formats/Impl/ParallelFormattingOutputFormat.h +++ b/src/Processors/Formats/Impl/ParallelFormattingOutputFormat.h @@ -122,7 +122,7 @@ public: started_prefix = true; } - void onCancel() override + void onCancel() noexcept override { finishAndWait(); } @@ -268,7 +268,7 @@ private: bool collected_suffix = false; bool collected_finalize = false; - void finishAndWait(); + void finishAndWait() noexcept; void onBackgroundException() { @@ -313,6 +313,12 @@ private: statistics.rows_before_limit = rows_before_limit; statistics.applied_limit = true; } + void setRowsBeforeAggregation(size_t rows_before_aggregation) override + { + std::lock_guard lock(statistics_mutex); + statistics.rows_before_aggregation = rows_before_aggregation; + statistics.applied_aggregation = true; + } }; } diff --git a/src/Processors/Formats/Impl/ParallelParsingInputFormat.h b/src/Processors/Formats/Impl/ParallelParsingInputFormat.h index 963ccd88def..b97bf5213e6 100644 --- a/src/Processors/Formats/Impl/ParallelParsingInputFormat.h +++ b/src/Processors/Formats/Impl/ParallelParsingInputFormat.h @@ -137,7 +137,7 @@ private: Chunk read() final; - void onCancel() final + void onCancel() noexcept final { /* * The format parsers themselves are not being cancelled here, so we'll @@ -292,7 +292,7 @@ private: first_parser_finished.wait(); } - void finishAndWait() + void finishAndWait() noexcept { /// Defending concurrent segmentator thread join std::lock_guard finish_and_wait_lock(finish_and_wait_mutex); diff --git a/src/Processors/Formats/Impl/ParquetBlockInputFormat.h b/src/Processors/Formats/Impl/ParquetBlockInputFormat.h index 24735ee4371..ed528cc077c 100644 --- a/src/Processors/Formats/Impl/ParquetBlockInputFormat.h +++ b/src/Processors/Formats/Impl/ParquetBlockInputFormat.h @@ -68,7 +68,7 @@ public: private: Chunk read() override; - void onCancel() override + void onCancel() noexcept override { is_stopped = 1; } diff --git a/src/Processors/Formats/Impl/ParquetBlockOutputFormat.cpp b/src/Processors/Formats/Impl/ParquetBlockOutputFormat.cpp index 9e499e2c400..645f16c6abe 100644 --- a/src/Processors/Formats/Impl/ParquetBlockOutputFormat.cpp +++ b/src/Processors/Formats/Impl/ParquetBlockOutputFormat.cpp @@ -270,7 +270,7 @@ void ParquetBlockOutputFormat::resetFormatterImpl() staging_bytes = 0; } -void ParquetBlockOutputFormat::onCancel() +void ParquetBlockOutputFormat::onCancel() noexcept { is_stopped = true; } diff --git a/src/Processors/Formats/Impl/ParquetBlockOutputFormat.h b/src/Processors/Formats/Impl/ParquetBlockOutputFormat.h index 422bae5c315..f8f5d2556a5 100644 --- a/src/Processors/Formats/Impl/ParquetBlockOutputFormat.h +++ b/src/Processors/Formats/Impl/ParquetBlockOutputFormat.h @@ -112,7 +112,7 @@ private: void consume(Chunk) override; void finalizeImpl() override; void resetFormatterImpl() override; - void onCancel() override; + void onCancel() noexcept override; void writeRowGroup(std::vector chunks); void writeUsingArrow(std::vector chunks); diff --git a/src/Processors/Formats/Impl/ParquetMetadataInputFormat.h b/src/Processors/Formats/Impl/ParquetMetadataInputFormat.h index ff63d78fa44..5d2d8989859 100644 --- a/src/Processors/Formats/Impl/ParquetMetadataInputFormat.h +++ b/src/Processors/Formats/Impl/ParquetMetadataInputFormat.h @@ -65,7 +65,7 @@ public: private: Chunk read() override; - void onCancel() override + void onCancel() noexcept override { is_stopped = 1; } diff --git a/src/Processors/Formats/Impl/PrometheusTextOutputFormat.cpp b/src/Processors/Formats/Impl/PrometheusTextOutputFormat.cpp index 3578401a0f8..b43c195f201 100644 --- a/src/Processors/Formats/Impl/PrometheusTextOutputFormat.cpp +++ b/src/Processors/Formats/Impl/PrometheusTextOutputFormat.cpp @@ -286,10 +286,10 @@ static void columnMapToContainer(const ColumnMap * col_map, size_t row_num, Cont { Field field; col_map->get(row_num, field); - const auto & map_field = field.get(); + const auto & map_field = field.safeGet(); for (const auto & map_element : map_field) { - const auto & map_entry = map_element.get(); + const auto & map_entry = map_element.safeGet(); String entry_key; String entry_value; diff --git a/src/Processors/Formats/Impl/TemplateBlockOutputFormat.cpp b/src/Processors/Formats/Impl/TemplateBlockOutputFormat.cpp index 1c43a0fa331..5d6db17aaa2 100644 --- a/src/Processors/Formats/Impl/TemplateBlockOutputFormat.cpp +++ b/src/Processors/Formats/Impl/TemplateBlockOutputFormat.cpp @@ -42,9 +42,11 @@ TemplateBlockOutputFormat::TemplateBlockOutputFormat(const Block & header_, Writ case static_cast(ResultsetPart::TimeElapsed): case static_cast(ResultsetPart::RowsRead): case static_cast(ResultsetPart::BytesRead): + case static_cast(ResultsetPart::RowsBeforeAggregation): if (format.escaping_rules[i] == EscapingRule::None) - format.throwInvalidFormat("Serialization type for output part rows, rows_before_limit, time, " - "rows_read or bytes_read is not specified", i); + format.throwInvalidFormat( + "Serialization type for output part rows, rows, time, " + "rows_read or bytes_read is not specified", i); break; default: format.throwInvalidFormat("Invalid output part", i); @@ -88,6 +90,8 @@ TemplateBlockOutputFormat::ResultsetPart TemplateBlockOutputFormat::stringToResu return ResultsetPart::RowsRead; else if (part == "bytes_read") return ResultsetPart::BytesRead; + else if (part == "rows_before_aggregation") + return ResultsetPart::RowsBeforeAggregation; else throw Exception(ErrorCodes::SYNTAX_ERROR, "Unknown output part {}", part); } @@ -173,6 +177,11 @@ void TemplateBlockOutputFormat::finalizeImpl() case ResultsetPart::BytesRead: writeValue(statistics.progress.read_bytes.load(), format.escaping_rules[i]); break; + case ResultsetPart::RowsBeforeAggregation: + if (!statistics.applied_aggregation) + format.throwInvalidFormat("Cannot print rows_before_aggregation for this request", i); + writeValue(statistics.rows_before_aggregation, format.escaping_rules[i]); + break; default: break; } diff --git a/src/Processors/Formats/Impl/TemplateBlockOutputFormat.h b/src/Processors/Formats/Impl/TemplateBlockOutputFormat.h index 53d98849482..5e88d79b4a8 100644 --- a/src/Processors/Formats/Impl/TemplateBlockOutputFormat.h +++ b/src/Processors/Formats/Impl/TemplateBlockOutputFormat.h @@ -21,6 +21,11 @@ public: String getName() const override { return "TemplateBlockOutputFormat"; } void setRowsBeforeLimit(size_t rows_before_limit_) override { statistics.rows_before_limit = rows_before_limit_; statistics.applied_limit = true; } + void setRowsBeforeAggregation(size_t rows_before_aggregation_) override + { + statistics.rows_before_aggregation = rows_before_aggregation_; + statistics.applied_aggregation = true; + } void onProgress(const Progress & progress_) override { statistics.progress.incrementPiecewiseAtomically(progress_); } enum class ResultsetPart : size_t @@ -33,7 +38,8 @@ public: RowsBeforeLimit, TimeElapsed, RowsRead, - BytesRead + BytesRead, + RowsBeforeAggregation }; static ResultsetPart stringToResultsetPart(const String & part); diff --git a/src/Processors/Formats/Impl/ValuesBlockInputFormat.cpp b/src/Processors/Formats/Impl/ValuesBlockInputFormat.cpp index de34a8aa04f..9839f64b947 100644 --- a/src/Processors/Formats/Impl/ValuesBlockInputFormat.cpp +++ b/src/Processors/Formats/Impl/ValuesBlockInputFormat.cpp @@ -333,7 +333,7 @@ namespace { const DataTypeTuple & type_tuple = static_cast(data_type); - Tuple & tuple_value = value.get(); + Tuple & tuple_value = value.safeGet(); size_t src_tuple_size = tuple_value.size(); size_t dst_tuple_size = type_tuple.getElements().size(); @@ -360,7 +360,7 @@ namespace if (element_type.isNullable()) return; - Array & array_value = value.get(); + Array & array_value = value.safeGet(); size_t array_value_size = array_value.size(); for (size_t i = 0; i < array_value_size; ++i) @@ -378,12 +378,12 @@ namespace const auto & key_type = *type_map.getKeyType(); const auto & value_type = *type_map.getValueType(); - auto & map = value.get(); + auto & map = value.safeGet(); size_t map_size = map.size(); for (size_t i = 0; i < map_size; ++i) { - auto & map_entry = map[i].get(); + auto & map_entry = map[i].safeGet(); auto & entry_key = map_entry[0]; auto & entry_value = map_entry[1]; diff --git a/src/Processors/Formats/Impl/XMLRowOutputFormat.cpp b/src/Processors/Formats/Impl/XMLRowOutputFormat.cpp index 52c161c3208..b19fcfd4a4a 100644 --- a/src/Processors/Formats/Impl/XMLRowOutputFormat.cpp +++ b/src/Processors/Formats/Impl/XMLRowOutputFormat.cpp @@ -191,6 +191,7 @@ void XMLRowOutputFormat::finalizeImpl() writeRowsBeforeLimitAtLeast(); + writeRowsBeforeAggregationAtLeast(); if (!exception_message.empty()) writeException(); @@ -219,6 +220,16 @@ void XMLRowOutputFormat::writeRowsBeforeLimitAtLeast() } } +void XMLRowOutputFormat::writeRowsBeforeAggregationAtLeast() +{ + if (statistics.applied_aggregation) + { + writeCString("\t", *ostr); + writeIntText(statistics.rows_before_aggregation, *ostr); + writeCString("\n", *ostr); + } +} + void XMLRowOutputFormat::writeStatistics() { writeCString("\t\n", *ostr); diff --git a/src/Processors/Formats/Impl/XMLRowOutputFormat.h b/src/Processors/Formats/Impl/XMLRowOutputFormat.h index daf03539d0b..792acd118c8 100644 --- a/src/Processors/Formats/Impl/XMLRowOutputFormat.h +++ b/src/Processors/Formats/Impl/XMLRowOutputFormat.h @@ -48,6 +48,11 @@ private: statistics.rows_before_limit = rows_before_limit_; } + void setRowsBeforeAggregation(size_t rows_before_aggregation_) override + { + statistics.applied_aggregation = true; + statistics.rows_before_aggregation = rows_before_aggregation_; + } void onRowsReadBeforeUpdate() override { row_count = getRowsReadBefore(); } void onProgress(const Progress & value) override; @@ -56,6 +61,7 @@ private: void writeExtremesElement(const char * title, const Columns & columns, size_t row_num); void writeRowsBeforeLimitAtLeast(); + void writeRowsBeforeAggregationAtLeast(); void writeStatistics(); void writeException(); diff --git a/src/Processors/Formats/LazyOutputFormat.cpp b/src/Processors/Formats/LazyOutputFormat.cpp index 4f6b10dd068..dc099765870 100644 --- a/src/Processors/Formats/LazyOutputFormat.cpp +++ b/src/Processors/Formats/LazyOutputFormat.cpp @@ -45,4 +45,8 @@ void LazyOutputFormat::setRowsBeforeLimit(size_t rows_before_limit) info.setRowsBeforeLimit(rows_before_limit); } +void LazyOutputFormat::setRowsBeforeAggregation(size_t rows_before_aggregation) +{ + info.setRowsBeforeAggregation(rows_before_aggregation); +} } diff --git a/src/Processors/Formats/LazyOutputFormat.h b/src/Processors/Formats/LazyOutputFormat.h index 9cf609ed2d7..5acb6cf3bf3 100644 --- a/src/Processors/Formats/LazyOutputFormat.h +++ b/src/Processors/Formats/LazyOutputFormat.h @@ -28,8 +28,9 @@ public: ProfileInfo & getProfileInfo() { return info; } void setRowsBeforeLimit(size_t rows_before_limit) override; + void setRowsBeforeAggregation(size_t rows_before_aggregation) override; - void onCancel() override + void onCancel() noexcept override { queue.clearAndFinish(); } diff --git a/src/Processors/Formats/PullingOutputFormat.cpp b/src/Processors/Formats/PullingOutputFormat.cpp index b2378e62d34..37050fb9675 100644 --- a/src/Processors/Formats/PullingOutputFormat.cpp +++ b/src/Processors/Formats/PullingOutputFormat.cpp @@ -42,5 +42,8 @@ void PullingOutputFormat::setRowsBeforeLimit(size_t rows_before_limit) { info.setRowsBeforeLimit(rows_before_limit); } - +void PullingOutputFormat::setRowsBeforeAggregation(size_t rows_before_aggregation) +{ + info.setRowsBeforeAggregation(rows_before_aggregation); +} } diff --git a/src/Processors/Formats/PullingOutputFormat.h b/src/Processors/Formats/PullingOutputFormat.h index a8efb8dd962..f2546cca180 100644 --- a/src/Processors/Formats/PullingOutputFormat.h +++ b/src/Processors/Formats/PullingOutputFormat.h @@ -22,6 +22,7 @@ public: ProfileInfo & getProfileInfo() { return info; } void setRowsBeforeLimit(size_t rows_before_limit) override; + void setRowsBeforeAggregation(size_t rows_before_aggregation) override; bool expectMaterializedColumns() const override { return false; } diff --git a/src/Processors/IProcessor.cpp b/src/Processors/IProcessor.cpp index f403aca2280..edb4d662d8b 100644 --- a/src/Processors/IProcessor.cpp +++ b/src/Processors/IProcessor.cpp @@ -9,7 +9,7 @@ namespace DB { -void IProcessor::cancel() +void IProcessor::cancel() noexcept { bool already_cancelled = is_cancelled.exchange(true, std::memory_order_acq_rel); diff --git a/src/Processors/IProcessor.h b/src/Processors/IProcessor.h index 4fd00d5e164..f1ce044d92f 100644 --- a/src/Processors/IProcessor.h +++ b/src/Processors/IProcessor.h @@ -21,8 +21,8 @@ class IQueryPlanStep; struct StorageLimits; using StorageLimitsList = std::list; -class RowsBeforeLimitCounter; -using RowsBeforeLimitCounterPtr = std::shared_ptr; +class RowsBeforeStepCounter; +using RowsBeforeStepCounterPtr = std::shared_ptr; class IProcessor; using ProcessorPtr = std::shared_ptr; @@ -255,7 +255,7 @@ public: /// In case if query was cancelled executor will wait till all processors finish their jobs. /// Generally, there is no reason to check this flag. However, it may be reasonable for long operations (e.g. i/o). bool isCancelled() const { return is_cancelled.load(std::memory_order_acquire); } - void cancel(); + void cancel() noexcept; /// Additional method which is called in case if ports were updated while work() method. /// May be used to stop execution in rare cases. @@ -377,10 +377,14 @@ public: /// Set rows_before_limit counter for current processor. /// This counter is used to calculate the number of rows right before any filtration of LimitTransform. - virtual void setRowsBeforeLimitCounter(RowsBeforeLimitCounterPtr /* counter */) {} + virtual void setRowsBeforeLimitCounter(RowsBeforeStepCounterPtr /* counter */) { } + + /// Set rows_before_aggregation counter for current processor. + /// This counter is used to calculate the number of rows right before AggregatingTransform. + virtual void setRowsBeforeAggregationCounter(RowsBeforeStepCounterPtr /* counter */) { } protected: - virtual void onCancel() {} + virtual void onCancel() noexcept {} std::atomic is_cancelled{false}; diff --git a/src/Processors/LimitTransform.h b/src/Processors/LimitTransform.h index 33ff968985f..45ae5b0ce81 100644 --- a/src/Processors/LimitTransform.h +++ b/src/Processors/LimitTransform.h @@ -1,8 +1,8 @@ #pragma once -#include -#include #include +#include +#include namespace DB { @@ -30,7 +30,7 @@ private: std::vector sort_column_positions; UInt64 rows_read = 0; /// including the last read block - RowsBeforeLimitCounterPtr rows_before_limit_at_least; + RowsBeforeStepCounterPtr rows_before_limit_at_least; /// State of port's pair. /// Chunks from different port pairs are not mixed for better cache locality. @@ -71,7 +71,7 @@ public: InputPort & getInputPort() { return inputs.front(); } OutputPort & getOutputPort() { return outputs.front(); } - void setRowsBeforeLimitCounter(RowsBeforeLimitCounterPtr counter) override { rows_before_limit_at_least.swap(counter); } + void setRowsBeforeLimitCounter(RowsBeforeStepCounterPtr counter) override { rows_before_limit_at_least.swap(counter); } void setInputPortHasCounter(size_t pos) { ports_data[pos].input_port_has_counter = true; } }; diff --git a/src/Processors/Merges/AggregatingSortedTransform.h b/src/Processors/Merges/AggregatingSortedTransform.h index c6d7e844c65..c96ad3db525 100644 --- a/src/Processors/Merges/AggregatingSortedTransform.h +++ b/src/Processors/Merges/AggregatingSortedTransform.h @@ -3,6 +3,11 @@ #include #include +namespace ProfileEvents +{ + extern const Event AggregatingSortedMilliseconds; +} + namespace DB { @@ -29,6 +34,11 @@ public: } String getName() const override { return "AggregatingSortedTransform"; } + + void onFinish() override + { + logMergedStats(ProfileEvents::AggregatingSortedMilliseconds, "Aggregated sorted", getLogger("AggregatingSortedTransform")); + } }; } diff --git a/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.h b/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.h index 53c103e7038..908994e1851 100644 --- a/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.h +++ b/src/Processors/Merges/Algorithms/AggregatingSortedAlgorithm.h @@ -30,6 +30,8 @@ public: void consume(Input & input, size_t source_num) override; Status merge() override; + MergedStats getMergedStats() const override { return merged_data.getMergedStats(); } + /// Stores information for aggregation of SimpleAggregateFunction columns struct SimpleAggregateDescription { diff --git a/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.cpp b/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.cpp index 86675bcb237..477566d8a94 100644 --- a/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.cpp +++ b/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.cpp @@ -126,6 +126,9 @@ IMergingAlgorithm::Status FinishAggregatingInOrderAlgorithm::merge() Chunk FinishAggregatingInOrderAlgorithm::prepareToMerge() { + total_merged_rows += accumulated_rows; + total_merged_bytes += accumulated_bytes; + accumulated_rows = 0; accumulated_bytes = 0; diff --git a/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.h b/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.h index cc6578e79be..c34028b1cba 100644 --- a/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.h +++ b/src/Processors/Merges/Algorithms/FinishAggregatingInOrderAlgorithm.h @@ -50,6 +50,8 @@ public: void consume(Input & input, size_t source_num) override; Status merge() override; + MergedStats getMergedStats() const override { return {.bytes = accumulated_bytes, .rows = accumulated_rows, .blocks = chunk_num}; } + private: Chunk prepareToMerge(); void addToAggregation(); @@ -92,6 +94,9 @@ private: UInt64 chunk_num = 0; size_t accumulated_rows = 0; size_t accumulated_bytes = 0; + + size_t total_merged_rows = 0; + size_t total_merged_bytes = 0; }; } diff --git a/src/Processors/Merges/Algorithms/GraphiteRollupSortedAlgorithm.h b/src/Processors/Merges/Algorithms/GraphiteRollupSortedAlgorithm.h index aaa3859efb6..cb2775c968d 100644 --- a/src/Processors/Merges/Algorithms/GraphiteRollupSortedAlgorithm.h +++ b/src/Processors/Merges/Algorithms/GraphiteRollupSortedAlgorithm.h @@ -33,6 +33,8 @@ public: const char * getName() const override { return "GraphiteRollupSortedAlgorithm"; } Status merge() override; + MergedStats getMergedStats() const override { return merged_data->getMergedStats(); } + struct ColumnsDefinition { size_t path_column_num; diff --git a/src/Processors/Merges/Algorithms/IMergingAlgorithm.h b/src/Processors/Merges/Algorithms/IMergingAlgorithm.h index 9a1c7c24270..83f11232b71 100644 --- a/src/Processors/Merges/Algorithms/IMergingAlgorithm.h +++ b/src/Processors/Merges/Algorithms/IMergingAlgorithm.h @@ -1,7 +1,7 @@ #pragma once #include -#include +#include namespace DB { @@ -65,6 +65,15 @@ public: IMergingAlgorithm() = default; virtual ~IMergingAlgorithm() = default; + + struct MergedStats + { + UInt64 bytes = 0; + UInt64 rows = 0; + UInt64 blocks = 0; + }; + + virtual MergedStats getMergedStats() const = 0; }; // TODO: use when compile with clang which could support it diff --git a/src/Processors/Merges/Algorithms/IMergingAlgorithmWithSharedChunks.h b/src/Processors/Merges/Algorithms/IMergingAlgorithmWithSharedChunks.h index bc1aafe93f7..1725108ac5d 100644 --- a/src/Processors/Merges/Algorithms/IMergingAlgorithmWithSharedChunks.h +++ b/src/Processors/Merges/Algorithms/IMergingAlgorithmWithSharedChunks.h @@ -16,6 +16,8 @@ public: void initialize(Inputs inputs) override; void consume(Input & input, size_t source_num) override; + MergedStats getMergedStats() const override { return merged_data->getMergedStats(); } + private: Block header; SortDescription description; diff --git a/src/Processors/Merges/Algorithms/MergedData.h b/src/Processors/Merges/Algorithms/MergedData.h index c5bb074bb0c..8f47f89d8ee 100644 --- a/src/Processors/Merges/Algorithms/MergedData.h +++ b/src/Processors/Merges/Algorithms/MergedData.h @@ -183,6 +183,8 @@ public: UInt64 totalAllocatedBytes() const { return total_allocated_bytes; } UInt64 maxBlockSize() const { return max_block_size; } + IMergingAlgorithm::MergedStats getMergedStats() const { return {.bytes = total_allocated_bytes, .rows = total_merged_rows, .blocks = total_chunks}; } + virtual ~MergedData() = default; protected: diff --git a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.h b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.h index bcb111baadf..c889668a38e 100644 --- a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.h +++ b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.h @@ -31,7 +31,7 @@ public: void consume(Input & input, size_t source_num) override; Status merge() override; - const MergedData & getMergedData() const { return merged_data; } + MergedStats getMergedStats() const override { return merged_data.getMergedStats(); } private: Block header; diff --git a/src/Processors/Merges/Algorithms/SummingSortedAlgorithm.cpp b/src/Processors/Merges/Algorithms/SummingSortedAlgorithm.cpp index e2c6371c44f..80c00f91d82 100644 --- a/src/Processors/Merges/Algorithms/SummingSortedAlgorithm.cpp +++ b/src/Processors/Merges/Algorithms/SummingSortedAlgorithm.cpp @@ -127,14 +127,14 @@ static bool mergeMap(const SummingSortedAlgorithm::MapDescription & desc, Row right(left.size()); for (size_t col_num : desc.key_col_nums) - right[col_num] = (*raw_columns[col_num])[row_number].template get(); + right[col_num] = (*raw_columns[col_num])[row_number].template safeGet(); for (size_t col_num : desc.val_col_nums) - right[col_num] = (*raw_columns[col_num])[row_number].template get(); + right[col_num] = (*raw_columns[col_num])[row_number].template safeGet(); auto at_ith_column_jth_row = [&](const Row & matrix, size_t i, size_t j) -> const Field & { - return matrix[i].get()[j]; + return matrix[i].safeGet()[j]; }; auto tuple_of_nth_columns_at_jth_row = [&](const Row & matrix, const ColumnNumbers & col_nums, size_t j) -> Array @@ -160,7 +160,7 @@ static bool mergeMap(const SummingSortedAlgorithm::MapDescription & desc, auto merge = [&](const Row & matrix) { - size_t rows = matrix[desc.key_col_nums[0]].get().size(); + size_t rows = matrix[desc.key_col_nums[0]].safeGet().size(); for (size_t j = 0; j < rows; ++j) { @@ -190,10 +190,10 @@ static bool mergeMap(const SummingSortedAlgorithm::MapDescription & desc, for (const auto & key_value : merged) { for (size_t col_num_index = 0, size = desc.key_col_nums.size(); col_num_index < size; ++col_num_index) - row[desc.key_col_nums[col_num_index]].get()[row_num] = key_value.first[col_num_index]; + row[desc.key_col_nums[col_num_index]].safeGet()[row_num] = key_value.first[col_num_index]; for (size_t col_num_index = 0, size = desc.val_col_nums.size(); col_num_index < size; ++col_num_index) - row[desc.val_col_nums[col_num_index]].get()[row_num] = key_value.second[col_num_index]; + row[desc.val_col_nums[col_num_index]].safeGet()[row_num] = key_value.second[col_num_index]; ++row_num; } diff --git a/src/Processors/Merges/Algorithms/SummingSortedAlgorithm.h b/src/Processors/Merges/Algorithms/SummingSortedAlgorithm.h index 664b171c4b9..74b4e397831 100644 --- a/src/Processors/Merges/Algorithms/SummingSortedAlgorithm.h +++ b/src/Processors/Merges/Algorithms/SummingSortedAlgorithm.h @@ -30,6 +30,8 @@ public: void consume(Input & input, size_t source_num) override; Status merge() override; + MergedStats getMergedStats() const override { return merged_data.getMergedStats(); } + struct AggregateDescription; struct MapDescription; diff --git a/src/Processors/Merges/CollapsingSortedTransform.h b/src/Processors/Merges/CollapsingSortedTransform.h index 4479ac82f66..99fb700abf1 100644 --- a/src/Processors/Merges/CollapsingSortedTransform.h +++ b/src/Processors/Merges/CollapsingSortedTransform.h @@ -3,6 +3,11 @@ #include #include +namespace ProfileEvents +{ + extern const Event CollapsingSortedMilliseconds; +} + namespace DB { @@ -36,6 +41,11 @@ public: } String getName() const override { return "CollapsingSortedTransform"; } + + void onFinish() override + { + logMergedStats(ProfileEvents::CollapsingSortedMilliseconds, "Collapsed sorted", getLogger("CollapsingSortedTransform")); + } }; } diff --git a/src/Processors/Merges/IMergingTransform.h b/src/Processors/Merges/IMergingTransform.h index be629271736..e5cd3bdde46 100644 --- a/src/Processors/Merges/IMergingTransform.h +++ b/src/Processors/Merges/IMergingTransform.h @@ -2,7 +2,10 @@ #include #include +#include #include +#include +#include namespace DB { @@ -110,6 +113,8 @@ public: void work() override { + Stopwatch watch{CLOCK_MONOTONIC_COARSE}; + if (!state.init_chunks.empty()) algorithm.initialize(std::move(state.init_chunks)); @@ -147,6 +152,8 @@ public: // std::cerr << "Finished" << std::endl; state.is_finished = true; } + + merging_elapsed_ns += watch.elapsedNanoseconds(); } protected: @@ -156,7 +163,33 @@ protected: Algorithm algorithm; /// Profile info. - Stopwatch total_stopwatch {CLOCK_MONOTONIC_COARSE}; + UInt64 merging_elapsed_ns = 0; + + void logMergedStats(ProfileEvents::Event elapsed_ms_event, std::string_view transform_message, LoggerPtr log) const + { + auto stats = algorithm.getMergedStats(); + + UInt64 elapsed_ms = merging_elapsed_ns / 1000000LL; + ProfileEvents::increment(elapsed_ms_event, elapsed_ms); + + /// Don't print info for small parts (< 1M rows) + if (stats.rows < 1000000) + return; + + double seconds = static_cast(merging_elapsed_ns) / 1000000000ULL; + + if (seconds == 0.0) + { + LOG_DEBUG(log, "{}, {} blocks, {} rows, {} bytes in 0 sec.", + transform_message, stats.blocks, stats.rows, stats.bytes); + } + else + { + LOG_DEBUG(log, "{}, {} blocks, {} rows, {} bytes in {} sec., {} rows/sec., {}/sec.", + transform_message, stats.blocks, stats.rows, stats.bytes, + seconds, stats.rows / seconds, ReadableSize(stats.bytes / seconds)); + } + } private: using IMergingTransformBase::state; diff --git a/src/Processors/Merges/MergingSortedTransform.cpp b/src/Processors/Merges/MergingSortedTransform.cpp index 338b1ff7935..d2895a2a2e9 100644 --- a/src/Processors/Merges/MergingSortedTransform.cpp +++ b/src/Processors/Merges/MergingSortedTransform.cpp @@ -1,9 +1,12 @@ #include #include #include - #include -#include + +namespace ProfileEvents +{ + extern const Event MergingSortedMilliseconds; +} namespace DB { @@ -18,7 +21,6 @@ MergingSortedTransform::MergingSortedTransform( UInt64 limit_, bool always_read_till_end_, WriteBuffer * out_row_sources_buf_, - bool quiet_, bool use_average_block_sizes, bool have_all_inputs_) : IMergingTransform( @@ -37,7 +39,6 @@ MergingSortedTransform::MergingSortedTransform( limit_, out_row_sources_buf_, use_average_block_sizes) - , quiet(quiet_) { } @@ -48,22 +49,7 @@ void MergingSortedTransform::onNewInput() void MergingSortedTransform::onFinish() { - if (quiet) - return; - - const auto & merged_data = algorithm.getMergedData(); - - auto log = getLogger("MergingSortedTransform"); - - double seconds = total_stopwatch.elapsedSeconds(); - - if (seconds == 0.0) - LOG_DEBUG(log, "Merge sorted {} blocks, {} rows in 0 sec.", merged_data.totalChunks(), merged_data.totalMergedRows()); - else - LOG_DEBUG(log, "Merge sorted {} blocks, {} rows in {} sec., {} rows/sec., {}/sec", - merged_data.totalChunks(), merged_data.totalMergedRows(), seconds, - merged_data.totalMergedRows() / seconds, - ReadableSize(merged_data.totalAllocatedBytes() / seconds)); + logMergedStats(ProfileEvents::MergingSortedMilliseconds, "Merged sorted", getLogger("MergingSortedTransform")); } } diff --git a/src/Processors/Merges/MergingSortedTransform.h b/src/Processors/Merges/MergingSortedTransform.h index 2b53939f309..6e52450efa7 100644 --- a/src/Processors/Merges/MergingSortedTransform.h +++ b/src/Processors/Merges/MergingSortedTransform.h @@ -21,7 +21,6 @@ public: UInt64 limit_ = 0, bool always_read_till_end_ = false, WriteBuffer * out_row_sources_buf_ = nullptr, - bool quiet_ = false, bool use_average_block_sizes = false, bool have_all_inputs_ = true); @@ -30,9 +29,6 @@ public: protected: void onNewInput() override; void onFinish() override; - -private: - bool quiet = false; }; } diff --git a/src/Processors/Merges/ReplacingSortedTransform.h b/src/Processors/Merges/ReplacingSortedTransform.h index 2657987f161..dc262aab9ee 100644 --- a/src/Processors/Merges/ReplacingSortedTransform.h +++ b/src/Processors/Merges/ReplacingSortedTransform.h @@ -3,6 +3,10 @@ #include #include +namespace ProfileEvents +{ + extern const Event ReplacingSortedMilliseconds; +} namespace DB { @@ -38,6 +42,11 @@ public: } String getName() const override { return "ReplacingSorted"; } + + void onFinish() override + { + logMergedStats(ProfileEvents::ReplacingSortedMilliseconds, "Replaced sorted", getLogger("ReplacingSortedTransform")); + } }; } diff --git a/src/Processors/Merges/SummingSortedTransform.h b/src/Processors/Merges/SummingSortedTransform.h index 70ddebfea95..d7c20223d7e 100644 --- a/src/Processors/Merges/SummingSortedTransform.h +++ b/src/Processors/Merges/SummingSortedTransform.h @@ -3,6 +3,11 @@ #include #include +namespace ProfileEvents +{ + extern const Event SummingSortedMilliseconds; +} + namespace DB { @@ -33,6 +38,11 @@ public: } String getName() const override { return "SummingSortedTransform"; } + + void onFinish() override + { + logMergedStats(ProfileEvents::SummingSortedMilliseconds, "Summed sorted", getLogger("SummingSortedTransform")); + } }; } diff --git a/src/Processors/Merges/VersionedCollapsingTransform.h b/src/Processors/Merges/VersionedCollapsingTransform.h index 18244469bd7..32b5d7bf343 100644 --- a/src/Processors/Merges/VersionedCollapsingTransform.h +++ b/src/Processors/Merges/VersionedCollapsingTransform.h @@ -3,6 +3,10 @@ #include #include +namespace ProfileEvents +{ + extern const Event VersionedCollapsingSortedMilliseconds; +} namespace DB { @@ -33,6 +37,11 @@ public: } String getName() const override { return "VersionedCollapsingTransform"; } + + void onFinish() override + { + logMergedStats(ProfileEvents::VersionedCollapsingSortedMilliseconds, "Versioned collapsed sorted", getLogger("VersionedCollapsingTransform")); + } }; } diff --git a/src/Processors/OffsetTransform.h b/src/Processors/OffsetTransform.h index 79a7d15fe0b..04486a6c940 100644 --- a/src/Processors/OffsetTransform.h +++ b/src/Processors/OffsetTransform.h @@ -1,8 +1,8 @@ #pragma once -#include -#include #include +#include +#include namespace DB { @@ -16,7 +16,7 @@ private: UInt64 offset; UInt64 rows_read = 0; /// including the last read block - RowsBeforeLimitCounterPtr rows_before_limit_at_least; + RowsBeforeStepCounterPtr rows_before_limit_at_least; /// State of port's pair. /// Chunks from different port pairs are not mixed for better cache locality. @@ -45,7 +45,7 @@ public: InputPort & getInputPort() { return inputs.front(); } OutputPort & getOutputPort() { return outputs.front(); } - void setRowsBeforeLimitCounter(RowsBeforeLimitCounterPtr counter) override { rows_before_limit_at_least.swap(counter); } + void setRowsBeforeLimitCounter(RowsBeforeStepCounterPtr counter) override { rows_before_limit_at_least.swap(counter); } }; } diff --git a/src/Processors/QueryPlan/Optimizations/filterPushDown.cpp b/src/Processors/QueryPlan/Optimizations/filterPushDown.cpp index 73314f005b6..b71326ff75b 100644 --- a/src/Processors/QueryPlan/Optimizations/filterPushDown.cpp +++ b/src/Processors/QueryPlan/Optimizations/filterPushDown.cpp @@ -442,6 +442,15 @@ size_t tryPushDownFilter(QueryPlan::Node * parent_node, QueryPlan::Nodes & nodes const auto & params = aggregating->getParams(); const auto & keys = params.keys; + /** The filter is applied either to aggregation keys or aggregation result + * (columns under aggregation is not available in outer scope, so we can't have a filter for them). + * The filter for the aggregation result is not pushed down, so the only valid case is filtering aggregation keys. + * In case keys are empty, do not push down the filter. + * Also with empty keys we can have an issue with `empty_result_for_aggregation_by_empty_set`, + * since we can gen a result row when everything is filtered. + */ + if (keys.empty()) + return 0; const bool filter_column_is_not_among_aggregation_keys = std::find(keys.begin(), keys.end(), filter->getFilterColumnName()) == keys.end(); diff --git a/src/Processors/QueryPlan/Optimizations/optimizeUseAggregateProjection.cpp b/src/Processors/QueryPlan/Optimizations/optimizeUseAggregateProjection.cpp index ad89cec5f79..b31ee7ea53c 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeUseAggregateProjection.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeUseAggregateProjection.cpp @@ -255,20 +255,13 @@ static void appendAggregateFunctions( const auto * node = input; - if (node->result_name != aggregate.column_name) - { - if (DataTypeAggregateFunction::strictEquals(type, node->result_type)) - { - node = &proj_dag.addAlias(*node, aggregate.column_name); - } - else - { - /// Cast to aggregate types specified in query if it's not - /// strictly the same as the one specified in projection. This - /// is required to generate correct results during finalization. - node = &proj_dag.addCast(*node, type, aggregate.column_name); - } - } + if (!DataTypeAggregateFunction::strictEquals(type, node->result_type)) + /// Cast to aggregate types specified in query if it's not + /// strictly the same as the one specified in projection. This + /// is required to generate correct results during finalization. + node = &proj_dag.addCast(*node, type, aggregate.column_name); + else if (node->result_name != aggregate.column_name) + node = &proj_dag.addAlias(*node, aggregate.column_name); proj_dag_outputs.push_back(node); } @@ -757,9 +750,7 @@ std::optional optimizeUseAggregateProjections(QueryPlan::Node & node, Qu else { auto storage_snapshot = reading->getStorageSnapshot(); - auto proj_snapshot = std::make_shared(storage_snapshot->storage, storage_snapshot->metadata); - proj_snapshot->addProjection(best_candidate->projection); - + auto proj_snapshot = std::make_shared(storage_snapshot->storage, best_candidate->projection->metadata); auto projection_query_info = query_info; projection_query_info.prewhere_info = nullptr; projection_query_info.filter_actions_dag = nullptr; diff --git a/src/Processors/QueryPlan/Optimizations/optimizeUseNormalProjection.cpp b/src/Processors/QueryPlan/Optimizations/optimizeUseNormalProjection.cpp index e9aee65675f..b15f913fc19 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeUseNormalProjection.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeUseNormalProjection.cpp @@ -193,9 +193,7 @@ std::optional optimizeUseNormalProjections(Stack & stack, QueryPlan::Nod } auto storage_snapshot = reading->getStorageSnapshot(); - auto proj_snapshot = std::make_shared(storage_snapshot->storage, storage_snapshot->metadata); - proj_snapshot->addProjection(best_candidate->projection); - + auto proj_snapshot = std::make_shared(storage_snapshot->storage, best_candidate->projection->metadata); auto query_info_copy = query_info; query_info_copy.prewhere_info = nullptr; diff --git a/src/Processors/QueryPlan/Optimizations/removeRedundantSorting.cpp b/src/Processors/QueryPlan/Optimizations/removeRedundantSorting.cpp index 7cac7bee6ec..f0094f0f8d2 100644 --- a/src/Processors/QueryPlan/Optimizations/removeRedundantSorting.cpp +++ b/src/Processors/QueryPlan/Optimizations/removeRedundantSorting.cpp @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include @@ -59,9 +60,10 @@ public: if (typeid_cast(current_step) || typeid_cast(current_step) /// (1) if there are LIMITs on top of ORDER BY, the ORDER BY is non-removable - || typeid_cast(current_step) /// (2) if ORDER BY is with FILL WITH, it is non-removable - || typeid_cast(current_step) /// (3) ORDER BY will change order of previous sorting - || typeid_cast(current_step)) /// (4) aggregation change order + || typeid_cast(current_step) /// (2) OFFSET on top of ORDER BY, the ORDER BY is non-removable + || typeid_cast(current_step) /// (3) if ORDER BY is with FILL WITH, it is non-removable + || typeid_cast(current_step) /// (4) ORDER BY will change order of previous sorting + || typeid_cast(current_step)) /// (5) aggregation change order { logStep("nodes_affect_order/push", current_node); nodes_affect_order.push_back(current_node); diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 51143a519a1..734e67bda24 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -24,8 +24,8 @@ #include #include #include -#include -#include +#include +#include #include #include #include @@ -52,6 +52,8 @@ #include #include +#include "config.h" + using namespace DB; namespace @@ -285,7 +287,6 @@ ReadFromMergeTree::ReadFromMergeTree( , all_column_names(std::move(all_column_names_)) , data(data_) , actions_settings(ExpressionActionsSettings::fromContext(context_)) - , metadata_for_reading(storage_snapshot->getMetadataForQuery()) , block_size{ .max_block_size_rows = max_block_size_, .preferred_block_size_bytes = context->getSettingsRef().preferred_block_size_bytes, @@ -327,7 +328,7 @@ ReadFromMergeTree::ReadFromMergeTree( updateSortDescriptionForOutputStream( *output_stream, - storage_snapshot->getMetadataForQuery()->getSortingKeyColumns(), + storage_snapshot->metadata->getSortingKeyColumns(), getSortDirection(), query_info.input_order_info, prewhere_info, @@ -351,7 +352,15 @@ Pipe ReadFromMergeTree::readFromPoolParallelReplicas( /// We have a special logic for local replica. It has to read less data, because in some cases it should /// merge states of aggregate functions or do some other important stuff other than reading from Disk. - const auto multiplier = context->getSettingsRef().parallel_replicas_single_task_marks_count_multiplier; + auto multiplier = context->getSettingsRef().parallel_replicas_single_task_marks_count_multiplier; + const auto min_marks_for_concurrent_read_limit = std::numeric_limits::max() >> 1; + if (pool_settings.min_marks_for_concurrent_read > min_marks_for_concurrent_read_limit) + { + /// limit min marks to read in case it's big, happened in test since due to settings randomzation + pool_settings.min_marks_for_concurrent_read = min_marks_for_concurrent_read_limit; + multiplier = 1.0f; + } + if (auto result = pool_settings.min_marks_for_concurrent_read * multiplier; canConvertTo(result)) pool_settings.min_marks_for_concurrent_read = static_cast(result); else @@ -520,7 +529,15 @@ Pipe ReadFromMergeTree::readInOrder( .number_of_current_replica = client_info.number_of_current_replica, }; - const auto multiplier = context->getSettingsRef().parallel_replicas_single_task_marks_count_multiplier; + auto multiplier = context->getSettingsRef().parallel_replicas_single_task_marks_count_multiplier; + const auto min_marks_for_concurrent_read_limit = std::numeric_limits::max() >> 1; + if (pool_settings.min_marks_for_concurrent_read > min_marks_for_concurrent_read_limit) + { + /// limit min marks to read in case it's big, happened in test since due to settings randomzation + pool_settings.min_marks_for_concurrent_read = min_marks_for_concurrent_read_limit; + multiplier = 1.0f; + } + if (auto result = pool_settings.min_marks_for_concurrent_read * multiplier; canConvertTo(result)) pool_settings.min_marks_for_concurrent_read = static_cast(result); else @@ -782,7 +799,7 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreams(RangesInDataParts && parts_ Names in_order_column_names_to_read(column_names); /// Add columns needed to calculate the sorting expression - for (const auto & column_name : metadata_for_reading->getColumnsRequiredForSortingKey()) + for (const auto & column_name : storage_snapshot->metadata->getColumnsRequiredForSortingKey()) { if (column_names_set.contains(column_name)) continue; @@ -802,10 +819,10 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreams(RangesInDataParts && parts_ info.use_uncompressed_cache); }; - auto sorting_expr = metadata_for_reading->getSortingKey().expression; + auto sorting_expr = storage_snapshot->metadata->getSortingKey().expression; SplitPartsWithRangesByPrimaryKeyResult split_ranges_result = splitPartsWithRangesByPrimaryKey( - metadata_for_reading->getPrimaryKey(), + storage_snapshot->metadata->getPrimaryKey(), std::move(sorting_expr), std::move(parts_with_ranges), num_streams, @@ -883,7 +900,7 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( if (prewhere_info) { NameSet sorting_columns; - for (const auto & column : metadata_for_reading->getSortingKey().expression->getRequiredColumnsWithTypes()) + for (const auto & column : storage_snapshot->metadata->getSortingKey().expression->getRequiredColumnsWithTypes()) sorting_columns.insert(column.name); have_input_columns_removed_after_prewhere = restorePrewhereInputs(*prewhere_info, sorting_columns); @@ -1038,12 +1055,12 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( if (need_preliminary_merge || output_each_partition_through_separate_port) { size_t prefix_size = input_order_info->used_prefix_of_sorting_key_size; - auto order_key_prefix_ast = metadata_for_reading->getSortingKey().expression_list_ast->clone(); + auto order_key_prefix_ast = storage_snapshot->metadata->getSortingKey().expression_list_ast->clone(); order_key_prefix_ast->children.resize(prefix_size); - auto syntax_result = TreeRewriter(context).analyze(order_key_prefix_ast, metadata_for_reading->getColumns().getAllPhysical()); + auto syntax_result = TreeRewriter(context).analyze(order_key_prefix_ast, storage_snapshot->metadata->getColumns().getAllPhysical()); auto sorting_key_prefix_expr = ExpressionAnalyzer(order_key_prefix_ast, syntax_result, context).getActionsDAG(false); - const auto & sorting_columns = metadata_for_reading->getSortingKey().column_names; + const auto & sorting_columns = storage_snapshot->metadata->getSortingKey().column_names; SortDescription sort_description; sort_description.compile_sort_description = settings.compile_sort_description; @@ -1150,7 +1167,7 @@ bool ReadFromMergeTree::doNotMergePartsAcrossPartitionsFinal() const if (settings.do_not_merge_across_partitions_select_final.changed) return settings.do_not_merge_across_partitions_select_final; - if (!metadata_for_reading->hasPrimaryKey() || !metadata_for_reading->hasPartitionKey()) + if (!storage_snapshot->metadata->hasPrimaryKey() || !storage_snapshot->metadata->hasPartitionKey()) return false; /** To avoid merging parts across partitions we want result of partition key expression for @@ -1160,11 +1177,11 @@ bool ReadFromMergeTree::doNotMergePartsAcrossPartitionsFinal() const * in primary key, then for same primary key column values, result of partition key expression * will be the same. */ - const auto & partition_key_expression = metadata_for_reading->getPartitionKey().expression; + const auto & partition_key_expression = storage_snapshot->metadata->getPartitionKey().expression; if (partition_key_expression->getActionsDAG().hasNonDeterministic()) return false; - const auto & primary_key_columns = metadata_for_reading->getPrimaryKey().column_names; + const auto & primary_key_columns = storage_snapshot->metadata->getPrimaryKey().column_names; NameSet primary_key_columns_set(primary_key_columns.begin(), primary_key_columns.end()); const auto & partition_key_required_columns = partition_key_expression->getRequiredColumns(); @@ -1217,12 +1234,12 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsFinal( /// we will store lonely parts with level > 0 to use parallel select on them. RangesInDataParts non_intersecting_parts_by_primary_key; - auto sorting_expr = metadata_for_reading->getSortingKey().expression; + auto sorting_expr = storage_snapshot->metadata->getSortingKey().expression; if (prewhere_info) { NameSet sorting_columns; - for (const auto & column : metadata_for_reading->getSortingKey().expression->getRequiredColumnsWithTypes()) + for (const auto & column : storage_snapshot->metadata->getSortingKey().expression->getRequiredColumnsWithTypes()) sorting_columns.insert(column.name); restorePrewhereInputs(*prewhere_info, sorting_columns); } @@ -1253,7 +1270,7 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsFinal( if (new_parts.empty()) continue; - if (num_streams > 1 && metadata_for_reading->hasPrimaryKey()) + if (num_streams > 1 && storage_snapshot->metadata->hasPrimaryKey()) { // Let's split parts into non intersecting parts ranges and layers to ensure data parallelism of FINAL. auto in_order_reading_step_getter = [this, &column_names, &info](auto parts) @@ -1273,7 +1290,7 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsFinal( data.merging_params.is_deleted_column.empty() && !reader_settings.read_in_order; SplitPartsWithRangesByPrimaryKeyResult split_ranges_result = splitPartsWithRangesByPrimaryKey( - metadata_for_reading->getPrimaryKey(), + storage_snapshot->metadata->getPrimaryKey(), sorting_expr, std::move(new_parts), num_streams, @@ -1305,7 +1322,7 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsFinal( if (pipes.empty()) continue; - Names sort_columns = metadata_for_reading->getSortingKeyColumns(); + Names sort_columns = storage_snapshot->metadata->getSortingKeyColumns(); SortDescription sort_description; sort_description.compile_sort_description = settings.compile_sort_description; sort_description.min_count_to_compile_sort_description = settings.min_count_to_compile_sort_description; @@ -1313,7 +1330,7 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsFinal( size_t sort_columns_size = sort_columns.size(); sort_description.reserve(sort_columns_size); - Names partition_key_columns = metadata_for_reading->getPartitionKey().column_names; + Names partition_key_columns = storage_snapshot->metadata->getPartitionKey().column_names; for (size_t i = 0; i < sort_columns_size; ++i) sort_description.emplace_back(sort_columns[i], 1, 1); @@ -1370,7 +1387,7 @@ ReadFromMergeTree::AnalysisResultPtr ReadFromMergeTree::selectRangesToRead( return selectRangesToRead( std::move(parts), std::move(alter_conversions), - metadata_for_reading, + storage_snapshot->metadata, query_info, context, requested_num_streams, @@ -1475,16 +1492,14 @@ static void buildIndexes( else { MergeTreeIndexConditionPtr condition; - if (index_helper->isVectorSearch()) + if (index_helper->isVectorSimilarityIndex()) { -#ifdef ENABLE_ANNOY - if (const auto * annoy = typeid_cast(index_helper.get())) - condition = annoy->createIndexCondition(query_info, context); -#endif -#ifdef ENABLE_USEARCH - if (const auto * usearch = typeid_cast(index_helper.get())) - condition = usearch->createIndexCondition(query_info, context); +#if USE_USEARCH + if (const auto * vector_similarity_index = typeid_cast(index_helper.get())) + condition = vector_similarity_index->createIndexCondition(query_info, context); #endif + if (const auto * legacy_vector_similarity_index = typeid_cast(index_helper.get())) + condition = legacy_vector_similarity_index->createIndexCondition(query_info, context); if (!condition) throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown vector search index {}", index_helper->index.name); } @@ -1534,7 +1549,7 @@ void ReadFromMergeTree::applyFilters(ActionDAGNodes added_filter_nodes) prepared_parts, context, query_info, - metadata_for_reading); + storage_snapshot->metadata); } } @@ -1703,7 +1718,7 @@ bool ReadFromMergeTree::requestReadingInOrder(size_t prefix_size, int direction, /// update sort info for output stream SortDescription sort_description; - const Names & sorting_key_columns = metadata_for_reading->getSortingKeyColumns(); + const Names & sorting_key_columns = storage_snapshot->metadata->getSortingKeyColumns(); const Block & header = output_stream->header; const int sort_direction = getSortDirection(); for (const auto & column_name : sorting_key_columns) @@ -1745,7 +1760,7 @@ void ReadFromMergeTree::updatePrewhereInfo(const PrewhereInfoPtr & prewhere_info updateSortDescriptionForOutputStream( *output_stream, - storage_snapshot->getMetadataForQuery()->getSortingKeyColumns(), + storage_snapshot->metadata->getSortingKeyColumns(), getSortDirection(), query_info.input_order_info, prewhere_info, @@ -1871,7 +1886,7 @@ Pipe ReadFromMergeTree::spreadMarkRanges( throw Exception(ErrorCodes::LOGICAL_ERROR, "Optimization isn't supposed to be used for queries with final"); /// Add columns needed to calculate the sorting expression and the sign. - for (const auto & column : metadata_for_reading->getColumnsRequiredForSortingKey()) + for (const auto & column : storage_snapshot->metadata->getColumnsRequiredForSortingKey()) { if (!names.contains(column)) { @@ -1965,10 +1980,6 @@ void ReadFromMergeTree::initializePipeline(QueryPipelineBuilder & pipeline, cons fmt::format("{}.{}", data.getStorageID().getFullNameNotQuoted(), part.data_part->info.partition_id)); } context->getQueryContext()->addQueryAccessInfo(partition_names); - - if (storage_snapshot->projection) - context->getQueryContext()->addQueryAccessInfo( - Context::QualifiedProjectionName{.storage_id = data.getStorageID(), .projection_name = storage_snapshot->projection->name}); } ProfileEvents::increment(ProfileEvents::SelectedParts, result.selected_parts); diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.h b/src/Processors/QueryPlan/ReadFromMergeTree.h index a12f53924c3..f12da5d10bc 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.h +++ b/src/Processors/QueryPlan/ReadFromMergeTree.h @@ -171,7 +171,7 @@ public: AnalysisResultPtr selectRangesToRead(bool find_exact_ranges = false) const; - StorageMetadataPtr getStorageMetadata() const { return metadata_for_reading; } + StorageMetadataPtr getStorageMetadata() const { return storage_snapshot->metadata; } /// Returns `false` if requested reading cannot be performed. bool requestReadingInOrder(size_t prefix_size, int direction, size_t limit); @@ -216,8 +216,6 @@ private: const MergeTreeData & data; ExpressionActionsSettings actions_settings; - StorageMetadataPtr metadata_for_reading; - const MergeTreeReadTask::BlockSizeParams block_size; size_t requested_num_streams; diff --git a/src/Processors/QueryPlan/ReadFromSystemNumbersStep.cpp b/src/Processors/QueryPlan/ReadFromSystemNumbersStep.cpp index 2080e29ceba..596d08845e1 100644 --- a/src/Processors/QueryPlan/ReadFromSystemNumbersStep.cpp +++ b/src/Processors/QueryPlan/ReadFromSystemNumbersStep.cpp @@ -119,23 +119,23 @@ using RangesWithStep = std::vector; std::optional steppedRangeFromRange(const Range & r, UInt64 step, UInt64 remainder) { - if ((r.right.get() == 0) && (!r.right_included)) + if ((r.right.safeGet() == 0) && (!r.right_included)) return std::nullopt; - UInt64 begin = (r.left.get() / step) * step; + UInt64 begin = (r.left.safeGet() / step) * step; if (begin > std::numeric_limits::max() - remainder) return std::nullopt; begin += remainder; - while ((r.left_included <= r.left.get()) && (begin <= r.left.get() - r.left_included)) + while ((r.left_included <= r.left.safeGet()) && (begin <= r.left.safeGet() - r.left_included)) { if (std::numeric_limits::max() - step < begin) return std::nullopt; begin += step; } - if ((begin >= r.right_included) && (begin - r.right_included >= r.right.get())) + if ((begin >= r.right_included) && (begin - r.right_included >= r.right.safeGet())) return std::nullopt; - UInt64 right_edge_included = r.right.get() - (1 - r.right_included); + UInt64 right_edge_included = r.right.safeGet() - (1 - r.right_included); return std::optional{RangeWithStep{begin, step, static_cast(right_edge_included - begin) / step + 1}}; } diff --git a/src/Processors/RowsBeforeLimitCounter.h b/src/Processors/RowsBeforeLimitCounter.h deleted file mode 100644 index f5eb40ff84a..00000000000 --- a/src/Processors/RowsBeforeLimitCounter.h +++ /dev/null @@ -1,36 +0,0 @@ -#pragma once -#include -#include - -namespace DB -{ - -/// This class helps to calculate rows_before_limit_at_least. -class RowsBeforeLimitCounter -{ -public: - void add(uint64_t rows) - { - setAppliedLimit(); - rows_before_limit.fetch_add(rows, std::memory_order_release); - } - - void set(uint64_t rows) - { - setAppliedLimit(); - rows_before_limit.store(rows, std::memory_order_release); - } - - uint64_t get() const { return rows_before_limit.load(std::memory_order_acquire); } - - void setAppliedLimit() { has_applied_limit.store(true, std::memory_order_release); } - bool hasAppliedLimit() const { return has_applied_limit.load(std::memory_order_acquire); } - -private: - std::atomic rows_before_limit = 0; - std::atomic_bool has_applied_limit = false; -}; - -using RowsBeforeLimitCounterPtr = std::shared_ptr; - -} diff --git a/src/Processors/RowsBeforeStepCounter.h b/src/Processors/RowsBeforeStepCounter.h new file mode 100644 index 00000000000..789731f82bd --- /dev/null +++ b/src/Processors/RowsBeforeStepCounter.h @@ -0,0 +1,36 @@ +#pragma once +#include +#include + +namespace DB +{ + +/// This class helps to calculate rows_before_limit_at_least and rows_before_aggregation. +class RowsBeforeStepCounter +{ +public: + void add(uint64_t rows) + { + setAppliedStep(); + rows_before_step.fetch_add(rows, std::memory_order_release); + } + + void set(uint64_t rows) + { + setAppliedStep(); + rows_before_step.store(rows, std::memory_order_release); + } + + uint64_t get() const { return rows_before_step.load(std::memory_order_acquire); } + + void setAppliedStep() { has_applied_step.store(true, std::memory_order_release); } + bool hasAppliedStep() const { return has_applied_step.load(std::memory_order_acquire); } + +private: + std::atomic rows_before_step = 0; + std::atomic_bool has_applied_step = false; +}; + +using RowsBeforeStepCounterPtr = std::shared_ptr; + +} diff --git a/src/Processors/Sources/DelayedSource.cpp b/src/Processors/Sources/DelayedSource.cpp index f7928f89015..788017e3df0 100644 --- a/src/Processors/Sources/DelayedSource.cpp +++ b/src/Processors/Sources/DelayedSource.cpp @@ -139,6 +139,12 @@ void DelayedSource::work() processor->setRowsBeforeLimitCounter(rows_before_limit); } + if (rows_before_aggregation) + { + for (auto & processor : processors) + processor->setRowsBeforeAggregationCounter(rows_before_aggregation); + } + synchronizePorts(totals_output, totals, header, processors); synchronizePorts(extremes_output, extremes, header, processors); } diff --git a/src/Processors/Sources/DelayedSource.h b/src/Processors/Sources/DelayedSource.h index 0b2751e18a6..4ee90e34599 100644 --- a/src/Processors/Sources/DelayedSource.h +++ b/src/Processors/Sources/DelayedSource.h @@ -30,13 +30,15 @@ public: OutputPort * getTotalsPort() { return totals; } OutputPort * getExtremesPort() { return extremes; } - void setRowsBeforeLimitCounter(RowsBeforeLimitCounterPtr counter) override { rows_before_limit.swap(counter); } + void setRowsBeforeLimitCounter(RowsBeforeStepCounterPtr counter) override { rows_before_limit.swap(counter); } + void setRowsBeforeAggregationCounter(RowsBeforeStepCounterPtr counter) override { rows_before_aggregation.swap(counter); } private: QueryPlanResourceHolder resources; Creator creator; Processors processors; - RowsBeforeLimitCounterPtr rows_before_limit; + RowsBeforeStepCounterPtr rows_before_limit; + RowsBeforeStepCounterPtr rows_before_aggregation; /// Outputs for DelayedSource. OutputPort * main = nullptr; diff --git a/src/Processors/Sources/MySQLSource.cpp b/src/Processors/Sources/MySQLSource.cpp index 5d533a7747e..52be9a6e84a 100644 --- a/src/Processors/Sources/MySQLSource.cpp +++ b/src/Processors/Sources/MySQLSource.cpp @@ -219,11 +219,11 @@ namespace read_bytes_size += 8; break; case ValueType::vtEnum8: - assert_cast(column).insertValue(assert_cast &>(data_type).castToValue(value.data()).get()); + assert_cast(column).insertValue(assert_cast &>(data_type).castToValue(value.data()).safeGet()); read_bytes_size += assert_cast(column).byteSize(); break; case ValueType::vtEnum16: - assert_cast(column).insertValue(assert_cast &>(data_type).castToValue(value.data()).get()); + assert_cast(column).insertValue(assert_cast &>(data_type).castToValue(value.data()).safeGet()); read_bytes_size += assert_cast(column).byteSize(); break; case ValueType::vtString: diff --git a/src/Processors/Sources/PostgreSQLSource.cpp b/src/Processors/Sources/PostgreSQLSource.cpp index a3d6fd691d8..b9bda46bd10 100644 --- a/src/Processors/Sources/PostgreSQLSource.cpp +++ b/src/Processors/Sources/PostgreSQLSource.cpp @@ -35,9 +35,9 @@ PostgreSQLSource::PostgreSQLSource( const Block & sample_block, UInt64 max_block_size_) : ISource(sample_block.cloneEmpty()) - , query_str(query_str_) , max_block_size(max_block_size_) , connection_holder(std::move(connection_holder_)) + , query_str(query_str_) { init(sample_block); } @@ -51,10 +51,10 @@ PostgreSQLSource::PostgreSQLSource( UInt64 max_block_size_, bool auto_commit_) : ISource(sample_block.cloneEmpty()) - , query_str(query_str_) - , tx(std::move(tx_)) , max_block_size(max_block_size_) , auto_commit(auto_commit_) + , query_str(query_str_) + , tx(std::move(tx_)) { init(sample_block); } @@ -204,15 +204,15 @@ PostgreSQLSource::~PostgreSQLSource() */ stream->close(); } - - stream.reset(); - tx.reset(); } catch (...) { tryLogCurrentException(__PRETTY_FUNCTION__); } + stream.reset(); + tx.reset(); + if (connection_holder) connection_holder->setBroken(); } diff --git a/src/Processors/Sources/PostgreSQLSource.h b/src/Processors/Sources/PostgreSQLSource.h index 8a648ae8bb5..319c5d8d7c2 100644 --- a/src/Processors/Sources/PostgreSQLSource.h +++ b/src/Processors/Sources/PostgreSQLSource.h @@ -38,14 +38,12 @@ protected: UInt64 max_block_size_, bool auto_commit_); - String query_str; - std::shared_ptr tx; - std::unique_ptr stream; - Status prepare() override; - void onStart(); Chunk generate() override; + + void onStart(); + void onFinish(); private: @@ -61,6 +59,12 @@ private: postgres::ConnectionHolderPtr connection_holder; std::unordered_map array_info; + +protected: + String query_str; + /// tx and stream must be destroyed before connection_holder. + std::shared_ptr tx; + std::unique_ptr stream; }; diff --git a/src/Processors/Sources/RemoteSource.cpp b/src/Processors/Sources/RemoteSource.cpp index 2f9a30296be..c2da5753a27 100644 --- a/src/Processors/Sources/RemoteSource.cpp +++ b/src/Processors/Sources/RemoteSource.cpp @@ -4,6 +4,8 @@ #include #include #include +#include +#include namespace DB { @@ -35,16 +37,23 @@ RemoteSource::RemoteSource(RemoteQueryExecutorPtr executor, bool add_aggregation progress(value.read_rows, value.read_bytes); }); - query_executor->setProfileInfoCallback([this](const ProfileInfo & info) - { - if (rows_before_limit) + query_executor->setProfileInfoCallback( + [this](const ProfileInfo & info) { - if (info.hasAppliedLimit()) - rows_before_limit->add(info.getRowsBeforeLimit()); - else - manually_add_rows_before_limit_counter = true; /// Remote subquery doesn't contain a limit - } - }); + if (rows_before_limit) + { + if (info.hasAppliedLimit()) + rows_before_limit->add(info.getRowsBeforeLimit()); + else + manually_add_rows_before_limit_counter = true; /// Remote subquery doesn't contain a limit + } + + if (rows_before_aggregation) + { + if (info.hasAppliedAggregation()) + rows_before_aggregation->add(info.getRowsBeforeAggregation()); + } + }); } RemoteSource::~RemoteSource() = default; @@ -182,7 +191,6 @@ std::optional RemoteSource::tryGenerate() { if (manually_add_rows_before_limit_counter) rows_before_limit->add(rows); - query_executor->finish(); return {}; } @@ -202,9 +210,16 @@ std::optional RemoteSource::tryGenerate() return chunk; } -void RemoteSource::onCancel() +void RemoteSource::onCancel() noexcept { - query_executor->cancel(); + try + { + query_executor->cancel(); + } + catch (...) + { + tryLogCurrentException(getLogger("RemoteSource"), "Error occurs on cancellation."); + } } void RemoteSource::onUpdatePorts() diff --git a/src/Processors/Sources/RemoteSource.h b/src/Processors/Sources/RemoteSource.h index 22d3921708b..c9d1a647779 100644 --- a/src/Processors/Sources/RemoteSource.h +++ b/src/Processors/Sources/RemoteSource.h @@ -1,10 +1,10 @@ #pragma once #include -#include +#include #include -#include +#include namespace DB { @@ -25,7 +25,8 @@ public: void work() override; String getName() const override { return "Remote"; } - void setRowsBeforeLimitCounter(RowsBeforeLimitCounterPtr counter) override { rows_before_limit.swap(counter); } + void setRowsBeforeLimitCounter(RowsBeforeStepCounterPtr counter) override { rows_before_limit.swap(counter); } + void setRowsBeforeAggregationCounter(RowsBeforeStepCounterPtr counter) override { rows_before_aggregation.swap(counter); } /// Stop reading from stream if output port is finished. void onUpdatePorts() override; @@ -38,7 +39,7 @@ public: protected: std::optional tryGenerate() override; - void onCancel() override; + void onCancel() noexcept override; private: bool was_query_sent = false; @@ -46,7 +47,8 @@ private: bool executor_finished = false; bool add_aggregation_info = false; RemoteQueryExecutorPtr query_executor; - RowsBeforeLimitCounterPtr rows_before_limit; + RowsBeforeStepCounterPtr rows_before_limit; + RowsBeforeStepCounterPtr rows_before_aggregation; const bool async_read; const bool async_query_sending; diff --git a/src/Processors/Sources/ShellCommandSource.cpp b/src/Processors/Sources/ShellCommandSource.cpp index 55eaf67eb3b..f55a3713215 100644 --- a/src/Processors/Sources/ShellCommandSource.cpp +++ b/src/Processors/Sources/ShellCommandSource.cpp @@ -8,13 +8,15 @@ #include #include -#include -#include -#include -#include #include +#include +#include +#include +#include + #include +#include namespace DB { @@ -68,11 +70,17 @@ static void makeFdBlocking(int fd) static int pollWithTimeout(pollfd * pfds, size_t num, size_t timeout_milliseconds) { + auto logger = getLogger("TimeoutReadBufferFromFileDescriptor"); + auto describe_fd = [](const auto & pollfd) { return fmt::format("(fd={}, flags={})", pollfd.fd, fcntl(pollfd.fd, F_GETFL)); }; + int res; while (true) { Stopwatch watch; + + LOG_TEST(logger, "Polling descriptors: {}", fmt::join(std::span(pfds, pfds + num) | std::views::transform(describe_fd), ", ")); + res = poll(pfds, static_cast(num), static_cast(timeout_milliseconds)); if (res < 0) @@ -82,7 +90,10 @@ static int pollWithTimeout(pollfd * pfds, size_t num, size_t timeout_millisecond const auto elapsed = watch.elapsedMilliseconds(); if (timeout_milliseconds <= elapsed) + { + LOG_TEST(logger, "Timeout exceeded: elapsed={}, timeout={}", elapsed, timeout_milliseconds); break; + } timeout_milliseconds -= elapsed; } else @@ -91,6 +102,12 @@ static int pollWithTimeout(pollfd * pfds, size_t num, size_t timeout_millisecond } } + LOG_TEST( + logger, + "Poll for descriptors: {} returned {}", + fmt::join(std::span(pfds, pfds + num) | std::views::transform(describe_fd), ", "), + res); + return res; } @@ -200,12 +217,6 @@ public: return true; } - void reset() const - { - makeFdBlocking(stdout_fd); - makeFdBlocking(stderr_fd); - } - ~TimeoutReadBufferFromFileDescriptor() override { tryMakeFdBlocking(stdout_fd); diff --git a/src/Processors/Transforms/AggregatingInOrderTransform.cpp b/src/Processors/Transforms/AggregatingInOrderTransform.cpp index 45b0960ec8f..f8bc419b623 100644 --- a/src/Processors/Transforms/AggregatingInOrderTransform.cpp +++ b/src/Processors/Transforms/AggregatingInOrderTransform.cpp @@ -81,6 +81,8 @@ void AggregatingInOrderTransform::consume(Chunk chunk) is_consume_started = true; } + if (rows_before_aggregation) + rows_before_aggregation->add(rows); src_rows += rows; src_bytes += chunk.bytes(); diff --git a/src/Processors/Transforms/AggregatingInOrderTransform.h b/src/Processors/Transforms/AggregatingInOrderTransform.h index 41a0d7fc7f1..2ac7ffba1aa 100644 --- a/src/Processors/Transforms/AggregatingInOrderTransform.h +++ b/src/Processors/Transforms/AggregatingInOrderTransform.h @@ -45,6 +45,7 @@ public: void work() override; void consume(Chunk chunk); + void setRowsBeforeAggregationCounter(RowsBeforeStepCounterPtr counter) override { rows_before_aggregation.swap(counter); } private: void generate(); @@ -86,6 +87,8 @@ private: Chunk current_chunk; Chunk to_push_chunk; + RowsBeforeStepCounterPtr rows_before_aggregation; + LoggerPtr log = getLogger("AggregatingInOrderTransform"); }; diff --git a/src/Processors/Transforms/AggregatingTransform.cpp b/src/Processors/Transforms/AggregatingTransform.cpp index fda03174c53..c9ada32b839 100644 --- a/src/Processors/Transforms/AggregatingTransform.cpp +++ b/src/Processors/Transforms/AggregatingTransform.cpp @@ -8,7 +8,6 @@ #include #include #include - #include @@ -367,7 +366,7 @@ public: return prepareTwoLevel(); } - void onCancel() override + void onCancel() noexcept override { shared_data->is_cancelled.store(true, std::memory_order_seq_cst); } @@ -676,7 +675,8 @@ void AggregatingTransform::consume(Chunk chunk) LOG_TRACE(log, "Aggregating"); is_consume_started = true; } - + if (rows_before_aggregation) + rows_before_aggregation->add(num_rows); src_rows += num_rows; src_bytes += chunk.bytes(); diff --git a/src/Processors/Transforms/AggregatingTransform.h b/src/Processors/Transforms/AggregatingTransform.h index 95983c39d1e..b9212375c91 100644 --- a/src/Processors/Transforms/AggregatingTransform.h +++ b/src/Processors/Transforms/AggregatingTransform.h @@ -4,11 +4,13 @@ #include #include #include -#include -#include -#include +#include #include #include +#include +#include +#include + namespace CurrentMetrics { @@ -168,6 +170,7 @@ public: Status prepare() override; void work() override; Processors expandPipeline() override; + void setRowsBeforeAggregationCounter(RowsBeforeStepCounterPtr counter) override { rows_before_aggregation.swap(counter); } protected: void consume(Chunk chunk); @@ -211,6 +214,8 @@ private: bool is_consume_started = false; + RowsBeforeStepCounterPtr rows_before_aggregation; + void initGenerate(); }; diff --git a/src/Processors/Transforms/ColumnGathererTransform.cpp b/src/Processors/Transforms/ColumnGathererTransform.cpp index 15f8355bdc7..52fa42fdb51 100644 --- a/src/Processors/Transforms/ColumnGathererTransform.cpp +++ b/src/Processors/Transforms/ColumnGathererTransform.cpp @@ -1,11 +1,15 @@ #include +#include #include #include #include #include #include -#include +namespace ProfileEvents +{ + extern const Event GatheringColumnMilliseconds; +} namespace DB { @@ -33,6 +37,13 @@ ColumnGathererStream::ColumnGathererStream( throw Exception(ErrorCodes::EMPTY_DATA_PASSED, "There are no streams to gather"); } +void ColumnGathererStream::updateStats(const IColumn & column) +{ + merged_rows += column.size(); + merged_bytes += column.allocatedBytes(); + ++merged_blocks; +} + void ColumnGathererStream::initialize(Inputs inputs) { Columns source_columns; @@ -82,7 +93,9 @@ IMergingAlgorithm::Status ColumnGathererStream::merge() { res.addColumn(source_to_fully_copy->column); } - merged_rows += source_to_fully_copy->size; + + updateStats(*source_to_fully_copy->column); + source_to_fully_copy->pos = source_to_fully_copy->size; source_to_fully_copy = nullptr; return Status(std::move(res)); @@ -96,8 +109,7 @@ IMergingAlgorithm::Status ColumnGathererStream::merge() { next_required_source = 0; Chunk res; - merged_rows += sources.front().column->size(); - merged_bytes += sources.front().column->allocatedBytes(); + updateStats(*sources.front().column); res.addColumn(std::move(sources.front().column)); sources.front().pos = sources.front().size = 0; return Status(std::move(res)); @@ -123,8 +135,8 @@ IMergingAlgorithm::Status ColumnGathererStream::merge() if (source_to_fully_copy && result_column->empty()) { Chunk res; - merged_rows += source_to_fully_copy->column->size(); - merged_bytes += source_to_fully_copy->column->allocatedBytes(); + updateStats(*source_to_fully_copy->column); + if (result_column->hasDynamicStructure()) { auto col = result_column->cloneEmpty(); @@ -140,13 +152,13 @@ IMergingAlgorithm::Status ColumnGathererStream::merge() return Status(std::move(res)); } - auto col = result_column->cloneEmpty(); - result_column.swap(col); + auto return_column = result_column->cloneEmpty(); + result_column.swap(return_column); Chunk res; - merged_rows += col->size(); - merged_bytes += col->allocatedBytes(); - res.addColumn(std::move(col)); + updateStats(*return_column); + + res.addColumn(std::move(return_column)); return Status(std::move(res), row_sources_buf.eof() && !source_to_fully_copy); } @@ -185,31 +197,10 @@ ColumnGathererTransform::ColumnGathererTransform( toString(header.columns())); } -void ColumnGathererTransform::work() -{ - Stopwatch stopwatch; - IMergingTransform::work(); - elapsed_ns += stopwatch.elapsedNanoseconds(); -} - void ColumnGathererTransform::onFinish() { - auto merged_rows = algorithm.getMergedRows(); - auto merged_bytes = algorithm.getMergedRows(); - /// Don't print info for small parts (< 10M rows) - if (merged_rows < 10000000) - return; - - double seconds = static_cast(elapsed_ns) / 1000000000ULL; const auto & column_name = getOutputPort().getHeader().getByPosition(0).name; - - if (seconds == 0.0) - LOG_DEBUG(log, "Gathered column {} ({} bytes/elem.) in 0 sec.", - column_name, static_cast(merged_bytes) / merged_rows); - else - LOG_DEBUG(log, "Gathered column {} ({} bytes/elem.) in {} sec., {} rows/sec., {}/sec.", - column_name, static_cast(merged_bytes) / merged_rows, seconds, - merged_rows / seconds, ReadableSize(merged_bytes / seconds)); + logMergedStats(ProfileEvents::GatheringColumnMilliseconds, fmt::format("Gathered column {}", column_name), log); } } diff --git a/src/Processors/Transforms/ColumnGathererTransform.h b/src/Processors/Transforms/ColumnGathererTransform.h index ec5691316ce..fbc9a6bfcc6 100644 --- a/src/Processors/Transforms/ColumnGathererTransform.h +++ b/src/Processors/Transforms/ColumnGathererTransform.h @@ -72,10 +72,11 @@ public: template void gather(Column & column_res); - UInt64 getMergedRows() const { return merged_rows; } - UInt64 getMergedBytes() const { return merged_bytes; } + MergedStats getMergedStats() const override { return {.bytes = merged_bytes, .rows = merged_rows, .blocks = merged_blocks}; } private: + void updateStats(const IColumn & column); + /// Cache required fields struct Source { @@ -105,6 +106,7 @@ private: ssize_t next_required_source = -1; UInt64 merged_rows = 0; UInt64 merged_bytes = 0; + UInt64 merged_blocks = 0; }; class ColumnGathererTransform final : public IMergingTransform @@ -120,12 +122,8 @@ public: String getName() const override { return "ColumnGathererTransform"; } - void work() override; - protected: void onFinish() override; - UInt64 elapsed_ns = 0; - LoggerPtr log; }; diff --git a/src/Processors/Transforms/ExceptionKeepingTransform.h b/src/Processors/Transforms/ExceptionKeepingTransform.h index 000b5da798a..9aa33a8cbe5 100644 --- a/src/Processors/Transforms/ExceptionKeepingTransform.h +++ b/src/Processors/Transforms/ExceptionKeepingTransform.h @@ -52,7 +52,7 @@ protected: virtual void onConsume(Chunk chunk) = 0; virtual GenerateResult onGenerate() = 0; virtual void onFinish() {} - virtual void onException(std::exception_ptr /* exception */) {} + virtual void onException(std::exception_ptr /* exception */) { } public: ExceptionKeepingTransform(const Block & in_header, const Block & out_header, bool ignore_on_start_and_finish_ = true); diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index 9601f821cc8..95f4a674ebb 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -62,7 +62,7 @@ static FillColumnDescription::StepFunction getStepFunction( case IntervalKind::Kind::NAME: \ return [step, scale, &date_lut](Field & field) { \ field = Add##NAME##sImpl::execute(static_cast(\ - field.get()), static_cast(step), date_lut, utc_time_zone, scale); }; + field.safeGet()), static_cast(step), date_lut, utc_time_zone, scale); }; FOR_EACH_INTERVAL_KIND(DECLARE_CASE) #undef DECLARE_CASE @@ -139,21 +139,21 @@ static bool tryConvertFields(FillColumnDescription & descr, const DataTypePtr & { if (which.isDate() || which.isDate32()) { - Int64 avg_seconds = descr.fill_step.get() * descr.step_kind->toAvgSeconds(); + Int64 avg_seconds = descr.fill_step.safeGet() * descr.step_kind->toAvgSeconds(); if (std::abs(avg_seconds) < 86400) throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, "Value of step is to low ({} seconds). Must be >= 1 day", std::abs(avg_seconds)); } if (which.isDate()) - descr.step_func = getStepFunction(*descr.step_kind, descr.fill_step.get(), DateLUT::instance()); + descr.step_func = getStepFunction(*descr.step_kind, descr.fill_step.safeGet(), DateLUT::instance()); else if (which.isDate32()) - descr.step_func = getStepFunction(*descr.step_kind, descr.fill_step.get(), DateLUT::instance()); + descr.step_func = getStepFunction(*descr.step_kind, descr.fill_step.safeGet(), DateLUT::instance()); else if (const auto * date_time = checkAndGetDataType(type.get())) - descr.step_func = getStepFunction(*descr.step_kind, descr.fill_step.get(), date_time->getTimeZone()); + descr.step_func = getStepFunction(*descr.step_kind, descr.fill_step.safeGet(), date_time->getTimeZone()); else if (const auto * date_time64 = checkAndGetDataType(type.get())) { - const auto & step_dec = descr.fill_step.get &>(); + const auto & step_dec = descr.fill_step.safeGet &>(); Int64 step = DecimalUtils::convertTo(step_dec.getValue(), step_dec.getScale()); static const DateLUTImpl & utc_time_zone = DateLUT::instance("UTC"); @@ -163,7 +163,7 @@ static bool tryConvertFields(FillColumnDescription & descr, const DataTypePtr & case IntervalKind::Kind::NAME: \ descr.step_func = [step, &time_zone = date_time64->getTimeZone()](Field & field) \ { \ - auto field_decimal = field.get>(); \ + auto field_decimal = field.safeGet>(); \ auto res = Add##NAME##sImpl::execute(field_decimal.getValue(), step, time_zone, utc_time_zone, field_decimal.getScale()); \ field = DecimalField(res, field_decimal.getScale()); \ }; \ diff --git a/src/Processors/Transforms/MergeJoinTransform.cpp b/src/Processors/Transforms/MergeJoinTransform.cpp index e96a75d277b..6abfa0fccd0 100644 --- a/src/Processors/Transforms/MergeJoinTransform.cpp +++ b/src/Processors/Transforms/MergeJoinTransform.cpp @@ -511,6 +511,16 @@ void MergeJoinAlgorithm::logElapsed(double seconds) stat.max_blocks_loaded); } +IMergingAlgorithm::MergedStats MergeJoinAlgorithm::getMergedStats() const +{ + return + { + .bytes = stat.num_bytes[0] + stat.num_bytes[1], + .rows = stat.num_rows[0] + stat.num_rows[1], + .blocks = stat.num_blocks[0] + stat.num_blocks[1], + }; +} + static void prepareChunk(Chunk & chunk) { if (!chunk) @@ -547,6 +557,7 @@ void MergeJoinAlgorithm::consume(Input & input, size_t source_num) { stat.num_blocks[source_num] += 1; stat.num_rows[source_num] += input.chunk.getNumRows(); + stat.num_bytes[source_num] += input.chunk.allocatedBytes(); } prepareChunk(input.chunk); @@ -1271,7 +1282,7 @@ MergeJoinTransform::MergeJoinTransform( void MergeJoinTransform::onFinish() { - algorithm.logElapsed(total_stopwatch.elapsedSeconds()); + algorithm.logElapsed(static_cast(merging_elapsed_ns) / 1000000000ULL); } } diff --git a/src/Processors/Transforms/MergeJoinTransform.h b/src/Processors/Transforms/MergeJoinTransform.h index d37a0b9f3ae..8f74974af0f 100644 --- a/src/Processors/Transforms/MergeJoinTransform.h +++ b/src/Processors/Transforms/MergeJoinTransform.h @@ -245,6 +245,8 @@ public: void setAsofInequality(ASOFJoinInequality asof_inequality_); void logElapsed(double seconds); + MergedStats getMergedStats() const override; + private: std::optional handleAnyJoinState(); Status anyJoin(); @@ -280,6 +282,7 @@ private: { size_t num_blocks[2] = {0, 0}; size_t num_rows[2] = {0, 0}; + size_t num_bytes[2] = {0, 0}; size_t max_blocks_loaded = 0; }; diff --git a/src/Processors/Transforms/MergeSortingTransform.cpp b/src/Processors/Transforms/MergeSortingTransform.cpp index ede13b29219..c45192e7118 100644 --- a/src/Processors/Transforms/MergeSortingTransform.cpp +++ b/src/Processors/Transforms/MergeSortingTransform.cpp @@ -185,7 +185,6 @@ void MergeSortingTransform::consume(Chunk chunk) if (!external_merging_sorted) { - bool quiet = false; bool have_all_inputs = false; bool use_average_block_sizes = false; @@ -199,7 +198,6 @@ void MergeSortingTransform::consume(Chunk chunk) limit, /*always_read_till_end_=*/ false, nullptr, - quiet, use_average_block_sizes, have_all_inputs); diff --git a/src/Processors/Transforms/PartialSortingTransform.h b/src/Processors/Transforms/PartialSortingTransform.h index 8f25c93037f..73c490d5b92 100644 --- a/src/Processors/Transforms/PartialSortingTransform.h +++ b/src/Processors/Transforms/PartialSortingTransform.h @@ -1,7 +1,7 @@ #pragma once -#include -#include #include +#include +#include #include namespace DB @@ -20,7 +20,7 @@ public: String getName() const override { return "PartialSortingTransform"; } - void setRowsBeforeLimitCounter(RowsBeforeLimitCounterPtr counter) override { read_rows.swap(counter); } + void setRowsBeforeLimitCounter(RowsBeforeStepCounterPtr counter) override { read_rows.swap(counter); } protected: void transform(Chunk & chunk) override; @@ -29,7 +29,7 @@ private: const SortDescription description; SortDescriptionWithPositions description_with_positions; const UInt64 limit; - RowsBeforeLimitCounterPtr read_rows; + RowsBeforeStepCounterPtr read_rows; Columns sort_description_threshold_columns; diff --git a/src/Processors/Transforms/PasteJoinTransform.cpp b/src/Processors/Transforms/PasteJoinTransform.cpp index d2fa7eed256..982a347a70f 100644 --- a/src/Processors/Transforms/PasteJoinTransform.cpp +++ b/src/Processors/Transforms/PasteJoinTransform.cpp @@ -58,6 +58,16 @@ static void prepareChunk(Chunk & chunk) chunk.setColumns(std::move(columns), num_rows); } +IMergingAlgorithm::MergedStats PasteJoinAlgorithm::getMergedStats() const +{ + return + { + .bytes = stat.num_bytes[0] + stat.num_bytes[1], + .rows = stat.num_rows[0] + stat.num_rows[1], + .blocks = stat.num_blocks[0] + stat.num_blocks[1], + }; +} + void PasteJoinAlgorithm::initialize(Inputs inputs) { if (inputs.size() != 2) diff --git a/src/Processors/Transforms/PasteJoinTransform.h b/src/Processors/Transforms/PasteJoinTransform.h index 6a7e65ee27c..c184f20362d 100644 --- a/src/Processors/Transforms/PasteJoinTransform.h +++ b/src/Processors/Transforms/PasteJoinTransform.h @@ -35,8 +35,7 @@ public: void initialize(Inputs inputs) override; void consume(Input & input, size_t source_num) override; Status merge() override; - - void logElapsed(double seconds); + MergedStats getMergedStats() const override; private: Chunk createBlockWithDefaults(size_t source_num); @@ -55,6 +54,7 @@ private: { size_t num_blocks[2] = {0, 0}; size_t num_rows[2] = {0, 0}; + size_t num_bytes[2] = {0, 0}; size_t max_blocks_loaded = 0; }; diff --git a/src/Processors/Transforms/WindowTransform.cpp b/src/Processors/Transforms/WindowTransform.cpp index 8de248a9c95..bd11aa4cd28 100644 --- a/src/Processors/Transforms/WindowTransform.cpp +++ b/src/Processors/Transforms/WindowTransform.cpp @@ -65,28 +65,6 @@ namespace ErrorCodes extern const int TOO_MANY_ARGUMENTS_FOR_FUNCTION; } -// Interface for true window functions. It's not much of an interface, they just -// accept the guts of WindowTransform and do 'something'. Given a small number of -// true window functions, and the fact that the WindowTransform internals are -// pretty much well-defined in domain terms (e.g. frame boundaries), this is -// somewhat acceptable. -class IWindowFunction -{ -public: - virtual ~IWindowFunction() = default; - - // Must insert the result for current_row. - virtual void windowInsertResultInto(const WindowTransform * transform, - size_t function_index) const = 0; - - virtual std::optional getDefaultFrame() const { return {}; } - - virtual ColumnPtr castColumn(const Columns &, const std::vector &) { return nullptr; } - - /// 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: // [compared] with [reference] +/- offset. Return value is -1/0/+1, like in // sorting predicates -- -1 means [compared] is less than [reference] +/- offset. @@ -107,7 +85,7 @@ static int compareValuesWithOffset(const IColumn * _compared_column, using ValueType = typename ColumnType::ValueType; // Note that the storage type of offset returned by get<> is different, so // we need to specify the type explicitly. - const ValueType offset = static_cast(_offset.get()); + const ValueType offset = static_cast(_offset.safeGet()); assert(offset >= 0); const auto compared_value_data = compared_column->getDataAt(compared_row); @@ -162,7 +140,7 @@ static int compareValuesWithOffsetFloat(const IColumn * _compared_column, _compared_column); const auto * reference_column = assert_cast( _reference_column); - const auto offset = _offset.get(); + const auto offset = _offset.safeGet(); chassert(offset >= 0); const auto compared_value_data = compared_column->getDataAt(compared_row); @@ -631,7 +609,7 @@ void WindowTransform::advanceFrameStartRowsOffset() { // Just recalculate it each time by walking blocks. const auto [moved_row, offset_left] = moveRowNumber(current_row, - window_description.frame.begin_offset.get() + window_description.frame.begin_offset.safeGet() * (window_description.frame.begin_preceding ? -1 : 1)); frame_start = moved_row; @@ -870,7 +848,7 @@ void WindowTransform::advanceFrameEndRowsOffset() // Walk the specified offset from the current row. The "+1" is needed // because the frame_end is a past-the-end pointer. const auto [moved_row, offset_left] = moveRowNumber(current_row, - window_description.frame.end_offset.get() + window_description.frame.end_offset.safeGet() * (window_description.frame.end_preceding ? -1 : 1) + 1); @@ -1179,8 +1157,7 @@ void WindowTransform::appendChunk(Chunk & chunk) // Initialize output columns. for (auto & ws : workspaces) { - if (ws.window_function_impl) - block.casted_columns.push_back(ws.window_function_impl->castColumn(block.input_columns, ws.argument_column_indices)); + block.casted_columns.push_back(ws.window_function_impl ? ws.window_function_impl->castColumn(block.input_columns, ws.argument_column_indices) : nullptr); block.output_columns.push_back(ws.aggregate_function->getResultType() ->createColumn()); @@ -1523,41 +1500,6 @@ void WindowTransform::work() } } -// A basic implementation for a true window function. It pretends to be an -// aggregate function, but refuses to work as such. -struct WindowFunction - : public IAggregateFunctionHelper - , public IWindowFunction -{ - std::string name; - - WindowFunction(const std::string & name_, const DataTypes & argument_types_, const Array & parameters_, const DataTypePtr & result_type_) - : IAggregateFunctionHelper(argument_types_, parameters_, result_type_) - , name(name_) - {} - - bool isOnlyWindowFunction() const override { return true; } - - [[noreturn]] void fail() const - { - throw Exception(ErrorCodes::BAD_ARGUMENTS, - "The function '{}' can only be used as a window function, not as an aggregate function", - getName()); - } - - String getName() const override { return name; } - void create(AggregateDataPtr __restrict) const override {} - void destroy(AggregateDataPtr __restrict) const noexcept override {} - bool hasTrivialDestructor() const override { return true; } - size_t sizeOfData() const override { return 0; } - size_t alignOfData() const override { return 1; } - void add(AggregateDataPtr __restrict, const IColumn **, size_t, Arena *) const override { fail(); } - void merge(AggregateDataPtr __restrict, ConstAggregateDataPtr, Arena *) const override { fail(); } - void serialize(ConstAggregateDataPtr __restrict, WriteBuffer &, std::optional) const override { fail(); } - void deserialize(AggregateDataPtr __restrict, ReadBuffer &, std::optional, Arena *) const override { fail(); } - void insertResultInto(AggregateDataPtr __restrict, IColumn &, Arena *) const override { fail(); } -}; - struct WindowFunctionRank final : public WindowFunction { WindowFunctionRank(const std::string & name_, @@ -1669,36 +1611,6 @@ struct WindowFunctionHelpers } }; -template -struct StatefulWindowFunction : public WindowFunction -{ - StatefulWindowFunction(const std::string & name_, - const DataTypes & argument_types_, const Array & parameters_, const DataTypePtr & result_type_) - : WindowFunction(name_, argument_types_, parameters_, result_type_) - { - } - - size_t sizeOfData() const override { return sizeof(State); } - size_t alignOfData() const override { return 1; } - - void create(AggregateDataPtr __restrict place) const override - { - new (place) State(); - } - - void destroy(AggregateDataPtr __restrict place) const noexcept override - { - reinterpret_cast(place)->~State(); - } - - bool hasTrivialDestructor() const override { return std::is_trivially_destructible_v; } - - State & getState(const WindowFunctionWorkspace & workspace) const - { - return *reinterpret_cast(workspace.aggregate_function_state.data()); - } -}; - struct ExponentialTimeDecayedSumState { Float64 previous_time; @@ -2192,13 +2104,13 @@ namespace throw Exception(ErrorCodes::BAD_ARGUMENTS, "Argument of 'ntile' function must be a constant"); auto type_id = argument_types[0]->getTypeId(); if (type_id == TypeIndex::UInt8) - buckets = arg_col[transform->current_row.row].get(); + buckets = arg_col[transform->current_row.row].safeGet(); else if (type_id == TypeIndex::UInt16) - buckets = arg_col[transform->current_row.row].get(); + buckets = arg_col[transform->current_row.row].safeGet(); else if (type_id == TypeIndex::UInt32) - buckets = arg_col[transform->current_row.row].get(); + buckets = arg_col[transform->current_row.row].safeGet(); else if (type_id == TypeIndex::UInt64) - buckets = arg_col[transform->current_row.row].get(); + buckets = arg_col[transform->current_row.row].safeGet(); if (!buckets) { @@ -2278,14 +2190,13 @@ public: bool checkWindowFrameType(const WindowTransform * transform) const override { - 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; + auto default_window_frame = getDefaultFrame(); + if (transform->window_description.frame != default_window_frame) + { + LOG_ERROR( + getLogger("WindowFunctionPercentRank"), + "Window frame for function 'percent_rank' should be '{}'", default_window_frame->toString()); + return false; } return true; } @@ -2295,7 +2206,7 @@ public: WindowFrame frame; frame.type = WindowFrame::FrameType::RANGE; frame.begin_type = WindowFrame::BoundaryType::Unbounded; - frame.end_type = WindowFrame::BoundaryType::Current; + frame.end_type = WindowFrame::BoundaryType::Unbounded; return frame; } @@ -2425,22 +2336,9 @@ struct WindowFunctionLagLeadInFrame final : public WindowFunction argument_types[2]->getName()); } - const auto from_name = argument_types[2]->getName(); - const auto to_name = argument_types[0]->getName(); - ColumnsWithTypeAndName arguments + auto get_cast_func = [from = argument_types[2], to = argument_types[0]] { - { argument_types[2], "" }, - { - DataTypeString().createColumnConst(0, to_name), - std::make_shared(), - "" - } - }; - - auto get_cast_func = [&arguments] - { - FunctionOverloadResolverPtr func_builder_cast = createInternalCastOverloadResolver(CastType::accurate, {}); - return func_builder_cast->build(arguments); + return createInternalCast({from, {}}, to, CastType::accurate, {}); }; func_cast = get_cast_func(); @@ -2490,7 +2388,7 @@ struct WindowFunctionLagLeadInFrame final : public WindowFunction { offset = (*current_block.input_columns[ workspace.argument_column_indices[1]])[ - transform->current_row.row].get(); + transform->current_row.row].safeGet(); /// Either overflow or really negative value, both is not acceptable. if (offset < 0) @@ -2576,7 +2474,7 @@ struct WindowFunctionNthValue final : public WindowFunction Int64 offset = (*current_block.input_columns[ workspace.argument_column_indices[1]])[ - transform->current_row.row].get(); + transform->current_row.row].safeGet(); /// Either overflow or really negative value, both is not acceptable. if (offset <= 0) @@ -2860,5 +2758,4 @@ void registerWindowFunctions(AggregateFunctionFactory & factory) name, argument_types, parameters); }, properties}); } - } diff --git a/src/Processors/Transforms/WindowTransform.h b/src/Processors/Transforms/WindowTransform.h index fe4f79e997c..cb672ad6841 100644 --- a/src/Processors/Transforms/WindowTransform.h +++ b/src/Processors/Transforms/WindowTransform.h @@ -1,6 +1,7 @@ #pragma once #include +#include #include @@ -21,30 +22,6 @@ using ExpressionActionsPtr = std::shared_ptr; class Arena; -class IWindowFunction; - -// Runtime data for computing one window function. -struct WindowFunctionWorkspace -{ - AggregateFunctionPtr aggregate_function; - - // Cached value of aggregate function isState virtual method - bool is_aggregate_function_state = false; - - // This field is set for pure window functions. When set, we ignore the - // window_function.aggregate_function, and work through this interface - // instead. - IWindowFunction * window_function_impl = nullptr; - - std::vector argument_column_indices; - - // Will not be initialized for a pure window function. - mutable AlignedBuffer aggregate_function_state; - - // Argument columns. Be careful, this is a per-block cache. - std::vector argument_columns; - UInt64 cached_block_number = std::numeric_limits::max(); -}; struct WindowTransformBlock { diff --git a/src/Processors/tests/gtest_full_sorting_join.cpp b/src/Processors/tests/gtest_full_sorting_join.cpp index f678d7984e8..befe5e28b5d 100644 --- a/src/Processors/tests/gtest_full_sorting_join.cpp +++ b/src/Processors/tests/gtest_full_sorting_join.cpp @@ -208,6 +208,12 @@ Block executePipeline(QueryPipeline && pipeline) template void assertColumnVectorEq(const typename ColumnVector::Container & expected, const Block & block, const std::string & name) { + if (expected.empty()) + { + ASSERT_TRUE(block.columns() == 0); + return; + } + 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(); @@ -230,6 +236,12 @@ void assertColumnVectorEq(const typename ColumnVector::Container & expected, template void assertColumnEq(const IColumn & expected, const Block & block, const std::string & name) { + if (expected.empty()) + { + ASSERT_TRUE(block.columns() == 0); + return; + } + const ColumnPtr & actual = block.getByName(name).column; ASSERT_TRUE(checkColumn(*actual)); ASSERT_TRUE(checkColumn(expected)); diff --git a/src/QueryPipeline/ProfileInfo.cpp b/src/QueryPipeline/ProfileInfo.cpp index ee0ff8c69bf..69575939edc 100644 --- a/src/QueryPipeline/ProfileInfo.cpp +++ b/src/QueryPipeline/ProfileInfo.cpp @@ -1,14 +1,14 @@ #include +#include +#include #include #include -#include - namespace DB { -void ProfileInfo::read(ReadBuffer & in) +void ProfileInfo::read(ReadBuffer & in, UInt64 server_revision) { readVarUInt(rows, in); readVarUInt(blocks, in); @@ -16,10 +16,15 @@ void ProfileInfo::read(ReadBuffer & in) readBinary(applied_limit, in); readVarUInt(rows_before_limit, in); readBinary(calculated_rows_before_limit, in); + if (server_revision >= DBMS_MIN_REVISION_WITH_ROWS_BEFORE_AGGREGATION) + { + readBinary(applied_aggregation, in); + readVarUInt(rows_before_aggregation, in); + } } -void ProfileInfo::write(WriteBuffer & out) const +void ProfileInfo::write(WriteBuffer & out, UInt64 client_revision) const { writeVarUInt(rows, out); writeVarUInt(blocks, out); @@ -27,6 +32,11 @@ void ProfileInfo::write(WriteBuffer & out) const writeBinary(hasAppliedLimit(), out); writeVarUInt(getRowsBeforeLimit(), out); writeBinary(calculated_rows_before_limit, out); + if (client_revision >= DBMS_MIN_REVISION_WITH_ROWS_BEFORE_AGGREGATION) + { + writeBinary(hasAppliedAggregation(), out); + writeVarUInt(getRowsBeforeAggregation(), out); + } } @@ -41,6 +51,8 @@ void ProfileInfo::setFrom(const ProfileInfo & rhs, bool skip_block_size_info) applied_limit = rhs.applied_limit; rows_before_limit = rhs.rows_before_limit; calculated_rows_before_limit = rhs.calculated_rows_before_limit; + applied_aggregation = rhs.applied_aggregation; + rows_before_aggregation = rhs.rows_before_aggregation; } @@ -57,6 +69,17 @@ bool ProfileInfo::hasAppliedLimit() const return applied_limit; } +size_t ProfileInfo::getRowsBeforeAggregation() const +{ + return rows_before_aggregation; +} + + +bool ProfileInfo::hasAppliedAggregation() const +{ + return applied_aggregation; +} + void ProfileInfo::update(Block & block) { diff --git a/src/QueryPipeline/ProfileInfo.h b/src/QueryPipeline/ProfileInfo.h index 7a0a0c304e2..92c83c8c3be 100644 --- a/src/QueryPipeline/ProfileInfo.h +++ b/src/QueryPipeline/ProfileInfo.h @@ -32,13 +32,16 @@ struct ProfileInfo size_t getRowsBeforeLimit() const; bool hasAppliedLimit() const; + size_t getRowsBeforeAggregation() const; + bool hasAppliedAggregation() const; + void update(Block & block); void update(size_t num_rows, size_t num_bytes); /// Binary serialization and deserialization of main fields. /// Writes only main fields i.e. fields that required by internal transmission protocol. - void read(ReadBuffer & in); - void write(WriteBuffer & out) const; + void read(ReadBuffer & in, UInt64 server_revision); + void write(WriteBuffer & out, UInt64 client_revision) const; /// Sets main fields from other object (see methods above). /// If skip_block_size_info if true, then rows, bytes and block fields are ignored. @@ -51,11 +54,21 @@ struct ProfileInfo rows_before_limit = rows_before_limit_; } + /// Only for Processors. + void setRowsBeforeAggregation(size_t rows_before_aggregation_) + { + applied_aggregation = true; + rows_before_aggregation = rows_before_aggregation_; + } + private: /// For these fields we make accessors, because they must be calculated beforehand. mutable bool applied_limit = false; /// Whether LIMIT was applied mutable size_t rows_before_limit = 0; - mutable bool calculated_rows_before_limit = false; /// Whether the field rows_before_limit was calculated + mutable bool calculated_rows_before_limit = false; /// Whether the field rows was calculated + + mutable bool applied_aggregation = false; /// Whether GROUP BY was applied + mutable size_t rows_before_aggregation = 0; }; } diff --git a/src/QueryPipeline/QueryPipeline.cpp b/src/QueryPipeline/QueryPipeline.cpp index 935c006c217..c9c0bad7553 100644 --- a/src/QueryPipeline/QueryPipeline.cpp +++ b/src/QueryPipeline/QueryPipeline.cpp @@ -1,15 +1,14 @@ #include #include -#include -#include -#include -#include +#include #include #include -#include -#include -#include +#include +#include +#include +#include +#include #include #include #include @@ -17,15 +16,19 @@ #include #include #include -#include +#include +#include #include +#include #include #include #include #include -#include #include -#include +#include +#include +#include +#include namespace DB @@ -139,7 +142,7 @@ static void checkCompleted(Processors & processors) static void initRowsBeforeLimit(IOutputFormat * output_format) { - RowsBeforeLimitCounterPtr rows_before_limit_at_least; + RowsBeforeStepCounterPtr rows_before_limit_at_least; std::vector processors; std::map> limit_candidates; std::unordered_set visited; @@ -261,7 +264,7 @@ static void initRowsBeforeLimit(IOutputFormat * output_format) if (!processors.empty()) { - rows_before_limit_at_least = std::make_shared(); + rows_before_limit_at_least = std::make_shared(); for (auto & processor : processors) processor->setRowsBeforeLimitCounter(rows_before_limit_at_least); @@ -273,7 +276,28 @@ static void initRowsBeforeLimit(IOutputFormat * output_format) output_format->setRowsBeforeLimitCounter(rows_before_limit_at_least); } } +static void initRowsBeforeAggregation(std::shared_ptr processors, IOutputFormat * output_format) +{ + bool has_aggregation = false; + if (!processors->empty()) + { + RowsBeforeStepCounterPtr rows_before_aggregation = std::make_shared(); + for (const auto & processor : *processors) + { + if (typeid_cast(processor.get()) || typeid_cast(processor.get())) + { + processor->setRowsBeforeAggregationCounter(rows_before_aggregation); + has_aggregation = true; + } + if (typeid_cast(processor.get()) || typeid_cast(processor.get())) + processor->setRowsBeforeAggregationCounter(rows_before_aggregation); + } + if (has_aggregation) + rows_before_aggregation->add(0); + output_format->setRowsBeforeAggregationCounter(rows_before_aggregation); + } +} QueryPipeline::QueryPipeline( QueryPlanResourceHolder resources_, @@ -521,6 +545,14 @@ void QueryPipeline::complete(std::shared_ptr format) extremes = nullptr; initRowsBeforeLimit(format.get()); + for (const auto & context : resources.interpreter_context) + { + if (context->getSettingsRef().rows_before_aggregation) + { + initRowsBeforeAggregation(processors, format.get()); + break; + } + } output_format = format.get(); processors->emplace_back(std::move(format)); diff --git a/src/QueryPipeline/QueryPipelineBuilder.cpp b/src/QueryPipeline/QueryPipelineBuilder.cpp index 803d1686ad7..d276fed60a2 100644 --- a/src/QueryPipeline/QueryPipelineBuilder.cpp +++ b/src/QueryPipeline/QueryPipelineBuilder.cpp @@ -15,7 +15,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/src/QueryPipeline/RemoteQueryExecutor.cpp b/src/QueryPipeline/RemoteQueryExecutor.cpp index 61a512bcfc5..6f8b3931803 100644 --- a/src/QueryPipeline/RemoteQueryExecutor.cpp +++ b/src/QueryPipeline/RemoteQueryExecutor.cpp @@ -89,12 +89,12 @@ RemoteQueryExecutor::RemoteQueryExecutor( auto table_name = main_table.getQualifiedName(); ConnectionEstablisher connection_establisher(pool, &timeouts, current_settings, log, &table_name); - connection_establisher.run(result, fail_message); + connection_establisher.run(result, fail_message, /*force_connected=*/ true); } else { ConnectionEstablisher connection_establisher(pool, &timeouts, current_settings, log, nullptr); - connection_establisher.run(result, fail_message); + connection_establisher.run(result, fail_message, /*force_connected=*/ true); } std::vector connection_entries; diff --git a/src/QueryPipeline/tests/gtest_blocks_size_merging_streams.cpp b/src/QueryPipeline/tests/gtest_blocks_size_merging_streams.cpp index bc22f249f97..f41a447049c 100644 --- a/src/QueryPipeline/tests/gtest_blocks_size_merging_streams.cpp +++ b/src/QueryPipeline/tests/gtest_blocks_size_merging_streams.cpp @@ -83,7 +83,7 @@ TEST(MergingSortedTest, SimpleBlockSizeTest) EXPECT_EQ(pipe.numOutputPorts(), 3); auto transform = std::make_shared(pipe.getHeader(), pipe.numOutputPorts(), sort_description, - 8192, /*max_block_size_bytes=*/0, SortingQueueStrategy::Batch, 0, false, nullptr, false, true); + 8192, /*max_block_size_bytes=*/0, SortingQueueStrategy::Batch, 0, false, nullptr, true); pipe.addTransform(std::move(transform)); @@ -125,7 +125,7 @@ TEST(MergingSortedTest, MoreInterestingBlockSizes) EXPECT_EQ(pipe.numOutputPorts(), 3); auto transform = std::make_shared(pipe.getHeader(), pipe.numOutputPorts(), sort_description, - 8192, /*max_block_size_bytes=*/0, SortingQueueStrategy::Batch, 0, false, nullptr, false, true); + 8192, /*max_block_size_bytes=*/0, SortingQueueStrategy::Batch, 0, false, nullptr, true); pipe.addTransform(std::move(transform)); diff --git a/src/Server/GRPCServer.cpp b/src/Server/GRPCServer.cpp index d8a4d7f0e1f..9c8e0c6bf73 100644 --- a/src/Server/GRPCServer.cpp +++ b/src/Server/GRPCServer.cpp @@ -1577,6 +1577,8 @@ namespace stats.set_allocated_bytes(info.bytes); stats.set_applied_limit(info.hasAppliedLimit()); stats.set_rows_before_limit(info.getRowsBeforeLimit()); + stats.set_applied_aggregation(info.hasAppliedAggregation()); + stats.set_rows_before_aggregation(info.getRowsBeforeAggregation()); } void Call::addLogsToResult() diff --git a/src/Server/HTTP/HTTPServerConnection.cpp b/src/Server/HTTP/HTTPServerConnection.cpp index 047db014560..39e066005b9 100644 --- a/src/Server/HTTP/HTTPServerConnection.cpp +++ b/src/Server/HTTP/HTTPServerConnection.cpp @@ -2,6 +2,7 @@ #include #include +#include namespace DB { @@ -97,6 +98,21 @@ void HTTPServerConnection::run() { sendErrorResponse(session, Poco::Net::HTTPResponse::HTTP_BAD_REQUEST); } + catch (const Poco::Net::NetException & e) + { + /// Do not spam logs with messages related to connection reset by peer. + if (e.code() == POCO_ENOTCONN) + { + LOG_DEBUG(LogFrequencyLimiter(getLogger("HTTPServerConnection"), 10), "Connection reset by peer while processing HTTP request: {}", e.message()); + break; + } + + if (session.networkException()) + session.networkException()->rethrow(); + else + throw; + } + catch (const Poco::Exception &) { if (session.networkException()) diff --git a/src/Server/HTTP/HTTPServerResponse.h b/src/Server/HTTP/HTTPServerResponse.h index ac4f52e7766..51f5814556d 100644 --- a/src/Server/HTTP/HTTPServerResponse.h +++ b/src/Server/HTTP/HTTPServerResponse.h @@ -248,6 +248,8 @@ public: void attachRequest(HTTPServerRequest * request_) { request = request_; } + const Poco::Net::HTTPServerSession & getSession() const { return session; } + private: Poco::Net::HTTPServerSession & session; HTTPServerRequest * request = nullptr; diff --git a/src/Server/HTTP/WriteBufferFromHTTPServerResponse.cpp b/src/Server/HTTP/WriteBufferFromHTTPServerResponse.cpp index e2098b284bf..2fcb66ae606 100644 --- a/src/Server/HTTP/WriteBufferFromHTTPServerResponse.cpp +++ b/src/Server/HTTP/WriteBufferFromHTTPServerResponse.cpp @@ -30,7 +30,7 @@ void WriteBufferFromHTTPServerResponse::startSendHeaders() if (add_cors_header) response.set("Access-Control-Allow-Origin", "*"); - setResponseDefaultHeaders(response, keep_alive_timeout); + setResponseDefaultHeaders(response); std::stringstream header; //STYLE_CHECK_ALLOW_STD_STRING_STREAM response.beginWrite(header); @@ -119,12 +119,10 @@ void WriteBufferFromHTTPServerResponse::nextImpl() WriteBufferFromHTTPServerResponse::WriteBufferFromHTTPServerResponse( HTTPServerResponse & response_, bool is_http_method_head_, - UInt64 keep_alive_timeout_, const ProfileEvents::Event & write_event_) : HTTPWriteBuffer(response_.getSocket(), write_event_) , response(response_) , is_http_method_head(is_http_method_head_) - , keep_alive_timeout(keep_alive_timeout_) { } diff --git a/src/Server/HTTP/WriteBufferFromHTTPServerResponse.h b/src/Server/HTTP/WriteBufferFromHTTPServerResponse.h index a3952b7c553..f0c80f24582 100644 --- a/src/Server/HTTP/WriteBufferFromHTTPServerResponse.h +++ b/src/Server/HTTP/WriteBufferFromHTTPServerResponse.h @@ -29,7 +29,6 @@ public: WriteBufferFromHTTPServerResponse( HTTPServerResponse & response_, bool is_http_method_head_, - UInt64 keep_alive_timeout_, const ProfileEvents::Event & write_event_ = ProfileEvents::end()); ~WriteBufferFromHTTPServerResponse() override; @@ -91,7 +90,6 @@ private: bool is_http_method_head; bool add_cors_header = false; - size_t keep_alive_timeout = 0; bool initialized = false; diff --git a/src/Server/HTTP/checkHTTPHeader.cpp b/src/Server/HTTP/checkHTTPHeader.cpp new file mode 100644 index 00000000000..812adde022a --- /dev/null +++ b/src/Server/HTTP/checkHTTPHeader.cpp @@ -0,0 +1,22 @@ +#include + +#include + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int UNEXPECTED_HTTP_HEADERS; +} + +void checkHTTPHeader(const HTTPRequest & request, const String & header_name, const String & expected_value) +{ + if (!request.has(header_name)) + throw Exception(ErrorCodes::UNEXPECTED_HTTP_HEADERS, "No HTTP header {}", header_name); + if (request.get(header_name) != expected_value) + throw Exception(ErrorCodes::UNEXPECTED_HTTP_HEADERS, "HTTP header {} has unexpected value '{}' (instead of '{}')", header_name, request.get(header_name), expected_value); +} + +} diff --git a/src/Server/HTTP/checkHTTPHeader.h b/src/Server/HTTP/checkHTTPHeader.h new file mode 100644 index 00000000000..956599ae66b --- /dev/null +++ b/src/Server/HTTP/checkHTTPHeader.h @@ -0,0 +1,13 @@ +#pragma once + +#include +#include + + +namespace DB +{ + +/// Checks that the HTTP request has a specified header with a specified value. +void checkHTTPHeader(const HTTPRequest & request, const String & header_name, const String & expected_value); + +} diff --git a/src/Server/HTTP/sendExceptionToHTTPClient.cpp b/src/Server/HTTP/sendExceptionToHTTPClient.cpp index 022a763a9a2..07a649dc396 100644 --- a/src/Server/HTTP/sendExceptionToHTTPClient.cpp +++ b/src/Server/HTTP/sendExceptionToHTTPClient.cpp @@ -29,7 +29,7 @@ void sendExceptionToHTTPClient( if (!out) { /// If nothing was sent yet. - WriteBufferFromHTTPServerResponse out_for_message{response, request.getMethod() == HTTPRequest::HTTP_HEAD, DEFAULT_HTTP_KEEP_ALIVE_TIMEOUT}; + WriteBufferFromHTTPServerResponse out_for_message{response, request.getMethod() == HTTPRequest::HTTP_HEAD}; out_for_message.writeln(exception_message); out_for_message.finalize(); @@ -43,7 +43,6 @@ void sendExceptionToHTTPClient( out->position() = out->buffer().begin(); out->writeln(exception_message); - out->finalize(); } } diff --git a/src/Server/HTTPHandler.cpp b/src/Server/HTTPHandler.cpp index c8a58527f2c..d2bc22e98cc 100644 --- a/src/Server/HTTPHandler.cpp +++ b/src/Server/HTTPHandler.cpp @@ -266,7 +266,6 @@ void HTTPHandler::processQuery( std::make_shared( response, request.getMethod() == HTTPRequest::HTTP_HEAD, - context->getServerSettings().keep_alive_timeout.totalSeconds(), write_event); used_output.out = used_output.out_holder; used_output.out_maybe_compressed = used_output.out_holder; @@ -558,7 +557,7 @@ try if (!used_output.out_holder && !used_output.exception_is_written) { /// If nothing was sent yet and we don't even know if we must compress the response. - WriteBufferFromHTTPServerResponse(response, request.getMethod() == HTTPRequest::HTTP_HEAD, DEFAULT_HTTP_KEEP_ALIVE_TIMEOUT).writeln(s); + WriteBufferFromHTTPServerResponse(response, request.getMethod() == HTTPRequest::HTTP_HEAD).writeln(s); } else if (used_output.out_maybe_compressed) { diff --git a/src/Server/HTTPHandlerFactory.cpp b/src/Server/HTTPHandlerFactory.cpp index 5344b2d024b..fc31ad2874e 100644 --- a/src/Server/HTTPHandlerFactory.cpp +++ b/src/Server/HTTPHandlerFactory.cpp @@ -1,18 +1,16 @@ -#include #include #include +#include +#include #include -#include #include #include "HTTPHandler.h" -#include "Server/PrometheusMetricsWriter.h" #include "StaticRequestHandler.h" #include "ReplicasStatusHandler.h" #include "InterserverIOHTTPHandler.h" -#include "PrometheusRequestHandler.h" #include "WebUIRequestHandler.h" @@ -124,7 +122,8 @@ static inline auto createHandlersFactoryFromConfig( } else if (handler_type == "prometheus") { - main_handler_factory->addHandler(createPrometheusHandlerFactory(server, config, async_metrics, prefix + "." + key)); + main_handler_factory->addHandler( + createPrometheusHandlerFactoryForHTTPRule(server, config, prefix + "." + key, async_metrics)); } else if (handler_type == "replicas_status") { @@ -201,10 +200,7 @@ HTTPRequestHandlerFactoryPtr createHandlerFactory(IServer & server, const Poco:: else if (name == "InterserverIOHTTPHandler-factory" || name == "InterserverIOHTTPSHandler-factory") return createInterserverHTTPHandlerFactory(server, name); else if (name == "PrometheusHandler-factory") - { - auto metrics_writer = std::make_shared(config, "prometheus", async_metrics); - return createPrometheusMainHandlerFactory(server, config, metrics_writer, name); - } + return createPrometheusHandlerFactory(server, config, async_metrics, name); throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown HTTP handler factory name."); } @@ -291,20 +287,9 @@ void addDefaultHandlersFactory( ); factory.addHandler(query_handler); - /// We check that prometheus handler will be served on current (default) port. - /// Otherwise it will be created separately, see createHandlerFactory(...). - if (config.has("prometheus") && config.getInt("prometheus.port", 0) == 0) - { - auto writer = std::make_shared(config, "prometheus", async_metrics); - auto creator = [&server, writer] () -> std::unique_ptr - { - return std::make_unique(server, writer); - }; - auto prometheus_handler = std::make_shared>(std::move(creator)); - prometheus_handler->attachStrictPath(config.getString("prometheus.endpoint", "/metrics")); - prometheus_handler->allowGetAndHeadRequest(); + /// createPrometheusHandlerFactoryForHTTPRuleDefaults() can return nullptr if prometheus protocols must not be served on http port. + if (auto prometheus_handler = createPrometheusHandlerFactoryForHTTPRuleDefaults(server, config, async_metrics)) factory.addHandler(prometheus_handler); - } } } diff --git a/src/Server/HTTPHandlerFactory.h b/src/Server/HTTPHandlerFactory.h index b4c32366463..db4bb73cbc4 100644 --- a/src/Server/HTTPHandlerFactory.h +++ b/src/Server/HTTPHandlerFactory.h @@ -1,15 +1,12 @@ #pragma once -#include -#include #include #include #include #include -#include - #include + namespace DB { @@ -19,6 +16,7 @@ namespace ErrorCodes } class IServer; +class AsynchronousMetrics; template class HandlingRuleHTTPHandlerFactory : public HTTPRequestHandlerFactory @@ -126,18 +124,6 @@ HTTPRequestHandlerFactoryPtr createReplicasStatusHandlerFactory(IServer & server const Poco::Util::AbstractConfiguration & config, const std::string & config_prefix); -HTTPRequestHandlerFactoryPtr -createPrometheusHandlerFactory(IServer & server, - const Poco::Util::AbstractConfiguration & config, - AsynchronousMetrics & async_metrics, - const std::string & config_prefix); - -HTTPRequestHandlerFactoryPtr createPrometheusMainHandlerFactory( - IServer & server, - const Poco::Util::AbstractConfiguration & config, - PrometheusMetricsWriterPtr metrics_writer, - const std::string & name); - /// @param server - used in handlers to check IServer::isCancelled() /// @param config - not the same as server.config(), since it can be newer /// @param async_metrics - used for prometheus (in case of prometheus.asynchronous_metrics=true) diff --git a/src/Server/InterserverIOHTTPHandler.cpp b/src/Server/InterserverIOHTTPHandler.cpp index e46021c8e68..59852c79139 100644 --- a/src/Server/InterserverIOHTTPHandler.cpp +++ b/src/Server/InterserverIOHTTPHandler.cpp @@ -87,9 +87,8 @@ void InterserverIOHTTPHandler::handleRequest(HTTPServerRequest & request, HTTPSe response.setChunkedTransferEncoding(true); Output used_output; - const auto keep_alive_timeout = server.context()->getServerSettings().keep_alive_timeout.totalSeconds(); used_output.out = std::make_shared( - response, request.getMethod() == Poco::Net::HTTPRequest::HTTP_HEAD, keep_alive_timeout, write_event); + response, request.getMethod() == Poco::Net::HTTPRequest::HTTP_HEAD, write_event); auto finalize_output = [&] { diff --git a/src/Server/PrometheusMetricsWriter.cpp b/src/Server/PrometheusMetricsWriter.cpp index 85eafbe4808..43370116015 100644 --- a/src/Server/PrometheusMetricsWriter.cpp +++ b/src/Server/PrometheusMetricsWriter.cpp @@ -1,13 +1,27 @@ #include "PrometheusMetricsWriter.h" -#include +#include +#include #include #include - -#include +#include #include "config.h" + +#if USE_NURAFT +namespace ProfileEvents +{ + extern const std::vector keeper_profile_events; +} + +namespace CurrentMetrics +{ + extern const std::vector keeper_metrics; +} +#endif + + namespace { @@ -107,100 +121,84 @@ void writeAsyncMetrics(DB::WriteBuffer & wb, const DB::AsynchronousMetricValues } -#if USE_NURAFT -namespace ProfileEvents -{ - extern const std::vector keeper_profile_events; -} - -namespace CurrentMetrics -{ - extern const std::vector keeper_metrics; -} -#endif - namespace DB { -PrometheusMetricsWriter::PrometheusMetricsWriter( - const Poco::Util::AbstractConfiguration & config, const std::string & config_name, - const AsynchronousMetrics & async_metrics_) - : async_metrics(async_metrics_) - , send_events(config.getBool(config_name + ".events", true)) - , send_metrics(config.getBool(config_name + ".metrics", true)) - , send_asynchronous_metrics(config.getBool(config_name + ".asynchronous_metrics", true)) - , send_errors(config.getBool(config_name + ".errors", true)) +void PrometheusMetricsWriter::writeEvents(WriteBuffer & wb) const { + for (ProfileEvents::Event i = ProfileEvents::Event(0), end = ProfileEvents::end(); i < end; ++i) + writeEvent(wb, i); } -void PrometheusMetricsWriter::write(WriteBuffer & wb) const +void PrometheusMetricsWriter::writeMetrics(WriteBuffer & wb) const { - if (send_events) + for (size_t i = 0, end = CurrentMetrics::end(); i < end; ++i) + writeMetric(wb, i); +} + +void PrometheusMetricsWriter::writeAsynchronousMetrics(WriteBuffer & wb, const AsynchronousMetrics & async_metrics) const +{ + writeAsyncMetrics(wb, async_metrics.getValues()); +} + +void PrometheusMetricsWriter::writeErrors(WriteBuffer & wb) const +{ + size_t total_count = 0; + + for (size_t i = 0, end = ErrorCodes::end(); i < end; ++i) { - for (ProfileEvents::Event i = ProfileEvents::Event(0), end = ProfileEvents::end(); i < end; ++i) - writeEvent(wb, i); - } + const auto & error = ErrorCodes::values[i].get(); + std::string_view name = ErrorCodes::getName(static_cast(i)); - if (send_metrics) - { - for (size_t i = 0, end = CurrentMetrics::end(); i < end; ++i) - writeMetric(wb, i); - } + if (name.empty()) + continue; - if (send_asynchronous_metrics) - writeAsyncMetrics(wb, async_metrics.getValues()); + std::string key{error_metrics_prefix + toString(name)}; + std::string help = fmt::format("The number of {} errors since last server restart", name); - if (send_errors) - { - size_t total_count = 0; - - for (size_t i = 0, end = ErrorCodes::end(); i < end; ++i) - { - const auto & error = ErrorCodes::values[i].get(); - std::string_view name = ErrorCodes::getName(static_cast(i)); - - if (name.empty()) - continue; - - std::string key{error_metrics_prefix + toString(name)}; - std::string help = fmt::format("The number of {} errors since last server restart", name); - - writeOutLine(wb, "# HELP", key, help); - writeOutLine(wb, "# TYPE", key, "counter"); - /// We are interested in errors which are happened only on this server. - writeOutLine(wb, key, error.local.count); - - total_count += error.local.count; - } - - /// Write the total number of errors as a separate metric - std::string key{error_metrics_prefix + toString("ALL")}; - writeOutLine(wb, "# HELP", key, "The total number of errors since last server restart"); + writeOutLine(wb, "# HELP", key, help); writeOutLine(wb, "# TYPE", key, "counter"); - writeOutLine(wb, key, total_count); + /// We are interested in errors which are happened only on this server. + writeOutLine(wb, key, error.local.count); + + total_count += error.local.count; } + /// Write the total number of errors as a separate metric + std::string key{error_metrics_prefix + toString("ALL")}; + writeOutLine(wb, "# HELP", key, "The total number of errors since last server restart"); + writeOutLine(wb, "# TYPE", key, "counter"); + writeOutLine(wb, key, total_count); } -void KeeperPrometheusMetricsWriter::write([[maybe_unused]] WriteBuffer & wb) const + +void KeeperPrometheusMetricsWriter::writeEvents([[maybe_unused]] WriteBuffer & wb) const { #if USE_NURAFT - if (send_events) - { - for (auto event : ProfileEvents::keeper_profile_events) - writeEvent(wb, event); - } - - if (send_metrics) - { - for (auto metric : CurrentMetrics::keeper_metrics) - writeMetric(wb, metric); - } - - if (send_asynchronous_metrics) - writeAsyncMetrics(wb, async_metrics.getValues()); + for (auto event : ProfileEvents::keeper_profile_events) + writeEvent(wb, event); #endif } +void KeeperPrometheusMetricsWriter::writeMetrics([[maybe_unused]] WriteBuffer & wb) const +{ +#if USE_NURAFT + for (auto metric : CurrentMetrics::keeper_metrics) + writeMetric(wb, metric); +#endif +} + +void KeeperPrometheusMetricsWriter::writeAsynchronousMetrics([[maybe_unused]] WriteBuffer & wb, + [[maybe_unused]] const AsynchronousMetrics & async_metrics) const +{ +#if USE_NURAFT + writeAsyncMetrics(wb, async_metrics.getValues()); +#endif +} + +void KeeperPrometheusMetricsWriter::writeErrors(WriteBuffer &) const +{ +} + } diff --git a/src/Server/PrometheusMetricsWriter.h b/src/Server/PrometheusMetricsWriter.h index 933ad909ee0..cf2587d80b8 100644 --- a/src/Server/PrometheusMetricsWriter.h +++ b/src/Server/PrometheusMetricsWriter.h @@ -1,44 +1,33 @@ #pragma once -#include - -#include -#include -#include - -#include +#include namespace DB { +class AsynchronousMetrics; +class WriteBuffer; /// Write metrics in Prometheus format class PrometheusMetricsWriter { public: - PrometheusMetricsWriter( - const Poco::Util::AbstractConfiguration & config, const std::string & config_name, - const AsynchronousMetrics & async_metrics_); - - virtual void write(WriteBuffer & wb) const; - virtual ~PrometheusMetricsWriter() = default; -protected: - const AsynchronousMetrics & async_metrics; - const bool send_events; - const bool send_metrics; - const bool send_asynchronous_metrics; - const bool send_errors; + virtual void writeMetrics(WriteBuffer & wb) const; + virtual void writeAsynchronousMetrics(WriteBuffer & wb, const AsynchronousMetrics & async_metrics) const; + virtual void writeEvents(WriteBuffer & wb) const; + virtual void writeErrors(WriteBuffer & wb) const; }; + class KeeperPrometheusMetricsWriter : public PrometheusMetricsWriter { - using PrometheusMetricsWriter::PrometheusMetricsWriter; - - void write(WriteBuffer & wb) const override; +public: + void writeMetrics(WriteBuffer & wb) const override; + void writeAsynchronousMetrics(WriteBuffer & wb, const AsynchronousMetrics & async_metrics) const override; + void writeEvents(WriteBuffer & wb) const override; + void writeErrors(WriteBuffer & wb) const override; }; -using PrometheusMetricsWriterPtr = std::shared_ptr; - } diff --git a/src/Server/PrometheusRequestHandler.cpp b/src/Server/PrometheusRequestHandler.cpp index 87c106c3fc0..ae1fb6d629e 100644 --- a/src/Server/PrometheusRequestHandler.cpp +++ b/src/Server/PrometheusRequestHandler.cpp @@ -1,74 +1,447 @@ #include +#include +#include #include #include -#include +#include #include -#include -#include -#include -#include "Server/PrometheusMetricsWriter.h" +#include +#include "config.h" -#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include namespace DB { -void PrometheusRequestHandler::handleRequest(HTTPServerRequest & request, HTTPServerResponse & response, const ProfileEvents::Event & write_event) + +namespace ErrorCodes { + extern const int BAD_ARGUMENTS; + extern const int SUPPORT_IS_DISABLED; + extern const int LOGICAL_ERROR; +} + +/// Base implementation of a prometheus protocol. +class PrometheusRequestHandler::Impl +{ +public: + explicit Impl(PrometheusRequestHandler & parent) : parent_ref(parent) {} + virtual ~Impl() = default; + virtual void beforeHandlingRequest(HTTPServerRequest & /* request */) {} + virtual void handleRequest(HTTPServerRequest & request, HTTPServerResponse & response) = 0; + virtual void onException() {} + +protected: + PrometheusRequestHandler & parent() { return parent_ref; } + IServer & server() { return parent().server; } + const PrometheusRequestHandlerConfig & config() { return parent().config; } + PrometheusMetricsWriter & metrics_writer() { return *parent().metrics_writer; } + LoggerPtr log() { return parent().log; } + WriteBuffer & getOutputStream(HTTPServerResponse & response) { return parent().getOutputStream(response); } + +private: + PrometheusRequestHandler & parent_ref; +}; + + +/// Implementation of the exposing metrics protocol. +class PrometheusRequestHandler::ExposeMetricsImpl : public Impl +{ +public: + explicit ExposeMetricsImpl(PrometheusRequestHandler & parent) : Impl(parent) {} + + void beforeHandlingRequest(HTTPServerRequest & request) override + { + LOG_INFO(log(), "Handling metrics request from {}", request.get("User-Agent")); + chassert(config().type == PrometheusRequestHandlerConfig::Type::ExposeMetrics); + } + + void handleRequest(HTTPServerRequest & /* request */, HTTPServerResponse & response) override + { + response.setContentType("text/plain; version=0.0.4; charset=UTF-8"); + auto & out = getOutputStream(response); + + if (config().expose_events) + metrics_writer().writeEvents(out); + + if (config().expose_metrics) + metrics_writer().writeMetrics(out); + + if (config().expose_asynchronous_metrics) + metrics_writer().writeAsynchronousMetrics(out, parent().async_metrics); + + if (config().expose_errors) + metrics_writer().writeErrors(out); + } +}; + + +/// Base implementation of a protocol with Context and authentication. +class PrometheusRequestHandler::ImplWithContext : public Impl +{ +public: + explicit ImplWithContext(PrometheusRequestHandler & parent) : Impl(parent), default_settings(server().context()->getSettingsRef()) { } + + virtual void handlingRequestWithContext(HTTPServerRequest & request, HTTPServerResponse & response) = 0; + +protected: + void handleRequest(HTTPServerRequest & request, HTTPServerResponse & response) override + { + SCOPE_EXIT({ + request_credentials.reset(); + context.reset(); + session.reset(); + params.reset(); + }); + + params = std::make_unique(default_settings, request); + parent().send_stacktrace = config().is_stacktrace_enabled && params->getParsed("stacktrace", false); + + if (!authenticateUserAndMakeContext(request, response)) + return; /// The user is not authenticated yet, and the HTTP_UNAUTHORIZED response is sent with the "WWW-Authenticate" header, + /// and `request_credentials` must be preserved until the next request or until any exception. + + /// Initialize query scope. + std::optional query_scope; + if (context) + query_scope.emplace(context); + + handlingRequestWithContext(request, response); + } + + bool authenticateUserAndMakeContext(HTTPServerRequest & request, HTTPServerResponse & response) + { + session = std::make_unique(server().context(), ClientInfo::Interface::PROMETHEUS, request.isSecure()); + + if (!authenticateUser(request, response)) + return false; + + makeContext(request); + return true; + } + + bool authenticateUser(HTTPServerRequest & request, HTTPServerResponse & response) + { + return authenticateUserByHTTP(request, *params, response, *session, request_credentials, server().context(), log()); + } + + void makeContext(HTTPServerRequest & request) + { + context = session->makeQueryContext(); + + /// Anything else beside HTTP POST should be readonly queries. + setReadOnlyIfHTTPMethodIdempotent(context, request.getMethod()); + + auto roles = params->getAll("role"); + if (!roles.empty()) + context->setCurrentRoles(roles); + + /// Settings can be overridden in the URL query. + auto is_setting_like_parameter = [&] (const String & name) + { + /// Empty parameter appears when URL like ?&a=b or a=b&&c=d. Just skip them for user's convenience. + if (name.empty()) + return false; + + /// Some parameters (database, default_format, everything used in the code above) do not + /// belong to the Settings class. + static const NameSet reserved_param_names{"user", "password", "quota_key", "stacktrace", "role", "query_id"}; + return !reserved_param_names.contains(name); + }; + + SettingsChanges settings_changes; + for (const auto & [key, value] : *params) + { + if (is_setting_like_parameter(key)) + { + /// This query parameter should be considered as a ClickHouse setting. + settings_changes.push_back({key, value}); + } + } + + context->checkSettingsConstraints(settings_changes, SettingSource::QUERY); + context->applySettingsChanges(settings_changes); + + /// 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", ""))); + } + + void onException() override + { + // So that the next requests on the connection have to always start afresh in case of exceptions. + request_credentials.reset(); + } + + const Settings & default_settings; + std::unique_ptr params; + std::unique_ptr session; + std::unique_ptr request_credentials; + ContextMutablePtr context; +}; + + +/// Implementation of the remote-write protocol. +class PrometheusRequestHandler::RemoteWriteImpl : public ImplWithContext +{ +public: + using ImplWithContext::ImplWithContext; + + void beforeHandlingRequest(HTTPServerRequest & request) override + { + LOG_INFO(log(), "Handling remote write request from {}", request.get("User-Agent", "")); + chassert(config().type == PrometheusRequestHandlerConfig::Type::RemoteWrite); + } + + void handlingRequestWithContext([[maybe_unused]] HTTPServerRequest & request, [[maybe_unused]] HTTPServerResponse & response) override + { +#if USE_PROMETHEUS_PROTOBUFS + checkHTTPHeader(request, "Content-Type", "application/x-protobuf"); + checkHTTPHeader(request, "Content-Encoding", "snappy"); + + ProtobufZeroCopyInputStreamFromReadBuffer zero_copy_input_stream{ + std::make_unique(wrapReadBufferReference(request.getStream()))}; + + prometheus::WriteRequest write_request; + if (!write_request.ParsePartialFromZeroCopyStream(&zero_copy_input_stream)) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Cannot parse WriteRequest"); + + auto table = DatabaseCatalog::instance().getTable(StorageID{config().time_series_table_name}, context); + PrometheusRemoteWriteProtocol protocol{table, context}; + + if (write_request.timeseries_size()) + protocol.writeTimeSeries(write_request.timeseries()); + + if (write_request.metadata_size()) + protocol.writeMetricsMetadata(write_request.metadata()); + + response.setContentType("text/plain; charset=UTF-8"); + response.send(); + +#else + throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, "Prometheus remote write protocol is disabled"); +#endif + } +}; + +/// Implementation of the remote-read protocol. +class PrometheusRequestHandler::RemoteReadImpl : public ImplWithContext +{ +public: + using ImplWithContext::ImplWithContext; + + void beforeHandlingRequest(HTTPServerRequest & request) override + { + LOG_INFO(log(), "Handling remote read request from {}", request.get("User-Agent", "")); + chassert(config().type == PrometheusRequestHandlerConfig::Type::RemoteRead); + } + + void handlingRequestWithContext([[maybe_unused]] HTTPServerRequest & request, [[maybe_unused]] HTTPServerResponse & response) override + { +#if USE_PROMETHEUS_PROTOBUFS + checkHTTPHeader(request, "Content-Type", "application/x-protobuf"); + checkHTTPHeader(request, "Content-Encoding", "snappy"); + + auto table = DatabaseCatalog::instance().getTable(StorageID{config().time_series_table_name}, context); + PrometheusRemoteReadProtocol protocol{table, context}; + + ProtobufZeroCopyInputStreamFromReadBuffer zero_copy_input_stream{ + std::make_unique(wrapReadBufferReference(request.getStream()))}; + + prometheus::ReadRequest read_request; + if (!read_request.ParseFromZeroCopyStream(&zero_copy_input_stream)) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Cannot parse ReadRequest"); + + prometheus::ReadResponse read_response; + + size_t num_queries = read_request.queries_size(); + for (size_t i = 0; i != num_queries; ++i) + { + const auto & query = read_request.queries(static_cast(i)); + auto & new_query_result = *read_response.add_results(); + protocol.readTimeSeries( + *new_query_result.mutable_timeseries(), + query.start_timestamp_ms(), + query.end_timestamp_ms(), + query.matchers(), + query.hints()); + } + +# if 0 + LOG_DEBUG(log, "ReadResponse = {}", read_response.DebugString()); +# endif + + response.setContentType("application/x-protobuf"); + response.set("Content-Encoding", "snappy"); + + ProtobufZeroCopyOutputStreamFromWriteBuffer zero_copy_output_stream{std::make_unique(getOutputStream(response))}; + read_response.SerializeToZeroCopyStream(&zero_copy_output_stream); + zero_copy_output_stream.finalize(); + +#else + throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, "Prometheus remote read protocol is disabled"); +#endif + } +}; + + +PrometheusRequestHandler::PrometheusRequestHandler( + IServer & server_, + const PrometheusRequestHandlerConfig & config_, + const AsynchronousMetrics & async_metrics_, + std::shared_ptr metrics_writer_) + : server(server_) + , config(config_) + , async_metrics(async_metrics_) + , metrics_writer(metrics_writer_) + , log(getLogger("PrometheusRequestHandler")) +{ + createImpl(); +} + +PrometheusRequestHandler::~PrometheusRequestHandler() = default; + +void PrometheusRequestHandler::createImpl() +{ + switch (config.type) + { + case PrometheusRequestHandlerConfig::Type::ExposeMetrics: + { + impl = std::make_unique(*this); + return; + } + case PrometheusRequestHandlerConfig::Type::RemoteWrite: + { + impl = std::make_unique(*this); + return; + } + case PrometheusRequestHandlerConfig::Type::RemoteRead: + { + impl = std::make_unique(*this); + return; + } + } + UNREACHABLE(); +} + +void PrometheusRequestHandler::handleRequest(HTTPServerRequest & request, HTTPServerResponse & response, const ProfileEvents::Event & write_event_) +{ + setThreadName("PrometheusHndlr"); + try { - const auto & config = server.config(); - unsigned keep_alive_timeout = config.getUInt("keep_alive_timeout", DEFAULT_HTTP_KEEP_ALIVE_TIMEOUT); + response_finalized = false; + write_event = write_event_; + http_method = request.getMethod(); + chassert(!write_buffer_from_response); /// Nothing is written to the response yet. - /// In order to make keep-alive works. + /// Make keep-alive works. if (request.getVersion() == HTTPServerRequest::HTTP_1_1) response.setChunkedTransferEncoding(true); - setResponseDefaultHeaders(response, keep_alive_timeout); + setResponseDefaultHeaders(response); - response.setContentType("text/plain; version=0.0.4; charset=UTF-8"); + impl->beforeHandlingRequest(request); + impl->handleRequest(request, response); - WriteBufferFromHTTPServerResponse wb(response, request.getMethod() == Poco::Net::HTTPRequest::HTTP_HEAD, keep_alive_timeout, write_event); - metrics_writer->write(wb); - wb.finalize(); + finalizeResponse(response); } catch (...) { - tryLogCurrentException("PrometheusRequestHandler"); + tryLogCurrentException(log); + + ExecutionStatus status = ExecutionStatus::fromCurrentException("", send_stacktrace); + trySendExceptionToClient(status.message, status.code, request, response); + tryFinalizeResponse(response); + + tryCallOnException(); } } -HTTPRequestHandlerFactoryPtr createPrometheusHandlerFactory( - IServer & server, - const Poco::Util::AbstractConfiguration & config, - AsynchronousMetrics & async_metrics, - const std::string & config_prefix) +WriteBufferFromHTTPServerResponse & PrometheusRequestHandler::getOutputStream(HTTPServerResponse & response) { - auto writer = std::make_shared(config, config_prefix + ".handler", async_metrics); - auto creator = [&server, writer]() -> std::unique_ptr - { - return std::make_unique(server, writer); - }; - - auto factory = std::make_shared>(std::move(creator)); - factory->addFiltersFromConfig(config, config_prefix); - return factory; + if (response_finalized) + throw Exception(ErrorCodes::LOGICAL_ERROR, "PrometheusRequestHandler: Response already sent"); + if (write_buffer_from_response) + return *write_buffer_from_response; + write_buffer_from_response = std::make_unique( + response, http_method == HTTPRequest::HTTP_HEAD, write_event); + return *write_buffer_from_response; } -HTTPRequestHandlerFactoryPtr createPrometheusMainHandlerFactory( - IServer & server, const Poco::Util::AbstractConfiguration & config, PrometheusMetricsWriterPtr metrics_writer, const std::string & name) +void PrometheusRequestHandler::finalizeResponse(HTTPServerResponse & response) { - auto factory = std::make_shared(name); - auto creator = [&server, metrics_writer] + if (response_finalized) { - return std::make_unique(server, metrics_writer); - }; + /// Response is already finalized or at least tried to. We don't need the write buffer anymore in either case. + write_buffer_from_response = nullptr; + } + else + { + /// We set `response_finalized = true` before actually calling `write_buffer_from_response->finalize()` + /// because we shouldn't call finalize() again even if finalize() throws an exception. + response_finalized = true; - auto handler = std::make_shared>(std::move(creator)); - handler->attachStrictPath(config.getString("prometheus.endpoint", "/metrics")); - handler->allowGetAndHeadRequest(); - factory->addHandler(handler); - return factory; + if (write_buffer_from_response) + std::exchange(write_buffer_from_response, {})->finalize(); + else + WriteBufferFromHTTPServerResponse{response, http_method == HTTPRequest::HTTP_HEAD, write_event}.finalize(); + } + chassert(response_finalized && !write_buffer_from_response); } + +void PrometheusRequestHandler::trySendExceptionToClient(const String & exception_message, int exception_code, HTTPServerRequest & request, HTTPServerResponse & response) +{ + if (response_finalized) + return; /// Response is already finalized (or tried to). We can't write the error message to the response in either case. + + try + { + sendExceptionToHTTPClient(exception_message, exception_code, request, response, &getOutputStream(response), log); + } + catch (...) + { + tryLogCurrentException(log, "Couldn't send exception to client"); + } +} + +void PrometheusRequestHandler::tryFinalizeResponse(HTTPServerResponse & response) +{ + try + { + finalizeResponse(response); + } + catch (...) + { + tryLogCurrentException(log, "Cannot flush data to client (after sending exception)"); + } +} + +void PrometheusRequestHandler::tryCallOnException() +{ + try + { + if (impl) + impl->onException(); + } + catch (...) + { + tryLogCurrentException(log, "onException"); + } +} + } diff --git a/src/Server/PrometheusRequestHandler.h b/src/Server/PrometheusRequestHandler.h index d120752c8c5..281ecf5260e 100644 --- a/src/Server/PrometheusRequestHandler.h +++ b/src/Server/PrometheusRequestHandler.h @@ -1,28 +1,64 @@ #pragma once #include +#include -#include "PrometheusMetricsWriter.h" namespace DB { - +class AsynchronousMetrics; class IServer; +class PrometheusMetricsWriter; +class WriteBufferFromHTTPServerResponse; +/// Handles requests for prometheus protocols (expose_metrics, remote_write, remote-read). class PrometheusRequestHandler : public HTTPRequestHandler { -private: - IServer & server; - PrometheusMetricsWriterPtr metrics_writer; - public: - PrometheusRequestHandler(IServer & server_, PrometheusMetricsWriterPtr metrics_writer_) - : server(server_) - , metrics_writer(std::move(metrics_writer_)) - { - } + PrometheusRequestHandler( + IServer & server_, + const PrometheusRequestHandlerConfig & config_, + const AsynchronousMetrics & async_metrics_, + std::shared_ptr metrics_writer_); + ~PrometheusRequestHandler() override; - void handleRequest(HTTPServerRequest & request, HTTPServerResponse & response, const ProfileEvents::Event & write_event) override; + void handleRequest(HTTPServerRequest & request, HTTPServerResponse & response, const ProfileEvents::Event & write_event_) override; + +private: + /// Creates an internal implementation based on which PrometheusRequestHandlerConfig::Type is used. + void createImpl(); + + /// Returns the write buffer used for the current HTTP response. + WriteBufferFromHTTPServerResponse & getOutputStream(HTTPServerResponse & response); + + /// Finalizes the output stream and sends the response to the client. + void finalizeResponse(HTTPServerResponse & response); + void tryFinalizeResponse(HTTPServerResponse & response); + + /// Writes the current exception to the response. + void trySendExceptionToClient(const String & exception_message, int exception_code, HTTPServerRequest & request, HTTPServerResponse & response); + + /// Calls onException() in a try-catch block. + void tryCallOnException(); + + IServer & server; + const PrometheusRequestHandlerConfig config; + const AsynchronousMetrics & async_metrics; + const std::shared_ptr metrics_writer; + const LoggerPtr log; + + class Impl; + class ImplWithContext; + class ExposeMetricsImpl; + class RemoteWriteImpl; + class RemoteReadImpl; + std::unique_ptr impl; + + String http_method; + bool send_stacktrace = false; + std::unique_ptr write_buffer_from_response; + bool response_finalized = false; + ProfileEvents::Event write_event; }; } diff --git a/src/Server/PrometheusRequestHandlerConfig.h b/src/Server/PrometheusRequestHandlerConfig.h new file mode 100644 index 00000000000..d01d28f702c --- /dev/null +++ b/src/Server/PrometheusRequestHandlerConfig.h @@ -0,0 +1,39 @@ +#pragma once + +#include + + +namespace DB +{ + +/// Configuration of a Prometheus protocol handler after it's parsed from a configuration file. +struct PrometheusRequestHandlerConfig +{ + enum class Type + { + /// Exposes ClickHouse metrics for scraping by Prometheus. + ExposeMetrics, + + /// Handles Prometheus remote-write protocol. + RemoteWrite, + + /// Handles Prometheus remote-read protocol. + RemoteRead, + }; + + Type type = Type::ExposeMetrics; + + /// Settings for type ExposeMetrics: + bool expose_metrics = false; + bool expose_asynchronous_metrics = false; + bool expose_events = false; + bool expose_errors = false; + + /// Settings for types RemoteWrite, RemoteRead: + QualifiedTableName time_series_table_name; + + size_t keep_alive_timeout = 0; + bool is_stacktrace_enabled = true; +}; + +} diff --git a/src/Server/PrometheusRequestHandlerFactory.cpp b/src/Server/PrometheusRequestHandlerFactory.cpp new file mode 100644 index 00000000000..52f1d3b64c1 --- /dev/null +++ b/src/Server/PrometheusRequestHandlerFactory.cpp @@ -0,0 +1,242 @@ +#include + +#include +#include +#include +#include + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int UNKNOWN_ELEMENT_IN_CONFIG; +} + +namespace +{ + /// Parses common configuration which is attached to any other configuration. The common configuration looks like this: + /// + /// true + /// + /// 30 + void parseCommonConfig(const Poco::Util::AbstractConfiguration & config, PrometheusRequestHandlerConfig & res) + { + res.is_stacktrace_enabled = config.getBool("prometheus.enable_stacktrace", true); + res.keep_alive_timeout = config.getUInt("keep_alive_timeout", DEFAULT_HTTP_KEEP_ALIVE_TIMEOUT); + } + + /// Parses a configuration like this: + /// + /// true + /// true + /// true + /// true + PrometheusRequestHandlerConfig parseExposeMetricsConfig(const Poco::Util::AbstractConfiguration & config, const String & config_prefix) + { + PrometheusRequestHandlerConfig res; + res.type = PrometheusRequestHandlerConfig::Type::ExposeMetrics; + res.expose_metrics = config.getBool(config_prefix + ".metrics", true); + res.expose_asynchronous_metrics = config.getBool(config_prefix + ".asynchronous_metrics", true); + res.expose_events = config.getBool(config_prefix + ".events", true); + res.expose_errors = config.getBool(config_prefix + ".errors", true); + parseCommonConfig(config, res); + return res; + } + + /// Extracts a qualified table name from the config. It can be set either as + /// mydb.prometheus
+ /// or + /// mydb + /// prometheus
+ QualifiedTableName parseTableNameFromConfig(const Poco::Util::AbstractConfiguration & config, const String & config_prefix) + { + QualifiedTableName res; + res.table = config.getString(config_prefix + ".table", "prometheus"); + res.database = config.getString(config_prefix + ".database", ""); + if (res.database.empty()) + res = QualifiedTableName::parseFromString(res.table); + if (res.database.empty()) + res.database = "default"; + return res; + } + + /// Parses a configuration like this: + /// + /// db.time_series_table_name
+ PrometheusRequestHandlerConfig parseRemoteWriteConfig(const Poco::Util::AbstractConfiguration & config, const String & config_prefix) + { + PrometheusRequestHandlerConfig res; + res.type = PrometheusRequestHandlerConfig::Type::RemoteWrite; + res.time_series_table_name = parseTableNameFromConfig(config, config_prefix); + parseCommonConfig(config, res); + return res; + } + + /// Parses a configuration like this: + /// + /// db.time_series_table_name
+ PrometheusRequestHandlerConfig parseRemoteReadConfig(const Poco::Util::AbstractConfiguration & config, const String & config_prefix) + { + PrometheusRequestHandlerConfig res; + res.type = PrometheusRequestHandlerConfig::Type::RemoteRead; + res.time_series_table_name = parseTableNameFromConfig(config, config_prefix); + parseCommonConfig(config, res); + return res; + } + + /// Parses a configuration like this: + /// expose_metrics + /// true + /// true + /// true + /// true + /// -OR- + /// remote_write + /// db.time_series_table_name
+ PrometheusRequestHandlerConfig parseHandlerConfig(const Poco::Util::AbstractConfiguration & config, const String & config_prefix) + { + String type = config.getString(config_prefix + ".type"); + + if (type == "expose_metrics") + return parseExposeMetricsConfig(config, config_prefix); + else if (type == "remote_write") + return parseRemoteWriteConfig(config, config_prefix); + else if (type == "remote_read") + return parseRemoteReadConfig(config, config_prefix); + else + throw Exception(ErrorCodes::UNKNOWN_ELEMENT_IN_CONFIG, "Unknown type {} is specified in the configuration for a prometheus protocol", type); + } + + /// Returns true if the protocol represented by a passed config can be handled. + bool canBeHandled(const PrometheusRequestHandlerConfig & config, bool for_keeper) + { + /// The standalone ClickHouse Keeper can only expose its metrics. + /// It can't handle other Prometheus protocols. + return !for_keeper || (config.type == PrometheusRequestHandlerConfig::Type::ExposeMetrics); + } + + /// Creates a writer which serializes exposing metrics. + std::shared_ptr createPrometheusMetricWriter(bool for_keeper) + { + if (for_keeper) + return std::make_unique(); + else + return std::make_unique(); + } + + /// Base function for making a factory for PrometheusRequestHandler. This function can return nullptr. + std::shared_ptr> createPrometheusHandlerFactoryFromConfig( + IServer & server, + const AsynchronousMetrics & async_metrics, + const PrometheusRequestHandlerConfig & config, + bool for_keeper) + { + if (!canBeHandled(config, for_keeper)) + return nullptr; + auto metric_writer = createPrometheusMetricWriter(for_keeper); + auto creator = [&server, &async_metrics, config, metric_writer]() -> std::unique_ptr + { + return std::make_unique(server, config, async_metrics, metric_writer); + }; + return std::make_shared>(std::move(creator)); + } + + /// Generic function for createPrometheusHandlerFactory() and createKeeperPrometheusHandlerFactory(). + HTTPRequestHandlerFactoryPtr createPrometheusHandlerFactoryImpl( + IServer & server, + const Poco::Util::AbstractConfiguration & config, + const AsynchronousMetrics & asynchronous_metrics, + const String & name, + bool for_keeper) + { + auto factory = std::make_shared(name); + + if (config.has("prometheus.handlers")) + { + Strings keys; + config.keys("prometheus.handlers", keys); + for (const String & key : keys) + { + String prefix = "prometheus.handlers." + key; + auto parsed_config = parseHandlerConfig(config, prefix + ".handler"); + if (auto handler = createPrometheusHandlerFactoryFromConfig(server, asynchronous_metrics, parsed_config, for_keeper)) + { + handler->addFiltersFromConfig(config, prefix); + factory->addHandler(handler); + } + } + } + else + { + auto parsed_config = parseExposeMetricsConfig(config, "prometheus"); + if (auto handler = createPrometheusHandlerFactoryFromConfig(server, asynchronous_metrics, parsed_config, for_keeper)) + { + String endpoint = config.getString("prometheus.endpoint", "/metrics"); + handler->attachStrictPath(endpoint); + handler->allowGetAndHeadRequest(); + factory->addHandler(handler); + } + } + + return factory; + } + +} + + +HTTPRequestHandlerFactoryPtr createPrometheusHandlerFactory( + IServer & server, + const Poco::Util::AbstractConfiguration & config, + const AsynchronousMetrics & asynchronous_metrics, + const String & name) +{ + return createPrometheusHandlerFactoryImpl(server, config, asynchronous_metrics, name, /* for_keeper= */ false); +} + + +HTTPRequestHandlerFactoryPtr createPrometheusHandlerFactoryForHTTPRule( + IServer & server, + const Poco::Util::AbstractConfiguration & config, + const String & config_prefix, + const AsynchronousMetrics & asynchronous_metrics) +{ + auto parsed_config = parseExposeMetricsConfig(config, config_prefix + ".handler"); + auto handler = createPrometheusHandlerFactoryFromConfig(server, asynchronous_metrics, parsed_config, /* for_keeper= */ false); + chassert(handler); /// `handler` can't be nullptr here because `for_keeper` is false. + handler->addFiltersFromConfig(config, config_prefix); + return handler; +} + + +HTTPRequestHandlerFactoryPtr createPrometheusHandlerFactoryForHTTPRuleDefaults( + IServer & server, + const Poco::Util::AbstractConfiguration & config, + const AsynchronousMetrics & asynchronous_metrics) +{ + /// The "defaults" HTTP handler should serve the prometheus exposing metrics protocol on the http port + /// only if it isn't already served on its own port and if there is no section. + if (!config.has("prometheus") || config.getInt("prometheus.port", 0) || config.has("prometheus.handlers")) + return nullptr; + + auto parsed_config = parseExposeMetricsConfig(config, "prometheus"); + String endpoint = config.getString("prometheus.endpoint", "/metrics"); + auto handler = createPrometheusHandlerFactoryFromConfig(server, asynchronous_metrics, parsed_config, /* for_keeper= */ false); + chassert(handler); /// `handler` can't be nullptr here because `for_keeper` is false. + handler->attachStrictPath(endpoint); + handler->allowGetAndHeadRequest(); + return handler; +} + + +HTTPRequestHandlerFactoryPtr createKeeperPrometheusHandlerFactory( + IServer & server, + const Poco::Util::AbstractConfiguration & config, + const AsynchronousMetrics & asynchronous_metrics, + const String & name) +{ + return createPrometheusHandlerFactoryImpl(server, config, asynchronous_metrics, name, /* for_keeper= */ true); +} + +} diff --git a/src/Server/PrometheusRequestHandlerFactory.h b/src/Server/PrometheusRequestHandlerFactory.h new file mode 100644 index 00000000000..c52395ca93f --- /dev/null +++ b/src/Server/PrometheusRequestHandlerFactory.h @@ -0,0 +1,130 @@ +#pragma once + +#include +#include + + +namespace Poco::Util { class AbstractConfiguration; } + +namespace DB +{ + +class IServer; +class HTTPRequestHandlerFactory; +using HTTPRequestHandlerFactoryPtr = std::shared_ptr; +class AsynchronousMetrics; + +/// Makes a handler factory to handle prometheus protocols. +/// Expects a configuration like this: +/// +/// +/// 1234 +/// /metric +/// true +/// true +/// true +/// true +/// +/// +/// More prometheus protocols can be supported with using a different configuration +/// (which is similar to the section): +/// +/// +/// 1234 +/// +/// +/// /metrics +/// +/// expose_metrics +/// true +/// true +/// true +/// true +/// +/// +/// +/// +/// +/// An alternative port to serve prometheus protocols can be specified in the section: +/// +/// +/// +/// 4321 +/// prometheus +/// +/// +HTTPRequestHandlerFactoryPtr createPrometheusHandlerFactory( + IServer & server, + const Poco::Util::AbstractConfiguration & config, + const AsynchronousMetrics & asynchronous_metrics, + const String & name); + +/// Makes a HTTP handler factory to handle requests for prometheus metrics for a HTTP rule in the section. +/// Expects a configuration like this: +/// +/// 8123 +/// +/// +/// /metrics +/// +/// prometheus +/// true +/// true +/// true +/// true +/// +/// +/// +/// /write +/// +/// remote_write +/// db.time_series_table_name
+/// +/// +/// +/// /read +/// +/// remote_read +/// db.time_series_table_name
+/// +/// +/// +HTTPRequestHandlerFactoryPtr createPrometheusHandlerFactoryForHTTPRule( + IServer & server, + const Poco::Util::AbstractConfiguration & config, + const String & config_prefix, /// path to "http_handlers.my_handler_1" + const AsynchronousMetrics & asynchronous_metrics); + +/// Makes a HTTP Handler factory to handle requests for prometheus metrics as a part of the default HTTP rule in the section. +/// Expects a configuration like this: +/// +/// 8123 +/// +/// +/// +/// +/// /metric +/// true +/// true +/// true +/// true +/// +/// +/// The "defaults" HTTP handler should serve the prometheus exposing metrics protocol on the http port +/// only if it isn't already served on its own port , +/// and also if there is no section in the configuration +/// (because if that section exists then it must be in charge of how prometheus protocols are handled). +HTTPRequestHandlerFactoryPtr createPrometheusHandlerFactoryForHTTPRuleDefaults( + IServer & server, + const Poco::Util::AbstractConfiguration & config, + const AsynchronousMetrics & asynchronous_metrics); + +/// Makes a handler factory to handle prometheus protocols. +/// Supports the "expose_metrics" protocol only. +HTTPRequestHandlerFactoryPtr createKeeperPrometheusHandlerFactory( + IServer & server, + const Poco::Util::AbstractConfiguration & config, + const AsynchronousMetrics & asynchronous_metrics, + const String & name); + +} diff --git a/src/Server/ProtocolServerAdapter.cpp b/src/Server/ProtocolServerAdapter.cpp index b41ad2376f1..6b723bc8d87 100644 --- a/src/Server/ProtocolServerAdapter.cpp +++ b/src/Server/ProtocolServerAdapter.cpp @@ -20,6 +20,7 @@ public: UInt16 portNumber() const override { return tcp_server->portNumber(); } size_t currentConnections() const override { return tcp_server->currentConnections(); } size_t currentThreads() const override { return tcp_server->currentThreads(); } + size_t refusedConnections() const override { return tcp_server->refusedConnections(); } private: std::unique_ptr tcp_server; @@ -54,6 +55,7 @@ public: UInt16 portNumber() const override { return grpc_server->portNumber(); } size_t currentConnections() const override { return grpc_server->currentConnections(); } size_t currentThreads() const override { return grpc_server->currentThreads(); } + size_t refusedConnections() const override { return 0; } private: std::unique_ptr grpc_server; diff --git a/src/Server/ProtocolServerAdapter.h b/src/Server/ProtocolServerAdapter.h index 76a6776ed9c..4a0b0cae8e7 100644 --- a/src/Server/ProtocolServerAdapter.h +++ b/src/Server/ProtocolServerAdapter.h @@ -38,6 +38,8 @@ public: /// Returns the number of currently handled connections. size_t currentConnections() const { return impl->currentConnections(); } + size_t refusedConnections() const { return impl->refusedConnections(); } + /// Returns the number of current threads. size_t currentThreads() const { return impl->currentThreads(); } @@ -61,6 +63,7 @@ private: virtual UInt16 portNumber() const = 0; virtual size_t currentConnections() const = 0; virtual size_t currentThreads() const = 0; + virtual size_t refusedConnections() const = 0; }; class TCPServerAdapterImpl; class GRPCServerAdapterImpl; diff --git a/src/Server/ReplicasStatusHandler.cpp b/src/Server/ReplicasStatusHandler.cpp index f43357db0a8..419ad635d0d 100644 --- a/src/Server/ReplicasStatusHandler.cpp +++ b/src/Server/ReplicasStatusHandler.cpp @@ -89,8 +89,7 @@ void ReplicasStatusHandler::handleRequest(HTTPServerRequest & request, HTTPServe } } - const auto & server_settings = getContext()->getServerSettings(); - setResponseDefaultHeaders(response, server_settings.keep_alive_timeout.totalSeconds()); + setResponseDefaultHeaders(response); if (!ok) { diff --git a/src/Server/StaticRequestHandler.cpp b/src/Server/StaticRequestHandler.cpp index f3981dea9fb..d8c0765bca4 100644 --- a/src/Server/StaticRequestHandler.cpp +++ b/src/Server/StaticRequestHandler.cpp @@ -35,10 +35,9 @@ namespace ErrorCodes extern const int INVALID_CONFIG_PARAMETER; } -static inline std::unique_ptr -responseWriteBuffer(HTTPServerRequest & request, HTTPServerResponse & response, UInt64 keep_alive_timeout) +static inline std::unique_ptr responseWriteBuffer(HTTPServerRequest & request, HTTPServerResponse & response) { - auto buf = std::unique_ptr(new WriteBufferFromHTTPServerResponse(response, request.getMethod() == HTTPRequest::HTTP_HEAD, keep_alive_timeout)); + auto buf = std::unique_ptr(new WriteBufferFromHTTPServerResponse(response, request.getMethod() == HTTPRequest::HTTP_HEAD)); /// The client can pass a HTTP header indicating supported compression method (gzip or deflate). String http_response_compression_methods = request.get("Accept-Encoding", ""); @@ -91,8 +90,7 @@ static inline void trySendExceptionToClient( void StaticRequestHandler::handleRequest(HTTPServerRequest & request, HTTPServerResponse & response, const ProfileEvents::Event & /*write_event*/) { - auto keep_alive_timeout = server.context()->getServerSettings().keep_alive_timeout.totalSeconds(); - auto out = responseWriteBuffer(request, response, keep_alive_timeout); + auto out = responseWriteBuffer(request, response); try { @@ -107,7 +105,7 @@ void StaticRequestHandler::handleRequest(HTTPServerRequest & request, HTTPServer "The Transfer-Encoding is not chunked and there " "is no Content-Length header for POST request"); - setResponseDefaultHeaders(response, keep_alive_timeout); + setResponseDefaultHeaders(response); response.setStatusAndReason(Poco::Net::HTTPResponse::HTTPStatus(status)); writeResponse(*out); } diff --git a/src/Server/TCPHandler.cpp b/src/Server/TCPHandler.cpp index c5dfe3e6e5f..2b9a7295198 100644 --- a/src/Server/TCPHandler.cpp +++ b/src/Server/TCPHandler.cpp @@ -103,6 +103,7 @@ namespace DB::ErrorCodes extern const int SUPPORT_IS_DISABLED; extern const int UNSUPPORTED_METHOD; extern const int USER_EXPIRED; + extern const int NETWORK_ERROR; } namespace @@ -254,8 +255,8 @@ void TCPHandler::runImpl() socket().setSendTimeout(send_timeout); socket().setNoDelay(true); - in = std::make_shared(socket(), read_event); - out = std::make_shared(socket(), write_event); + in = std::make_shared(socket(), read_event); + out = std::make_shared(socket(), write_event); /// Support for PROXY protocol if (parse_proxy_protocol && !receiveProxyHeader()) @@ -280,6 +281,48 @@ void TCPHandler::runImpl() if (client_tcp_protocol_version >= DBMS_MIN_PROTOCOL_VERSION_WITH_ADDENDUM) receiveAddendum(); + { + /// Server side of chunked protocol negotiation. + /// Server advertises its protocol capabilities (separate for send and receive channels) by sending + /// in its 'Hello' response one of four types - chunked, notchunked, chunked_optional, notchunked_optional. + /// Not optional types are strict meaning that server only supports this type, optional means that + /// server prefer this type but capable to work in opposite. + /// Client selects which type it is going to communicate based on the settings from config or arguments, + /// and sends either "chunked" or "notchunked" protocol request in addendum section of handshake. + /// Client can detect if server's protocol capabilities are not compatible with client's settings (for example + /// server strictly requires chunked protocol but client's settings only allows notchunked protocol) - in such case + /// client should interrupt this connection. However if client continues with incompatible protocol type request, server + /// will send appropriate exception and disconnect client. + + auto is_chunked = [](const String & chunked_srv_str, const String & chunked_cl_str, const String & direction) + { + bool chunked_srv = chunked_srv_str.starts_with("chunked"); + bool optional_srv = chunked_srv_str.ends_with("_optional"); + bool chunked_cl = chunked_cl_str.starts_with("chunked"); + + if (optional_srv) + return chunked_cl; + + if (chunked_cl != chunked_srv) + throw NetException( + ErrorCodes::NETWORK_ERROR, + "Incompatible protocol: {} is {}, client requested {}", + direction, + chunked_srv ? "chunked" : "notchunked", + chunked_cl ? "chunked" : "notchunked"); + + return chunked_srv; + }; + + bool out_chunked = is_chunked(server.config().getString("proto_caps.send", "notchunked"), proto_recv_chunked_cl, "send"); + bool in_chunked = is_chunked(server.config().getString("proto_caps.recv", "notchunked"), proto_send_chunked_cl, "recv"); + + if (out_chunked) + out->enableChunked(); + if (in_chunked) + in->enableChunked(); + } + if (!is_interserver_mode) { /// If session created, then settings in session context has been updated. @@ -321,7 +364,7 @@ void TCPHandler::runImpl() { Stopwatch idle_time; UInt64 timeout_ms = std::min(poll_interval, idle_connection_timeout) * 1000000; - while (tcp_server.isOpen() && !server.isCancelled() && !static_cast(*in).poll(timeout_ms)) + while (tcp_server.isOpen() && !server.isCancelled() && !in->poll(timeout_ms)) { if (idle_time.elapsedSeconds() > idle_connection_timeout) { @@ -365,9 +408,6 @@ void TCPHandler::runImpl() try { - /// If a user passed query-local timeouts, reset socket to initial state at the end of the query - SCOPE_EXIT({state.timeout_setter.reset();}); - /** If Query - process it. If Ping or Cancel - go back to the beginning. * There may come settings for a separate query that modify `query_context`. * It's possible to receive part uuids packet before the query, so then receivePacket has to be called twice. @@ -397,7 +437,8 @@ void TCPHandler::runImpl() /// So it's better to update the connection settings for flexibility. extractConnectionSettingsFromContext(query_context); - /// Sync timeouts on client and server during current query to avoid dangling queries on server + /// Sync timeouts on client and server during current query to avoid dangling queries on server. + /// It should be reset at the end of query. state.timeout_setter = std::make_unique(socket(), send_timeout, receive_timeout); /// Should we send internal logs to client? @@ -558,7 +599,9 @@ void TCPHandler::runImpl() std::scoped_lock lock(out_mutex, task_callback_mutex); if (getQueryCancellationStatus() == CancellationStatus::FULLY_CANCELLED) + { return true; + } sendProgress(); sendSelectProfileEvents(); @@ -605,6 +648,7 @@ void TCPHandler::runImpl() /// QueryState should be cleared before QueryScope, since otherwise /// the MemoryTracker will be wrong for possible deallocations. /// (i.e. deallocations from the Aggregator with two-level aggregation) + /// Also it resets socket's timeouts. state.reset(); last_sent_snapshots = ProfileEvents::ThreadIdToCountersSnapshot{}; query_scope.reset(); @@ -630,6 +674,9 @@ void TCPHandler::runImpl() state.io.onException(); exception.reset(e.clone()); + /// In case of exception state was not reset, so socket's timouts must be reset explicitly + state.timeout_setter.reset(); + if (e.code() == ErrorCodes::UNKNOWN_PACKET_FROM_CLIENT) throw; @@ -688,6 +735,9 @@ void TCPHandler::runImpl() exception = std::make_unique(Exception(ErrorCodes::UNKNOWN_EXCEPTION, "Unknown exception")); } + /// In case of exception state was not reset, so socket's timouts must be reset explicitly + state.timeout_setter.reset(); + try { if (exception) @@ -789,7 +839,7 @@ bool TCPHandler::readDataNext() /// We are waiting for a packet from the client. Thus, every `POLL_INTERVAL` seconds check whether we need to shut down. while (true) { - if (static_cast(*in).poll(timeout_us)) + if (in->poll(timeout_us)) { /// If client disconnected. if (in->eof()) @@ -1179,6 +1229,8 @@ void TCPHandler::processTablesStatusRequest() } response.write(*out, client_tcp_protocol_version); + + out->finishChunk(); } void TCPHandler::receiveUnexpectedTablesStatusRequest() @@ -1199,6 +1251,8 @@ void TCPHandler::sendPartUUIDs() writeVarUInt(Protocol::Server::PartUUIDs, *out); writeVectorBinary(uuids, *out); + + out->finishChunk(); out->next(); } } @@ -1207,6 +1261,8 @@ void TCPHandler::sendPartUUIDs() void TCPHandler::sendReadTaskRequestAssumeLocked() { writeVarUInt(Protocol::Server::ReadTaskRequest, *out); + + out->finishChunk(); out->next(); } @@ -1215,6 +1271,8 @@ void TCPHandler::sendMergeTreeAllRangesAnnouncementAssumeLocked(InitialAllRanges { writeVarUInt(Protocol::Server::MergeTreeAllRangesAnnouncement, *out); announcement.serialize(*out); + + out->finishChunk(); out->next(); } @@ -1223,6 +1281,8 @@ void TCPHandler::sendMergeTreeReadTaskRequestAssumeLocked(ParallelReadRequest re { writeVarUInt(Protocol::Server::MergeTreeReadTaskRequest, *out); request.serialize(*out); + + out->finishChunk(); out->next(); } @@ -1230,7 +1290,9 @@ void TCPHandler::sendMergeTreeReadTaskRequestAssumeLocked(ParallelReadRequest re void TCPHandler::sendProfileInfo(const ProfileInfo & info) { writeVarUInt(Protocol::Server::ProfileInfo, *out); - info.write(*out); + info.write(*out, client_tcp_protocol_version); + + out->finishChunk(); out->next(); } @@ -1246,6 +1308,8 @@ void TCPHandler::sendTotals(const Block & totals) state.block_out->write(totals); state.maybe_compressed_out->next(); + + out->finishChunk(); out->next(); } } @@ -1262,6 +1326,8 @@ void TCPHandler::sendExtremes(const Block & extremes) state.block_out->write(extremes); state.maybe_compressed_out->next(); + + out->finishChunk(); out->next(); } } @@ -1279,6 +1345,8 @@ void TCPHandler::sendProfileEvents() writeStringBinary("", *out); state.profile_events_block_out->write(block); + + out->finishChunk(); out->next(); auto elapsed_milliseconds = stopwatch.elapsedMilliseconds(); @@ -1316,6 +1384,8 @@ void TCPHandler::sendTimezone() LOG_DEBUG(log, "TCPHandler::sendTimezone(): {}", tz); writeVarUInt(Protocol::Server::TimezoneUpdate, *out); writeStringBinary(tz, *out); + + out->finishChunk(); out->next(); } @@ -1576,6 +1646,12 @@ void TCPHandler::receiveAddendum() if (!is_interserver_mode) session->setQuotaClientKey(quota_key); + + if (client_tcp_protocol_version >= DBMS_MIN_PROTOCOL_VERSION_WITH_CHUNKED_PACKETS) + { + readStringBinary(proto_send_chunked_cl, *in); + readStringBinary(proto_recv_chunked_cl, *in); + } } @@ -1609,6 +1685,11 @@ void TCPHandler::sendHello() writeStringBinary(server_display_name, *out); if (client_tcp_protocol_version >= DBMS_MIN_REVISION_WITH_VERSION_PATCH) writeVarUInt(VERSION_PATCH, *out); + if (client_tcp_protocol_version >= DBMS_MIN_PROTOCOL_VERSION_WITH_CHUNKED_PACKETS) + { + writeStringBinary(server.config().getString("proto_caps.send", "notchunked"), *out); + writeStringBinary(server.config().getString("proto_caps.recv", "notchunked"), *out); + } if (client_tcp_protocol_version >= DBMS_MIN_PROTOCOL_VERSION_WITH_PASSWORD_COMPLEXITY_RULES) { auto rules = server.context()->getAccessControl().getPasswordComplexityRules(); @@ -1661,6 +1742,7 @@ bool TCPHandler::receivePacket() case Protocol::Client::Ping: writeVarUInt(Protocol::Server::Pong, *out); + out->finishChunk(); out->next(); return false; @@ -2190,7 +2272,7 @@ QueryState::CancellationStatus TCPHandler::getQueryCancellationStatus() after_check_cancelled.restart(); /// During request execution the only packet that can come from the client is stopping the query. - if (static_cast(*in).poll(0)) + if (in->poll(0)) { if (in->eof()) { @@ -2241,19 +2323,33 @@ void TCPHandler::sendData(const Block & block) } writeVarUInt(Protocol::Server::Data, *out); - /// Send external table name (empty name is the main table) - writeStringBinary("", *out); /// For testing hedged requests if (block.rows() > 0 && query_context->getSettingsRef().sleep_in_send_data_ms.totalMilliseconds()) { + /// This strange sequence is needed in case of chunked protocol is enabled, in order for client not to + /// hang on receiving of at least packet type - chunk will not be processed unless either chunk footer + /// or chunk continuation header is received - first 'next' is sending starting chunk containing packet type + /// and second 'next' is sending chunk continuation header. + out->next(); + /// Send external table name (empty name is the main table) + writeStringBinary("", *out); out->next(); std::chrono::milliseconds ms(query_context->getSettingsRef().sleep_in_send_data_ms.totalMilliseconds()); std::this_thread::sleep_for(ms); } + else + { + /// Send external table name (empty name is the main table) + writeStringBinary("", *out); + } state.block_out->write(block); - state.maybe_compressed_out->next(); + + if (state.maybe_compressed_out != out) + state.maybe_compressed_out->next(); + + out->finishChunk(); out->next(); } catch (...) @@ -2289,6 +2385,8 @@ void TCPHandler::sendLogData(const Block & block) writeStringBinary("", *out); state.logs_block_out->write(block); + + out->finishChunk(); out->next(); } @@ -2300,6 +2398,7 @@ void TCPHandler::sendTableColumns(const ColumnsDescription & columns) writeStringBinary("", *out); writeStringBinary(columns.toString(), *out); + out->finishChunk(); out->next(); } @@ -2309,6 +2408,8 @@ void TCPHandler::sendException(const Exception & e, bool with_stack_trace) writeVarUInt(Protocol::Server::Exception, *out); writeException(e, *out, with_stack_trace); + + out->finishChunk(); out->next(); } @@ -2319,6 +2420,8 @@ void TCPHandler::sendEndOfStream() state.io.setAllDataSent(); writeVarUInt(Protocol::Server::EndOfStream, *out); + + out->finishChunk(); out->next(); } @@ -2337,6 +2440,8 @@ void TCPHandler::sendProgress() increment.elapsed_ns = current_elapsed_ns - state.prev_elapsed_ns; state.prev_elapsed_ns = current_elapsed_ns; increment.write(*out, client_tcp_protocol_version); + + out->finishChunk(); out->next(); } diff --git a/src/Server/TCPHandler.h b/src/Server/TCPHandler.h index 74afb5a14a5..dca40e98920 100644 --- a/src/Server/TCPHandler.h +++ b/src/Server/TCPHandler.h @@ -18,6 +18,8 @@ #include #include #include +#include +#include #include "Core/Types.h" #include "IServer.h" @@ -186,6 +188,8 @@ private: UInt64 client_version_minor = 0; UInt64 client_version_patch = 0; UInt32 client_tcp_protocol_version = 0; + String proto_send_chunked_cl = "notchunked"; + String proto_recv_chunked_cl = "notchunked"; String quota_key; /// Connection settings, which are extracted from a context. @@ -204,8 +208,8 @@ private: ClientInfo::QueryKind query_kind = ClientInfo::QueryKind::NO_QUERY; /// Streams for reading/writing from/to client connection socket. - std::shared_ptr in; - std::shared_ptr out; + std::shared_ptr in; + std::shared_ptr out; ProfileEvents::Event read_event; ProfileEvents::Event write_event; diff --git a/src/Server/WebUIRequestHandler.cpp b/src/Server/WebUIRequestHandler.cpp index a3d098014e7..c04d7a3f2a0 100644 --- a/src/Server/WebUIRequestHandler.cpp +++ b/src/Server/WebUIRequestHandler.cpp @@ -30,23 +30,20 @@ DashboardWebUIRequestHandler::DashboardWebUIRequestHandler(IServer & server_) : BinaryWebUIRequestHandler::BinaryWebUIRequestHandler(IServer & server_) : server(server_) {} JavaScriptWebUIRequestHandler::JavaScriptWebUIRequestHandler(IServer & server_) : server(server_) {} -static void handle(const IServer & server, HTTPServerRequest & request, HTTPServerResponse & response, std::string_view html) +static void handle(HTTPServerRequest & request, HTTPServerResponse & response, std::string_view html) { - auto keep_alive_timeout = server.context()->getServerSettings().keep_alive_timeout.totalSeconds(); - response.setContentType("text/html; charset=UTF-8"); if (request.getVersion() == HTTPServerRequest::HTTP_1_1) response.setChunkedTransferEncoding(true); - setResponseDefaultHeaders(response, keep_alive_timeout); + setResponseDefaultHeaders(response); response.setStatusAndReason(Poco::Net::HTTPResponse::HTTP_OK); - WriteBufferFromHTTPServerResponse(response, request.getMethod() == HTTPRequest::HTTP_HEAD, keep_alive_timeout).write(html.data(), html.size()); - + WriteBufferFromHTTPServerResponse(response, request.getMethod() == HTTPRequest::HTTP_HEAD).write(html.data(), html.size()); } void PlayWebUIRequestHandler::handleRequest(HTTPServerRequest & request, HTTPServerResponse & response, const ProfileEvents::Event &) { - handle(server, request, response, {reinterpret_cast(gresource_play_htmlData), gresource_play_htmlSize}); + handle(request, response, {reinterpret_cast(gresource_play_htmlData), gresource_play_htmlSize}); } void DashboardWebUIRequestHandler::handleRequest(HTTPServerRequest & request, HTTPServerResponse & response, const ProfileEvents::Event &) @@ -64,23 +61,23 @@ void DashboardWebUIRequestHandler::handleRequest(HTTPServerRequest & request, HT static re2::RE2 lz_string_url = R"(https://[^\s"'`]+lz-string[^\s"'`]*\.js)"; RE2::Replace(&html, lz_string_url, "/js/lz-string.js"); - handle(server, request, response, html); + handle(request, response, html); } void BinaryWebUIRequestHandler::handleRequest(HTTPServerRequest & request, HTTPServerResponse & response, const ProfileEvents::Event &) { - handle(server, request, response, {reinterpret_cast(gresource_binary_htmlData), gresource_binary_htmlSize}); + handle(request, response, {reinterpret_cast(gresource_binary_htmlData), gresource_binary_htmlSize}); } void JavaScriptWebUIRequestHandler::handleRequest(HTTPServerRequest & request, HTTPServerResponse & response, const ProfileEvents::Event &) { if (request.getURI() == "/js/uplot.js") { - handle(server, request, response, {reinterpret_cast(gresource_uplot_jsData), gresource_uplot_jsSize}); + handle(request, response, {reinterpret_cast(gresource_uplot_jsData), gresource_uplot_jsSize}); } else if (request.getURI() == "/js/lz-string.js") { - handle(server, request, response, {reinterpret_cast(gresource_lz_string_jsData), gresource_lz_string_jsSize}); + handle(request, response, {reinterpret_cast(gresource_lz_string_jsData), gresource_lz_string_jsSize}); } else { @@ -88,7 +85,7 @@ void JavaScriptWebUIRequestHandler::handleRequest(HTTPServerRequest & request, H *response.send() << "Not found.\n"; } - handle(server, request, response, {reinterpret_cast(gresource_binary_htmlData), gresource_binary_htmlSize}); + handle(request, response, {reinterpret_cast(gresource_binary_htmlData), gresource_binary_htmlSize}); } } diff --git a/src/Server/grpc_protos/clickhouse_grpc.proto b/src/Server/grpc_protos/clickhouse_grpc.proto index dc17570f833..7836e88f2af 100644 --- a/src/Server/grpc_protos/clickhouse_grpc.proto +++ b/src/Server/grpc_protos/clickhouse_grpc.proto @@ -180,6 +180,8 @@ message Stats { uint64 allocated_bytes = 3; bool applied_limit = 4; uint64 rows_before_limit = 5; + bool applied_aggregation = 6; + uint64 rows_before_aggregation = 7; } message Exception { diff --git a/src/Storages/AlterCommands.cpp b/src/Storages/AlterCommands.cpp index 7891042bb96..d5780e32db3 100644 --- a/src/Storages/AlterCommands.cpp +++ b/src/Storages/AlterCommands.cpp @@ -19,6 +19,7 @@ #include #include #include +#include #include #include #include @@ -109,7 +110,7 @@ std::optional AlterCommand::parse(const ASTAlterCommand * command_ if (ast_col_decl.comment) { const auto & ast_comment = typeid_cast(*ast_col_decl.comment); - command.comment = ast_comment.value.get(); + command.comment = ast_comment.value.safeGet(); } if (ast_col_decl.codec) @@ -167,7 +168,7 @@ std::optional AlterCommand::parse(const ASTAlterCommand * command_ if (ast_col_decl.comment) { const auto & ast_comment = ast_col_decl.comment->as(); - command.comment.emplace(ast_comment.value.get()); + command.comment.emplace(ast_comment.value.safeGet()); } if (ast_col_decl.ttl) @@ -210,7 +211,7 @@ std::optional AlterCommand::parse(const ASTAlterCommand * command_ command.type = COMMENT_COLUMN; command.column_name = getIdentifierName(command_ast->column); const auto & ast_comment = command_ast->comment->as(); - command.comment = ast_comment.value.get(); + command.comment = ast_comment.value.safeGet(); command.if_exists = command_ast->if_exists; return command; } @@ -220,7 +221,7 @@ std::optional AlterCommand::parse(const ASTAlterCommand * command_ command.ast = command_ast->clone(); command.type = COMMENT_TABLE; const auto & ast_comment = command_ast->comment->as(); - command.comment = ast_comment.value.get(); + command.comment = ast_comment.value.safeGet(); return command; } else if (command_ast->type == ASTAlterCommand::MODIFY_ORDER_BY) @@ -1316,6 +1317,8 @@ void AlterCommands::validate(const StoragePtr & table, ContextPtr context) const throw Exception(ErrorCodes::BAD_ARGUMENTS, "Data type have to be specified for column {} to add", backQuote(column_name)); + validateDataType(command.data_type, DataTypeValidationSettings(context->getSettingsRef())); + /// FIXME: Adding a new column of type Object(JSON) is broken. /// Looks like there is something around default expression for this column (method `getDefault` is not implemented for the data type Object). /// But after ALTER TABLE ADD COLUMN we need to fill existing rows with something (exactly the default value). @@ -1395,6 +1398,8 @@ void AlterCommands::validate(const StoragePtr & table, ContextPtr context) const /// So we don't allow to do it for now. if (command.data_type) { + validateDataType(command.data_type, DataTypeValidationSettings(context->getSettingsRef())); + const GetColumnsOptions options(GetColumnsOptions::All); const auto old_data_type = all_columns.getColumn(options, column_name).type; diff --git a/src/Storages/Cache/ExternalDataSourceCache.cpp b/src/Storages/Cache/ExternalDataSourceCache.cpp index cffb1dc9ca3..8c778fd511a 100644 --- a/src/Storages/Cache/ExternalDataSourceCache.cpp +++ b/src/Storages/Cache/ExternalDataSourceCache.cpp @@ -57,8 +57,15 @@ LocalFileHolder::~LocalFileHolder() { if (original_readbuffer) { - assert_cast(original_readbuffer.get())->seek(0, SEEK_SET); - file_cache_controller->value().startBackgroundDownload(std::move(original_readbuffer), *thread_pool); + try + { + assert_cast(original_readbuffer.get())->seek(0, SEEK_SET); + file_cache_controller->value().startBackgroundDownload(std::move(original_readbuffer), *thread_pool); + } + catch (...) + { + tryLogCurrentException(getLogger("LocalFileHolder"), "Exception during destructor of LocalFileHolder."); + } } } diff --git a/src/Storages/ColumnsDescription.cpp b/src/Storages/ColumnsDescription.cpp index da749812167..0d724245b49 100644 --- a/src/Storages/ColumnsDescription.cpp +++ b/src/Storages/ColumnsDescription.cpp @@ -197,7 +197,7 @@ void ColumnDescription::readText(ReadBuffer & buf) } if (col_ast->comment) - comment = col_ast->comment->as().value.get(); + comment = col_ast->comment->as().value.safeGet(); if (col_ast->codec) codec = CompressionCodecFactory::instance().validateCodecAndGetPreprocessedAST(col_ast->codec, type, false, true, true, true); diff --git a/src/Storages/Distributed/DistributedAsyncInsertDirectoryQueue.cpp b/src/Storages/Distributed/DistributedAsyncInsertDirectoryQueue.cpp index fdb4cfcb371..7616b384860 100644 --- a/src/Storages/Distributed/DistributedAsyncInsertDirectoryQueue.cpp +++ b/src/Storages/Distributed/DistributedAsyncInsertDirectoryQueue.cpp @@ -273,6 +273,8 @@ ConnectionPoolWithFailoverPtr DistributedAsyncInsertDirectoryQueue::createPool(c address.default_database, address.user, address.password, + address.proto_send_chunked, + address.proto_recv_chunked, address.quota_key, address.cluster, address.cluster_secret, diff --git a/src/Storages/Distributed/DistributedSink.cpp b/src/Storages/Distributed/DistributedSink.cpp index e0baefd5838..e3e73e42096 100644 --- a/src/Storages/Distributed/DistributedSink.cpp +++ b/src/Storages/Distributed/DistributedSink.cpp @@ -607,7 +607,7 @@ void DistributedSink::onFinish() } } -void DistributedSink::onCancel() +void DistributedSink::onCancel() noexcept { std::lock_guard lock(execution_mutex); if (pool && !pool->finished()) @@ -618,14 +618,26 @@ void DistributedSink::onCancel() } catch (...) { - tryLogCurrentException(storage.log); + tryLogCurrentException(storage.log, "Error occurs on cancellation."); } } for (auto & shard_jobs : per_shard_jobs) + { for (JobReplica & job : shard_jobs.replicas_jobs) - if (job.executor) - job.executor->cancel(); + { + try + { + if (job.executor) + job.executor->cancel(); + } + catch (...) + { + tryLogCurrentException(storage.log, "Error occurs on cancellation."); + } + } + } + } diff --git a/src/Storages/Distributed/DistributedSink.h b/src/Storages/Distributed/DistributedSink.h index 5b7396f2c6f..33e580e00fa 100644 --- a/src/Storages/Distributed/DistributedSink.h +++ b/src/Storages/Distributed/DistributedSink.h @@ -53,7 +53,7 @@ public: void onFinish() override; private: - void onCancel() override; + void onCancel() noexcept override; IColumn::Selector createSelector(const Block & source_block) const; diff --git a/src/Storages/Hive/StorageHive.cpp b/src/Storages/Hive/StorageHive.cpp index 089fb5c585c..ae2e8cffe28 100644 --- a/src/Storages/Hive/StorageHive.cpp +++ b/src/Storages/Hive/StorageHive.cpp @@ -445,7 +445,7 @@ StorageHive::StorageHive( storage_metadata.partition_key = KeyDescription::getKeyFromAST(partition_by_ast, storage_metadata.columns, getContext()); setInMemoryMetadata(storage_metadata); - setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(storage_metadata.getColumns())); + setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(storage_metadata.getColumns(), getContext())); } void StorageHive::lazyInitialize() diff --git a/src/Storages/IndicesDescription.cpp b/src/Storages/IndicesDescription.cpp index cef8fd85f97..753fbf1d635 100644 --- a/src/Storages/IndicesDescription.cpp +++ b/src/Storages/IndicesDescription.cpp @@ -3,6 +3,7 @@ #include #include +#include #include #include #include @@ -130,10 +131,15 @@ IndexDescription IndexDescription::getIndexFromAST(const ASTPtr & definition_ast { for (size_t i = 0; i < index_type->arguments->children.size(); ++i) { - const auto * argument = index_type->arguments->children[i]->as(); - if (!argument) + const auto & child = index_type->arguments->children[i]; + if (const auto * ast_literal = child->as(); ast_literal != nullptr) + /// E.g. INDEX index_name column_name TYPE vector_similarity('hnsw', 'f32') + result.arguments.emplace_back(ast_literal->value); + else if (const auto * ast_identifier = child->as(); ast_identifier != nullptr) + /// E.g. INDEX index_name column_name TYPE vector_similarity(hnsw, f32) + result.arguments.emplace_back(ast_identifier->name()); + else throw Exception(ErrorCodes::INCORRECT_QUERY, "Only literals can be skip index arguments"); - result.arguments.emplace_back(argument->value); } } diff --git a/src/Storages/Kafka/KafkaConfigLoader.cpp b/src/Storages/Kafka/KafkaConfigLoader.cpp new file mode 100644 index 00000000000..df6ccec4b7f --- /dev/null +++ b/src/Storages/Kafka/KafkaConfigLoader.cpp @@ -0,0 +1,480 @@ +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace CurrentMetrics +{ +extern const Metric KafkaLibrdkafkaThreads; +} + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +} + +template +struct KafkaInterceptors +{ + static rd_kafka_resp_err_t rdKafkaOnThreadStart(rd_kafka_t *, rd_kafka_thread_type_t thread_type, const char *, void * ctx); + + static rd_kafka_resp_err_t rdKafkaOnThreadExit(rd_kafka_t *, rd_kafka_thread_type_t, const char *, void * ctx); + + static rd_kafka_resp_err_t + rdKafkaOnNew(rd_kafka_t * rk, const rd_kafka_conf_t *, void * ctx, char * /*errstr*/, size_t /*errstr_size*/); + + static rd_kafka_resp_err_t rdKafkaOnConfDup( + rd_kafka_conf_t * new_conf, const rd_kafka_conf_t * /*old_conf*/, size_t /*filter_cnt*/, const char ** /*filter*/, void * ctx); +}; + +template +rd_kafka_resp_err_t +KafkaInterceptors::rdKafkaOnThreadStart(rd_kafka_t *, rd_kafka_thread_type_t thread_type, const char *, void * ctx) +{ + TStorageKafka * self = reinterpret_cast(ctx); + CurrentMetrics::add(CurrentMetrics::KafkaLibrdkafkaThreads, 1); + + const auto & storage_id = self->getStorageID(); + const auto & table = storage_id.getTableName(); + + switch (thread_type) + { + case RD_KAFKA_THREAD_MAIN: + setThreadName(("rdk:m/" + table.substr(0, 9)).c_str()); + break; + case RD_KAFKA_THREAD_BACKGROUND: + setThreadName(("rdk:bg/" + table.substr(0, 8)).c_str()); + break; + case RD_KAFKA_THREAD_BROKER: + setThreadName(("rdk:b/" + table.substr(0, 9)).c_str()); + break; + } + + /// Create ThreadStatus to track memory allocations from librdkafka threads. + // + /// And store them in a separate list (thread_statuses) to make sure that they will be destroyed, + /// regardless how librdkafka calls the hooks. + /// But this can trigger use-after-free if librdkafka will not destroy threads after rd_kafka_wait_destroyed() + auto thread_status = std::make_shared(); + std::lock_guard lock(self->thread_statuses_mutex); + self->thread_statuses.emplace_back(std::move(thread_status)); + + return RD_KAFKA_RESP_ERR_NO_ERROR; +} + +template +rd_kafka_resp_err_t KafkaInterceptors::rdKafkaOnThreadExit(rd_kafka_t *, rd_kafka_thread_type_t, const char *, void * ctx) +{ + TStorageKafka * self = reinterpret_cast(ctx); + CurrentMetrics::sub(CurrentMetrics::KafkaLibrdkafkaThreads, 1); + + std::lock_guard lock(self->thread_statuses_mutex); + const auto it = std::find_if( + self->thread_statuses.begin(), + self->thread_statuses.end(), + [](const auto & thread_status_ptr) { return thread_status_ptr.get() == current_thread; }); + if (it == self->thread_statuses.end()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "No thread status for this librdkafka thread."); + + self->thread_statuses.erase(it); + + return RD_KAFKA_RESP_ERR_NO_ERROR; +} + +template +rd_kafka_resp_err_t KafkaInterceptors::rdKafkaOnNew( + rd_kafka_t * rk, const rd_kafka_conf_t *, void * ctx, char * /*errstr*/, size_t /*errstr_size*/) +{ + TStorageKafka * self = reinterpret_cast(ctx); + rd_kafka_resp_err_t status; + + status = rd_kafka_interceptor_add_on_thread_start(rk, "init-thread", rdKafkaOnThreadStart, ctx); + if (status != RD_KAFKA_RESP_ERR_NO_ERROR) + { + LOG_ERROR(self->log, "Cannot set on thread start interceptor due to {} error", status); + return status; + } + + status = rd_kafka_interceptor_add_on_thread_exit(rk, "exit-thread", rdKafkaOnThreadExit, ctx); + if (status != RD_KAFKA_RESP_ERR_NO_ERROR) + LOG_ERROR(self->log, "Cannot set on thread exit interceptor due to {} error", status); + + return status; +} + +template +rd_kafka_resp_err_t KafkaInterceptors::rdKafkaOnConfDup( + rd_kafka_conf_t * new_conf, const rd_kafka_conf_t * /*old_conf*/, size_t /*filter_cnt*/, const char ** /*filter*/, void * ctx) +{ + TStorageKafka * self = reinterpret_cast(ctx); + rd_kafka_resp_err_t status; + + // cppkafka copies configuration multiple times + status = rd_kafka_conf_interceptor_add_on_conf_dup(new_conf, "init", rdKafkaOnConfDup, ctx); + if (status != RD_KAFKA_RESP_ERR_NO_ERROR) + { + LOG_ERROR(self->log, "Cannot set on conf dup interceptor due to {} error", status); + return status; + } + + status = rd_kafka_conf_interceptor_add_on_new(new_conf, "init", rdKafkaOnNew, ctx); + if (status != RD_KAFKA_RESP_ERR_NO_ERROR) + LOG_ERROR(self->log, "Cannot set on conf new interceptor due to {} error", status); + + return status; +} + +template struct KafkaInterceptors; +template struct KafkaInterceptors; + +namespace +{ + +void setKafkaConfigValue(cppkafka::Configuration & kafka_config, const String & key, const String & value) +{ + /// "log_level" has valid underscore, the remaining librdkafka setting use dot.separated.format which isn't acceptable for XML. + /// See https://github.com/edenhill/librdkafka/blob/master/CONFIGURATION.md + const String setting_name_in_kafka_config = (key == "log_level") ? key : boost::replace_all_copy(key, "_", "."); + kafka_config.set(setting_name_in_kafka_config, value); +} + +void loadConfigProperty( + cppkafka::Configuration & kafka_config, + const Poco::Util::AbstractConfiguration & config, + const String & config_prefix, + const String & tag) +{ + const String property_path = config_prefix + "." + tag; + const String property_value = config.getString(property_path); + + setKafkaConfigValue(kafka_config, tag, property_value); +} + +void loadNamedCollectionConfig(cppkafka::Configuration & kafka_config, const String & collection_name, const String & config_prefix) +{ + const auto & collection = NamedCollectionFactory::instance().get(collection_name); + for (const auto & key : collection->getKeys(-1, config_prefix)) + { + // Cut prefix with '.' before actual config tag. + const auto param_name = key.substr(config_prefix.size() + 1); + setKafkaConfigValue(kafka_config, param_name, collection->get(key)); + } +} + +void loadLegacyTopicConfig( + cppkafka::Configuration & kafka_config, + const Poco::Util::AbstractConfiguration & config, + const String & collection_name, + const String & config_prefix) +{ + if (!collection_name.empty()) + { + loadNamedCollectionConfig(kafka_config, collection_name, config_prefix); + return; + } + + Poco::Util::AbstractConfiguration::Keys tags; + config.keys(config_prefix, tags); + + for (const auto & tag : tags) + { + loadConfigProperty(kafka_config, config, config_prefix, tag); + } +} + +/// Read server configuration into cppkafa configuration, used by new per-topic configuration +void loadTopicConfig( + cppkafka::Configuration & kafka_config, + const Poco::Util::AbstractConfiguration & config, + const String & collection_name, + const String & config_prefix, + const String & topic) +{ + if (!collection_name.empty()) + { + const auto topic_prefix = fmt::format("{}.{}", config_prefix, KafkaConfigLoader::CONFIG_KAFKA_TOPIC_TAG); + const auto & collection = NamedCollectionFactory::instance().get(collection_name); + for (const auto & key : collection->getKeys(1, config_prefix)) + { + /// Only consider key . Multiple occurrences given as "kafka_topic", "kafka_topic[1]", etc. + if (!key.starts_with(topic_prefix)) + continue; + + const String kafka_topic_path = config_prefix + "." + key; + const String kafka_topic_name_path = kafka_topic_path + "." + KafkaConfigLoader::CONFIG_NAME_TAG; + if (topic == collection->get(kafka_topic_name_path)) + /// Found it! Now read the per-topic configuration into cppkafka. + loadNamedCollectionConfig(kafka_config, collection_name, kafka_topic_path); + } + } + else + { + /// Read all tags one level below + Poco::Util::AbstractConfiguration::Keys tags; + config.keys(config_prefix, tags); + + for (const auto & tag : tags) + { + if (tag == KafkaConfigLoader::CONFIG_NAME_TAG) + continue; // ignore , it is used to match topic configurations + loadConfigProperty(kafka_config, config, config_prefix, tag); + } + } +} + +/// Read server configuration into cppkafka configuration, used by global configuration and by legacy per-topic configuration +void loadFromConfig( + cppkafka::Configuration & kafka_config, const KafkaConfigLoader::LoadConfigParams & params, const String & config_prefix) +{ + if (!params.collection_name.empty()) + { + loadNamedCollectionConfig(kafka_config, params.collection_name, config_prefix); + return; + } + + /// Read all tags one level below + Poco::Util::AbstractConfiguration::Keys tags; + params.config.keys(config_prefix, tags); + + for (const auto & tag : tags) + { + if (tag == KafkaConfigLoader::CONFIG_KAFKA_PRODUCER_TAG || tag == KafkaConfigLoader::CONFIG_KAFKA_CONSUMER_TAG) + /// Do not load consumer/producer properties, since they should be separated by different configuration objects. + continue; + + if (tag.starts_with( + KafkaConfigLoader::CONFIG_KAFKA_TOPIC_TAG)) /// multiple occurrences given as "kafka_topic", "kafka_topic[1]", etc. + { + // Update consumer topic-specific configuration (new syntax). Example with topics "football" and "baseball": + // + // + // football + // 250 + // 5000 + // + // + // baseball + // 300 + // 2000 + // + // + // Advantages: The period restriction no longer applies (e.g. sports.football will work), everything + // Kafka-related is below . + for (const auto & topic : params.topics) + { + /// Read topic name between ... + const String kafka_topic_path = config_prefix + "." + tag; + const String kafka_topic_name_path = kafka_topic_path + "." + KafkaConfigLoader::CONFIG_NAME_TAG; + const String topic_name = params.config.getString(kafka_topic_name_path); + + if (topic_name != topic) + continue; + loadTopicConfig(kafka_config, params.config, params.collection_name, kafka_topic_path, topic); + } + continue; + } + if (tag.starts_with(KafkaConfigLoader::CONFIG_KAFKA_TAG)) + /// skip legacy configuration per topic e.g. . + /// it will be processed is a separate function + continue; + // Update configuration from the configuration. Example: + // + // 250 + // 100000 + // + loadConfigProperty(kafka_config, params.config, config_prefix, tag); + } +} + +void loadLegacyConfigSyntax( + cppkafka::Configuration & kafka_config, + const Poco::Util::AbstractConfiguration & config, + const String & collection_name, + const Names & topics) +{ + for (const auto & topic : topics) + { + const String kafka_topic_path = KafkaConfigLoader::CONFIG_KAFKA_TAG + "." + KafkaConfigLoader::CONFIG_KAFKA_TAG + "_" + topic; + loadLegacyTopicConfig(kafka_config, config, collection_name, kafka_topic_path); + } +} + +void loadConsumerConfig(cppkafka::Configuration & kafka_config, const KafkaConfigLoader::LoadConfigParams & params) +{ + const String consumer_path = KafkaConfigLoader::CONFIG_KAFKA_TAG + "." + KafkaConfigLoader::CONFIG_KAFKA_CONSUMER_TAG; + loadLegacyConfigSyntax(kafka_config, params.config, params.collection_name, params.topics); + // A new syntax has higher priority + loadFromConfig(kafka_config, params, consumer_path); +} + +void loadProducerConfig(cppkafka::Configuration & kafka_config, const KafkaConfigLoader::LoadConfigParams & params) +{ + const String producer_path = KafkaConfigLoader::CONFIG_KAFKA_TAG + "." + KafkaConfigLoader::CONFIG_KAFKA_PRODUCER_TAG; + loadLegacyConfigSyntax(kafka_config, params.config, params.collection_name, params.topics); + // A new syntax has higher priority + loadFromConfig(kafka_config, params, producer_path); +} + +template +void updateGlobalConfiguration( + cppkafka::Configuration & kafka_config, TKafkaStorage & storage, const KafkaConfigLoader::LoadConfigParams & params) +{ + loadFromConfig(kafka_config, params, KafkaConfigLoader::CONFIG_KAFKA_TAG); + +#if USE_KRB5 + if (kafka_config.has_property("sasl.kerberos.kinit.cmd")) + LOG_WARNING(params.log, "sasl.kerberos.kinit.cmd configuration parameter is ignored."); + + kafka_config.set("sasl.kerberos.kinit.cmd", ""); + kafka_config.set("sasl.kerberos.min.time.before.relogin", "0"); + + if (kafka_config.has_property("sasl.kerberos.keytab") && kafka_config.has_property("sasl.kerberos.principal")) + { + String keytab = kafka_config.get("sasl.kerberos.keytab"); + String principal = kafka_config.get("sasl.kerberos.principal"); + LOG_DEBUG(params.log, "Running KerberosInit"); + try + { + kerberosInit(keytab, principal); + } + catch (const Exception & e) + { + LOG_ERROR(params.log, "KerberosInit failure: {}", getExceptionMessage(e, false)); + } + LOG_DEBUG(params.log, "Finished KerberosInit"); + } +#else // USE_KRB5 + if (kafka_config.has_property("sasl.kerberos.keytab") || kafka_config.has_property("sasl.kerberos.principal")) + LOG_WARNING(params.log, "Ignoring Kerberos-related parameters because ClickHouse was built without krb5 library support."); +#endif // USE_KRB5 + // No need to add any prefix, messages can be distinguished + kafka_config.set_log_callback( + [log = params.log](cppkafka::KafkaHandleBase & handle, int level, const std::string & facility, const std::string & message) + { + auto [poco_level, client_logs_level] = parseSyslogLevel(level); + const auto & kafka_object_config = handle.get_configuration(); + const std::string client_id_key{"client.id"}; + chassert(kafka_object_config.has_property(client_id_key) && "Kafka configuration doesn't have expected client.id set"); + LOG_IMPL( + log, + client_logs_level, + poco_level, + "[client.id:{}] [rdk:{}] {}", + kafka_object_config.get(client_id_key), + facility, + message); + }); + + /// NOTE: statistics should be consumed, otherwise it creates too much + /// entries in the queue, that leads to memory leak and slow shutdown. + if (!kafka_config.has_property("statistics.interval.ms")) + { + // every 3 seconds by default. set to 0 to disable. + kafka_config.set("statistics.interval.ms", "3000"); + } + // Configure interceptor to change thread name + // + // TODO: add interceptors support into the cppkafka. + // XXX: rdkafka uses pthread_set_name_np(), but glibc-compatibility overrides it to noop. + { + // This should be safe, since we wait the rdkafka object anyway. + void * self = static_cast(&storage); + + int status; + + status + = rd_kafka_conf_interceptor_add_on_new(kafka_config.get_handle(), "init", KafkaInterceptors::rdKafkaOnNew, self); + if (status != RD_KAFKA_RESP_ERR_NO_ERROR) + LOG_ERROR(params.log, "Cannot set new interceptor due to {} error", status); + + // cppkafka always copy the configuration + status = rd_kafka_conf_interceptor_add_on_conf_dup( + kafka_config.get_handle(), "init", KafkaInterceptors::rdKafkaOnConfDup, self); + if (status != RD_KAFKA_RESP_ERR_NO_ERROR) + LOG_ERROR(params.log, "Cannot set dup conf interceptor due to {} error", status); + } +} + +} + +template +cppkafka::Configuration KafkaConfigLoader::getConsumerConfiguration(TKafkaStorage & storage, const ConsumerConfigParams & params) +{ + cppkafka::Configuration conf; + + conf.set("metadata.broker.list", params.brokers); + conf.set("group.id", params.group); + if (params.multiple_consumers) + conf.set("client.id", fmt::format("{}-{}", params.client_id, params.consumer_number)); + else + conf.set("client.id", params.client_id); + conf.set("client.software.name", VERSION_NAME); + conf.set("client.software.version", VERSION_DESCRIBE); + conf.set("auto.offset.reset", "earliest"); // If no offset stored for this group, read all messages from the start + + // that allows to prevent fast draining of the librdkafka queue + // during building of single insert block. Improves performance + // significantly, but may lead to bigger memory consumption. + size_t default_queued_min_messages = 100000; // must be greater than or equal to default + size_t max_allowed_queued_min_messages = 10000000; // must be less than or equal to max allowed value + conf.set( + "queued.min.messages", std::min(std::max(params.max_block_size, default_queued_min_messages), max_allowed_queued_min_messages)); + + updateGlobalConfiguration(conf, storage, params); + loadConsumerConfig(conf, params); + + // those settings should not be changed by users. + conf.set("enable.auto.commit", "false"); // We manually commit offsets after a stream successfully finished + conf.set("enable.auto.offset.store", "false"); // Update offset automatically - to commit them all at once. + conf.set("enable.partition.eof", "false"); // Ignore EOF messages + + for (auto & property : conf.get_all()) + { + LOG_TRACE(params.log, "Consumer set property {}:{}", property.first, property.second); + } + + return conf; +} + +template cppkafka::Configuration +KafkaConfigLoader::getConsumerConfiguration(StorageKafka & storage, const ConsumerConfigParams & params); +template cppkafka::Configuration +KafkaConfigLoader::getConsumerConfiguration(StorageKafka2 & storage, const ConsumerConfigParams & params); + +template +cppkafka::Configuration KafkaConfigLoader::getProducerConfiguration(TKafkaStorage & storage, const ProducerConfigParams & params) +{ + cppkafka::Configuration conf; + conf.set("metadata.broker.list", params.brokers); + conf.set("client.id", params.client_id); + conf.set("client.software.name", VERSION_NAME); + conf.set("client.software.version", VERSION_DESCRIBE); + + updateGlobalConfiguration(conf, storage, params); + loadProducerConfig(conf, params); + + for (auto & property : conf.get_all()) + { + LOG_TRACE(params.log, "Producer set property {}:{}", property.first, property.second); + } + + return conf; +} + +template cppkafka::Configuration +KafkaConfigLoader::getProducerConfiguration(StorageKafka & storage, const ProducerConfigParams & params); +template cppkafka::Configuration +KafkaConfigLoader::getProducerConfiguration(StorageKafka2 & storage, const ProducerConfigParams & params); + +} diff --git a/src/Storages/Kafka/KafkaConfigLoader.h b/src/Storages/Kafka/KafkaConfigLoader.h new file mode 100644 index 00000000000..f18683c17f0 --- /dev/null +++ b/src/Storages/Kafka/KafkaConfigLoader.h @@ -0,0 +1,54 @@ +#pragma once + +#include +#include +#include +#include +#include + +namespace DB +{ +struct KafkaSettings; +class VirtualColumnsDescription; + +struct KafkaConfigLoader +{ + static inline const String CONFIG_KAFKA_TAG = "kafka"; + static inline const String CONFIG_KAFKA_TOPIC_TAG = "kafka_topic"; + static inline const String CONFIG_NAME_TAG = "name"; + static inline const String CONFIG_KAFKA_CONSUMER_TAG = "consumer"; + static inline const String CONFIG_KAFKA_PRODUCER_TAG = "producer"; + using LogCallback = cppkafka::Configuration::LogCallback; + + + struct LoadConfigParams + { + const Poco::Util::AbstractConfiguration & config; + String & collection_name; + const Names & topics; + LoggerPtr & log; + }; + + struct ConsumerConfigParams : public LoadConfigParams + { + String brokers; + String group; + bool multiple_consumers; + size_t consumer_number; + String client_id; + size_t max_block_size; + }; + + struct ProducerConfigParams : public LoadConfigParams + { + String brokers; + String client_id; + }; + + template + static cppkafka::Configuration getConsumerConfiguration(TKafkaStorage & storage, const ConsumerConfigParams & params); + + template + static cppkafka::Configuration getProducerConfiguration(TKafkaStorage & storage, const ProducerConfigParams & params); +}; +} diff --git a/src/Storages/Kafka/KafkaConsumer.cpp b/src/Storages/Kafka/KafkaConsumer.cpp index 9ba42b9875e..d9256cf39ce 100644 --- a/src/Storages/Kafka/KafkaConsumer.cpp +++ b/src/Storages/Kafka/KafkaConsumer.cpp @@ -9,6 +9,7 @@ #include #include +#include #include #include @@ -20,13 +21,12 @@ namespace CurrentMetrics namespace ProfileEvents { - extern const Event KafkaRebalanceRevocations; - extern const Event KafkaRebalanceAssignments; - extern const Event KafkaRebalanceErrors; - extern const Event KafkaMessagesPolled; - extern const Event KafkaCommitFailures; - extern const Event KafkaCommits; - extern const Event KafkaConsumerErrors; +extern const Event KafkaRebalanceRevocations; +extern const Event KafkaRebalanceAssignments; +extern const Event KafkaRebalanceErrors; +extern const Event KafkaMessagesPolled; +extern const Event KafkaCommitFailures; +extern const Event KafkaCommits; } namespace DB @@ -199,44 +199,9 @@ KafkaConsumer::~KafkaConsumer() // https://github.com/confluentinc/confluent-kafka-go/issues/189 etc. void KafkaConsumer::drain() { - auto start_time = std::chrono::steady_clock::now(); - cppkafka::Error last_error(RD_KAFKA_RESP_ERR_NO_ERROR); - - while (true) - { - auto msg = consumer->poll(100ms); - if (!msg) - break; - - auto error = msg.get_error(); - - if (error) - { - if (msg.is_eof() || error == last_error) - { - break; - } - else - { - LOG_ERROR(log, "Error during draining: {}", error); - setExceptionInfo(error); - } - } - - // i don't stop draining on first error, - // only if it repeats once again sequentially - last_error = error; - - auto ts = std::chrono::steady_clock::now(); - if (std::chrono::duration_cast(ts-start_time) > DRAIN_TIMEOUT_MS) - { - LOG_ERROR(log, "Timeout during draining."); - break; - } - } + StorageKafkaUtils::drainConsumer(*consumer, DRAIN_TIMEOUT_MS, log, [this](const cppkafka::Error & err) { setExceptionInfo(err); }); } - void KafkaConsumer::commit() { auto print_offsets = [this] (const char * prefix, const cppkafka::TopicPartitionList & offsets) @@ -409,7 +374,7 @@ void KafkaConsumer::resetToLastCommitted(const char * msg) { if (!assignment.has_value() || assignment->empty()) { - LOG_TRACE(log, "Not assignned. Can't reset to last committed position."); + LOG_TRACE(log, "Not assigned. Can't reset to last committed position."); return; } auto committed_offset = consumer->get_offsets_committed(consumer->get_assignment()); @@ -473,7 +438,7 @@ ReadBufferPtr KafkaConsumer::consume() // If we're doing a manual select then it's better to get something after a wait, then immediate nothing. if (!assignment.has_value()) { - waited_for_assignment += poll_timeout; // slightly innaccurate, but rough calculation is ok. + waited_for_assignment += poll_timeout; // slightly inaccurate, but rough calculation is ok. if (waited_for_assignment < MAX_TIME_TO_WAIT_FOR_ASSIGNMENT_MS) { continue; @@ -535,26 +500,12 @@ ReadBufferPtr KafkaConsumer::getNextMessage() return getNextMessage(); } -size_t KafkaConsumer::filterMessageErrors() +void KafkaConsumer::filterMessageErrors() { assert(current == messages.begin()); - size_t skipped = std::erase_if(messages, [this](auto & message) - { - if (auto error = message.get_error()) - { - ProfileEvents::increment(ProfileEvents::KafkaConsumerErrors); - LOG_ERROR(log, "Consumer error: {}", error); - setExceptionInfo(error); - return true; - } - return false; - }); - - if (skipped) - LOG_ERROR(log, "There were {} messages with an error", skipped); - - return skipped; + StorageKafkaUtils::eraseMessageErrors(messages, log, [this](const cppkafka::Error & err) { setExceptionInfo(err); }); + current = messages.begin(); } void KafkaConsumer::resetIfStopped() diff --git a/src/Storages/Kafka/KafkaConsumer.h b/src/Storages/Kafka/KafkaConsumer.h index 4daf8652c3b..285f3680213 100644 --- a/src/Storages/Kafka/KafkaConsumer.h +++ b/src/Storages/Kafka/KafkaConsumer.h @@ -1,14 +1,12 @@ #pragma once #include -#include #include #include #include #include -#include #include namespace CurrentMetrics @@ -193,12 +191,8 @@ private: void drain(); void cleanUnprocessed(); void resetIfStopped(); - /// Return number of messages with an error. - size_t filterMessageErrors(); + void filterMessageErrors(); ReadBufferPtr getNextMessage(); }; } - -template <> struct fmt::formatter : fmt::ostream_formatter {}; -template <> struct fmt::formatter : fmt::ostream_formatter {}; diff --git a/src/Storages/Kafka/KafkaConsumer2.cpp b/src/Storages/Kafka/KafkaConsumer2.cpp new file mode 100644 index 00000000000..60626dfa402 --- /dev/null +++ b/src/Storages/Kafka/KafkaConsumer2.cpp @@ -0,0 +1,384 @@ +#include + +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +#include +#include + + +namespace CurrentMetrics +{ +extern const Metric KafkaAssignedPartitions; +extern const Metric KafkaConsumersWithAssignment; +} + +namespace ProfileEvents +{ +extern const Event KafkaRebalanceRevocations; +extern const Event KafkaRebalanceAssignments; +extern const Event KafkaRebalanceErrors; +extern const Event KafkaMessagesPolled; +extern const Event KafkaCommitFailures; +extern const Event KafkaCommits; +} + +namespace DB +{ + +using namespace std::chrono_literals; +static constexpr auto EVENT_POLL_TIMEOUT = 50ms; +static constexpr auto DRAIN_TIMEOUT_MS = 5000ms; + + +bool KafkaConsumer2::TopicPartition::operator<(const TopicPartition & other) const +{ + return std::tie(topic, partition_id, offset) < std::tie(other.topic, other.partition_id, other.offset); +} + + +KafkaConsumer2::KafkaConsumer2( + ConsumerPtr consumer_, + LoggerPtr log_, + size_t max_batch_size, + size_t poll_timeout_, + const std::atomic & stopped_, + const Names & topics_) + : consumer(consumer_) + , log(log_) + , batch_size(max_batch_size) + , poll_timeout(poll_timeout_) + , stopped(stopped_) + , current(messages.begin()) + , topics(topics_) +{ + // called (synchronously, during poll) when we enter the consumer group + consumer->set_assignment_callback( + [this](const cppkafka::TopicPartitionList & topic_partitions) + { + CurrentMetrics::add(CurrentMetrics::KafkaAssignedPartitions, topic_partitions.size()); + ProfileEvents::increment(ProfileEvents::KafkaRebalanceAssignments); + + if (topic_partitions.empty()) + { + LOG_INFO(log, "Got empty assignment: Not enough partitions in the topic for all consumers?"); + } + else + { + LOG_TRACE(log, "Topics/partitions assigned: {}", topic_partitions); + CurrentMetrics::add(CurrentMetrics::KafkaConsumersWithAssignment, 1); + } + + chassert(!assignment.has_value()); + + assignment.emplace(); + assignment->reserve(topic_partitions.size()); + needs_offset_update = true; + for (const auto & topic_partition : topic_partitions) + { + assignment->push_back( + TopicPartition{topic_partition.get_topic(), topic_partition.get_partition(), topic_partition.get_offset()}); + } + + // We need to initialize the queues here in order to detach them from the consumer queue. Otherwise `pollEvents` might eventually poll actual messages also. + initializeQueues(topic_partitions); + }); + + // called (synchronously, during poll) when we leave the consumer group + consumer->set_revocation_callback( + [this](const cppkafka::TopicPartitionList & topic_partitions) + { + CurrentMetrics::sub(CurrentMetrics::KafkaAssignedPartitions, topic_partitions.size()); + ProfileEvents::increment(ProfileEvents::KafkaRebalanceRevocations); + + // Rebalance is happening now, and now we have a chance to finish the work + // with topics/partitions we were working with before rebalance + LOG_TRACE(log, "Rebalance initiated. Revoking partitions: {}", topic_partitions); + + if (!topic_partitions.empty()) + { + CurrentMetrics::sub(CurrentMetrics::KafkaConsumersWithAssignment, 1); + } + + assignment.reset(); + queues.clear(); + needs_offset_update = true; + }); + + consumer->set_rebalance_error_callback( + [this](cppkafka::Error err) + { + LOG_ERROR(log, "Rebalance error: {}", err); + ProfileEvents::increment(ProfileEvents::KafkaRebalanceErrors); + }); +} + +KafkaConsumer2::~KafkaConsumer2() +{ + try + { + if (!consumer->get_subscription().empty()) + { + try + { + consumer->unsubscribe(); + } + catch (const cppkafka::HandleException & e) + { + LOG_ERROR(log, "Error during unsubscribe: {}", e.what()); + } + drainConsumerQueue(); + } + } + catch (const cppkafka::HandleException & e) + { + LOG_ERROR(log, "Error while destructing consumer: {}", e.what()); + } +} + +// Needed to drain rest of the messages / queued callback calls from the consumer after unsubscribe, otherwise consumer +// will hang on destruction. Partition queues doesn't have to be attached as events are not handled by those queues. +// see https://github.com/edenhill/librdkafka/issues/2077 +// https://github.com/confluentinc/confluent-kafka-go/issues/189 etc. +void KafkaConsumer2::drainConsumerQueue() +{ + StorageKafkaUtils::drainConsumer(*consumer, DRAIN_TIMEOUT_MS, log); +} + +void KafkaConsumer2::pollEvents() +{ + static constexpr auto max_tries = 5; + for (auto i = 0; i < max_tries; ++i) + { + auto msg = consumer->poll(EVENT_POLL_TIMEOUT); + if (!msg) + return; + // All the partition queues are detached, so the consumer shouldn't be able to poll any real messages + const auto err = msg.get_error(); + chassert(RD_KAFKA_RESP_ERR_NO_ERROR != err.get_error() && "Consumer returned a message when it was not expected"); + LOG_ERROR(log, "Consumer received error while polling events, code {}, error '{}'", err.get_error(), err.to_string()); + } +}; + +bool KafkaConsumer2::polledDataUnusable(const TopicPartition & topic_partition) const +{ + const auto different_topic_partition = current == messages.end() + ? false + : (current->get_topic() != topic_partition.topic || current->get_partition() != topic_partition.partition_id); + return different_topic_partition; +} + +KafkaConsumer2::TopicPartitions const * KafkaConsumer2::getKafkaAssignment() const +{ + if (assignment.has_value()) + { + return &*assignment; + } + + return nullptr; +} + +void KafkaConsumer2::updateOffsets(const TopicPartitions & topic_partitions) +{ + cppkafka::TopicPartitionList original_topic_partitions; + original_topic_partitions.reserve(topic_partitions.size()); + std::transform( + topic_partitions.begin(), + topic_partitions.end(), + std::back_inserter(original_topic_partitions), + [](const TopicPartition & tp) + { + return cppkafka::TopicPartition{tp.topic, tp.partition_id, tp.offset}; + }); + initializeQueues(original_topic_partitions); + needs_offset_update = false; + stalled_status = StalledStatus::NOT_STALLED; +} + +void KafkaConsumer2::initializeQueues(const cppkafka::TopicPartitionList & topic_partitions) +{ + queues.clear(); + messages.clear(); + current = messages.end(); + // cppkafka itself calls assign(), but in order to detach the queues here we have to do the assignment manually. Later on we have to reassign the topic partitions with correct offsets. + consumer->assign(topic_partitions); + for (const auto & topic_partition : topic_partitions) + // This will also detach the partition queues from the consumer, thus the messages won't be forwarded without attaching them manually + queues.emplace( + TopicPartition{topic_partition.get_topic(), topic_partition.get_partition(), topic_partition.get_offset()}, + consumer->get_partition_queue(topic_partition)); +} + +// it do the poll when needed +ReadBufferPtr KafkaConsumer2::consume(const TopicPartition & topic_partition, const std::optional & message_count) +{ + resetIfStopped(); + + if (polledDataUnusable(topic_partition)) + return nullptr; + + if (hasMorePolledMessages()) + { + if (auto next_message = getNextMessage(); next_message) + return next_message; + } + + while (true) + { + stalled_status = StalledStatus::NO_MESSAGES_RETURNED; + + auto & queue_to_poll_from = queues.at(topic_partition); + LOG_TRACE(log, "Batch size {}, offset {}", batch_size, topic_partition.offset); + const auto messages_to_pull = message_count.value_or(batch_size); + /// Don't drop old messages immediately, since we may need them for virtual columns. + auto new_messages = queue_to_poll_from.consume_batch(messages_to_pull, std::chrono::milliseconds(poll_timeout)); + + resetIfStopped(); + if (stalled_status == StalledStatus::CONSUMER_STOPPED) + { + return nullptr; + } + + if (new_messages.empty()) + { + LOG_TRACE(log, "Stalled"); + return nullptr; + } + else + { + messages = std::move(new_messages); + current = messages.begin(); + LOG_TRACE( + log, + "Polled batch of {} messages. Offsets position: {}", + messages.size(), + consumer->get_offsets_position(consumer->get_assignment())); + break; + } + } + + filterMessageErrors(); + if (current == messages.end()) + { + LOG_ERROR(log, "Only errors left"); + stalled_status = StalledStatus::ERRORS_RETURNED; + return nullptr; + } + + ProfileEvents::increment(ProfileEvents::KafkaMessagesPolled, messages.size()); + + stalled_status = StalledStatus::NOT_STALLED; + return getNextMessage(); +} + +void KafkaConsumer2::commit(const TopicPartition & topic_partition) +{ + static constexpr auto max_retries = 5; + bool committed = false; + + LOG_TEST( + log, + "Trying to commit offset {} to Kafka for topic-partition [{}:{}]", + topic_partition.offset, + topic_partition.topic, + topic_partition.partition_id); + + const auto topic_partition_list = std::vector{cppkafka::TopicPartition{ + topic_partition.topic, + topic_partition.partition_id, + topic_partition.offset, + }}; + for (auto try_count = 0; try_count < max_retries && !committed; ++try_count) + { + try + { + // See https://github.com/edenhill/librdkafka/issues/1470 + // broker may reject commit if during offsets.commit.timeout.ms (5000 by default), + // there were not enough replicas available for the __consumer_offsets topic. + // also some other temporary issues like client-server connectivity problems are possible + + consumer->commit(topic_partition_list); + committed = true; + LOG_INFO( + log, + "Committed offset {} to Kafka for topic-partition [{}:{}]", + topic_partition.offset, + topic_partition.topic, + topic_partition.partition_id); + } + catch (const cppkafka::HandleException & e) + { + // If there were actually no offsets to commit, return. Retrying won't solve + // anything here + if (e.get_error() == RD_KAFKA_RESP_ERR__NO_OFFSET) + committed = true; + else + LOG_ERROR(log, "Exception during attempt to commit to Kafka: {}", e.what()); + } + } + + if (!committed) + { + // The failure is not the biggest issue, it only counts when a table is dropped and recreated, otherwise the offsets are taken from keeper. + ProfileEvents::increment(ProfileEvents::KafkaCommitFailures); + LOG_ERROR(log, "All commit attempts failed"); + } + else + { + ProfileEvents::increment(ProfileEvents::KafkaCommits); + } +} + +void KafkaConsumer2::subscribeIfNotSubscribedYet() +{ + if (likely(is_subscribed)) + return; + + consumer->subscribe(topics); + is_subscribed = true; + LOG_DEBUG(log, "Subscribed."); +} + +ReadBufferPtr KafkaConsumer2::getNextMessage() +{ + while (current != messages.end()) + { + const auto * data = current->get_payload().get_data(); + size_t size = current->get_payload().get_size(); + ++current; + + // `data` can be nullptr on case of the Kafka message has empty payload + if (data) + return std::make_shared(data, size); + } + + return nullptr; +} + +void KafkaConsumer2::filterMessageErrors() +{ + assert(current == messages.begin()); + + StorageKafkaUtils::eraseMessageErrors(messages, log); + current = messages.begin(); +} + +void KafkaConsumer2::resetIfStopped() +{ + if (stopped) + { + stalled_status = StalledStatus::CONSUMER_STOPPED; + } +} +} diff --git a/src/Storages/Kafka/KafkaConsumer2.h b/src/Storages/Kafka/KafkaConsumer2.h new file mode 100644 index 00000000000..dd2cfe87aa0 --- /dev/null +++ b/src/Storages/Kafka/KafkaConsumer2.h @@ -0,0 +1,162 @@ +#pragma once + +#include +#include +#include +#include + +#include +#include +#include +#include + +namespace CurrentMetrics +{ +extern const Metric KafkaConsumers; +} + +namespace Poco +{ +class Logger; +} + +namespace DB +{ + +using ConsumerPtr = std::shared_ptr; + +class KafkaConsumer2 +{ +public: + static inline constexpr int INVALID_OFFSET = RD_KAFKA_OFFSET_INVALID; + static inline constexpr int BEGINNING_OFFSET = RD_KAFKA_OFFSET_BEGINNING; + static inline constexpr int END_OFFSET = RD_KAFKA_OFFSET_END; + + struct TopicPartition + { + String topic; + int32_t partition_id; + int64_t offset{INVALID_OFFSET}; + + bool operator==(const TopicPartition &) const = default; + bool operator<(const TopicPartition & other) const; + }; + + using TopicPartitions = std::vector; + + struct OnlyTopicNameAndPartitionIdHash + { + std::size_t operator()(const TopicPartition & tp) const + { + SipHash s; + s.update(tp.topic); + s.update(tp.partition_id); + return s.get64(); + } + }; + + struct OnlyTopicNameAndPartitionIdEquality + { + bool operator()(const TopicPartition & lhs, const TopicPartition & rhs) const + { + return lhs.topic == rhs.topic && lhs.partition_id == rhs.partition_id; + } + }; + + struct TopicPartitionCount + { + String topic; + size_t partition_count; + }; + + using TopicPartitionCounts = std::vector; + + KafkaConsumer2( + ConsumerPtr consumer_, + LoggerPtr log_, + size_t max_batch_size, + size_t poll_timeout_, + const std::atomic & stopped_, + const Names & topics_); + + ~KafkaConsumer2(); + + // Poll only the main consumer queue without any topic-partition queues. This is useful to get notified about events, such as rebalance, + // new assignment, etc.. + void pollEvents(); + + auto pollTimeout() const { return poll_timeout; } + + inline bool hasMorePolledMessages() const { return (stalled_status == StalledStatus::NOT_STALLED) && (current != messages.end()); } + + inline bool isStalled() const { return stalled_status != StalledStatus::NOT_STALLED; } + + // Returns the topic partitions that the consumer got from rebalancing the consumer group. If the consumer received + // no topic partitions or all of them were revoked, it returns a null pointer. + TopicPartitions const * getKafkaAssignment() const; + + // As the main source of offsets is not Kafka, the offsets needs to be pushed to the consumer from outside + // Returns true if it received new assignment and internal state should be updated by updateOffsets + bool needsOffsetUpdate() const { return needs_offset_update; } + void updateOffsets(const TopicPartitions & topic_partitions); + + /// Polls batch of messages from the given topic-partition and returns read buffer containing the next message or + /// nullptr when there are no messages to process. + ReadBufferPtr consume(const TopicPartition & topic_partition, const std::optional & message_count); + + void commit(const TopicPartition & topic_partition); + + // Return values for the message that's being read. + String currentTopic() const { return current[-1].get_topic(); } + String currentKey() const { return current[-1].get_key(); } + auto currentOffset() const { return current[-1].get_offset(); } + auto currentPartition() const { return current[-1].get_partition(); } + auto currentTimestamp() const { return current[-1].get_timestamp(); } + const auto & currentHeaderList() const { return current[-1].get_header_list(); } + String currentPayload() const { return current[-1].get_payload(); } + + void subscribeIfNotSubscribedYet(); + +private: + using Messages = std::vector; + CurrentMetrics::Increment metric_increment{CurrentMetrics::KafkaConsumers}; + + enum class StalledStatus + { + NOT_STALLED, + NO_MESSAGES_RETURNED, + CONSUMER_STOPPED, + NO_ASSIGNMENT, + ERRORS_RETURNED + }; + + ConsumerPtr consumer; + LoggerPtr log; + const size_t batch_size = 1; + const size_t poll_timeout = 0; + + StalledStatus stalled_status = StalledStatus::NO_MESSAGES_RETURNED; + + const std::atomic & stopped; + bool is_subscribed = false; + + // order is important, need to be destructed before consumer + Messages messages; + Messages::const_iterator current; + + // order is important, need to be destructed before consumer + std::optional assignment; + bool needs_offset_update{false}; + std::unordered_map queues; + const Names topics; + + bool polledDataUnusable(const TopicPartition & topic_partition) const; + void drainConsumerQueue(); + void resetIfStopped(); + void filterMessageErrors(); + ReadBufferPtr getNextMessage(); + + void initializeQueues(const cppkafka::TopicPartitionList & topic_partitions); +}; + +} diff --git a/src/Storages/Kafka/KafkaSettings.h b/src/Storages/Kafka/KafkaSettings.h index 0addaf9e3b3..9ca5e189f0e 100644 --- a/src/Storages/Kafka/KafkaSettings.h +++ b/src/Storages/Kafka/KafkaSettings.h @@ -38,6 +38,8 @@ const auto KAFKA_CONSUMERS_POOL_TTL_MS_MAX = 600'000; M(StreamingHandleErrorMode, kafka_handle_error_mode, StreamingHandleErrorMode::DEFAULT, "How to handle errors for Kafka engine. Possible values: default (throw an exception after rabbitmq_skip_broken_messages broken messages), stream (save broken messages and errors in virtual columns _raw_message, _error).", 0) \ M(Bool, kafka_commit_on_select, false, "Commit messages when select query is made", 0) \ M(UInt64, kafka_max_rows_per_message, 1, "The maximum number of rows produced in one kafka message for row-based formats.", 0) \ + M(String, kafka_keeper_path, "", "The path to the table in ClickHouse Keeper", 0) \ + M(String, kafka_replica_name, "", "The replica name in ClickHouse Keeper", 0) \ #define OBSOLETE_KAFKA_SETTINGS(M, ALIAS) \ MAKE_OBSOLETE(M, Char, kafka_row_delimiter, '\0') \ diff --git a/src/Storages/Kafka/KafkaSource.cpp b/src/Storages/Kafka/KafkaSource.cpp index 9c68107872e..3ddd0d1be8c 100644 --- a/src/Storages/Kafka/KafkaSource.cpp +++ b/src/Storages/Kafka/KafkaSource.cpp @@ -262,7 +262,7 @@ Chunk KafkaSource::generateImpl() // they are not needed here: // and it's misleading to use them here, // as columns 'materialized' that way stays 'ephemeral' - // i.e. will not be stored anythere + // i.e. will not be stored anywhere // If needed any extra columns can be added using DEFAULT they can be added at MV level if needed. auto result_block = non_virtual_header.cloneWithColumns(executor.getResultColumns()); diff --git a/src/Storages/Kafka/StorageKafka.cpp b/src/Storages/Kafka/StorageKafka.cpp index 3aad64a0cfb..f4f641d1c68 100644 --- a/src/Storages/Kafka/StorageKafka.cpp +++ b/src/Storages/Kafka/StorageKafka.cpp @@ -1,13 +1,5 @@ #include -#include -#include -#include -#include -#include -#include -#include -#include #include #include #include @@ -21,18 +13,19 @@ #include #include #include +#include +#include #include #include #include #include #include #include +#include #include #include -#include #include #include -#include #include #include #include @@ -41,10 +34,10 @@ #include #include #include +#include #include #include #include -#include #include #include #include @@ -55,13 +48,8 @@ #include #include -#if USE_KRB5 -#include -#endif // USE_KRB5 - namespace CurrentMetrics { - extern const Metric KafkaLibrdkafkaThreads; extern const Metric KafkaBackgroundReads; extern const Metric KafkaConsumersInUse; extern const Metric KafkaWrites; @@ -82,104 +70,10 @@ namespace ErrorCodes { extern const int NOT_IMPLEMENTED; extern const int LOGICAL_ERROR; - extern const int BAD_ARGUMENTS; - extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; extern const int QUERY_NOT_ALLOWED; extern const int ABORTED; } -struct StorageKafkaInterceptors -{ - static rd_kafka_resp_err_t rdKafkaOnThreadStart(rd_kafka_t *, rd_kafka_thread_type_t thread_type, const char *, void * ctx) - { - StorageKafka * self = reinterpret_cast(ctx); - CurrentMetrics::add(CurrentMetrics::KafkaLibrdkafkaThreads, 1); - - const auto & storage_id = self->getStorageID(); - const auto & table = storage_id.getTableName(); - - switch (thread_type) - { - case RD_KAFKA_THREAD_MAIN: - setThreadName(("rdk:m/" + table.substr(0, 9)).c_str()); - break; - case RD_KAFKA_THREAD_BACKGROUND: - setThreadName(("rdk:bg/" + table.substr(0, 8)).c_str()); - break; - case RD_KAFKA_THREAD_BROKER: - setThreadName(("rdk:b/" + table.substr(0, 9)).c_str()); - break; - } - - /// Create ThreadStatus to track memory allocations from librdkafka threads. - // - /// And store them in a separate list (thread_statuses) to make sure that they will be destroyed, - /// regardless how librdkafka calls the hooks. - /// But this can trigger use-after-free if librdkafka will not destroy threads after rd_kafka_wait_destroyed() - auto thread_status = std::make_shared(); - std::lock_guard lock(self->thread_statuses_mutex); - self->thread_statuses.emplace_back(std::move(thread_status)); - - return RD_KAFKA_RESP_ERR_NO_ERROR; - } - static rd_kafka_resp_err_t rdKafkaOnThreadExit(rd_kafka_t *, rd_kafka_thread_type_t, const char *, void * ctx) - { - StorageKafka * self = reinterpret_cast(ctx); - CurrentMetrics::sub(CurrentMetrics::KafkaLibrdkafkaThreads, 1); - - std::lock_guard lock(self->thread_statuses_mutex); - const auto it = std::find_if(self->thread_statuses.begin(), self->thread_statuses.end(), [](const auto & thread_status_ptr) - { - return thread_status_ptr.get() == current_thread; - }); - if (it == self->thread_statuses.end()) - throw Exception(ErrorCodes::LOGICAL_ERROR, "No thread status for this librdkafka thread."); - - self->thread_statuses.erase(it); - - return RD_KAFKA_RESP_ERR_NO_ERROR; - } - - static rd_kafka_resp_err_t rdKafkaOnNew(rd_kafka_t * rk, const rd_kafka_conf_t *, void * ctx, char * /*errstr*/, size_t /*errstr_size*/) - { - StorageKafka * self = reinterpret_cast(ctx); - rd_kafka_resp_err_t status; - - status = rd_kafka_interceptor_add_on_thread_start(rk, "init-thread", rdKafkaOnThreadStart, ctx); - if (status != RD_KAFKA_RESP_ERR_NO_ERROR) - { - LOG_ERROR(self->log, "Cannot set on thread start interceptor due to {} error", status); - return status; - } - - status = rd_kafka_interceptor_add_on_thread_exit(rk, "exit-thread", rdKafkaOnThreadExit, ctx); - if (status != RD_KAFKA_RESP_ERR_NO_ERROR) - LOG_ERROR(self->log, "Cannot set on thread exit interceptor due to {} error", status); - - return status; - } - - static rd_kafka_resp_err_t rdKafkaOnConfDup(rd_kafka_conf_t * new_conf, const rd_kafka_conf_t * /*old_conf*/, size_t /*filter_cnt*/, const char ** /*filter*/, void * ctx) - { - StorageKafka * self = reinterpret_cast(ctx); - rd_kafka_resp_err_t status; - - // cppkafka copies configuration multiple times - status = rd_kafka_conf_interceptor_add_on_conf_dup(new_conf, "init", rdKafkaOnConfDup, ctx); - if (status != RD_KAFKA_RESP_ERR_NO_ERROR) - { - LOG_ERROR(self->log, "Cannot set on conf dup interceptor due to {} error", status); - return status; - } - - status = rd_kafka_conf_interceptor_add_on_new(new_conf, "init", rdKafkaOnNew, ctx); - if (status != RD_KAFKA_RESP_ERR_NO_ERROR) - LOG_ERROR(self->log, "Cannot set on conf new interceptor due to {} error", status); - - return status; - } -}; - class ReadFromStorageKafka final : public ReadFromStreamLikeEngine { public: @@ -241,182 +135,6 @@ private: StorageSnapshotPtr storage_snapshot; }; -namespace -{ - const String CONFIG_KAFKA_TAG = "kafka"; - const String CONFIG_KAFKA_TOPIC_TAG = "kafka_topic"; - const String CONFIG_KAFKA_CONSUMER_TAG = "consumer"; - const String CONFIG_KAFKA_PRODUCER_TAG = "producer"; - const String CONFIG_NAME_TAG = "name"; - - void setKafkaConfigValue(cppkafka::Configuration & kafka_config, const String & key, const String & value) - { - /// "log_level" has valid underscore, the remaining librdkafka setting use dot.separated.format which isn't acceptable for XML. - /// See https://github.com/edenhill/librdkafka/blob/master/CONFIGURATION.md - const String setting_name_in_kafka_config = (key == "log_level") ? key : boost::replace_all_copy(key, "_", "."); - kafka_config.set(setting_name_in_kafka_config, value); - } - - void loadConfigProperty(cppkafka::Configuration & kafka_config, const Poco::Util::AbstractConfiguration & config, const String & config_prefix, const String & tag) - { - const String property_path = config_prefix + "." + tag; - const String property_value = config.getString(property_path); - - setKafkaConfigValue(kafka_config, tag, property_value); - } - - void loadNamedCollectionConfig(cppkafka::Configuration & kafka_config, const String & collection_name, const String & config_prefix) - { - const auto & collection = NamedCollectionFactory::instance().get(collection_name); - for (const auto & key : collection->getKeys(-1, config_prefix)) - { - // Cut prefix with '.' before actual config tag. - const auto param_name = key.substr(config_prefix.size() + 1); - setKafkaConfigValue(kafka_config, param_name, collection->get(key)); - } - } - - void loadLegacyTopicConfig(cppkafka::Configuration & kafka_config, const Poco::Util::AbstractConfiguration & config, const String & collection_name, const String & config_prefix) - { - if (!collection_name.empty()) - { - loadNamedCollectionConfig(kafka_config, collection_name, config_prefix); - return; - } - - Poco::Util::AbstractConfiguration::Keys tags; - config.keys(config_prefix, tags); - - for (const auto & tag : tags) - { - loadConfigProperty(kafka_config, config, config_prefix, tag); - } - } - - /// Read server configuration into cppkafa configuration, used by new per-topic configuration - void loadTopicConfig(cppkafka::Configuration & kafka_config, const Poco::Util::AbstractConfiguration & config, const String & collection_name, const String & config_prefix, const String & topic) - { - if (!collection_name.empty()) - { - const auto topic_prefix = fmt::format("{}.{}", config_prefix, CONFIG_KAFKA_TOPIC_TAG); - const auto & collection = NamedCollectionFactory::instance().get(collection_name); - for (const auto & key : collection->getKeys(1, config_prefix)) - { - /// Only consider key . Multiple occurrences given as "kafka_topic", "kafka_topic[1]", etc. - if (!key.starts_with(topic_prefix)) - continue; - - const String kafka_topic_path = config_prefix + "." + key; - const String kafka_topic_name_path = kafka_topic_path + "." + CONFIG_NAME_TAG; - if (topic == collection->get(kafka_topic_name_path)) - /// Found it! Now read the per-topic configuration into cppkafka. - loadNamedCollectionConfig(kafka_config, collection_name, kafka_topic_path); - } - } - else - { - /// Read all tags one level below - Poco::Util::AbstractConfiguration::Keys tags; - config.keys(config_prefix, tags); - - for (const auto & tag : tags) - { - if (tag == CONFIG_NAME_TAG) - continue; // ignore , it is used to match topic configurations - loadConfigProperty(kafka_config, config, config_prefix, tag); - } - } - } - - /// Read server configuration into cppkafka configuration, used by global configuration and by legacy per-topic configuration - void loadFromConfig(cppkafka::Configuration & kafka_config, const Poco::Util::AbstractConfiguration & config, const String & collection_name, const String & config_prefix, const Names & topics) - { - if (!collection_name.empty()) - { - loadNamedCollectionConfig(kafka_config, collection_name, config_prefix); - return; - } - - /// Read all tags one level below - Poco::Util::AbstractConfiguration::Keys tags; - config.keys(config_prefix, tags); - - for (const auto & tag : tags) - { - if (tag == CONFIG_KAFKA_PRODUCER_TAG || tag == CONFIG_KAFKA_CONSUMER_TAG) - /// Do not load consumer/producer properties, since they should be separated by different configuration objects. - continue; - - if (tag.starts_with(CONFIG_KAFKA_TOPIC_TAG)) /// multiple occurrences given as "kafka_topic", "kafka_topic[1]", etc. - { - // Update consumer topic-specific configuration (new syntax). Example with topics "football" and "baseball": - // - // - // football - // 250 - // 5000 - // - // - // baseball - // 300 - // 2000 - // - // - // Advantages: The period restriction no longer applies (e.g. sports.football will work), everything - // Kafka-related is below . - for (const auto & topic : topics) - { - /// Read topic name between ... - const String kafka_topic_path = config_prefix + "." + tag; - const String kafka_topic_name_path = kafka_topic_path + "." + CONFIG_NAME_TAG; - const String topic_name = config.getString(kafka_topic_name_path); - - if (topic_name != topic) - continue; - loadTopicConfig(kafka_config, config, collection_name, kafka_topic_path, topic); - } - continue; - } - if (tag.starts_with(CONFIG_KAFKA_TAG)) - /// skip legacy configuration per topic e.g. . - /// it will be processed is a separate function - continue; - // Update configuration from the configuration. Example: - // - // 250 - // 100000 - // - loadConfigProperty(kafka_config, config, config_prefix, tag); - } - } - - void loadLegacyConfigSyntax(cppkafka::Configuration & kafka_config, const Poco::Util::AbstractConfiguration & config, const String & collection_name, const String & prefix, const Names & topics) - { - for (const auto & topic : topics) - { - const String kafka_topic_path = prefix + "." + CONFIG_KAFKA_TAG + "_" + topic; - loadLegacyTopicConfig(kafka_config, config, collection_name, kafka_topic_path); - } - } - - void loadConsumerConfig(cppkafka::Configuration & kafka_config, const Poco::Util::AbstractConfiguration & config, const String & collection_name, const String & prefix, const Names & topics) - { - const String consumer_path = prefix + "." + CONFIG_KAFKA_CONSUMER_TAG; - loadLegacyConfigSyntax(kafka_config, config, collection_name, prefix, topics); - // A new syntax has higher priority - loadFromConfig(kafka_config, config, collection_name, consumer_path, topics); - } - - void loadProducerConfig(cppkafka::Configuration & kafka_config, const Poco::Util::AbstractConfiguration & config, const String & collection_name, const String & prefix, const Names & topics) - { - const String producer_path = prefix + "." + CONFIG_KAFKA_PRODUCER_TAG; - loadLegacyConfigSyntax(kafka_config, config, collection_name, prefix, topics); - // A new syntax has higher priority - loadFromConfig(kafka_config, config, collection_name, producer_path, topics); - - } -} - StorageKafka::StorageKafka( const StorageID & table_id_, ContextPtr context_, @@ -428,19 +146,20 @@ StorageKafka::StorageKafka( , WithContext(context_->getGlobalContext()) , kafka_settings(std::move(kafka_settings_)) , macros_info{.table_id = table_id_} - , topics(parseTopics(getContext()->getMacros()->expand(kafka_settings->kafka_topic_list.value, macros_info))) + , topics(StorageKafkaUtils::parseTopics(getContext()->getMacros()->expand(kafka_settings->kafka_topic_list.value, macros_info))) , brokers(getContext()->getMacros()->expand(kafka_settings->kafka_broker_list.value, macros_info)) , group(getContext()->getMacros()->expand(kafka_settings->kafka_group_name.value, macros_info)) , client_id( - kafka_settings->kafka_client_id.value.empty() ? getDefaultClientId(table_id_) - : getContext()->getMacros()->expand(kafka_settings->kafka_client_id.value, macros_info)) + kafka_settings->kafka_client_id.value.empty() + ? StorageKafkaUtils::getDefaultClientId(table_id_) + : getContext()->getMacros()->expand(kafka_settings->kafka_client_id.value, macros_info)) , format_name(getContext()->getMacros()->expand(kafka_settings->kafka_format.value)) , max_rows_per_message(kafka_settings->kafka_max_rows_per_message.value) , schema_name(getContext()->getMacros()->expand(kafka_settings->kafka_schema.value, macros_info)) , num_consumers(kafka_settings->kafka_num_consumers.value) , log(getLogger("StorageKafka (" + table_id_.table_name + ")")) , intermediate_commit(kafka_settings->kafka_commit_every_batch.value) - , settings_adjustments(createSettingsAdjustments()) + , settings_adjustments(StorageKafkaUtils::createSettingsAdjustments(*kafka_settings, schema_name)) , thread_per_consumer(kafka_settings->kafka_thread_per_consumer.value) , collection_name(collection_name_) { @@ -456,7 +175,7 @@ StorageKafka::StorageKafka( storage_metadata.setColumns(columns_); storage_metadata.setComment(comment); setInMemoryMetadata(storage_metadata); - setVirtuals(createVirtuals(kafka_settings->kafka_handle_error_mode)); + setVirtuals(StorageKafkaUtils::createVirtuals(kafka_settings->kafka_handle_error_mode)); auto task_count = thread_per_consumer ? num_consumers : 1; for (size_t i = 0; i < task_count; ++i) @@ -481,76 +200,6 @@ StorageKafka::StorageKafka( StorageKafka::~StorageKafka() = default; -VirtualColumnsDescription StorageKafka::createVirtuals(StreamingHandleErrorMode handle_error_mode) -{ - VirtualColumnsDescription desc; - - desc.addEphemeral("_topic", std::make_shared(std::make_shared()), ""); - desc.addEphemeral("_key", std::make_shared(), ""); - desc.addEphemeral("_offset", std::make_shared(), ""); - desc.addEphemeral("_partition", std::make_shared(), ""); - desc.addEphemeral("_timestamp", std::make_shared(std::make_shared()), ""); - desc.addEphemeral("_timestamp_ms", std::make_shared(std::make_shared(3)), ""); - desc.addEphemeral("_headers.name", std::make_shared(std::make_shared()), ""); - desc.addEphemeral("_headers.value", std::make_shared(std::make_shared()), ""); - - if (handle_error_mode == StreamingHandleErrorMode::STREAM) - { - desc.addEphemeral("_raw_message", std::make_shared(), ""); - desc.addEphemeral("_error", std::make_shared(), ""); - } - - return desc; -} - -SettingsChanges StorageKafka::createSettingsAdjustments() -{ - SettingsChanges result; - // Needed for backward compatibility - if (!kafka_settings->input_format_skip_unknown_fields.changed) - { - // Always skip unknown fields regardless of the context (JSON or TSKV) - kafka_settings->input_format_skip_unknown_fields = true; - } - - if (!kafka_settings->input_format_allow_errors_ratio.changed) - { - kafka_settings->input_format_allow_errors_ratio = 0.; - } - - if (!kafka_settings->input_format_allow_errors_num.changed) - { - kafka_settings->input_format_allow_errors_num = kafka_settings->kafka_skip_broken_messages.value; - } - - if (!schema_name.empty()) - result.emplace_back("format_schema", schema_name); - - for (const auto & setting : *kafka_settings) - { - const auto & name = setting.getName(); - if (name.find("kafka_") == std::string::npos) - result.emplace_back(name, setting.getValue()); - } - return result; -} - -Names StorageKafka::parseTopics(String topic_list) -{ - Names result; - boost::split(result,topic_list,[](char c){ return c == ','; }); - for (String & topic : result) - { - boost::trim(topic); - } - return result; -} - -String StorageKafka::getDefaultClientId(const StorageID & table_id_) -{ - return fmt::format("{}-{}-{}-{}", VERSION_NAME, getFQDNOrHostName(), table_id_.database_name, table_id_.table_name); -} - void StorageKafka::read( QueryPlan & query_plan, const Names & column_names, @@ -751,65 +400,26 @@ KafkaConsumerPtr StorageKafka::createKafkaConsumer(size_t consumer_number) topics); return kafka_consumer_ptr; } - cppkafka::Configuration StorageKafka::getConsumerConfiguration(size_t consumer_number) { - cppkafka::Configuration conf; - - conf.set("metadata.broker.list", brokers); - conf.set("group.id", group); - if (num_consumers > 1) - { - conf.set("client.id", fmt::format("{}-{}", client_id, consumer_number)); - } - else - { - conf.set("client.id", client_id); - } - conf.set("client.software.name", VERSION_NAME); - conf.set("client.software.version", VERSION_DESCRIBE); - conf.set("auto.offset.reset", "earliest"); // If no offset stored for this group, read all messages from the start - - // that allows to prevent fast draining of the librdkafka queue - // during building of single insert block. Improves performance - // significantly, but may lead to bigger memory consumption. - size_t default_queued_min_messages = 100000; // must be greater than or equal to default - size_t max_allowed_queued_min_messages = 10000000; // must be less than or equal to max allowed value - conf.set("queued.min.messages", std::min(std::max(getMaxBlockSize(), default_queued_min_messages), max_allowed_queued_min_messages)); - - updateGlobalConfiguration(conf); - updateConsumerConfiguration(conf); - - // those settings should not be changed by users. - conf.set("enable.auto.commit", "false"); // We manually commit offsets after a stream successfully finished - conf.set("enable.auto.offset.store", "false"); // Update offset automatically - to commit them all at once. - conf.set("enable.partition.eof", "false"); // Ignore EOF messages - - for (auto & property : conf.get_all()) - { - LOG_TRACE(log, "Consumer set property {}:{}", property.first, property.second); - } - - return conf; + KafkaConfigLoader::ConsumerConfigParams params{ + {getContext()->getConfigRef(), collection_name, topics, log}, + brokers, + group, + num_consumers > 1, + consumer_number, + client_id, + getMaxBlockSize()}; + return KafkaConfigLoader::getConsumerConfiguration(*this, params); } cppkafka::Configuration StorageKafka::getProducerConfiguration() { - cppkafka::Configuration conf; - conf.set("metadata.broker.list", brokers); - conf.set("client.id", client_id); - conf.set("client.software.name", VERSION_NAME); - conf.set("client.software.version", VERSION_DESCRIBE); - - updateGlobalConfiguration(conf); - updateProducerConfiguration(conf); - - for (auto & property : conf.get_all()) - { - LOG_TRACE(log, "Producer set property {}:{}", property.first, property.second); - } - - return conf; + KafkaConfigLoader::ProducerConfigParams params{ + {getContext()->getConfigRef(), collection_name, topics, log}, + brokers, + client_id}; + return KafkaConfigLoader::getProducerConfiguration(*this, params); } void StorageKafka::cleanConsumers() @@ -887,126 +497,6 @@ size_t StorageKafka::getPollTimeoutMillisecond() const : getContext()->getSettingsRef().stream_poll_timeout_ms.totalMilliseconds(); } -void StorageKafka::updateGlobalConfiguration(cppkafka::Configuration & kafka_config) -{ - const auto & config = getContext()->getConfigRef(); - loadFromConfig(kafka_config, config, collection_name, CONFIG_KAFKA_TAG, topics); - -#if USE_KRB5 - if (kafka_config.has_property("sasl.kerberos.kinit.cmd")) - LOG_WARNING(log, "sasl.kerberos.kinit.cmd configuration parameter is ignored."); - - kafka_config.set("sasl.kerberos.kinit.cmd",""); - kafka_config.set("sasl.kerberos.min.time.before.relogin","0"); - - if (kafka_config.has_property("sasl.kerberos.keytab") && kafka_config.has_property("sasl.kerberos.principal")) - { - String keytab = kafka_config.get("sasl.kerberos.keytab"); - String principal = kafka_config.get("sasl.kerberos.principal"); - LOG_DEBUG(log, "Running KerberosInit"); - try - { - kerberosInit(keytab,principal); - } - catch (const Exception & e) - { - LOG_ERROR(log, "KerberosInit failure: {}", getExceptionMessage(e, false)); - } - LOG_DEBUG(log, "Finished KerberosInit"); - } -#else // USE_KRB5 - if (kafka_config.has_property("sasl.kerberos.keytab") || kafka_config.has_property("sasl.kerberos.principal")) - LOG_WARNING(log, "Ignoring Kerberos-related parameters because ClickHouse was built without krb5 library support."); -#endif // USE_KRB5 - - // No need to add any prefix, messages can be distinguished - kafka_config.set_log_callback( - [this](cppkafka::KafkaHandleBase & handle, int level, const std::string & facility, const std::string & message) - { - auto [poco_level, client_logs_level] = parseSyslogLevel(level); - const auto & kafka_object_config = handle.get_configuration(); - const std::string client_id_key{"client.id"}; - chassert(kafka_object_config.has_property(client_id_key) && "Kafka configuration doesn't have expected client.id set"); - LOG_IMPL( - log, - client_logs_level, - poco_level, - "[client.id:{}] [rdk:{}] {}", - kafka_object_config.get(client_id_key), - facility, - message); - }); - - /// NOTE: statistics should be consumed, otherwise it creates too much - /// entries in the queue, that leads to memory leak and slow shutdown. - if (!kafka_config.has_property("statistics.interval.ms")) - { - // every 3 seconds by default. set to 0 to disable. - kafka_config.set("statistics.interval.ms", "3000"); - } - - // Configure interceptor to change thread name - // - // TODO: add interceptors support into the cppkafka. - // XXX: rdkafka uses pthread_set_name_np(), but glibc-compatibliity overrides it to noop. - { - // This should be safe, since we wait the rdkafka object anyway. - void * self = static_cast(this); - - int status; - - status = rd_kafka_conf_interceptor_add_on_new(kafka_config.get_handle(), - "init", StorageKafkaInterceptors::rdKafkaOnNew, self); - if (status != RD_KAFKA_RESP_ERR_NO_ERROR) - LOG_ERROR(log, "Cannot set new interceptor due to {} error", status); - - // cppkafka always copy the configuration - status = rd_kafka_conf_interceptor_add_on_conf_dup(kafka_config.get_handle(), - "init", StorageKafkaInterceptors::rdKafkaOnConfDup, self); - if (status != RD_KAFKA_RESP_ERR_NO_ERROR) - LOG_ERROR(log, "Cannot set dup conf interceptor due to {} error", status); - } -} - -void StorageKafka::updateConsumerConfiguration(cppkafka::Configuration & kafka_config) -{ - const auto & config = getContext()->getConfigRef(); - loadConsumerConfig(kafka_config, config, collection_name, CONFIG_KAFKA_TAG, topics); -} - -void StorageKafka::updateProducerConfiguration(cppkafka::Configuration & kafka_config) -{ - const auto & config = getContext()->getConfigRef(); - loadProducerConfig(kafka_config, config, collection_name, CONFIG_KAFKA_TAG, topics); -} - -bool StorageKafka::checkDependencies(const StorageID & table_id) -{ - // Check if all dependencies are attached - auto view_ids = DatabaseCatalog::instance().getDependentViews(table_id); - if (view_ids.empty()) - return true; - - // Check the dependencies are ready? - for (const auto & view_id : view_ids) - { - auto view = DatabaseCatalog::instance().tryGetTable(view_id, getContext()); - if (!view) - return false; - - // If it materialized view, check it's target table - auto * materialized_view = dynamic_cast(view.get()); - if (materialized_view && !materialized_view->tryGetTargetTable()) - return false; - - // Check all its dependencies - if (!checkDependencies(view_id)) - return false; - } - - return true; -} - void StorageKafka::threadFunc(size_t idx) { assert(idx < tasks.size()); @@ -1027,7 +517,7 @@ void StorageKafka::threadFunc(size_t idx) // Keep streaming as long as there are attached views and streaming is not cancelled while (!task->stream_cancelled) { - if (!checkDependencies(table_id)) + if (!StorageKafkaUtils::checkDependencies(table_id, getContext())) break; LOG_DEBUG(log, "Started streaming to {} attached views", num_views); @@ -1109,7 +599,7 @@ bool StorageKafka::streamToViews() /* allow_materialized */ false, /* no_squash */ true, /* no_destination */ true, - /* async_isnert */ false); + /* async_insert */ false); auto block_io = interpreter.execute(); // Create a stream for each consumer and join them in a union stream @@ -1167,164 +657,4 @@ bool StorageKafka::streamToViews() return some_stream_is_stalled; } - -void registerStorageKafka(StorageFactory & factory) -{ - auto creator_fn = [](const StorageFactory::Arguments & args) - { - ASTs & engine_args = args.engine_args; - size_t args_count = engine_args.size(); - const bool has_settings = args.storage_def->settings; - - auto kafka_settings = std::make_unique(); - String collection_name; - if (auto named_collection = tryGetNamedCollectionWithOverrides(args.engine_args, args.getLocalContext())) - { - for (const auto & setting : kafka_settings->all()) - { - const auto & setting_name = setting.getName(); - if (named_collection->has(setting_name)) - kafka_settings->set(setting_name, named_collection->get(setting_name)); - } - collection_name = assert_cast(args.engine_args[0].get())->name(); - } - - if (has_settings) - { - kafka_settings->loadFromQuery(*args.storage_def); - } - - // Check arguments and settings - #define CHECK_KAFKA_STORAGE_ARGUMENT(ARG_NUM, PAR_NAME, EVAL) \ - /* One of the four required arguments is not specified */ \ - if (args_count < (ARG_NUM) && (ARG_NUM) <= 4 && \ - !kafka_settings->PAR_NAME.changed) \ - { \ - throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH,\ - "Required parameter '{}' " \ - "for storage Kafka not specified", \ - #PAR_NAME); \ - } \ - if (args_count >= (ARG_NUM)) \ - { \ - /* The same argument is given in two places */ \ - if (has_settings && \ - kafka_settings->PAR_NAME.changed) \ - { \ - throw Exception(ErrorCodes::BAD_ARGUMENTS, \ - "The argument ā„–{} of storage Kafka " \ - "and the parameter '{}' " \ - "in SETTINGS cannot be specified at the same time", \ - #ARG_NUM, #PAR_NAME); \ - } \ - /* move engine args to settings */ \ - else \ - { \ - if ((EVAL) == 1) \ - { \ - engine_args[(ARG_NUM)-1] = \ - evaluateConstantExpressionAsLiteral( \ - engine_args[(ARG_NUM)-1], \ - args.getLocalContext()); \ - } \ - if ((EVAL) == 2) \ - { \ - engine_args[(ARG_NUM)-1] = \ - evaluateConstantExpressionOrIdentifierAsLiteral( \ - engine_args[(ARG_NUM)-1], \ - args.getLocalContext()); \ - } \ - kafka_settings->PAR_NAME = \ - engine_args[(ARG_NUM)-1]->as().value; \ - } \ - } - - /** Arguments of engine is following: - * - Kafka broker list - * - List of topics - * - Group ID (may be a constant expression with a string result) - * - Message format (string) - * - Row delimiter - * - Schema (optional, if the format supports it) - * - Number of consumers - * - Max block size for background consumption - * - Skip (at least) unreadable messages number - * - Do intermediate commits when the batch consumed and handled - */ - - /* 0 = raw, 1 = evaluateConstantExpressionAsLiteral, 2=evaluateConstantExpressionOrIdentifierAsLiteral */ - /// In case of named collection we already validated the arguments. - if (collection_name.empty()) - { - CHECK_KAFKA_STORAGE_ARGUMENT(1, kafka_broker_list, 0) - CHECK_KAFKA_STORAGE_ARGUMENT(2, kafka_topic_list, 1) - CHECK_KAFKA_STORAGE_ARGUMENT(3, kafka_group_name, 2) - CHECK_KAFKA_STORAGE_ARGUMENT(4, kafka_format, 2) - CHECK_KAFKA_STORAGE_ARGUMENT(5, kafka_row_delimiter, 2) - CHECK_KAFKA_STORAGE_ARGUMENT(6, kafka_schema, 2) - CHECK_KAFKA_STORAGE_ARGUMENT(7, kafka_num_consumers, 0) - CHECK_KAFKA_STORAGE_ARGUMENT(8, kafka_max_block_size, 0) - CHECK_KAFKA_STORAGE_ARGUMENT(9, kafka_skip_broken_messages, 0) - CHECK_KAFKA_STORAGE_ARGUMENT(10, kafka_commit_every_batch, 0) - CHECK_KAFKA_STORAGE_ARGUMENT(11, kafka_client_id, 2) - CHECK_KAFKA_STORAGE_ARGUMENT(12, kafka_poll_timeout_ms, 0) - CHECK_KAFKA_STORAGE_ARGUMENT(13, kafka_flush_interval_ms, 0) - CHECK_KAFKA_STORAGE_ARGUMENT(14, kafka_thread_per_consumer, 0) - CHECK_KAFKA_STORAGE_ARGUMENT(15, kafka_handle_error_mode, 0) - CHECK_KAFKA_STORAGE_ARGUMENT(16, kafka_commit_on_select, 0) - CHECK_KAFKA_STORAGE_ARGUMENT(17, kafka_max_rows_per_message, 0) - } - - #undef CHECK_KAFKA_STORAGE_ARGUMENT - - auto num_consumers = kafka_settings->kafka_num_consumers.value; - auto max_consumers = std::max(getNumberOfPhysicalCPUCores(), 16); - - if (!args.getLocalContext()->getSettingsRef().kafka_disable_num_consumers_limit && num_consumers > max_consumers) - { - throw Exception(ErrorCodes::BAD_ARGUMENTS, "The number of consumers can not be bigger than {}. " - "A single consumer can read any number of partitions. " - "Extra consumers are relatively expensive, " - "and using a lot of them can lead to high memory and CPU usage. " - "To achieve better performance " - "of getting data from Kafka, consider using a setting kafka_thread_per_consumer=1, " - "and ensure you have enough threads " - "in MessageBrokerSchedulePool (background_message_broker_schedule_pool_size). " - "See also https://clickhouse.com/docs/en/integrations/kafka#tuning-performance", max_consumers); - } - else if (num_consumers < 1) - { - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Number of consumers can not be lower than 1"); - } - - if (kafka_settings->kafka_max_block_size.changed && kafka_settings->kafka_max_block_size.value < 1) - { - throw Exception(ErrorCodes::BAD_ARGUMENTS, "kafka_max_block_size can not be lower than 1"); - } - - if (kafka_settings->kafka_poll_max_batch_size.changed && kafka_settings->kafka_poll_max_batch_size.value < 1) - { - throw Exception(ErrorCodes::BAD_ARGUMENTS, "kafka_poll_max_batch_size can not be lower than 1"); - } - NamesAndTypesList supported_columns; - for (const auto & column : args.columns) - { - if (column.default_desc.kind == ColumnDefaultKind::Alias) - supported_columns.emplace_back(column.name, column.type); - if (column.default_desc.kind == ColumnDefaultKind::Default && !column.default_desc.expression) - supported_columns.emplace_back(column.name, column.type); - } - // Kafka engine allows only ordinary columns without default expression or alias columns. - if (args.columns.getAll() != supported_columns) - { - throw Exception(ErrorCodes::BAD_ARGUMENTS, "KafkaEngine doesn't support DEFAULT/MATERIALIZED/EPHEMERAL expressions for columns. " - "See https://clickhouse.com/docs/en/engines/table-engines/integrations/kafka/#configuration"); - } - - 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 31e1a6076b6..966d818d675 100644 --- a/src/Storages/Kafka/StorageKafka.h +++ b/src/Storages/Kafka/StorageKafka.h @@ -23,7 +23,8 @@ class ReadFromStorageKafka; class StorageSystemKafkaConsumers; class ThreadStatus; -struct StorageKafkaInterceptors; +template +struct KafkaInterceptors; using KafkaConsumerPtr = std::shared_ptr; using ConsumerPtr = std::shared_ptr; @@ -33,7 +34,8 @@ using ConsumerPtr = std::shared_ptr; */ class StorageKafka final : public IStorage, WithContext { - friend struct StorageKafkaInterceptors; + using KafkaInterceptors = KafkaInterceptors; + friend KafkaInterceptors; public: StorageKafka( @@ -133,7 +135,6 @@ private: std::mutex thread_statuses_mutex; std::list> thread_statuses; - SettingsChanges createSettingsAdjustments(); /// Creates KafkaConsumer object without real consumer (cppkafka::Consumer) KafkaConsumerPtr createKafkaConsumer(size_t consumer_number); /// Returns full consumer related configuration, also the configuration @@ -148,33 +149,15 @@ private: std::atomic shutdown_called = false; - // Load Kafka global configuration - // https://github.com/confluentinc/librdkafka/blob/master/CONFIGURATION.md#global-configuration-properties - void updateGlobalConfiguration(cppkafka::Configuration & kafka_config); - // Load Kafka properties from consumer configuration - // NOTE: librdkafka allow to set a consumer property to a producer and vice versa, - // but a warning will be generated e.g: - // "Configuration property session.timeout.ms is a consumer property and - // will be ignored by this producer instance" - void updateConsumerConfiguration(cppkafka::Configuration & kafka_config); - // Load Kafka properties from producer configuration - void updateProducerConfiguration(cppkafka::Configuration & kafka_config); - void threadFunc(size_t idx); size_t getPollMaxBatchSize() const; size_t getMaxBlockSize() const; size_t getPollTimeoutMillisecond() const; - static Names parseTopics(String topic_list); - static String getDefaultClientId(const StorageID & table_id_); - bool streamToViews(); - bool checkDependencies(const StorageID & table_id); void cleanConsumers(); - - static VirtualColumnsDescription createVirtuals(StreamingHandleErrorMode handle_error_mode); }; } diff --git a/src/Storages/Kafka/StorageKafka2.cpp b/src/Storages/Kafka/StorageKafka2.cpp new file mode 100644 index 00000000000..3574b46e3b0 --- /dev/null +++ b/src/Storages/Kafka/StorageKafka2.cpp @@ -0,0 +1,1289 @@ +#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 CurrentMetrics +{ +// TODO: Add proper metrics, similar to old StorageKafka +extern const Metric KafkaBackgroundReads; +extern const Metric KafkaWrites; +} + +namespace ProfileEvents +{ +extern const Event KafkaBackgroundReads; +extern const Event KafkaMessagesRead; +extern const Event KafkaMessagesFailed; +extern const Event KafkaRowsRead; +extern const Event KafkaWrites; +} + + +namespace DB +{ + +namespace fs = std::filesystem; + +namespace ErrorCodes +{ +extern const int NOT_IMPLEMENTED; +extern const int LOGICAL_ERROR; +extern const int REPLICA_ALREADY_EXISTS; +extern const int TABLE_IS_DROPPED; +extern const int NO_ZOOKEEPER; +extern const int REPLICA_IS_ALREADY_ACTIVE; +} + +namespace +{ +constexpr auto MAX_FAILED_POLL_ATTEMPTS = 10; +constexpr auto MAX_TIME_TO_WAIT_FOR_ASSIGNMENT_MS = 15000; +} + +StorageKafka2::StorageKafka2( + 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()) + , keeper(getContext()->getZooKeeper()) + , keeper_path(kafka_settings_->kafka_keeper_path.value) + , replica_path(keeper_path + "/replicas/" + kafka_settings_->kafka_replica_name.value) + , kafka_settings(std::move(kafka_settings_)) + , macros_info{.table_id = table_id_} + , topics(StorageKafkaUtils::parseTopics(getContext()->getMacros()->expand(kafka_settings->kafka_topic_list.value, macros_info))) + , brokers(getContext()->getMacros()->expand(kafka_settings->kafka_broker_list.value, macros_info)) + , group(getContext()->getMacros()->expand(kafka_settings->kafka_group_name.value, macros_info)) + , client_id( + kafka_settings->kafka_client_id.value.empty() + ? StorageKafkaUtils::getDefaultClientId(table_id_) + : getContext()->getMacros()->expand(kafka_settings->kafka_client_id.value, macros_info)) + , format_name(getContext()->getMacros()->expand(kafka_settings->kafka_format.value)) + , max_rows_per_message(kafka_settings->kafka_max_rows_per_message.value) + , schema_name(getContext()->getMacros()->expand(kafka_settings->kafka_schema.value, macros_info)) + , num_consumers(kafka_settings->kafka_num_consumers.value) + , log(getLogger("StorageKafka2 (" + table_id_.getNameForLogs() + ")")) + , semaphore(0, static_cast(num_consumers)) + , settings_adjustments(StorageKafkaUtils::createSettingsAdjustments(*kafka_settings, schema_name)) + , thread_per_consumer(kafka_settings->kafka_thread_per_consumer.value) + , collection_name(collection_name_) + , active_node_identifier(toString(ServerUUID::get())) +{ + if (kafka_settings->kafka_num_consumers > 1 && !thread_per_consumer) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "With multiple consumers, it is required to use `kafka_thread_per_consumer` setting"); + + if (kafka_settings->kafka_handle_error_mode == StreamingHandleErrorMode::STREAM) + { + kafka_settings->input_format_allow_errors_num = 0; + kafka_settings->input_format_allow_errors_ratio = 0; + } + StorageInMemoryMetadata storage_metadata; + storage_metadata.setColumns(columns_); + storage_metadata.setComment(comment); + setInMemoryMetadata(storage_metadata); + setVirtuals(StorageKafkaUtils::createVirtuals(kafka_settings->kafka_handle_error_mode)); + + auto task_count = thread_per_consumer ? num_consumers : 1; + for (size_t i = 0; i < task_count; ++i) + { + auto task = getContext()->getMessageBrokerSchedulePool().createTask(log->name(), [this, i] { threadFunc(i); }); + task->deactivate(); + tasks.emplace_back(std::make_shared(std::move(task))); + } + + const auto first_replica = createTableIfNotExists(); + + if (!first_replica) + createReplica(); + + activating_task = getContext()->getSchedulePool().createTask(log->name() + "(activating task)", [this]() { activateAndReschedule(); }); + activating_task->deactivate(); +} + +void StorageKafka2::partialShutdown() +{ + // This is called in a background task within a catch block, thus this function shouldn't throw + LOG_TRACE(log, "Cancelling streams"); + for (auto & task : tasks) + { + task->stream_cancelled = true; + } + + LOG_TRACE(log, "Waiting for cleanup"); + for (auto & task : tasks) + { + task->holder->deactivate(); + } + is_active = false; +} + +bool StorageKafka2::activate() +{ + LOG_TEST(log, "Activate task"); + if (is_active && !getZooKeeper()->expired()) + { + LOG_TEST(log, "No need to activate"); + return true; + } + + if (!is_active) + { + LOG_WARNING(log, "Table was not active. Will try to activate it"); + } + else if (getZooKeeper()->expired()) + { + LOG_WARNING(log, "ZooKeeper session has expired. Switching to a new session"); + partialShutdown(); + } + else + { + UNREACHABLE(); + } + + try + { + setZooKeeper(); + } + catch (const Coordination::Exception &) + { + /// The exception when you try to zookeeper_init usually happens if DNS does not work or the connection with ZK fails + tryLogCurrentException(log, "Failed to establish a new ZK connection. Will try again"); + assert(!is_active); + return false; + } + + if (shutdown_called) + return false; + + auto activate_in_keeper = [this]() + { + try + { + auto zookeeper = getZooKeeper(); + + String is_active_path = fs::path(replica_path) / "is_active"; + zookeeper->deleteEphemeralNodeIfContentMatches(is_active_path, active_node_identifier); + + try + { + /// Simultaneously declare that this replica is active, and update the host. + zookeeper->create(is_active_path, active_node_identifier, zkutil::CreateMode::Ephemeral); + } + catch (const Coordination::Exception & e) + { + if (e.code == Coordination::Error::ZNODEEXISTS) + throw Exception( + ErrorCodes::REPLICA_IS_ALREADY_ACTIVE, + "Replica {} appears to be already active. If you're sure it's not, " + "try again in a minute or remove znode {}/is_active manually", + replica_path, + replica_path); + + throw; + } + replica_is_active_node = zkutil::EphemeralNodeHolder::existing(is_active_path, *zookeeper); + + return true; + } + catch (const Coordination::Exception & e) + { + replica_is_active_node = nullptr; + LOG_ERROR(log, "Couldn't start replica: {}. {}", e.what(), DB::getCurrentExceptionMessage(true)); + return false; + + } + catch (const Exception & e) + { + replica_is_active_node = nullptr; + if (e.code() != ErrorCodes::REPLICA_IS_ALREADY_ACTIVE) + throw; + + LOG_ERROR(log, "Couldn't start replica: {}. {}", e.what(), DB::getCurrentExceptionMessage(true)); + return false; + } + }; + + if (!activate_in_keeper()) + { + assert(!is_active); + return false; + } + + is_active = true; + + // Start the reader threads + for (auto & task : tasks) + { + task->stream_cancelled = false; + task->holder->activateAndSchedule(); + } + + LOG_DEBUG(log, "Table activated successfully"); + return true; +} + +void StorageKafka2::activateAndReschedule() +{ + if (shutdown_called) + return; + + /// It would be ideal to introduce a setting for this + constexpr static size_t check_period_ms = 60000; + /// In case of any exceptions we want to rerun the this task as fast as possible but we also don't want to keep retrying immediately + /// in a close loop (as fast as tasks can be processed), so we'll retry in between 100 and 10000 ms + const size_t backoff_ms = 100 * ((consecutive_activate_failures + 1) * (consecutive_activate_failures + 2)) / 2; + const size_t next_failure_retry_ms = std::min(size_t{10000}, backoff_ms); + + try + { + bool replica_is_active = activate(); + if (replica_is_active) + { + consecutive_activate_failures = 0; + activating_task->scheduleAfter(check_period_ms); + } + else + { + consecutive_activate_failures++; + activating_task->scheduleAfter(next_failure_retry_ms); + } + } + catch (...) + { + consecutive_activate_failures++; + activating_task->scheduleAfter(next_failure_retry_ms); + + /// We couldn't activate table let's set it into readonly mode if necessary + /// We do this after scheduling the task in case it throws + partialShutdown(); + tryLogCurrentException(log, "Failed to restart the table. Will try again"); + } +} + +void StorageKafka2::assertActive() const +{ + // TODO(antaljanosbenjamin): change LOGICAL_ERROR to something sensible + if (!is_active) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Table is not active (replica path: {})", replica_path); +} + + +Pipe StorageKafka2::read( + const Names & /*column_names */, + const StorageSnapshotPtr & /* storage_snapshot */, + SelectQueryInfo & /* query_info */, + ContextPtr /* local_context */, + QueryProcessingStage::Enum /* processed_stage */, + size_t /* max_block_size */, + size_t /* num_streams */) +{ + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Direct read from the new Kafka storage is not implemented"); +} + + +SinkToStoragePtr +StorageKafka2::write(const ASTPtr &, const StorageMetadataPtr & metadata_snapshot, ContextPtr local_context, bool /*async_insert*/) +{ + auto modified_context = Context::createCopy(local_context); + modified_context->applySettingsChanges(settings_adjustments); + + CurrentMetrics::Increment metric_increment{CurrentMetrics::KafkaWrites}; + ProfileEvents::increment(ProfileEvents::KafkaWrites); + + if (topics.size() > 1) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Can't write to Kafka table with multiple topics!"); + + cppkafka::Configuration conf = getProducerConfiguration(); + + const Settings & settings = getContext()->getSettingsRef(); + size_t poll_timeout = settings.stream_poll_timeout_ms.totalMilliseconds(); + const auto & header = metadata_snapshot->getSampleBlockNonMaterialized(); + + auto producer = std::make_unique( + std::make_shared(conf), topics[0], std::chrono::milliseconds(poll_timeout), shutdown_called, header); + + LOG_TRACE(log, "Kafka producer created"); + + size_t max_rows = max_rows_per_message; + /// Need for backward compatibility. + if (format_name == "Avro" && local_context->getSettingsRef().output_format_avro_rows_in_file.changed) + max_rows = local_context->getSettingsRef().output_format_avro_rows_in_file.value; + return std::make_shared(header, getFormatName(), max_rows, std::move(producer), getName(), modified_context); +} + +void StorageKafka2::startup() +{ + for (size_t i = 0; i < num_consumers; ++i) + { + try + { + consumers.push_back(ConsumerAndAssignmentInfo{.consumer = createConsumer(i), .keeper = getZooKeeper()}); + LOG_DEBUG(log, "Created #{} consumer", num_created_consumers); + ++num_created_consumers; + + consumers.back().consumer->subscribeIfNotSubscribedYet(); + } + catch (const cppkafka::Exception &) + { + tryLogCurrentException(log); + } + } + activating_task->activateAndSchedule(); +} + + +void StorageKafka2::shutdown(bool) +{ + shutdown_called = true; + activating_task->deactivate(); + partialShutdown(); + LOG_TRACE(log, "Closing consumers"); + consumers.clear(); + LOG_TRACE(log, "Consumers closed"); +} + +void StorageKafka2::drop() +{ + dropReplica(); +} + +KafkaConsumer2Ptr StorageKafka2::createConsumer(size_t consumer_number) +{ + // Create a consumer and subscribe to topics + auto consumer_impl = std::make_shared(getConsumerConfiguration(consumer_number)); + consumer_impl->set_destroy_flags(RD_KAFKA_DESTROY_F_NO_CONSUMER_CLOSE); + + /// NOTE: we pass |stream_cancelled| by reference here, so the buffers should not outlive the storage. + chassert((thread_per_consumer || num_consumers == 1) && "StorageKafka2 cannot handle multiple consumers on a single thread"); + auto & stream_cancelled = tasks[consumer_number]->stream_cancelled; + return std::make_shared( + consumer_impl, log, getPollMaxBatchSize(), getPollTimeoutMillisecond(), stream_cancelled, topics); + +} + + +cppkafka::Configuration StorageKafka2::getConsumerConfiguration(size_t consumer_number) +{ + KafkaConfigLoader::ConsumerConfigParams params{ + {getContext()->getConfigRef(), collection_name, topics, log}, + brokers, + group, + num_consumers > 1, + consumer_number, + client_id, + getMaxBlockSize()}; + auto kafka_config = KafkaConfigLoader::getConsumerConfiguration(*this, params); + // It is disabled, because in case of no materialized views are attached, it can cause live memory leak. To enable it, a similar cleanup mechanism must be introduced as for StorageKafka. + kafka_config.set("statistics.interval.ms", "0"); + return kafka_config; +} + +cppkafka::Configuration StorageKafka2::getProducerConfiguration() +{ + KafkaConfigLoader::ProducerConfigParams params{ + {getContext()->getConfigRef(), collection_name, topics, log}, + brokers, + client_id}; + return KafkaConfigLoader::getProducerConfiguration(*this, params); +} + +size_t StorageKafka2::getMaxBlockSize() const +{ + return kafka_settings->kafka_max_block_size.changed ? kafka_settings->kafka_max_block_size.value + : (getContext()->getSettingsRef().max_insert_block_size.value / num_consumers); +} + +size_t StorageKafka2::getPollMaxBatchSize() const +{ + size_t batch_size = kafka_settings->kafka_poll_max_batch_size.changed ? kafka_settings->kafka_poll_max_batch_size.value + : getContext()->getSettingsRef().max_block_size.value; + + return std::min(batch_size, getMaxBlockSize()); +} + +size_t StorageKafka2::getPollTimeoutMillisecond() const +{ + return kafka_settings->kafka_poll_timeout_ms.changed ? kafka_settings->kafka_poll_timeout_ms.totalMilliseconds() + : getContext()->getSettingsRef().stream_poll_timeout_ms.totalMilliseconds(); +} + +namespace +{ +const std::string lock_file_name{"lock"}; +const std::string commit_file_name{"committed"}; +const std::string intent_file_name{"intention"}; + +std::optional getNumber(zkutil::ZooKeeper & keeper, const fs::path & path) +{ + std::string result; + if (!keeper.tryGet(path, result)) + return std::nullopt; + + return DB::parse(result); +} +} + +bool StorageKafka2::createTableIfNotExists() +{ + // Heavily based on StorageReplicatedMergeTree::createTableIfNotExists + const auto my_keeper_path = fs::path(keeper_path); + const auto replicas_path = my_keeper_path / "replicas"; + + for (auto i = 0; i < 1000; ++i) + { + if (keeper->exists(replicas_path)) + { + LOG_DEBUG(log, "This table {} is already created, will add new replica", keeper_path); + return false; + } + + /// There are leftovers from incompletely dropped table. + if (keeper->exists(my_keeper_path / "dropped")) + { + /// This condition may happen when the previous drop attempt was not completed + /// or when table is dropped by another replica right now. + /// This is Ok because another replica is definitely going to drop the table. + + LOG_WARNING(log, "Removing leftovers from table {}", keeper_path); + String drop_lock_path = my_keeper_path / "dropped" / "lock"; + Coordination::Error code = keeper->tryCreate(drop_lock_path, "", zkutil::CreateMode::Ephemeral); + + if (code == Coordination::Error::ZNONODE || code == Coordination::Error::ZNODEEXISTS) + { + LOG_WARNING(log, "The leftovers from table {} were removed by another replica", keeper_path); + } + else if (code != Coordination::Error::ZOK) + { + throw Coordination::Exception::fromPath(code, drop_lock_path); + } + else + { + auto metadata_drop_lock = zkutil::EphemeralNodeHolder::existing(drop_lock_path, *keeper); + if (!removeTableNodesFromZooKeeper(keeper, metadata_drop_lock)) + { + /// Someone is recursively removing table right now, we cannot create new table until old one is removed + continue; + } + } + } + + keeper->createAncestors(keeper_path); + Coordination::Requests ops; + + ops.emplace_back(zkutil::makeCreateRequest(keeper_path, "", zkutil::CreateMode::Persistent)); + + const auto topics_path = my_keeper_path / "topics"; + ops.emplace_back(zkutil::makeCreateRequest(topics_path, "", zkutil::CreateMode::Persistent)); + + for (const auto & topic : topics) + { + LOG_DEBUG(log, "Creating path in keeper for topic {}", topic); + + const auto topic_path = topics_path / topic; + ops.emplace_back(zkutil::makeCreateRequest(topic_path, "", zkutil::CreateMode::Persistent)); + + const auto partitions_path = topic_path / "partitions"; + ops.emplace_back(zkutil::makeCreateRequest(partitions_path, "", zkutil::CreateMode::Persistent)); + } + + // Create the first replica + ops.emplace_back(zkutil::makeCreateRequest(replicas_path, "", zkutil::CreateMode::Persistent)); + ops.emplace_back(zkutil::makeCreateRequest(replica_path, "", zkutil::CreateMode::Persistent)); + + + Coordination::Responses responses; + const auto code = keeper->tryMulti(ops, responses); + if (code == Coordination::Error::ZNODEEXISTS) + { + LOG_INFO(log, "It looks like the table {} was created by another replica at the same moment, will retry", keeper_path); + continue; + } + else if (code != Coordination::Error::ZOK) + { + zkutil::KeeperMultiException::check(code, ops, responses); + } + + LOG_INFO(log, "Table {} created successfully ", keeper_path); + + return true; + } + + throw Exception( + ErrorCodes::REPLICA_ALREADY_EXISTS, + "Cannot create table, because it is created concurrently every time or because " + "of wrong zookeeper_path or because of logical error"); +} + + +bool StorageKafka2::removeTableNodesFromZooKeeper(zkutil::ZooKeeperPtr keeper_to_use, const zkutil::EphemeralNodeHolder::Ptr & drop_lock) +{ + bool completely_removed = false; + + Strings children; + if (const auto code = keeper_to_use->tryGetChildren(keeper_path, children); code == Coordination::Error::ZNONODE) + throw Exception(ErrorCodes::LOGICAL_ERROR, "There is a race condition between creation and removal. It's a bug"); + + const auto my_keeper_path = fs::path(keeper_path); + for (const auto & child : children) + if (child != "dropped") + keeper_to_use->tryRemoveRecursive(my_keeper_path / child); + + Coordination::Requests ops; + Coordination::Responses responses; + ops.emplace_back(zkutil::makeRemoveRequest(drop_lock->getPath(), -1)); + ops.emplace_back(zkutil::makeRemoveRequest(my_keeper_path / "dropped", -1)); + ops.emplace_back(zkutil::makeRemoveRequest(my_keeper_path, -1)); + const auto code = keeper_to_use->tryMulti(ops, responses, /* check_session_valid */ true); + + if (code == Coordination::Error::ZNONODE) + { + throw Exception( + ErrorCodes::LOGICAL_ERROR, "There is a race condition between creation and removal of replicated table. It's a bug"); + } + else if (code == Coordination::Error::ZNOTEMPTY) + { + LOG_ERROR( + log, + "Table was not completely removed from Keeper, {} still exists and may contain some garbage," + "but someone is removing it right now.", + keeper_path); + } + else if (code != Coordination::Error::ZOK) + { + /// It is still possible that ZooKeeper session is expired or server is killed in the middle of the delete operation. + zkutil::KeeperMultiException::check(code, ops, responses); + } + else + { + drop_lock->setAlreadyRemoved(); + completely_removed = true; + LOG_INFO(log, "Table {} was successfully removed from ZooKeeper", keeper_path); + } + + return completely_removed; +} + +void StorageKafka2::createReplica() +{ + LOG_INFO(log, "Creating replica {}", replica_path); + // TODO: This can cause issues if a new table is created with the same path. To make this work, we should store some metadata + // about the table to be able to identify that the same table is created, not a new one. + const auto code = keeper->tryCreate(replica_path, "", zkutil::CreateMode::Persistent); + + switch (code) + { + case Coordination::Error::ZNODEEXISTS: + LOG_INFO(log, "Replica {} already exists, will try to use it", replica_path); + break; + case Coordination::Error::ZOK: + LOG_INFO(log, "Replica {} created", replica_path); + break; + case Coordination::Error::ZNONODE: + throw Exception(ErrorCodes::TABLE_IS_DROPPED, "Table {} was suddenly removed", keeper_path); + default: + throw Coordination::Exception::fromPath(code, replica_path); + } +} + + +void StorageKafka2::dropReplica() +{ + LOG_INFO(log, "Trying to drop replica {}", replica_path); + auto my_keeper = getZooKeeperIfTableShutDown(); + + LOG_INFO(log, "Removing replica {}", replica_path); + + if (!my_keeper->exists(replica_path)) + { + LOG_INFO(log, "Removing replica {} does not exist", replica_path); + return; + } + + my_keeper->tryRemoveChildrenRecursive(replica_path); + + if (my_keeper->tryRemove(replica_path) != Coordination::Error::ZOK) + LOG_ERROR(log, "Replica was not completely removed from Keeper, {} still exists and may contain some garbage.", replica_path); + + /// Check that `zookeeper_path` exists: it could have been deleted by another replica after execution of previous line. + Strings replicas; + if (Coordination::Error::ZOK != my_keeper->tryGetChildren(keeper_path + "/replicas", replicas) || !replicas.empty()) + return; + + LOG_INFO(log, "{} is the last replica, will remove table", replica_path); + + /** At this moment, another replica can be created and we cannot remove the table. + * Try to remove /replicas node first. If we successfully removed it, + * it guarantees that we are the only replica that proceed to remove the table + * and no new replicas can be created after that moment (it requires the existence of /replicas node). + * and table cannot be recreated with new /replicas node on another servers while we are removing data, + * because table creation is executed in single transaction that will conflict with remaining nodes. + */ + + /// Node /dropped works like a lock that protects from concurrent removal of old table and creation of new table. + /// But recursive removal may fail in the middle of operation leaving some garbage in zookeeper_path, so + /// we remove it on table creation if there is /dropped node. Creating thread may remove /dropped node created by + /// removing thread, and it causes race condition if removing thread is not finished yet. + /// To avoid this we also create ephemeral child before starting recursive removal. + /// (The existence of child node does not allow to remove parent node). + Coordination::Requests ops; + Coordination::Responses responses; + fs::path my_keeper_path = keeper_path; + String drop_lock_path = my_keeper_path / "dropped" / "lock"; + ops.emplace_back(zkutil::makeRemoveRequest(my_keeper_path / "replicas", -1)); + ops.emplace_back(zkutil::makeCreateRequest(my_keeper_path / "dropped", "", zkutil::CreateMode::Persistent)); + ops.emplace_back(zkutil::makeCreateRequest(drop_lock_path, "", zkutil::CreateMode::Ephemeral)); + Coordination::Error code = my_keeper->tryMulti(ops, responses); + + if (code == Coordination::Error::ZNONODE || code == Coordination::Error::ZNODEEXISTS) + { + LOG_WARNING(log, "Table {} is already started to be removing by another replica right now", replica_path); + } + else if (code == Coordination::Error::ZNOTEMPTY) + { + LOG_WARNING(log, "Another replica was suddenly created, will keep the table {}", replica_path); + } + else if (code != Coordination::Error::ZOK) + { + zkutil::KeeperMultiException::check(code, ops, responses); + } + else + { + auto drop_lock = zkutil::EphemeralNodeHolder::existing(drop_lock_path, *my_keeper); + LOG_INFO(log, "Removing table {} (this might take several minutes)", keeper_path); + removeTableNodesFromZooKeeper(my_keeper, drop_lock); + } +} + +std::optional +StorageKafka2::lockTopicPartitions(zkutil::ZooKeeper & keeper_to_use, const TopicPartitions & topic_partitions) +{ + std::vector topic_partition_paths; + topic_partition_paths.reserve(topic_partitions.size()); + for (const auto & topic_partition : topic_partitions) + topic_partition_paths.emplace_back(getTopicPartitionPath(topic_partition)); + + Coordination::Requests ops; + + static constexpr auto ignore_if_exists = true; + + for (const auto & topic_partition_path : topic_partition_paths) + { + const auto lock_file_path = String(topic_partition_path / lock_file_name); + LOG_TRACE(log, "Creating locking ops for: {}", lock_file_path); + ops.push_back(zkutil::makeCreateRequest(topic_partition_path, "", zkutil::CreateMode::Persistent, ignore_if_exists)); + ops.push_back(zkutil::makeCreateRequest(lock_file_path, kafka_settings->kafka_replica_name.value, zkutil::CreateMode::Ephemeral)); + } + Coordination::Responses responses; + + if (const auto code = keeper_to_use.tryMulti(ops, responses); code != Coordination::Error::ZOK) + { + if (code != Coordination::Error::ZNODEEXISTS) + zkutil::KeeperMultiException::check(code, ops, responses); + + // Possible optimization: check the content of lock files, if we locked them, then we can clean them up and retry to lock them. + return std::nullopt; + } + + // We have the locks, let's gather the information we needed + TopicPartitionLocks locks; + { + auto tp_it = topic_partitions.begin(); + auto path_it = topic_partition_paths.begin(); + for (; tp_it != topic_partitions.end(); ++tp_it, ++path_it) + { + using zkutil::EphemeralNodeHolder; + LockedTopicPartitionInfo lock_info{ + EphemeralNodeHolder::existing(*path_it / lock_file_name, keeper_to_use), + getNumber(keeper_to_use, *path_it / commit_file_name), + getNumber(keeper_to_use, *path_it / intent_file_name)}; + + LOG_TRACE( + log, + "Locked topic partition: {}:{} at offset {} with intent size {}", + tp_it->topic, + tp_it->partition_id, + lock_info.committed_offset.value_or(0), + lock_info.intent_size.value_or(0)); + locks.emplace(TopicPartition(*tp_it), std::move(lock_info)); + } + } + + return locks; +} + + +void StorageKafka2::saveCommittedOffset(zkutil::ZooKeeper & keeper_to_use, const TopicPartition & topic_partition) +{ + const auto partition_prefix = getTopicPartitionPath(topic_partition); + keeper_to_use.createOrUpdate(partition_prefix / commit_file_name, toString(topic_partition.offset), zkutil::CreateMode::Persistent); + // This is best effort, if it fails we will try to remove in the next round + keeper_to_use.tryRemove(partition_prefix / intent_file_name, -1); + LOG_TEST( + log, "Saved offset {} for topic-partition [{}:{}]", topic_partition.offset, topic_partition.topic, topic_partition.partition_id); +} + +void StorageKafka2::saveIntent(zkutil::ZooKeeper & keeper_to_use, const TopicPartition & topic_partition, int64_t intent) +{ + LOG_TEST( + log, + "Saving intent of {} for topic-partition [{}:{}] at offset {}", + intent, + topic_partition.topic, + topic_partition.partition_id, + topic_partition.offset); + keeper_to_use.createOrUpdate( + getTopicPartitionPath(topic_partition) / intent_file_name, toString(intent), zkutil::CreateMode::Persistent); +} + + +StorageKafka2::PolledBatchInfo StorageKafka2::pollConsumer( + KafkaConsumer2 & consumer, + const TopicPartition & topic_partition, + std::optional message_count, + Stopwatch & total_stopwatch, + const ContextPtr & modified_context) +{ + LOG_TEST(log, "Polling consumer"); + PolledBatchInfo batch_info; + auto storage_snapshot = getStorageSnapshot(getInMemoryMetadataPtr(), getContext()); + Block non_virtual_header(storage_snapshot->metadata->getSampleBlockNonMaterialized()); + auto virtual_header = getVirtualsHeader(); + + // now it's one-time usage InputStream + // one block of the needed size (or with desired flush timeout) is formed in one internal iteration + // otherwise external iteration will reuse that and logic will became even more fuzzy + MutableColumns virtual_columns = virtual_header.cloneEmptyColumns(); + + auto put_error_to_stream = kafka_settings->kafka_handle_error_mode == StreamingHandleErrorMode::STREAM; + + EmptyReadBuffer empty_buf; + auto input_format = FormatFactory::instance().getInput( + getFormatName(), empty_buf, non_virtual_header, modified_context, getMaxBlockSize(), std::nullopt, 1); + + std::optional exception_message; + size_t total_rows = 0; + size_t failed_poll_attempts = 0; + + auto on_error = [&](const MutableColumns & result_columns, Exception & e) + { + ProfileEvents::increment(ProfileEvents::KafkaMessagesFailed); + + if (put_error_to_stream) + { + exception_message = e.message(); + for (const auto & column : result_columns) + { + // read_kafka_message could already push some rows to result_columns + // before exception, we need to fix it. + auto cur_rows = column->size(); + if (cur_rows > total_rows) + column->popBack(cur_rows - total_rows); + + // all data columns will get default value in case of error + column->insertDefault(); + } + + return 1; + } + else + { + e.addMessage( + "while parsing Kafka message (topic: {}, partition: {}, offset: {})'", + consumer.currentTopic(), + consumer.currentPartition(), + consumer.currentOffset()); + throw std::move(e); + } + }; + + StreamingFormatExecutor executor(non_virtual_header, input_format, std::move(on_error)); + + + Poco::Timespan max_execution_time = kafka_settings->kafka_flush_interval_ms.changed + ? kafka_settings->kafka_flush_interval_ms + : getContext()->getSettingsRef().stream_flush_interval_ms; + + const auto check_time_limit = [&max_execution_time, &total_stopwatch]() + { + if (max_execution_time != 0) + { + auto elapsed_ns = total_stopwatch.elapsed(); + + if (elapsed_ns > static_cast(max_execution_time.totalMicroseconds()) * 1000) + return false; + } + + return true; + }; + + while (true) + { + size_t new_rows = 0; + exception_message.reset(); + if (auto buf = consumer.consume(topic_partition, message_count)) + { + ProfileEvents::increment(ProfileEvents::KafkaMessagesRead); + new_rows = executor.execute(*buf); + } + + if (new_rows) + { + ProfileEvents::increment(ProfileEvents::KafkaRowsRead, new_rows); + + const auto & header_list = consumer.currentHeaderList(); + + Array headers_names; + Array headers_values; + + if (!header_list.empty()) + { + headers_names.reserve(header_list.size()); + headers_values.reserve(header_list.size()); + for (const auto & header : header_list) + { + headers_names.emplace_back(header.get_name()); + headers_values.emplace_back(static_cast(header.get_value())); + } + } + + for (size_t i = 0; i < new_rows; ++i) + { + virtual_columns[0]->insert(consumer.currentTopic()); + virtual_columns[1]->insert(consumer.currentKey()); + virtual_columns[2]->insert(consumer.currentOffset()); + virtual_columns[3]->insert(consumer.currentPartition()); + + + auto timestamp_raw = consumer.currentTimestamp(); + if (timestamp_raw) + { + auto ts = timestamp_raw->get_timestamp(); + virtual_columns[4]->insert(std::chrono::duration_cast(ts).count()); + virtual_columns[5]->insert( + DecimalField(std::chrono::duration_cast(ts).count(), 3)); + } + else + { + virtual_columns[4]->insertDefault(); + virtual_columns[5]->insertDefault(); + } + virtual_columns[6]->insert(headers_names); + virtual_columns[7]->insert(headers_values); + if (put_error_to_stream) + { + if (exception_message) + { + virtual_columns[8]->insert(consumer.currentPayload()); + virtual_columns[9]->insert(*exception_message); + } + else + { + virtual_columns[8]->insertDefault(); + virtual_columns[9]->insertDefault(); + } + } + } + + total_rows = total_rows + new_rows; + batch_info.last_offset = consumer.currentOffset(); + } + else if (consumer.isStalled()) + { + ++failed_poll_attempts; + } + else + { + // We came here in case of tombstone (or sometimes zero-length) messages, and it is not something abnormal + // TODO: it seems like in case of put_error_to_stream=true we may need to process those differently + // currently we just skip them with note in logs. + LOG_DEBUG( + log, + "Parsing of message (topic: {}, partition: {}, offset: {}) return no rows.", + consumer.currentTopic(), + consumer.currentPartition(), + consumer.currentOffset()); + } + + if (!consumer.hasMorePolledMessages() + && (total_rows >= getMaxBlockSize() || !check_time_limit() || failed_poll_attempts >= MAX_FAILED_POLL_ATTEMPTS + || consumer.needsOffsetUpdate())) + { + LOG_TRACE( + log, + "Stopped collecting message for current batch. There are {} failed polled attempts, {} total rows and consumer needs " + "offset update is {}", + failed_poll_attempts, + total_rows, + consumer.needsOffsetUpdate()); + break; + } + } + + if (total_rows == 0) + return {}; + + /// MATERIALIZED columns can be added here, but I think + // they are not needed here: + // and it's misleading to use them here, + // as columns 'materialized' that way stays 'ephemeral' + // i.e. will not be stored anywhere + // If needed any extra columns can be added using DEFAULT they can be added at MV level if needed. + + auto result_block = non_virtual_header.cloneWithColumns(executor.getResultColumns()); + auto virtual_block = virtual_header.cloneWithColumns(std::move(virtual_columns)); + + for (const auto & column : virtual_block.getColumnsWithTypeAndName()) + result_block.insert(column); + + batch_info.blocks.emplace_back(std::move(result_block)); + return batch_info; +} + +void StorageKafka2::threadFunc(size_t idx) +{ + chassert(idx < tasks.size()); + auto task = tasks[idx]; + std::optional maybe_stall_reason; + try + { + auto table_id = getStorageID(); + // Check if at least one direct dependency is attached + size_t num_views = DatabaseCatalog::instance().getDependentViews(table_id).size(); + if (num_views) + { + auto start_time = std::chrono::steady_clock::now(); + + // Keep streaming as long as there are attached views and streaming is not cancelled + while (!task->stream_cancelled && num_created_consumers > 0) + { + maybe_stall_reason.reset(); + if (!StorageKafkaUtils::checkDependencies(table_id, getContext())) + break; + + LOG_DEBUG(log, "Started streaming to {} attached views", num_views); + + // Exit the loop & reschedule if some stream stalled + if (maybe_stall_reason = streamToViews(idx); maybe_stall_reason.has_value()) + { + LOG_TRACE(log, "Stream stalled."); + break; + } + + auto ts = std::chrono::steady_clock::now(); + auto duration = std::chrono::duration_cast(ts - start_time); + if (duration.count() > KAFKA_MAX_THREAD_WORK_DURATION_MS) + { + LOG_TRACE(log, "Thread work duration limit exceeded. Reschedule."); + break; + } + } + } + } + catch (...) + { + tryLogCurrentException(__PRETTY_FUNCTION__); + } + + if (!task->stream_cancelled) + { + // Keeper related problems should be solved relatively fast, it makes sense wait less time + if (maybe_stall_reason.has_value() + && (*maybe_stall_reason == StallReason::KeeperSessionEnded || *maybe_stall_reason == StallReason::CouldNotAcquireLocks)) + task->holder->scheduleAfter(KAFKA_RESCHEDULE_MS / 10); + else + task->holder->scheduleAfter(KAFKA_RESCHEDULE_MS); + } +} + +std::optional StorageKafka2::streamToViews(size_t idx) +{ + // This function is written assuming that each consumer has their own thread. This means once this is changed, this function should be revisited. + // The return values should be revisited, as stalling all consumers because of a single one stalled is not a good idea. + auto table_id = getStorageID(); + auto table = DatabaseCatalog::instance().getTable(table_id, getContext()); + if (!table) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Engine table {} doesn't exist.", table_id.getNameForLogs()); + + CurrentMetrics::Increment metric_increment{CurrentMetrics::KafkaBackgroundReads}; + ProfileEvents::increment(ProfileEvents::KafkaBackgroundReads); + + auto & consumer_info = consumers[idx]; + consumer_info.watch.restart(); + auto & consumer = consumer_info.consumer; + // In case the initial subscribe in startup failed, let's subscribe now + consumer->subscribeIfNotSubscribedYet(); + + // To keep the consumer alive + const auto wait_for_assignment = consumer_info.locks.empty(); + LOG_TRACE(log, "Polling consumer {} for events", idx); + consumer->pollEvents(); + + if (wait_for_assignment) + { + while (nullptr == consumer->getKafkaAssignment() && consumer_info.watch.elapsedMilliseconds() < MAX_TIME_TO_WAIT_FOR_ASSIGNMENT_MS) + consumer->pollEvents(); + LOG_INFO(log, "Consumer has assignment: {}", nullptr == consumer->getKafkaAssignment()); + } + + try + { + if (consumer->needsOffsetUpdate() || consumer_info.locks.empty()) + { + LOG_TRACE(log, "Consumer needs update offset"); + // First release the locks so let other consumers acquire them ASAP + consumer_info.locks.clear(); + consumer_info.topic_partitions.clear(); + + const auto * current_assignment = consumer->getKafkaAssignment(); + if (current_assignment == nullptr) + { + // The consumer lost its assignment and haven't received a new one. + // By returning true this function reports the current consumer as a "stalled" stream, which + LOG_TRACE(log, "No assignment"); + return StallReason::NoAssignment; + } + consumer_info.consume_from_topic_partition_index = 0; + + if (consumer_info.keeper->expired()) + { + consumer_info.keeper = getZooKeeperAndAssertActive(); + LOG_TEST(log, "Got new zookeeper"); + } + + auto maybe_locks = lockTopicPartitions(*consumer_info.keeper, *current_assignment); + + if (!maybe_locks.has_value()) + { + // We couldn't acquire locks, probably some other consumers are still holding them. + LOG_TRACE(log, "Couldn't acquire locks"); + return StallReason::CouldNotAcquireLocks; + } + + consumer_info.locks = std::move(*maybe_locks); + + consumer_info.topic_partitions.reserve(current_assignment->size()); + for (const auto & topic_partition : *current_assignment) + { + TopicPartition topic_partition_copy{topic_partition}; + if (const auto & maybe_committed_offset = consumer_info.locks.at(topic_partition).committed_offset; + maybe_committed_offset.has_value()) + { + topic_partition_copy.offset = *maybe_committed_offset; + } + // in case no saved offset, we will get the offset from Kafka as a best effort. This is important to not to duplicate message when recreating the table. + + consumer_info.topic_partitions.push_back(std::move(topic_partition_copy)); + } + consumer_info.consumer->updateOffsets(consumer_info.topic_partitions); + } + + if (consumer_info.topic_partitions.empty()) + { + LOG_TRACE(log, "Consumer {} has assignment, but has no partitions, probably because there are more consumers in the consumer group than partitions.", idx); + return StallReason::NoPartitions; + } + LOG_TRACE(log, "Trying to consume from consumer {}", idx); + const auto maybe_rows = streamFromConsumer(consumer_info); + if (maybe_rows.has_value()) + { + const auto milliseconds = consumer_info.watch.elapsedMilliseconds(); + LOG_DEBUG( + log, "Pushing {} rows to {} took {} ms.", formatReadableQuantity(*maybe_rows), table_id.getNameForLogs(), milliseconds); + } + else + { + LOG_DEBUG(log, "Couldn't stream any messages"); + return StallReason::NoMessages; + } + } + catch (const zkutil::KeeperException & e) + { + if (Coordination::isHardwareError(e.code)) + { + LOG_INFO(log, "Cleaning up topic-partitions locks because of exception: {}", e.displayText()); + consumer_info.locks.clear(); + activating_task->schedule(); + return StallReason::KeeperSessionEnded; + } + + throw; + } + return {}; +} + + +std::optional StorageKafka2::streamFromConsumer(ConsumerAndAssignmentInfo & consumer_info) +{ + // Create an INSERT query for streaming data + auto insert = std::make_shared(); + insert->table_id = getStorageID(); + + auto kafka_context = Context::createCopy(getContext()); + kafka_context->makeQueryContext(); + kafka_context->applySettingsChanges(settings_adjustments); + + // 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, + /* allow_materialized */ false, + /* no_squash */ true, + /* no_destination */ true, + /* async_insert */ false); + auto block_io = interpreter.execute(); + + auto & topic_partition = consumer_info.topic_partitions[consumer_info.consume_from_topic_partition_index]; + LOG_TRACE( + log, + "Will fetch {}:{} (consume_from_topic_partition_index is {})", + topic_partition.topic, + topic_partition.partition_id, + consumer_info.consume_from_topic_partition_index); + consumer_info.consume_from_topic_partition_index + = (consumer_info.consume_from_topic_partition_index + 1) % consumer_info.topic_partitions.size(); + + bool needs_offset_reset = true; + SCOPE_EXIT({ + if (!needs_offset_reset) + return; + consumer_info.consumer->updateOffsets(consumer_info.topic_partitions); + }); + auto [blocks, last_read_offset] = pollConsumer( + *consumer_info.consumer, topic_partition, consumer_info.locks[topic_partition].intent_size, consumer_info.watch, kafka_context); + + if (blocks.empty()) + { + LOG_TRACE(log, "Didn't get any messages"); + needs_offset_reset = false; + return std::nullopt; + } + + auto converting_dag = ActionsDAG::makeConvertingActions( + blocks.front().cloneEmpty().getColumnsWithTypeAndName(), + block_io.pipeline.getHeader().getColumnsWithTypeAndName(), + ActionsDAG::MatchColumnsMode::Name); + + auto converting_actions = std::make_shared(std::move(converting_dag)); + + for (auto & block : blocks) + converting_actions->execute(block); + + // We can't cancel during copyData, as it's not aware of commits and other kafka-related stuff. + // It will be cancelled on underlying layer (kafka buffer) + + auto & keeper_to_use = *consumer_info.keeper; + auto & lock_info = consumer_info.locks.at(topic_partition); + lock_info.intent_size = last_read_offset - lock_info.committed_offset.value_or(0); + saveIntent(keeper_to_use, topic_partition, *lock_info.intent_size); + std::atomic_size_t rows = 0; + { + block_io.pipeline.complete(Pipe{std::make_shared(std::move(blocks))}); + + block_io.pipeline.setProgressCallback([&](const Progress & progress) { rows += progress.read_rows.load(); }); + CompletedPipelineExecutor executor(block_io.pipeline); + executor.execute(); + } + lock_info.committed_offset = last_read_offset + 1; + topic_partition.offset = last_read_offset + 1; + saveCommittedOffset(keeper_to_use, topic_partition); + consumer_info.consumer->commit(topic_partition); + lock_info.intent_size.reset(); + needs_offset_reset = false; + + return rows; +} + +void StorageKafka2::setZooKeeper() +{ + std::unique_lock lock{keeper_mutex}; + keeper = getContext()->getZooKeeper(); +} + +zkutil::ZooKeeperPtr StorageKafka2::tryGetZooKeeper() const +{ + std::unique_lock lock{keeper_mutex}; + return keeper; +} + +zkutil::ZooKeeperPtr StorageKafka2::getZooKeeper() const +{ + auto res = tryGetZooKeeper(); + if (!res) + throw Exception(ErrorCodes::NO_ZOOKEEPER, "Cannot get ZooKeeper"); + return res; +} + +zkutil::ZooKeeperPtr StorageKafka2::getZooKeeperAndAssertActive() const +{ + auto res = getZooKeeper(); + assertActive(); + return res; +} + +zkutil::ZooKeeperPtr StorageKafka2::getZooKeeperIfTableShutDown() const +{ + zkutil::ZooKeeperPtr new_zookeeper = getContext()->getZooKeeper(); + new_zookeeper->sync(keeper_path); + return new_zookeeper; +} + +fs::path StorageKafka2::getTopicPartitionPath(const TopicPartition & topic_partition) +{ + return fs::path(keeper_path) / "topics" / topic_partition.topic / "partitions" / std::to_string(topic_partition.partition_id); +} + +} diff --git a/src/Storages/Kafka/StorageKafka2.h b/src/Storages/Kafka/StorageKafka2.h new file mode 100644 index 00000000000..f85fedb316a --- /dev/null +++ b/src/Storages/Kafka/StorageKafka2.h @@ -0,0 +1,241 @@ +#pragma once + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#include +#include +#include +#include +#include + +namespace cppkafka +{ + +class Configuration; + +} + +namespace DB +{ + +template +struct KafkaInterceptors; + +using KafkaConsumer2Ptr = std::shared_ptr; + +/// Implements a Kafka queue table engine that can be used as a persistent queue / buffer, +/// or as a basic building block for creating pipelines with a continuous insertion / ETL. +/// +/// It is similar to the already existing StorageKafka, it instead of storing the offsets +/// in Kafka, its main source of information about offsets is Keeper. On top of the +/// offsets, it also stores the number of messages (intent size) it tried to insert from +/// each topic. By storing the intent sizes it is possible to retry the same batch of +/// messages in case of any errors and giving deduplication a chance to deduplicate +/// blocks. +/// +/// To not complicate things too much, the current implementation makes sure to fetch +/// messages only from a single topic-partition on a single thread at a time by +/// manipulating the queues of librdkafka. By pulling from multiple topic-partitions +/// the order of messages are not guaranteed, therefore they would have different +/// hashes for deduplication. +class StorageKafka2 final : public IStorage, WithContext +{ + using KafkaInterceptors = KafkaInterceptors; + friend KafkaInterceptors; + +public: + StorageKafka2( + const StorageID & table_id_, + ContextPtr context_, + const ColumnsDescription & columns_, + const String & comment, + std::unique_ptr kafka_settings_, + const String & collection_name_); + + std::string getName() const override { return "Kafka"; } + + bool noPushingToViews() const override { return true; } + + void startup() override; + void shutdown(bool is_drop) override; + + void drop() override; + + Pipe read( + 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; + + SinkToStoragePtr + write(const ASTPtr & query, const StorageMetadataPtr & /*metadata_snapshot*/, ContextPtr context, bool async_insert) override; + + /// We want to control the number of rows in a chunk inserted into Kafka + bool prefersLargeBlocks() const override { return false; } + + const auto & getFormatName() const { return format_name; } + + StreamingHandleErrorMode getHandleKafkaErrorMode() const { return kafka_settings->kafka_handle_error_mode; } + +private: + using TopicPartition = KafkaConsumer2::TopicPartition; + using TopicPartitions = KafkaConsumer2::TopicPartitions; + + struct LockedTopicPartitionInfo + { + zkutil::EphemeralNodeHolderPtr lock; + std::optional committed_offset; + std::optional intent_size; + }; + + using TopicPartitionLocks = std::unordered_map< + TopicPartition, + LockedTopicPartitionInfo, + KafkaConsumer2::OnlyTopicNameAndPartitionIdHash, + KafkaConsumer2::OnlyTopicNameAndPartitionIdEquality>; + + struct ConsumerAndAssignmentInfo + { + KafkaConsumer2Ptr consumer; + size_t consume_from_topic_partition_index{0}; + TopicPartitions topic_partitions{}; + zkutil::ZooKeeperPtr keeper; + TopicPartitionLocks locks{}; + Stopwatch watch{CLOCK_MONOTONIC_COARSE}; + }; + + struct PolledBatchInfo + { + BlocksList blocks; + int64_t last_offset; + }; + + // Stream thread + struct TaskContext + { + BackgroundSchedulePool::TaskHolder holder; + std::atomic stream_cancelled{false}; + explicit TaskContext(BackgroundSchedulePool::TaskHolder && task_) : holder(std::move(task_)) { } + }; + + enum class AssignmentChange + { + NotChanged, + Updated, + Lost + }; + + // Configuration and state + mutable std::mutex keeper_mutex; + zkutil::ZooKeeperPtr keeper; + String keeper_path; + String replica_path; + std::unique_ptr kafka_settings; + Macros::MacroExpansionInfo macros_info; + const Names topics; + const String brokers; + const String group; + const String client_id; + const String format_name; + const size_t max_rows_per_message; + const String schema_name; + const size_t num_consumers; /// total number of consumers + LoggerPtr log; + Poco::Semaphore semaphore; + const SettingsChanges settings_adjustments; + /// Can differ from num_consumers in case of exception in startup() (or if startup() hasn't been called). + /// In this case we still need to be able to shutdown() properly. + size_t num_created_consumers = 0; /// number of actually created consumers. + std::vector consumers; + std::vector> tasks; + bool thread_per_consumer = false; + /// For memory accounting in the librdkafka threads. + std::mutex thread_statuses_mutex; + std::list> thread_statuses; + /// If named_collection is specified. + String collection_name; + std::atomic shutdown_called = false; + + // Handling replica activation. + std::atomic is_active = false; + zkutil::EphemeralNodeHolderPtr replica_is_active_node; + BackgroundSchedulePool::TaskHolder activating_task; + String active_node_identifier; + UInt64 consecutive_activate_failures = 0; + bool activate(); + void activateAndReschedule(); + void partialShutdown(); + + void assertActive() const; + KafkaConsumer2Ptr createConsumer(size_t consumer_number); + // Returns full consumer related configuration, also the configuration + // contains global kafka properties. + cppkafka::Configuration getConsumerConfiguration(size_t consumer_number); + // Returns full producer related configuration, also the configuration + // contains global kafka properties. + cppkafka::Configuration getProducerConfiguration(); + + void threadFunc(size_t idx); + + size_t getPollMaxBatchSize() const; + size_t getMaxBlockSize() const; + size_t getPollTimeoutMillisecond() const; + + enum class StallReason + { + NoAssignment, + CouldNotAcquireLocks, + NoPartitions, + NoMessages, + KeeperSessionEnded, + }; + + std::optional streamToViews(size_t idx); + + std::optional streamFromConsumer(ConsumerAndAssignmentInfo & consumer_info); + + // Returns true if this is the first replica + bool createTableIfNotExists(); + // Returns true if all of the nodes were cleaned up + bool removeTableNodesFromZooKeeper(zkutil::ZooKeeperPtr keeper_to_use, const zkutil::EphemeralNodeHolder::Ptr & drop_lock); + // Creates only the replica in ZooKeeper. Shouldn't be called on the first replica as it is created in createTableIfNotExists + void createReplica(); + void dropReplica(); + + // Takes lock over topic partitions and sets the committed offset in topic_partitions. + std::optional lockTopicPartitions(zkutil::ZooKeeper & keeper_to_use, const TopicPartitions & topic_partitions); + void saveCommittedOffset(zkutil::ZooKeeper & keeper_to_use, const TopicPartition & topic_partition); + void saveIntent(zkutil::ZooKeeper & keeper_to_use, const TopicPartition & topic_partition, int64_t intent); + + PolledBatchInfo pollConsumer( + KafkaConsumer2 & consumer, + const TopicPartition & topic_partition, + std::optional message_count, + Stopwatch & watch, + const ContextPtr & context); + + void setZooKeeper(); + zkutil::ZooKeeperPtr tryGetZooKeeper() const; + zkutil::ZooKeeperPtr getZooKeeper() const; + zkutil::ZooKeeperPtr getZooKeeperAndAssertActive() const; + zkutil::ZooKeeperPtr getZooKeeperIfTableShutDown() const; + + + std::filesystem::path getTopicPartitionPath(const TopicPartition & topic_partition); +}; + +} diff --git a/src/Storages/Kafka/StorageKafkaUtils.cpp b/src/Storages/Kafka/StorageKafkaUtils.cpp new file mode 100644 index 00000000000..cdc32d775eb --- /dev/null +++ b/src/Storages/Kafka/StorageKafkaUtils.cpp @@ -0,0 +1,452 @@ +#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 + +#if USE_KRB5 +# include +#endif // USE_KRB5 + +namespace ProfileEvents +{ +extern const Event KafkaConsumerErrors; +} + +namespace DB +{ + +using namespace std::chrono_literals; + +namespace ErrorCodes +{ + extern const int BAD_ARGUMENTS; + extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; + extern const int SUPPORT_IS_DISABLED; +} + + +void registerStorageKafka(StorageFactory & factory) +{ + auto creator_fn = [](const StorageFactory::Arguments & args) -> std::shared_ptr + { + ASTs & engine_args = args.engine_args; + size_t args_count = engine_args.size(); + const bool has_settings = args.storage_def->settings; + + auto kafka_settings = std::make_unique(); + String collection_name; + if (auto named_collection = tryGetNamedCollectionWithOverrides(args.engine_args, args.getLocalContext())) + { + for (const auto & setting : kafka_settings->all()) + { + const auto & setting_name = setting.getName(); + if (named_collection->has(setting_name)) + kafka_settings->set(setting_name, named_collection->get(setting_name)); + } + collection_name = assert_cast(args.engine_args[0].get())->name(); + } + + if (has_settings) + { + kafka_settings->loadFromQuery(*args.storage_def); + } + +// Check arguments and settings +#define CHECK_KAFKA_STORAGE_ARGUMENT(ARG_NUM, PAR_NAME, EVAL) \ + /* One of the four required arguments is not specified */ \ + if (args_count < (ARG_NUM) && (ARG_NUM) <= 4 && !kafka_settings->PAR_NAME.changed) \ + { \ + throw Exception( \ + ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, \ + "Required parameter '{}' " \ + "for storage Kafka not specified", \ + #PAR_NAME); \ + } \ + if (args_count >= (ARG_NUM)) \ + { \ + /* The same argument is given in two places */ \ + if (has_settings && kafka_settings->PAR_NAME.changed) \ + { \ + throw Exception( \ + ErrorCodes::BAD_ARGUMENTS, \ + "The argument ā„–{} of storage Kafka " \ + "and the parameter '{}' " \ + "in SETTINGS cannot be specified at the same time", \ + #ARG_NUM, \ + #PAR_NAME); \ + } \ + /* move engine args to settings */ \ + else \ + { \ + if constexpr ((EVAL) == 1) \ + { \ + engine_args[(ARG_NUM)-1] = evaluateConstantExpressionAsLiteral(engine_args[(ARG_NUM)-1], args.getLocalContext()); \ + } \ + if constexpr ((EVAL) == 2) \ + { \ + engine_args[(ARG_NUM)-1] \ + = evaluateConstantExpressionOrIdentifierAsLiteral(engine_args[(ARG_NUM)-1], args.getLocalContext()); \ + } \ + kafka_settings->PAR_NAME = engine_args[(ARG_NUM)-1]->as().value; \ + } \ + } + + /** Arguments of engine is following: + * - Kafka broker list + * - List of topics + * - Group ID (may be a constant expression with a string result) + * - Message format (string) + * - Row delimiter + * - Schema (optional, if the format supports it) + * - Number of consumers + * - Max block size for background consumption + * - Skip (at least) unreadable messages number + * - Do intermediate commits when the batch consumed and handled + */ + + /* 0 = raw, 1 = evaluateConstantExpressionAsLiteral, 2=evaluateConstantExpressionOrIdentifierAsLiteral */ + /// In case of named collection we already validated the arguments. + if (collection_name.empty()) + { + CHECK_KAFKA_STORAGE_ARGUMENT(1, kafka_broker_list, 0) + CHECK_KAFKA_STORAGE_ARGUMENT(2, kafka_topic_list, 1) + CHECK_KAFKA_STORAGE_ARGUMENT(3, kafka_group_name, 2) + CHECK_KAFKA_STORAGE_ARGUMENT(4, kafka_format, 2) + CHECK_KAFKA_STORAGE_ARGUMENT(5, kafka_row_delimiter, 2) + CHECK_KAFKA_STORAGE_ARGUMENT(6, kafka_schema, 2) + CHECK_KAFKA_STORAGE_ARGUMENT(7, kafka_num_consumers, 0) + CHECK_KAFKA_STORAGE_ARGUMENT(8, kafka_max_block_size, 0) + CHECK_KAFKA_STORAGE_ARGUMENT(9, kafka_skip_broken_messages, 0) + CHECK_KAFKA_STORAGE_ARGUMENT(10, kafka_commit_every_batch, 0) + CHECK_KAFKA_STORAGE_ARGUMENT(11, kafka_client_id, 2) + CHECK_KAFKA_STORAGE_ARGUMENT(12, kafka_poll_timeout_ms, 0) + CHECK_KAFKA_STORAGE_ARGUMENT(13, kafka_flush_interval_ms, 0) + CHECK_KAFKA_STORAGE_ARGUMENT(14, kafka_thread_per_consumer, 0) + CHECK_KAFKA_STORAGE_ARGUMENT(15, kafka_handle_error_mode, 0) + CHECK_KAFKA_STORAGE_ARGUMENT(16, kafka_commit_on_select, 0) + CHECK_KAFKA_STORAGE_ARGUMENT(17, kafka_max_rows_per_message, 0) + } + +#undef CHECK_KAFKA_STORAGE_ARGUMENT + + auto num_consumers = kafka_settings->kafka_num_consumers.value; + auto max_consumers = std::max(getNumberOfPhysicalCPUCores(), 16); + + if (!args.getLocalContext()->getSettingsRef().kafka_disable_num_consumers_limit && num_consumers > max_consumers) + { + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "The number of consumers can not be bigger than {}. " + "A single consumer can read any number of partitions. " + "Extra consumers are relatively expensive, " + "and using a lot of them can lead to high memory and CPU usage. " + "To achieve better performance " + "of getting data from Kafka, consider using a setting kafka_thread_per_consumer=1, " + "and ensure you have enough threads " + "in MessageBrokerSchedulePool (background_message_broker_schedule_pool_size). " + "See also https://clickhouse.com/docs/integrations/kafka/kafka-table-engine#tuning-performance", + max_consumers); + } + else if (num_consumers < 1) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Number of consumers can not be lower than 1"); + } + + if (kafka_settings->kafka_max_block_size.changed && kafka_settings->kafka_max_block_size.value < 1) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "kafka_max_block_size can not be lower than 1"); + } + + if (kafka_settings->kafka_poll_max_batch_size.changed && kafka_settings->kafka_poll_max_batch_size.value < 1) + { + throw Exception(ErrorCodes::BAD_ARGUMENTS, "kafka_poll_max_batch_size can not be lower than 1"); + } + NamesAndTypesList supported_columns; + for (const auto & column : args.columns) + { + if (column.default_desc.kind == ColumnDefaultKind::Alias) + supported_columns.emplace_back(column.name, column.type); + if (column.default_desc.kind == ColumnDefaultKind::Default && !column.default_desc.expression) + supported_columns.emplace_back(column.name, column.type); + } + // Kafka engine allows only ordinary columns without default expression or alias columns. + if (args.columns.getAll() != supported_columns) + { + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "KafkaEngine doesn't support DEFAULT/MATERIALIZED/EPHEMERAL expressions for columns. " + "See https://clickhouse.com/docs/en/engines/table-engines/integrations/kafka/#configuration"); + } + + const auto has_keeper_path = kafka_settings->kafka_keeper_path.changed && !kafka_settings->kafka_keeper_path.value.empty(); + const auto has_replica_name = kafka_settings->kafka_replica_name.changed && !kafka_settings->kafka_replica_name.value.empty(); + + if (!has_keeper_path && !has_replica_name) + return std::make_shared( + args.table_id, args.getContext(), args.columns, args.comment, std::move(kafka_settings), collection_name); + + if (!args.getLocalContext()->getSettingsRef().allow_experimental_kafka_offsets_storage_in_keeper && !args.query.attach) + throw Exception( + ErrorCodes::SUPPORT_IS_DISABLED, + "Storing the Kafka offsets in Keeper is experimental. Set `allow_experimental_kafka_offsets_storage_in_keeper` setting " + "to enable it"); + + if (!has_keeper_path || !has_replica_name) + throw Exception( + ErrorCodes::BAD_ARGUMENTS, "Either specify both zookeeper path and replica name or none of them"); + + const auto is_on_cluster = args.getLocalContext()->getClientInfo().query_kind == ClientInfo::QueryKind::SECONDARY_QUERY; + const auto is_replicated_database = args.getLocalContext()->getClientInfo().query_kind == ClientInfo::QueryKind::SECONDARY_QUERY + && DatabaseCatalog::instance().getDatabase(args.table_id.database_name)->getEngineName() == "Replicated"; + + // UUID macro is only allowed: + // - with Atomic database only with ON CLUSTER queries, otherwise it is easy to misuse: each replica would have separate uuid generated. + // - with Replicated database + // - with attach queries, as those are used on server startup + const auto allow_uuid_macro = is_on_cluster || is_replicated_database || args.query.attach; + + auto context = args.getContext(); + // Unfold {database} and {table} macro on table creation, so table can be renamed. + if (args.mode < LoadingStrictnessLevel::ATTACH) + { + Macros::MacroExpansionInfo info; + /// NOTE: it's not recursive + info.expand_special_macros_only = true; + info.table_id = args.table_id; + // We could probably unfold UUID here too, but let's keep it similar to ReplicatedMergeTree, which doesn't do the unfolding. + info.table_id.uuid = UUIDHelpers::Nil; + kafka_settings->kafka_keeper_path.value = context->getMacros()->expand(kafka_settings->kafka_keeper_path.value, info); + + info.level = 0; + kafka_settings->kafka_replica_name.value = context->getMacros()->expand(kafka_settings->kafka_replica_name.value, info); + } + + + auto * settings_query = args.storage_def->settings; + chassert(has_settings && "Unexpected settings query in StorageKafka"); + + settings_query->changes.setSetting("kafka_keeper_path", kafka_settings->kafka_keeper_path.value); + settings_query->changes.setSetting("kafka_replica_name", kafka_settings->kafka_replica_name.value); + + // Expand other macros (such as {replica}). We do not expand them on previous step to make possible copying metadata files between replicas. + // Disable expanding {shard} macro, because it can lead to incorrect behavior and it doesn't make sense to shard Kafka tables. + Macros::MacroExpansionInfo info; + info.table_id = args.table_id; + if (is_replicated_database) + { + auto database = DatabaseCatalog::instance().getDatabase(args.table_id.database_name); + info.shard.reset(); + info.replica = getReplicatedDatabaseReplicaName(database); + } + if (!allow_uuid_macro) + info.table_id.uuid = UUIDHelpers::Nil; + kafka_settings->kafka_keeper_path.value = context->getMacros()->expand(kafka_settings->kafka_keeper_path.value, info); + + info.level = 0; + info.table_id.uuid = UUIDHelpers::Nil; + kafka_settings->kafka_replica_name.value = context->getMacros()->expand(kafka_settings->kafka_replica_name.value, info); + + 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, + }); +} + +namespace StorageKafkaUtils +{ +Names parseTopics(String topic_list) +{ + Names result; + boost::split(result, topic_list, [](char c) { return c == ','; }); + for (String & topic : result) + boost::trim(topic); + return result; +} + +String getDefaultClientId(const StorageID & table_id) +{ + return fmt::format("{}-{}-{}-{}", VERSION_NAME, getFQDNOrHostName(), table_id.database_name, table_id.table_name); +} + +void drainConsumer( + cppkafka::Consumer & consumer, const std::chrono::milliseconds drain_timeout, const LoggerPtr & log, ErrorHandler error_handler) +{ + auto start_time = std::chrono::steady_clock::now(); + cppkafka::Error last_error(RD_KAFKA_RESP_ERR_NO_ERROR); + + while (true) + { + auto msg = consumer.poll(100ms); + if (!msg) + break; + + auto error = msg.get_error(); + + if (error) + { + if (msg.is_eof() || error == last_error) + { + break; + } + else + { + LOG_ERROR(log, "Error during draining: {}", error); + error_handler(error); + } + } + + // i don't stop draining on first error, + // only if it repeats once again sequentially + last_error = error; + + auto ts = std::chrono::steady_clock::now(); + if (std::chrono::duration_cast(ts - start_time) > drain_timeout) + { + LOG_ERROR(log, "Timeout during draining."); + break; + } + } +} + +void eraseMessageErrors(Messages & messages, const LoggerPtr & log, ErrorHandler error_handler) +{ + size_t skipped = std::erase_if( + messages, + [&](auto & message) + { + if (auto error = message.get_error()) + { + ProfileEvents::increment(ProfileEvents::KafkaConsumerErrors); + LOG_ERROR(log, "Consumer error: {}", error); + error_handler(error); + return true; + } + return false; + }); + + if (skipped) + LOG_ERROR(log, "There were {} messages with an error", skipped); +} + +SettingsChanges createSettingsAdjustments(KafkaSettings & kafka_settings, const String & schema_name) +{ + SettingsChanges result; + // Needed for backward compatibility + if (!kafka_settings.input_format_skip_unknown_fields.changed) + { + // Always skip unknown fields regardless of the context (JSON or TSKV) + kafka_settings.input_format_skip_unknown_fields = true; + } + + if (!kafka_settings.input_format_allow_errors_ratio.changed) + { + kafka_settings.input_format_allow_errors_ratio = 0.; + } + + if (!kafka_settings.input_format_allow_errors_num.changed) + { + kafka_settings.input_format_allow_errors_num = kafka_settings.kafka_skip_broken_messages.value; + } + + if (!schema_name.empty()) + result.emplace_back("format_schema", schema_name); + + for (const auto & setting : kafka_settings) + { + const auto & name = setting.getName(); + if (name.find("kafka_") == std::string::npos) + result.emplace_back(name, setting.getValue()); + } + return result; +} + + +bool checkDependencies(const StorageID & table_id, const ContextPtr& context) +{ + // Check if all dependencies are attached + auto view_ids = DatabaseCatalog::instance().getDependentViews(table_id); + if (view_ids.empty()) + return true; + + // Check the dependencies are ready? + for (const auto & view_id : view_ids) + { + auto view = DatabaseCatalog::instance().tryGetTable(view_id, context); + if (!view) + return false; + + // If it materialized view, check it's target table + auto * materialized_view = dynamic_cast(view.get()); + if (materialized_view && !materialized_view->tryGetTargetTable()) + return false; + + // Check all its dependencies + if (!checkDependencies(view_id, context)) + return false; + } + + return true; +} + + +VirtualColumnsDescription createVirtuals(StreamingHandleErrorMode handle_error_mode) +{ + VirtualColumnsDescription desc; + + desc.addEphemeral("_topic", std::make_shared(std::make_shared()), ""); + desc.addEphemeral("_key", std::make_shared(), ""); + desc.addEphemeral("_offset", std::make_shared(), ""); + desc.addEphemeral("_partition", std::make_shared(), ""); + desc.addEphemeral("_timestamp", std::make_shared(std::make_shared()), ""); + desc.addEphemeral("_timestamp_ms", std::make_shared(std::make_shared(3)), ""); + desc.addEphemeral("_headers.name", std::make_shared(std::make_shared()), ""); + desc.addEphemeral("_headers.value", std::make_shared(std::make_shared()), ""); + + if (handle_error_mode == StreamingHandleErrorMode::STREAM) + { + desc.addEphemeral("_raw_message", std::make_shared(), ""); + desc.addEphemeral("_error", std::make_shared(), ""); + } + + return desc; +} +} +} diff --git a/src/Storages/Kafka/StorageKafkaUtils.h b/src/Storages/Kafka/StorageKafkaUtils.h new file mode 100644 index 00000000000..cc956dde78d --- /dev/null +++ b/src/Storages/Kafka/StorageKafkaUtils.h @@ -0,0 +1,61 @@ +#pragma once + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace Poco +{ +namespace Util +{ + class AbstractConfiguration; +} +} + +namespace DB +{ + +class VirtualColumnsDescription; +struct KafkaSettings; + +namespace StorageKafkaUtils +{ +Names parseTopics(String topic_list); +String getDefaultClientId(const StorageID & table_id); + +using ErrorHandler = std::function; + +void drainConsumer( + cppkafka::Consumer & consumer, + std::chrono::milliseconds drain_timeout, + const LoggerPtr & log, + ErrorHandler error_handler = [](const cppkafka::Error & /*err*/) {}); + +using Messages = std::vector; +void eraseMessageErrors(Messages & messages, const LoggerPtr & log, ErrorHandler error_handler = [](const cppkafka::Error & /*err*/) {}); + +SettingsChanges createSettingsAdjustments(KafkaSettings & kafka_settings, const String & schema_name); + +bool checkDependencies(const StorageID & table_id, const ContextPtr& context); + +VirtualColumnsDescription createVirtuals(StreamingHandleErrorMode handle_error_mode); +} +} + +template <> +struct fmt::formatter : fmt::ostream_formatter +{ +}; +template <> +struct fmt::formatter : fmt::ostream_formatter +{ +}; diff --git a/src/Storages/Kafka/parseSyslogLevel.cpp b/src/Storages/Kafka/parseSyslogLevel.cpp index 43630a5001f..828cffc311b 100644 --- a/src/Storages/Kafka/parseSyslogLevel.cpp +++ b/src/Storages/Kafka/parseSyslogLevel.cpp @@ -1,4 +1,5 @@ -#include "parseSyslogLevel.h" +#include + #include /// Must be in a separate compilation unit due to macros overlaps: diff --git a/src/Storages/LiveView/LiveViewEventsSource.h b/src/Storages/LiveView/LiveViewEventsSource.h index de10a98e1a2..4210acbc5bc 100644 --- a/src/Storages/LiveView/LiveViewEventsSource.h +++ b/src/Storages/LiveView/LiveViewEventsSource.h @@ -54,7 +54,7 @@ public: String getName() const override { return "LiveViewEventsSource"; } - void onCancel() override + void onCancel() noexcept override { if (storage->shutdown_called) return; diff --git a/src/Storages/LiveView/LiveViewSource.h b/src/Storages/LiveView/LiveViewSource.h index f8b428fc04d..81dd5620e57 100644 --- a/src/Storages/LiveView/LiveViewSource.h +++ b/src/Storages/LiveView/LiveViewSource.h @@ -36,7 +36,7 @@ public: String getName() const override { return "LiveViewSource"; } - void onCancel() override + void onCancel() noexcept override { if (storage->shutdown_called) return; diff --git a/src/Storages/MaterializedView/RefreshSet.cpp b/src/Storages/MaterializedView/RefreshSet.cpp index a3ef327dc24..7536f59c1e4 100644 --- a/src/Storages/MaterializedView/RefreshSet.cpp +++ b/src/Storages/MaterializedView/RefreshSet.cpp @@ -9,11 +9,6 @@ namespace CurrentMetrics namespace DB { -namespace ErrorCodes -{ - extern const int LOGICAL_ERROR; -} - RefreshSet::Handle::Handle(Handle && other) noexcept { *this = std::move(other); @@ -27,6 +22,7 @@ RefreshSet::Handle & RefreshSet::Handle::operator=(Handle && other) noexcept parent_set = std::exchange(other.parent_set, nullptr); id = std::move(other.id); dependencies = std::move(other.dependencies); + iter = std::move(other.iter); metric_increment = std::move(other.metric_increment); return *this; } @@ -39,21 +35,21 @@ RefreshSet::Handle::~Handle() void RefreshSet::Handle::rename(StorageID new_id) { std::lock_guard lock(parent_set->mutex); - parent_set->removeDependenciesLocked(id, dependencies); - auto it = parent_set->tasks.find(id); - auto task = it->second; - parent_set->tasks.erase(it); + RefreshTaskHolder task = *iter; + parent_set->removeDependenciesLocked(task, dependencies); + parent_set->removeTaskLocked(id, iter); id = new_id; - parent_set->tasks.emplace(id, task); - parent_set->addDependenciesLocked(id, dependencies); + iter = parent_set->addTaskLocked(id, task); + parent_set->addDependenciesLocked(task, dependencies); } void RefreshSet::Handle::changeDependencies(std::vector deps) { std::lock_guard lock(parent_set->mutex); - parent_set->removeDependenciesLocked(id, dependencies); + RefreshTaskHolder task = *iter; + parent_set->removeDependenciesLocked(task, dependencies); dependencies = std::move(deps); - parent_set->addDependenciesLocked(id, dependencies); + parent_set->addDependenciesLocked(task, dependencies); } void RefreshSet::Handle::reset() @@ -63,8 +59,8 @@ void RefreshSet::Handle::reset() { std::lock_guard lock(parent_set->mutex); - parent_set->removeDependenciesLocked(id, dependencies); - parent_set->tasks.erase(id); + parent_set->removeDependenciesLocked(*iter, dependencies); + parent_set->removeTaskLocked(id, iter); } parent_set = nullptr; @@ -76,37 +72,50 @@ RefreshSet::RefreshSet() = default; void RefreshSet::emplace(StorageID id, const std::vector & dependencies, RefreshTaskHolder task) { std::lock_guard guard(mutex); - auto [it, is_inserted] = tasks.emplace(id, task); - if (!is_inserted) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Refresh set entry already exists for table {}", id.getFullTableName()); - addDependenciesLocked(id, dependencies); + const auto iter = addTaskLocked(id, task); + addDependenciesLocked(task, dependencies); - task->setRefreshSetHandleUnlock(Handle(this, id, dependencies)); + task->setRefreshSetHandleUnlock(Handle(this, id, iter, dependencies)); } -void RefreshSet::addDependenciesLocked(const StorageID & id, const std::vector & dependencies) +RefreshTaskList::iterator RefreshSet::addTaskLocked(StorageID id, RefreshTaskHolder task) +{ + RefreshTaskList & list = tasks[id]; + list.push_back(task); + return std::prev(list.end()); +} + +void RefreshSet::removeTaskLocked(StorageID id, RefreshTaskList::iterator iter) +{ + const auto it = tasks.find(id); + it->second.erase(iter); + if (it->second.empty()) + tasks.erase(it); +} + +void RefreshSet::addDependenciesLocked(RefreshTaskHolder task, const std::vector & dependencies) { for (const StorageID & dep : dependencies) - dependents[dep].insert(id); + dependents[dep].insert(task); } -void RefreshSet::removeDependenciesLocked(const StorageID & id, const std::vector & dependencies) +void RefreshSet::removeDependenciesLocked(RefreshTaskHolder task, const std::vector & dependencies) { for (const StorageID & dep : dependencies) { auto & set = dependents[dep]; - set.erase(id); + set.erase(task); if (set.empty()) dependents.erase(dep); } } -RefreshTaskHolder RefreshSet::getTask(const StorageID & id) const +RefreshTaskList RefreshSet::findTasks(const StorageID & id) const { std::lock_guard lock(mutex); - if (auto task = tasks.find(id); task != tasks.end()) - return task->second; - return nullptr; + if (auto it = tasks.find(id); it != tasks.end()) + return it->second; + return {}; } RefreshSet::InfoContainer RefreshSet::getInfo() const @@ -116,26 +125,23 @@ RefreshSet::InfoContainer RefreshSet::getInfo() const lock.unlock(); InfoContainer res; - for (const auto & [id, task] : tasks_copy) - res.push_back(task->getInfo()); + for (const auto & [id, list] : tasks_copy) + for (const auto & task : list) + res.push_back(task->getInfo()); return res; } std::vector RefreshSet::getDependents(const StorageID & id) const { std::lock_guard lock(mutex); - std::vector res; auto it = dependents.find(id); if (it == dependents.end()) return {}; - for (const StorageID & dep_id : it->second) - if (auto task = tasks.find(dep_id); task != tasks.end()) - res.push_back(task->second); - return res; + return std::vector(it->second.begin(), it->second.end()); } -RefreshSet::Handle::Handle(RefreshSet * parent_set_, StorageID id_, std::vector dependencies_) +RefreshSet::Handle::Handle(RefreshSet * parent_set_, StorageID id_, RefreshTaskList::iterator iter_, std::vector dependencies_) : parent_set(parent_set_), id(std::move(id_)), dependencies(std::move(dependencies_)) - , metric_increment(CurrentMetrics::Increment(CurrentMetrics::RefreshableViews)) {} + , iter(iter_), metric_increment(CurrentMetrics::Increment(CurrentMetrics::RefreshableViews)) {} } diff --git a/src/Storages/MaterializedView/RefreshSet.h b/src/Storages/MaterializedView/RefreshSet.h index eff445023a6..7fb583fd316 100644 --- a/src/Storages/MaterializedView/RefreshSet.h +++ b/src/Storages/MaterializedView/RefreshSet.h @@ -5,12 +5,11 @@ #include #include #include +#include namespace DB { -using DatabaseAndTableNameSet = std::unordered_set; - enum class RefreshState : RefreshTaskStateUnderlying { Disabled = 0, @@ -46,8 +45,7 @@ struct RefreshInfo class RefreshSet { public: - /// RAII thing that unregisters a task and its dependencies in destructor. - /// Storage IDs must be unique. Not thread safe. + /// RAII thing that unregisters a task and its dependencies in destructor. Not thread safe. class Handle { friend class RefreshSet; @@ -73,9 +71,10 @@ public: RefreshSet * parent_set = nullptr; StorageID id = StorageID::createEmpty(); std::vector dependencies; + RefreshTaskList::iterator iter; // in parent_set->tasks[id] std::optional metric_increment; - Handle(RefreshSet * parent_set_, StorageID id_, std::vector dependencies_); + Handle(RefreshSet * parent_set_, StorageID id_, RefreshTaskList::iterator iter_, std::vector dependencies_); }; using InfoContainer = std::vector; @@ -84,7 +83,9 @@ public: void emplace(StorageID id, const std::vector & dependencies, RefreshTaskHolder task); - RefreshTaskHolder getTask(const StorageID & id) const; + /// Finds active refreshable view(s) by database and table name. + /// Normally there's at most one, but we allow name collisions here, just in case. + RefreshTaskList findTasks(const StorageID & id) const; InfoContainer getInfo() const; @@ -92,8 +93,8 @@ public: std::vector getDependents(const StorageID & id) const; private: - using TaskMap = std::unordered_map; - using DependentsMap = std::unordered_map; + using TaskMap = std::unordered_map; + using DependentsMap = std::unordered_map, StorageID::DatabaseAndTableNameHash, StorageID::DatabaseAndTableNameEqual>; /// Protects the two maps below, not locked for any nontrivial operations (e.g. operations that /// block or lock other mutexes). @@ -102,8 +103,10 @@ private: TaskMap tasks; DependentsMap dependents; - void addDependenciesLocked(const StorageID & id, const std::vector & dependencies); - void removeDependenciesLocked(const StorageID & id, const std::vector & dependencies); + RefreshTaskList::iterator addTaskLocked(StorageID id, RefreshTaskHolder task); + void removeTaskLocked(StorageID id, RefreshTaskList::iterator iter); + void addDependenciesLocked(RefreshTaskHolder task, const std::vector & dependencies); + void removeDependenciesLocked(RefreshTaskHolder task, const std::vector & dependencies); }; } diff --git a/src/Storages/MaterializedView/RefreshTask.cpp b/src/Storages/MaterializedView/RefreshTask.cpp index aa8f51d5295..0837eaf97fd 100644 --- a/src/Storages/MaterializedView/RefreshTask.cpp +++ b/src/Storages/MaterializedView/RefreshTask.cpp @@ -33,7 +33,6 @@ RefreshTask::RefreshTask( {} RefreshTaskHolder RefreshTask::create( - const StorageMaterializedView & view, ContextMutablePtr context, const DB::ASTRefreshStrategy & strategy) { @@ -46,12 +45,9 @@ RefreshTaskHolder RefreshTask::create( t->refreshTask(); }); - std::vector deps; if (strategy.dependencies) for (auto && dependency : strategy.dependencies->children) - deps.emplace_back(dependency->as()); - - context->getRefreshSet().emplace(view.getStorageID(), deps, task); + task->initial_dependencies.emplace_back(dependency->as()); return task; } @@ -61,6 +57,7 @@ void RefreshTask::initializeAndStart(std::shared_ptr vi view_to_refresh = view; if (view->getContext()->getSettingsRef().stop_refreshable_materialized_views_on_startup) stop_requested = true; + view->getContext()->getRefreshSet().emplace(view->getStorageID(), initial_dependencies, shared_from_this()); populateDependencies(); advanceNextRefreshTime(currentTime()); refresh_task->schedule(); @@ -69,7 +66,8 @@ void RefreshTask::initializeAndStart(std::shared_ptr vi void RefreshTask::rename(StorageID new_id) { std::lock_guard guard(mutex); - set_handle.rename(new_id); + if (set_handle) + set_handle.rename(new_id); } void RefreshTask::alterRefreshParams(const DB::ASTRefreshStrategy & new_strategy) diff --git a/src/Storages/MaterializedView/RefreshTask.h b/src/Storages/MaterializedView/RefreshTask.h index 1f050a97cd9..623493f6aec 100644 --- a/src/Storages/MaterializedView/RefreshTask.h +++ b/src/Storages/MaterializedView/RefreshTask.h @@ -26,7 +26,6 @@ public: /// The only proper way to construct task static RefreshTaskHolder create( - const StorageMaterializedView & view, ContextMutablePtr context, const DB::ASTRefreshStrategy & strategy); @@ -84,9 +83,11 @@ private: RefreshSchedule refresh_schedule; RefreshSettings refresh_settings; // TODO: populate, use, update on alter + std::vector initial_dependencies; RefreshSet::Handle set_handle; /// StorageIDs of our dependencies that we're waiting for. + using DatabaseAndTableNameSet = std::unordered_set; DatabaseAndTableNameSet remaining_dependencies; bool time_arrived = false; diff --git a/src/Storages/MaterializedView/RefreshTask_fwd.h b/src/Storages/MaterializedView/RefreshTask_fwd.h index 1f366962eb6..9a0a122381e 100644 --- a/src/Storages/MaterializedView/RefreshTask_fwd.h +++ b/src/Storages/MaterializedView/RefreshTask_fwd.h @@ -11,5 +11,6 @@ class RefreshTask; using RefreshTaskStateUnderlying = UInt8; using RefreshTaskHolder = std::shared_ptr; using RefreshTaskObserver = std::weak_ptr; +using RefreshTaskList = std::list; } diff --git a/src/Storages/MergeTree/ApproximateNearestNeighborIndexesCommon.cpp b/src/Storages/MergeTree/ApproximateNearestNeighborIndexesCommon.cpp deleted file mode 100644 index 7354243732c..00000000000 --- a/src/Storages/MergeTree/ApproximateNearestNeighborIndexesCommon.cpp +++ /dev/null @@ -1,507 +0,0 @@ -#include - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -namespace DB -{ - -namespace ErrorCodes -{ - extern const int LOGICAL_ERROR; - extern const int INCORRECT_QUERY; -} - -namespace -{ - -template -void extractReferenceVectorFromLiteral(ApproximateNearestNeighborInformation::Embedding & reference_vector, Literal literal) -{ - Float64 float_element_of_reference_vector; - Int64 int_element_of_reference_vector; - - for (const auto & value : literal.value()) - { - if (value.tryGet(float_element_of_reference_vector)) - reference_vector.emplace_back(float_element_of_reference_vector); - else if (value.tryGet(int_element_of_reference_vector)) - reference_vector.emplace_back(static_cast(int_element_of_reference_vector)); - else - throw Exception(ErrorCodes::INCORRECT_QUERY, "Wrong type of elements in reference vector. Only float or int are supported."); - } -} - -ApproximateNearestNeighborInformation::Metric stringToMetric(std::string_view metric) -{ - if (metric == "L2Distance") - return ApproximateNearestNeighborInformation::Metric::L2; - else if (metric == "LpDistance") - return ApproximateNearestNeighborInformation::Metric::Lp; - else - return ApproximateNearestNeighborInformation::Metric::Unknown; -} - -} - -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->getSettingsRef().max_limit_for_ann_queries) - , index_is_useful(checkQueryStructure(query_info)) -{} - -bool ApproximateNearestNeighborCondition::alwaysUnknownOrTrue(String metric) const -{ - if (!index_is_useful) - return true; // Query isn't supported - // If query is supported, check metrics for match - return !(stringToMetric(metric) == query_information->metric); -} - -float ApproximateNearestNeighborCondition::getComparisonDistanceForWhereQuery() const -{ - if (index_is_useful && query_information.has_value() - && query_information->type == ApproximateNearestNeighborInformation::Type::Where) - return query_information->distance; - throw Exception(ErrorCodes::LOGICAL_ERROR, "Not supported method for this query type"); -} - -UInt64 ApproximateNearestNeighborCondition::getLimit() const -{ - if (index_is_useful && query_information.has_value()) - return query_information->limit; - throw Exception(ErrorCodes::LOGICAL_ERROR, "No LIMIT section in query, not supported"); -} - -std::vector ApproximateNearestNeighborCondition::getReferenceVector() const -{ - if (index_is_useful && query_information.has_value()) - return query_information->reference_vector; - throw Exception(ErrorCodes::LOGICAL_ERROR, "Reference vector was requested for useless or uninitialized index."); -} - -size_t ApproximateNearestNeighborCondition::getDimensions() const -{ - if (index_is_useful && query_information.has_value()) - return query_information->reference_vector.size(); - throw Exception(ErrorCodes::LOGICAL_ERROR, "Number of dimensions was requested for useless or uninitialized index."); -} - -String ApproximateNearestNeighborCondition::getColumnName() const -{ - if (index_is_useful && query_information.has_value()) - return query_information->column_name; - throw Exception(ErrorCodes::LOGICAL_ERROR, "Column name was requested for useless or uninitialized index."); -} - -ApproximateNearestNeighborInformation::Metric ApproximateNearestNeighborCondition::getMetricType() const -{ - if (index_is_useful && query_information.has_value()) - return query_information->metric; - throw Exception(ErrorCodes::LOGICAL_ERROR, "Metric name was requested for useless or uninitialized index."); -} - -float ApproximateNearestNeighborCondition::getPValueForLpDistance() const -{ - if (index_is_useful && query_information.has_value()) - return query_information->p_for_lp_dist; - throw Exception(ErrorCodes::LOGICAL_ERROR, "P from LPDistance was requested for useless or uninitialized index."); -} - -ApproximateNearestNeighborInformation::Type ApproximateNearestNeighborCondition::getQueryType() const -{ - if (index_is_useful && query_information.has_value()) - return query_information->type; - throw Exception(ErrorCodes::LOGICAL_ERROR, "Query type was requested for useless or uninitialized index."); -} - -bool ApproximateNearestNeighborCondition::checkQueryStructure(const SelectQueryInfo & query) -{ - /// RPN-s for different sections of the query - RPN rpn_prewhere_clause; - RPN rpn_where_clause; - RPN rpn_order_by_clause; - RPNElement rpn_limit; - UInt64 limit; - - ApproximateNearestNeighborInformation prewhere_info; - ApproximateNearestNeighborInformation where_info; - ApproximateNearestNeighborInformation order_by_info; - - /// Build rpns for query sections - const auto & select = query.query->as(); - - /// If query has PREWHERE clause - if (select.prewhere()) - traverseAST(select.prewhere(), rpn_prewhere_clause); - - /// If query has WHERE clause - if (select.where()) - traverseAST(select.where(), rpn_where_clause); - - /// If query has LIMIT clause - if (select.limitLength()) - traverseAtomAST(select.limitLength(), rpn_limit); - - if (select.orderBy()) // If query has ORDERBY clause - traverseOrderByAST(select.orderBy(), rpn_order_by_clause); - - /// Reverse RPNs for conveniences during parsing - std::reverse(rpn_prewhere_clause.begin(), rpn_prewhere_clause.end()); - std::reverse(rpn_where_clause.begin(), rpn_where_clause.end()); - std::reverse(rpn_order_by_clause.begin(), rpn_order_by_clause.end()); - - /// Match rpns with supported types and extract information - const bool prewhere_is_valid = matchRPNWhere(rpn_prewhere_clause, prewhere_info); - const bool where_is_valid = matchRPNWhere(rpn_where_clause, where_info); - const bool order_by_is_valid = matchRPNOrderBy(rpn_order_by_clause, order_by_info); - const bool limit_is_valid = matchRPNLimit(rpn_limit, limit); - - /// Query without a LIMIT clause or with a limit greater than a restriction is not supported - if (!limit_is_valid || max_limit_for_ann_queries < limit) - return false; - - /// Search type query in both sections isn't supported - if (prewhere_is_valid && where_is_valid) - return false; - - /// Search type should be in WHERE or PREWHERE clause - if (prewhere_is_valid || where_is_valid) - query_information = std::move(prewhere_is_valid ? prewhere_info : where_info); - - if (order_by_is_valid) - { - /// Query with valid where and order by type is not supported - if (query_information.has_value()) - return false; - - query_information = std::move(order_by_info); - } - - if (query_information) - query_information->limit = limit; - - return query_information.has_value(); -} - -void ApproximateNearestNeighborCondition::traverseAST(const ASTPtr & node, RPN & rpn) -{ - // If the node is ASTFunction, it may have children nodes - if (const auto * func = node->as()) - { - const ASTs & children = func->arguments->children; - // Traverse children nodes - for (const auto& child : children) - traverseAST(child, rpn); - } - - RPNElement element; - /// Get the data behind node - if (!traverseAtomAST(node, element)) - element.function = RPNElement::FUNCTION_UNKNOWN; - - rpn.emplace_back(std::move(element)); -} - -bool ApproximateNearestNeighborCondition::traverseAtomAST(const ASTPtr & node, RPNElement & out) -{ - /// Match Functions - if (const auto * function = node->as()) - { - /// Set the name - out.func_name = function->name; - - if (function->name == "L1Distance" || - function->name == "L2Distance" || - function->name == "LinfDistance" || - function->name == "cosineDistance" || - function->name == "dotProduct" || - function->name == "LpDistance") - out.function = RPNElement::FUNCTION_DISTANCE; - else if (function->name == "tuple") - out.function = RPNElement::FUNCTION_TUPLE; - else if (function->name == "array") - out.function = RPNElement::FUNCTION_ARRAY; - else if (function->name == "less" || - function->name == "greater" || - function->name == "lessOrEquals" || - function->name == "greaterOrEquals") - out.function = RPNElement::FUNCTION_COMPARISON; - else if (function->name == "_CAST") - out.function = RPNElement::FUNCTION_CAST; - else - return false; - - return true; - } - /// Match identifier - else if (const auto * identifier = node->as()) - { - out.function = RPNElement::FUNCTION_IDENTIFIER; - out.identifier.emplace(identifier->name()); - out.func_name = "column identifier"; - - return true; - } - - /// Check if we have constants behind the node - return tryCastToConstType(node, out); -} - -bool ApproximateNearestNeighborCondition::tryCastToConstType(const ASTPtr & node, RPNElement & out) -{ - Field const_value; - DataTypePtr const_type; - - if (KeyCondition::getConstant(node, block_with_constants, const_value, const_type)) - { - /// Check for constant types - if (const_value.getType() == Field::Types::Float64) - { - out.function = RPNElement::FUNCTION_FLOAT_LITERAL; - out.float_literal.emplace(const_value.get()); - out.func_name = "Float literal"; - return true; - } - - if (const_value.getType() == Field::Types::UInt64) - { - out.function = RPNElement::FUNCTION_INT_LITERAL; - out.int_literal.emplace(const_value.get()); - out.func_name = "Int literal"; - return true; - } - - if (const_value.getType() == Field::Types::Int64) - { - out.function = RPNElement::FUNCTION_INT_LITERAL; - out.int_literal.emplace(const_value.get()); - out.func_name = "Int literal"; - return true; - } - - if (const_value.getType() == Field::Types::Tuple) - { - out.function = RPNElement::FUNCTION_LITERAL_TUPLE; - out.tuple_literal = const_value.get(); - out.func_name = "Tuple literal"; - return true; - } - - if (const_value.getType() == Field::Types::Array) - { - out.function = RPNElement::FUNCTION_LITERAL_ARRAY; - out.array_literal = const_value.get(); - out.func_name = "Array literal"; - return true; - } - - if (const_value.getType() == Field::Types::String) - { - out.function = RPNElement::FUNCTION_STRING_LITERAL; - out.func_name = const_value.get(); - return true; - } - } - - return false; -} - -void ApproximateNearestNeighborCondition::traverseOrderByAST(const ASTPtr & node, RPN & rpn) -{ - if (const auto * expr_list = node->as()) - if (const auto * order_by_element = expr_list->children.front()->as()) - traverseAST(order_by_element->children.front(), rpn); -} - -/// Returns true and stores ApproximateNearestNeighborInformation if the query has valid WHERE clause -bool ApproximateNearestNeighborCondition::matchRPNWhere(RPN & rpn, ApproximateNearestNeighborInformation & ann_info) -{ - /// Fill query type field - ann_info.type = ApproximateNearestNeighborInformation::Type::Where; - - /// WHERE section must have at least 5 expressions - /// Operator->Distance(float)->DistanceFunc->Column->Tuple(Array)Func(ReferenceVector(floats)) - if (rpn.size() < 5) - return false; - - auto iter = rpn.begin(); - - /// Query starts from operator less - if (iter->function != RPNElement::FUNCTION_COMPARISON) - return false; - - const bool greater_case = iter->func_name == "greater" || iter->func_name == "greaterOrEquals"; - const bool less_case = iter->func_name == "less" || iter->func_name == "lessOrEquals"; - - ++iter; - - if (less_case) - { - if (iter->function != RPNElement::FUNCTION_FLOAT_LITERAL) - return false; - - ann_info.distance = getFloatOrIntLiteralOrPanic(iter); - if (ann_info.distance < 0) - throw Exception(ErrorCodes::INCORRECT_QUERY, "Distance can't be negative. Got {}", ann_info.distance); - - ++iter; - - } - else if (!greater_case) - return false; - - auto end = rpn.end(); - if (!matchMainParts(iter, end, ann_info)) - return false; - - if (greater_case) - { - if (ann_info.reference_vector.size() < 2) - return false; - ann_info.distance = ann_info.reference_vector.back(); - if (ann_info.distance < 0) - throw Exception(ErrorCodes::INCORRECT_QUERY, "Distance can't be negative. Got {}", ann_info.distance); - ann_info.reference_vector.pop_back(); - } - - /// query is ok - return true; -} - -/// Returns true and stores ANNExpr if the query has valid ORDERBY clause -bool ApproximateNearestNeighborCondition::matchRPNOrderBy(RPN & rpn, ApproximateNearestNeighborInformation & ann_info) -{ - /// Fill query type field - ann_info.type = ApproximateNearestNeighborInformation::Type::OrderBy; - - // ORDER BY clause must have at least 3 expressions - if (rpn.size() < 3) - return false; - - auto iter = rpn.begin(); - auto end = rpn.end(); - - return ApproximateNearestNeighborCondition::matchMainParts(iter, end, ann_info); -} - -/// Returns true and stores Length if we have valid LIMIT clause in query -bool ApproximateNearestNeighborCondition::matchRPNLimit(RPNElement & rpn, UInt64 & limit) -{ - if (rpn.function == RPNElement::FUNCTION_INT_LITERAL) - { - limit = rpn.int_literal.value(); - return true; - } - - return false; -} - -/// Matches dist function, referencer vector, column name -bool ApproximateNearestNeighborCondition::matchMainParts(RPN::iterator & iter, const RPN::iterator & end, ApproximateNearestNeighborInformation & ann_info) -{ - bool identifier_found = false; - - /// Matches DistanceFunc->[Column]->[Tuple(array)Func]->ReferenceVector(floats)->[Column] - if (iter->function != RPNElement::FUNCTION_DISTANCE) - return false; - - ann_info.metric = stringToMetric(iter->func_name); - ++iter; - - if (ann_info.metric == ApproximateNearestNeighborInformation::Metric::Lp) - { - if (iter->function != RPNElement::FUNCTION_FLOAT_LITERAL && - iter->function != RPNElement::FUNCTION_INT_LITERAL) - return false; - ann_info.p_for_lp_dist = getFloatOrIntLiteralOrPanic(iter); - ++iter; - } - - if (iter->function == RPNElement::FUNCTION_IDENTIFIER) - { - identifier_found = true; - ann_info.column_name = std::move(iter->identifier.value()); - ++iter; - } - - if (iter->function == RPNElement::FUNCTION_TUPLE || iter->function == RPNElement::FUNCTION_ARRAY) - ++iter; - - if (iter->function == RPNElement::FUNCTION_LITERAL_TUPLE) - { - extractReferenceVectorFromLiteral(ann_info.reference_vector, iter->tuple_literal); - ++iter; - } - - if (iter->function == RPNElement::FUNCTION_LITERAL_ARRAY) - { - extractReferenceVectorFromLiteral(ann_info.reference_vector, iter->array_literal); - ++iter; - } - - /// further conditions are possible if there is no tuple or array, or no identifier is found - /// the tuple or array can be inside a cast function. For other cases, see the loop after this condition - if (iter != end && iter->function == RPNElement::FUNCTION_CAST) - { - ++iter; - /// Cast should be made to array or tuple - if (!iter->func_name.starts_with("Array") && !iter->func_name.starts_with("Tuple")) - return false; - ++iter; - if (iter->function == RPNElement::FUNCTION_LITERAL_TUPLE) - { - extractReferenceVectorFromLiteral(ann_info.reference_vector, iter->tuple_literal); - ++iter; - } - else if (iter->function == RPNElement::FUNCTION_LITERAL_ARRAY) - { - extractReferenceVectorFromLiteral(ann_info.reference_vector, iter->array_literal); - ++iter; - } - else - return false; - } - - while (iter != end) - { - if (iter->function == RPNElement::FUNCTION_FLOAT_LITERAL || - iter->function == RPNElement::FUNCTION_INT_LITERAL) - ann_info.reference_vector.emplace_back(getFloatOrIntLiteralOrPanic(iter)); - else if (iter->function == RPNElement::FUNCTION_IDENTIFIER) - { - if (identifier_found) - return false; - ann_info.column_name = std::move(iter->identifier.value()); - identifier_found = true; - } - else - return false; - - ++iter; - } - - /// Final checks of correctness - return identifier_found && !ann_info.reference_vector.empty(); -} - -/// Gets float or int from AST node -float ApproximateNearestNeighborCondition::getFloatOrIntLiteralOrPanic(const RPN::iterator& iter) -{ - if (iter->float_literal.has_value()) - return iter->float_literal.value(); - if (iter->int_literal.has_value()) - return static_cast(iter->int_literal.value()); - throw Exception(ErrorCodes::INCORRECT_QUERY, "Wrong parsed AST in buildRPN\n"); -} - -} diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.cpp b/src/Storages/MergeTree/IMergeTreeDataPart.cpp index 3a44359b537..195aa4fdc10 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPart.cpp +++ b/src/Storages/MergeTree/IMergeTreeDataPart.cpp @@ -428,7 +428,7 @@ std::pair IMergeTreeDataPart::getMinMaxDate() const if (storage.minmax_idx_date_column_pos != -1 && minmax_idx->initialized) { const auto & hyperrectangle = minmax_idx->hyperrectangle[storage.minmax_idx_date_column_pos]; - return {DayNum(hyperrectangle.left.get()), DayNum(hyperrectangle.right.get())}; + return {DayNum(hyperrectangle.left.safeGet()), DayNum(hyperrectangle.right.safeGet())}; } else return {}; @@ -444,15 +444,15 @@ std::pair IMergeTreeDataPart::getMinMaxTime() const if (hyperrectangle.left.getType() == Field::Types::UInt64) { assert(hyperrectangle.right.getType() == Field::Types::UInt64); - return {hyperrectangle.left.get(), hyperrectangle.right.get()}; + return {hyperrectangle.left.safeGet(), hyperrectangle.right.safeGet()}; } /// The case of DateTime64 else if (hyperrectangle.left.getType() == Field::Types::Decimal64) { assert(hyperrectangle.right.getType() == Field::Types::Decimal64); - auto left = hyperrectangle.left.get>(); - auto right = hyperrectangle.right.get>(); + auto left = hyperrectangle.left.safeGet>(); + auto right = hyperrectangle.right.safeGet>(); assert(left.getScale() == right.getScale()); @@ -749,8 +749,16 @@ void IMergeTreeDataPart::loadColumnsChecksumsIndexes(bool require_columns_checks /// 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())); + { + std::string file_info; + if (!getDataPartStorage().isDirectory(it->name())) + file_info = fmt::format(" ({} bytes)", getDataPartStorage().getFileSize(it->name())); + + files_in_part += fmt::format("{}{}{}", (files_in_part.empty() ? "" : ", "), it->name(), file_info); + + } if (!files_in_part.empty()) e->addMessage("Part contains files: {}", files_in_part); if (isEmpty()) @@ -1654,11 +1662,9 @@ void IMergeTreeDataPart::loadColumns(bool require) } -/// Project part / part with project parts / compact part doesn't support LWD. bool IMergeTreeDataPart::supportLightweightDeleteMutate() const { - return (part_type == MergeTreeDataPartType::Wide || part_type == MergeTreeDataPartType::Compact) && - parent_part == nullptr && projection_parts.empty(); + return (part_type == MergeTreeDataPartType::Wide || part_type == MergeTreeDataPartType::Compact); } bool IMergeTreeDataPart::hasLightweightDelete() const @@ -2141,7 +2147,27 @@ void IMergeTreeDataPart::checkConsistencyBase() const } } - checksums.checkSizes(getDataPartStorage()); + const auto & data_part_storage = getDataPartStorage(); + for (const auto & [filename, checksum] : checksums.files) + { + try + { + checksum.checkSize(data_part_storage, filename); + } + catch (const Exception & ex) + { + /// For projection parts check will mark them broken in loadProjections + if (!parent_part && filename.ends_with(".proj")) + { + std::string projection_name = fs::path(filename).stem(); + LOG_INFO(storage.log, "Projection {} doesn't exist on start for part {}, marking it as broken", projection_name, name); + if (hasProjection(projection_name)) + markProjectionPartAsBroken(projection_name, ex.message(), ex.code()); + } + else + throw; + } + } } else { 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/IMergeTreeReader.cpp b/src/Storages/MergeTree/IMergeTreeReader.cpp index 264b2b397f4..1f46d6b8e1b 100644 --- a/src/Storages/MergeTree/IMergeTreeReader.cpp +++ b/src/Storages/MergeTree/IMergeTreeReader.cpp @@ -3,6 +3,7 @@ #include #include #include +#include #include #include #include @@ -43,6 +44,7 @@ IMergeTreeReader::IMergeTreeReader( , alter_conversions(data_part_info_for_read->getAlterConversions()) /// For wide parts convert plain arrays of Nested to subcolumns /// to allow to use shared offset column from cache. + , original_requested_columns(columns_) , requested_columns(data_part_info_for_read->isWidePart() ? Nested::convertToSubcolumns(columns_) : columns_) @@ -75,7 +77,7 @@ void IMergeTreeReader::fillVirtualColumns(Columns & columns, size_t rows) const throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Filling of virtual columns is supported only for LoadedMergeTreeDataPartInfoForReader"); const auto & data_part = loaded_part_info->getDataPart(); - const auto & storage_columns = storage_snapshot->getMetadataForQuery()->getColumns(); + const auto & storage_columns = storage_snapshot->metadata->getColumns(); const auto & virtual_columns = storage_snapshot->virtual_columns; auto it = requested_columns.begin(); @@ -138,25 +140,33 @@ void IMergeTreeReader::evaluateMissingDefaults(Block additional_columns, Columns { try { - size_t num_columns = requested_columns.size(); + size_t num_columns = original_requested_columns.size(); if (res_columns.size() != num_columns) throw Exception(ErrorCodes::LOGICAL_ERROR, "invalid number of columns passed to MergeTreeReader::fillMissingColumns. " "Expected {}, got {}", num_columns, res_columns.size()); - /// Convert columns list to block. - /// TODO: rewrite with columns interface. It will be possible after changes in ExpressionActions. - auto name_and_type = requested_columns.begin(); - for (size_t pos = 0; pos < num_columns; ++pos, ++name_and_type) - { - if (res_columns[pos] == nullptr) - continue; + NameSet full_requested_columns_set; + NamesAndTypesList full_requested_columns; - additional_columns.insert({res_columns[pos], name_and_type->type, name_and_type->name}); + /// Convert columns list to block. And convert subcolumns to full columns. + /// Defaults should be executed on full columns to get correct values for subcolumns. + /// TODO: rewrite with columns interface. It will be possible after changes in ExpressionActions. + + auto it = original_requested_columns.begin(); + for (size_t pos = 0; pos < num_columns; ++pos, ++it) + { + auto name_in_storage = it->getNameInStorage(); + + if (full_requested_columns_set.emplace(name_in_storage).second) + full_requested_columns.emplace_back(name_in_storage, it->getTypeInStorage()); + + if (res_columns[pos]) + additional_columns.insert({res_columns[pos], it->type, it->name}); } auto dag = DB::evaluateMissingDefaults( - additional_columns, requested_columns, + additional_columns, full_requested_columns, storage_snapshot->metadata->getColumns(), data_part_info_for_read->getContext()); @@ -170,9 +180,18 @@ void IMergeTreeReader::evaluateMissingDefaults(Block additional_columns, Columns } /// Move columns from block. - name_and_type = requested_columns.begin(); - for (size_t pos = 0; pos < num_columns; ++pos, ++name_and_type) - res_columns[pos] = std::move(additional_columns.getByName(name_and_type->name).column); + it = original_requested_columns.begin(); + for (size_t pos = 0; pos < num_columns; ++pos, ++it) + { + auto name_in_storage = it->getNameInStorage(); + res_columns[pos] = additional_columns.getByName(name_in_storage).column; + + if (it->isSubcolumn()) + { + const auto & type_in_storage = it->getTypeInStorage(); + res_columns[pos] = type_in_storage->getSubcolumn(it->getSubcolumnName(), res_columns[pos]); + } + } } catch (Exception & e) { @@ -192,7 +211,12 @@ bool IMergeTreeReader::isSubcolumnOffsetsOfNested(const String & name_in_storage if (!data_part_info_for_read->isWidePart() || subcolumn_name != "size0") return false; - return Nested::isSubcolumnOfNested(name_in_storage, part_columns); + auto split = Nested::splitName(name_in_storage); + if (split.second.empty()) + return false; + + auto nested_column = part_columns.tryGetColumn(GetColumnsOptions::All, split.first); + return nested_column && isNested(nested_column->type); } String IMergeTreeReader::getColumnNameInPart(const NameAndTypePair & required_column) const diff --git a/src/Storages/MergeTree/IMergeTreeReader.h b/src/Storages/MergeTree/IMergeTreeReader.h index a1ec0339fd6..d799ce57b40 100644 --- a/src/Storages/MergeTree/IMergeTreeReader.h +++ b/src/Storages/MergeTree/IMergeTreeReader.h @@ -112,6 +112,9 @@ protected: private: /// Columns that are requested to read. + NamesAndTypesList original_requested_columns; + + /// The same as above but with converted Arrays to subcolumns of Nested. NamesAndTypesList requested_columns; /// Actual columns description in part. diff --git a/src/Storages/MergeTree/KeyCondition.cpp b/src/Storages/MergeTree/KeyCondition.cpp index dfb43c4e75d..2e57e172a7f 100644 --- a/src/Storages/MergeTree/KeyCondition.cpp +++ b/src/Storages/MergeTree/KeyCondition.cpp @@ -349,7 +349,7 @@ const KeyCondition::AtomMap KeyCondition::atom_map if (value.getType() != Field::Types::String) return false; - String prefix = extractFixedPrefixFromLikePattern(value.get(), /*requires_perfect_prefix*/ false); + String prefix = extractFixedPrefixFromLikePattern(value.safeGet(), /*requires_perfect_prefix*/ false); if (prefix.empty()) return false; @@ -370,7 +370,7 @@ const KeyCondition::AtomMap KeyCondition::atom_map if (value.getType() != Field::Types::String) return false; - String prefix = extractFixedPrefixFromLikePattern(value.get(), /*requires_perfect_prefix*/ true); + String prefix = extractFixedPrefixFromLikePattern(value.safeGet(), /*requires_perfect_prefix*/ true); if (prefix.empty()) return false; @@ -391,7 +391,7 @@ const KeyCondition::AtomMap KeyCondition::atom_map if (value.getType() != Field::Types::String) return false; - String prefix = value.get(); + String prefix = value.safeGet(); if (prefix.empty()) return false; @@ -412,7 +412,7 @@ const KeyCondition::AtomMap KeyCondition::atom_map if (value.getType() != Field::Types::String) return false; - const String & expression = value.get(); + const String & expression = value.safeGet(); /// This optimization can't process alternation - this would require /// a comprehensive parsing of regular expression. @@ -1956,11 +1956,8 @@ bool KeyCondition::extractAtomFromTree(const RPNBuilderTreeNode & node, RPNEleme auto common_type_maybe_nullable = (key_expr_type_is_nullable && !common_type->isNullable()) ? DataTypePtr(std::make_shared(common_type)) : common_type; - ColumnsWithTypeAndName arguments{ - {nullptr, key_expr_type, ""}, - {DataTypeString().createColumnConst(1, common_type_maybe_nullable->getName()), common_type_maybe_nullable, ""}}; - FunctionOverloadResolverPtr func_builder_cast = createInternalCastOverloadResolver(CastType::nonAccurate, {}); - auto func_cast = func_builder_cast->build(arguments); + + auto func_cast = createInternalCast({key_expr_type, {}}, common_type_maybe_nullable, CastType::nonAccurate, {}); /// If we know the given range only contains one value, then we treat all functions as positive monotonic. if (!single_point && !func_cast->hasInformationAboutMonotonicity()) @@ -2931,8 +2928,8 @@ BoolMask KeyCondition::checkInHyperrectangle( /// Let's support only the case of 2d, because I'm not confident in other cases. if (num_dimensions == 2) { - UInt64 left = key_range.left.get(); - UInt64 right = key_range.right.get(); + UInt64 left = key_range.left.safeGet(); + UInt64 right = key_range.right.safeGet(); BoolMask mask(false, true); auto hyperrectangle_intersection_callback = [&](std::array, 2> curve_hyperrectangle) diff --git a/src/Storages/MergeTree/KeyCondition.h b/src/Storages/MergeTree/KeyCondition.h index e9343ec08ea..8bbb86aba43 100644 --- a/src/Storages/MergeTree/KeyCondition.h +++ b/src/Storages/MergeTree/KeyCondition.h @@ -6,6 +6,8 @@ #include #include +#include + #include #include @@ -14,7 +16,6 @@ #include #include -#include "DataTypes/Serializations/ISerialization.h" namespace DB diff --git a/src/Storages/MergeTree/MergeProgress.h b/src/Storages/MergeTree/MergeProgress.h index dd4922051b5..8562e81e761 100644 --- a/src/Storages/MergeTree/MergeProgress.h +++ b/src/Storages/MergeTree/MergeProgress.h @@ -8,10 +8,10 @@ namespace ProfileEvents { - extern const Event MergesTimeMilliseconds; extern const Event MergedUncompressedBytes; extern const Event MergedRows; - extern const Event Merge; + extern const Event MutatedRows; + extern const Event MutatedUncompressedBytes; } namespace DB @@ -63,18 +63,17 @@ public: void updateWatch() { UInt64 watch_curr_elapsed = merge_list_element_ptr->watch.elapsed(); - ProfileEvents::increment(ProfileEvents::MergesTimeMilliseconds, (watch_curr_elapsed - watch_prev_elapsed) / 1000000); watch_prev_elapsed = watch_curr_elapsed; } - void operator() (const Progress & value) + void operator()(const Progress & value) { - ProfileEvents::increment(ProfileEvents::MergedUncompressedBytes, value.read_bytes); - if (stage.is_first) - { - ProfileEvents::increment(ProfileEvents::MergedRows, value.read_rows); - ProfileEvents::increment(ProfileEvents::Merge); - } + if (merge_list_element_ptr->is_mutation) + updateProfileEvents(value, ProfileEvents::MutatedRows, ProfileEvents::MutatedUncompressedBytes); + else + updateProfileEvents(value, ProfileEvents::MergedRows, ProfileEvents::MergedUncompressedBytes); + + updateWatch(); merge_list_element_ptr->bytes_read_uncompressed += value.read_bytes; @@ -90,6 +89,14 @@ public: std::memory_order_relaxed); } } + +private: + void updateProfileEvents(const Progress & value, ProfileEvents::Event rows_event, ProfileEvents::Event bytes_event) const + { + ProfileEvents::increment(bytes_event, value.read_bytes); + if (stage.is_first) + ProfileEvents::increment(rows_event, value.read_rows); + } }; } diff --git a/src/Storages/MergeTree/MergeTask.cpp b/src/Storages/MergeTree/MergeTask.cpp index fc64fae9a58..95e00773bae 100644 --- a/src/Storages/MergeTree/MergeTask.cpp +++ b/src/Storages/MergeTree/MergeTask.cpp @@ -8,6 +8,7 @@ #include #include #include +#include #include #include #include @@ -39,6 +40,18 @@ #include #include +namespace ProfileEvents +{ + extern const Event Merge; + extern const Event MergedColumns; + extern const Event GatheredColumns; + extern const Event MergeTotalMilliseconds; + extern const Event MergeExecuteMilliseconds; + extern const Event MergeHorizontalStageExecuteMilliseconds; + extern const Event MergeVerticalStageExecuteMilliseconds; + extern const Event MergeProjectionStageExecuteMilliseconds; +} + namespace DB { @@ -169,6 +182,8 @@ void MergeTask::ExecuteAndFinalizeHorizontalPart::extractMergingAndGatheringColu bool MergeTask::ExecuteAndFinalizeHorizontalPart::prepare() { + ProfileEvents::increment(ProfileEvents::Merge); + String local_tmp_prefix; if (global_ctx->need_prefix) { @@ -446,6 +461,13 @@ void MergeTask::addGatheringColumn(GlobalRuntimeContextPtr global_ctx, const Str MergeTask::StageRuntimeContextPtr MergeTask::ExecuteAndFinalizeHorizontalPart::getContextForNextStage() { + /// Do not increment for projection stage because time is already accounted in main task. + if (global_ctx->parent_part == nullptr) + { + ProfileEvents::increment(ProfileEvents::MergeExecuteMilliseconds, ctx->elapsed_execute_ns / 1000000UL); + ProfileEvents::increment(ProfileEvents::MergeHorizontalStageExecuteMilliseconds, ctx->elapsed_execute_ns / 1000000UL); + } + auto new_ctx = std::make_shared(); new_ctx->rows_sources_write_buf = std::move(ctx->rows_sources_write_buf); @@ -463,8 +485,14 @@ MergeTask::StageRuntimeContextPtr MergeTask::ExecuteAndFinalizeHorizontalPart::g MergeTask::StageRuntimeContextPtr MergeTask::VerticalMergeStage::getContextForNextStage() { - auto new_ctx = std::make_shared(); + /// Do not increment for projection stage because time is already accounted in main task. + if (global_ctx->parent_part == nullptr) + { + ProfileEvents::increment(ProfileEvents::MergeExecuteMilliseconds, ctx->elapsed_execute_ns / 1000000UL); + ProfileEvents::increment(ProfileEvents::MergeVerticalStageExecuteMilliseconds, ctx->elapsed_execute_ns / 1000000UL); + } + auto new_ctx = std::make_shared(); new_ctx->need_sync = std::move(ctx->need_sync); ctx.reset(); @@ -474,9 +502,14 @@ MergeTask::StageRuntimeContextPtr MergeTask::VerticalMergeStage::getContextForNe bool MergeTask::ExecuteAndFinalizeHorizontalPart::execute() { - assert(subtasks_iterator != subtasks.end()); - if ((this->**subtasks_iterator)()) - return true; + chassert(subtasks_iterator != subtasks.end()); + + Stopwatch watch; + bool res = (this->**subtasks_iterator)(); + ctx->elapsed_execute_ns += watch.elapsedNanoseconds(); + + if (res) + return res; /// Move to the next subtask in an array of subtasks ++subtasks_iterator; @@ -534,7 +567,7 @@ bool MergeTask::ExecuteAndFinalizeHorizontalPart::executeImpl() bool MergeTask::VerticalMergeStage::prepareVerticalMergeForAllColumns() const { - /// No need to execute this part if it is horizontal merge. + /// No need to execute this part if it is horizontal merge. if (global_ctx->chosen_merge_algorithm != MergeAlgorithm::Vertical) return false; @@ -784,6 +817,9 @@ bool MergeTask::MergeProjectionsStage::mergeMinMaxIndexAndPrepareProjections() c /// Print overall profiling info. NOTE: it may duplicates previous messages { + ProfileEvents::increment(ProfileEvents::MergedColumns, global_ctx->merging_columns.size()); + ProfileEvents::increment(ProfileEvents::GatheredColumns, global_ctx->gathering_columns.size()); + double elapsed_seconds = global_ctx->merge_list_element_ptr->watch.elapsedSeconds(); LOG_DEBUG(ctx->log, "Merge sorted {} rows, containing {} columns ({} merged, {} gathered) in {} sec., {} rows/sec., {}/sec.", @@ -797,6 +833,16 @@ bool MergeTask::MergeProjectionsStage::mergeMinMaxIndexAndPrepareProjections() c } + const auto mode = global_ctx->data->getSettings()->deduplicate_merge_projection_mode; + /// Under throw mode, we still choose to drop projections due to backward compatibility since some + /// users might have projections before this change. + if (global_ctx->data->merging_params.mode != MergeTreeData::MergingParams::Ordinary + && (mode == DeduplicateMergeProjectionMode::THROW || mode == DeduplicateMergeProjectionMode::DROP)) + { + ctx->projections_iterator = ctx->tasks_for_projections.begin(); + return false; + } + const auto & projections = global_ctx->metadata_snapshot->getProjections(); for (const auto & projection : projections) @@ -896,12 +942,29 @@ bool MergeTask::MergeProjectionsStage::finalizeProjectionsAndWholeMerge() const return false; } +MergeTask::StageRuntimeContextPtr MergeTask::MergeProjectionsStage::getContextForNextStage() +{ + /// Do not increment for projection stage because time is already accounted in main task. + /// The projection stage has its own empty projection stage which may add a drift of several milliseconds. + if (global_ctx->parent_part == nullptr) + { + ProfileEvents::increment(ProfileEvents::MergeExecuteMilliseconds, ctx->elapsed_execute_ns / 1000000UL); + ProfileEvents::increment(ProfileEvents::MergeProjectionStageExecuteMilliseconds, ctx->elapsed_execute_ns / 1000000UL); + } + + return nullptr; +} bool MergeTask::VerticalMergeStage::execute() { - assert(subtasks_iterator != subtasks.end()); - if ((this->**subtasks_iterator)()) - return true; + chassert(subtasks_iterator != subtasks.end()); + + Stopwatch watch; + bool res = (this->**subtasks_iterator)(); + ctx->elapsed_execute_ns += watch.elapsedNanoseconds(); + + if (res) + return res; /// Move to the next subtask in an array of subtasks ++subtasks_iterator; @@ -910,9 +973,14 @@ bool MergeTask::VerticalMergeStage::execute() bool MergeTask::MergeProjectionsStage::execute() { - assert(subtasks_iterator != subtasks.end()); - if ((this->**subtasks_iterator)()) - return true; + chassert(subtasks_iterator != subtasks.end()); + + Stopwatch watch; + bool res = (this->**subtasks_iterator)(); + ctx->elapsed_execute_ns += watch.elapsedNanoseconds(); + + if (res) + return res; /// Move to the next subtask in an array of subtasks ++subtasks_iterator; @@ -959,12 +1027,26 @@ bool MergeTask::VerticalMergeStage::executeVerticalMergeForAllColumns() const bool MergeTask::execute() { - assert(stages_iterator != stages.end()); - if ((*stages_iterator)->execute()) + chassert(stages_iterator != stages.end()); + const auto & current_stage = *stages_iterator; + + if (current_stage->execute()) return true; - /// Stage is finished, need initialize context for the next stage - auto next_stage_context = (*stages_iterator)->getContextForNextStage(); + /// Stage is finished, need to initialize context for the next stage and update profile events. + + UInt64 current_elapsed_ms = global_ctx->merge_list_element_ptr->watch.elapsedMilliseconds(); + UInt64 stage_elapsed_ms = current_elapsed_ms - global_ctx->prev_elapsed_ms; + global_ctx->prev_elapsed_ms = current_elapsed_ms; + + auto next_stage_context = current_stage->getContextForNextStage(); + + /// Do not increment for projection stage because time is already accounted in main task. + if (global_ctx->parent_part == nullptr) + { + ProfileEvents::increment(current_stage->getTotalTimeProfileEvent(), stage_elapsed_ms); + ProfileEvents::increment(ProfileEvents::MergeTotalMilliseconds, stage_elapsed_ms); + } /// Move to the next stage in an array of stages ++stages_iterator; @@ -1089,7 +1171,6 @@ void MergeTask::ExecuteAndFinalizeHorizontalPart::createMergedStream() /* limit_= */0, /* always_read_till_end_= */false, ctx->rows_sources_write_buf.get(), - true, ctx->blocks_are_granules_size); break; diff --git a/src/Storages/MergeTree/MergeTask.h b/src/Storages/MergeTree/MergeTask.h index 8b0f2130e8e..38ccc287187 100644 --- a/src/Storages/MergeTree/MergeTask.h +++ b/src/Storages/MergeTree/MergeTask.h @@ -3,6 +3,7 @@ #include #include +#include #include #include @@ -26,6 +27,12 @@ #include #include +namespace ProfileEvents +{ + extern const Event MergeHorizontalStageTotalMilliseconds; + extern const Event MergeVerticalStageTotalMilliseconds; + extern const Event MergeProjectionStageTotalMilliseconds; +} namespace DB { @@ -134,6 +141,7 @@ private: { virtual void setRuntimeContext(StageRuntimeContextPtr local, StageRuntimeContextPtr global) = 0; virtual StageRuntimeContextPtr getContextForNextStage() = 0; + virtual ProfileEvents::Event getTotalTimeProfileEvent() const = 0; virtual bool execute() = 0; virtual ~IStage() = default; }; @@ -195,6 +203,7 @@ private: bool need_prefix; scope_guard temporary_directory_lock; + UInt64 prev_elapsed_ms{0}; }; using GlobalRuntimeContextPtr = std::shared_ptr; @@ -233,6 +242,7 @@ private: /// Dependencies for next stages std::list::const_iterator it_name_and_type; bool need_sync{false}; + UInt64 elapsed_execute_ns{0}; }; using ExecuteAndFinalizeHorizontalPartRuntimeContextPtr = std::shared_ptr; @@ -256,7 +266,6 @@ private: ExecuteAndFinalizeHorizontalPartSubtasks::const_iterator subtasks_iterator = subtasks.begin(); - MergeAlgorithm chooseMergeAlgorithm() const; void createMergedStream(); void extractMergingAndGatheringColumns() const; @@ -268,6 +277,7 @@ private: } StageRuntimeContextPtr getContextForNextStage() override; + ProfileEvents::Event getTotalTimeProfileEvent() const override { return ProfileEvents::MergeHorizontalStageTotalMilliseconds; } ExecuteAndFinalizeHorizontalPartRuntimeContextPtr ctx; GlobalRuntimeContextPtr global_ctx; @@ -307,6 +317,7 @@ private: QueryPipeline column_parts_pipeline; std::unique_ptr executor; std::unique_ptr rows_sources_read_buf{nullptr}; + UInt64 elapsed_execute_ns{0}; }; using VerticalMergeRuntimeContextPtr = std::shared_ptr; @@ -321,6 +332,7 @@ private: global_ctx = static_pointer_cast(global); } StageRuntimeContextPtr getContextForNextStage() override; + ProfileEvents::Event getTotalTimeProfileEvent() const override { return ProfileEvents::MergeVerticalStageTotalMilliseconds; } bool prepareVerticalMergeForAllColumns() const; bool executeVerticalMergeForAllColumns() const; @@ -361,6 +373,7 @@ private: MergeTasks::iterator projections_iterator; LoggerPtr log{getLogger("MergeTask::MergeProjectionsStage")}; + UInt64 elapsed_execute_ns{0}; }; using MergeProjectionsRuntimeContextPtr = std::shared_ptr; @@ -368,12 +381,15 @@ private: struct MergeProjectionsStage : public IStage { bool execute() override; + void setRuntimeContext(StageRuntimeContextPtr local, StageRuntimeContextPtr global) override { ctx = static_pointer_cast(local); global_ctx = static_pointer_cast(global); } - StageRuntimeContextPtr getContextForNextStage() override { return nullptr; } + + StageRuntimeContextPtr getContextForNextStage() override; + ProfileEvents::Event getTotalTimeProfileEvent() const override { return ProfileEvents::MergeProjectionStageTotalMilliseconds; } bool mergeMinMaxIndexAndPrepareProjections() const; bool executeProjections() const; diff --git a/src/Storages/MergeTree/MergeTreeBlockReadUtils.cpp b/src/Storages/MergeTree/MergeTreeBlockReadUtils.cpp index aaa4ecd8eee..3dbb9d64f2f 100644 --- a/src/Storages/MergeTree/MergeTreeBlockReadUtils.cpp +++ b/src/Storages/MergeTree/MergeTreeBlockReadUtils.cpp @@ -71,8 +71,7 @@ bool injectRequiredColumnsRecursively( /// Column doesn't have default value and don't exist in part /// don't need to add to required set. - auto metadata_snapshot = storage_snapshot->getMetadataForQuery(); - const auto column_default = metadata_snapshot->getColumns().getDefault(column_name); + const auto column_default = storage_snapshot->metadata->getColumns().getDefault(column_name); if (!column_default) return false; diff --git a/src/Storages/MergeTree/MergeTreeData.cpp b/src/Storages/MergeTree/MergeTreeData.cpp index ce27ad24e10..94f6d196b99 100644 --- a/src/Storages/MergeTree/MergeTreeData.cpp +++ b/src/Storages/MergeTree/MergeTreeData.cpp @@ -1146,7 +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, /*allow_non_deterministic_functions=*/ false); + auto filter_dag = VirtualColumnUtils::splitFilterDagForAllowedInputs(filter_actions_dag.getOutputs().at(0), nullptr, /*allow_partial_result=*/ false); if (!filter_dag) return {}; @@ -2351,7 +2351,7 @@ size_t MergeTreeData::clearOldTemporaryDirectories(const String & root_path, siz /// We don't control the amount of refs for temporary parts so we cannot decide can we remove blobs /// or not. So we are not doing it bool keep_shared = false; - if (disk->supportZeroCopyReplication() && settings->allow_remote_fs_zero_copy_replication) + if (disk->supportZeroCopyReplication() && settings->allow_remote_fs_zero_copy_replication && supportsReplication()) { LOG_WARNING(log, "Since zero-copy replication is enabled we are not going to remove blobs from shared storage for {}", full_path); keep_shared = true; @@ -3213,6 +3213,17 @@ void MergeTreeData::checkAlterIsPossible(const AlterCommands & commands, Context queryToString(mutation_commands.ast())); } + /// Block the case of alter table add projection for special merge trees. + if (std::any_of(commands.begin(), commands.end(), [](const AlterCommand & c) { return c.type == AlterCommand::ADD_PROJECTION; })) + { + if (merging_params.mode != MergingParams::Mode::Ordinary + && settings_from_storage->deduplicate_merge_projection_mode == DeduplicateMergeProjectionMode::THROW) + throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, + "Projection is fully supported in {} with deduplicate_merge_projection_mode = throw. " + "Use 'drop' or 'rebuild' option of deduplicate_merge_projection_mode.", + getName()); + } + commands.apply(new_metadata, local_context); if (AlterCommands::hasFullTextIndex(new_metadata) && !settings.allow_experimental_full_text_index) @@ -3348,6 +3359,10 @@ void MergeTreeData::checkAlterIsPossible(const AlterCommands & commands, Context throw Exception(ErrorCodes::NOT_IMPLEMENTED, "ALTER MODIFY REFRESH is not supported by MergeTree engines family"); + if (command.type == AlterCommand::MODIFY_SQL_SECURITY) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, + "ALTER MODIFY SQL SECURITY is not supported by MergeTree engines family"); + if (command.type == AlterCommand::MODIFY_ORDER_BY && !is_custom_partitioned) { throw Exception(ErrorCodes::BAD_ARGUMENTS, @@ -3502,7 +3517,7 @@ void MergeTreeData::checkAlterIsPossible(const AlterCommands & commands, Context const auto & new_column = new_metadata.getColumns().get(command.column_name); if (!old_column.type->equals(*new_column.type)) throw Exception(ErrorCodes::ALTER_OF_COLUMN_IS_FORBIDDEN, - "ALTER types of column {} with statistics is not not safe " + "ALTER types of column {} with statistics is not safe " "because it can change the representation of statistics", backQuoteIfNeed(command.column_name)); } @@ -5873,7 +5888,7 @@ String MergeTreeData::getPartitionIDFromQuery(const ASTPtr & ast, ContextPtr loc if (partition_lit && partition_lit->value.getType() == Field::Types::String) { MergeTreePartInfo::validatePartitionID(partition_ast.value->clone(), format_version); - return partition_lit->value.get(); + return partition_lit->value.safeGet(); } } @@ -5936,7 +5951,7 @@ String MergeTreeData::getPartitionIDFromQuery(const ASTPtr & ast, ContextPtr loc throw Exception(ErrorCodes::INVALID_PARTITION_VALUE, "Expected tuple for complex partition key, got {}", partition_key_value.getTypeName()); - const Tuple & tuple = partition_key_value.get(); + const Tuple & tuple = partition_key_value.safeGet(); if (tuple.size() != fields_count) throw Exception(ErrorCodes::LOGICAL_ERROR, "Wrong number of fields in the partition expression: {}, must be: {}", tuple.size(), fields_count); @@ -6257,10 +6272,13 @@ void MergeTreeData::dropDetached(const ASTPtr & partition, bool part, ContextPtr } else { - String partition_id = getPartitionIDFromQuery(partition, local_context); + String partition_id; + bool all = partition->as()->all; + if (!all) + partition_id = getPartitionIDFromQuery(partition, local_context); DetachedPartsInfo detached_parts = getDetachedParts(); for (const auto & part_info : detached_parts) - if (part_info.valid_name && part_info.partition_id == partition_id + if (part_info.valid_name && (all || part_info.partition_id == partition_id) && part_info.prefix != "attaching" && part_info.prefix != "deleting") renamed_parts.addPart(part_info.dir_name, "deleting_" + part_info.dir_name, part_info.disk); } @@ -6872,7 +6890,7 @@ Block MergeTreeData::getMinMaxCountProjectionBlock( auto * place = arena.alignedAlloc(size_of_state, align_of_state); func->create(place); if (const AggregateFunctionCount * agg_count = typeid_cast(func.get())) - AggregateFunctionCount::set(place, value.get()); + AggregateFunctionCount::set(place, value.safeGet()); else { auto value_column = func->getArgumentTypes().front()->createColumnConst(1, value)->convertToFullColumnIfConst(); @@ -6914,7 +6932,8 @@ Block MergeTreeData::getMinMaxCountProjectionBlock( const auto * predicate = filter_dag->getOutputs().at(0); // Generate valid expressions for filtering - VirtualColumnUtils::filterBlockWithPredicate(predicate, virtual_columns_block, query_context); + VirtualColumnUtils::filterBlockWithPredicate( + predicate, virtual_columns_block, query_context, /*allow_filtering_with_partial_predicate =*/true); rows = virtual_columns_block.rows(); part_name_column = virtual_columns_block.getByName("_part").column; @@ -7141,12 +7160,7 @@ UInt64 MergeTreeData::estimateNumberOfRowsToRead( MergeTreeDataSelectExecutor reader(*this); auto result_ptr = reader.estimateNumMarksToRead( - parts, - storage_snapshot->getMetadataForQuery()->getColumns().getAll().getNames(), - storage_snapshot->metadata, - query_info, - query_context, - query_context->getSettingsRef().max_threads); + parts, {}, storage_snapshot->metadata, query_info, query_context, query_context->getSettingsRef().max_threads); UInt64 total_rows = result_ptr->selected_rows; if (query_info.trivial_limit > 0 && query_info.trivial_limit < total_rows) @@ -7514,7 +7528,7 @@ MergeTreeData::MatcherFn MergeTreeData::getPartitionMatcher(const ASTPtr & parti if (const auto * partition_lit = partition_ast->as().value->as()) { id = partition_lit->value.getType() == Field::Types::UInt64 - ? toString(partition_lit->value.get()) + ? toString(partition_lit->value.safeGet()) : partition_lit->value.safeGet(); prefixed = true; } diff --git a/src/Storages/MergeTree/MergeTreeDataPartChecksum.cpp b/src/Storages/MergeTree/MergeTreeDataPartChecksum.cpp index b327480fa92..3ef36ce364c 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartChecksum.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartChecksum.cpp @@ -100,12 +100,6 @@ void MergeTreeDataPartChecksums::checkEqual(const MergeTreeDataPartChecksums & r } } -void MergeTreeDataPartChecksums::checkSizes(const IDataPartStorage & storage) const -{ - for (const auto & [name, checksum] : files) - checksum.checkSize(storage, name); -} - UInt64 MergeTreeDataPartChecksums::getTotalSizeOnDisk() const { UInt64 res = 0; diff --git a/src/Storages/MergeTree/MergeTreeDataPartChecksum.h b/src/Storages/MergeTree/MergeTreeDataPartChecksum.h index 05178dc3a60..dc52f1ada2b 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartChecksum.h +++ b/src/Storages/MergeTree/MergeTreeDataPartChecksum.h @@ -65,9 +65,6 @@ struct MergeTreeDataPartChecksums static bool isBadChecksumsErrorCode(int code); - /// Checks that the directory contains all the needed files of the correct size. Does not check the checksum. - void checkSizes(const IDataPartStorage & storage) const; - /// Returns false if the checksum is too old. bool read(ReadBuffer & in); /// Assume that header with version (the first line) is read diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp index 46dd766139a..b0e70e94b73 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp @@ -255,6 +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()); } } @@ -300,22 +306,30 @@ 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 (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(); - } + if (!metadata_snapshot->hasPrimaryKey()) + return; { /** While filling index (index_columns), disable memory tracker. @@ -326,25 +340,20 @@ void MergeTreeDataPartWriterOnDisk::calculateAndSerializePrimaryIndex(const Bloc */ MemoryTrackerBlockerInThread temporarily_disable_memory_tracker; + if (index_columns.empty()) + index_columns = primary_index_block.cloneEmptyColumns(); + /// 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) @@ -421,19 +430,14 @@ 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) { index_source_hashing_stream->finalize(); 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/MergeTreeDataSelectExecutor.cpp b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp index a6ef0063069..59f3a299c99 100644 --- a/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp +++ b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp @@ -11,6 +11,7 @@ #include #include #include +#include #include #include #include @@ -48,7 +49,6 @@ #include #include -#include namespace CurrentMetrics { @@ -1406,11 +1406,10 @@ MarkRanges MergeTreeDataSelectExecutor::filterMarksUsingIndex( if (index_mark != index_range.begin || !granule || last_index_mark != index_range.begin) reader.read(granule); - auto ann_condition = std::dynamic_pointer_cast(condition); - if (ann_condition != nullptr) + if (index_helper->isVectorSimilarityIndex()) { /// An array of indices of useful ranges. - auto result = ann_condition->getUsefulRanges(granule); + auto result = condition->getUsefulRanges(granule); for (auto range : result) { diff --git a/src/Storages/MergeTree/MergeTreeDataWriter.cpp b/src/Storages/MergeTree/MergeTreeDataWriter.cpp index ee3ac4207cc..f29d715e791 100644 --- a/src/Storages/MergeTree/MergeTreeDataWriter.cpp +++ b/src/Storages/MergeTree/MergeTreeDataWriter.cpp @@ -240,6 +240,14 @@ std::vector scatterAsyncInsertInfoBySelector(AsyncInsertInfo ++offset_idx; } } + if (offset_idx != async_insert_info->offsets.size()) + { + LOG_ERROR( + getLogger("MergeTreeDataWriter"), + "ChunkInfo of async insert offsets doesn't match the selector size {}. Offsets content is ({})", + selector.size(), fmt::join(async_insert_info->offsets.begin(), async_insert_info->offsets.end(), ",")); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected error for async deduplicated insert, please check error logs"); + } return result; } @@ -317,8 +325,10 @@ BlocksWithPartition MergeTreeDataWriter::splitBlockIntoParts( { 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(), ",")); + 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 ({}). Offsets content is ({})", + i, partitions_count, fmt::join(selector.begin(), selector.end(), ","), fmt::join(async_insert_info->offsets.begin(), async_insert_info->offsets.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); @@ -441,8 +451,8 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeTempPartImpl( String part_name; if (data.format_version < MERGE_TREE_DATA_MIN_FORMAT_VERSION_WITH_CUSTOM_PARTITIONING) { - DayNum min_date(minmax_idx->hyperrectangle[data.minmax_idx_date_column_pos].left.get()); - DayNum max_date(minmax_idx->hyperrectangle[data.minmax_idx_date_column_pos].right.get()); + DayNum min_date(minmax_idx->hyperrectangle[data.minmax_idx_date_column_pos].left.safeGet()); + DayNum max_date(minmax_idx->hyperrectangle[data.minmax_idx_date_column_pos].right.safeGet()); const auto & date_lut = DateLUT::serverTimezoneInstance(); diff --git a/src/Storages/MergeTree/MergeTreeIOSettings.cpp b/src/Storages/MergeTree/MergeTreeIOSettings.cpp index 58c3bd28d6a..24cb25afe47 100644 --- a/src/Storages/MergeTree/MergeTreeIOSettings.cpp +++ b/src/Storages/MergeTree/MergeTreeIOSettings.cpp @@ -27,7 +27,6 @@ MergeTreeWriterSettings::MergeTreeWriterSettings( , 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 c79ca1e66ee..47b174b2e29 100644 --- a/src/Storages/MergeTree/MergeTreeIOSettings.h +++ b/src/Storages/MergeTree/MergeTreeIOSettings.h @@ -77,8 +77,6 @@ struct MergeTreeWriterSettings bool blocks_are_granules_size; WriteSettings query_write_settings; - size_t max_threads_for_annoy_index_creation; - 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 deleted file mode 100644 index 497e86334f3..00000000000 --- a/src/Storages/MergeTree/MergeTreeIndexAnnoy.cpp +++ /dev/null @@ -1,416 +0,0 @@ -#ifdef ENABLE_ANNOY - -#include - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - - -namespace DB -{ - -namespace ErrorCodes -{ - extern const int ILLEGAL_COLUMN; - extern const int INCORRECT_DATA; - extern const int INCORRECT_NUMBER_OF_COLUMNS; - extern const int INCORRECT_QUERY; - extern const int LOGICAL_ERROR; - extern const int NOT_IMPLEMENTED; -} - -template -AnnoyIndexWithSerialization::AnnoyIndexWithSerialization(size_t dimensions) - : Base::AnnoyIndex(static_cast(dimensions)) -{ -} - -template -void AnnoyIndexWithSerialization::serialize(WriteBuffer & ostr) const -{ - chassert(Base::_built); - writeIntBinary(Base::_s, ostr); - writeIntBinary(Base::_n_items, ostr); - writeIntBinary(Base::_n_nodes, ostr); - writeIntBinary(Base::_nodes_size, ostr); - writeIntBinary(Base::_K, ostr); - writeIntBinary(Base::_seed, ostr); - writeVectorBinary(Base::_roots, ostr); - ostr.write(reinterpret_cast(Base::_nodes), Base::_s * Base::_n_nodes); -} - -template -void AnnoyIndexWithSerialization::deserialize(ReadBuffer & istr) -{ - chassert(!Base::_built); - readIntBinary(Base::_s, istr); - readIntBinary(Base::_n_items, istr); - readIntBinary(Base::_n_nodes, istr); - readIntBinary(Base::_nodes_size, istr); - readIntBinary(Base::_K, istr); - readIntBinary(Base::_seed, istr); - readVectorBinary(Base::_roots, istr); - Base::_nodes = realloc(Base::_nodes, Base::_s * Base::_n_nodes); - istr.readStrict(reinterpret_cast(Base::_nodes), Base::_s * Base::_n_nodes); - - Base::_fd = 0; - // set flags - Base::_loaded = false; - Base::_verbose = false; - Base::_on_disk = false; - Base::_built = true; -} - -template -size_t AnnoyIndexWithSerialization::getDimensions() const -{ - return Base::get_f(); -} - - -template -MergeTreeIndexGranuleAnnoy::MergeTreeIndexGranuleAnnoy(const String & index_name_, const Block & index_sample_block_) - : index_name(index_name_) - , index_sample_block(index_sample_block_) - , index(nullptr) -{} - -template -MergeTreeIndexGranuleAnnoy::MergeTreeIndexGranuleAnnoy( - const String & index_name_, - const Block & index_sample_block_, - AnnoyIndexWithSerializationPtr index_) - : index_name(index_name_) - , index_sample_block(index_sample_block_) - , index(std::move(index_)) -{} - -template -void MergeTreeIndexGranuleAnnoy::serializeBinary(WriteBuffer & ostr) const -{ - /// Number of dimensions is required in the index constructor, - /// so it must be written and read separately from the other part - writeIntBinary(static_cast(index->getDimensions()), ostr); // write dimension - index->serialize(ostr); -} - -template -void MergeTreeIndexGranuleAnnoy::deserializeBinary(ReadBuffer & istr, MergeTreeIndexVersion /*version*/) -{ - UInt64 dimension; - readIntBinary(dimension, istr); - index = std::make_shared>(dimension); - index->deserialize(istr); -} - -template -MergeTreeIndexAggregatorAnnoy::MergeTreeIndexAggregatorAnnoy( - const String & index_name_, - const Block & index_sample_block_, - UInt64 trees_, - size_t max_threads_for_creation_) - : index_name(index_name_) - , index_sample_block(index_sample_block_) - , trees(trees_) - , max_threads_for_creation(max_threads_for_creation_) -{} - -template -MergeTreeIndexGranulePtr MergeTreeIndexAggregatorAnnoy::getGranuleAndReset() -{ - int threads = (max_threads_for_creation == 0) ? -1 : static_cast(max_threads_for_creation); - /// clang-tidy reports a false positive: it considers %p with an outdated pointer in fprintf() (used by logging which we don't do) dereferencing - index->build(static_cast(trees), threads); - auto granule = std::make_shared>(index_name, index_sample_block, index); - index = nullptr; - return granule; -} - -template -void MergeTreeIndexAggregatorAnnoy::update(const Block & block, size_t * pos, size_t limit) -{ - if (*pos >= block.rows()) - throw Exception( - ErrorCodes::LOGICAL_ERROR, - "The provided position is not less than the number of block rows. Position: {}, Block rows: {}.", - *pos, block.rows()); - - size_t rows_read = std::min(limit, block.rows() - *pos); - - if (rows_read == 0) - return; - - if (rows_read > std::numeric_limits::max()) - throw Exception(ErrorCodes::INCORRECT_DATA, "Index granularity is too big: more than 4B rows per index granule."); - - if (index_sample_block.columns() > 1) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected block with single column"); - - const String & index_column_name = index_sample_block.getByPosition(0).name; - ColumnPtr column_cut = block.getByName(index_column_name).column->cut(*pos, rows_read); - - if (const auto & column_array = typeid_cast(column_cut.get())) - { - const auto & column_array_data = column_array->getData(); - const auto & column_array_data_float = typeid_cast(column_array_data); - const auto & column_array_data_float_data = column_array_data_float.getData(); - - const auto & column_array_offsets = column_array->getOffsets(); - const size_t num_rows = column_array_offsets.size(); - - if (column_array->empty()) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Array is unexpectedly empty"); - - /// The Annoy algorithm naturally assumes that the indexed vectors have dimension >= 1. This condition is violated if empty arrays - /// are INSERTed into an Annoy-indexed column or if no value was specified at all in which case the arrays take on their default - /// value which is also empty. - if (column_array->isDefaultAt(0)) - throw Exception(ErrorCodes::INCORRECT_DATA, "The arrays in column '{}' must not be empty. Did you try to INSERT default values?", index_column_name); - - /// Check all sizes are the same - size_t dimension = column_array_offsets[0]; - for (size_t i = 0; i < num_rows - 1; ++i) - if (column_array_offsets[i + 1] - column_array_offsets[i] != dimension) - throw Exception(ErrorCodes::INCORRECT_DATA, "All arrays in column '{}' must have equal length", index_column_name); - - /// Also check that previously inserted blocks have the same size as this block. - /// Note that this guarantees consistency of dimension only within parts. We are unable to detect inconsistent dimensions across - /// parts - for this, a little help from the user is needed, e.g. CONSTRAINT cnstr CHECK length(array) = 42. - if (index && index->getDimensions() != dimension) - throw Exception(ErrorCodes::INCORRECT_DATA, "All arrays in column '{}' must have equal length", index_column_name); - - if (!index) - index = std::make_shared>(dimension); - - /// Add all rows of block - index->add_item(index->get_n_items(), column_array_data_float_data.data()); - for (size_t current_row = 1; current_row < num_rows; ++current_row) - index->add_item(index->get_n_items(), &column_array_data_float_data[column_array_offsets[current_row - 1]]); - } - else if (const auto & column_tuple = typeid_cast(column_cut.get())) - { - const auto & column_tuple_columns = column_tuple->getColumns(); - - /// TODO check if calling index->add_item() directly on the block's tuples is faster than materializing everything - std::vector> data(column_tuple->size(), std::vector()); - for (const auto & column : column_tuple_columns) - { - const auto & pod_array = typeid_cast(column.get())->getData(); - for (size_t i = 0; i < pod_array.size(); ++i) - data[i].push_back(pod_array[i]); - } - - if (data.empty()) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Tuple has 0 rows, {} rows expected", rows_read); - - if (!index) - index = std::make_shared>(data[0].size()); - - for (const auto & item : data) - index->add_item(index->get_n_items(), item.data()); - } - else - throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected Array or Tuple column"); - - *pos += rows_read; -} - - -MergeTreeIndexConditionAnnoy::MergeTreeIndexConditionAnnoy( - const IndexDescription & /*index_description*/, - const SelectQueryInfo & query, - const String & distance_function_, - ContextPtr context) - : ann_condition(query, context) - , distance_function(distance_function_) - , search_k(context->getSettingsRef().annoy_index_search_k_nodes) -{} - -bool MergeTreeIndexConditionAnnoy::mayBeTrueOnGranule(MergeTreeIndexGranulePtr /*idx_granule*/) const -{ - throw Exception(ErrorCodes::LOGICAL_ERROR, "mayBeTrueOnGranule is not supported for ANN skip indexes"); -} - -bool MergeTreeIndexConditionAnnoy::alwaysUnknownOrTrue() const -{ - return ann_condition.alwaysUnknownOrTrue(distance_function); -} - -std::vector MergeTreeIndexConditionAnnoy::getUsefulRanges(MergeTreeIndexGranulePtr idx_granule) const -{ - if (distance_function == DISTANCE_FUNCTION_L2) - return getUsefulRangesImpl(idx_granule); - else if (distance_function == DISTANCE_FUNCTION_COSINE) - return getUsefulRangesImpl(idx_granule); - std::unreachable(); -} - -template -std::vector MergeTreeIndexConditionAnnoy::getUsefulRangesImpl(MergeTreeIndexGranulePtr idx_granule) const -{ - const UInt64 limit = ann_condition.getLimit(); - const UInt64 index_granularity = ann_condition.getIndexGranularity(); - const std::optional comparison_distance = ann_condition.getQueryType() == ApproximateNearestNeighborInformation::Type::Where - ? std::optional(ann_condition.getComparisonDistanceForWhereQuery()) - : std::nullopt; - - if (comparison_distance && comparison_distance.value() < 0) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Attempt to optimize query with where without distance"); - - const std::vector reference_vector = ann_condition.getReferenceVector(); - - const auto granule = std::dynamic_pointer_cast>(idx_granule); - if (granule == nullptr) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Granule has the wrong type"); - - const AnnoyIndexWithSerializationPtr annoy = granule->index; - - if (ann_condition.getDimensions() != annoy->getDimensions()) - throw Exception(ErrorCodes::INCORRECT_QUERY, "The dimension of the space in the request ({}) " - "does not match the dimension in the index ({})", - ann_condition.getDimensions(), annoy->getDimensions()); - - std::vector neighbors; /// indexes of dots which were closest to the reference vector - std::vector distances; - neighbors.reserve(limit); - distances.reserve(limit); - - annoy->get_nns_by_vector(reference_vector.data(), limit, static_cast(search_k), &neighbors, &distances); - - chassert(neighbors.size() == distances.size()); - - std::vector granules; - granules.reserve(neighbors.size()); - for (size_t i = 0; i < neighbors.size(); ++i) - { - if (comparison_distance && distances[i] > comparison_distance) - continue; - granules.push_back(neighbors[i] / index_granularity); - } - - /// make unique - std::sort(granules.begin(), granules.end()); - granules.erase(std::unique(granules.begin(), granules.end()), granules.end()); - - return granules; -} - -MergeTreeIndexAnnoy::MergeTreeIndexAnnoy(const IndexDescription & index_, UInt64 trees_, const String & distance_function_) - : IMergeTreeIndex(index_) - , trees(trees_) - , distance_function(distance_function_) -{} - -MergeTreeIndexGranulePtr MergeTreeIndexAnnoy::createIndexGranule() const -{ - if (distance_function == DISTANCE_FUNCTION_L2) - return std::make_shared>(index.name, index.sample_block); - else if (distance_function == DISTANCE_FUNCTION_COSINE) - return std::make_shared>(index.name, index.sample_block); - std::unreachable(); -} - -MergeTreeIndexAggregatorPtr MergeTreeIndexAnnoy::createIndexAggregator(const MergeTreeWriterSettings & settings) const -{ - /// TODO: Support more metrics. Available metrics: https://github.com/spotify/annoy/blob/master/src/annoymodule.cc#L151-L171 - if (distance_function == DISTANCE_FUNCTION_L2) - return std::make_shared>(index.name, index.sample_block, trees, settings.max_threads_for_annoy_index_creation); - else if (distance_function == DISTANCE_FUNCTION_COSINE) - return std::make_shared>(index.name, index.sample_block, trees, settings.max_threads_for_annoy_index_creation); - std::unreachable(); -} - -MergeTreeIndexConditionPtr MergeTreeIndexAnnoy::createIndexCondition(const SelectQueryInfo & query, ContextPtr context) const -{ - return std::make_shared(index, query, distance_function, context); -}; - -MergeTreeIndexConditionPtr MergeTreeIndexAnnoy::createIndexCondition(const ActionsDAG *, ContextPtr) const -{ - throw Exception(ErrorCodes::NOT_IMPLEMENTED, "MergeTreeIndexAnnoy cannot be created with ActionsDAG"); -} - -MergeTreeIndexPtr annoyIndexCreator(const IndexDescription & index) -{ - static constexpr auto DEFAULT_DISTANCE_FUNCTION = DISTANCE_FUNCTION_L2; - String distance_function = DEFAULT_DISTANCE_FUNCTION; - if (!index.arguments.empty()) - distance_function = index.arguments[0].get(); - - static constexpr auto DEFAULT_TREES = 100uz; - UInt64 trees = DEFAULT_TREES; - if (index.arguments.size() > 1) - trees = index.arguments[1].get(); - - return std::make_shared(index, trees, distance_function); -} - -void annoyIndexValidator(const IndexDescription & index, bool /* attach */) -{ - /// Check number and type of Annoy index arguments: - - if (index.arguments.size() > 2) - throw Exception(ErrorCodes::INCORRECT_QUERY, "Annoy index must not have more than two parameters"); - - if (!index.arguments.empty() && index.arguments[0].getType() != Field::Types::String) - throw Exception(ErrorCodes::INCORRECT_QUERY, "Distance function argument of Annoy index must be of type String"); - - if (index.arguments.size() > 1 && index.arguments[1].getType() != Field::Types::UInt64) - throw Exception(ErrorCodes::INCORRECT_QUERY, "Number of trees argument of Annoy index must be of type UInt64"); - - /// Check that the index is created on a single column - - if (index.column_names.size() != 1 || index.data_types.size() != 1) - throw Exception(ErrorCodes::INCORRECT_NUMBER_OF_COLUMNS, "Annoy indexes must be created on a single column"); - - /// Check that a supported metric was passed as first argument - - if (!index.arguments.empty()) - { - String distance_name = index.arguments[0].get(); - if (distance_name != DISTANCE_FUNCTION_L2 && distance_name != DISTANCE_FUNCTION_COSINE) - throw Exception(ErrorCodes::INCORRECT_DATA, "Annoy index only supports distance functions '{}' and '{}'", DISTANCE_FUNCTION_L2, DISTANCE_FUNCTION_COSINE); - } - - /// Check data type of indexed column: - - auto throw_unsupported_underlying_column_exception = []() - { - throw Exception( - ErrorCodes::ILLEGAL_COLUMN, - "Annoy indexes can only be created on columns of type Array(Float32) and Tuple(Float32[, Float32[, ...]])"); - }; - - DataTypePtr data_type = index.sample_block.getDataTypes()[0]; - - if (const auto * data_type_array = typeid_cast(data_type.get())) - { - TypeIndex nested_type_index = data_type_array->getNestedType()->getTypeId(); - if (!WhichDataType(nested_type_index).isFloat32()) - throw_unsupported_underlying_column_exception(); - } - else if (const auto * data_type_tuple = typeid_cast(data_type.get())) - { - const DataTypes & inner_types = data_type_tuple->getElements(); - for (const auto & inner_type : inner_types) - { - TypeIndex nested_type_index = inner_type->getTypeId(); - if (!WhichDataType(nested_type_index).isFloat32()) - throw_unsupported_underlying_column_exception(); - } - } - else - throw_unsupported_underlying_column_exception(); -} - -} - -#endif diff --git a/src/Storages/MergeTree/MergeTreeIndexAnnoy.h b/src/Storages/MergeTree/MergeTreeIndexAnnoy.h deleted file mode 100644 index 282920c608e..00000000000 --- a/src/Storages/MergeTree/MergeTreeIndexAnnoy.h +++ /dev/null @@ -1,112 +0,0 @@ -#pragma once - -#ifdef ENABLE_ANNOY - -#include - -#include -#include - -namespace DB -{ - -template -class AnnoyIndexWithSerialization : public Annoy::AnnoyIndex -{ - using Base = Annoy::AnnoyIndex; - -public: - explicit AnnoyIndexWithSerialization(size_t dimensions); - void serialize(WriteBuffer & ostr) const; - void deserialize(ReadBuffer & istr); - size_t getDimensions() const; -}; - -template -using AnnoyIndexWithSerializationPtr = std::shared_ptr>; - - -template -struct MergeTreeIndexGranuleAnnoy final : public IMergeTreeIndexGranule -{ - MergeTreeIndexGranuleAnnoy(const String & index_name_, const Block & index_sample_block_); - MergeTreeIndexGranuleAnnoy(const String & index_name_, const Block & index_sample_block_, AnnoyIndexWithSerializationPtr index_); - - ~MergeTreeIndexGranuleAnnoy() override = default; - - void serializeBinary(WriteBuffer & ostr) const override; - void deserializeBinary(ReadBuffer & istr, MergeTreeIndexVersion version) override; - - bool empty() const override { return !index.get(); } - - const String index_name; - const Block index_sample_block; - AnnoyIndexWithSerializationPtr index; -}; - - -template -struct MergeTreeIndexAggregatorAnnoy final : IMergeTreeIndexAggregator -{ - MergeTreeIndexAggregatorAnnoy(const String & index_name_, const Block & index_sample_block, UInt64 trees, size_t max_threads_for_creation); - ~MergeTreeIndexAggregatorAnnoy() override = default; - - bool empty() const override { return !index || index->get_n_items() == 0; } - MergeTreeIndexGranulePtr getGranuleAndReset() override; - void update(const Block & block, size_t * pos, size_t limit) override; - - const String index_name; - const Block index_sample_block; - const UInt64 trees; - const size_t max_threads_for_creation; - AnnoyIndexWithSerializationPtr index; -}; - - -class MergeTreeIndexConditionAnnoy final : public IMergeTreeIndexConditionApproximateNearestNeighbor -{ -public: - MergeTreeIndexConditionAnnoy( - const IndexDescription & index_description, - const SelectQueryInfo & query, - const String & distance_function, - ContextPtr context); - - ~MergeTreeIndexConditionAnnoy() override = default; - - bool alwaysUnknownOrTrue() const override; - bool mayBeTrueOnGranule(MergeTreeIndexGranulePtr idx_granule) const override; - std::vector getUsefulRanges(MergeTreeIndexGranulePtr idx_granule) const override; - -private: - template - std::vector getUsefulRangesImpl(MergeTreeIndexGranulePtr idx_granule) const; - - const ApproximateNearestNeighborCondition ann_condition; - const String distance_function; - const Int64 search_k; -}; - - -class MergeTreeIndexAnnoy final : public IMergeTreeIndex -{ -public: - - MergeTreeIndexAnnoy(const IndexDescription & index_, UInt64 trees_, const String & distance_function_); - - ~MergeTreeIndexAnnoy() override = default; - - MergeTreeIndexGranulePtr createIndexGranule() const override; - MergeTreeIndexAggregatorPtr createIndexAggregator(const MergeTreeWriterSettings & settings) const override; - MergeTreeIndexConditionPtr createIndexCondition(const SelectQueryInfo & query, ContextPtr context) const; - MergeTreeIndexConditionPtr createIndexCondition(const ActionsDAG *, ContextPtr) const override; - bool isVectorSearch() const override { return true; } - -private: - const UInt64 trees; - const String distance_function; -}; - -} - -#endif diff --git a/src/Storages/MergeTree/MergeTreeIndexBloomFilter.cpp b/src/Storages/MergeTree/MergeTreeIndexBloomFilter.cpp index c6a00751f25..b796ed7114e 100644 --- a/src/Storages/MergeTree/MergeTreeIndexBloomFilter.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexBloomFilter.cpp @@ -348,19 +348,19 @@ bool MergeTreeIndexConditionBloomFilter::extractAtomFromTree(const RPNBuilderTre { if (const_value.getType() == Field::Types::UInt64) { - out.function = const_value.get() ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; + out.function = const_value.safeGet() ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; return true; } if (const_value.getType() == Field::Types::Int64) { - out.function = const_value.get() ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; + out.function = const_value.safeGet() ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; return true; } if (const_value.getType() == Field::Types::Float64) { - out.function = const_value.get() != 0.0 ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; + out.function = const_value.safeGet() != 0.0 ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; return true; } } @@ -371,67 +371,78 @@ bool MergeTreeIndexConditionBloomFilter::extractAtomFromTree(const RPNBuilderTre bool MergeTreeIndexConditionBloomFilter::traverseFunction(const RPNBuilderTreeNode & node, RPNElement & out, const RPNBuilderTreeNode * parent) { - bool maybe_useful = false; + if (!node.isFunction()) + return false; - if (node.isFunction()) + const auto function = node.toFunctionNode(); + auto arguments_size = function.getArgumentsSize(); + auto function_name = function.getFunctionName(); + + if (parent == nullptr) { - const auto function = node.toFunctionNode(); - auto arguments_size = function.getArgumentsSize(); - auto function_name = function.getFunctionName(); - + /// Recurse a little bit for indexOf(). for (size_t i = 0; i < arguments_size; ++i) { auto argument = function.getArgumentAt(i); if (traverseFunction(argument, out, &node)) - maybe_useful = true; - } - - if (arguments_size != 2) - return false; - - auto lhs_argument = function.getArgumentAt(0); - auto rhs_argument = function.getArgumentAt(1); - - if (functionIsInOrGlobalInOperator(function_name)) - { - if (auto future_set = rhs_argument.tryGetPreparedSet(); future_set) - { - if (auto prepared_set = future_set->buildOrderedSetInplace(rhs_argument.getTreeContext().getQueryContext()); prepared_set) - { - if (prepared_set->hasExplicitSetElements()) - { - const auto prepared_info = getPreparedSetInfo(prepared_set); - if (traverseTreeIn(function_name, lhs_argument, prepared_set, prepared_info.type, prepared_info.column, out)) - maybe_useful = true; - } - } - } - } - else if (function_name == "equals" || - function_name == "notEquals" || - function_name == "has" || - function_name == "mapContains" || - function_name == "indexOf" || - function_name == "hasAny" || - function_name == "hasAll") - { - Field const_value; - DataTypePtr const_type; - - if (rhs_argument.tryGetConstant(const_value, const_type)) - { - if (traverseTreeEquals(function_name, lhs_argument, const_type, const_value, out, parent)) - maybe_useful = true; - } - else if (lhs_argument.tryGetConstant(const_value, const_type)) - { - if (traverseTreeEquals(function_name, rhs_argument, const_type, const_value, out, parent)) - maybe_useful = true; - } + return true; } } - return maybe_useful; + if (arguments_size != 2) + return false; + + /// indexOf() should be inside comparison function, e.g. greater(indexOf(key, 42), 0). + /// Other conditions should be at top level, e.g. equals(key, 42), not equals(equals(key, 42), 1). + if ((function_name == "indexOf") != (parent != nullptr)) + return false; + + auto lhs_argument = function.getArgumentAt(0); + auto rhs_argument = function.getArgumentAt(1); + + if (functionIsInOrGlobalInOperator(function_name)) + { + if (auto future_set = rhs_argument.tryGetPreparedSet(); future_set) + { + if (auto prepared_set = future_set->buildOrderedSetInplace(rhs_argument.getTreeContext().getQueryContext()); prepared_set) + { + if (prepared_set->hasExplicitSetElements()) + { + const auto prepared_info = getPreparedSetInfo(prepared_set); + if (traverseTreeIn(function_name, lhs_argument, prepared_set, prepared_info.type, prepared_info.column, out)) + return true; + } + } + } + return false; + } + + if (function_name == "equals" || + function_name == "notEquals" || + function_name == "has" || + function_name == "mapContains" || + function_name == "indexOf" || + function_name == "hasAny" || + function_name == "hasAll") + { + Field const_value; + DataTypePtr const_type; + + if (rhs_argument.tryGetConstant(const_value, const_type)) + { + if (traverseTreeEquals(function_name, lhs_argument, const_type, const_value, out, parent)) + return true; + } + else if (lhs_argument.tryGetConstant(const_value, const_type) && (function_name == "equals" || function_name == "notEquals")) + { + if (traverseTreeEquals(function_name, rhs_argument, const_type, const_value, out, parent)) + return true; + } + + return false; + } + + return false; } bool MergeTreeIndexConditionBloomFilter::traverseTreeIn( @@ -692,7 +703,7 @@ bool MergeTreeIndexConditionBloomFilter::traverseTreeEquals( const bool is_nullable = actual_type->isNullable(); auto mutable_column = actual_type->createColumn(); - for (const auto & f : value_field.get()) + for (const auto & f : value_field.safeGet()) { if ((f.isNull() && !is_nullable) || f.isDecimal(f.getType())) /// NOLINT(readability-static-accessed-through-instance) return false; @@ -763,7 +774,7 @@ bool MergeTreeIndexConditionBloomFilter::traverseTreeEquals( if (which.isTuple() && key_node_function_name == "tuple") { - const Tuple & tuple = value_field.get(); + const Tuple & tuple = value_field.safeGet(); const auto * value_tuple_data_type = typeid_cast(value_type.get()); if (tuple.size() != key_node_function_arguments_size) @@ -952,7 +963,7 @@ void bloomFilterIndexValidator(const IndexDescription & index, bool attach) { const auto & argument = index.arguments[0]; - if (!attach && (argument.getType() != Field::Types::Float64 || argument.get() < 0 || argument.get() > 1)) + if (!attach && (argument.getType() != Field::Types::Float64 || argument.safeGet() < 0 || argument.safeGet() > 1)) throw Exception(ErrorCodes::BAD_ARGUMENTS, "The BloomFilter false positive must be a double number between 0 and 1."); } } diff --git a/src/Storages/MergeTree/MergeTreeIndexBloomFilterText.cpp b/src/Storages/MergeTree/MergeTreeIndexBloomFilterText.cpp index 5b6813d12e3..857b7903588 100644 --- a/src/Storages/MergeTree/MergeTreeIndexBloomFilterText.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexBloomFilterText.cpp @@ -341,19 +341,19 @@ bool MergeTreeConditionBloomFilterText::extractAtomFromTree(const RPNBuilderTree if (const_value.getType() == Field::Types::UInt64) { - out.function = const_value.get() ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; + out.function = const_value.safeGet() ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; return true; } if (const_value.getType() == Field::Types::Int64) { - out.function = const_value.get() ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; + out.function = const_value.safeGet() ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; return true; } if (const_value.getType() == Field::Types::Float64) { - out.function = const_value.get() != 0.0 ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; + out.function = const_value.safeGet() != 0.0 ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; return true; } } @@ -493,7 +493,7 @@ bool MergeTreeConditionBloomFilterText::traverseTreeEquals( out.function = RPNElement::FUNCTION_EQUALS; out.bloom_filter = std::make_unique(params); - auto value = const_value.get(); + auto value = const_value.safeGet(); if (is_case_insensitive_scenario) std::ranges::transform(value, value.begin(), [](const auto & c) { return static_cast(std::tolower(c)); }); @@ -509,7 +509,7 @@ bool MergeTreeConditionBloomFilterText::traverseTreeEquals( out.key_column = *key_index; out.function = RPNElement::FUNCTION_HAS; out.bloom_filter = std::make_unique(params); - auto & value = const_value.get(); + auto & value = const_value.safeGet(); token_extractor->stringToBloomFilter(value.data(), value.size(), *out.bloom_filter); return true; } @@ -519,7 +519,7 @@ bool MergeTreeConditionBloomFilterText::traverseTreeEquals( out.key_column = *key_index; out.function = RPNElement::FUNCTION_HAS; out.bloom_filter = std::make_unique(params); - auto & value = const_value.get(); + auto & value = const_value.safeGet(); token_extractor->stringToBloomFilter(value.data(), value.size(), *out.bloom_filter); return true; } @@ -529,7 +529,7 @@ bool MergeTreeConditionBloomFilterText::traverseTreeEquals( out.key_column = *key_index; out.function = RPNElement::FUNCTION_NOT_EQUALS; out.bloom_filter = std::make_unique(params); - const auto & value = const_value.get(); + const auto & value = const_value.safeGet(); token_extractor->stringToBloomFilter(value.data(), value.size(), *out.bloom_filter); return true; } @@ -538,7 +538,7 @@ bool MergeTreeConditionBloomFilterText::traverseTreeEquals( out.key_column = *key_index; out.function = RPNElement::FUNCTION_EQUALS; out.bloom_filter = std::make_unique(params); - const auto & value = const_value.get(); + const auto & value = const_value.safeGet(); token_extractor->stringToBloomFilter(value.data(), value.size(), *out.bloom_filter); return true; } @@ -547,7 +547,7 @@ bool MergeTreeConditionBloomFilterText::traverseTreeEquals( out.key_column = *key_index; out.function = RPNElement::FUNCTION_EQUALS; out.bloom_filter = std::make_unique(params); - const auto & value = const_value.get(); + const auto & value = const_value.safeGet(); token_extractor->stringLikeToBloomFilter(value.data(), value.size(), *out.bloom_filter); return true; } @@ -556,7 +556,7 @@ bool MergeTreeConditionBloomFilterText::traverseTreeEquals( out.key_column = *key_index; out.function = RPNElement::FUNCTION_NOT_EQUALS; out.bloom_filter = std::make_unique(params); - const auto & value = const_value.get(); + const auto & value = const_value.safeGet(); token_extractor->stringLikeToBloomFilter(value.data(), value.size(), *out.bloom_filter); return true; } @@ -565,7 +565,7 @@ bool MergeTreeConditionBloomFilterText::traverseTreeEquals( out.key_column = *key_index; out.function = RPNElement::FUNCTION_EQUALS; out.bloom_filter = std::make_unique(params); - const auto & value = const_value.get(); + const auto & value = const_value.safeGet(); token_extractor->substringToBloomFilter(value.data(), value.size(), *out.bloom_filter, true, false); return true; } @@ -574,7 +574,7 @@ bool MergeTreeConditionBloomFilterText::traverseTreeEquals( out.key_column = *key_index; out.function = RPNElement::FUNCTION_EQUALS; out.bloom_filter = std::make_unique(params); - const auto & value = const_value.get(); + const auto & value = const_value.safeGet(); token_extractor->substringToBloomFilter(value.data(), value.size(), *out.bloom_filter, false, true); return true; } @@ -589,13 +589,13 @@ bool MergeTreeConditionBloomFilterText::traverseTreeEquals( /// 2d vector is not needed here but is used because already exists for FUNCTION_IN std::vector> bloom_filters; bloom_filters.emplace_back(); - for (const auto & element : const_value.get()) + for (const auto & element : const_value.safeGet()) { if (element.getType() != Field::Types::String) return false; bloom_filters.back().emplace_back(params); - const auto & value = element.get(); + const auto & value = element.safeGet(); if (function_name == "multiSearchAny") { @@ -615,7 +615,7 @@ bool MergeTreeConditionBloomFilterText::traverseTreeEquals( out.function = RPNElement::FUNCTION_MATCH; out.bloom_filter = std::make_unique(params); - auto & value = const_value.get(); + auto & value = const_value.safeGet(); String required_substring; bool dummy_is_trivial, dummy_required_substring_is_prefix; std::vector alternatives; @@ -743,11 +743,11 @@ MergeTreeIndexPtr bloomFilterIndexTextCreator( { if (index.type == NgramTokenExtractor::getName()) { - size_t n = index.arguments[0].get(); + size_t n = index.arguments[0].safeGet(); BloomFilterParameters params( - index.arguments[1].get(), - index.arguments[2].get(), - index.arguments[3].get()); + index.arguments[1].safeGet(), + index.arguments[2].safeGet(), + index.arguments[3].safeGet()); auto tokenizer = std::make_unique(n); @@ -756,9 +756,9 @@ MergeTreeIndexPtr bloomFilterIndexTextCreator( else if (index.type == SplitTokenExtractor::getName()) { BloomFilterParameters params( - index.arguments[0].get(), - index.arguments[1].get(), - index.arguments[2].get()); + index.arguments[0].safeGet(), + index.arguments[1].safeGet(), + index.arguments[2].safeGet()); auto tokenizer = std::make_unique(); @@ -815,9 +815,9 @@ void bloomFilterIndexTextValidator(const IndexDescription & index, bool /*attach /// Just validate BloomFilterParameters params( - index.arguments[0].get(), - index.arguments[1].get(), - index.arguments[2].get()); + index.arguments[0].safeGet(), + index.arguments[1].safeGet(), + index.arguments[2].safeGet()); } } diff --git a/src/Storages/MergeTree/MergeTreeIndexFullText.cpp b/src/Storages/MergeTree/MergeTreeIndexFullText.cpp index cd6af68ebcc..b5c6bb95d37 100644 --- a/src/Storages/MergeTree/MergeTreeIndexFullText.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexFullText.cpp @@ -74,7 +74,7 @@ void MergeTreeIndexGranuleFullText::deserializeBinary(ReadBuffer & istr, MergeTr for (auto & gin_filter : gin_filters) { size_serialization->deserializeBinary(field_rows, istr, {}); - size_t filter_size = field_rows.get(); + size_t filter_size = field_rows.safeGet(); gin_filter.getFilter().resize(filter_size); if (filter_size == 0) @@ -379,19 +379,19 @@ bool MergeTreeConditionFullText::traverseAtomAST(const RPNBuilderTreeNode & node /// Check constant like in KeyCondition if (const_value.getType() == Field::Types::UInt64) { - out.function = const_value.get() ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; + out.function = const_value.safeGet() ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; return true; } if (const_value.getType() == Field::Types::Int64) { - out.function = const_value.get() ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; + out.function = const_value.safeGet() ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; return true; } if (const_value.getType() == Field::Types::Float64) { - out.function = const_value.get() != 0.00 ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; + out.function = const_value.safeGet() != 0.00 ? RPNElement::ALWAYS_TRUE : RPNElement::ALWAYS_FALSE; return true; } } @@ -530,7 +530,7 @@ bool MergeTreeConditionFullText::traverseASTEquals( out.key_column = key_column_num; out.function = RPNElement::FUNCTION_HAS; out.gin_filter = std::make_unique(params); - auto & value = const_value.get(); + auto & value = const_value.safeGet(); token_extractor->stringToGinFilter(value.data(), value.size(), *out.gin_filter); return true; } @@ -539,7 +539,7 @@ bool MergeTreeConditionFullText::traverseASTEquals( out.key_column = key_column_num; out.function = RPNElement::FUNCTION_HAS; out.gin_filter = std::make_unique(params); - auto & value = const_value.get(); + auto & value = const_value.safeGet(); token_extractor->stringToGinFilter(value.data(), value.size(), *out.gin_filter); return true; } @@ -549,7 +549,7 @@ bool MergeTreeConditionFullText::traverseASTEquals( out.key_column = key_column_num; out.function = RPNElement::FUNCTION_NOT_EQUALS; out.gin_filter = std::make_unique(params); - const auto & value = const_value.get(); + const auto & value = const_value.safeGet(); token_extractor->stringToGinFilter(value.data(), value.size(), *out.gin_filter); return true; } @@ -558,7 +558,7 @@ bool MergeTreeConditionFullText::traverseASTEquals( out.key_column = key_column_num; out.function = RPNElement::FUNCTION_EQUALS; out.gin_filter = std::make_unique(params); - const auto & value = const_value.get(); + const auto & value = const_value.safeGet(); token_extractor->stringToGinFilter(value.data(), value.size(), *out.gin_filter); return true; } @@ -567,7 +567,7 @@ bool MergeTreeConditionFullText::traverseASTEquals( out.key_column = key_column_num; out.function = RPNElement::FUNCTION_EQUALS; out.gin_filter = std::make_unique(params); - const auto & value = const_value.get(); + const auto & value = const_value.safeGet(); token_extractor->stringLikeToGinFilter(value.data(), value.size(), *out.gin_filter); return true; } @@ -576,7 +576,7 @@ bool MergeTreeConditionFullText::traverseASTEquals( out.key_column = key_column_num; out.function = RPNElement::FUNCTION_NOT_EQUALS; out.gin_filter = std::make_unique(params); - const auto & value = const_value.get(); + const auto & value = const_value.safeGet(); token_extractor->stringLikeToGinFilter(value.data(), value.size(), *out.gin_filter); return true; } @@ -585,7 +585,7 @@ bool MergeTreeConditionFullText::traverseASTEquals( out.key_column = key_column_num; out.function = RPNElement::FUNCTION_EQUALS; out.gin_filter = std::make_unique(params); - const auto & value = const_value.get(); + const auto & value = const_value.safeGet(); token_extractor->stringToGinFilter(value.data(), value.size(), *out.gin_filter); return true; } @@ -594,7 +594,7 @@ bool MergeTreeConditionFullText::traverseASTEquals( out.key_column = key_column_num; out.function = RPNElement::FUNCTION_EQUALS; out.gin_filter = std::make_unique(params); - const auto & value = const_value.get(); + const auto & value = const_value.safeGet(); token_extractor->substringToGinFilter(value.data(), value.size(), *out.gin_filter, true, false); return true; } @@ -603,7 +603,7 @@ bool MergeTreeConditionFullText::traverseASTEquals( out.key_column = key_column_num; out.function = RPNElement::FUNCTION_EQUALS; out.gin_filter = std::make_unique(params); - const auto & value = const_value.get(); + const auto & value = const_value.safeGet(); token_extractor->substringToGinFilter(value.data(), value.size(), *out.gin_filter, false, true); return true; } @@ -615,13 +615,13 @@ bool MergeTreeConditionFullText::traverseASTEquals( /// 2d vector is not needed here but is used because already exists for FUNCTION_IN std::vector gin_filters; gin_filters.emplace_back(); - for (const auto & element : const_value.get()) + for (const auto & element : const_value.safeGet()) { if (element.getType() != Field::Types::String) return false; gin_filters.back().emplace_back(params); - const auto & value = element.get(); + const auto & value = element.safeGet(); token_extractor->substringToGinFilter(value.data(), value.size(), gin_filters.back().back(), false, false); } out.set_gin_filters = std::move(gin_filters); @@ -632,7 +632,7 @@ bool MergeTreeConditionFullText::traverseASTEquals( out.key_column = key_column_num; out.function = RPNElement::FUNCTION_MATCH; - auto & value = const_value.get(); + auto & value = const_value.safeGet(); String required_substring; bool dummy_is_trivial, dummy_required_substring_is_prefix; std::vector alternatives; @@ -776,8 +776,8 @@ MergeTreeIndexConditionPtr MergeTreeIndexFullText::createIndexCondition( MergeTreeIndexPtr fullTextIndexCreator( const IndexDescription & index) { - size_t n = index.arguments.empty() ? 0 : index.arguments[0].get(); - UInt64 max_rows = index.arguments.size() < 2 ? DEFAULT_MAX_ROWS_PER_POSTINGS_LIST : index.arguments[1].get(); + size_t n = index.arguments.empty() ? 0 : index.arguments[0].safeGet(); + UInt64 max_rows = index.arguments.size() < 2 ? DEFAULT_MAX_ROWS_PER_POSTINGS_LIST : index.arguments[1].safeGet(); GinFilterParameters params(n, max_rows); /// Use SplitTokenExtractor when n is 0, otherwise use NgramTokenExtractor @@ -826,12 +826,12 @@ void fullTextIndexValidator(const IndexDescription & index, bool /*attach*/) { if (index.arguments[1].getType() != Field::Types::UInt64) throw Exception(ErrorCodes::INCORRECT_QUERY, "The second full text index argument must be UInt64"); - if (index.arguments[1].get() != UNLIMITED_ROWS_PER_POSTINGS_LIST && index.arguments[1].get() < MIN_ROWS_PER_POSTINGS_LIST) + if (index.arguments[1].safeGet() != UNLIMITED_ROWS_PER_POSTINGS_LIST && index.arguments[1].safeGet() < MIN_ROWS_PER_POSTINGS_LIST) throw Exception(ErrorCodes::INCORRECT_QUERY, "The maximum rows per postings list must be no less than {}", MIN_ROWS_PER_POSTINGS_LIST); } /// Just validate - size_t ngrams = index.arguments.empty() ? 0 : index.arguments[0].get(); - UInt64 max_rows_per_postings_list = index.arguments.size() < 2 ? DEFAULT_MAX_ROWS_PER_POSTINGS_LIST : index.arguments[1].get(); + size_t ngrams = index.arguments.empty() ? 0 : index.arguments[0].safeGet(); + UInt64 max_rows_per_postings_list = index.arguments.size() < 2 ? DEFAULT_MAX_ROWS_PER_POSTINGS_LIST : index.arguments[1].safeGet(); GinFilterParameters params(ngrams, max_rows_per_postings_list); } } diff --git a/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp b/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp index 2a45ab1d927..2b924284857 100644 --- a/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp @@ -103,8 +103,19 @@ size_t MergeTreeIndexGranularity::countMarksForRows(size_t from_mark, size_t num /// This is a heuristic to respect min_marks_to_read which is ignored by MergeTreeReadPool in case of remote disk. /// See comment in IMergeTreeSelectAlgorithm. - if (min_marks_to_read && from_mark + 2 * min_marks_to_read <= to_mark) - to_mark = from_mark + min_marks_to_read; + if (min_marks_to_read) + { + // check overflow + size_t min_marks_to_read_2 = 0; + bool overflow = common::mulOverflow(min_marks_to_read, 2, min_marks_to_read_2); + + size_t to_mark_overwrite = 0; + if (!overflow) + overflow = common::addOverflow(from_mark, min_marks_to_read_2, to_mark_overwrite); + + if (!overflow && to_mark_overwrite < to_mark) + to_mark = to_mark_overwrite; + } return getRowsCountInRange(from_mark, std::max(1UL, to_mark)) - offset_in_rows; } diff --git a/src/Storages/MergeTree/MergeTreeIndexHypothesis.cpp b/src/Storages/MergeTree/MergeTreeIndexHypothesis.cpp index cd8065ecadf..abf3ae56376 100644 --- a/src/Storages/MergeTree/MergeTreeIndexHypothesis.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexHypothesis.cpp @@ -37,7 +37,7 @@ void MergeTreeIndexGranuleHypothesis::deserializeBinary(ReadBuffer & istr, Merge Field field_met; const auto & size_type = DataTypePtr(std::make_shared()); size_type->getDefaultSerialization()->deserializeBinary(field_met, istr, {}); - met = field_met.get(); + met = field_met.safeGet(); is_empty = false; } diff --git a/src/Storages/MergeTree/MergeTreeIndexLegacyVectorSimilarity.cpp b/src/Storages/MergeTree/MergeTreeIndexLegacyVectorSimilarity.cpp new file mode 100644 index 00000000000..29de109d4fc --- /dev/null +++ b/src/Storages/MergeTree/MergeTreeIndexLegacyVectorSimilarity.cpp @@ -0,0 +1,45 @@ +#include + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int ILLEGAL_INDEX; +} + +MergeTreeIndexLegacyVectorSimilarity::MergeTreeIndexLegacyVectorSimilarity(const IndexDescription & index_) + : IMergeTreeIndex(index_) +{ +} + +MergeTreeIndexGranulePtr MergeTreeIndexLegacyVectorSimilarity::createIndexGranule() const +{ + throw Exception(ErrorCodes::ILLEGAL_INDEX, "Indexes of type 'annoy' or 'usearch' are no longer supported. Please drop and recreate the index as type 'vector_similarity'"); +} + +MergeTreeIndexAggregatorPtr MergeTreeIndexLegacyVectorSimilarity::createIndexAggregator(const MergeTreeWriterSettings &) const +{ + throw Exception(ErrorCodes::ILLEGAL_INDEX, "Indexes of type 'annoy' or 'usearch' are no longer supported. Please drop and recreate the index as type 'vector_similarity'"); +} + +MergeTreeIndexConditionPtr MergeTreeIndexLegacyVectorSimilarity::createIndexCondition(const SelectQueryInfo &, ContextPtr) const +{ + throw Exception(ErrorCodes::ILLEGAL_INDEX, "Indexes of type 'annoy' or 'usearch' are no longer supported. Please drop and recreate the index as type 'vector_similarity'"); +}; + +MergeTreeIndexConditionPtr MergeTreeIndexLegacyVectorSimilarity::createIndexCondition(const ActionsDAG *, ContextPtr) const +{ + throw Exception(ErrorCodes::ILLEGAL_INDEX, "Indexes of type 'annoy' or 'usearch' are no longer supported. Please drop and recreate the index as type 'vector_similarity'"); +} + +MergeTreeIndexPtr legacyVectorSimilarityIndexCreator(const IndexDescription & index) +{ + return std::make_shared(index); +} + +void legacyVectorSimilarityIndexValidator(const IndexDescription &, bool) +{ +} + +} diff --git a/src/Storages/MergeTree/MergeTreeIndexLegacyVectorSimilarity.h b/src/Storages/MergeTree/MergeTreeIndexLegacyVectorSimilarity.h new file mode 100644 index 00000000000..1015401823d --- /dev/null +++ b/src/Storages/MergeTree/MergeTreeIndexLegacyVectorSimilarity.h @@ -0,0 +1,26 @@ +#pragma once + +#include + +/// Walking corpse implementation for removed skipping index of type "annoy" and "usearch". +/// Its only purpose is to allow loading old tables with indexes of these types. +/// Data insertion and index usage/search will throw an exception, suggesting to migrate to "vector_similarity" indexes. + +namespace DB +{ + +class MergeTreeIndexLegacyVectorSimilarity : public IMergeTreeIndex +{ +public: + explicit MergeTreeIndexLegacyVectorSimilarity(const IndexDescription & index_); + ~MergeTreeIndexLegacyVectorSimilarity() override = default; + + MergeTreeIndexGranulePtr createIndexGranule() const override; + MergeTreeIndexAggregatorPtr createIndexAggregator(const MergeTreeWriterSettings &) const override; + MergeTreeIndexConditionPtr createIndexCondition(const SelectQueryInfo &, ContextPtr) const; + MergeTreeIndexConditionPtr createIndexCondition(const ActionsDAG *, ContextPtr) const override; + + bool isVectorSimilarityIndex() const override { return true; } +}; + +} diff --git a/src/Storages/MergeTree/MergeTreeIndexSet.cpp b/src/Storages/MergeTree/MergeTreeIndexSet.cpp index a92df4ac72d..fa242fccbc1 100644 --- a/src/Storages/MergeTree/MergeTreeIndexSet.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexSet.cpp @@ -97,7 +97,7 @@ void MergeTreeIndexGranuleSet::deserializeBinary(ReadBuffer & istr, MergeTreeInd Field field_rows; const auto & size_type = DataTypePtr(std::make_shared()); size_type->getDefaultSerialization()->deserializeBinary(field_rows, istr, {}); - size_t rows_to_read = field_rows.get(); + size_t rows_to_read = field_rows.safeGet(); if (rows_to_read == 0) return; @@ -591,7 +591,7 @@ MergeTreeIndexConditionPtr MergeTreeIndexSet::createIndexCondition( MergeTreeIndexPtr setIndexCreator(const IndexDescription & index) { - size_t max_rows = index.arguments[0].get(); + size_t max_rows = index.arguments[0].safeGet(); return std::make_shared(index, max_rows); } diff --git a/src/Storages/MergeTree/MergeTreeIndexUSearch.cpp b/src/Storages/MergeTree/MergeTreeIndexUSearch.cpp deleted file mode 100644 index 59a4b0fbf9c..00000000000 --- a/src/Storages/MergeTree/MergeTreeIndexUSearch.cpp +++ /dev/null @@ -1,463 +0,0 @@ -#ifdef ENABLE_USEARCH - -#pragma clang diagnostic push -#pragma clang diagnostic ignored "-Wpass-failed" - -#include - -#include -#include -#include -#include -#include -#include -#include -#include -#include - -namespace ProfileEvents -{ - extern const Event USearchAddCount; - extern const Event USearchAddVisitedMembers; - extern const Event USearchAddComputedDistances; - extern const Event USearchSearchCount; - extern const Event USearchSearchVisitedMembers; - extern const Event USearchSearchComputedDistances; -} - -namespace DB -{ - -namespace ErrorCodes -{ - extern const int CANNOT_ALLOCATE_MEMORY; - extern const int ILLEGAL_COLUMN; - extern const int INCORRECT_DATA; - extern const int INCORRECT_NUMBER_OF_COLUMNS; - extern const int INCORRECT_QUERY; - extern const int LOGICAL_ERROR; - extern const int NOT_IMPLEMENTED; -} - -namespace -{ - -std::unordered_map nameToScalarKind = { - {"f64", unum::usearch::scalar_kind_t::f64_k}, - {"f32", unum::usearch::scalar_kind_t::f32_k}, - {"f16", unum::usearch::scalar_kind_t::f16_k}, - {"i8", unum::usearch::scalar_kind_t::i8_k}}; - -} - -template -USearchIndexWithSerialization::USearchIndexWithSerialization(size_t dimensions, unum::usearch::scalar_kind_t scalar_kind) - : Base(Base::make(unum::usearch::metric_punned_t(dimensions, Metric, scalar_kind))) -{ -} - -template -void USearchIndexWithSerialization::serialize(WriteBuffer & ostr) const -{ - auto callback = [&ostr](void * from, size_t n) - { - ostr.write(reinterpret_cast(from), n); - return true; - }; - - Base::save_to_stream(callback); -} - -template -void USearchIndexWithSerialization::deserialize(ReadBuffer & istr) -{ - auto callback = [&istr](void * from, size_t n) - { - istr.readStrict(reinterpret_cast(from), n); - return true; - }; - - Base::load_from_stream(callback); -} - -template -size_t USearchIndexWithSerialization::getDimensions() const -{ - return Base::dimensions(); -} - -template -MergeTreeIndexGranuleUSearch::MergeTreeIndexGranuleUSearch( - const String & index_name_, - const Block & index_sample_block_, - unum::usearch::scalar_kind_t scalar_kind_) - : index_name(index_name_) - , index_sample_block(index_sample_block_) - , scalar_kind(scalar_kind_) - , index(nullptr) -{ -} - -template -MergeTreeIndexGranuleUSearch::MergeTreeIndexGranuleUSearch( - const String & index_name_, - const Block & index_sample_block_, - unum::usearch::scalar_kind_t scalar_kind_, - USearchIndexWithSerializationPtr index_) - : index_name(index_name_) - , index_sample_block(index_sample_block_) - , scalar_kind(scalar_kind_) - , index(std::move(index_)) -{ -} - -template -void MergeTreeIndexGranuleUSearch::serializeBinary(WriteBuffer & ostr) const -{ - /// Number of dimensions is required in the index constructor, - /// so it must be written and read separately from the other part - writeIntBinary(static_cast(index->getDimensions()), ostr); // write dimension - index->serialize(ostr); -} - -template -void MergeTreeIndexGranuleUSearch::deserializeBinary(ReadBuffer & istr, MergeTreeIndexVersion /*version*/) -{ - UInt64 dimension; - readIntBinary(dimension, istr); - index = std::make_shared>(dimension, scalar_kind); - index->deserialize(istr); -} - -template -MergeTreeIndexAggregatorUSearch::MergeTreeIndexAggregatorUSearch( - const String & index_name_, - const Block & index_sample_block_, - unum::usearch::scalar_kind_t scalar_kind_) - : index_name(index_name_) - , index_sample_block(index_sample_block_) - , scalar_kind(scalar_kind_) -{ -} - -template -MergeTreeIndexGranulePtr MergeTreeIndexAggregatorUSearch::getGranuleAndReset() -{ - auto granule = std::make_shared>(index_name, index_sample_block, scalar_kind, index); - index = nullptr; - return granule; -} - -template -void MergeTreeIndexAggregatorUSearch::update(const Block & block, size_t * pos, size_t limit) -{ - if (*pos >= block.rows()) - throw Exception( - ErrorCodes::LOGICAL_ERROR, - "The provided position is not less than the number of block rows. Position: {}, Block rows: {}.", - *pos, - block.rows()); - - size_t rows_read = std::min(limit, block.rows() - *pos); - - if (rows_read == 0) - return; - - if (rows_read > std::numeric_limits::max()) - throw Exception(ErrorCodes::INCORRECT_DATA, "Index granularity is too big: more than 4B rows per index granule."); - - if (index_sample_block.columns() > 1) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected block with single column"); - - const String & index_column_name = index_sample_block.getByPosition(0).name; - ColumnPtr column_cut = block.getByName(index_column_name).column->cut(*pos, rows_read); - - if (const auto & column_array = typeid_cast(column_cut.get())) - { - const auto & column_array_data = column_array->getData(); - const auto & column_array_data_float = typeid_cast(column_array_data); - const auto & column_array_data_float_data = column_array_data_float.getData(); - - const auto & column_array_offsets = column_array->getOffsets(); - const size_t num_rows = column_array_offsets.size(); - - if (column_array->empty()) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Array is unexpectedly empty"); - - /// The Usearch algorithm naturally assumes that the indexed vectors have dimension >= 1. This condition is violated if empty arrays - /// are INSERTed into an Usearch-indexed column or if no value was specified at all in which case the arrays take on their default - /// values which is also empty. - if (column_array->isDefaultAt(0)) - throw Exception(ErrorCodes::INCORRECT_DATA, "The arrays in column '{}' must not be empty. Did you try to INSERT default values?", index_column_name); - - /// Check all sizes are the same - size_t dimension = column_array_offsets[0]; - for (size_t i = 0; i < num_rows - 1; ++i) - if (column_array_offsets[i + 1] - column_array_offsets[i] != dimension) - throw Exception(ErrorCodes::INCORRECT_DATA, "All arrays in column '{}' must have equal length", index_column_name); - - /// Also check that previously inserted blocks have the same size as this block. - /// Note that this guarantees consistency of dimension only within parts. We are unable to detect inconsistent dimensions across - /// parts - for this, a little help from the user is needed, e.g. CONSTRAINT cnstr CHECK length(array) = 42. - if (index && index->getDimensions() != dimension) - throw Exception(ErrorCodes::INCORRECT_DATA, "All arrays in column '{}' must have equal length", index_column_name); - - if (!index) - index = std::make_shared>(dimension, scalar_kind); - - /// Add all rows of block - if (!index->reserve(unum::usearch::ceil2(index->size() + num_rows))) - throw Exception(ErrorCodes::CANNOT_ALLOCATE_MEMORY, "Could not reserve memory for usearch index"); - - for (size_t current_row = 0; current_row < num_rows; ++current_row) - { - auto rc = index->add(static_cast(index->size()), &column_array_data_float_data[column_array_offsets[current_row - 1]]); - if (!rc) - throw Exception::createRuntime(ErrorCodes::INCORRECT_DATA, rc.error.release()); - - ProfileEvents::increment(ProfileEvents::USearchAddCount); - ProfileEvents::increment(ProfileEvents::USearchAddVisitedMembers, rc.visited_members); - ProfileEvents::increment(ProfileEvents::USearchAddComputedDistances, rc.computed_distances); - } - } - else if (const auto & column_tuple = typeid_cast(column_cut.get())) - { - const auto & column_tuple_columns = column_tuple->getColumns(); - std::vector> data(column_tuple->size(), std::vector()); - for (const auto & column : column_tuple_columns) - { - const auto & pod_array = typeid_cast(column.get())->getData(); - for (size_t i = 0; i < pod_array.size(); ++i) - data[i].push_back(pod_array[i]); - } - - if (data.empty()) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Tuple has 0 rows, {} rows expected", rows_read); - - if (!index) - index = std::make_shared>(data[0].size(), scalar_kind); - - if (!index->reserve(unum::usearch::ceil2(index->size() + data.size()))) - throw Exception(ErrorCodes::CANNOT_ALLOCATE_MEMORY, "Could not reserve memory for usearch index"); - - for (const auto & item : data) - { - auto rc = index->add(static_cast(index->size()), item.data()); - if (!rc) - throw Exception::createRuntime(ErrorCodes::INCORRECT_DATA, rc.error.release()); - - ProfileEvents::increment(ProfileEvents::USearchAddCount); - ProfileEvents::increment(ProfileEvents::USearchAddVisitedMembers, rc.visited_members); - ProfileEvents::increment(ProfileEvents::USearchAddComputedDistances, rc.computed_distances); - } - } - else - throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected Array or Tuple column"); - - *pos += rows_read; -} - -MergeTreeIndexConditionUSearch::MergeTreeIndexConditionUSearch( - const IndexDescription & /*index_description*/, - const SelectQueryInfo & query, - const String & distance_function_, - ContextPtr context) - : ann_condition(query, context) - , distance_function(distance_function_) -{ -} - -bool MergeTreeIndexConditionUSearch::mayBeTrueOnGranule(MergeTreeIndexGranulePtr /*idx_granule*/) const -{ - throw Exception(ErrorCodes::LOGICAL_ERROR, "mayBeTrueOnGranule is not supported for ANN skip indexes"); -} - -bool MergeTreeIndexConditionUSearch::alwaysUnknownOrTrue() const -{ - return ann_condition.alwaysUnknownOrTrue(distance_function); -} - -std::vector MergeTreeIndexConditionUSearch::getUsefulRanges(MergeTreeIndexGranulePtr idx_granule) const -{ - if (distance_function == DISTANCE_FUNCTION_L2) - return getUsefulRangesImpl(idx_granule); - else if (distance_function == DISTANCE_FUNCTION_COSINE) - return getUsefulRangesImpl(idx_granule); - std::unreachable(); -} - -template -std::vector MergeTreeIndexConditionUSearch::getUsefulRangesImpl(MergeTreeIndexGranulePtr idx_granule) const -{ - const UInt64 limit = ann_condition.getLimit(); - const UInt64 index_granularity = ann_condition.getIndexGranularity(); - const std::optional comparison_distance = ann_condition.getQueryType() == ApproximateNearestNeighborInformation::Type::Where - ? std::optional(ann_condition.getComparisonDistanceForWhereQuery()) - : std::nullopt; - - if (comparison_distance && comparison_distance.value() < 0) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Attempt to optimize query with where without distance"); - - const std::vector reference_vector = ann_condition.getReferenceVector(); - - const auto granule = std::dynamic_pointer_cast>(idx_granule); - if (granule == nullptr) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Granule has the wrong type"); - - const USearchIndexWithSerializationPtr index = granule->index; - - if (ann_condition.getDimensions() != index->dimensions()) - throw Exception(ErrorCodes::INCORRECT_QUERY, "The dimension of the space in the request ({}) " - "does not match the dimension in the index ({})", - ann_condition.getDimensions(), index->dimensions()); - - auto result = index->search(reference_vector.data(), limit); - - ProfileEvents::increment(ProfileEvents::USearchSearchCount); - ProfileEvents::increment(ProfileEvents::USearchSearchVisitedMembers, result.visited_members); - ProfileEvents::increment(ProfileEvents::USearchSearchComputedDistances, result.computed_distances); - - std::vector neighbors(result.size()); /// indexes of dots which were closest to the reference vector - std::vector distances(result.size()); - result.dump_to(neighbors.data(), distances.data()); - - std::vector granules; - granules.reserve(neighbors.size()); - for (size_t i = 0; i < neighbors.size(); ++i) - { - if (comparison_distance && distances[i] > comparison_distance) - continue; - granules.push_back(neighbors[i] / index_granularity); - } - - /// make unique - std::sort(granules.begin(), granules.end()); - granules.erase(std::unique(granules.begin(), granules.end()), granules.end()); - - return granules; -} - -MergeTreeIndexUSearch::MergeTreeIndexUSearch(const IndexDescription & index_, const String & distance_function_, unum::usearch::scalar_kind_t scalar_kind_) - : IMergeTreeIndex(index_) - , distance_function(distance_function_) - , scalar_kind(scalar_kind_) -{ -} - -MergeTreeIndexGranulePtr MergeTreeIndexUSearch::createIndexGranule() const -{ - if (distance_function == DISTANCE_FUNCTION_L2) - return std::make_shared>(index.name, index.sample_block, scalar_kind); - else if (distance_function == DISTANCE_FUNCTION_COSINE) - return std::make_shared>(index.name, index.sample_block, scalar_kind); - std::unreachable(); -} - -MergeTreeIndexAggregatorPtr MergeTreeIndexUSearch::createIndexAggregator(const MergeTreeWriterSettings & /*settings*/) const -{ - if (distance_function == DISTANCE_FUNCTION_L2) - return std::make_shared>(index.name, index.sample_block, scalar_kind); - else if (distance_function == DISTANCE_FUNCTION_COSINE) - return std::make_shared>(index.name, index.sample_block, scalar_kind); - std::unreachable(); -} - -MergeTreeIndexConditionPtr MergeTreeIndexUSearch::createIndexCondition(const SelectQueryInfo & query, ContextPtr context) const -{ - return std::make_shared(index, query, distance_function, context); -}; - -MergeTreeIndexConditionPtr MergeTreeIndexUSearch::createIndexCondition(const ActionsDAG *, ContextPtr) const -{ - throw Exception(ErrorCodes::NOT_IMPLEMENTED, "MergeTreeIndexAnnoy cannot be created with ActionsDAG"); -} - -MergeTreeIndexPtr usearchIndexCreator(const IndexDescription & index) -{ - static constexpr auto default_distance_function = DISTANCE_FUNCTION_L2; - String distance_function = default_distance_function; - if (!index.arguments.empty()) - distance_function = index.arguments[0].get(); - - static constexpr auto default_scalar_kind = unum::usearch::scalar_kind_t::f16_k; - auto scalar_kind = default_scalar_kind; - if (index.arguments.size() > 1) - scalar_kind = nameToScalarKind.at(index.arguments[1].get()); - - return std::make_shared(index, distance_function, scalar_kind); -} - -void usearchIndexValidator(const IndexDescription & index, bool /* attach */) -{ - /// Check number and type of USearch index arguments: - - if (index.arguments.size() > 2) - throw Exception(ErrorCodes::INCORRECT_QUERY, "USearch index must not have more than one parameters"); - - if (!index.arguments.empty() && index.arguments[0].getType() != Field::Types::String) - throw Exception(ErrorCodes::INCORRECT_QUERY, "First argument of USearch index (distance function) must be of type String"); - if (index.arguments.size() > 1 && index.arguments[1].getType() != Field::Types::String) - throw Exception(ErrorCodes::INCORRECT_QUERY, "Second argument of USearch index (scalar type) must be of type String"); - - /// Check that the index is created on a single column - - if (index.column_names.size() != 1 || index.data_types.size() != 1) - throw Exception(ErrorCodes::INCORRECT_NUMBER_OF_COLUMNS, "USearch indexes must be created on a single column"); - - /// Check that a supported metric was passed as first argument - - if (!index.arguments.empty()) - { - String distance_name = index.arguments[0].get(); - if (distance_name != DISTANCE_FUNCTION_L2 && distance_name != DISTANCE_FUNCTION_COSINE) - throw Exception(ErrorCodes::INCORRECT_DATA, "USearch index only supports distance functions '{}' and '{}'", DISTANCE_FUNCTION_L2, DISTANCE_FUNCTION_COSINE); - } - - /// Check that a supported kind was passed as a second argument - - if (index.arguments.size() > 1 && !nameToScalarKind.contains(index.arguments[1].get())) - { - String supported_kinds; - for (const auto & [name, kind] : nameToScalarKind) - { - if (!supported_kinds.empty()) - supported_kinds += ", "; - supported_kinds += name; - } - throw Exception(ErrorCodes::INCORRECT_DATA, "Unrecognized scalar kind (second argument) for USearch index. Supported kinds are: {}", supported_kinds); - } - - /// Check data type of indexed column: - - auto throw_unsupported_underlying_column_exception = []() - { - throw Exception( - ErrorCodes::ILLEGAL_COLUMN, - "USearch can only be created on columns of type Array(Float32) and Tuple(Float32[, Float32[, ...]])"); - }; - - DataTypePtr data_type = index.sample_block.getDataTypes()[0]; - - if (const auto * data_type_array = typeid_cast(data_type.get())) - { - TypeIndex nested_type_index = data_type_array->getNestedType()->getTypeId(); - if (!WhichDataType(nested_type_index).isFloat32()) - throw_unsupported_underlying_column_exception(); - } - else if (const auto * data_type_tuple = typeid_cast(data_type.get())) - { - const DataTypes & inner_types = data_type_tuple->getElements(); - for (const auto & inner_type : inner_types) - { - TypeIndex nested_type_index = inner_type->getTypeId(); - if (!WhichDataType(nested_type_index).isFloat32()) - throw_unsupported_underlying_column_exception(); - } - } - else - throw_unsupported_underlying_column_exception(); -} - -} - -#endif diff --git a/src/Storages/MergeTree/MergeTreeIndexUSearch.h b/src/Storages/MergeTree/MergeTreeIndexUSearch.h deleted file mode 100644 index 41de94402c9..00000000000 --- a/src/Storages/MergeTree/MergeTreeIndexUSearch.h +++ /dev/null @@ -1,116 +0,0 @@ -#pragma once - -#ifdef ENABLE_USEARCH - -#include - -#pragma clang diagnostic push -#pragma clang diagnostic ignored "-Wpass-failed" -#include -#pragma clang diagnostic pop - -namespace DB -{ - -using USearchImplType = unum::usearch::index_dense_gt; - -template -class USearchIndexWithSerialization : public USearchImplType -{ - using Base = USearchImplType; - -public: - USearchIndexWithSerialization(size_t dimensions, unum::usearch::scalar_kind_t scalar_kind); - void serialize(WriteBuffer & ostr) const; - void deserialize(ReadBuffer & istr); - size_t getDimensions() const; -}; - -template -using USearchIndexWithSerializationPtr = std::shared_ptr>; - - -template -struct MergeTreeIndexGranuleUSearch final : public IMergeTreeIndexGranule -{ - MergeTreeIndexGranuleUSearch(const String & index_name_, const Block & index_sample_block_, unum::usearch::scalar_kind_t scalar_kind_); - MergeTreeIndexGranuleUSearch(const String & index_name_, const Block & index_sample_block_, unum::usearch::scalar_kind_t scalar_kind_, USearchIndexWithSerializationPtr index_); - - ~MergeTreeIndexGranuleUSearch() override = default; - - void serializeBinary(WriteBuffer & ostr) const override; - void deserializeBinary(ReadBuffer & istr, MergeTreeIndexVersion version) override; - - bool empty() const override { return !index.get(); } - - const String index_name; - const Block index_sample_block; - const unum::usearch::scalar_kind_t scalar_kind; - USearchIndexWithSerializationPtr index; -}; - - -template -struct MergeTreeIndexAggregatorUSearch final : IMergeTreeIndexAggregator -{ - MergeTreeIndexAggregatorUSearch(const String & index_name_, const Block & index_sample_block, unum::usearch::scalar_kind_t scalar_kind_); - ~MergeTreeIndexAggregatorUSearch() override = default; - - bool empty() const override { return !index || index->size() == 0; } - MergeTreeIndexGranulePtr getGranuleAndReset() override; - void update(const Block & block, size_t * pos, size_t limit) override; - - const String index_name; - const Block index_sample_block; - const unum::usearch::scalar_kind_t scalar_kind; - USearchIndexWithSerializationPtr index; -}; - - -class MergeTreeIndexConditionUSearch final : public IMergeTreeIndexConditionApproximateNearestNeighbor -{ -public: - MergeTreeIndexConditionUSearch( - const IndexDescription & index_description, - const SelectQueryInfo & query, - const String & distance_function, - ContextPtr context); - - ~MergeTreeIndexConditionUSearch() override = default; - - bool alwaysUnknownOrTrue() const override; - bool mayBeTrueOnGranule(MergeTreeIndexGranulePtr idx_granule) const override; - std::vector getUsefulRanges(MergeTreeIndexGranulePtr idx_granule) const override; - -private: - template - std::vector getUsefulRangesImpl(MergeTreeIndexGranulePtr idx_granule) const; - - const ApproximateNearestNeighborCondition ann_condition; - const String distance_function; -}; - - -class MergeTreeIndexUSearch : public IMergeTreeIndex -{ -public: - MergeTreeIndexUSearch(const IndexDescription & index_, const String & distance_function_, unum::usearch::scalar_kind_t scalar_kind_); - - ~MergeTreeIndexUSearch() override = default; - - MergeTreeIndexGranulePtr createIndexGranule() const override; - MergeTreeIndexAggregatorPtr createIndexAggregator(const MergeTreeWriterSettings & settings) const override; - MergeTreeIndexConditionPtr createIndexCondition(const SelectQueryInfo & query, ContextPtr context) const; - MergeTreeIndexConditionPtr createIndexCondition(const ActionsDAG *, ContextPtr) const override; - bool isVectorSearch() const override { return true; } - -private: - const String distance_function; - const unum::usearch::scalar_kind_t scalar_kind; -}; - -} - - -#endif - diff --git a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp new file mode 100644 index 00000000000..fbbc66bd8db --- /dev/null +++ b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp @@ -0,0 +1,520 @@ +#include + +#if USE_USEARCH + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace ProfileEvents +{ + extern const Event USearchAddCount; + extern const Event USearchAddVisitedMembers; + extern const Event USearchAddComputedDistances; + extern const Event USearchSearchCount; + extern const Event USearchSearchVisitedMembers; + extern const Event USearchSearchComputedDistances; +} + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int CANNOT_ALLOCATE_MEMORY; + extern const int FORMAT_VERSION_TOO_OLD; + extern const int ILLEGAL_COLUMN; + extern const int INCORRECT_DATA; + extern const int INCORRECT_NUMBER_OF_COLUMNS; + extern const int INCORRECT_QUERY; + extern const int LOGICAL_ERROR; + extern const int NOT_IMPLEMENTED; +} + +namespace +{ + +/// The only indexing method currently supported by USearch +const std::set methods = {"hnsw"}; + +/// Maps from user-facing name to internal name +const std::unordered_map distanceFunctionToMetricKind = { + {"L2Distance", unum::usearch::metric_kind_t::l2sq_k}, + {"cosineDistance", unum::usearch::metric_kind_t::cos_k}}; + +/// Maps from user-facing name to internal name +const std::unordered_map quantizationToScalarKind = { + {"f32", unum::usearch::scalar_kind_t::f32_k}, + {"f16", unum::usearch::scalar_kind_t::f16_k}, + {"i8", unum::usearch::scalar_kind_t::i8_k}}; + +template +concept is_set = std::same_as>; + +template +concept is_unordered_map = std::same_as>; + +template +String joinByComma(const T & t) +{ + if constexpr (is_set) + { + return fmt::format("{}", fmt::join(t, ", ")); + } + else if constexpr (is_unordered_map) + { + String joined_keys; + for (const auto & [k, _] : t) + { + if (!joined_keys.empty()) + joined_keys += ", "; + joined_keys += k; + } + return joined_keys; + } + /// TODO once our libcxx is recent enough, replace above by + /// return fmt::format("{}", fmt::join(std::views::keys(t)), ", ")); + std::unreachable(); +} + +} + +USearchIndexWithSerialization::USearchIndexWithSerialization( + size_t dimensions, + unum::usearch::metric_kind_t metric_kind, + unum::usearch::scalar_kind_t scalar_kind, + UsearchHnswParams usearch_hnsw_params) +{ + USearchIndex::metric_t metric(dimensions, metric_kind, scalar_kind); + + unum::usearch::index_dense_config_t config(usearch_hnsw_params.m, usearch_hnsw_params.ef_construction, usearch_hnsw_params.ef_search); + config.enable_key_lookups = false; /// we don't do row-to-vector lookups + + if (auto error = config.validate(); error) /// already called in vectorSimilarityIndexValidator, call again because usearch may change the config in-place + throw Exception(ErrorCodes::INCORRECT_DATA, "Invalid parameters passed to vector similarity index. Error: {}", String(error.release())); + + if (auto result = USearchIndex::make(metric, config); !result) + throw Exception(ErrorCodes::INCORRECT_DATA, "Could not create vector similarity index. Error: {}", String(result.error.release())); + else + swap(result.index); +} + +void USearchIndexWithSerialization::serialize(WriteBuffer & ostr) const +{ + auto callback = [&ostr](void * from, size_t n) + { + ostr.write(reinterpret_cast(from), n); + return true; + }; + + if (auto result = Base::save_to_stream(callback); !result) + throw Exception(ErrorCodes::INCORRECT_DATA, "Could not save vector similarity index. Error: {}", String(result.error.release())); +} + +void USearchIndexWithSerialization::deserialize(ReadBuffer & istr) +{ + auto callback = [&istr](void * from, size_t n) + { + istr.readStrict(reinterpret_cast(from), n); + return true; + }; + + if (auto result = Base::load_from_stream(callback); !result) + /// See the comment in MergeTreeIndexGranuleVectorSimilarity::deserializeBinary why we throw here + throw Exception(ErrorCodes::INCORRECT_DATA, "Could not load vector similarity index. Please drop the index and create it again. Error: {}", String(result.error.release())); + + if (!try_reserve(limits())) + throw Exception(ErrorCodes::CANNOT_ALLOCATE_MEMORY, "Could not reserve memory for usearch index"); +} + +USearchIndexWithSerialization::Statistics USearchIndexWithSerialization::getStatistics() const +{ + USearchIndex::stats_t global_stats = Base::stats(); + + Statistics statistics = { + .max_level = max_level(), + .connectivity = connectivity(), + .size = size(), + .capacity = capacity(), + .memory_usage = memory_usage(), + .bytes_per_vector = bytes_per_vector(), + .scalar_words = scalar_words(), + .nodes = global_stats.nodes, + .edges = global_stats.edges, + .max_edges = global_stats.max_edges, + .level_stats = {}}; + + for (size_t i = 0; i < statistics.max_level; ++i) + statistics.level_stats.push_back(Base::stats(i)); + + return statistics; +} + +String USearchIndexWithSerialization::Statistics::toString() const +{ + return fmt::format("max_level = {}, connectivity = {}, size = {}, capacity = {}, memory_usage = {}, bytes_per_vector = {}, scalar_words = {}, nodes = {}, edges = {}, max_edges = {}", + max_level, connectivity, size, capacity, ReadableSize(memory_usage), bytes_per_vector, scalar_words, nodes, edges, max_edges); + +} +MergeTreeIndexGranuleVectorSimilarity::MergeTreeIndexGranuleVectorSimilarity( + const String & index_name_, + const Block & index_sample_block_, + unum::usearch::metric_kind_t metric_kind_, + unum::usearch::scalar_kind_t scalar_kind_, + UsearchHnswParams usearch_hnsw_params_) + : MergeTreeIndexGranuleVectorSimilarity(index_name_, index_sample_block_, metric_kind_, scalar_kind_, usearch_hnsw_params_, nullptr) +{ +} + +MergeTreeIndexGranuleVectorSimilarity::MergeTreeIndexGranuleVectorSimilarity( + const String & index_name_, + const Block & index_sample_block_, + unum::usearch::metric_kind_t metric_kind_, + unum::usearch::scalar_kind_t scalar_kind_, + UsearchHnswParams usearch_hnsw_params_, + USearchIndexWithSerializationPtr index_) + : index_name(index_name_) + , index_sample_block(index_sample_block_) + , metric_kind(metric_kind_) + , scalar_kind(scalar_kind_) + , usearch_hnsw_params(usearch_hnsw_params_) + , index(std::move(index_)) +{ +} + +void MergeTreeIndexGranuleVectorSimilarity::serializeBinary(WriteBuffer & ostr) const +{ + if (empty()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Attempt to write empty minmax index {}", backQuote(index_name)); + + writeIntBinary(FILE_FORMAT_VERSION, ostr); + + /// Number of dimensions is required in the index constructor, + /// so it must be written and read separately from the other part + writeIntBinary(static_cast(index->dimensions()), ostr); + + index->serialize(ostr); + + auto statistics = index->getStatistics(); + LOG_TRACE(logger, "Wrote vector similarity index: {}", statistics.toString()); +} + +void MergeTreeIndexGranuleVectorSimilarity::deserializeBinary(ReadBuffer & istr, MergeTreeIndexVersion /*version*/) +{ + UInt64 file_version; + readIntBinary(file_version, istr); + if (file_version != FILE_FORMAT_VERSION) + throw Exception( + ErrorCodes::FORMAT_VERSION_TOO_OLD, + "Vector similarity index could not be loaded because its version is too old (current version: {}, persisted version: {}). Please drop the index and create it again.", + FILE_FORMAT_VERSION, file_version); + /// More fancy error handling would be: Set a flag on the index that it failed to load. During usage return all granules, i.e. + /// behave as if the index does not exist. Since format changes are expected to happen only rarely and it is "only" an index, keep it simple for now. + + UInt64 dimension; + readIntBinary(dimension, istr); + index = std::make_shared(dimension, metric_kind, scalar_kind, usearch_hnsw_params); + + index->deserialize(istr); + + auto statistics = index->getStatistics(); + LOG_TRACE(logger, "Loaded vector similarity index: {}", statistics.toString()); +} + +MergeTreeIndexAggregatorVectorSimilarity::MergeTreeIndexAggregatorVectorSimilarity( + const String & index_name_, + const Block & index_sample_block_, + unum::usearch::metric_kind_t metric_kind_, + unum::usearch::scalar_kind_t scalar_kind_, + UsearchHnswParams usearch_hnsw_params_) + : index_name(index_name_) + , index_sample_block(index_sample_block_) + , metric_kind(metric_kind_) + , scalar_kind(scalar_kind_) + , usearch_hnsw_params(usearch_hnsw_params_) +{ +} + +MergeTreeIndexGranulePtr MergeTreeIndexAggregatorVectorSimilarity::getGranuleAndReset() +{ + auto granule = std::make_shared(index_name, index_sample_block, metric_kind, scalar_kind, usearch_hnsw_params, index); + index = nullptr; + return granule; +} + +void MergeTreeIndexAggregatorVectorSimilarity::update(const Block & block, size_t * pos, size_t limit) +{ + if (*pos >= block.rows()) + throw Exception( + ErrorCodes::LOGICAL_ERROR, + "The provided position is not less than the number of block rows. Position: {}, Block rows: {}.", + *pos, + block.rows()); + + size_t rows_read = std::min(limit, block.rows() - *pos); + + if (rows_read == 0) + return; + + if (rows_read > std::numeric_limits::max()) + throw Exception(ErrorCodes::INCORRECT_DATA, "Index granularity is too big: more than {} rows per index granule.", std::numeric_limits::max()); + + if (index_sample_block.columns() > 1) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected block with single column"); + + const String & index_column_name = index_sample_block.getByPosition(0).name; + ColumnPtr column_cut = block.getByName(index_column_name).column->cut(*pos, rows_read); + + if (const auto & column_array = typeid_cast(column_cut.get())) + { + const auto & column_array_data = column_array->getData(); + const auto & column_array_data_float = typeid_cast(column_array_data); + const auto & column_array_data_float_data = column_array_data_float.getData(); + + const auto & column_array_offsets = column_array->getOffsets(); + const size_t num_rows = column_array_offsets.size(); + + if (column_array->empty()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Array is unexpectedly empty"); + + /// The vector similarity algorithm naturally assumes that the indexed vectors have dimension >= 1. This condition is violated if empty arrays + /// are INSERTed into an vector-similarity-indexed column or if no value was specified at all in which case the arrays take on their default + /// values which is also empty. + if (column_array->isDefaultAt(0)) + throw Exception(ErrorCodes::INCORRECT_DATA, "The arrays in column '{}' must not be empty. Did you try to INSERT default values?", index_column_name); + + /// Check all sizes are the same + const size_t dimensions = column_array_offsets[0]; + for (size_t i = 0; i < num_rows - 1; ++i) + if (column_array_offsets[i + 1] - column_array_offsets[i] != dimensions) + throw Exception(ErrorCodes::INCORRECT_DATA, "All arrays in column '{}' must have equal length", index_column_name); + + /// Also check that previously inserted blocks have the same size as this block. + /// Note that this guarantees consistency of dimension only within parts. We are unable to detect inconsistent dimensions across + /// parts - for this, a little help from the user is needed, e.g. CONSTRAINT cnstr CHECK length(array) = 42. + if (index && index->dimensions() != dimensions) + throw Exception(ErrorCodes::INCORRECT_DATA, "All arrays in column '{}' must have equal length", index_column_name); + + if (!index) + index = std::make_shared(dimensions, metric_kind, scalar_kind, usearch_hnsw_params); + + /// We use Usearch's index_dense_t as index type which supports only 4 bio entries according to https://github.com/unum-cloud/usearch/tree/main/cpp + if (index->size() + num_rows > std::numeric_limits::max()) + throw Exception(ErrorCodes::INCORRECT_DATA, "Size of vector similarity index in column {} would exceed 4 billion entries", index_column_name); + + /// Reserving space is mandatory + if (!index->try_reserve(roundUpToPowerOfTwoOrZero(index->size() + num_rows))) + throw Exception(ErrorCodes::CANNOT_ALLOCATE_MEMORY, "Could not reserve memory for vector similarity index"); + + for (size_t row = 0; row < num_rows; ++row) + { + if (auto result = index->add(static_cast(index->size()), &column_array_data_float_data[column_array_offsets[row - 1]]); !result) + throw Exception(ErrorCodes::INCORRECT_DATA, "Could not add data to vector similarity index. Error: {}", String(result.error.release())); + else + { + ProfileEvents::increment(ProfileEvents::USearchAddCount); + ProfileEvents::increment(ProfileEvents::USearchAddVisitedMembers, result.visited_members); + ProfileEvents::increment(ProfileEvents::USearchAddComputedDistances, result.computed_distances); + } + } + } + else + throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected Array(Float32) column"); + + *pos += rows_read; +} + +MergeTreeIndexConditionVectorSimilarity::MergeTreeIndexConditionVectorSimilarity( + const IndexDescription & /*index_description*/, + const SelectQueryInfo & query, + unum::usearch::metric_kind_t metric_kind_, + ContextPtr context) + : vector_similarity_condition(query, context) + , metric_kind(metric_kind_) +{ +} + +bool MergeTreeIndexConditionVectorSimilarity::mayBeTrueOnGranule(MergeTreeIndexGranulePtr) const +{ + throw Exception(ErrorCodes::LOGICAL_ERROR, "mayBeTrueOnGranule is not supported for ANN skip indexes"); +} + +bool MergeTreeIndexConditionVectorSimilarity::alwaysUnknownOrTrue() const +{ + String index_distance_function; + switch (metric_kind) + { + case unum::usearch::metric_kind_t::l2sq_k: index_distance_function = "L2Distance"; break; + case unum::usearch::metric_kind_t::cos_k: index_distance_function = "cosineDistance"; break; + default: std::unreachable(); + } + return vector_similarity_condition.alwaysUnknownOrTrue(index_distance_function); +} + +std::vector MergeTreeIndexConditionVectorSimilarity::getUsefulRanges(MergeTreeIndexGranulePtr granule_) const +{ + const UInt64 limit = vector_similarity_condition.getLimit(); + const UInt64 index_granularity = vector_similarity_condition.getIndexGranularity(); + + const auto granule = std::dynamic_pointer_cast(granule_); + if (granule == nullptr) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Granule has the wrong type"); + + const USearchIndexWithSerializationPtr index = granule->index; + + if (vector_similarity_condition.getDimensions() != index->dimensions()) + throw Exception(ErrorCodes::INCORRECT_QUERY, "The dimension of the space in the request ({}) " + "does not match the dimension in the index ({})", + vector_similarity_condition.getDimensions(), index->dimensions()); + + const std::vector reference_vector = vector_similarity_condition.getReferenceVector(); + + auto search_result = index->search(reference_vector.data(), limit); + if (!search_result) + throw Exception(ErrorCodes::INCORRECT_DATA, "Could not search in vector similarity index. Error: {}", String(search_result.error.release())); + + ProfileEvents::increment(ProfileEvents::USearchSearchCount); + ProfileEvents::increment(ProfileEvents::USearchSearchVisitedMembers, search_result.visited_members); + ProfileEvents::increment(ProfileEvents::USearchSearchComputedDistances, search_result.computed_distances); + + std::vector neighbors(search_result.size()); /// indexes of vectors which were closest to the reference vector + search_result.dump_to(neighbors.data()); + + std::vector granules; + granules.reserve(neighbors.size()); + for (auto neighbor : neighbors) + granules.push_back(neighbor / index_granularity); + + /// make unique + std::sort(granules.begin(), granules.end()); + granules.erase(std::unique(granules.begin(), granules.end()), granules.end()); + + return granules; +} + +MergeTreeIndexVectorSimilarity::MergeTreeIndexVectorSimilarity( + const IndexDescription & index_, + unum::usearch::metric_kind_t metric_kind_, + unum::usearch::scalar_kind_t scalar_kind_, + UsearchHnswParams usearch_hnsw_params_) + : IMergeTreeIndex(index_) + , metric_kind(metric_kind_) + , scalar_kind(scalar_kind_) + , usearch_hnsw_params(usearch_hnsw_params_) +{ +} + +MergeTreeIndexGranulePtr MergeTreeIndexVectorSimilarity::createIndexGranule() const +{ + return std::make_shared(index.name, index.sample_block, metric_kind, scalar_kind, usearch_hnsw_params); +} + +MergeTreeIndexAggregatorPtr MergeTreeIndexVectorSimilarity::createIndexAggregator(const MergeTreeWriterSettings & /*settings*/) const +{ + return std::make_shared(index.name, index.sample_block, metric_kind, scalar_kind, usearch_hnsw_params); +} + +MergeTreeIndexConditionPtr MergeTreeIndexVectorSimilarity::createIndexCondition(const SelectQueryInfo & query, ContextPtr context) const +{ + return std::make_shared(index, query, metric_kind, context); +}; + +MergeTreeIndexConditionPtr MergeTreeIndexVectorSimilarity::createIndexCondition(const ActionsDAG *, ContextPtr) const +{ + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "MergeTreeIndexAnnoy cannot be created with ActionsDAG"); +} + +MergeTreeIndexPtr vectorSimilarityIndexCreator(const IndexDescription & index) +{ + /// Default parameters: + unum::usearch::metric_kind_t metric_kind = distanceFunctionToMetricKind.at(index.arguments[1].safeGet()); + unum::usearch::scalar_kind_t scalar_kind = unum::usearch::scalar_kind_t::f32_k; + UsearchHnswParams usearch_hnsw_params; + + /// Optional parameters: + const bool has_six_args = (index.arguments.size() == 6); + if (has_six_args) + { + scalar_kind = quantizationToScalarKind.at(index.arguments[2].safeGet()); + usearch_hnsw_params = {.m = index.arguments[3].safeGet(), + .ef_construction = index.arguments[4].safeGet(), + .ef_search = index.arguments[5].safeGet()}; + } + + return std::make_shared(index, metric_kind, scalar_kind, usearch_hnsw_params); +} + +void vectorSimilarityIndexValidator(const IndexDescription & index, bool /* attach */) +{ + const bool has_two_args = (index.arguments.size() == 2); + const bool has_six_args = (index.arguments.size() == 6); + + /// Check number and type of arguments + if (!has_two_args && !has_six_args) + throw Exception(ErrorCodes::INCORRECT_QUERY, "Vector similarity index must have two or six arguments"); + if (index.arguments[0].getType() != Field::Types::String) + throw Exception(ErrorCodes::INCORRECT_QUERY, "First argument of vector similarity index (method) must be of type String"); + if (index.arguments[1].getType() != Field::Types::String) + throw Exception(ErrorCodes::INCORRECT_QUERY, "Second argument of vector similarity index (metric) must be of type String"); + if (has_six_args) + { + if (index.arguments[2].getType() != Field::Types::String) + throw Exception(ErrorCodes::INCORRECT_QUERY, "Third argument of vector similarity index (quantization) must be of type String"); + if (index.arguments[3].getType() != Field::Types::UInt64) + throw Exception(ErrorCodes::INCORRECT_QUERY, "Fourth argument of vector similarity index (M) must be of type UInt64"); + if (index.arguments[4].getType() != Field::Types::UInt64) + throw Exception(ErrorCodes::INCORRECT_QUERY, "Fifth argument of vector similarity index (ef_construction) must be of type UInt64"); + if (index.arguments[5].getType() != Field::Types::UInt64) + throw Exception(ErrorCodes::INCORRECT_QUERY, "Sixth argument of vector similarity index (ef_search) must be of type UInt64"); + } + + /// Check that passed arguments are supported + if (!methods.contains(index.arguments[0].safeGet())) + throw Exception(ErrorCodes::INCORRECT_DATA, "First argument (method) of vector similarity index is not supported. Supported methods are: {}", joinByComma(methods)); + if (!distanceFunctionToMetricKind.contains(index.arguments[1].safeGet())) + throw Exception(ErrorCodes::INCORRECT_DATA, "Second argument (distance function) of vector similarity index is not supported. Supported distance function are: {}", joinByComma(distanceFunctionToMetricKind)); + if (has_six_args) + { + if (!quantizationToScalarKind.contains(index.arguments[2].safeGet())) + throw Exception(ErrorCodes::INCORRECT_DATA, "Third argument (quantization) of vector similarity index is not supported. Supported quantizations are: {}", joinByComma(quantizationToScalarKind)); + + /// Call Usearche's own parameter validation method for HNSW-specific parameters + UInt64 m = index.arguments[3].safeGet(); + UInt64 ef_construction = index.arguments[4].safeGet(); + UInt64 ef_search = index.arguments[5].safeGet(); + + unum::usearch::index_dense_config_t config(m, ef_construction, ef_search); + + if (auto error = config.validate(); error) + throw Exception(ErrorCodes::INCORRECT_DATA, "Invalid parameters passed to vector similarity index. Error: {}", String(error.release())); + } + + /// Check that the index is created on a single column + if (index.column_names.size() != 1 || index.data_types.size() != 1) + throw Exception(ErrorCodes::INCORRECT_NUMBER_OF_COLUMNS, "Vector similarity indexes must be created on a single column"); + + /// Check data type of the indexed column: + DataTypePtr data_type = index.sample_block.getDataTypes()[0]; + if (const auto * data_type_array = typeid_cast(data_type.get())) + { + TypeIndex nested_type_index = data_type_array->getNestedType()->getTypeId(); + if (!WhichDataType(nested_type_index).isFloat32()) + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Vector similarity indexes can only be created on columns of type Array(Float32)"); + } + else + { + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Vector similarity indexes can only be created on columns of type Array(Float32)"); + } +} + +} + +#endif diff --git a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.h b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.h new file mode 100644 index 00000000000..c4c03254d2d --- /dev/null +++ b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.h @@ -0,0 +1,174 @@ +#pragma once + +#include "config.h" + +#if USE_USEARCH + +#include +#include +#include + +namespace DB +{ + +struct UsearchHnswParams +{ + size_t m = unum::usearch::default_connectivity(); + size_t ef_construction = unum::usearch::default_expansion_add(); + size_t ef_search = unum::usearch::default_expansion_search(); +}; + +using USearchIndex = unum::usearch::index_dense_t; + +class USearchIndexWithSerialization : public USearchIndex +{ + using Base = USearchIndex; + +public: + USearchIndexWithSerialization( + size_t dimensions, + unum::usearch::metric_kind_t metric_kind, + unum::usearch::scalar_kind_t scalar_kind, + UsearchHnswParams usearch_hnsw_params); + + void serialize(WriteBuffer & ostr) const; + void deserialize(ReadBuffer & istr); + + struct Statistics + { + size_t max_level; + size_t connectivity; + size_t size; /// number of indexed vectors + size_t capacity; /// reserved number of indexed vectors + size_t memory_usage; /// byte size (not exact) + size_t bytes_per_vector; + size_t scalar_words; + size_t nodes; + size_t edges; + size_t max_edges; + + std::vector level_stats; /// for debugging, excluded from getStatistics() + + String toString() const; + }; + + Statistics getStatistics() const; +}; + +using USearchIndexWithSerializationPtr = std::shared_ptr; + + +struct MergeTreeIndexGranuleVectorSimilarity final : public IMergeTreeIndexGranule +{ + MergeTreeIndexGranuleVectorSimilarity( + const String & index_name_, + const Block & index_sample_block_, + unum::usearch::metric_kind_t metric_kind_, + unum::usearch::scalar_kind_t scalar_kind_, + UsearchHnswParams usearch_hnsw_params_); + + MergeTreeIndexGranuleVectorSimilarity( + const String & index_name_, + const Block & index_sample_block_, + unum::usearch::metric_kind_t metric_kind_, + unum::usearch::scalar_kind_t scalar_kind_, + UsearchHnswParams usearch_hnsw_params_, + USearchIndexWithSerializationPtr index_); + + ~MergeTreeIndexGranuleVectorSimilarity() override = default; + + void serializeBinary(WriteBuffer & ostr) const override; + void deserializeBinary(ReadBuffer & istr, MergeTreeIndexVersion version) override; + + bool empty() const override { return !index || index->size() == 0; } + + const String index_name; + const Block index_sample_block; + const unum::usearch::metric_kind_t metric_kind; + const unum::usearch::scalar_kind_t scalar_kind; + const UsearchHnswParams usearch_hnsw_params; + USearchIndexWithSerializationPtr index; + + LoggerPtr logger = getLogger("VectorSimilarityIndex"); + +private: + /// The version of the persistence format of USearch index. Increment whenever you change the format. + /// Note: USearch prefixes the serialized data with its own version header. We can't rely on that because 1. the index in ClickHouse + /// is (at least in theory) agnostic of specific vector search libraries, and 2. additional data (e.g. the number of dimensions) + /// outside USearch exists which we should version separately. + static constexpr UInt64 FILE_FORMAT_VERSION = 1; +}; + + +struct MergeTreeIndexAggregatorVectorSimilarity final : IMergeTreeIndexAggregator +{ + MergeTreeIndexAggregatorVectorSimilarity( + const String & index_name_, + const Block & index_sample_block, + unum::usearch::metric_kind_t metric_kind_, + unum::usearch::scalar_kind_t scalar_kind_, + UsearchHnswParams usearch_hnsw_params_); + + ~MergeTreeIndexAggregatorVectorSimilarity() override = default; + + bool empty() const override { return !index || index->size() == 0; } + MergeTreeIndexGranulePtr getGranuleAndReset() override; + void update(const Block & block, size_t * pos, size_t limit) override; + + const String index_name; + const Block index_sample_block; + const unum::usearch::metric_kind_t metric_kind; + const unum::usearch::scalar_kind_t scalar_kind; + const UsearchHnswParams usearch_hnsw_params; + USearchIndexWithSerializationPtr index; +}; + + +class MergeTreeIndexConditionVectorSimilarity final : public IMergeTreeIndexCondition +{ +public: + MergeTreeIndexConditionVectorSimilarity( + const IndexDescription & index_description, + const SelectQueryInfo & query, + unum::usearch::metric_kind_t metric_kind_, + ContextPtr context); + + ~MergeTreeIndexConditionVectorSimilarity() override = default; + + bool alwaysUnknownOrTrue() const override; + bool mayBeTrueOnGranule(MergeTreeIndexGranulePtr granule) const override; + std::vector getUsefulRanges(MergeTreeIndexGranulePtr granule) const override; + +private: + const VectorSimilarityCondition vector_similarity_condition; + const unum::usearch::metric_kind_t metric_kind; +}; + + +class MergeTreeIndexVectorSimilarity : public IMergeTreeIndex +{ +public: + MergeTreeIndexVectorSimilarity( + const IndexDescription & index_, + unum::usearch::metric_kind_t metric_kind_, + unum::usearch::scalar_kind_t scalar_kind_, + UsearchHnswParams usearch_hnsw_params_); + + ~MergeTreeIndexVectorSimilarity() override = default; + + MergeTreeIndexGranulePtr createIndexGranule() const override; + MergeTreeIndexAggregatorPtr createIndexAggregator(const MergeTreeWriterSettings & settings) const override; + MergeTreeIndexConditionPtr createIndexCondition(const SelectQueryInfo & query, ContextPtr context) const; + MergeTreeIndexConditionPtr createIndexCondition(const ActionsDAG *, ContextPtr) const override; + bool isVectorSimilarityIndex() const override { return true; } + +private: + const unum::usearch::metric_kind_t metric_kind; + const unum::usearch::scalar_kind_t scalar_kind; + const UsearchHnswParams usearch_hnsw_params; +}; + +} + + +#endif diff --git a/src/Storages/MergeTree/MergeTreeIndices.cpp b/src/Storages/MergeTree/MergeTreeIndices.cpp index bded961db8e..d2fc0e84b56 100644 --- a/src/Storages/MergeTree/MergeTreeIndices.cpp +++ b/src/Storages/MergeTree/MergeTreeIndices.cpp @@ -127,15 +127,21 @@ MergeTreeIndexFactory::MergeTreeIndexFactory() registerCreator("hypothesis", hypothesisIndexCreator); registerValidator("hypothesis", hypothesisIndexValidator); -#ifdef ENABLE_ANNOY - registerCreator("annoy", annoyIndexCreator); - registerValidator("annoy", annoyIndexValidator); -#endif -#ifdef ENABLE_USEARCH - registerCreator("usearch", usearchIndexCreator); - registerValidator("usearch", usearchIndexValidator); +#if USE_USEARCH + registerCreator("vector_similarity", vectorSimilarityIndexCreator); + registerValidator("vector_similarity", vectorSimilarityIndexValidator); #endif + /// ------ + /// TODO: remove this block at the end of 2024. + /// Index types 'annoy' and 'usearch' are no longer supported as of June 2024. Their successor is index type 'vector_similarity'. + /// To support loading tables with old indexes during a transition period, register dummy indexes which allow load/attaching but + /// throw an exception when the user attempts to use them. + registerCreator("annoy", legacyVectorSimilarityIndexCreator); + registerValidator("annoy", legacyVectorSimilarityIndexValidator); + registerCreator("usearch", legacyVectorSimilarityIndexCreator); + registerValidator("usearch", legacyVectorSimilarityIndexValidator); + /// ------ registerCreator("inverted", fullTextIndexCreator); registerValidator("inverted", fullTextIndexValidator); diff --git a/src/Storages/MergeTree/MergeTreeIndices.h b/src/Storages/MergeTree/MergeTreeIndices.h index 1be73e1c811..c52d7ffe131 100644 --- a/src/Storages/MergeTree/MergeTreeIndices.h +++ b/src/Storages/MergeTree/MergeTreeIndices.h @@ -15,6 +15,7 @@ #include #include +#include "config.h" constexpr auto INDEX_FILE_PREFIX = "skp_idx_"; @@ -92,6 +93,13 @@ public: virtual bool alwaysUnknownOrTrue() const = 0; virtual bool mayBeTrueOnGranule(MergeTreeIndexGranulePtr granule) const = 0; + + /// Special stuff for vector similarity indexes + /// - Returns vector of indexes of ranges in granule which are useful for query. + virtual std::vector getUsefulRanges(MergeTreeIndexGranulePtr) const + { + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Not implemented for non-vector-similarity indexes."); + } }; using MergeTreeIndexConditionPtr = std::shared_ptr; @@ -169,7 +177,7 @@ struct IMergeTreeIndex virtual MergeTreeIndexConditionPtr createIndexCondition( const ActionsDAG * filter_actions_dag, ContextPtr context) const = 0; - virtual bool isVectorSearch() const { return false; } + virtual bool isVectorSimilarityIndex() const { return false; } virtual MergeTreeIndexMergedConditionPtr createIndexMergedCondition( const SelectQueryInfo & /*query_info*/, StorageMetadataPtr /*storage_metadata*/) const @@ -230,17 +238,15 @@ void bloomFilterIndexValidator(const IndexDescription & index, bool attach); MergeTreeIndexPtr hypothesisIndexCreator(const IndexDescription & index); void hypothesisIndexValidator(const IndexDescription & index, bool attach); -#ifdef ENABLE_ANNOY -MergeTreeIndexPtr annoyIndexCreator(const IndexDescription & index); -void annoyIndexValidator(const IndexDescription & index, bool attach); +#if USE_USEARCH +MergeTreeIndexPtr vectorSimilarityIndexCreator(const IndexDescription & index); +void vectorSimilarityIndexValidator(const IndexDescription & index, bool attach); #endif -#ifdef ENABLE_USEARCH -MergeTreeIndexPtr usearchIndexCreator(const IndexDescription& index); -void usearchIndexValidator(const IndexDescription& index, bool attach); -#endif +MergeTreeIndexPtr legacyVectorSimilarityIndexCreator(const IndexDescription & index); +void legacyVectorSimilarityIndexValidator(const IndexDescription & index, bool attach); -MergeTreeIndexPtr fullTextIndexCreator(const IndexDescription& index); -void fullTextIndexValidator(const IndexDescription& index, bool attach); +MergeTreeIndexPtr fullTextIndexCreator(const IndexDescription & index); +void fullTextIndexValidator(const IndexDescription & index, bool attach); } diff --git a/src/Storages/MergeTree/MergeTreePartition.cpp b/src/Storages/MergeTree/MergeTreePartition.cpp index b240f80ee13..5b5bc244f92 100644 --- a/src/Storages/MergeTree/MergeTreePartition.cpp +++ b/src/Storages/MergeTree/MergeTreePartition.cpp @@ -241,7 +241,7 @@ String MergeTreePartition::getID(const Block & partition_key_sample) const if (typeid_cast(partition_key_sample.getByPosition(i).type.get())) result += toString(DateLUT::serverTimezoneInstance().toNumYYYYMMDD(DayNum(value[i].safeGet()))); else if (typeid_cast(partition_key_sample.getByPosition(i).type.get())) - result += toString(value[i].get().toUnderType()); + result += toString(value[i].safeGet().toUnderType()); else result += applyVisitor(to_string_visitor, value[i]); diff --git a/src/Storages/MergeTree/MergeTreeReaderCompact.cpp b/src/Storages/MergeTree/MergeTreeReaderCompact.cpp index a2b8f0ad96f..69dc2e4b2bb 100644 --- a/src/Storages/MergeTree/MergeTreeReaderCompact.cpp +++ b/src/Storages/MergeTree/MergeTreeReaderCompact.cpp @@ -60,39 +60,25 @@ void MergeTreeReaderCompact::fillColumnPositions() for (size_t i = 0; i < columns_num; ++i) { - const auto & column_to_read = columns_to_read[i]; - + auto & column_to_read = columns_to_read[i]; auto position = data_part_info_for_read->getColumnPosition(column_to_read.getNameInStorage()); - bool is_array = isArray(column_to_read.type); if (column_to_read.isSubcolumn()) { - auto storage_column_from_part = getColumnInPart( - {column_to_read.getNameInStorage(), column_to_read.getTypeInStorage()}); + NameAndTypePair column_in_storage{column_to_read.getNameInStorage(), column_to_read.getTypeInStorage()}; + auto storage_column_from_part = getColumnInPart(column_in_storage); auto subcolumn_name = column_to_read.getSubcolumnName(); if (!storage_column_from_part.type->hasSubcolumn(subcolumn_name)) position.reset(); } + column_positions[i] = std::move(position); + /// If array of Nested column is missing in part, /// we have to read its offsets if they exist. - if (!position && is_array) - { - auto column_to_read_with_subcolumns = getColumnConvertedToSubcolumnOfNested(column_to_read); - auto name_level_for_offsets = findColumnForOffsets(column_to_read_with_subcolumns); - - if (name_level_for_offsets.has_value()) - { - column_positions[i] = data_part_info_for_read->getColumnPosition(name_level_for_offsets->first); - columns_for_offsets[i] = name_level_for_offsets; - partially_read_columns.insert(column_to_read.name); - } - } - else - { - column_positions[i] = std::move(position); - } + if (!column_positions[i]) + findPositionForMissedNested(i); } } @@ -115,7 +101,7 @@ NameAndTypePair MergeTreeReaderCompact::getColumnConvertedToSubcolumnOfNested(co if (!storage_columns_with_collected_nested) { - auto options = GetColumnsOptions(GetColumnsOptions::AllPhysical).withExtendedObjects(); + auto options = GetColumnsOptions(GetColumnsOptions::All).withExtendedObjects(); auto storage_columns_list = Nested::collect(storage_snapshot->getColumns(options)); storage_columns_with_collected_nested = ColumnsDescription(std::move(storage_columns_list)); } @@ -125,11 +111,44 @@ NameAndTypePair MergeTreeReaderCompact::getColumnConvertedToSubcolumnOfNested(co Nested::concatenateName(name_in_storage, subcolumn_name)); } +void MergeTreeReaderCompact::findPositionForMissedNested(size_t pos) +{ + auto & column = columns_to_read[pos]; + + bool is_array = isArray(column.type); + bool is_offsets_subcolumn = isArray(column.getTypeInStorage()) && column.getSubcolumnName() == "size0"; + + if (!is_array && !is_offsets_subcolumn) + return; + + NameAndTypePair column_in_storage{column.getNameInStorage(), column.getTypeInStorage()}; + + auto column_to_read_with_subcolumns = getColumnConvertedToSubcolumnOfNested(column_in_storage); + auto name_level_for_offsets = findColumnForOffsets(column_to_read_with_subcolumns); + + if (!name_level_for_offsets) + return; + + column_positions[pos] = data_part_info_for_read->getColumnPosition(name_level_for_offsets->first); + + if (is_offsets_subcolumn) + { + /// Read offsets from antoher array from the same Nested column. + column = {name_level_for_offsets->first, column.getSubcolumnName(), column.getTypeInStorage(), column.type}; + } + else + { + columns_for_offsets[pos] = std::move(name_level_for_offsets); + partially_read_columns.insert(column.name); + } +} + void MergeTreeReaderCompact::readData( const NameAndTypePair & name_and_type, ColumnPtr & column, size_t rows_to_read, - const InputStreamGetter & getter) + const InputStreamGetter & getter, + ISerialization::SubstreamsCache & cache) { try { @@ -140,6 +159,13 @@ void MergeTreeReaderCompact::readData( deserialize_settings.getter = getter; deserialize_settings.avg_value_size_hint = avg_value_size_hints[name]; + auto it = cache.find(name); + if (it != cache.end() && it->second != nullptr) + { + column = it->second; + return; + } + if (name_and_type.isSubcolumn()) { const auto & type_in_storage = name_and_type.getTypeInStorage(); @@ -163,6 +189,8 @@ void MergeTreeReaderCompact::readData( serialization->deserializeBinaryBulkWithMultipleStreams(column, rows_to_read, deserialize_settings, deserialize_binary_bulk_state_map[name], nullptr); } + cache[name] = column; + size_t read_rows_in_column = column->size() - column_size_before_reading; if (read_rows_in_column != rows_to_read) throw Exception(ErrorCodes::CANNOT_READ_ALL_DATA, diff --git a/src/Storages/MergeTree/MergeTreeReaderCompact.h b/src/Storages/MergeTree/MergeTreeReaderCompact.h index a783e595af5..1c6bd1474e3 100644 --- a/src/Storages/MergeTree/MergeTreeReaderCompact.h +++ b/src/Storages/MergeTree/MergeTreeReaderCompact.h @@ -36,6 +36,7 @@ public: protected: void fillColumnPositions(); NameAndTypePair getColumnConvertedToSubcolumnOfNested(const NameAndTypePair & column); + void findPositionForMissedNested(size_t pos); using InputStreamGetter = ISerialization::InputStreamGetter; @@ -43,7 +44,8 @@ protected: const NameAndTypePair & name_and_type, ColumnPtr & column, size_t rows_to_read, - const InputStreamGetter & getter); + const InputStreamGetter & getter, + ISerialization::SubstreamsCache & cache); void readPrefix( const NameAndTypePair & name_and_type, diff --git a/src/Storages/MergeTree/MergeTreeReaderCompactSingleBuffer.cpp b/src/Storages/MergeTree/MergeTreeReaderCompactSingleBuffer.cpp index 2b2cf493bb5..649bcce1188 100644 --- a/src/Storages/MergeTree/MergeTreeReaderCompactSingleBuffer.cpp +++ b/src/Storages/MergeTree/MergeTreeReaderCompactSingleBuffer.cpp @@ -26,6 +26,10 @@ try { size_t rows_to_read = data_part_info_for_read->getIndexGranularity().getMarkRows(from_mark); + /// Use cache to avoid reading the column with the same name twice. + /// It may happen if there are empty array Nested in the part. + ISerialization::SubstreamsCache cache; + for (size_t pos = 0; pos < num_columns; ++pos) { if (!res_columns[pos]) @@ -52,7 +56,7 @@ try }; readPrefix(columns_to_read[pos], buffer_getter, buffer_getter_for_prefix, columns_for_offsets[pos]); - readData(columns_to_read[pos], column, rows_to_read, buffer_getter); + readData(columns_to_read[pos], column, rows_to_read, buffer_getter, cache); } ++from_mark; diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.h b/src/Storages/MergeTree/MergeTreeSelectProcessor.h index 03ca30dd5b3..7a9cebbcb2e 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.h +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.h @@ -48,7 +48,7 @@ public: ChunkAndProgress read(); - void cancel() { is_cancelled = true; } + void cancel() noexcept { is_cancelled = true; } const MergeTreeReaderSettings & getSettings() const { return reader_settings; } diff --git a/src/Storages/MergeTree/MergeTreeSettings.cpp b/src/Storages/MergeTree/MergeTreeSettings.cpp index c968ad84936..dabb6991b0b 100644 --- a/src/Storages/MergeTree/MergeTreeSettings.cpp +++ b/src/Storages/MergeTree/MergeTreeSettings.cpp @@ -1,6 +1,6 @@ #include #include -#include +#include #include #include #include @@ -59,15 +59,19 @@ void MergeTreeSettings::loadFromQuery(ASTStorage & storage_def, ContextPtr conte CustomType custom; if (name == "disk") { + ASTPtr value_as_custom_ast = nullptr; if (value.tryGet(custom) && 0 == strcmp(custom.getTypeName(), "AST")) + value_as_custom_ast = dynamic_cast(custom.getImpl()).ast; + + if (value_as_custom_ast && isDiskFunction(value_as_custom_ast)) { - auto ast = dynamic_cast(custom.getImpl()).ast; - if (ast && isDiskFunction(ast)) - { - auto disk_name = getOrCreateDiskFromDiskAST(ast, context, is_attach); - LOG_TRACE(getLogger("MergeTreeSettings"), "Created custom disk {}", disk_name); - value = disk_name; - } + auto disk_name = DiskFomAST::createCustomDisk(value_as_custom_ast, context, is_attach); + LOG_DEBUG(getLogger("MergeTreeSettings"), "Created custom disk {}", disk_name); + value = disk_name; + } + else + { + DiskFomAST::ensureDiskIsNotCustom(value.safeGet(), context); } if (has("storage_policy")) diff --git a/src/Storages/MergeTree/MergeTreeSettings.h b/src/Storages/MergeTree/MergeTreeSettings.h index f5ada81cf55..d6d56272d7b 100644 --- a/src/Storages/MergeTree/MergeTreeSettings.h +++ b/src/Storages/MergeTree/MergeTreeSettings.h @@ -215,6 +215,8 @@ struct Settings; M(Float, primary_key_ratio_of_unique_prefix_values_to_skip_suffix_columns, 0.9f, "If the value of a column of the primary key in data part changes at least in this ratio of times, skip loading next columns in memory. This allows to save memory usage by not loading useless columns of the primary key.", 0) \ /** Projection settings. */ \ M(UInt64, max_projections, 25, "The maximum number of merge tree projections.", 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 projections of this table's relevant parts, or rebuild the projections.", 0) \ + M(DeduplicateMergeProjectionMode, deduplicate_merge_projection_mode, DeduplicateMergeProjectionMode::THROW, "Whether to allow create projection for the table with non-classic MergeTree, if allowed, what is the action when merge, drop or rebuild.", 0) \ #define MAKE_OBSOLETE_MERGE_TREE_SETTING(M, TYPE, NAME, DEFAULT) \ M(TYPE, NAME, DEFAULT, "Obsolete setting, does nothing.", BaseSettingsHelpers::Flags::OBSOLETE) diff --git a/src/Storages/MergeTree/MergeTreeSink.cpp b/src/Storages/MergeTree/MergeTreeSink.cpp index 0ca4276b858..459c5d5f243 100644 --- a/src/Storages/MergeTree/MergeTreeSink.cpp +++ b/src/Storages/MergeTree/MergeTreeSink.cpp @@ -34,7 +34,18 @@ struct MergeTreeSink::DelayedChunk }; -MergeTreeSink::~MergeTreeSink() = default; +MergeTreeSink::~MergeTreeSink() +{ + if (!delayed_chunk) + return; + + for (auto & partition : delayed_chunk->partitions) + { + partition.temp_part.cancel(); + } + + delayed_chunk.reset(); +} MergeTreeSink::MergeTreeSink( StorageMergeTree & storage_, @@ -59,13 +70,10 @@ void MergeTreeSink::onStart() void MergeTreeSink::onFinish() { + chassert(!isCancelled()); finishDelayedChunk(); } -void MergeTreeSink::onCancel() -{ -} - void MergeTreeSink::consume(Chunk & chunk) { if (num_blocks_processed > 0) diff --git a/src/Storages/MergeTree/MergeTreeSink.h b/src/Storages/MergeTree/MergeTreeSink.h index 90976020d52..8f065773d6a 100644 --- a/src/Storages/MergeTree/MergeTreeSink.h +++ b/src/Storages/MergeTree/MergeTreeSink.h @@ -28,7 +28,6 @@ public: void consume(Chunk & chunk) override; void onStart() override; void onFinish() override; - void onCancel() override; private: StorageMergeTree & storage; diff --git a/src/Storages/MergeTree/MergeTreeSource.cpp b/src/Storages/MergeTree/MergeTreeSource.cpp index e323b9f9ee7..380c47723bc 100644 --- a/src/Storages/MergeTree/MergeTreeSource.cpp +++ b/src/Storages/MergeTree/MergeTreeSource.cpp @@ -149,7 +149,7 @@ std::string MergeTreeSource::getName() const return processor->getName(); } -void MergeTreeSource::onCancel() +void MergeTreeSource::onCancel() noexcept { processor->cancel(); } diff --git a/src/Storages/MergeTree/MergeTreeSource.h b/src/Storages/MergeTree/MergeTreeSource.h index fc39b4f9b09..7506af4f9b8 100644 --- a/src/Storages/MergeTree/MergeTreeSource.h +++ b/src/Storages/MergeTree/MergeTreeSource.h @@ -26,7 +26,7 @@ public: protected: std::optional tryGenerate() override; - void onCancel() override; + void onCancel() noexcept override; private: MergeTreeSelectProcessorPtr processor; diff --git a/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp b/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp index 36ff6c0a4bd..9c82817e8cb 100644 --- a/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp +++ b/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp @@ -152,23 +152,15 @@ const ActionsDAG::Node & addFunction( const ActionsDAG::Node & addCast( const ActionsDAGPtr & dag, const ActionsDAG::Node & node_to_cast, - const String & type_name, + const DataTypePtr & to_type, OriginalToNewNodeMap & node_remap) { - if (node_to_cast.result_type->getName() == type_name) + if (!node_to_cast.result_type->equals(*to_type)) return node_to_cast; - Field cast_type_constant_value(type_name); - - ColumnWithTypeAndName column; - column.column = DataTypeString().createColumnConst(0, cast_type_constant_value); - column.type = std::make_shared(); - - const auto * cast_type_constant_node = &dag->addColumn(std::move(column)); - ActionsDAG::NodeRawConstPtrs children = {&node_to_cast, cast_type_constant_node}; - FunctionOverloadResolverPtr func_builder_cast = createInternalCastOverloadResolver(CastType::nonAccurate, {}); - - return addFunction(dag, func_builder_cast, std::move(children), node_remap); + const auto & new_node = dag->addCast(node_to_cast, to_type, {}); + node_remap[new_node.result_name] = {dag.get(), &new_node}; + return new_node; } /// Normalizes the filter node by adding AND with a constant true. @@ -332,7 +324,7 @@ bool tryBuildPrewhereSteps(PrewhereInfoPtr prewhere_info, const ExpressionAction /// Build AND(last_step_result_node, true) const auto & and_node = addAndTrue(last_step_dag, *last_step_result_node_info.node, node_remap); /// Build CAST(and_node, type of PREWHERE column) - const auto & cast_node = addCast(last_step_dag, and_node, output->result_type->getName(), node_remap); + const auto & cast_node = addCast(last_step_dag, and_node, output->result_type, node_remap); /// Add alias for the result with the name of the PREWHERE column const auto & prewhere_result_node = last_step_dag->addAlias(cast_node, output->result_name); last_step_dag->addOrReplaceInOutputs(prewhere_result_node); diff --git a/src/Storages/MergeTree/MergeTreeWhereOptimizer.cpp b/src/Storages/MergeTree/MergeTreeWhereOptimizer.cpp index 43c40dee77d..f0c26c302e1 100644 --- a/src/Storages/MergeTree/MergeTreeWhereOptimizer.cpp +++ b/src/Storages/MergeTree/MergeTreeWhereOptimizer.cpp @@ -222,17 +222,17 @@ static bool isConditionGood(const RPNBuilderTreeNode & condition, const NameSet /// check the value with respect to threshold if (type == Field::Types::UInt64) { - const auto value = output_value.get(); + const auto value = output_value.safeGet(); return value > threshold; } else if (type == Field::Types::Int64) { - const auto value = output_value.get(); + const auto value = output_value.safeGet(); return value < -threshold || threshold < value; } else if (type == Field::Types::Float64) { - const auto value = output_value.get(); + const auto value = output_value.safeGet(); return value < -threshold || threshold < value; } diff --git a/src/Storages/MergeTree/MutateFromLogEntryTask.cpp b/src/Storages/MergeTree/MutateFromLogEntryTask.cpp index 73084f487b9..56f68fd265a 100644 --- a/src/Storages/MergeTree/MutateFromLogEntryTask.cpp +++ b/src/Storages/MergeTree/MutateFromLogEntryTask.cpp @@ -254,6 +254,7 @@ bool MutateFromLogEntryTask::finalize(ReplicatedMergeMutateTaskBase::PartLogWrit LOG_ERROR(log, "{}. Data after mutation is not byte-identical to data on another replicas. " "We will download merged part from replica to force byte-identical result.", getCurrentExceptionMessage(false)); + mutate_task->updateProfileEvents(); write_part_log(ExecutionStatus::fromCurrentException("", true)); if (storage.getSettings()->detach_not_byte_identical_parts) @@ -281,6 +282,7 @@ bool MutateFromLogEntryTask::finalize(ReplicatedMergeMutateTaskBase::PartLogWrit */ finish_callback = [storage_ptr = &storage]() { storage_ptr->merge_selecting_task->schedule(); }; ProfileEvents::increment(ProfileEvents::ReplicatedPartMutations); + mutate_task->updateProfileEvents(); write_part_log({}); return true; diff --git a/src/Storages/MergeTree/MutatePlainMergeTreeTask.cpp b/src/Storages/MergeTree/MutatePlainMergeTreeTask.cpp index 9aec074deae..10461eb5942 100644 --- a/src/Storages/MergeTree/MutatePlainMergeTreeTask.cpp +++ b/src/Storages/MergeTree/MutatePlainMergeTreeTask.cpp @@ -102,6 +102,7 @@ bool MutatePlainMergeTreeTask::executeStep() transaction.commit(); storage.updateMutationEntriesErrors(future_part, true, ""); + mutate_task->updateProfileEvents(); write_part_log({}); state = State::NEED_FINISH; @@ -114,6 +115,7 @@ bool MutatePlainMergeTreeTask::executeStep() PreformattedMessage exception_message = getCurrentExceptionMessageAndPattern(/* with_stacktrace */ false); LOG_ERROR(getLogger("MutatePlainMergeTreeTask"), exception_message); storage.updateMutationEntriesErrors(future_part, false, exception_message.text); + mutate_task->updateProfileEvents(); write_part_log(ExecutionStatus::fromCurrentException("", true)); tryLogCurrentException(__PRETTY_FUNCTION__); return false; diff --git a/src/Storages/MergeTree/MutateTask.cpp b/src/Storages/MergeTree/MutateTask.cpp index 9a775db73e2..14c274d7f64 100644 --- a/src/Storages/MergeTree/MutateTask.cpp +++ b/src/Storages/MergeTree/MutateTask.cpp @@ -38,7 +38,13 @@ namespace ProfileEvents { -extern const Event MutateTaskProjectionsCalculationMicroseconds; + extern const Event MutationTotalParts; + extern const Event MutationUntouchedParts; + extern const Event MutationTotalMilliseconds; + extern const Event MutationExecuteMilliseconds; + extern const Event MutationAllPartColumns; + extern const Event MutationSomePartColumns; + extern const Event MutateTaskProjectionsCalculationMicroseconds; } namespace CurrentMetrics @@ -659,7 +665,7 @@ static NameSet collectFilesToSkip( const Block & updated_header, const std::set & indices_to_recalc, const String & mrk_extension, - const std::set & projections_to_recalc, + const std::set & projections_to_skip, const std::set & stats_to_recalc) { NameSet files_to_skip = source_part->getFileNamesWithoutChecksums(); @@ -684,7 +690,7 @@ static NameSet collectFilesToSkip( } } - for (const auto & projection : projections_to_recalc) + for (const auto & projection : projections_to_skip) files_to_skip.insert(projection->getDirectoryName()); for (const auto & stat : stats_to_recalc) @@ -1046,6 +1052,7 @@ struct MutationContext /// Whether we need to count lightweight delete rows in this mutation bool count_lightweight_deleted_rows; + UInt64 execute_elapsed_ns = 0; }; using MutationContextPtr = std::shared_ptr; @@ -1554,6 +1561,10 @@ private: removed_projections.insert(command.column_name); } + bool lightweight_delete_mode = ctx->updated_header.has(RowExistsColumn::name); + bool lightweight_delete_drop = lightweight_delete_mode + && ctx->data->getSettings()->lightweight_mutation_projection_mode == LightweightMutationProjectionMode::DROP; + const auto & projections = ctx->metadata_snapshot->getProjections(); for (const auto & projection : projections) { @@ -1561,10 +1572,11 @@ private: continue; bool need_recalculate = - ctx->materialized_projections.contains(projection.name) + (ctx->materialized_projections.contains(projection.name) || (!is_full_part_storage && ctx->source_part->hasProjection(projection.name) - && !ctx->source_part->hasBrokenProjection(projection.name)); + && !ctx->source_part->hasBrokenProjection(projection.name))) + && !lightweight_delete_drop; if (need_recalculate) { @@ -1572,7 +1584,7 @@ private: } else { - if (ctx->source_part->checksums.has(projection.getDirectoryName())) + if (!lightweight_delete_mode && ctx->source_part->checksums.has(projection.getDirectoryName())) entries_to_hardlink.insert(projection.getDirectoryName()); } } @@ -2017,6 +2029,9 @@ MutateTask::MutateTask( bool MutateTask::execute() { + Stopwatch watch; + SCOPE_EXIT({ ctx->execute_elapsed_ns += watch.elapsedNanoseconds(); }); + switch (state) { case State::NEED_PREPARE: @@ -2050,6 +2065,15 @@ bool MutateTask::execute() return false; } +void MutateTask::updateProfileEvents() const +{ + UInt64 total_elapsed_ms = (*ctx->mutate_entry)->watch.elapsedMilliseconds(); + UInt64 execute_elapsed_ms = ctx->execute_elapsed_ns / 1000000UL; + + ProfileEvents::increment(ProfileEvents::MutationTotalMilliseconds, total_elapsed_ms); + ProfileEvents::increment(ProfileEvents::MutationExecuteMilliseconds, execute_elapsed_ms); +} + static bool canSkipConversionToNullable(const MergeTreeDataPartPtr & part, const MutationCommand & command) { if (command.type != MutationCommand::READ_COLUMN) @@ -2112,6 +2136,7 @@ static bool canSkipMutationCommandForPart(const MergeTreeDataPartPtr & part, con bool MutateTask::prepare() { + ProfileEvents::increment(ProfileEvents::MutationTotalParts); MutationHelpers::checkOperationIsNotCanceled(*ctx->merges_blocker, ctx->mutate_entry); if (ctx->future_part->parts.size() != 1) @@ -2134,7 +2159,7 @@ bool MutateTask::prepare() ctx->commands_for_part.emplace_back(command); if (ctx->source_part->isStoredOnDisk() && !isStorageTouchedByMutations( - *ctx->data, ctx->source_part, ctx->metadata_snapshot, ctx->commands_for_part, context_for_reading)) + ctx->source_part, ctx->metadata_snapshot, ctx->commands_for_part, context_for_reading)) { NameSet files_to_copy_instead_of_hardlinks; auto settings_ptr = ctx->data->getSettings(); @@ -2174,6 +2199,7 @@ bool MutateTask::prepare() ctx->temporary_directory_lock = std::move(lock); } + ProfileEvents::increment(ProfileEvents::MutationUntouchedParts); promise.set_value(std::move(part)); return false; } @@ -2198,6 +2224,8 @@ bool MutateTask::prepare() ctx->stage_progress = std::make_unique(1.0); + bool lightweight_delete_mode = false; + if (!ctx->for_interpreter.empty()) { /// Always disable filtering in mutations: we want to read and write all rows because for updates we rewrite only some of the @@ -2215,6 +2243,21 @@ bool MutateTask::prepare() ctx->mutating_pipeline_builder = ctx->interpreter->execute(); ctx->updated_header = ctx->interpreter->getUpdatedHeader(); ctx->progress_callback = MergeProgressCallback((*ctx->mutate_entry)->ptr(), ctx->watch_prev_elapsed, *ctx->stage_progress); + + lightweight_delete_mode = ctx->updated_header.has(RowExistsColumn::name); + /// If under the condition of lightweight delete mode with rebuild option, add projections again here as we can only know + /// the condition as early as from here. + if (lightweight_delete_mode + && ctx->data->getSettings()->lightweight_mutation_projection_mode == LightweightMutationProjectionMode::REBUILD) + { + for (const auto & projection : ctx->metadata_snapshot->getProjections()) + { + if (!ctx->source_part->hasProjection(projection.name)) + continue; + + ctx->materialized_projections.insert(projection.name); + } + } } auto single_disk_volume = std::make_shared("volume_" + ctx->future_part->name, ctx->space_reservation->getDisk(), 0); @@ -2256,7 +2299,7 @@ bool MutateTask::prepare() if (ctx->mutating_pipeline_builder.initialized()) ctx->execute_ttl_type = MutationHelpers::shouldExecuteTTL(ctx->metadata_snapshot, ctx->interpreter->getColumnDependencies()); - if (ctx->data->getSettings()->exclude_deleted_rows_for_part_size_in_merge && ctx->updated_header.has(RowExistsColumn::name)) + if (ctx->data->getSettings()->exclude_deleted_rows_for_part_size_in_merge && lightweight_delete_mode) { /// This mutation contains lightweight delete and we need to count the deleted rows, /// Reset existing_rows_count of new data part to 0 and it will be updated while writing _row_exists column @@ -2283,6 +2326,7 @@ bool MutateTask::prepare() ctx->new_data_part->remove_tmp_policy = IMergeTreeDataPart::BlobsRemovalPolicyForTemporaryParts::REMOVE_BLOBS; task = std::make_unique(ctx); + ProfileEvents::increment(ProfileEvents::MutationAllPartColumns); } else /// TODO: check that we modify only non-key columns in this case. { @@ -2293,10 +2337,30 @@ bool MutateTask::prepare() ctx->context, ctx->materialized_indices); - ctx->projections_to_recalc = MutationHelpers::getProjectionsToRecalculate( - ctx->source_part, - ctx->metadata_snapshot, - ctx->materialized_projections); + auto lightweight_mutation_projection_mode = ctx->data->getSettings()->lightweight_mutation_projection_mode; + bool lightweight_delete_drops_projections = + lightweight_mutation_projection_mode == LightweightMutationProjectionMode::DROP + || lightweight_mutation_projection_mode == LightweightMutationProjectionMode::THROW; + + std::set projections_to_skip_container; + auto * projections_to_skip = &projections_to_skip_container; + + bool should_create_projections = !(lightweight_delete_mode && lightweight_delete_drops_projections); + /// Under lightweight delete mode, if option is drop, projections_to_recalc should be empty. + if (should_create_projections) + { + ctx->projections_to_recalc = MutationHelpers::getProjectionsToRecalculate( + ctx->source_part, + ctx->metadata_snapshot, + ctx->materialized_projections); + + projections_to_skip = &ctx->projections_to_recalc; + } + else + { + for (const auto & projection : ctx->metadata_snapshot->getProjections()) + projections_to_skip->insert(&projection); + } ctx->stats_to_recalc = MutationHelpers::getStatisticsToRecalculate(ctx->metadata_snapshot, ctx->materialized_statistics); @@ -2306,7 +2370,7 @@ bool MutateTask::prepare() ctx->updated_header, ctx->indices_to_recalc, ctx->mrk_extension, - ctx->projections_to_recalc, + *projections_to_skip, ctx->stats_to_recalc); ctx->files_to_rename = MutationHelpers::collectFilesForRenames( @@ -2322,6 +2386,7 @@ bool MutateTask::prepare() ctx->new_data_part->remove_tmp_policy = IMergeTreeDataPart::BlobsRemovalPolicyForTemporaryParts::ASK_KEEPER; task = std::make_unique(ctx); + ProfileEvents::increment(ProfileEvents::MutationSomePartColumns); } return true; diff --git a/src/Storages/MergeTree/MutateTask.h b/src/Storages/MergeTree/MutateTask.h index dc22b90f0e9..08427bff6d8 100644 --- a/src/Storages/MergeTree/MutateTask.h +++ b/src/Storages/MergeTree/MutateTask.h @@ -39,6 +39,7 @@ public: bool need_prefix_); bool execute(); + void updateProfileEvents() const; std::future getFuture() { diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp index 63e1c12a15f..2f0b71b977e 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp @@ -3,8 +3,8 @@ #include #include #include +#include #include -#include "Common/Exception.h" #include #include #include @@ -155,7 +155,18 @@ ReplicatedMergeTreeSinkImpl::ReplicatedMergeTreeSinkImpl( } template -ReplicatedMergeTreeSinkImpl::~ReplicatedMergeTreeSinkImpl() = default; +ReplicatedMergeTreeSinkImpl::~ReplicatedMergeTreeSinkImpl() +{ + if (!delayed_chunk) + return; + + for (auto & partition : delayed_chunk->partitions) + { + partition.temp_part.cancel(); + } + + delayed_chunk.reset(); +} template size_t ReplicatedMergeTreeSinkImpl::checkQuorumPrecondition(const ZooKeeperWithFaultInjectionPtr & zookeeper) @@ -1155,8 +1166,9 @@ void ReplicatedMergeTreeSinkImpl::onStart() template void ReplicatedMergeTreeSinkImpl::onFinish() { - const auto & settings = context->getSettingsRef(); + chassert(!isCancelled()); + const auto & settings = context->getSettingsRef(); ZooKeeperWithFaultInjectionPtr zookeeper = ZooKeeperWithFaultInjection::createInstance( settings.insert_keeper_fault_injection_probability, settings.insert_keeper_fault_injection_seed, diff --git a/src/Storages/MergeTree/VectorSimilarityCondition.cpp b/src/Storages/MergeTree/VectorSimilarityCondition.cpp new file mode 100644 index 00000000000..2e53b4ecb3a --- /dev/null +++ b/src/Storages/MergeTree/VectorSimilarityCondition.cpp @@ -0,0 +1,350 @@ +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; + extern const int INCORRECT_QUERY; +} + +namespace +{ + +template +void extractReferenceVectorFromLiteral(std::vector & reference_vector, Literal literal) +{ + Float64 float_element_of_reference_vector; + Int64 int_element_of_reference_vector; + + for (const auto & value : literal.value()) + { + if (value.tryGet(float_element_of_reference_vector)) + reference_vector.emplace_back(float_element_of_reference_vector); + else if (value.tryGet(int_element_of_reference_vector)) + reference_vector.emplace_back(static_cast(int_element_of_reference_vector)); + else + throw Exception(ErrorCodes::INCORRECT_QUERY, "Wrong type of elements in reference vector. Only float or int are supported."); + } +} + +VectorSimilarityCondition::Info::DistanceFunction stringToDistanceFunction(std::string_view distance_function) +{ + if (distance_function == "L2Distance") + return VectorSimilarityCondition::Info::DistanceFunction::L2; + else + return VectorSimilarityCondition::Info::DistanceFunction::Unknown; +} + +} + +VectorSimilarityCondition::VectorSimilarityCondition(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->getSettingsRef().max_limit_for_ann_queries) + , index_is_useful(checkQueryStructure(query_info)) +{} + +bool VectorSimilarityCondition::alwaysUnknownOrTrue(String distance_function) const +{ + if (!index_is_useful) + return true; /// query isn't supported + /// If query is supported, check if distance function of index is the same as distance function in query + return !(stringToDistanceFunction(distance_function) == query_information->distance_function); +} + +UInt64 VectorSimilarityCondition::getLimit() const +{ + if (index_is_useful && query_information.has_value()) + return query_information->limit; + throw Exception(ErrorCodes::LOGICAL_ERROR, "No LIMIT section in query, not supported"); +} + +std::vector VectorSimilarityCondition::getReferenceVector() const +{ + if (index_is_useful && query_information.has_value()) + return query_information->reference_vector; + throw Exception(ErrorCodes::LOGICAL_ERROR, "Reference vector was requested for useless or uninitialized index."); +} + +size_t VectorSimilarityCondition::getDimensions() const +{ + if (index_is_useful && query_information.has_value()) + return query_information->reference_vector.size(); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Number of dimensions was requested for useless or uninitialized index."); +} + +String VectorSimilarityCondition::getColumnName() const +{ + if (index_is_useful && query_information.has_value()) + return query_information->column_name; + throw Exception(ErrorCodes::LOGICAL_ERROR, "Column name was requested for useless or uninitialized index."); +} + +VectorSimilarityCondition::Info::DistanceFunction VectorSimilarityCondition::getDistanceFunction() const +{ + if (index_is_useful && query_information.has_value()) + return query_information->distance_function; + throw Exception(ErrorCodes::LOGICAL_ERROR, "Distance function was requested for useless or uninitialized index."); +} + +bool VectorSimilarityCondition::checkQueryStructure(const SelectQueryInfo & query) +{ + Info order_by_info; + + /// Build rpns for query sections + const auto & select = query.query->as(); + + RPN rpn_order_by; + RPNElement rpn_limit; + UInt64 limit; + + if (select.limitLength()) + traverseAtomAST(select.limitLength(), rpn_limit); + + if (select.orderBy()) + traverseOrderByAST(select.orderBy(), rpn_order_by); + + /// Reverse RPNs for conveniences during parsing + std::reverse(rpn_order_by.begin(), rpn_order_by.end()); + + const bool order_by_is_valid = matchRPNOrderBy(rpn_order_by, order_by_info); + const bool limit_is_valid = matchRPNLimit(rpn_limit, limit); + + if (!limit_is_valid || limit > max_limit_for_ann_queries) + return false; + + if (order_by_is_valid) + { + query_information = std::move(order_by_info); + query_information->limit = limit; + return true; + } + + return false; +} + +void VectorSimilarityCondition::traverseAST(const ASTPtr & node, RPN & rpn) +{ + /// If the node is ASTFunction, it may have children nodes + if (const auto * func = node->as()) + { + const ASTs & children = func->arguments->children; + /// Traverse children nodes + for (const auto& child : children) + traverseAST(child, rpn); + } + + RPNElement element; + /// Get the data behind node + if (!traverseAtomAST(node, element)) + element.function = RPNElement::FUNCTION_UNKNOWN; + + rpn.emplace_back(std::move(element)); +} + +bool VectorSimilarityCondition::traverseAtomAST(const ASTPtr & node, RPNElement & out) +{ + /// Match Functions + if (const auto * function = node->as()) + { + /// Set the name + out.func_name = function->name; + + if (function->name == "L1Distance" || + function->name == "L2Distance" || + function->name == "LinfDistance" || + function->name == "cosineDistance" || + function->name == "dotProduct") + out.function = RPNElement::FUNCTION_DISTANCE; + else if (function->name == "array") + out.function = RPNElement::FUNCTION_ARRAY; + else if (function->name == "_CAST") + out.function = RPNElement::FUNCTION_CAST; + else + return false; + + return true; + } + /// Match identifier + else if (const auto * identifier = node->as()) + { + out.function = RPNElement::FUNCTION_IDENTIFIER; + out.identifier.emplace(identifier->name()); + out.func_name = "column identifier"; + + return true; + } + + /// Check if we have constants behind the node + return tryCastToConstType(node, out); +} + +bool VectorSimilarityCondition::tryCastToConstType(const ASTPtr & node, RPNElement & out) +{ + Field const_value; + DataTypePtr const_type; + + if (KeyCondition::getConstant(node, block_with_constants, const_value, const_type)) + { + /// Check for constant types + if (const_value.getType() == Field::Types::Float64) + { + out.function = RPNElement::FUNCTION_FLOAT_LITERAL; + out.float_literal.emplace(const_value.safeGet()); + out.func_name = "Float literal"; + return true; + } + + if (const_value.getType() == Field::Types::UInt64) + { + out.function = RPNElement::FUNCTION_INT_LITERAL; + out.int_literal.emplace(const_value.safeGet()); + out.func_name = "Int literal"; + return true; + } + + if (const_value.getType() == Field::Types::Int64) + { + out.function = RPNElement::FUNCTION_INT_LITERAL; + out.int_literal.emplace(const_value.safeGet()); + out.func_name = "Int literal"; + return true; + } + + if (const_value.getType() == Field::Types::Array) + { + out.function = RPNElement::FUNCTION_LITERAL_ARRAY; + out.array_literal = const_value.safeGet(); + out.func_name = "Array literal"; + return true; + } + + if (const_value.getType() == Field::Types::String) + { + out.function = RPNElement::FUNCTION_STRING_LITERAL; + out.func_name = const_value.safeGet(); + return true; + } + } + + return false; +} + +void VectorSimilarityCondition::traverseOrderByAST(const ASTPtr & node, RPN & rpn) +{ + if (const auto * expr_list = node->as()) + if (const auto * order_by_element = expr_list->children.front()->as()) + traverseAST(order_by_element->children.front(), rpn); +} + +/// Returns true and stores ANNExpr if the query has valid ORDERBY clause +bool VectorSimilarityCondition::matchRPNOrderBy(RPN & rpn, Info & info) +{ + /// ORDER BY clause must have at least 3 expressions + if (rpn.size() < 3) + return false; + + auto iter = rpn.begin(); + auto end = rpn.end(); + + bool identifier_found = false; + + /// Matches DistanceFunc->[Column]->[ArrayFunc]->ReferenceVector(floats)->[Column] + if (iter->function != RPNElement::FUNCTION_DISTANCE) + return false; + + info.distance_function = stringToDistanceFunction(iter->func_name); + ++iter; + + if (iter->function == RPNElement::FUNCTION_IDENTIFIER) + { + identifier_found = true; + info.column_name = std::move(iter->identifier.value()); + ++iter; + } + + if (iter->function == RPNElement::FUNCTION_ARRAY) + ++iter; + + if (iter->function == RPNElement::FUNCTION_LITERAL_ARRAY) + { + extractReferenceVectorFromLiteral(info.reference_vector, iter->array_literal); + ++iter; + } + + /// further conditions are possible if there is no array, or no identifier is found + /// the array can be inside a cast function. For other cases, see the loop after this condition + if (iter != end && iter->function == RPNElement::FUNCTION_CAST) + { + ++iter; + /// Cast should be made to array + if (!iter->func_name.starts_with("Array")) + return false; + ++iter; + if (iter->function == RPNElement::FUNCTION_LITERAL_ARRAY) + { + extractReferenceVectorFromLiteral(info.reference_vector, iter->array_literal); + ++iter; + } + else + return false; + } + + while (iter != end) + { + if (iter->function == RPNElement::FUNCTION_FLOAT_LITERAL || + iter->function == RPNElement::FUNCTION_INT_LITERAL) + info.reference_vector.emplace_back(getFloatOrIntLiteralOrPanic(iter)); + else if (iter->function == RPNElement::FUNCTION_IDENTIFIER) + { + if (identifier_found) + return false; + info.column_name = std::move(iter->identifier.value()); + identifier_found = true; + } + else + return false; + + ++iter; + } + + /// Final checks of correctness + return identifier_found && !info.reference_vector.empty(); +} + +/// Returns true and stores Length if we have valid LIMIT clause in query +bool VectorSimilarityCondition::matchRPNLimit(RPNElement & rpn, UInt64 & limit) +{ + if (rpn.function == RPNElement::FUNCTION_INT_LITERAL) + { + limit = rpn.int_literal.value(); + return true; + } + + return false; +} + +/// Gets float or int from AST node +float VectorSimilarityCondition::getFloatOrIntLiteralOrPanic(const RPN::iterator& iter) +{ + if (iter->float_literal.has_value()) + return iter->float_literal.value(); + if (iter->int_literal.has_value()) + return static_cast(iter->int_literal.value()); + throw Exception(ErrorCodes::INCORRECT_QUERY, "Wrong parsed AST in buildRPN\n"); +} + +} diff --git a/src/Storages/MergeTree/ApproximateNearestNeighborIndexesCommon.h b/src/Storages/MergeTree/VectorSimilarityCondition.h similarity index 54% rename from src/Storages/MergeTree/ApproximateNearestNeighborIndexesCommon.h rename to src/Storages/MergeTree/VectorSimilarityCondition.h index 5da2a714b02..fd339ed715d 100644 --- a/src/Storages/MergeTree/ApproximateNearestNeighborIndexesCommon.h +++ b/src/Storages/MergeTree/VectorSimilarityCondition.h @@ -9,52 +9,9 @@ namespace DB { -static constexpr auto DISTANCE_FUNCTION_L2 = "L2Distance"; -static constexpr auto DISTANCE_FUNCTION_COSINE = "cosineDistance"; - -/// Approximate Nearest Neighbour queries have a similar structure: -/// - reference vector from which all distances are calculated -/// - metric name (e.g L2Distance, LpDistance, etc.) -/// - name of column with embeddings -/// - type of query -/// - maximum number of returned elements (LIMIT) -/// -/// And two optional parameters: -/// - p for LpDistance function -/// - distance to compare with (only for where queries) -/// -/// This struct holds all these components. -struct ApproximateNearestNeighborInformation -{ - using Embedding = std::vector; - Embedding reference_vector; - - enum class Metric : uint8_t - { - Unknown, - L2, - Lp - }; - Metric metric; - - String column_name; - UInt64 limit; - - enum class Type : uint8_t - { - OrderBy, - Where - }; - Type type; - - float p_for_lp_dist = -1.0; - float distance = -1.0; -}; - - -// Class ANNCondition, is responsible for recognizing if the query is an ANN queries which can utilize ANN indexes. It parses the SQL query -/// and checks if it matches ANNIndexes. Method alwaysUnknownOrTrue returns false if we can speed up the query, and true otherwise. It has -/// only one argument, the name of the metric with which index was built. Two main patterns of queries are supported +/// Class VectorSimilarityCondition is responsible for recognizing if the query can utilize vector similarity indexes. +/// Method alwaysUnknownOrTrue returns false if we can speed up the query, and true otherwise. It has +/// only one argument, the name of the distance function with which index was built. Two main patterns of queries are supported /// /// - 1. WHERE queries: /// SELECT * FROM * WHERE DistanceFunc(column, reference_vector) < floatLiteral LIMIT count @@ -64,14 +21,14 @@ struct ApproximateNearestNeighborInformation /// /// Queries without LIMIT count are not supported /// If the query is both of type 1. and 2., than we can't use the index and alwaysUnknownOrTrue returns true. -/// reference_vector should have float coordinates, e.g. (0.2, 0.1, .., 0.5) +/// reference_vector should have float coordinates, e.g. [0.2, 0.1, .., 0.5] /// -/// If the query matches one of these two types, then this class extracts the main information needed for ANN indexes from the query. +/// If the query matches one of these two types, then this class extracts the main information needed for vector similarity indexes from the +/// query. /// /// From matching query it extracts /// - referenceVector -/// - metricName(DistanceFunction) -/// - dimension size if query uses LpDistance +/// - distance function /// - distance to compare(ONLY for search types, otherwise you get exception) /// - spaceDimension(which is referenceVector's components count) /// - column @@ -79,35 +36,45 @@ struct ApproximateNearestNeighborInformation /// - queryHasOrderByClause and queryHasWhereClause return true if query matches the type /// /// Search query type is also recognized for PREWHERE clause -class ApproximateNearestNeighborCondition +class VectorSimilarityCondition { public: - ApproximateNearestNeighborCondition(const SelectQueryInfo & query_info, ContextPtr context); + VectorSimilarityCondition(const SelectQueryInfo & query_info, ContextPtr context); + + /// Approximate nearest neighbour (ANN) / vector similarity queries have a similar structure: + /// - reference vector from which all distances are calculated + /// - distance function, e.g L2Distance + /// - name of column with embeddings + /// - type of query + /// - maximum number of returned elements (LIMIT) + /// + /// And one optional parameter: + /// - distance to compare with (only for where queries) + /// + /// This struct holds all these components. + struct Info + { + enum class DistanceFunction : uint8_t + { + Unknown, + L2 + }; + + std::vector reference_vector; + DistanceFunction distance_function; + String column_name; + UInt64 limit; + float distance = -1.0; + }; /// Returns false if query can be speeded up by an ANN index, true otherwise. - bool alwaysUnknownOrTrue(String metric) const; + bool alwaysUnknownOrTrue(String distance_function) const; - /// Returns the distance to compare with for search query - float getComparisonDistanceForWhereQuery() const; - - /// Distance should be calculated regarding to referenceVector std::vector getReferenceVector() const; - - /// Reference vector's dimension count size_t getDimensions() const; - String getColumnName() const; - - ApproximateNearestNeighborInformation::Metric getMetricType() const; - - /// The P- value if the metric is 'LpDistance' - float getPValueForLpDistance() const; - - ApproximateNearestNeighborInformation::Type getQueryType() const; - + Info::DistanceFunction getDistanceFunction() const; UInt64 getIndexGranularity() const { return index_granularity; } - - /// Length's value from LIMIT clause UInt64 getLimit() const; private: @@ -118,9 +85,6 @@ private: /// DistanceFunctions FUNCTION_DISTANCE, - //tuple(0.1, ..., 0.1) - FUNCTION_TUPLE, - //array(0.1, ..., 0.1) FUNCTION_ARRAY, @@ -139,9 +103,6 @@ private: /// Unknown, can be any value FUNCTION_UNKNOWN, - /// (0.1, ...., 0.1) vector without word 'tuple' - FUNCTION_LITERAL_TUPLE, - /// [0.1, ...., 0.1] vector without word 'array' FUNCTION_LITERAL_ARRAY, @@ -154,19 +115,14 @@ private: explicit RPNElement(Function function_ = FUNCTION_UNKNOWN) : function(function_) - , func_name("Unknown") - , float_literal(std::nullopt) - , identifier(std::nullopt) {} Function function; - String func_name; + String func_name = "Unknown"; std::optional float_literal; std::optional identifier; std::optional int_literal; - - std::optional tuple_literal; std::optional array_literal; UInt32 dim = 0; @@ -186,16 +142,16 @@ private: void traverseOrderByAST(const ASTPtr & node, RPN & rpn); /// Returns true and stores ANNExpr if the query has valid WHERE section - static bool matchRPNWhere(RPN & rpn, ApproximateNearestNeighborInformation & ann_info); + static bool matchRPNWhere(RPN & rpn, Info & info); /// Returns true and stores ANNExpr if the query has valid ORDERBY section - static bool matchRPNOrderBy(RPN & rpn, ApproximateNearestNeighborInformation & ann_info); + static bool matchRPNOrderBy(RPN & rpn, Info & info); /// Returns true and stores Length if we have valid LIMIT clause in query static bool matchRPNLimit(RPNElement & rpn, UInt64 & limit); - /* Matches dist function, reference vector, column name */ - static bool matchMainParts(RPN::iterator & iter, const RPN::iterator & end, ApproximateNearestNeighborInformation & ann_info); + /// Matches dist function, reference vector, column name + static bool matchMainParts(RPN::iterator & iter, const RPN::iterator & end, Info & info); /// Gets float or int from AST node static float getFloatOrIntLiteralOrPanic(const RPN::iterator& iter); @@ -203,7 +159,7 @@ private: Block block_with_constants; /// true if we have one of two supported query types - std::optional query_information; + std::optional query_information; // Get from settings ANNIndex parameters const UInt64 index_granularity; @@ -214,13 +170,4 @@ private: bool index_is_useful = false; }; - -/// Common interface of ANN indexes. -class IMergeTreeIndexConditionApproximateNearestNeighbor : public IMergeTreeIndexCondition -{ -public: - /// Returns vector of indexes of ranges in granule which are useful for query. - virtual std::vector getUsefulRanges(MergeTreeIndexGranulePtr idx_granule) const = 0; -}; - } diff --git a/src/Storages/MergeTree/registerStorageMergeTree.cpp b/src/Storages/MergeTree/registerStorageMergeTree.cpp index 3f0603f6900..44548e33d46 100644 --- a/src/Storages/MergeTree/registerStorageMergeTree.cpp +++ b/src/Storages/MergeTree/registerStorageMergeTree.cpp @@ -34,6 +34,7 @@ namespace ErrorCodes extern const int UNKNOWN_STORAGE; extern const int NO_REPLICA_NAME_GIVEN; extern const int CANNOT_EXTRACT_TABLE_STRUCTURE; + extern const int SUPPORT_IS_DISABLED; } @@ -591,7 +592,7 @@ static StoragePtr create(const StorageFactory::Arguments & args) if (ast->value.getType() != Field::Types::String) throw Exception(ErrorCodes::BAD_ARGUMENTS, format_str, error_msg); - graphite_config_name = ast->value.get(); + graphite_config_name = ast->value.safeGet(); } else throw Exception(ErrorCodes::BAD_ARGUMENTS, format_str, error_msg); @@ -829,6 +830,18 @@ static StoragePtr create(const StorageFactory::Arguments & args) "Floating point partition key is not supported: {}", metadata.partition_key.column_names[i]); } + if (metadata.hasProjections() && args.mode == LoadingStrictnessLevel::CREATE) + { + /// Now let's handle the merge tree family. Note we only handle in the mode of CREATE due to backward compatibility. + /// Otherwise, it would fail to start in the case of existing projections with special mergetree. + if (merging_params.mode != MergeTreeData::MergingParams::Mode::Ordinary + && storage_settings->deduplicate_merge_projection_mode == DeduplicateMergeProjectionMode::THROW) + throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, + "Projection is fully supported in {}MergeTree with deduplicate_merge_projection_mode = throw. " + "Use 'drop' or 'rebuild' option of deduplicate_merge_projection_mode.", + merging_params.getModeName()); + } + if (arg_num != arg_cnt) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Wrong number of engine arguments."); diff --git a/src/Storages/MessageQueueSink.cpp b/src/Storages/MessageQueueSink.cpp index 36899011e33..8190a375b97 100644 --- a/src/Storages/MessageQueueSink.cpp +++ b/src/Storages/MessageQueueSink.cpp @@ -1,6 +1,7 @@ #include #include #include +#include #include namespace DB @@ -46,6 +47,8 @@ void MessageQueueSink::consume(Chunk & chunk) if (columns.empty()) return; + /// The formatter might hold pointers to buffer (e.g. if PeekableWriteBuffer is used), which means the formatter + /// needs to be reset after buffer might reallocate its memory. In this exact case after restarting the buffer. if (row_format) { size_t row = 0; @@ -60,12 +63,12 @@ void MessageQueueSink::consume(Chunk & chunk) row_format->writeRow(columns, row); } row_format->finalize(); - row_format->resetFormatter(); producer->produce(buffer->str(), i, columns, row - 1); /// Reallocate buffer if it's capacity is large then DBMS_DEFAULT_BUFFER_SIZE, /// because most likely in this case we serialized abnormally large row /// and won't need this large allocated buffer anymore. buffer->restart(DBMS_DEFAULT_BUFFER_SIZE); + row_format->resetFormatter(); } } else @@ -73,10 +76,21 @@ void MessageQueueSink::consume(Chunk & chunk) format->write(getHeader().cloneWithColumns(chunk.detachColumns())); format->finalize(); producer->produce(buffer->str(), chunk.getNumRows(), columns, chunk.getNumRows() - 1); - format->resetFormatter(); buffer->restart(); + format->resetFormatter(); } } +void MessageQueueSink::onCancel() noexcept +{ + try + { + onFinish(); + } + catch (...) + { + tryLogCurrentException(getLogger("MessageQueueSink"), "Error occurs on cancellation."); + } +} } diff --git a/src/Storages/MessageQueueSink.h b/src/Storages/MessageQueueSink.h index 4a9248c6c4d..cb6562185c0 100644 --- a/src/Storages/MessageQueueSink.h +++ b/src/Storages/MessageQueueSink.h @@ -39,7 +39,7 @@ public: void onStart() override; void onFinish() override; - void onCancel() override { onFinish(); } + void onCancel() noexcept override; void onException(std::exception_ptr /* exception */) override { onFinish(); } protected: diff --git a/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.cpp b/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.cpp index c896a760597..ffb987bfa82 100644 --- a/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.cpp +++ b/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.cpp @@ -332,6 +332,8 @@ struct DeltaLakeMetadataImpl WhichDataType which(check_type->getTypeId()); if (which.isStringOrFixedString()) return value; + else if (isBool(check_type)) + return parse(value); else if (which.isInt8()) return parse(value); else if (which.isUInt8()) diff --git a/src/Storages/ObjectStorage/DataLakes/IStorageDataLake.h b/src/Storages/ObjectStorage/DataLakes/IStorageDataLake.h index c8603fccb86..087207d3860 100644 --- a/src/Storages/ObjectStorage/DataLakes/IStorageDataLake.h +++ b/src/Storages/ObjectStorage/DataLakes/IStorageDataLake.h @@ -91,8 +91,9 @@ public: { ConfigurationPtr configuration = base_configuration->clone(); configuration->setPaths(metadata->getDataFiles()); + std::string sample_path; return Storage::resolveSchemaFromData( - object_storage_, configuration, format_settings_, local_context); + object_storage_, configuration, format_settings_, sample_path, local_context); } } diff --git a/src/Storages/ObjectStorage/ReadBufferIterator.cpp b/src/Storages/ObjectStorage/ReadBufferIterator.cpp index df78f128c80..fadf683fce7 100644 --- a/src/Storages/ObjectStorage/ReadBufferIterator.cpp +++ b/src/Storages/ObjectStorage/ReadBufferIterator.cpp @@ -132,7 +132,7 @@ void ReadBufferIterator::setFormatName(const String & format_name) format = format_name; } -String ReadBufferIterator::getLastFileName() const +String ReadBufferIterator::getLastFilePath() const { if (current_object_info) return current_object_info->getPath(); diff --git a/src/Storages/ObjectStorage/ReadBufferIterator.h b/src/Storages/ObjectStorage/ReadBufferIterator.h index 6eeb52ec2ed..b81aebb7b07 100644 --- a/src/Storages/ObjectStorage/ReadBufferIterator.h +++ b/src/Storages/ObjectStorage/ReadBufferIterator.h @@ -33,7 +33,7 @@ public: void setResultingSchema(const ColumnsDescription & columns) override; - String getLastFileName() const override; + String getLastFilePath() const override; void setFormatName(const String & format_name) override; diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.cpp b/src/Storages/ObjectStorage/StorageObjectStorage.cpp index fe675a17d54..d9c82d68791 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorage.cpp @@ -1,4 +1,5 @@ #include +#include #include #include @@ -33,6 +34,33 @@ namespace ErrorCodes extern const int LOGICAL_ERROR; } +String StorageObjectStorage::getPathSample(StorageInMemoryMetadata metadata, ContextPtr context) +{ + auto query_settings = configuration->getQuerySettings(context); + /// We don't want to throw an exception if there are no files with specified path. + query_settings.throw_on_zero_files_match = false; + + bool local_distributed_processing = distributed_processing; + if (context->getSettingsRef().use_hive_partitioning) + local_distributed_processing = false; + + auto file_iterator = StorageObjectStorageSource::createFileIterator( + configuration, + query_settings, + object_storage, + local_distributed_processing, + context, + {}, // predicate + metadata.getColumns().getAll(), // virtual_columns + nullptr, // read_keys + {} // file_progress_callback + ); + + if (auto file = file_iterator->next(0)) + return file->getPath(); + return ""; +} + StorageObjectStorage::StorageObjectStorage( ConfigurationPtr configuration_, ObjectStoragePtr object_storage_, @@ -53,7 +81,9 @@ StorageObjectStorage::StorageObjectStorage( , log(getLogger(fmt::format("Storage{}({})", configuration->getEngineName(), table_id_.getFullTableName()))) { ColumnsDescription columns{columns_}; - resolveSchemaAndFormat(columns, configuration->format, object_storage, configuration, format_settings, context); + + std::string sample_path; + resolveSchemaAndFormat(columns, configuration->format, object_storage, configuration, format_settings, sample_path, context); configuration->check(context); StorageInMemoryMetadata metadata; @@ -61,7 +91,10 @@ StorageObjectStorage::StorageObjectStorage( metadata.setConstraints(constraints_); metadata.setComment(comment); - setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(metadata.getColumns())); + if (sample_path.empty() && context->getSettingsRef().use_hive_partitioning) + sample_path = getPathSample(metadata, context); + + setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(metadata.getColumns(), context, sample_path, format_settings)); setInMemoryMetadata(metadata); } @@ -198,7 +231,7 @@ private: return; auto context = getContext(); iterator_wrapper = StorageObjectStorageSource::createFileIterator( - configuration, object_storage, distributed_processing, + configuration, configuration->getQuerySettings(context), object_storage, distributed_processing, context, predicate, virtual_columns, nullptr, context->getFileProgressCallback()); } }; @@ -350,6 +383,7 @@ std::unique_ptr StorageObjectStorage::createReadBufferIterat { auto file_iterator = StorageObjectStorageSource::createFileIterator( configuration, + configuration->getQuerySettings(context), object_storage, false/* distributed_processing */, context, @@ -366,33 +400,41 @@ ColumnsDescription StorageObjectStorage::resolveSchemaFromData( const ObjectStoragePtr & object_storage, const ConfigurationPtr & configuration, const std::optional & format_settings, + std::string & sample_path, const ContextPtr & context) { ObjectInfos read_keys; auto iterator = createReadBufferIterator(object_storage, configuration, format_settings, read_keys, context); - return readSchemaFromFormat(configuration->format, format_settings, *iterator, context); + auto schema = readSchemaFromFormat(configuration->format, format_settings, *iterator, context); + sample_path = iterator->getLastFilePath(); + return schema; } std::string StorageObjectStorage::resolveFormatFromData( const ObjectStoragePtr & object_storage, const ConfigurationPtr & configuration, const std::optional & format_settings, + std::string & sample_path, const ContextPtr & context) { ObjectInfos read_keys; auto iterator = createReadBufferIterator(object_storage, configuration, format_settings, read_keys, context); - return detectFormatAndReadSchema(format_settings, *iterator, context).second; + auto format_and_schema = detectFormatAndReadSchema(format_settings, *iterator, context).second; + sample_path = iterator->getLastFilePath(); + return format_and_schema; } std::pair StorageObjectStorage::resolveSchemaAndFormatFromData( const ObjectStoragePtr & object_storage, const ConfigurationPtr & configuration, const std::optional & format_settings, + std::string & sample_path, const ContextPtr & context) { ObjectInfos read_keys; auto iterator = createReadBufferIterator(object_storage, configuration, format_settings, read_keys, context); auto [columns, format] = detectFormatAndReadSchema(format_settings, *iterator, context); + sample_path = iterator->getLastFilePath(); configuration->format = format; return std::pair(columns, format); } diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.h b/src/Storages/ObjectStorage/StorageObjectStorage.h index 818ce055c77..cae0db48f31 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.h +++ b/src/Storages/ObjectStorage/StorageObjectStorage.h @@ -102,23 +102,28 @@ public: const ObjectStoragePtr & object_storage, const ConfigurationPtr & configuration, const std::optional & format_settings, + std::string & sample_path, const ContextPtr & context); static std::string resolveFormatFromData( const ObjectStoragePtr & object_storage, const ConfigurationPtr & configuration, const std::optional & format_settings, + std::string & sample_path, const ContextPtr & context); static std::pair resolveSchemaAndFormatFromData( const ObjectStoragePtr & object_storage, const ConfigurationPtr & configuration, const std::optional & format_settings, + std::string & sample_path, const ContextPtr & context); protected: virtual void updateConfiguration(ContextPtr local_context); + String getPathSample(StorageInMemoryMetadata metadata, ContextPtr context); + virtual ReadFromFormatInfo prepareReadingFromFormat( const Strings & requested_columns, const StorageSnapshotPtr & storage_snapshot, diff --git a/src/Storages/ObjectStorage/StorageObjectStorageCluster.cpp b/src/Storages/ObjectStorage/StorageObjectStorageCluster.cpp index 78f568d8ae2..c214665f7e0 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorageCluster.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorageCluster.cpp @@ -1,6 +1,8 @@ #include "Storages/ObjectStorage/StorageObjectStorageCluster.h" #include +#include +#include #include #include #include @@ -19,6 +21,28 @@ namespace ErrorCodes extern const int LOGICAL_ERROR; } +String StorageObjectStorageCluster::getPathSample(StorageInMemoryMetadata metadata, ContextPtr context) +{ + auto query_settings = configuration->getQuerySettings(context); + /// We don't want to throw an exception if there are no files with specified path. + query_settings.throw_on_zero_files_match = false; + auto file_iterator = StorageObjectStorageSource::createFileIterator( + configuration, + query_settings, + object_storage, + false, // distributed_processing + context, + {}, // predicate + metadata.getColumns().getAll(), // virtual_columns + nullptr, // read_keys + {} // file_progress_callback + ); + + if (auto file = file_iterator->next(0)) + return file->getPath(); + return ""; +} + StorageObjectStorageCluster::StorageObjectStorageCluster( const String & cluster_name_, ConfigurationPtr configuration_, @@ -33,14 +57,18 @@ StorageObjectStorageCluster::StorageObjectStorageCluster( , object_storage(object_storage_) { ColumnsDescription columns{columns_}; - resolveSchemaAndFormat(columns, configuration->format, object_storage, configuration, {}, context_); + std::string sample_path; + resolveSchemaAndFormat(columns, configuration->format, object_storage, configuration, {}, sample_path, context_); configuration->check(context_); StorageInMemoryMetadata metadata; metadata.setColumns(columns); metadata.setConstraints(constraints_); - setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(metadata.getColumns())); + if (sample_path.empty() && context_->getSettingsRef().use_hive_partitioning) + sample_path = getPathSample(metadata, context_); + + setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(metadata.getColumns(), context_, sample_path)); setInMemoryMetadata(metadata); } @@ -83,8 +111,8 @@ RemoteQueryExecutor::Extension StorageObjectStorageCluster::getTaskIteratorExten const ActionsDAG::Node * predicate, const ContextPtr & local_context) const { auto iterator = StorageObjectStorageSource::createFileIterator( - configuration, object_storage, /* distributed_processing */false, local_context, - predicate, virtual_columns, nullptr, local_context->getFileProgressCallback()); + configuration, configuration->getQuerySettings(local_context), object_storage, /* distributed_processing */false, + local_context, predicate, virtual_columns, nullptr, local_context->getFileProgressCallback()); auto callback = std::make_shared>([iterator]() mutable -> String { diff --git a/src/Storages/ObjectStorage/StorageObjectStorageCluster.h b/src/Storages/ObjectStorage/StorageObjectStorageCluster.h index 108aa109616..0088ff28fc2 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorageCluster.h +++ b/src/Storages/ObjectStorage/StorageObjectStorageCluster.h @@ -27,6 +27,8 @@ public: RemoteQueryExecutor::Extension getTaskIteratorExtension( const ActionsDAG::Node * predicate, const ContextPtr & context) const override; + String getPathSample(StorageInMemoryMetadata metadata, ContextPtr context); + private: void updateQueryToSendIfNeeded( ASTPtr & query, diff --git a/src/Storages/ObjectStorage/StorageObjectStorageSink.cpp b/src/Storages/ObjectStorage/StorageObjectStorageSink.cpp index c864fb9a7e2..33c96fdfae9 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorageSink.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorageSink.cpp @@ -4,6 +4,7 @@ #include #include #include +#include namespace DB { @@ -42,31 +43,16 @@ StorageObjectStorageSink::StorageObjectStorageSink( void StorageObjectStorageSink::consume(Chunk & chunk) { - std::lock_guard lock(cancel_mutex); - if (cancelled) + if (isCancelled()) return; writer->write(getHeader().cloneWithColumns(chunk.getColumns())); } -void StorageObjectStorageSink::onCancel() -{ - std::lock_guard lock(cancel_mutex); - cancelBuffers(); - releaseBuffers(); - cancelled = true; -} - -void StorageObjectStorageSink::onException(std::exception_ptr) -{ - std::lock_guard lock(cancel_mutex); - cancelBuffers(); - releaseBuffers(); -} - void StorageObjectStorageSink::onFinish() { - std::lock_guard lock(cancel_mutex); + chassert(!isCancelled()); finalizeBuffers(); + releaseBuffers(); } void StorageObjectStorageSink::finalizeBuffers() @@ -120,6 +106,12 @@ PartitionedStorageObjectStorageSink::PartitionedStorageObjectStorageSink( { } +StorageObjectStorageSink::~StorageObjectStorageSink() +{ + if (isCancelled()) + cancelBuffers(); +} + SinkPtr PartitionedStorageObjectStorageSink::createSinkForPartition(const String & partition_id) { auto partition_bucket = replaceWildcards(configuration->getNamespace(), partition_id); diff --git a/src/Storages/ObjectStorage/StorageObjectStorageSink.h b/src/Storages/ObjectStorage/StorageObjectStorageSink.h index 6ab531bb21a..97fd3d9b417 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorageSink.h +++ b/src/Storages/ObjectStorage/StorageObjectStorageSink.h @@ -18,22 +18,18 @@ public: ContextPtr context, const std::string & blob_path = ""); + ~StorageObjectStorageSink() override; + String getName() const override { return "StorageObjectStorageSink"; } void consume(Chunk & chunk) override; - void onCancel() override; - - void onException(std::exception_ptr exception) override; - void onFinish() override; private: const Block sample_block; std::unique_ptr write_buf; OutputFormatPtr writer; - bool cancelled = false; - std::mutex cancel_mutex; void finalizeBuffers(); void releaseBuffers(); diff --git a/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp b/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp index e760098f10f..d8e26977e75 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp @@ -99,6 +99,7 @@ std::string StorageObjectStorageSource::getUniqueStoragePathIdentifier( std::shared_ptr StorageObjectStorageSource::createFileIterator( ConfigurationPtr configuration, + const StorageObjectStorage::QuerySettings & query_settings, ObjectStoragePtr object_storage, bool distributed_processing, const ContextPtr & local_context, @@ -116,7 +117,6 @@ std::shared_ptr StorageObjectStorageSourc throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expression can not have wildcards inside {} name", configuration->getNamespaceType()); - auto settings = configuration->getQuerySettings(local_context); const bool is_archive = configuration->isArchive(); std::unique_ptr iterator; @@ -125,8 +125,8 @@ std::shared_ptr StorageObjectStorageSourc /// Iterate through disclosed globs and make a source for each file iterator = std::make_unique( object_storage, configuration, predicate, virtual_columns, - local_context, is_archive ? nullptr : read_keys, settings.list_object_keys_size, - settings.throw_on_zero_files_match, file_progress_callback); + local_context, is_archive ? nullptr : read_keys, query_settings.list_object_keys_size, + query_settings.throw_on_zero_files_match, file_progress_callback); } else { @@ -148,7 +148,7 @@ std::shared_ptr StorageObjectStorageSourc iterator = std::make_unique( object_storage, copy_configuration, virtual_columns, is_archive ? nullptr : read_keys, - settings.ignore_non_existent_file, file_progress_callback); + query_settings.ignore_non_existent_file, file_progress_callback); } if (is_archive) @@ -198,13 +198,17 @@ Chunk StorageObjectStorageSource::generate() const auto & object_info = reader.getObjectInfo(); const auto & filename = object_info->getFileName(); chassert(object_info->metadata); + VirtualColumnUtils::addRequestedFileLikeStorageVirtualsToChunk( 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}); + { + .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, + .etag = &(object_info->metadata->etag) + }, getContext(), read_from_format_info.columns_description); const auto & partition_columns = configuration->getPartitionColumns(); if (!partition_columns.empty() && chunk_size && chunk.hasColumns()) diff --git a/src/Storages/ObjectStorage/StorageObjectStorageSource.h b/src/Storages/ObjectStorage/StorageObjectStorageSource.h index e466621e1e1..6681dbf4578 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorageSource.h +++ b/src/Storages/ObjectStorage/StorageObjectStorageSource.h @@ -52,6 +52,7 @@ public: static std::shared_ptr createFileIterator( ConfigurationPtr configuration, + const StorageObjectStorage::QuerySettings & query_settings, ObjectStoragePtr object_storage, bool distributed_processing, const ContextPtr & local_context, diff --git a/src/Storages/ObjectStorage/Utils.cpp b/src/Storages/ObjectStorage/Utils.cpp index e49e14d2a0c..73410d959e0 100644 --- a/src/Storages/ObjectStorage/Utils.cpp +++ b/src/Storages/ObjectStorage/Utils.cpp @@ -49,19 +49,20 @@ void resolveSchemaAndFormat( ObjectStoragePtr object_storage, const StorageObjectStorage::ConfigurationPtr & configuration, std::optional format_settings, + std::string & sample_path, const ContextPtr & context) { if (columns.empty()) { if (format == "auto") std::tie(columns, format) = - StorageObjectStorage::resolveSchemaAndFormatFromData(object_storage, configuration, format_settings, context); + StorageObjectStorage::resolveSchemaAndFormatFromData(object_storage, configuration, format_settings, sample_path, context); else - columns = StorageObjectStorage::resolveSchemaFromData(object_storage, configuration, format_settings, context); + columns = StorageObjectStorage::resolveSchemaFromData(object_storage, configuration, format_settings, sample_path, context); } else if (format == "auto") { - format = StorageObjectStorage::resolveFormatFromData(object_storage, configuration, format_settings, context); + format = StorageObjectStorage::resolveFormatFromData(object_storage, configuration, format_settings, sample_path, context); } if (!columns.hasOnlyOrdinary()) diff --git a/src/Storages/ObjectStorage/Utils.h b/src/Storages/ObjectStorage/Utils.h index 2077999df41..7ee14f50979 100644 --- a/src/Storages/ObjectStorage/Utils.h +++ b/src/Storages/ObjectStorage/Utils.h @@ -19,6 +19,7 @@ void resolveSchemaAndFormat( ObjectStoragePtr object_storage, const StorageObjectStorage::ConfigurationPtr & configuration, std::optional format_settings, + std::string & sample_path, const ContextPtr & context); } diff --git a/src/Storages/ObjectStorageQueue/ObjectStorageQueueSource.cpp b/src/Storages/ObjectStorageQueue/ObjectStorageQueueSource.cpp index 4d921003e04..2634a7b2f1e 100644 --- a/src/Storages/ObjectStorageQueue/ObjectStorageQueueSource.cpp +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueSource.cpp @@ -524,7 +524,7 @@ Chunk ObjectStorageQueueSource::generateImpl() { .path = path, .size = reader.getObjectInfo()->metadata->size_bytes - }); + }, getContext(), read_from_format_info.columns_description); return chunk; } diff --git a/src/Storages/ObjectStorageQueue/StorageObjectStorageQueue.cpp b/src/Storages/ObjectStorageQueue/StorageObjectStorageQueue.cpp index f51a7a913b8..5dc3e01962c 100644 --- a/src/Storages/ObjectStorageQueue/StorageObjectStorageQueue.cpp +++ b/src/Storages/ObjectStorageQueue/StorageObjectStorageQueue.cpp @@ -161,14 +161,15 @@ StorageObjectStorageQueue::StorageObjectStorageQueue( configuration->check(context_); ColumnsDescription columns{columns_}; - resolveSchemaAndFormat(columns, configuration->format, object_storage, configuration, format_settings, context_); + std::string sample_path; + resolveSchemaAndFormat(columns, configuration->format, object_storage, configuration, format_settings, sample_path, context_); configuration->check(context_); StorageInMemoryMetadata storage_metadata; storage_metadata.setColumns(columns); storage_metadata.setConstraints(constraints_); storage_metadata.setComment(comment); - setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(storage_metadata.getColumns())); + setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(storage_metadata.getColumns(), context_)); setInMemoryMetadata(storage_metadata); LOG_INFO(log, "Using zookeeper path: {}", zk_path.string()); diff --git a/src/Storages/PostgreSQL/PostgreSQLReplicationHandler.cpp b/src/Storages/PostgreSQL/PostgreSQLReplicationHandler.cpp index f632e553a0d..01f78673ed8 100644 --- a/src/Storages/PostgreSQL/PostgreSQLReplicationHandler.cpp +++ b/src/Storages/PostgreSQL/PostgreSQLReplicationHandler.cpp @@ -659,7 +659,7 @@ void PostgreSQLReplicationHandler::dropReplicationSlot(pqxx::nontransaction & tx void PostgreSQLReplicationHandler::dropPublication(pqxx::nontransaction & tx) { - std::string query_str = fmt::format("DROP PUBLICATION IF EXISTS {}", publication_name); + std::string query_str = fmt::format("DROP PUBLICATION IF EXISTS {}", doubleQuoteString(publication_name)); tx.exec(query_str); LOG_DEBUG(log, "Dropped publication: {}", publication_name); } @@ -667,7 +667,7 @@ void PostgreSQLReplicationHandler::dropPublication(pqxx::nontransaction & tx) void PostgreSQLReplicationHandler::addTableToPublication(pqxx::nontransaction & ntx, const String & table_name) { - std::string query_str = fmt::format("ALTER PUBLICATION {} ADD TABLE ONLY {}", publication_name, doubleQuoteWithSchema(table_name)); + std::string query_str = fmt::format("ALTER PUBLICATION {} ADD TABLE ONLY {}", doubleQuoteString(publication_name), doubleQuoteWithSchema(table_name)); ntx.exec(query_str); LOG_TRACE(log, "Added table {} to publication `{}`", doubleQuoteWithSchema(table_name), publication_name); } diff --git a/src/Storages/RocksDB/StorageEmbeddedRocksDB.cpp b/src/Storages/RocksDB/StorageEmbeddedRocksDB.cpp index fafc72da04e..50f6266cb2f 100644 --- a/src/Storages/RocksDB/StorageEmbeddedRocksDB.cpp +++ b/src/Storages/RocksDB/StorageEmbeddedRocksDB.cpp @@ -36,10 +36,12 @@ #include #include +#include +#include #include #include +#include #include -#include #include #include @@ -428,6 +430,7 @@ void StorageEmbeddedRocksDB::initDB() 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; @@ -439,7 +442,7 @@ void StorageEmbeddedRocksDB::initDB() if (config.has("rocksdb.options")) { auto config_options = getOptionsFromConfig(config, "rocksdb.options"); - status = rocksdb::GetDBOptionsFromMap(merged, config_options, &merged); + status = rocksdb::GetDBOptionsFromMap({}, merged, config_options, &merged); if (!status.ok()) { throw Exception(ErrorCodes::ROCKSDB_ERROR, "Fail to merge rocksdb options from 'rocksdb.options' at: {}: {}", @@ -449,7 +452,7 @@ void StorageEmbeddedRocksDB::initDB() 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); + 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: {}: {}", @@ -459,7 +462,7 @@ void StorageEmbeddedRocksDB::initDB() 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); + 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: {}: {}", @@ -484,7 +487,7 @@ void StorageEmbeddedRocksDB::initDB() if (config.has(config_key)) { auto table_config_options = getOptionsFromConfig(config, config_key); - status = rocksdb::GetDBOptionsFromMap(merged, table_config_options, &merged); + status = rocksdb::GetDBOptionsFromMap({}, merged, table_config_options, &merged); if (!status.ok()) { throw Exception(ErrorCodes::ROCKSDB_ERROR, "Fail to merge rocksdb options from '{}' at: {}: {}", @@ -496,7 +499,7 @@ void StorageEmbeddedRocksDB::initDB() if (config.has(config_key)) { auto table_column_family_options = getOptionsFromConfig(config, config_key); - status = rocksdb::GetColumnFamilyOptionsFromMap(merged, table_column_family_options, &merged); + status = rocksdb::GetColumnFamilyOptionsFromMap({}, merged, table_column_family_options, &merged); if (!status.ok()) { throw Exception(ErrorCodes::ROCKSDB_ERROR, "Fail to merge rocksdb options from '{}' at: {}: {}", @@ -508,7 +511,7 @@ void StorageEmbeddedRocksDB::initDB() if (config.has(config_key)) { auto block_based_table_options = getOptionsFromConfig(config, config_key); - status = rocksdb::GetBlockBasedTableOptionsFromMap(table_options, block_based_table_options, &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 '{}' at: {}: {}", diff --git a/src/Storages/RocksDB/StorageSystemRocksDB.cpp b/src/Storages/RocksDB/StorageSystemRocksDB.cpp index 744d43e1c77..ea5852de830 100644 --- a/src/Storages/RocksDB/StorageSystemRocksDB.cpp +++ b/src/Storages/RocksDB/StorageSystemRocksDB.cpp @@ -41,6 +41,14 @@ ColumnsDescription StorageSystemRocksDB::getColumnsDescription() } +Block StorageSystemRocksDB::getFilterSampleBlock() const +{ + return { + { {}, std::make_shared(), "database" }, + { {}, std::make_shared(), "table" }, + }; +} + void StorageSystemRocksDB::fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node * predicate, std::vector) const { const auto access = context->getAccess(); diff --git a/src/Storages/RocksDB/StorageSystemRocksDB.h b/src/Storages/RocksDB/StorageSystemRocksDB.h index ec351c75446..be3bfaa860c 100644 --- a/src/Storages/RocksDB/StorageSystemRocksDB.h +++ b/src/Storages/RocksDB/StorageSystemRocksDB.h @@ -22,6 +22,7 @@ protected: using IStorageSystemOneBlock::IStorageSystemOneBlock; void fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node * predicate, std::vector) const override; + Block getFilterSampleBlock() const override; }; } diff --git a/src/Storages/Statistics/ConditionSelectivityEstimator.cpp b/src/Storages/Statistics/ConditionSelectivityEstimator.cpp index 57dff958b9a..432659f51f8 100644 --- a/src/Storages/Statistics/ConditionSelectivityEstimator.cpp +++ b/src/Storages/Statistics/ConditionSelectivityEstimator.cpp @@ -19,7 +19,7 @@ void ConditionSelectivityEstimator::ColumnSelectivityEstimator::merge(String par Float64 ConditionSelectivityEstimator::ColumnSelectivityEstimator::estimateLess(const Field & val, Float64 rows) const { if (part_statistics.empty()) - return default_normal_cond_factor * rows; + return default_cond_range_factor * rows; Float64 result = 0; Float64 part_rows = 0; for (const auto & [key, estimator] : part_statistics) @@ -39,13 +39,7 @@ Float64 ConditionSelectivityEstimator::ColumnSelectivityEstimator::estimateEqual { if (part_statistics.empty()) { - 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; + return default_cond_equal_factor * rows; } Float64 result = 0; Float64 partial_cnt = 0; @@ -149,30 +143,22 @@ Float64 ConditionSelectivityEstimator::estimateRowCount(const RPNBuilderTreeNode auto [op, val] = extractBinaryOp(node, col); + if (dummy) + { + if (op == "equals") + return default_cond_equal_factor * total_rows; + else if (op == "less" || op == "lessOrEquals" || op == "greater" || op == "greaterOrEquals") + return default_cond_range_factor * total_rows; + else + return default_unknown_cond_factor * total_rows; + } + if (op == "equals") - { - if (dummy) - { - 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; - } return estimator.estimateEqual(val, total_rows); - } else if (op == "less" || op == "lessOrEquals") - { - if (dummy) - return default_normal_cond_factor * total_rows; return estimator.estimateLess(val, total_rows); - } else if (op == "greater" || op == "greaterOrEquals") - { - if (dummy) - return default_normal_cond_factor * total_rows; return estimator.estimateGreater(val, total_rows); - } else return default_unknown_cond_factor * total_rows; } diff --git a/src/Storages/Statistics/ConditionSelectivityEstimator.h b/src/Storages/Statistics/ConditionSelectivityEstimator.h index ce7fdd12e92..269ee9ac6cb 100644 --- a/src/Storages/Statistics/ConditionSelectivityEstimator.h +++ b/src/Storages/Statistics/ConditionSelectivityEstimator.h @@ -38,12 +38,10 @@ private: 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; - /// Conditions like "x = N" are considered good if abs(N) > threshold. - /// This is used to assume that condition is likely to have good selectivity. - static constexpr auto threshold = 2; + /// Used to estimate the selectivity of a condition when there is no statistics. + static constexpr auto default_cond_range_factor = 0.5; + static constexpr auto default_cond_equal_factor = 0.01; + static constexpr auto default_unknown_cond_factor = 1; UInt64 total_rows = 0; std::map column_estimators; diff --git a/src/Storages/Statistics/Statistics.cpp b/src/Storages/Statistics/Statistics.cpp index ade3326288a..fd686c5f0aa 100644 --- a/src/Storages/Statistics/Statistics.cpp +++ b/src/Storages/Statistics/Statistics.cpp @@ -1,14 +1,17 @@ #include +#include +#include +#include +#include #include #include +#include #include #include #include #include #include #include -#include -#include #include "config.h" /// USE_DATASKETCHES @@ -27,33 +30,26 @@ enum StatisticsFileVersion : UInt16 V0 = 0, }; -std::optional StatisticsUtils::tryConvertToFloat64(const Field & field) +std::optional StatisticsUtils::tryConvertToFloat64(const Field & value, const DataTypePtr & data_type) { - switch (field.getType()) + if (data_type->isValueRepresentedByNumber()) { - 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 {}; - } -} + Field value_converted; -std::optional StatisticsUtils::tryConvertToString(const DB::Field & field) -{ - if (field.getType() == Field::Types::String) - return field.get(); + if (isInteger(data_type) && (value.getType() == Field::Types::Float64 || value.getType() == Field::Types::String)) + /// For case val_int32 < 10.5 or val_int32 < '10.5' we should convert 10.5 to Float64. + value_converted = convertFieldToType(value, *DataTypeFactory::instance().get("Float64")); + else + /// We should convert value to the real column data type and then translate it to Float64. + /// For example for expression col_date > '2024-08-07', if we directly convert '2024-08-07' to Float64, we will get null. + value_converted = convertFieldToType(value, *data_type); + + if (value_converted.isNull()) + return {}; + + Float64 value_as_float = applyVisitor(FieldVisitorConvertToNumber(), value_converted); + return value_as_float; + } return {}; } @@ -89,21 +85,23 @@ 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. +/// Notes: +/// - Statistics object usually only support estimation for certain types of predicates, e.g. +/// - TDigest: '< X' (less-than predicates) +/// - Count-min sketches: '= X' (equal predicates) +/// - Uniq (HyperLogLog): 'count distinct(*)' (column cardinality) +/// +/// If multiple statistics objects in a column support estimating a predicate, we want to try statistics in order of descending accuracy +/// (e.g. MinMax statistics are simpler than TDigest statistics and thus worse for estimating 'less' predicates). +/// +/// Sometimes, it is possible to combine multiple statistics 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 stats.at(StatisticsType::TDigest)->estimateLess(val); - return rows * ConditionSelectivityEstimator::default_normal_cond_factor; + return rows * ConditionSelectivityEstimator::default_cond_range_factor; } Float64 ColumnStatistics::estimateGreater(const Field & val) const @@ -113,8 +111,7 @@ Float64 ColumnStatistics::estimateGreater(const Field & val) const Float64 ColumnStatistics::estimateEqual(const Field & val) const { - auto float_val = StatisticsUtils::tryConvertToFloat64(val); - if (float_val.has_value() && stats.contains(StatisticsType::Uniq) && stats.contains(StatisticsType::TDigest)) + if (stats_desc.data_type->isValueRepresentedByNumber() && stats.contains(StatisticsType::Uniq) && stats.contains(StatisticsType::TDigest)) { /// 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) @@ -124,10 +121,7 @@ Float64 ColumnStatistics::estimateEqual(const Field & val) const 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; + return rows * ConditionSelectivityEstimator::default_cond_equal_factor; } /// ------------------------------------- @@ -204,15 +198,15 @@ void MergeTreeStatisticsFactory::registerValidator(StatisticsType stats_type, Va MergeTreeStatisticsFactory::MergeTreeStatisticsFactory() { - registerValidator(StatisticsType::TDigest, tdigestValidator); - registerCreator(StatisticsType::TDigest, tdigestCreator); + registerValidator(StatisticsType::TDigest, tdigestStatisticsValidator); + registerCreator(StatisticsType::TDigest, tdigestStatisticsCreator); - registerValidator(StatisticsType::Uniq, uniqValidator); - registerCreator(StatisticsType::Uniq, uniqCreator); + registerValidator(StatisticsType::Uniq, uniqStatisticsValidator); + registerCreator(StatisticsType::Uniq, uniqStatisticsCreator); #if USE_DATASKETCHES - registerValidator(StatisticsType::CountMinSketch, countMinSketchValidator); - registerCreator(StatisticsType::CountMinSketch, countMinSketchCreator); + registerValidator(StatisticsType::CountMinSketch, countMinSketchStatisticsValidator); + registerCreator(StatisticsType::CountMinSketch, countMinSketchStatisticsCreator); #endif } @@ -222,7 +216,7 @@ MergeTreeStatisticsFactory & MergeTreeStatisticsFactory::instance() return instance; } -void MergeTreeStatisticsFactory::validate(const ColumnStatisticsDescription & stats, DataTypePtr data_type) const +void MergeTreeStatisticsFactory::validate(const ColumnStatisticsDescription & stats, const DataTypePtr & data_type) const { for (const auto & [type, desc] : stats.types_to_desc) { diff --git a/src/Storages/Statistics/Statistics.h b/src/Storages/Statistics/Statistics.h index 16f0c67eabd..2a30c0de315 100644 --- a/src/Storages/Statistics/Statistics.h +++ b/src/Storages/Statistics/Statistics.h @@ -14,12 +14,11 @@ namespace DB 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); + /// - `data_type` is the type of the column on which the statistics object was build on + static std::optional tryConvertToFloat64(const Field & value, const DataTypePtr & data_type); }; /// Statistics describe properties of the values in the column, @@ -87,10 +86,10 @@ class MergeTreeStatisticsFactory : private boost::noncopyable public: static MergeTreeStatisticsFactory & instance(); - void validate(const ColumnStatisticsDescription & stats, DataTypePtr data_type) const; + void validate(const ColumnStatisticsDescription & stats, const DataTypePtr & data_type) const; - using Validator = std::function; - using Creator = std::function; + using Validator = std::function; + using Creator = std::function; ColumnStatisticsPtr get(const ColumnStatisticsDescription & stats) const; ColumnsStatistics getMany(const ColumnsDescription & columns) const; diff --git a/src/Storages/Statistics/StatisticsCountMinSketch.cpp b/src/Storages/Statistics/StatisticsCountMinSketch.cpp index e69bbc1515b..6dbd0625d3d 100644 --- a/src/Storages/Statistics/StatisticsCountMinSketch.cpp +++ b/src/Storages/Statistics/StatisticsCountMinSketch.cpp @@ -25,8 +25,8 @@ extern const int ILLEGAL_STATISTICS; static constexpr auto num_hashes = 7uz; static constexpr auto num_buckets = 2718uz; -StatisticsCountMinSketch::StatisticsCountMinSketch(const SingleStatisticsDescription & stat_, DataTypePtr data_type_) - : IStatistics(stat_) +StatisticsCountMinSketch::StatisticsCountMinSketch(const SingleStatisticsDescription & description, const DataTypePtr & data_type_) + : IStatistics(description) , sketch(num_hashes, num_buckets) , data_type(data_type_) { @@ -48,7 +48,7 @@ Float64 StatisticsCountMinSketch::estimateEqual(const Field & val) const return sketch.get_estimate(&val_converted, data_type->getSizeOfValueInMemory()); if (isStringOrFixedString(data_type)) - return sketch.get_estimate(val.get()); + return sketch.get_estimate(val.safeGet()); throw Exception(ErrorCodes::LOGICAL_ERROR, "Statistics 'count_min' does not support estimate data type of {}", data_type->getName()); } @@ -84,17 +84,17 @@ void StatisticsCountMinSketch::deserialize(ReadBuffer & buf) } -void countMinSketchValidator(const SingleStatisticsDescription &, DataTypePtr data_type) +void countMinSketchStatisticsValidator(const SingleStatisticsDescription & /*description*/, const DataTypePtr & data_type) { - data_type = removeNullable(data_type); - data_type = removeLowCardinalityAndNullable(data_type); - if (!data_type->isValueRepresentedByNumber() && !isStringOrFixedString(data_type)) + DataTypePtr inner_data_type = removeNullable(data_type); + inner_data_type = removeLowCardinalityAndNullable(inner_data_type); + if (!inner_data_type->isValueRepresentedByNumber() && !isStringOrFixedString(inner_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) +StatisticsPtr countMinSketchStatisticsCreator(const SingleStatisticsDescription & description, const DataTypePtr & data_type) { - return std::make_shared(stat, data_type); + return std::make_shared(description, data_type); } } diff --git a/src/Storages/Statistics/StatisticsCountMinSketch.h b/src/Storages/Statistics/StatisticsCountMinSketch.h index 6c8b74f8c35..d1de1a3aea5 100644 --- a/src/Storages/Statistics/StatisticsCountMinSketch.h +++ b/src/Storages/Statistics/StatisticsCountMinSketch.h @@ -14,7 +14,7 @@ namespace DB class StatisticsCountMinSketch : public IStatistics { public: - StatisticsCountMinSketch(const SingleStatisticsDescription & stat_, DataTypePtr data_type_); + StatisticsCountMinSketch(const SingleStatisticsDescription & description, const DataTypePtr & data_type_); Float64 estimateEqual(const Field & val) const override; @@ -31,8 +31,8 @@ private: }; -void countMinSketchValidator(const SingleStatisticsDescription &, DataTypePtr data_type); -StatisticsPtr countMinSketchCreator(const SingleStatisticsDescription & stat, DataTypePtr); +void countMinSketchStatisticsValidator(const SingleStatisticsDescription & description, const DataTypePtr & data_type); +StatisticsPtr countMinSketchStatisticsCreator(const SingleStatisticsDescription & description, const DataTypePtr & data_type); } diff --git a/src/Storages/Statistics/StatisticsTDigest.cpp b/src/Storages/Statistics/StatisticsTDigest.cpp index 66150e00fdb..285b779036f 100644 --- a/src/Storages/Statistics/StatisticsTDigest.cpp +++ b/src/Storages/Statistics/StatisticsTDigest.cpp @@ -1,33 +1,29 @@ #include -#include #include +#include namespace DB { namespace ErrorCodes { extern const int ILLEGAL_STATISTICS; -extern const int LOGICAL_ERROR; } -StatisticsTDigest::StatisticsTDigest(const SingleStatisticsDescription & stat_) - : IStatistics(stat_) +StatisticsTDigest::StatisticsTDigest(const SingleStatisticsDescription & description, const DataTypePtr & data_type_) + : IStatistics(description) + , data_type(data_type_) { } void StatisticsTDigest::update(const ColumnPtr & column) { - size_t rows = column->size(); - for (size_t row = 0; row < rows; ++row) + for (size_t row = 0; row < column->size(); ++row) { - Field field; - column->get(row, field); - - if (field.isNull()) + if (column->isNullAt(row)) continue; - if (auto field_as_float = StatisticsUtils::tryConvertToFloat64(field)) - t_digest.add(*field_as_float, 1); + auto data = column->getFloat64(row); + t_digest.add(data, 1); } } @@ -43,31 +39,31 @@ void StatisticsTDigest::deserialize(ReadBuffer & 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()); + auto val_as_float = StatisticsUtils::tryConvertToFloat64(val, data_type); + if (!val_as_float.has_value()) + return 0; + return t_digest.getCountLessThan(*val_as_float); } 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()); + auto val_as_float = StatisticsUtils::tryConvertToFloat64(val, data_type); + if (!val_as_float.has_value()) + return 0; + return t_digest.getCountEqual(*val_as_float); } -void tdigestValidator(const SingleStatisticsDescription &, DataTypePtr data_type) +void tdigestStatisticsValidator(const SingleStatisticsDescription & /*description*/, const DataTypePtr & data_type) { - data_type = removeNullable(data_type); - data_type = removeLowCardinalityAndNullable(data_type); - if (!data_type->isValueRepresentedByNumber()) + DataTypePtr inner_data_type = removeNullable(data_type); + inner_data_type = removeLowCardinalityAndNullable(inner_data_type); + if (!inner_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) +StatisticsPtr tdigestStatisticsCreator(const SingleStatisticsDescription & description, const DataTypePtr & data_type) { - return std::make_shared(stat); + return std::make_shared(description, data_type); } } diff --git a/src/Storages/Statistics/StatisticsTDigest.h b/src/Storages/Statistics/StatisticsTDigest.h index 614973e5d8b..5e744fee2ce 100644 --- a/src/Storages/Statistics/StatisticsTDigest.h +++ b/src/Storages/Statistics/StatisticsTDigest.h @@ -9,7 +9,7 @@ namespace DB class StatisticsTDigest : public IStatistics { public: - explicit StatisticsTDigest(const SingleStatisticsDescription & stat_); + explicit StatisticsTDigest(const SingleStatisticsDescription & description, const DataTypePtr & data_type_); void update(const ColumnPtr & column) override; @@ -21,9 +21,10 @@ public: private: QuantileTDigest t_digest; + DataTypePtr data_type; }; -void tdigestValidator(const SingleStatisticsDescription &, DataTypePtr data_type); -StatisticsPtr tdigestCreator(const SingleStatisticsDescription & stat, DataTypePtr); +void tdigestStatisticsValidator(const SingleStatisticsDescription & description, const DataTypePtr & data_type); +StatisticsPtr tdigestStatisticsCreator(const SingleStatisticsDescription & description, const DataTypePtr & data_type); } diff --git a/src/Storages/Statistics/StatisticsUniq.cpp b/src/Storages/Statistics/StatisticsUniq.cpp index 8f60ffcf0b5..07311b5b86d 100644 --- a/src/Storages/Statistics/StatisticsUniq.cpp +++ b/src/Storages/Statistics/StatisticsUniq.cpp @@ -11,8 +11,8 @@ namespace ErrorCodes extern const int ILLEGAL_STATISTICS; } -StatisticsUniq::StatisticsUniq(const SingleStatisticsDescription & stat_, const DataTypePtr & data_type) - : IStatistics(stat_) +StatisticsUniq::StatisticsUniq(const SingleStatisticsDescription & description, const DataTypePtr & data_type) + : IStatistics(description) { arena = std::make_unique(); AggregateFunctionProperties properties; @@ -52,17 +52,17 @@ UInt64 StatisticsUniq::estimateCardinality() const return column->getUInt(0); } -void uniqValidator(const SingleStatisticsDescription &, DataTypePtr data_type) +void uniqStatisticsValidator(const SingleStatisticsDescription & /*description*/, const DataTypePtr & data_type) { - data_type = removeNullable(data_type); - data_type = removeLowCardinalityAndNullable(data_type); - if (!data_type->isValueRepresentedByNumber()) + DataTypePtr inner_data_type = removeNullable(data_type); + inner_data_type = removeLowCardinalityAndNullable(inner_data_type); + if (!inner_data_type->isValueRepresentedByNumber()) 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 uniqStatisticsCreator(const SingleStatisticsDescription & description, const DataTypePtr & data_type) { - return std::make_shared(stat, data_type); + return std::make_shared(description, data_type); } } diff --git a/src/Storages/Statistics/StatisticsUniq.h b/src/Storages/Statistics/StatisticsUniq.h index faabde8d47c..1fdcab8bd89 100644 --- a/src/Storages/Statistics/StatisticsUniq.h +++ b/src/Storages/Statistics/StatisticsUniq.h @@ -10,7 +10,7 @@ namespace DB class StatisticsUniq : public IStatistics { public: - StatisticsUniq(const SingleStatisticsDescription & stat_, const DataTypePtr & data_type); + StatisticsUniq(const SingleStatisticsDescription & description, const DataTypePtr & data_type); ~StatisticsUniq() override; void update(const ColumnPtr & column) override; @@ -27,7 +27,7 @@ private: }; -void uniqValidator(const SingleStatisticsDescription &, DataTypePtr data_type); -StatisticsPtr uniqCreator(const SingleStatisticsDescription & stat, DataTypePtr data_type); +void uniqStatisticsValidator(const SingleStatisticsDescription & description, const DataTypePtr & data_type); +StatisticsPtr uniqStatisticsCreator(const SingleStatisticsDescription & description, const DataTypePtr & data_type); } diff --git a/src/Storages/StorageBuffer.cpp b/src/Storages/StorageBuffer.cpp index 04e6d6676d1..f753d369d2d 100644 --- a/src/Storages/StorageBuffer.cpp +++ b/src/Storages/StorageBuffer.cpp @@ -1,3 +1,7 @@ +#include + +#include +#include #include #include #include @@ -23,7 +27,6 @@ #include #include #include -#include #include #include #include @@ -232,6 +235,12 @@ QueryProcessingStage::Enum StorageBuffer::getQueryProcessingStage( return QueryProcessingStage::FetchColumns; } +bool StorageBuffer::isRemote() const +{ + auto destination = getDestinationTable(); + return destination && destination->isRemote(); +} + void StorageBuffer::read( QueryPlan & query_plan, const Names & column_names, @@ -242,6 +251,29 @@ void StorageBuffer::read( size_t max_block_size, size_t num_streams) { + bool allow_experimental_analyzer = local_context->getSettingsRef().allow_experimental_analyzer; + + if (allow_experimental_analyzer && processed_stage > QueryProcessingStage::FetchColumns) + { + /** For query processing stages after FetchColumns, we do not allow using the same table more than once in the query. + * For example: SELECT * FROM buffer t1 JOIN buffer t2 USING (column) + * In that case, we will execute this query separately for the destination table and for the buffer, resulting in incorrect results. + */ + const auto & current_storage_id = getStorageID(); + auto table_nodes = extractAllTableReferences(query_info.query_tree); + size_t count_of_current_storage = 0; + for (const auto & node : table_nodes) + { + const auto & table_node = node->as(); + if (table_node.getStorageID().getFullNameNotQuoted() == current_storage_id.getFullNameNotQuoted()) + { + count_of_current_storage++; + if (count_of_current_storage > 1) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "StorageBuffer over Distributed does not support using the same table more than once in the query"); + } + } + } + const auto & metadata_snapshot = storage_snapshot->metadata; if (auto destination = getDestinationTable()) @@ -328,13 +360,17 @@ void StorageBuffer::read( } } + src_table_query_info.merge_storage_snapshot = storage_snapshot; destination->read( query_plan, columns_intersection, destination_snapshot, src_table_query_info, local_context, processed_stage, max_block_size, num_streams); - if (query_plan.isInitialized()) + if (query_plan.isInitialized() && processed_stage <= QueryProcessingStage::FetchColumns) { - + /** The code below converts columns from metadata_snapshot to columns from destination_metadata_snapshot. + * This conversion is not applicable for processed_stage > FetchColumns. + * Instead, we rely on the converting actions at the end of this function. + */ auto actions = addMissingDefaults( query_plan.getCurrentDataStream().header, header_after_adding_defaults.getNamesAndTypesList(), @@ -397,7 +433,7 @@ void StorageBuffer::read( /// TODO: Find a way to support projections for StorageBuffer if (processed_stage > QueryProcessingStage::FetchColumns) { - if (local_context->getSettingsRef().allow_experimental_analyzer) + if (allow_experimental_analyzer) { auto storage = std::make_shared( getStorageID(), diff --git a/src/Storages/StorageBuffer.h b/src/Storages/StorageBuffer.h index cd6dd7b933f..02376f286b1 100644 --- a/src/Storages/StorageBuffer.h +++ b/src/Storages/StorageBuffer.h @@ -84,6 +84,7 @@ public: QueryProcessingStage::Enum processed_stage, size_t max_block_size, size_t num_streams) override; + bool isRemote() const override; bool supportsParallelInsert() const override { return true; } diff --git a/src/Storages/StorageDistributed.cpp b/src/Storages/StorageDistributed.cpp index 3e38ddf830a..e146e95f89f 100644 --- a/src/Storages/StorageDistributed.cpp +++ b/src/Storages/StorageDistributed.cpp @@ -308,7 +308,8 @@ StorageDistributed::StorageDistributed( const DistributedSettings & distributed_settings_, LoadingStrictnessLevel mode, ClusterPtr owned_cluster_, - ASTPtr remote_table_function_ptr_) + ASTPtr remote_table_function_ptr_, + bool is_remote_function_) : IStorage(id_) , WithContext(context_->getGlobalContext()) , remote_database(remote_database_) @@ -322,6 +323,7 @@ StorageDistributed::StorageDistributed( , relative_data_path(relative_data_path_) , distributed_settings(distributed_settings_) , rng(randomSeed()) + , is_remote_function(is_remote_function_) { if (!distributed_settings.flush_on_detach && distributed_settings.background_insert_batch) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Settings flush_on_detach=0 and background_insert_batch=1 are incompatible"); @@ -373,38 +375,6 @@ StorageDistributed::StorageDistributed( } -StorageDistributed::StorageDistributed( - const StorageID & id_, - const ColumnsDescription & columns_, - const ConstraintsDescription & constraints_, - ASTPtr remote_table_function_ptr_, - const String & cluster_name_, - ContextPtr context_, - const ASTPtr & sharding_key_, - const String & storage_policy_name_, - const String & relative_data_path_, - const DistributedSettings & distributed_settings_, - LoadingStrictnessLevel mode, - ClusterPtr owned_cluster_) - : StorageDistributed( - id_, - columns_, - constraints_, - String{}, - String{}, - String{}, - cluster_name_, - context_, - sharding_key_, - storage_policy_name_, - relative_data_path_, - distributed_settings_, - mode, - std::move(owned_cluster_), - remote_table_function_ptr_) -{ -} - QueryProcessingStage::Enum StorageDistributed::getQueryProcessingStage( ContextPtr local_context, QueryProcessingStage::Enum to_stage, @@ -901,7 +871,7 @@ void StorageDistributed::read( sharding_key_column_name, distributed_settings, shard_filter_generator, - /* is_remote_function= */ static_cast(owned_cluster)); + is_remote_function); /// This is a bug, it is possible only when there is no shards to query, and this is handled earlier. if (!query_plan.isInitialized()) diff --git a/src/Storages/StorageDistributed.h b/src/Storages/StorageDistributed.h index 85a8de86953..8a5585e9fd0 100644 --- a/src/Storages/StorageDistributed.h +++ b/src/Storages/StorageDistributed.h @@ -61,21 +61,8 @@ public: const DistributedSettings & distributed_settings_, LoadingStrictnessLevel mode, ClusterPtr owned_cluster_ = {}, - ASTPtr remote_table_function_ptr_ = {}); - - StorageDistributed( - const StorageID & id_, - const ColumnsDescription & columns_, - const ConstraintsDescription & constraints_, - ASTPtr remote_table_function_ptr_, - const String & cluster_name_, - ContextPtr context_, - const ASTPtr & sharding_key_, - const String & storage_policy_name_, - const String & relative_data_path_, - const DistributedSettings & distributed_settings_, - LoadingStrictnessLevel mode, - ClusterPtr owned_cluster_ = {}); + ASTPtr remote_table_function_ptr_ = {}, + bool is_remote_function_ = false); ~StorageDistributed() override; @@ -287,6 +274,8 @@ private: // For random shard index generation mutable std::mutex rng_mutex; pcg64 rng; + + bool is_remote_function; }; } diff --git a/src/Storages/StorageExternalDistributed.cpp b/src/Storages/StorageExternalDistributed.cpp index 951c87807bb..9fc8b588c89 100644 --- a/src/Storages/StorageExternalDistributed.cpp +++ b/src/Storages/StorageExternalDistributed.cpp @@ -1,4 +1,4 @@ -#include "StorageExternalDistributed.h" +#include #include #include @@ -6,6 +6,8 @@ #include #include #include +#include +#include #include #include #include @@ -112,14 +114,39 @@ void registerStorageExternalDistributed(StorageFactory & factory) std::unordered_set shards; ASTs inner_engine_args(engine_args.begin() + 1, engine_args.end()); + ASTPtr * address_arg = nullptr; + + /// If there is a named collection argument, named `addresses_expr` + for (auto & node : inner_engine_args) + { + if (ASTFunction * func = node->as(); func && func->name == "equals" && func->arguments) + { + if (ASTExpressionList * func_args = func->arguments->as(); func_args && func_args->children.size() == 2) + { + if (ASTIdentifier * arg_name = func_args->children[0]->as(); arg_name && arg_name->name() == "addresses_expr") + { + address_arg = &func_args->children[1]; + break; + } + } + } + } + + /// Otherwise it is the first argument. + if (!address_arg) + address_arg = &inner_engine_args.at(0); + + String addresses_expr = checkAndGetLiteralArgument(*address_arg, "addresses"); + Strings shards_addresses = get_addresses(addresses_expr); + auto engine_name = checkAndGetLiteralArgument(engine_args[0], "engine_name"); if (engine_name == "URL") { - auto configuration = StorageURL::getConfiguration(inner_engine_args, context); - auto shards_addresses = get_addresses(configuration.addresses_expr); auto format_settings = StorageURL::getFormatSettingsFromArgs(args); for (const auto & shard_address : shards_addresses) { + *address_arg = std::make_shared(shard_address); + auto configuration = StorageURL::getConfiguration(inner_engine_args, context); auto uri_options = parseRemoteDescription(shard_address, 0, shard_address.size(), '|', max_addresses); if (uri_options.size() > 1) { @@ -140,13 +167,12 @@ void registerStorageExternalDistributed(StorageFactory & factory) else if (engine_name == "MySQL") { MySQLSettings mysql_settings; - auto configuration = StorageMySQL::getConfiguration(inner_engine_args, context, mysql_settings); - auto shards_addresses = get_addresses(configuration.addresses_expr); for (const auto & shard_address : shards_addresses) { - auto current_configuration{configuration}; - current_configuration.addresses = parseRemoteDescriptionForExternalDatabase(shard_address, max_addresses, 3306); - auto pool = createMySQLPoolWithFailover(current_configuration, mysql_settings); + *address_arg = std::make_shared(shard_address); + auto configuration = StorageMySQL::getConfiguration(inner_engine_args, context, mysql_settings); + configuration.addresses = parseRemoteDescriptionForExternalDatabase(shard_address, max_addresses, 3306); + auto pool = createMySQLPoolWithFailover(configuration, mysql_settings); shards.insert(std::make_shared( args.table_id, std::move(pool), configuration.database, configuration.table, /* replace_query = */ false, /* on_duplicate_clause = */ "", @@ -157,14 +183,13 @@ void registerStorageExternalDistributed(StorageFactory & factory) #if USE_LIBPQXX else if (engine_name == "PostgreSQL") { - auto configuration = StoragePostgreSQL::getConfiguration(inner_engine_args, context); - auto shards_addresses = get_addresses(configuration.addresses_expr); for (const auto & shard_address : shards_addresses) { - auto current_configuration{configuration}; - current_configuration.addresses = parseRemoteDescriptionForExternalDatabase(shard_address, max_addresses, 5432); + *address_arg = std::make_shared(shard_address); + auto configuration = StoragePostgreSQL::getConfiguration(inner_engine_args, context); + configuration.addresses = parseRemoteDescriptionForExternalDatabase(shard_address, max_addresses, 5432); auto pool = std::make_shared( - current_configuration, + configuration, settings.postgresql_connection_pool_size, settings.postgresql_connection_pool_wait_timeout, settings.postgresql_connection_pool_retries, diff --git a/src/Storages/StorageFactory.cpp b/src/Storages/StorageFactory.cpp index 060b271d8f4..b95ccedb093 100644 --- a/src/Storages/StorageFactory.cpp +++ b/src/Storages/StorageFactory.cpp @@ -203,7 +203,7 @@ StoragePtr StorageFactory::get( } if (query.comment) - comment = query.comment->as().value.get(); + comment = query.comment->as().value.safeGet(); ASTs empty_engine_args; Arguments arguments{ diff --git a/src/Storages/StorageFile.cpp b/src/Storages/StorageFile.cpp index 8c079aa4600..766b7722cdf 100644 --- a/src/Storages/StorageFile.cpp +++ b/src/Storages/StorageFile.cpp @@ -53,6 +53,8 @@ #include #include #include +#include +#include "base/defines.h" #include @@ -515,7 +517,7 @@ namespace StorageFile::getSchemaCache(getContext()).addManyColumns(cache_keys, columns); } - String getLastFileName() const override + String getLastFilePath() const override { if (current_index != 0) return paths[current_index - 1]; @@ -792,7 +794,7 @@ namespace format = format_name; } - String getLastFileName() const override + String getLastFilePath() const override { return last_read_file_path; } @@ -1111,7 +1113,8 @@ void StorageFile::setStorageMetadata(CommonArguments args) storage_metadata.setConstraints(args.constraints); storage_metadata.setComment(args.comment); setInMemoryMetadata(storage_metadata); - setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(storage_metadata.getColumns())); + + setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(storage_metadata.getColumns(), args.getContext(), paths.empty() ? "" : paths[0], format_settings)); } @@ -1465,7 +1468,7 @@ Chunk StorageFileSource::generate() .size = current_file_size, .filename = (filename_override.has_value() ? &filename_override.value() : nullptr), .last_modified = current_file_last_modified - }); + }, getContext(), columns_description); return chunk; } @@ -1767,6 +1770,12 @@ public: initialize(); } + ~StorageFileSink() override + { + if (isCancelled()) + cancelBuffers(); + } + void initialize() { std::unique_ptr naked_buffer; @@ -1800,37 +1809,14 @@ public: void consume(Chunk & chunk) override { - std::lock_guard cancel_lock(cancel_mutex); - if (cancelled) + if (isCancelled()) return; writer->write(getHeader().cloneWithColumns(chunk.getColumns())); } - void onCancel() override - { - std::lock_guard cancel_lock(cancel_mutex); - cancelBuffers(); - releaseBuffers(); - cancelled = true; - } - - void onException(std::exception_ptr exception) override - { - std::lock_guard cancel_lock(cancel_mutex); - try - { - std::rethrow_exception(exception); - } - catch (...) - { - /// An exception context is needed to proper delete write buffers without finalization - releaseBuffers(); - } - } - void onFinish() override { - std::lock_guard cancel_lock(cancel_mutex); + chassert(!isCancelled()); finalizeBuffers(); } @@ -1885,9 +1871,6 @@ private: int flags; std::unique_lock lock; - - std::mutex cancel_mutex; - bool cancelled = false; }; class PartitionedStorageFileSink : public PartitionedSink @@ -2204,12 +2187,12 @@ void registerStorageFile(StorageFactory & factory) { auto type = literal->value.getType(); if (type == Field::Types::Int64) - source_fd = static_cast(literal->value.get()); + source_fd = static_cast(literal->value.safeGet()); else if (type == Field::Types::UInt64) - source_fd = static_cast(literal->value.get()); + source_fd = static_cast(literal->value.safeGet()); else if (type == Field::Types::String) StorageFile::parseFileSource( - literal->value.get(), + literal->value.safeGet(), source_path, storage_args.path_to_archive, factory_args.getLocalContext()->getSettingsRef().allow_archive_path_syntax); diff --git a/src/Storages/StorageFileCluster.cpp b/src/Storages/StorageFileCluster.cpp index d43e242f70c..82ae0b761ae 100644 --- a/src/Storages/StorageFileCluster.cpp +++ b/src/Storages/StorageFileCluster.cpp @@ -61,7 +61,7 @@ StorageFileCluster::StorageFileCluster( storage_metadata.setConstraints(constraints_); setInMemoryMetadata(storage_metadata); - setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(storage_metadata.getColumns())); + setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(storage_metadata.getColumns(), context, paths.empty() ? "" : paths[0])); } void StorageFileCluster::updateQueryToSendIfNeeded(DB::ASTPtr & query, const StorageSnapshotPtr & storage_snapshot, const DB::ContextPtr & context) diff --git a/src/Storages/StorageFuzzJSON.cpp b/src/Storages/StorageFuzzJSON.cpp index 9950d41f1c2..fc73f246d35 100644 --- a/src/Storages/StorageFuzzJSON.cpp +++ b/src/Storages/StorageFuzzJSON.cpp @@ -419,7 +419,7 @@ void fuzzJSONObject( if (val.fixed->getType() == Field::Types::Which::String) { out << fuzzJSONStructure(config, rnd, "\""); - writeText(val.fixed->get(), out); + writeText(val.fixed->safeGet(), out); out << fuzzJSONStructure(config, rnd, "\""); } else diff --git a/src/Storages/StorageFuzzQuery.cpp b/src/Storages/StorageFuzzQuery.cpp new file mode 100644 index 00000000000..6e8f425f8dc --- /dev/null +++ b/src/Storages/StorageFuzzQuery.cpp @@ -0,0 +1,169 @@ +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int BAD_ARGUMENTS; + extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; +} + +ColumnPtr FuzzQuerySource::createColumn() +{ + auto column = ColumnString::create(); + ColumnString::Chars & data_to = column->getChars(); + ColumnString::Offsets & offsets_to = column->getOffsets(); + + offsets_to.resize(block_size); + IColumn::Offset offset = 0; + + auto fuzz_base = query; + size_t row_num = 0; + + while (row_num < block_size) + { + ASTPtr new_query = fuzz_base->clone(); + + auto base_before_fuzz = fuzz_base->formatForErrorMessage(); + fuzzer.fuzzMain(new_query); + auto fuzzed_text = new_query->formatForErrorMessage(); + + if (base_before_fuzz == fuzzed_text) + continue; + + /// AST is too long, will start from the original query. + if (config.max_query_length > 500) + { + fuzz_base = query; + continue; + } + + IColumn::Offset next_offset = offset + fuzzed_text.size() + 1; + data_to.resize(next_offset); + + std::copy(fuzzed_text.begin(), fuzzed_text.end(), &data_to[offset]); + + data_to[offset + fuzzed_text.size()] = 0; + offsets_to[row_num] = next_offset; + + offset = next_offset; + fuzz_base = new_query; + ++row_num; + } + + return column; +} + +StorageFuzzQuery::StorageFuzzQuery( + const StorageID & table_id_, const ColumnsDescription & columns_, const String & comment_, const Configuration & config_) + : IStorage(table_id_), config(config_) +{ + StorageInMemoryMetadata storage_metadata; + storage_metadata.setColumns(columns_); + storage_metadata.setComment(comment_); + setInMemoryMetadata(storage_metadata); +} + +Pipe StorageFuzzQuery::read( + 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) +{ + storage_snapshot->check(column_names); + + Pipes pipes; + pipes.reserve(num_streams); + + const ColumnsDescription & our_columns = storage_snapshot->metadata->getColumns(); + Block block_header; + for (const auto & name : column_names) + { + const auto & name_type = our_columns.get(name); + MutableColumnPtr column = name_type.type->createColumn(); + block_header.insert({std::move(column), name_type.type, name_type.name}); + } + + const char * begin = config.query.data(); + const char * end = begin + config.query.size(); + + ParserQuery parser(end, false); + auto query = parseQuery(parser, begin, end, "", 0, DBMS_DEFAULT_MAX_PARSER_DEPTH, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS); + + for (UInt64 i = 0; i < num_streams; ++i) + pipes.emplace_back(std::make_shared(max_block_size, block_header, config, query)); + + return Pipe::unitePipes(std::move(pipes)); +} + +StorageFuzzQuery::Configuration StorageFuzzQuery::getConfiguration(ASTs & engine_args, ContextPtr local_context) +{ + StorageFuzzQuery::Configuration configuration{}; + + // Supported signatures: + // + // FuzzQuery(query) + // FuzzQuery(query, max_query_length) + // FuzzQuery(query, max_query_length, random_seed) + if (engine_args.empty() || engine_args.size() > 3) + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "FuzzQuery requires 1 to 3 arguments: query, max_query_length, random_seed"); + + for (auto & engine_arg : engine_args) + engine_arg = evaluateConstantExpressionOrIdentifierAsLiteral(engine_arg, local_context); + + auto first_arg = checkAndGetLiteralArgument(engine_args[0], "query"); + configuration.query = std::move(first_arg); + + if (engine_args.size() >= 2) + { + const auto & literal = engine_args[1]->as(); + if (!literal.value.isNull()) + configuration.max_query_length = checkAndGetLiteralArgument(literal, "max_query_length"); + } + + if (engine_args.size() == 3) + { + const auto & literal = engine_args[2]->as(); + if (!literal.value.isNull()) + configuration.random_seed = checkAndGetLiteralArgument(literal, "random_seed"); + } + + return configuration; +} + +void registerStorageFuzzQuery(StorageFactory & factory) +{ + factory.registerStorage( + "FuzzQuery", + [](const StorageFactory::Arguments & args) -> std::shared_ptr + { + ASTs & engine_args = args.engine_args; + + if (engine_args.empty()) + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Storage FuzzQuery must have arguments."); + + StorageFuzzQuery::Configuration configuration = StorageFuzzQuery::getConfiguration(engine_args, args.getLocalContext()); + + for (const auto& col : args.columns) + if (col.type->getTypeId() != TypeIndex::String) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "'StorageFuzzQuery' supports only columns of String type, got {}.", col.type->getName()); + + return std::make_shared(args.table_id, args.columns, args.comment, configuration); + }); +} + +} diff --git a/src/Storages/StorageFuzzQuery.h b/src/Storages/StorageFuzzQuery.h new file mode 100644 index 00000000000..125ef960e74 --- /dev/null +++ b/src/Storages/StorageFuzzQuery.h @@ -0,0 +1,88 @@ +#pragma once + +#include +#include +#include +#include + +#include "config.h" + +namespace DB +{ + +class NamedCollection; + +class StorageFuzzQuery final : public IStorage +{ +public: + struct Configuration : public StatelessTableEngineConfiguration + { + String query; + UInt64 max_query_length = 500; + UInt64 random_seed = randomSeed(); + }; + + StorageFuzzQuery( + const StorageID & table_id_, const ColumnsDescription & columns_, const String & comment_, const Configuration & config_); + + std::string getName() const override { return "FuzzQuery"; } + + Pipe read( + 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; + + static StorageFuzzQuery::Configuration getConfiguration(ASTs & engine_args, ContextPtr local_context); + +private: + const Configuration config; +}; + + +class FuzzQuerySource : public ISource +{ +public: + FuzzQuerySource( + UInt64 block_size_, Block block_header_, const StorageFuzzQuery::Configuration & config_, ASTPtr query_) + : ISource(block_header_) + , block_size(block_size_) + , block_header(std::move(block_header_)) + , config(config_) + , query(query_) + , fuzzer(config_.random_seed) + { + } + + String getName() const override { return "FuzzQuery"; } + +protected: + Chunk generate() override + { + Columns columns; + columns.reserve(block_header.columns()); + for (const auto & col : block_header) + { + chassert(col.type->getTypeId() == TypeIndex::String); + columns.emplace_back(createColumn()); + } + + return {std::move(columns), block_size}; + } + +private: + ColumnPtr createColumn(); + + UInt64 block_size; + Block block_header; + + StorageFuzzQuery::Configuration config; + ASTPtr query; + + QueryFuzzer fuzzer; +}; + +} diff --git a/src/Storages/StorageInMemoryMetadata.cpp b/src/Storages/StorageInMemoryMetadata.cpp index 2226de3e64f..4a655cac566 100644 --- a/src/Storages/StorageInMemoryMetadata.cpp +++ b/src/Storages/StorageInMemoryMetadata.cpp @@ -16,6 +16,7 @@ #include #include #include +#include namespace DB @@ -334,10 +335,17 @@ ColumnDependencies StorageInMemoryMetadata::getColumnDependencies( NameSet required_ttl_columns; NameSet updated_ttl_columns; - auto add_dependent_columns = [&updated_columns](const Names & required_columns, auto & to_set) + auto add_dependent_columns = [&updated_columns](const Names & required_columns, auto & to_set, bool is_projection = false) { for (const auto & dependency : required_columns) { + /// useful in the case of lightweight delete with wide part and option of rebuild projection + if (is_projection && updated_columns.contains(RowExistsColumn::name)) + { + to_set.insert(required_columns.begin(), required_columns.end()); + return true; + } + if (updated_columns.contains(dependency)) { to_set.insert(required_columns.begin(), required_columns.end()); @@ -357,7 +365,7 @@ ColumnDependencies StorageInMemoryMetadata::getColumnDependencies( for (const auto & projection : getProjections()) { if (has_dependency(projection.name, ColumnDependency::PROJECTION)) - add_dependent_columns(projection.getRequiredColumns(), projections_columns); + add_dependent_columns(projection.getRequiredColumns(), projections_columns, true); } auto add_for_rows_ttl = [&](const auto & expression, auto & to_set) diff --git a/src/Storages/StorageJoin.cpp b/src/Storages/StorageJoin.cpp index a0d6cf11b64..efa15c382dd 100644 --- a/src/Storages/StorageJoin.cpp +++ b/src/Storages/StorageJoin.cpp @@ -75,6 +75,7 @@ StorageJoin::StorageJoin( table_join = std::make_shared(limits, use_nulls, kind, strictness, key_names); join = std::make_shared(table_join, getRightSampleBlock(), overwrite); restore(); + optimizeUnlocked(); } RWLockImpl::LockHolder StorageJoin::tryLockTimedWithContext(const RWLock & lock, RWLockImpl::Type type, ContextPtr context) const @@ -99,6 +100,47 @@ SinkToStoragePtr StorageJoin::write(const ASTPtr & query, const StorageMetadataP return StorageSetOrJoinBase::write(query, metadata_snapshot, context, /*async_insert=*/false); } +bool StorageJoin::optimize( + const ASTPtr & /*query*/, + const StorageMetadataPtr & /*metadata_snapshot*/, + const ASTPtr & partition, + bool final, + bool deduplicate, + const Names & /* deduplicate_by_columns */, + bool cleanup, + ContextPtr context) +{ + + if (partition) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Partition cannot be specified when optimizing table of type Join"); + + if (final) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "FINAL cannot be specified when optimizing table of type Join"); + + if (deduplicate) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "DEDUPLICATE cannot be specified when optimizing table of type Join"); + + if (cleanup) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "CLEANUP cannot be specified when optimizing table of type Join"); + + std::lock_guard mutate_lock(mutate_mutex); + TableLockHolder lock_holder = tryLockTimedWithContext(rwlock, RWLockImpl::Write, context); + + optimizeUnlocked(); + return true; +} + +void StorageJoin::optimizeUnlocked() +{ + size_t current_bytes = join->getTotalByteCount(); + size_t dummy = current_bytes; + join->shrinkStoredBlocksToFit(dummy, true); + + size_t optimized_bytes = join->getTotalByteCount(); + if (current_bytes > optimized_bytes) + LOG_INFO(getLogger("StorageJoin"), "Optimized Join storage from {} to {} bytes", current_bytes, optimized_bytes); +} + void StorageJoin::truncate(const ASTPtr &, const StorageMetadataPtr &, ContextPtr context, TableExclusiveLockHolder &) { std::lock_guard mutate_lock(mutate_mutex); @@ -341,10 +383,10 @@ void registerStorageJoin(StorageFactory & factory) else if (setting.name == "any_join_distinct_right_table_keys") old_any_join = setting.value; else if (setting.name == "disk") - disk_name = setting.value.get(); + disk_name = setting.value.safeGet(); else if (setting.name == "persistent") { - persistent = setting.value.get(); + persistent = setting.value.safeGet(); } else throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unknown setting {} for storage {}", setting.name, args.engine_name); @@ -504,7 +546,11 @@ protected: return {}; Chunk chunk; - if (!joinDispatch(join->kind, join->strictness, join->data->maps.front(), + if (!joinDispatch( + join->kind, + join->strictness, + join->data->maps.front(), + join->table_join->getMixedJoinExpression() != nullptr, [&](auto kind, auto strictness, auto & map) { chunk = createChunk(map); })) throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown JOIN strictness"); return chunk; diff --git a/src/Storages/StorageJoin.h b/src/Storages/StorageJoin.h index c76df0cb452..10a551b4063 100644 --- a/src/Storages/StorageJoin.h +++ b/src/Storages/StorageJoin.h @@ -61,6 +61,18 @@ public: SinkToStoragePtr write(const ASTPtr & query, const StorageMetadataPtr & metadata_snapshot, ContextPtr context, bool async_insert) override; + bool optimize( + const ASTPtr & /*query*/, + const StorageMetadataPtr & /*metadata_snapshot*/, + const ASTPtr & /*partition*/, + bool /*final*/, + bool /*deduplicate*/, + const Names & /* deduplicate_by_columns */, + bool /*cleanup*/, + ContextPtr /*context*/) override; + + void optimizeUnlocked(); + Pipe read( const Names & column_names, const StorageSnapshotPtr & storage_snapshot, diff --git a/src/Storages/StorageMaterializedView.cpp b/src/Storages/StorageMaterializedView.cpp index 696136834d4..4c6c2fff209 100644 --- a/src/Storages/StorageMaterializedView.cpp +++ b/src/Storages/StorageMaterializedView.cpp @@ -203,7 +203,6 @@ StorageMaterializedView::StorageMaterializedView( { fixed_uuid = false; refresher = RefreshTask::create( - *this, getContext(), *query.refresh_strategy); refresh_on_start = mode < LoadingStrictnessLevel::ATTACH && !query.is_create_empty; diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp index 7c268d36a7b..e88844e2d31 100644 --- a/src/Storages/StorageMerge.cpp +++ b/src/Storages/StorageMerge.cpp @@ -602,7 +602,7 @@ std::vector ReadFromMerge::createChildrenPlans(SelectQ ASTPtr required_columns_expr_list = std::make_shared(); ASTPtr column_expr; - auto sample_block = merge_storage_snapshot->getMetadataForQuery()->getSampleBlock(); + auto sample_block = merge_storage_snapshot->metadata->getSampleBlock(); for (const auto & column : real_column_names) { diff --git a/src/Storages/StorageMergeTree.cpp b/src/Storages/StorageMergeTree.cpp index f55f672fe5e..f7701a2aab8 100644 --- a/src/Storages/StorageMergeTree.cpp +++ b/src/Storages/StorageMergeTree.cpp @@ -895,7 +895,7 @@ void StorageMergeTree::loadDeduplicationLog() std::string path = fs::path(relative_data_path) / "deduplication_logs"; /// If either there is already a deduplication log, or we will be able to use it. - if (disk->exists(path) || !disk->isReadOnly()) + if (!disk->isReadOnly() || disk->exists(path)) { deduplication_log = std::make_unique(path, settings->non_replicated_deduplication_window, format_version, disk); deduplication_log->load(); @@ -1568,10 +1568,12 @@ bool StorageMergeTree::optimize( { assertNotReadonly(); - if (deduplicate && getInMemoryMetadataPtr()->hasProjections()) - throw Exception(ErrorCodes::NOT_IMPLEMENTED, + if (deduplicate && getInMemoryMetadataPtr()->hasProjections() + && getSettings()->deduplicate_merge_projection_mode == DeduplicateMergeProjectionMode::THROW) + throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, "OPTIMIZE DEDUPLICATE query is not supported for table {} as it has projections. " - "User should drop all the projections manually before running the query", + "User should drop all the projections manually before running the query, " + "or consider drop or rebuild option of deduplicate_merge_projection_mode", getStorageID().getTableName()); if (deduplicate) diff --git a/src/Storages/StorageMergeTreeIndex.cpp b/src/Storages/StorageMergeTreeIndex.cpp index 0b1ad02f8c9..15728290f19 100644 --- a/src/Storages/StorageMergeTreeIndex.cpp +++ b/src/Storages/StorageMergeTreeIndex.cpp @@ -275,7 +275,7 @@ public: private: std::shared_ptr storage; Poco::Logger * log; - const ActionsDAG::Node * predicate = nullptr; + ExpressionActionsPtr virtual_columns_filter; }; void ReadFromMergeTreeIndex::applyFilters(ActionDAGNodes added_filter_nodes) @@ -283,7 +283,16 @@ void ReadFromMergeTreeIndex::applyFilters(ActionDAGNodes added_filter_nodes) SourceStepWithFilter::applyFilters(std::move(added_filter_nodes)); if (filter_actions_dag) - predicate = filter_actions_dag->getOutputs().at(0); + { + Block block_to_filter + { + { {}, std::make_shared(), StorageMergeTreeIndex::part_name_column.name }, + }; + + auto dag = VirtualColumnUtils::splitFilterDagForAllowedInputs(filter_actions_dag->getOutputs().at(0), &block_to_filter); + if (dag) + virtual_columns_filter = VirtualColumnUtils::buildFilterExpression(std::move(*dag), context); + } } void StorageMergeTreeIndex::read( @@ -335,7 +344,7 @@ void StorageMergeTreeIndex::read( void ReadFromMergeTreeIndex::initializePipeline(QueryPipelineBuilder & pipeline, const BuildQueryPipelineSettings &) { - auto filtered_parts = storage->getFilteredDataParts(predicate, context); + auto filtered_parts = storage->getFilteredDataParts(virtual_columns_filter); LOG_DEBUG(log, "Reading index{}from {} parts of table {}", storage->with_marks ? " with marks " : " ", @@ -345,9 +354,9 @@ void ReadFromMergeTreeIndex::initializePipeline(QueryPipelineBuilder & pipeline, pipeline.init(Pipe(std::make_shared(getOutputStream().header, storage->key_sample_block, std::move(filtered_parts), context, storage->with_marks))); } -MergeTreeData::DataPartsVector StorageMergeTreeIndex::getFilteredDataParts(const ActionsDAG::Node * predicate, const ContextPtr & context) const +MergeTreeData::DataPartsVector StorageMergeTreeIndex::getFilteredDataParts(const ExpressionActionsPtr & virtual_columns_filter) const { - if (!predicate) + if (!virtual_columns_filter) return data_parts; auto all_part_names = ColumnString::create(); @@ -355,7 +364,7 @@ MergeTreeData::DataPartsVector StorageMergeTreeIndex::getFilteredDataParts(const all_part_names->insert(part->name); Block filtered_block{{std::move(all_part_names), std::make_shared(), part_name_column.name}}; - VirtualColumnUtils::filterBlockWithPredicate(predicate, filtered_block, context); + VirtualColumnUtils::filterBlockWithExpression(virtual_columns_filter, filtered_block); if (!filtered_block.rows()) return {}; diff --git a/src/Storages/StorageMergeTreeIndex.h b/src/Storages/StorageMergeTreeIndex.h index a1fb61d5a56..ed8274d7d92 100644 --- a/src/Storages/StorageMergeTreeIndex.h +++ b/src/Storages/StorageMergeTreeIndex.h @@ -36,7 +36,7 @@ public: private: friend class ReadFromMergeTreeIndex; - MergeTreeData::DataPartsVector getFilteredDataParts(const ActionsDAG::Node * predicate, const ContextPtr & context) const; + MergeTreeData::DataPartsVector getFilteredDataParts(const ExpressionActionsPtr & virtual_columns_filter) const; StoragePtr source_table; bool with_marks; diff --git a/src/Storages/StoragePostgreSQL.cpp b/src/Storages/StoragePostgreSQL.cpp index d3214e7ed13..e0a4af68824 100644 --- a/src/Storages/StoragePostgreSQL.cpp +++ b/src/Storages/StoragePostgreSQL.cpp @@ -294,7 +294,7 @@ public: { const auto * array_type = typeid_cast(data_type.get()); const auto & nested = array_type->getNestedType(); - const auto & array = array_field.get(); + const auto & array = array_field.safeGet(); if (!isArray(nested)) { @@ -312,7 +312,7 @@ public: if (!isArray(nested_array_type->getNestedType())) { - parseArrayContent(iter->get(), nested, ostr); + parseArrayContent(iter->safeGet(), nested, ostr); } else { diff --git a/src/Storages/StorageReplicatedMergeTree.cpp b/src/Storages/StorageReplicatedMergeTree.cpp index 2d826c6c2df..38e94835beb 100644 --- a/src/Storages/StorageReplicatedMergeTree.cpp +++ b/src/Storages/StorageReplicatedMergeTree.cpp @@ -5704,7 +5704,8 @@ std::optional StorageReplicatedMergeTree::distributedWriteFromClu { auto connection = std::make_shared( node.host_name, node.port, query_context->getGlobalContext()->getCurrentDatabase(), - node.user, node.password, SSHKey(), /*jwt*/"", node.quota_key, node.cluster, node.cluster_secret, + node.user, node.password, node.proto_send_chunked, node.proto_recv_chunked, + SSHKey(), /*jwt*/"", node.quota_key, node.cluster, node.cluster_secret, "ParallelInsertSelectInititiator", node.compression, node.secure @@ -5794,10 +5795,12 @@ 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, + if (deduplicate && getInMemoryMetadataPtr()->hasProjections() + && getSettings()->deduplicate_merge_projection_mode == DeduplicateMergeProjectionMode::THROW) + throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, "OPTIMIZE DEDUPLICATE query is not supported for table {} as it has projections. " - "User should drop all the projections manually before running the query", + "User should drop all the projections manually before running the query, " + "or consider drop or rebuild option of deduplicate_merge_projection_mode", getStorageID().getTableName()); if (cleanup) diff --git a/src/Storages/StorageSnapshot.cpp b/src/Storages/StorageSnapshot.cpp index aada25168f8..13a46990556 100644 --- a/src/Storages/StorageSnapshot.cpp +++ b/src/Storages/StorageSnapshot.cpp @@ -63,7 +63,6 @@ std::shared_ptr StorageSnapshot::clone(DataPtr data_) const { auto res = std::make_shared(storage, metadata, object_columns); - res->projection = projection; res->data = std::move(data_); return res; @@ -79,7 +78,7 @@ ColumnsDescription StorageSnapshot::getAllColumnsDescription() const NamesAndTypesList StorageSnapshot::getColumns(const GetColumnsOptions & options) const { - auto all_columns = getMetadataForQuery()->getColumns().get(options); + auto all_columns = metadata->getColumns().get(options); if (options.with_extended_objects) extendObjectColumns(all_columns, object_columns, options.with_subcolumns); @@ -113,7 +112,7 @@ NamesAndTypesList StorageSnapshot::getColumnsByNames(const GetColumnsOptions & o std::optional StorageSnapshot::tryGetColumn(const GetColumnsOptions & options, const String & column_name) const { - const auto & columns = getMetadataForQuery()->getColumns(); + const auto & columns = metadata->getColumns(); auto column = columns.tryGetColumn(options, column_name); if (column && (!column->type->hasDynamicSubcolumnsDeprecated() || !options.with_extended_objects)) return column; @@ -189,7 +188,7 @@ Block StorageSnapshot::getSampleBlockForColumns(const Names & column_names) cons { Block res; - const auto & columns = getMetadataForQuery()->getColumns(); + const auto & columns = metadata->getColumns(); for (const auto & column_name : column_names) { auto column = columns.tryGetColumnOrSubcolumn(GetColumnsOptions::All, column_name); @@ -221,7 +220,7 @@ Block StorageSnapshot::getSampleBlockForColumns(const Names & column_names) cons ColumnsDescription StorageSnapshot::getDescriptionForColumns(const Names & column_names) const { ColumnsDescription res; - const auto & columns = getMetadataForQuery()->getColumns(); + const auto & columns = metadata->getColumns(); for (const auto & name : column_names) { auto column = columns.tryGetColumnOrSubcolumnDescription(GetColumnsOptions::All, name); @@ -257,7 +256,7 @@ namespace void StorageSnapshot::check(const Names & column_names) const { - const auto & columns = getMetadataForQuery()->getColumns(); + const auto & columns = metadata->getColumns(); auto options = GetColumnsOptions(GetColumnsOptions::AllPhysical).withSubcolumns(); if (column_names.empty()) diff --git a/src/Storages/StorageSnapshot.h b/src/Storages/StorageSnapshot.h index 89e97f2abb8..e94d1fc9a26 100644 --- a/src/Storages/StorageSnapshot.h +++ b/src/Storages/StorageSnapshot.h @@ -30,9 +30,6 @@ struct StorageSnapshot using DataPtr = std::unique_ptr; DataPtr data; - /// Projection that is used in query. - mutable const ProjectionDescription * projection = nullptr; - StorageSnapshot( const IStorage & storage_, StorageMetadataPtr metadata_); @@ -82,11 +79,6 @@ struct StorageSnapshot void check(const Names & column_names) const; DataTypePtr getConcreteType(const String & column_name) const; - - void addProjection(const ProjectionDescription * projection_) const { projection = projection_; } - - /// If we have a projection then we should use its metadata. - StorageMetadataPtr getMetadataForQuery() const { return projection ? projection->metadata : metadata; } }; using StorageSnapshotPtr = std::shared_ptr; diff --git a/src/Storages/StorageTimeSeries.cpp b/src/Storages/StorageTimeSeries.cpp new file mode 100644 index 00000000000..3ff57aaf3e5 --- /dev/null +++ b/src/Storages/StorageTimeSeries.cpp @@ -0,0 +1,477 @@ +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int INCORRECT_QUERY; + extern const int LOGICAL_ERROR; + extern const int NOT_IMPLEMENTED; + extern const int SUPPORT_IS_DISABLED; + extern const int UNEXPECTED_TABLE_ENGINE; +} + + +namespace +{ + namespace fs = std::filesystem; + + /// Loads TimeSeries storage settings from a create query. + std::shared_ptr getTimeSeriesSettingsFromQuery(const ASTCreateQuery & query) + { + auto storage_settings = std::make_shared(); + if (query.storage) + storage_settings->loadFromQuery(*query.storage); + return storage_settings; + } + + /// Creates an inner target table or just makes its storage ID. + /// This function is used by the constructor of StorageTimeSeries to find (or create) its target tables. + StorageID initTarget( + ViewTarget::Kind kind, + const ViewTarget * target_info, + const ContextPtr & context, + const StorageID & time_series_storage_id, + const ColumnsDescription & time_series_columns, + const TimeSeriesSettings & time_series_settings, + LoadingStrictnessLevel mode) + { + StorageID target_table_id = StorageID::createEmpty(); + + bool is_external_target = target_info && !target_info->table_id.empty(); + if (is_external_target) + { + /// A target table is specified. + target_table_id = target_info->table_id; + + if (mode < LoadingStrictnessLevel::ATTACH) + { + /// If it's not an ATTACH request then + /// check that the specified target table has all the required columns. + auto target_table = DatabaseCatalog::instance().getTable(target_table_id, context); + auto target_metadata = target_table->getInMemoryMetadataPtr(); + const auto & target_columns = target_metadata->columns; + TimeSeriesColumnsValidator validator{time_series_storage_id, time_series_settings}; + validator.validateTargetColumns(kind, target_table_id, target_columns); + } + } + else + { + TimeSeriesInnerTablesCreator inner_tables_creator{context, time_series_storage_id, time_series_columns, time_series_settings}; + auto inner_uuid = target_info ? target_info->inner_uuid : UUIDHelpers::Nil; + + /// An inner target table should be used. + if (mode >= LoadingStrictnessLevel::ATTACH) + { + /// If it's an ATTACH request, then the inner target table must be already created. + target_table_id = inner_tables_creator.getInnerTableID(kind, inner_uuid); + } + else + { + /// Create the inner target table. + auto inner_table_engine = target_info ? target_info->inner_engine : nullptr; + target_table_id = inner_tables_creator.createInnerTable(kind, inner_uuid, inner_table_engine); + } + } + + return target_table_id; + } +} + + +void StorageTimeSeries::normalizeTableDefinition(ASTCreateQuery & create_query, const ContextPtr & local_context) +{ + StorageID time_series_storage_id{create_query.getDatabase(), create_query.getTable()}; + TimeSeriesSettings time_series_settings; + if (create_query.storage) + time_series_settings.loadFromQuery(*create_query.storage); + std::shared_ptr as_create_query; + if (!create_query.as_table.empty()) + { + auto as_database = local_context->resolveDatabase(create_query.as_database); + as_create_query = typeid_cast>( + DatabaseCatalog::instance().getDatabase(as_database)->getCreateTableQuery(create_query.as_table, local_context)); + } + TimeSeriesDefinitionNormalizer normalizer{time_series_storage_id, time_series_settings, as_create_query.get()}; + normalizer.normalize(create_query); +} + + +StorageTimeSeries::StorageTimeSeries( + const StorageID & table_id, + const ContextPtr & local_context, + LoadingStrictnessLevel mode, + const ASTCreateQuery & query, + const ColumnsDescription & columns, + const String & comment) + : IStorage(table_id) + , WithContext(local_context->getGlobalContext()) +{ + if (mode <= LoadingStrictnessLevel::CREATE && !local_context->getSettingsRef().allow_experimental_time_series_table) + { + throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, + "Experimental TimeSeries table engine " + "is not enabled (the setting 'allow_experimental_time_series_table')"); + } + + storage_settings = getTimeSeriesSettingsFromQuery(query); + + if (mode < LoadingStrictnessLevel::ATTACH) + { + TimeSeriesColumnsValidator validator{table_id, *storage_settings}; + validator.validateColumns(columns); + } + + StorageInMemoryMetadata storage_metadata; + storage_metadata.setColumns(columns); + if (!comment.empty()) + storage_metadata.setComment(comment); + setInMemoryMetadata(storage_metadata); + + has_inner_tables = false; + + for (auto target_kind : {ViewTarget::Data, ViewTarget::Tags, ViewTarget::Metrics}) + { + const ViewTarget * target_info = query.targets ? query.targets->tryGetTarget(target_kind) : nullptr; + auto & target = targets.emplace_back(); + target.kind = target_kind; + target.table_id = initTarget(target_kind, target_info, local_context, getStorageID(), columns, *storage_settings, mode); + target.is_inner_table = target_info && target_info->table_id.empty(); + + if (target_kind == ViewTarget::Metrics && !target.is_inner_table) + { + auto table = DatabaseCatalog::instance().tryGetTable(target.table_id, getContext()); + auto metadata = table->getInMemoryMetadataPtr(); + + for (const auto & column : metadata->columns) + if (column.type->lowCardinality()) + throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, "External metrics table cannot have LowCardnality columns for now."); + } + + has_inner_tables |= target.is_inner_table; + } +} + + +StorageTimeSeries::~StorageTimeSeries() = default; + + +TimeSeriesSettings StorageTimeSeries::getStorageSettings() const +{ + return *getStorageSettingsPtr(); +} + +void StorageTimeSeries::startup() +{ +} + +void StorageTimeSeries::shutdown(bool) +{ +} + + +void StorageTimeSeries::drop() +{ + /// Sync flag and the setting make sense for Atomic databases only. + /// However, with Atomic databases, IStorage::drop() can be called only from a background task in DatabaseCatalog. + /// Running synchronous DROP from that task leads to deadlock. + dropInnerTableIfAny(/* sync= */ false, getContext()); +} + +void StorageTimeSeries::dropInnerTableIfAny(bool sync, ContextPtr local_context) +{ + if (!has_inner_tables) + return; + + for (const auto & target : targets) + { + if (target.is_inner_table && DatabaseCatalog::instance().tryGetTable(target.table_id, getContext())) + { + /// Best-effort to make them work: the inner table name is almost always less than the TimeSeries name (so it's safe to lock DDLGuard). + /// (See the comment in StorageMaterializedView::dropInnerTableIfAny.) + bool may_lock_ddl_guard = getStorageID().getQualifiedName() < target.table_id.getQualifiedName(); + InterpreterDropQuery::executeDropQuery(ASTDropQuery::Kind::Drop, getContext(), local_context, target.table_id, + sync, /* ignore_sync_setting= */ true, may_lock_ddl_guard); + } + } +} + +void StorageTimeSeries::truncate(const ASTPtr &, const StorageMetadataPtr &, ContextPtr local_context, TableExclusiveLockHolder &) +{ + if (!has_inner_tables) + return; + + for (const auto & target : targets) + { + /// We truncate only inner tables here. + if (target.is_inner_table) + InterpreterDropQuery::executeDropQuery(ASTDropQuery::Kind::Truncate, getContext(), local_context, target.table_id, /* sync= */ true); + } +} + + +StorageID StorageTimeSeries::getTargetTableId(ViewTarget::Kind target_kind) const +{ + for (const auto & target : targets) + { + if (target.kind == target_kind) + return target.table_id; + } + throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected target kind {}", toString(target_kind)); +} + +StoragePtr StorageTimeSeries::getTargetTable(ViewTarget::Kind target_kind, const ContextPtr & local_context) const +{ + return DatabaseCatalog::instance().getTable(getTargetTableId(target_kind), local_context); +} + +StoragePtr StorageTimeSeries::tryGetTargetTable(ViewTarget::Kind target_kind, const ContextPtr & local_context) const +{ + return DatabaseCatalog::instance().tryGetTable(getTargetTableId(target_kind), local_context); +} + + +std::optional StorageTimeSeries::totalRows(const Settings & settings) const +{ + UInt64 total_rows = 0; + if (has_inner_tables) + { + for (const auto & target : targets) + { + if (target.is_inner_table) + { + auto inner_table = DatabaseCatalog::instance().tryGetTable(target.table_id, getContext()); + if (!inner_table) + return std::nullopt; + + auto total_rows_in_inner_table = inner_table->totalRows(settings); + if (!total_rows_in_inner_table) + return std::nullopt; + + total_rows += *total_rows_in_inner_table; + } + } + } + return total_rows; +} + +std::optional StorageTimeSeries::totalBytes(const Settings & settings) const +{ + UInt64 total_bytes = 0; + if (has_inner_tables) + { + for (const auto & target : targets) + { + if (target.is_inner_table) + { + auto inner_table = DatabaseCatalog::instance().tryGetTable(target.table_id, getContext()); + if (!inner_table) + return std::nullopt; + + auto total_bytes_in_inner_table = inner_table->totalBytes(settings); + if (!total_bytes_in_inner_table) + return std::nullopt; + + total_bytes += *total_bytes_in_inner_table; + } + } + } + return total_bytes; +} + +std::optional StorageTimeSeries::totalBytesUncompressed(const Settings & settings) const +{ + UInt64 total_bytes = 0; + if (has_inner_tables) + { + for (const auto & target : targets) + { + if (target.is_inner_table) + { + auto inner_table = DatabaseCatalog::instance().tryGetTable(target.table_id, getContext()); + if (!inner_table) + return std::nullopt; + + auto total_bytes_in_inner_table = inner_table->totalBytesUncompressed(settings); + if (!total_bytes_in_inner_table) + return std::nullopt; + + total_bytes += *total_bytes_in_inner_table; + } + } + } + return total_bytes; +} + +Strings StorageTimeSeries::getDataPaths() const +{ + Strings data_paths; + for (const auto & target : targets) + { + auto table = DatabaseCatalog::instance().tryGetTable(target.table_id, getContext()); + if (!table) + continue; + + insertAtEnd(data_paths, table->getDataPaths()); + } + return data_paths; +} + + +bool StorageTimeSeries::optimize( + const ASTPtr & query, + const StorageMetadataPtr &, + const ASTPtr & partition, + bool final, + bool deduplicate, + const Names & deduplicate_by_columns, + bool cleanup, + ContextPtr local_context) +{ + if (!has_inner_tables) + { + throw Exception(ErrorCodes::INCORRECT_QUERY, "TimeSeries table {} targets only existing tables. Execute the statement directly on it.", + getStorageID().getNameForLogs()); + } + + bool optimized = false; + for (const auto & target : targets) + { + if (target.is_inner_table) + { + auto inner_table = DatabaseCatalog::instance().getTable(target.table_id, local_context); + optimized |= inner_table->optimize(query, inner_table->getInMemoryMetadataPtr(), partition, final, deduplicate, deduplicate_by_columns, cleanup, local_context); + } + } + + return optimized; +} + + +void StorageTimeSeries::checkAlterIsPossible(const AlterCommands & commands, ContextPtr) const +{ + for (const auto & command : commands) + { + if (!command.isCommentAlter() && command.type != AlterCommand::MODIFY_SQL_SECURITY) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Alter of type '{}' is not supported by storage {}", command.type, getName()); + } +} + +void StorageTimeSeries::alter(const AlterCommands & params, ContextPtr local_context, AlterLockHolder & table_lock_holder) +{ + IStorage::alter(params, local_context, table_lock_holder); +} + + +void StorageTimeSeries::renameInMemory(const StorageID & /* new_table_id */) +{ + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Renaming is not supported by storage {} yet", getName()); +} + + +void StorageTimeSeries::backupData(BackupEntriesCollector & backup_entries_collector, const String & data_path_in_backup, const std::optional &) +{ + for (const auto & target : targets) + { + /// We backup the target table's data only if it's inner. + if (target.is_inner_table) + { + auto table = DatabaseCatalog::instance().getTable(target.table_id, getContext()); + table->backupData(backup_entries_collector, fs::path{data_path_in_backup} / toString(target.kind), {}); + } + } +} + +void StorageTimeSeries::restoreDataFromBackup(RestorerFromBackup & restorer, const String & data_path_in_backup, const std::optional &) +{ + for (const auto & target : targets) + { + /// We backup the target table's data only if it's inner. + if (target.is_inner_table) + { + auto table = DatabaseCatalog::instance().getTable(target.table_id, getContext()); + table->restoreDataFromBackup(restorer, fs::path{data_path_in_backup} / toString(target.kind), {}); + } + } +} + + +void StorageTimeSeries::read( + QueryPlan & /* query_plan */, + const Names & /* column_names */, + const StorageSnapshotPtr & /* storage_snapshot */, + SelectQueryInfo & /* query_info */, + ContextPtr /* local_context */, + QueryProcessingStage::Enum /* processed_stage */, + size_t /* max_block_size */, + size_t /* num_streams */) +{ + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "SELECT is not supported by storage {} yet", getName()); +} + + +SinkToStoragePtr StorageTimeSeries::write( + const ASTPtr & /* query */, const StorageMetadataPtr & /* metadata_snapshot */, ContextPtr /* local_context */, bool /* async_insert */) +{ + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "INSERT is not supported by storage {} yet", getName()); +} + + +std::shared_ptr storagePtrToTimeSeries(StoragePtr storage) +{ + if (auto res = typeid_cast>(storage)) + return res; + + throw Exception( + ErrorCodes::UNEXPECTED_TABLE_ENGINE, + "This operation can be executed on a TimeSeries table only, the engine of table {} is not TimeSeries", + storage->getStorageID().getNameForLogs()); +} + +std::shared_ptr storagePtrToTimeSeries(ConstStoragePtr storage) +{ + if (auto res = typeid_cast>(storage)) + return res; + + throw Exception( + ErrorCodes::UNEXPECTED_TABLE_ENGINE, + "This operation can be executed on a TimeSeries table only, the engine of table {} is not TimeSeries", + storage->getStorageID().getNameForLogs()); +} + + +void registerStorageTimeSeries(StorageFactory & factory) +{ + factory.registerStorage("TimeSeries", [](const StorageFactory::Arguments & args) + { + /// Pass local_context here to convey setting to inner tables. + return std::make_shared( + args.table_id, args.getLocalContext(), args.mode, args.query, args.columns, args.comment); + } + , + { + .supports_settings = true, + .supports_schema_inference = true, + }); +} + +} diff --git a/src/Storages/StorageTimeSeries.h b/src/Storages/StorageTimeSeries.h new file mode 100644 index 00000000000..35db3131a0b --- /dev/null +++ b/src/Storages/StorageTimeSeries.h @@ -0,0 +1,111 @@ +#pragma once + +#include +#include +#include +#include + + +namespace DB +{ +struct TimeSeriesSettings; +using TimeSeriesSettingsPtr = std::shared_ptr; + +/// Represents a table engine to keep time series received by Prometheus protocols. +/// Examples of using this table engine: +/// +/// CREATE TABLE ts ENGINE = TimeSeries() +/// -OR- +/// CREATE TABLE ts ENGINE = TimeSeries() DATA [db].table1 TAGS [db].table2 METRICS [db].table3 +/// -OR- +/// CREATE TABLE ts ENGINE = TimeSeries() DATA ENGINE = MergeTree TAGS ENGINE = ReplacingMergeTree METRICS ENGINE = ReplacingMergeTree +/// -OR- +/// CREATE TABLE ts ( +/// id UUID DEFAULT reinterpretAsUUID(sipHash128(metric_name, all_tags)) CODEC(ZSTD(3)), +/// instance LowCardinality(String), +/// job String +/// ) ENGINE = TimeSeries() +/// SETTINGS tags_to_columns = {'instance': 'instance', 'job': 'job'} +/// DATA ENGINE = ReplicatedMergeTree('zkpath', 'replica'), ... +/// +class StorageTimeSeries final : public IStorage, WithContext +{ +public: + /// Adds missing columns and reorder columns, and also adds inner table engines if they aren't specified. + static void normalizeTableDefinition(ASTCreateQuery & create_query, const ContextPtr & local_context); + + StorageTimeSeries(const StorageID & table_id, const ContextPtr & local_context, LoadingStrictnessLevel mode, + const ASTCreateQuery & query, const ColumnsDescription & columns, const String & comment); + + ~StorageTimeSeries() override; + + std::string getName() const override { return "TimeSeries"; } + + TimeSeriesSettings getStorageSettings() const; + TimeSeriesSettingsPtr getStorageSettingsPtr() const { return storage_settings; } + + StorageID getTargetTableId(ViewTarget::Kind target_kind) const; + StoragePtr getTargetTable(ViewTarget::Kind target_kind, const ContextPtr & local_context) const; + StoragePtr tryGetTargetTable(ViewTarget::Kind target_kind, const ContextPtr & local_context) const; + + void startup() override; + void shutdown(bool is_drop) override; + + 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; + + SinkToStoragePtr write(const ASTPtr & query, const StorageMetadataPtr & metadata_snapshot, ContextPtr context, bool async_insert) override; + + bool optimize( + const ASTPtr & query, + const StorageMetadataPtr & metadata_snapshot, + const ASTPtr & partition, + bool final, + bool deduplicate, + const Names & deduplicate_by_columns, + bool cleanup, + ContextPtr local_context) override; + + void drop() override; + void dropInnerTableIfAny(bool sync, ContextPtr local_context) override; + + void truncate(const ASTPtr &, const StorageMetadataPtr &, ContextPtr, TableExclusiveLockHolder &) override; + + void renameInMemory(const StorageID & new_table_id) override; + + void checkAlterIsPossible(const AlterCommands & commands, ContextPtr local_context) const override; + void alter(const AlterCommands & params, ContextPtr local_context, AlterLockHolder & table_lock_holder) override; + + void backupData(BackupEntriesCollector & backup_entries_collector, const String & data_path_in_backup, const std::optional & partitions) override; + void restoreDataFromBackup(RestorerFromBackup & restorer, const String & data_path_in_backup, const std::optional & partitions) override; + + std::optional totalRows(const Settings & settings) const override; + std::optional totalBytes(const Settings & settings) const override; + std::optional totalBytesUncompressed(const Settings & settings) const override; + Strings getDataPaths() const override; + +private: + TimeSeriesSettingsPtr storage_settings; + + struct Target + { + ViewTarget::Kind kind; + StorageID table_id = StorageID::createEmpty(); + bool is_inner_table; + }; + + std::vector targets; + bool has_inner_tables; +}; + +std::shared_ptr storagePtrToTimeSeries(StoragePtr storage); +std::shared_ptr storagePtrToTimeSeries(ConstStoragePtr storage); + +} diff --git a/src/Storages/StorageURL.cpp b/src/Storages/StorageURL.cpp index 362aaf1363c..6442891cf23 100644 --- a/src/Storages/StorageURL.cpp +++ b/src/Storages/StorageURL.cpp @@ -36,6 +36,8 @@ #include #include #include + +#include #include #include #include @@ -90,11 +92,22 @@ static const std::vector> optional_regex_keys = { std::make_shared(R"(headers.header\[[0-9]*\].value)"), }; -static bool urlWithGlobs(const String & uri) +bool urlWithGlobs(const String & uri) { return (uri.find('{') != std::string::npos && uri.find('}') != std::string::npos) || uri.find('|') != std::string::npos; } +String getSampleURI(String uri, ContextPtr context) +{ + if (urlWithGlobs(uri)) + { + auto uris = parseRemoteDescription(uri, 0, uri.size(), ',', context->getSettingsRef().glob_expansion_max_elements); + if (!uris.empty()) + return uris[0]; + } + return uri; +} + static ConnectionTimeouts getHTTPTimeouts(ContextPtr context) { return ConnectionTimeouts::getHTTPTimeouts(context->getSettingsRef(), context->getServerSettings().keep_alive_timeout); @@ -153,7 +166,8 @@ IStorageURLBase::IStorageURLBase( storage_metadata.setConstraints(constraints_); storage_metadata.setComment(comment); setInMemoryMetadata(storage_metadata); - setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(storage_metadata.getColumns())); + + setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(storage_metadata.getColumns(), context_, getSampleURI(uri, context_), format_settings)); } @@ -414,13 +428,14 @@ Chunk StorageURLSource::generate() size_t chunk_size = 0; if (input_format) chunk_size = input_format->getApproxBytesReadForChunk(); + progress(num_rows, chunk_size ? chunk_size : chunk.bytes()); VirtualColumnUtils::addRequestedFileLikeStorageVirtualsToChunk( chunk, requested_virtual_columns, { .path = curr_uri.getPath(), - .size = current_file_size - }); + .size = current_file_size, + }, getContext(), columns_description); return chunk; } @@ -571,31 +586,15 @@ StorageURLSink::StorageURLSink( void StorageURLSink::consume(Chunk & chunk) { - std::lock_guard lock(cancel_mutex); - if (cancelled) + if (isCancelled()) return; writer->write(getHeader().cloneWithColumns(chunk.getColumns())); } -void StorageURLSink::onCancel() -{ - std::lock_guard lock(cancel_mutex); - cancelBuffers(); - releaseBuffers(); - cancelled = true; -} - -void StorageURLSink::onException(std::exception_ptr) -{ - std::lock_guard lock(cancel_mutex); - cancelBuffers(); - releaseBuffers(); -} - void StorageURLSink::onFinish() { - std::lock_guard lock(cancel_mutex); finalizeBuffers(); + releaseBuffers(); } void StorageURLSink::finalizeBuffers() @@ -855,7 +854,7 @@ namespace format = format_name; } - String getLastFileName() const override { return current_url_option; } + String getLastFilePath() const override { return current_url_option; } bool supportsLastReadBufferRecreation() const override { return true; } @@ -1176,6 +1175,7 @@ void ReadFromURL::createIterator(const ActionsDAG::Node * predicate) void ReadFromURL::initializePipeline(QueryPipelineBuilder & pipeline, const BuildQueryPipelineSettings &) { createIterator(nullptr); + const auto & settings = context->getSettingsRef(); if (is_empty_glob) { @@ -1186,7 +1186,6 @@ void ReadFromURL::initializePipeline(QueryPipelineBuilder & pipeline, const Buil Pipes pipes; pipes.reserve(num_streams); - const auto & settings = context->getSettingsRef(); const size_t max_parsing_threads = num_streams >= settings.max_parsing_threads ? 1 : (settings.max_parsing_threads / num_streams); for (size_t i = 0; i < num_streams; ++i) @@ -1402,6 +1401,11 @@ StorageURLWithFailover::StorageURLWithFailover( } } +StorageURLSink::~StorageURLSink() +{ + if (isCancelled()) + cancelBuffers(); +} FormatSettings StorageURL::getFormatSettingsFromArgs(const StorageFactory::Arguments & args) { @@ -1592,4 +1596,5 @@ void registerStorageURL(StorageFactory & factory) .source_access_type = AccessType::URL, }); } + } diff --git a/src/Storages/StorageURL.h b/src/Storages/StorageURL.h index 6c95cad474c..19daf843431 100644 --- a/src/Storages/StorageURL.h +++ b/src/Storages/StorageURL.h @@ -141,6 +141,9 @@ private: virtual Block getHeaderBlock(const Names & column_names, const StorageSnapshotPtr & storage_snapshot) const = 0; }; +bool urlWithGlobs(const String & uri); + +String getSampleURI(String uri, ContextPtr context); class StorageURLSource : public SourceWithKeyCondition, WithContext { @@ -250,10 +253,10 @@ public: const HTTPHeaderEntries & headers = {}, const String & method = Poco::Net::HTTPRequest::HTTP_POST); + ~StorageURLSink() override; + std::string getName() const override { return "StorageURLSink"; } void consume(Chunk & chunk) override; - void onCancel() override; - void onException(std::exception_ptr exception) override; void onFinish() override; private: @@ -263,8 +266,6 @@ private: std::unique_ptr write_buf; OutputFormatPtr writer; - std::mutex cancel_mutex; - bool cancelled = false; }; class StorageURL : public IStorageURLBase diff --git a/src/Storages/StorageURLCluster.cpp b/src/Storages/StorageURLCluster.cpp index 69041482fd8..7c7a299c64e 100644 --- a/src/Storages/StorageURLCluster.cpp +++ b/src/Storages/StorageURLCluster.cpp @@ -76,7 +76,7 @@ StorageURLCluster::StorageURLCluster( storage_metadata.setConstraints(constraints_); setInMemoryMetadata(storage_metadata); - setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(storage_metadata.getColumns())); + setVirtuals(VirtualColumnUtils::getVirtualsForFileLikeStorage(storage_metadata.getColumns(), context, getSampleURI(uri, context))); } void StorageURLCluster::updateQueryToSendIfNeeded(ASTPtr & query, const StorageSnapshotPtr & storage_snapshot, const ContextPtr & context) diff --git a/src/Storages/System/IStorageSystemOneBlock.cpp b/src/Storages/System/IStorageSystemOneBlock.cpp index 456b7c4f90b..b8f32fcdb83 100644 --- a/src/Storages/System/IStorageSystemOneBlock.cpp +++ b/src/Storages/System/IStorageSystemOneBlock.cpp @@ -5,6 +5,7 @@ // #include #include #include +#include #include #include #include @@ -44,7 +45,7 @@ public: private: std::shared_ptr storage; std::vector columns_mask; - const ActionsDAG::Node * predicate = nullptr; + std::optional filter; }; void IStorageSystemOneBlock::read( @@ -79,8 +80,9 @@ void IStorageSystemOneBlock::read( void ReadFromSystemOneBlock::initializePipeline(QueryPipelineBuilder & pipeline, const BuildQueryPipelineSettings &) { - const auto & sample_block = getOutputStream().header; + const Block & sample_block = getOutputStream().header; MutableColumns res_columns = sample_block.cloneEmptyColumns(); + const ActionsDAG::Node * predicate = filter ? filter->getOutputs().at(0) : nullptr; storage->fillData(res_columns, context, predicate, std::move(columns_mask)); UInt64 num_rows = res_columns.at(0)->size(); @@ -93,8 +95,18 @@ void ReadFromSystemOneBlock::applyFilters(ActionDAGNodes added_filter_nodes) { SourceStepWithFilter::applyFilters(std::move(added_filter_nodes)); - if (filter_actions_dag) - predicate = filter_actions_dag->getOutputs().at(0); + if (!filter_actions_dag) + return; + + Block sample = storage->getFilterSampleBlock(); + if (sample.columns() == 0) + return; + + filter = VirtualColumnUtils::splitFilterDagForAllowedInputs(filter_actions_dag->getOutputs().at(0), &sample); + + /// Must prepare sets here, initializePipeline() would be too late, see comment on FutureSetFromSubquery. + if (filter) + VirtualColumnUtils::buildSetsForDAG(*filter, context); } } diff --git a/src/Storages/System/IStorageSystemOneBlock.h b/src/Storages/System/IStorageSystemOneBlock.h index a20434fd97e..a47875c2537 100644 --- a/src/Storages/System/IStorageSystemOneBlock.h +++ b/src/Storages/System/IStorageSystemOneBlock.h @@ -22,8 +22,16 @@ class Context; class IStorageSystemOneBlock : public IStorage { protected: + /// If this method uses `predicate`, getFilterSampleBlock() must list all columns to which + /// it's applied. (Otherwise there'll be a LOGICAL_ERROR "Not-ready Set is passed" on subqueries.) virtual void fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node * predicate, std::vector columns_mask) const = 0; + /// Columns to which fillData() applies the `predicate`. + virtual Block getFilterSampleBlock() const + { + return {}; + } + virtual bool supportsColumnsMask() const { return false; } friend class ReadFromSystemOneBlock; diff --git a/src/Storages/System/StorageSystemBuildOptions.cpp.in b/src/Storages/System/StorageSystemBuildOptions.cpp.in index a81bcb08bfc..9e5adbfe825 100644 --- a/src/Storages/System/StorageSystemBuildOptions.cpp.in +++ b/src/Storages/System/StorageSystemBuildOptions.cpp.in @@ -21,7 +21,7 @@ const char * auto_config_build[] "BUILD_COMPILE_DEFINITIONS", "@BUILD_COMPILE_DEFINITIONS@", "USE_EMBEDDED_COMPILER", "@USE_EMBEDDED_COMPILER@", "USE_GLIBC_COMPATIBILITY", "@GLIBC_COMPATIBILITY@", - "USE_JEMALLOC", "@ENABLE_JEMALLOC@", + "USE_JEMALLOC", "@USE_JEMALLOC@", "USE_ICU", "@USE_ICU@", "USE_H3", "@USE_H3@", "USE_MYSQL", "@USE_MYSQL@", @@ -36,7 +36,7 @@ const char * auto_config_build[] "USE_SSL", "@USE_SSL@", "OPENSSL_VERSION", "@OPENSSL_VERSION@", "OPENSSL_IS_BORING_SSL", "@OPENSSL_IS_BORING_SSL@", - "USE_VECTORSCAN", "@ENABLE_VECTORSCAN@", + "USE_VECTORSCAN", "@USE_VECTORSCAN@", "USE_SIMDJSON", "@USE_SIMDJSON@", "USE_ODBC", "@USE_ODBC@", "USE_GRPC", "@USE_GRPC@", @@ -62,8 +62,8 @@ const char * auto_config_build[] "USE_ARROW", "@USE_ARROW@", "USE_ORC", "@USE_ORC@", "USE_MSGPACK", "@USE_MSGPACK@", - "USE_QPL", "@ENABLE_QPL@", - "USE_QAT", "@ENABLE_QATLIB@", + "USE_QPL", "@USE_QPL@", + "USE_QATLIB", "@USE_QATLIB@", "GIT_HASH", "@GIT_HASH@", "GIT_BRANCH", R"IRjaNsZIL9Yh7FQ4(@GIT_BRANCH@)IRjaNsZIL9Yh7FQ4", "GIT_DATE", "@GIT_DATE@", diff --git a/src/Storages/System/StorageSystemClusters.cpp b/src/Storages/System/StorageSystemClusters.cpp index 9c5c07ae49f..9493d2c97ab 100644 --- a/src/Storages/System/StorageSystemClusters.cpp +++ b/src/Storages/System/StorageSystemClusters.cpp @@ -1,3 +1,4 @@ +#include #include #include #include @@ -31,6 +32,8 @@ ColumnsDescription StorageSystemClusters::getColumnsDescription() {"database_shard_name", std::make_shared(), "The name of the `Replicated` database shard (for clusters that belong to a `Replicated` database)."}, {"database_replica_name", std::make_shared(), "The name of the `Replicated` database replica (for clusters that belong to a `Replicated` database)."}, {"is_active", std::make_shared(std::make_shared()), "The status of the Replicated database replica (for clusters that belong to a Replicated database): 1 means 'replica is online', 0 means 'replica is offline', NULL means 'unknown'."}, + {"replication_lag", std::make_shared(std::make_shared()), "The replication lag of the `Replicated` database replica (for clusters that belong to a Replicated database)."}, + {"recovery_time", std::make_shared(std::make_shared()), "The recovery time of the `Replicated` database replica (for clusters that belong to a Replicated database), in milliseconds."}, }; description.setAliases({ @@ -67,6 +70,11 @@ void StorageSystemClusters::writeCluster(MutableColumns & res_columns, const std const auto & shards_info = cluster->getShardsInfo(); const auto & addresses_with_failover = cluster->getShardsAddresses(); + size_t recovery_time_column_idx = columns_mask.size() - 1, replication_lag_column_idx = columns_mask.size() - 2, is_active_column_idx = columns_mask.size() - 3; + ReplicasInfo replicas_info; + if (replicated && (columns_mask[recovery_time_column_idx] || columns_mask[replication_lag_column_idx] || columns_mask[is_active_column_idx])) + replicas_info = replicated->tryGetReplicasInfo(name_and_cluster.second); + size_t replica_idx = 0; for (size_t shard_index = 0; shard_index < shards_info.size(); ++shard_index) { @@ -114,17 +122,46 @@ void StorageSystemClusters::writeCluster(MutableColumns & res_columns, const std res_columns[res_index++]->insert(address.database_shard_name); if (columns_mask[src_index++]) res_columns[res_index++]->insert(address.database_replica_name); + + /// make sure these three columns remain the last ones if (columns_mask[src_index++]) { - std::vector is_active; - if (replicated) - is_active = replicated->tryGetAreReplicasActive(name_and_cluster.second); - - if (is_active.empty()) + if (replicas_info.empty()) res_columns[res_index++]->insertDefault(); else - res_columns[res_index++]->insert(is_active[replica_idx++]); + { + const auto & replica_info = replicas_info[replica_idx]; + res_columns[res_index++]->insert(replica_info.is_active); + } } + if (columns_mask[src_index++]) + { + if (replicas_info.empty()) + res_columns[res_index++]->insertDefault(); + else + { + const auto & replica_info = replicas_info[replica_idx]; + if (replica_info.replication_lag != std::nullopt) + res_columns[res_index++]->insert(*replica_info.replication_lag); + else + res_columns[res_index++]->insertDefault(); + } + } + if (columns_mask[src_index++]) + { + if (replicas_info.empty()) + res_columns[res_index++]->insertDefault(); + else + { + const auto & replica_info = replicas_info[replica_idx]; + if (replica_info.recovery_time != 0) + res_columns[res_index++]->insert(replica_info.recovery_time); + else + res_columns[res_index++]->insertDefault(); + } + } + + ++replica_idx; } } } diff --git a/src/Storages/System/StorageSystemClusters.h b/src/Storages/System/StorageSystemClusters.h index f6adb902f43..a5f6d551ca1 100644 --- a/src/Storages/System/StorageSystemClusters.h +++ b/src/Storages/System/StorageSystemClusters.h @@ -1,10 +1,10 @@ #pragma once +#include #include #include #include - namespace DB { diff --git a/src/Storages/System/StorageSystemColumns.cpp b/src/Storages/System/StorageSystemColumns.cpp index 6a7810b97f9..bc13cb77d5e 100644 --- a/src/Storages/System/StorageSystemColumns.cpp +++ b/src/Storages/System/StorageSystemColumns.cpp @@ -105,8 +105,8 @@ protected: while (rows_count < max_block_size && db_table_num < total_tables) { - const std::string database_name = (*databases)[db_table_num].get(); - const std::string table_name = (*tables)[db_table_num].get(); + const std::string database_name = (*databases)[db_table_num].safeGet(); + const std::string table_name = (*tables)[db_table_num].safeGet(); ++db_table_num; ColumnsDescription columns; @@ -338,7 +338,7 @@ private: std::shared_ptr storage; std::vector columns_mask; const size_t max_block_size; - const ActionsDAG::Node * predicate = nullptr; + std::optional virtual_columns_filter; }; void ReadFromSystemColumns::applyFilters(ActionDAGNodes added_filter_nodes) @@ -346,7 +346,17 @@ void ReadFromSystemColumns::applyFilters(ActionDAGNodes added_filter_nodes) SourceStepWithFilter::applyFilters(std::move(added_filter_nodes)); if (filter_actions_dag) - predicate = filter_actions_dag->getOutputs().at(0); + { + Block block_to_filter; + block_to_filter.insert(ColumnWithTypeAndName(ColumnString::create(), std::make_shared(), "database")); + block_to_filter.insert(ColumnWithTypeAndName(ColumnString::create(), std::make_shared(), "table")); + + virtual_columns_filter = VirtualColumnUtils::splitFilterDagForAllowedInputs(filter_actions_dag->getOutputs().at(0), &block_to_filter); + + /// Must prepare sets here, initializePipeline() would be too late, see comment on FutureSetFromSubquery. + if (virtual_columns_filter) + VirtualColumnUtils::buildSetsForDAG(*virtual_columns_filter, context); + } } void StorageSystemColumns::read( @@ -408,7 +418,8 @@ void ReadFromSystemColumns::initializePipeline(QueryPipelineBuilder & pipeline, block_to_filter.insert(ColumnWithTypeAndName(std::move(database_column_mut), std::make_shared(), "database")); /// Filter block with `database` column. - VirtualColumnUtils::filterBlockWithPredicate(predicate, block_to_filter, context); + if (virtual_columns_filter) + VirtualColumnUtils::filterBlockWithPredicate(virtual_columns_filter->getOutputs().at(0), block_to_filter, context); if (!block_to_filter.rows()) { @@ -426,7 +437,7 @@ void ReadFromSystemColumns::initializePipeline(QueryPipelineBuilder & pipeline, for (size_t i = 0; i < num_databases; ++i) { - const std::string database_name = (*database_column)[i].get(); + const std::string database_name = (*database_column)[i].safeGet(); if (database_name.empty()) { for (auto & [table_name, table] : external_tables) @@ -456,7 +467,8 @@ void ReadFromSystemColumns::initializePipeline(QueryPipelineBuilder & pipeline, } /// Filter block with `database` and `table` columns. - VirtualColumnUtils::filterBlockWithPredicate(predicate, block_to_filter, context); + if (virtual_columns_filter) + VirtualColumnUtils::filterBlockWithPredicate(virtual_columns_filter->getOutputs().at(0), block_to_filter, context); if (!block_to_filter.rows()) { diff --git a/src/Storages/System/StorageSystemContributors.generated.cpp b/src/Storages/System/StorageSystemContributors.generated.cpp index 35b9c0008c6..67dfe3bfe86 100644 --- a/src/Storages/System/StorageSystemContributors.generated.cpp +++ b/src/Storages/System/StorageSystemContributors.generated.cpp @@ -457,6 +457,7 @@ const char * auto_contributors[] { "Gleb-Tretyakov", "GoGoWen2021", "Gosha Letov", + "Graham Campbell", "Gregory", "Grigorii Sokolik", "Grigory", @@ -472,6 +473,7 @@ const char * auto_contributors[] { "Habibullah Oladepo", "HaiBo Li", "Hakob Saghatelyan", + "Halersson Paris", "Hamoon", "Han Fei", "Han Shukai", @@ -541,6 +543,7 @@ const char * auto_contributors[] { "JackyWoo", "Jacob Hayes", "Jacob Herrington", + "Jacob Reckhard", "Jai Jhala", "Jake Bamrah", "Jake Liu", @@ -661,6 +664,7 @@ const char * auto_contributors[] { "LaurieLY", "Lee sungju", "Lemore", + "Lennard Eijsackers", "Leonardo Cecchi", "Leonardo Maciel", "Leonid Krylov", @@ -922,6 +926,7 @@ const char * auto_contributors[] { "Pervakov Grigorii", "Pervakov Grigory", "Peter", + "Peter Nguyen", "Petr Vasilev", "Pham Anh Tuan", "Philip Hallstrom", @@ -981,6 +986,7 @@ const char * auto_contributors[] { "Ronald Bradford", "Rory Crispin", "Roy Bellingan", + "Ruihang Xia", "Ruslan", "Ruslan Mardugalliamov", "Ruslan Savchenko", @@ -1000,9 +1006,11 @@ const char * auto_contributors[] { "Sami Kerola", "Samuel Chou", "Samuel Colvin", + "Samuele Guerrini", "San", "Sanjam Panda", "Sariel", + "Sasha Sheikin", "Saulius Valatka", "Sean Haynes", "Sean Lafferty", @@ -1202,6 +1210,7 @@ const char * auto_contributors[] { "Vladimir Makarov", "Vladimir Mihailenco", "Vladimir Smirnov", + "Vladimir Varankin", "Vladislav Rassokhin", "Vladislav Smirnov", "Vladislav V", @@ -1275,6 +1284,7 @@ const char * auto_contributors[] { "Zhichun Wu", "Zhiguo Zhou", "Zhipeng", + "Zhukova, Maria", "Zhuo Qiu", "Zijie Lu", "Zimu Li", @@ -1502,6 +1512,7 @@ const char * auto_contributors[] { "hchen9", "hcz", "hdhoang", + "heguangnan", "heleihelei", "helifu", "hendrik-m", @@ -1572,6 +1583,7 @@ const char * auto_contributors[] { "kevinyhzou", "kgurjev", "khamadiev", + "khodyrevyurii", "kigerzhang", "kirillikoff", "kmeaw", @@ -1787,6 +1799,7 @@ const char * auto_contributors[] { "ruslandoga", "ryzuo", "s-kat", + "sakulali", "sanjam", "santaux", "santrancisco", @@ -1804,6 +1817,7 @@ const char * auto_contributors[] { "shabroo", "shangshujie", "shedx", + "shiyer7474", "shuai-xu", "shuchaome", "shuyang", @@ -1901,6 +1915,7 @@ const char * auto_contributors[] { "wzl", "xPoSx", "xbthink", + "xc0derx", "xiao", "xiaolei565", "xiebin", @@ -1964,6 +1979,7 @@ const char * auto_contributors[] { "zkun", "zlx19950903", "zombee0", + "zoomxi", "zvonand", "zvrr", "zvvr", diff --git a/src/Storages/System/StorageSystemDataSkippingIndices.cpp b/src/Storages/System/StorageSystemDataSkippingIndices.cpp index 093adc59cc6..a41771df406 100644 --- a/src/Storages/System/StorageSystemDataSkippingIndices.cpp +++ b/src/Storages/System/StorageSystemDataSkippingIndices.cpp @@ -214,7 +214,7 @@ private: std::shared_ptr storage; std::vector columns_mask; const size_t max_block_size; - const ActionsDAG::Node * predicate = nullptr; + ExpressionActionsPtr virtual_columns_filter; }; void ReadFromSystemDataSkippingIndices::applyFilters(ActionDAGNodes added_filter_nodes) @@ -222,7 +222,16 @@ void ReadFromSystemDataSkippingIndices::applyFilters(ActionDAGNodes added_filter SourceStepWithFilter::applyFilters(std::move(added_filter_nodes)); if (filter_actions_dag) - predicate = filter_actions_dag->getOutputs().at(0); + { + Block block_to_filter + { + { ColumnString::create(), std::make_shared(), "database" }, + }; + + auto dag = VirtualColumnUtils::splitFilterDagForAllowedInputs(filter_actions_dag->getOutputs().at(0), &block_to_filter); + if (dag) + virtual_columns_filter = VirtualColumnUtils::buildFilterExpression(std::move(*dag), context); + } } void StorageSystemDataSkippingIndices::read( @@ -268,7 +277,8 @@ void ReadFromSystemDataSkippingIndices::initializePipeline(QueryPipelineBuilder /// Condition on "database" in a query acts like an index. Block block { ColumnWithTypeAndName(std::move(column), std::make_shared(), "database") }; - VirtualColumnUtils::filterBlockWithPredicate(predicate, block, context); + if (virtual_columns_filter) + VirtualColumnUtils::filterBlockWithExpression(virtual_columns_filter, block); ColumnPtr & filtered_databases = block.getByPosition(0).column; pipeline.init(Pipe(std::make_shared( diff --git a/src/Storages/System/StorageSystemDatabases.cpp b/src/Storages/System/StorageSystemDatabases.cpp index 1dbb187c418..0585506a661 100644 --- a/src/Storages/System/StorageSystemDatabases.cpp +++ b/src/Storages/System/StorageSystemDatabases.cpp @@ -73,6 +73,14 @@ static String getEngineFull(const ContextPtr & ctx, const DatabasePtr & database return engine_full; } +Block StorageSystemDatabases::getFilterSampleBlock() const +{ + return { + { {}, std::make_shared(), "engine" }, + { {}, std::make_shared(), "uuid" }, + }; +} + static ColumnPtr getFilteredDatabases(const Databases & databases, const ActionsDAG::Node * predicate, ContextPtr context) { MutableColumnPtr name_column = ColumnString::create(); diff --git a/src/Storages/System/StorageSystemDatabases.h b/src/Storages/System/StorageSystemDatabases.h index fa55f0aea32..d10b350435b 100644 --- a/src/Storages/System/StorageSystemDatabases.h +++ b/src/Storages/System/StorageSystemDatabases.h @@ -27,6 +27,7 @@ protected: bool supportsColumnsMask() const override { return true; } void fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node * predicate, std::vector columns_mask) const override; + Block getFilterSampleBlock() const override; }; } diff --git a/src/Storages/System/StorageSystemDistributionQueue.cpp b/src/Storages/System/StorageSystemDistributionQueue.cpp index e2058448904..dab318a9c1c 100644 --- a/src/Storages/System/StorageSystemDistributionQueue.cpp +++ b/src/Storages/System/StorageSystemDistributionQueue.cpp @@ -107,6 +107,13 @@ ColumnsDescription StorageSystemDistributionQueue::getColumnsDescription() }; } +Block StorageSystemDistributionQueue::getFilterSampleBlock() const +{ + return { + { {}, std::make_shared(), "database" }, + { {}, std::make_shared(), "table" }, + }; +} void StorageSystemDistributionQueue::fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node * predicate, std::vector) const { diff --git a/src/Storages/System/StorageSystemDistributionQueue.h b/src/Storages/System/StorageSystemDistributionQueue.h index 159a86bf082..27d777a4762 100644 --- a/src/Storages/System/StorageSystemDistributionQueue.h +++ b/src/Storages/System/StorageSystemDistributionQueue.h @@ -22,6 +22,7 @@ protected: using IStorageSystemOneBlock::IStorageSystemOneBlock; void fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node * predicate, std::vector) const override; + Block getFilterSampleBlock() const override; }; } diff --git a/src/Storages/System/StorageSystemKafkaConsumers.cpp b/src/Storages/System/StorageSystemKafkaConsumers.cpp index 86713632339..db6804d3ad7 100644 --- a/src/Storages/System/StorageSystemKafkaConsumers.cpp +++ b/src/Storages/System/StorageSystemKafkaConsumers.cpp @@ -42,7 +42,7 @@ ColumnsDescription StorageSystemKafkaConsumers::getColumnsDescription() {"num_rebalance_revocations", std::make_shared(), "Number of times the consumer was revoked its partitions."}, {"num_rebalance_assignments", std::make_shared(), "Number of times the consumer was assigned to Kafka cluster."}, {"is_currently_used", std::make_shared(), "The flag which shows whether the consumer is in use."}, - {"last_used", std::make_shared(6), "The last time this consumer was in use."}, + {"last_used", std::make_shared(), "The last time this consumer was in use, unix time in microseconds."}, {"rdkafka_stat", std::make_shared(), "Library internal statistic. Set statistics_interval_ms to 0 disable, default is 3000 (once in three seconds)."}, }; } @@ -79,7 +79,7 @@ void StorageSystemKafkaConsumers::fillData(MutableColumns & res_columns, Context auto & num_rebalance_revocations = assert_cast(*res_columns[index++]); auto & num_rebalance_assigments = assert_cast(*res_columns[index++]); auto & is_currently_used = assert_cast(*res_columns[index++]); - auto & last_used = assert_cast(*res_columns[index++]); + auto & last_used = assert_cast(*res_columns[index++]); auto & rdkafka_stat = assert_cast(*res_columns[index++]); const auto access = context->getAccess(); diff --git a/src/Storages/System/StorageSystemMutations.cpp b/src/Storages/System/StorageSystemMutations.cpp index 94656008029..df9a71310e5 100644 --- a/src/Storages/System/StorageSystemMutations.cpp +++ b/src/Storages/System/StorageSystemMutations.cpp @@ -46,6 +46,13 @@ ColumnsDescription StorageSystemMutations::getColumnsDescription() }; } +Block StorageSystemMutations::getFilterSampleBlock() const +{ + return { + { {}, std::make_shared(), "database" }, + { {}, std::make_shared(), "table" }, + }; +} void StorageSystemMutations::fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node * predicate, std::vector) const { diff --git a/src/Storages/System/StorageSystemMutations.h b/src/Storages/System/StorageSystemMutations.h index c60157cd853..5341838a65e 100644 --- a/src/Storages/System/StorageSystemMutations.h +++ b/src/Storages/System/StorageSystemMutations.h @@ -22,6 +22,7 @@ protected: using IStorageSystemOneBlock::IStorageSystemOneBlock; void fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node * predicate, std::vector) const override; + Block getFilterSampleBlock() const override; }; } diff --git a/src/Storages/System/StorageSystemPartMovesBetweenShards.cpp b/src/Storages/System/StorageSystemPartMovesBetweenShards.cpp index 9cba92bca12..ab74b205a96 100644 --- a/src/Storages/System/StorageSystemPartMovesBetweenShards.cpp +++ b/src/Storages/System/StorageSystemPartMovesBetweenShards.cpp @@ -43,6 +43,14 @@ ColumnsDescription StorageSystemPartMovesBetweenShards::getColumnsDescription() } +Block StorageSystemPartMovesBetweenShards::getFilterSampleBlock() const +{ + return { + { {}, std::make_shared(), "database" }, + { {}, std::make_shared(), "table" }, + }; +} + void StorageSystemPartMovesBetweenShards::fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node * predicate, std::vector) const { const auto access = context->getAccess(); diff --git a/src/Storages/System/StorageSystemPartMovesBetweenShards.h b/src/Storages/System/StorageSystemPartMovesBetweenShards.h index 6a859d4de80..bc6133fcaaa 100644 --- a/src/Storages/System/StorageSystemPartMovesBetweenShards.h +++ b/src/Storages/System/StorageSystemPartMovesBetweenShards.h @@ -20,6 +20,7 @@ protected: using IStorageSystemOneBlock::IStorageSystemOneBlock; void fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node * predicate, std::vector) const override; + Block getFilterSampleBlock() const override; }; } diff --git a/src/Storages/System/StorageSystemPartsBase.cpp b/src/Storages/System/StorageSystemPartsBase.cpp index 7ace8ee24aa..c87bdb6d26a 100644 --- a/src/Storages/System/StorageSystemPartsBase.cpp +++ b/src/Storages/System/StorageSystemPartsBase.cpp @@ -138,7 +138,7 @@ StoragesInfoStream::StoragesInfoStream(std::optional filter_by_datab for (size_t i = 0; i < rows; ++i) { - String database_name = (*database_column_for_filter)[i].get(); + String database_name = (*database_column_for_filter)[i].safeGet(); const DatabasePtr database = databases.at(database_name); offsets[i] = i ? offsets[i - 1] : 0; diff --git a/src/Storages/System/StorageSystemPartsBase.h b/src/Storages/System/StorageSystemPartsBase.h index 806af4a7bf8..3be73aeda17 100644 --- a/src/Storages/System/StorageSystemPartsBase.h +++ b/src/Storages/System/StorageSystemPartsBase.h @@ -52,13 +52,13 @@ public: { StoragesInfo info; - info.database = (*database_column)[next_row].get(); - info.table = (*table_column)[next_row].get(); - UUID storage_uuid = (*storage_uuid_column)[next_row].get(); + info.database = (*database_column)[next_row].safeGet(); + info.table = (*table_column)[next_row].safeGet(); + UUID storage_uuid = (*storage_uuid_column)[next_row].safeGet(); auto is_same_table = [&storage_uuid, this] (size_t row) -> bool { - return (*storage_uuid_column)[row].get() == storage_uuid; + return (*storage_uuid_column)[row].safeGet() == storage_uuid; }; /// We may have two rows per table which differ in 'active' value. @@ -66,7 +66,7 @@ public: /// must collect the inactive parts. Remember this fact in StoragesInfo. for (; next_row < rows && is_same_table(next_row); ++next_row) { - const auto active = (*active_column)[next_row].get(); + const auto active = (*active_column)[next_row].safeGet(); if (active == 0) info.need_inactive_parts = true; } diff --git a/src/Storages/System/StorageSystemQueryCache.cpp b/src/Storages/System/StorageSystemQueryCache.cpp index 4c54d4ae16f..b3532ba40a7 100644 --- a/src/Storages/System/StorageSystemQueryCache.cpp +++ b/src/Storages/System/StorageSystemQueryCache.cpp @@ -1,6 +1,7 @@ #include "StorageSystemQueryCache.h" #include #include +#include #include #include #include @@ -15,6 +16,7 @@ ColumnsDescription StorageSystemQueryCache::getColumnsDescription() { {"query", std::make_shared(), "Query string."}, {"result_size", std::make_shared(), "Size of the query cache entry."}, + {"tag", std::make_shared(std::make_shared()), "Tag of the query cache entry."}, {"stale", std::make_shared(), "If the query cache entry is stale."}, {"shared", std::make_shared(), "If the query cache entry is shared between multiple users."}, {"compressed", std::make_shared(), "If the query cache entry is compressed."}, @@ -51,11 +53,12 @@ void StorageSystemQueryCache::fillData(MutableColumns & res_columns, ContextPtr res_columns[0]->insert(key.query_string); /// approximates the original query string res_columns[1]->insert(QueryCache::QueryCacheEntryWeight()(*query_result)); - res_columns[2]->insert(key.expires_at < std::chrono::system_clock::now()); - res_columns[3]->insert(key.is_shared); - res_columns[4]->insert(key.is_compressed); - res_columns[5]->insert(std::chrono::system_clock::to_time_t(key.expires_at)); - res_columns[6]->insert(key.ast_hash.low64); /// query cache considers aliases (issue #56258) + res_columns[2]->insert(key.tag); + res_columns[3]->insert(key.expires_at < std::chrono::system_clock::now()); + res_columns[4]->insert(key.is_shared); + res_columns[5]->insert(key.is_compressed); + res_columns[6]->insert(std::chrono::system_clock::to_time_t(key.expires_at)); + res_columns[7]->insert(key.ast_hash.low64); /// query cache considers aliases (issue #56258) } } diff --git a/src/Storages/System/StorageSystemReplicas.cpp b/src/Storages/System/StorageSystemReplicas.cpp index 3bd5fd290db..724e4bd3f77 100644 --- a/src/Storages/System/StorageSystemReplicas.cpp +++ b/src/Storages/System/StorageSystemReplicas.cpp @@ -285,7 +285,7 @@ private: const bool with_zk_fields; const size_t max_block_size; std::shared_ptr impl; - const ActionsDAG::Node * predicate = nullptr; + ExpressionActionsPtr virtual_columns_filter; }; void ReadFromSystemReplicas::applyFilters(ActionDAGNodes added_filter_nodes) @@ -293,7 +293,18 @@ void ReadFromSystemReplicas::applyFilters(ActionDAGNodes added_filter_nodes) SourceStepWithFilter::applyFilters(std::move(added_filter_nodes)); if (filter_actions_dag) - predicate = filter_actions_dag->getOutputs().at(0); + { + Block block_to_filter + { + { ColumnString::create(), std::make_shared(), "database" }, + { ColumnString::create(), std::make_shared(), "table" }, + { ColumnString::create(), std::make_shared(), "engine" }, + }; + + auto dag = VirtualColumnUtils::splitFilterDagForAllowedInputs(filter_actions_dag->getOutputs().at(0), &block_to_filter); + if (dag) + virtual_columns_filter = VirtualColumnUtils::buildFilterExpression(std::move(*dag), context); + } } void StorageSystemReplicas::read( @@ -430,7 +441,8 @@ void ReadFromSystemReplicas::initializePipeline(QueryPipelineBuilder & pipeline, { col_engine, std::make_shared(), "engine" }, }; - VirtualColumnUtils::filterBlockWithPredicate(predicate, filtered_block, context); + if (virtual_columns_filter) + VirtualColumnUtils::filterBlockWithExpression(virtual_columns_filter, filtered_block); if (!filtered_block.rows()) { diff --git a/src/Storages/System/StorageSystemReplicationQueue.cpp b/src/Storages/System/StorageSystemReplicationQueue.cpp index 14b641f46c7..a50982de5f0 100644 --- a/src/Storages/System/StorageSystemReplicationQueue.cpp +++ b/src/Storages/System/StorageSystemReplicationQueue.cpp @@ -62,6 +62,14 @@ ColumnsDescription StorageSystemReplicationQueue::getColumnsDescription() } +Block StorageSystemReplicationQueue::getFilterSampleBlock() const +{ + return { + { {}, std::make_shared(), "database" }, + { {}, std::make_shared(), "table" }, + }; +} + void StorageSystemReplicationQueue::fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node * predicate, std::vector) const { const auto access = context->getAccess(); diff --git a/src/Storages/System/StorageSystemReplicationQueue.h b/src/Storages/System/StorageSystemReplicationQueue.h index a9e57851be1..82a4d68f300 100644 --- a/src/Storages/System/StorageSystemReplicationQueue.h +++ b/src/Storages/System/StorageSystemReplicationQueue.h @@ -21,6 +21,7 @@ public: protected: using IStorageSystemOneBlock::IStorageSystemOneBlock; void fillData(MutableColumns & res_columns, ContextPtr context, const ActionsDAG::Node * predicate, std::vector) const override; + Block getFilterSampleBlock() const override; }; } diff --git a/src/Storages/System/StorageSystemViewRefreshes.cpp b/src/Storages/System/StorageSystemViewRefreshes.cpp index 30539ed6b6a..061201017a7 100644 --- a/src/Storages/System/StorageSystemViewRefreshes.cpp +++ b/src/Storages/System/StorageSystemViewRefreshes.cpp @@ -5,6 +5,7 @@ #include #include #include +#include #include #include #include @@ -19,6 +20,7 @@ ColumnsDescription StorageSystemViewRefreshes::getColumnsDescription() { {"database", std::make_shared(), "The name of the database the table is in."}, {"view", std::make_shared(), "Table name."}, + {"uuid", std::make_shared(), "Table uuid (Atomic database)."}, {"status", std::make_shared(), "Current state of the refresh."}, {"last_refresh_result", std::make_shared(), "Outcome of the latest refresh attempt."}, {"last_refresh_time", std::make_shared(std::make_shared()), @@ -63,6 +65,7 @@ void StorageSystemViewRefreshes::fillData( std::size_t i = 0; res_columns[i++]->insert(refresh.view_id.getDatabaseName()); res_columns[i++]->insert(refresh.view_id.getTableName()); + res_columns[i++]->insert(refresh.view_id.uuid); res_columns[i++]->insert(toString(refresh.state)); res_columns[i++]->insert(toString(refresh.last_refresh_result)); diff --git a/src/Storages/TimeSeries/PrometheusRemoteReadProtocol.cpp b/src/Storages/TimeSeries/PrometheusRemoteReadProtocol.cpp new file mode 100644 index 00000000000..d6d258f5ff6 --- /dev/null +++ b/src/Storages/TimeSeries/PrometheusRemoteReadProtocol.cpp @@ -0,0 +1,472 @@ +#include + +#include "config.h" +#if USE_PROMETHEUS_PROTOBUFS + +#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 BAD_REQUEST_PARAMETER; +} + + +namespace +{ + /// Makes an ASTIdentifier for a column of the specified table. + ASTPtr makeASTColumn(const StorageID & table_id, const String & column_name) + { + return std::make_shared(Strings{table_id.database_name, table_id.table_name, column_name}); + } + + /// Makes an AST for condition `data_table.timestamp >= min_timestamp_ms` + ASTPtr makeASTTimestampGreaterOrEquals(Int64 min_timestamp_ms, const StorageID & data_table_id) + { + return makeASTFunction("greaterOrEquals", + makeASTColumn(data_table_id, TimeSeriesColumnNames::Timestamp), + std::make_shared(Field{DecimalField{DateTime64{min_timestamp_ms}, 3}})); + } + + /// Makes an AST for condition `data_table.timestamp <= max_timestamp_ms` + ASTPtr makeASTTimestampLessOrEquals(Int64 max_timestamp_ms, const StorageID & data_table_id) + { + return makeASTFunction("lessOrEquals", + makeASTColumn(data_table_id, TimeSeriesColumnNames::Timestamp), + std::make_shared(Field{DecimalField{DateTime64{max_timestamp_ms}, 3}})); + } + + /// Makes an AST for condition `tags_table.max_time >= min_timestamp_ms` + ASTPtr makeASTMaxTimeGreaterOrEquals(Int64 min_timestamp_ms, const StorageID & tags_table_id) + { + return makeASTFunction("greaterOrEquals", + makeASTColumn(tags_table_id, TimeSeriesColumnNames::MaxTime), + std::make_shared(Field{DecimalField{DateTime64{min_timestamp_ms}, 3}})); + } + + /// Makes an AST for condition `tags_table.min_time <= max_timestamp_ms` + ASTPtr makeASTMinTimeLessOrEquals(Int64 max_timestamp_ms, const StorageID & tags_table_id) + { + return makeASTFunction("lessOrEquals", + makeASTColumn(tags_table_id, TimeSeriesColumnNames::MinTime), + std::make_shared(Field{DecimalField{DateTime64{max_timestamp_ms}, 3}})); + } + + /// Makes an AST for the expression referencing a tag value. + ASTPtr makeASTLabelName(const String & label_name, const StorageID & tags_table_id, const std::unordered_map & column_name_by_tag_name) + { + if (label_name == TimeSeriesTagNames::MetricName) + return makeASTColumn(tags_table_id, TimeSeriesColumnNames::MetricName); + + auto it = column_name_by_tag_name.find(label_name); + if (it != column_name_by_tag_name.end()) + return makeASTColumn(tags_table_id, it->second); + + /// arrayElement() can be used to extract a value from a Map too. + return makeASTFunction("arrayElement", makeASTColumn(tags_table_id, TimeSeriesColumnNames::Tags), std::make_shared(label_name)); + } + + /// Makes an AST for a label matcher, for example `metric_name == 'value'` or `NOT match(labels['label_name'], 'regexp')`. + ASTPtr makeASTLabelMatcher( + const prometheus::LabelMatcher & label_matcher, + const StorageID & tags_table_id, + const std::unordered_map & column_name_by_tag_name) + { + const auto & label_name = label_matcher.name(); + const auto & label_value = label_matcher.value(); + auto type = label_matcher.type(); + + if (type == prometheus::LabelMatcher::EQ) + return makeASTFunction("equals", makeASTLabelName(label_name, tags_table_id, column_name_by_tag_name), std::make_shared(label_value)); + else if (type == prometheus::LabelMatcher::NEQ) + return makeASTFunction("notEquals", makeASTLabelName(label_name, tags_table_id, column_name_by_tag_name), std::make_shared(label_value)); + else if (type == prometheus::LabelMatcher::RE) + return makeASTFunction("match", makeASTLabelName(label_name, tags_table_id, column_name_by_tag_name), std::make_shared(label_value)); + else if (type == prometheus::LabelMatcher::NRE) + return makeASTFunction("not", makeASTFunction("match", makeASTLabelName(label_name, tags_table_id, column_name_by_tag_name), std::make_shared(label_value))); + else + throw Exception(ErrorCodes::BAD_REQUEST_PARAMETER, "Unexpected type of label matcher: {}", type); + } + + /// Makes an AST checking that tags match a specified label matcher and that timestamp is in range [min_timestamp_ms, max_timestamp_ms]. + ASTPtr makeASTFilterForReadingTimeSeries( + const google::protobuf::RepeatedPtrField & label_matcher, + Int64 min_timestamp_ms, + Int64 max_timestamp_ms, + const StorageID & data_table_id, + const StorageID & tags_table_id, + const std::unordered_map & column_name_by_tag_name, + bool filter_by_min_time_and_max_time) + { + ASTs filters; + + if (min_timestamp_ms) + { + filters.push_back(makeASTTimestampGreaterOrEquals(min_timestamp_ms, data_table_id)); + if (filter_by_min_time_and_max_time) + filters.push_back(makeASTMaxTimeGreaterOrEquals(min_timestamp_ms, tags_table_id)); + } + + if (max_timestamp_ms) + { + filters.push_back(makeASTTimestampLessOrEquals(max_timestamp_ms, data_table_id)); + if (filter_by_min_time_and_max_time) + filters.push_back(makeASTMinTimeLessOrEquals(max_timestamp_ms, tags_table_id)); + } + + for (const auto & label_matcher_element : label_matcher) + filters.push_back(makeASTLabelMatcher(label_matcher_element, tags_table_id, column_name_by_tag_name)); + + if (filters.empty()) + return nullptr; + + return makeASTForLogicalAnd(std::move(filters)); + } + + /// Makes a mapping from a tag name to a column name. + std::unordered_map makeColumnNameByTagNameMap(const TimeSeriesSettings & storage_settings) + { + std::unordered_map res; + const Map & tags_to_columns = storage_settings.tags_to_columns; + for (const auto & tag_name_and_column_name : tags_to_columns) + { + const auto & tuple = tag_name_and_column_name.safeGet(); + const auto & tag_name = tuple.at(0).safeGet(); + const auto & column_name = tuple.at(1).safeGet(); + res[tag_name] = column_name; + } + return res; + } + + /// The function builds a SELECT query for reading time series: + /// SELECT tags_table.metric_name, tags_table.tag_column1, ... tags_table.tag_columnN, tags_table.tags, + /// groupArray(CAST(data_table.timestamp, 'DateTime64(3)'), CAST(data_table.value, 'Float64')) + /// FROM data_table + /// SEMI LEFT JOIN tag_table ON data_table.id = tags_table.id + /// WHERE filter + /// GROUP BY tags_table.tag_column1, ..., tags_table.tag_columnN, tags_table.tags + ASTPtr buildSelectQueryForReadingTimeSeries( + Int64 min_timestamp_ms, + Int64 max_timestamp_ms, + const google::protobuf::RepeatedPtrField & label_matcher, + const TimeSeriesSettings & time_series_settings, + const StorageID & data_table_id, + const StorageID & tags_table_id) + { + auto select_query = std::make_shared(); + + /// SELECT tags_table.metric_name, any(tags_table.tag_column1), ... any(tags_table.tag_columnN), any(tags_table.tags), + /// groupArray(data_table.timestamp, data_table.value) + { + auto exp_list = std::make_shared(); + + exp_list->children.push_back( + makeASTColumn(tags_table_id, TimeSeriesColumnNames::MetricName)); + + const Map & tags_to_columns = time_series_settings.tags_to_columns; + for (const auto & tag_name_and_column_name : tags_to_columns) + { + const auto & tuple = tag_name_and_column_name.safeGet(); + const auto & column_name = tuple.at(1).safeGet(); + exp_list->children.push_back( + makeASTColumn(tags_table_id, column_name)); + } + + exp_list->children.push_back( + makeASTColumn(tags_table_id, TimeSeriesColumnNames::Tags)); + + exp_list->children.push_back( + makeASTFunction("groupArray", + makeASTFunction("tuple", + makeASTFunction("CAST", makeASTColumn(data_table_id, TimeSeriesColumnNames::Timestamp), std::make_shared("DateTime64(3)")), + makeASTFunction("CAST", makeASTColumn(data_table_id, TimeSeriesColumnNames::Value), std::make_shared("Float64"))))); + + select_query->setExpression(ASTSelectQuery::Expression::SELECT, exp_list); + } + + /// FROM data_table + auto tables = std::make_shared(); + + { + auto table = std::make_shared(); + auto table_exp = std::make_shared(); + table_exp->database_and_table_name = std::make_shared(data_table_id); + table_exp->children.emplace_back(table_exp->database_and_table_name); + + table->table_expression = table_exp; + tables->children.push_back(table); + } + + /// SEMI LEFT JOIN tags_table ON data_table.id = tags_table.id + { + auto table = std::make_shared(); + + auto table_join = std::make_shared(); + table_join->kind = JoinKind::Left; + table_join->strictness = JoinStrictness::Semi; + + table_join->on_expression = makeASTFunction("equals", makeASTColumn(data_table_id, TimeSeriesColumnNames::ID), makeASTColumn(tags_table_id, TimeSeriesColumnNames::ID)); + table->table_join = table_join; + + auto table_exp = std::make_shared(); + table_exp->database_and_table_name = std::make_shared(tags_table_id); + table_exp->children.emplace_back(table_exp->database_and_table_name); + + table->table_expression = table_exp; + tables->children.push_back(table); + + select_query->setExpression(ASTSelectQuery::Expression::TABLES, tables); + } + + auto column_name_by_tag_name = makeColumnNameByTagNameMap(time_series_settings); + + /// WHERE + if (auto where = makeASTFilterForReadingTimeSeries(label_matcher, min_timestamp_ms, max_timestamp_ms, data_table_id, tags_table_id, + column_name_by_tag_name, time_series_settings.filter_by_min_time_and_max_time)) + { + select_query->setExpression(ASTSelectQuery::Expression::WHERE, std::move(where)); + } + + /// GROUP BY tags_table.metric_name, tags_table.tag_column1, ..., tags_table.tag_columnN, tags_table.tags + { + auto exp_list = std::make_shared(); + + exp_list->children.push_back( + makeASTColumn(tags_table_id, TimeSeriesColumnNames::MetricName)); + + const Map & tags_to_columns = time_series_settings.tags_to_columns; + for (const auto & tag_name_and_column_name : tags_to_columns) + { + const auto & tuple = tag_name_and_column_name.safeGet(); + const auto & column_name = tuple.at(1).safeGet(); + exp_list->children.push_back( + makeASTColumn(tags_table_id, column_name)); + } + + exp_list->children.push_back(makeASTColumn(tags_table_id, TimeSeriesColumnNames::Tags)); + + select_query->setExpression(ASTSelectQuery::Expression::GROUP_BY, exp_list); + } + + return select_query; + } + + /// Sorts a list of pairs {tag_name, tag_value} by tag name. + void sortLabelsByName(std::vector> & labels) + { + auto less_by_label_name = [](const std::pair & left, const std::pair & right) + { + return left.first < right.first; + }; + std::sort(labels.begin(), labels.end(), less_by_label_name); + } + + /// Sorts a list of pairs {timestamp, value} by timestamp. + void sortTimeSeriesByTimestamp(std::vector> & time_series) + { + auto less_by_timestamp = [](const std::pair & left, const std::pair & right) + { + return left.first < right.first; + }; + std::sort(time_series.begin(), time_series.end(), less_by_timestamp); + } + + /// Converts a block generated by the SELECT query for converting time series to the protobuf format. + void convertBlockToProtobuf( + Block && block, + google::protobuf::RepeatedPtrField & out_time_series, + const StorageID & time_series_storage_id, + const TimeSeriesSettings & time_series_settings) + { + size_t num_rows = block.rows(); + if (!num_rows) + return; + + size_t column_index = 0; + + /// We analyze columns sequentially. + auto get_next_column_with_type = [&] -> const ColumnWithTypeAndName & { return block.getByPosition(column_index++); }; + auto get_next_column = [&] -> const IColumn & { return *(get_next_column_with_type().column); }; + + /// Column "metric_name". + const auto & metric_name_column_with_type = get_next_column_with_type(); + TimeSeriesColumnsValidator validator{time_series_storage_id, time_series_settings}; + validator.validateColumnForMetricName(metric_name_column_with_type); + const auto & metric_name_column = *metric_name_column_with_type.column; + + /// Columns corresponding to specific tags specified in the "tags_to_columns" setting. + std::unordered_map column_by_tag_name; + const Map & tags_to_columns = time_series_settings.tags_to_columns; + for (const auto & tag_name_and_column_name : tags_to_columns) + { + const auto & tuple = tag_name_and_column_name.safeGet(); + const auto & tag_name = tuple.at(0).safeGet(); + const auto & column_with_type = get_next_column_with_type(); + validator.validateColumnForTagValue(column_with_type); + const auto & column = *column_with_type.column; + column_by_tag_name[tag_name] = &column; + } + + /// Column "tags". + const auto & tags_column_with_type = get_next_column_with_type(); + validator.validateColumnForTagsMap(tags_column_with_type); + const auto & tags_column = checkAndGetColumn(*tags_column_with_type.column); + const auto & tags_names = tags_column.getNestedData().getColumn(0); + const auto & tags_values = tags_column.getNestedData().getColumn(1); + const auto & tags_offsets = tags_column.getNestedColumn().getOffsets(); + + /// Column containing time series: groupArray(CAST(data_table.timestamp, 'DateTime64(3)'), CAST(data_table.value, 'Float64')) + const auto & time_series_column = checkAndGetColumn(get_next_column()); + const auto & time_series_timestamps = checkAndGetColumn>(checkAndGetColumn(time_series_column.getData()).getColumn(0)); + const auto & time_series_values = checkAndGetColumn(checkAndGetColumn(time_series_column.getData()).getColumn(1)); + const auto & time_series_offsets = time_series_column.getOffsets(); + + /// We will sort labels lexicographically and time series by timestamp before sending them to a client. + std::vector> labels; + std::vector> time_series; + + for (size_t i = 0; i != num_rows; ++i) + { + /// Collect labels. + size_t num_labels = 1; /* 1 for a metric name */ + + for (const auto & [_, column] : column_by_tag_name) + { + if (!column->isNullAt(i) && !column->getDataAt(i).empty()) + ++num_labels; + } + + size_t tags_start_offset = tags_offsets[i - 1]; + size_t tags_end_offset = tags_offsets[i]; + num_labels += tags_end_offset - tags_start_offset; + + labels.clear(); + labels.reserve(num_labels); + + labels.emplace_back(TimeSeriesTagNames::MetricName, metric_name_column.getDataAt(i)); + + for (const auto & [tag_name, column] : column_by_tag_name) + { + if (!column->isNullAt(i) && !column->getDataAt(i).empty()) + labels.emplace_back(tag_name, column->getDataAt(i)); + } + + for (size_t j = tags_start_offset; j != tags_end_offset; ++j) + { + std::string_view tag_name{tags_names.getDataAt(j)}; + std::string_view tag_value{tags_values.getDataAt(j)}; + labels.emplace_back(tag_name, tag_value); + } + + /// Sort labels. + sortLabelsByName(labels); + + /// Collect time series. + size_t time_series_start_offset = time_series_offsets[i - 1]; + size_t time_series_end_offset = time_series_offsets[i]; + size_t num_time_series = time_series_end_offset - time_series_start_offset; + + time_series.clear(); + time_series.reserve(num_time_series); + + for (size_t j = time_series_start_offset; j != time_series_end_offset; ++j) + time_series.emplace_back(time_series_timestamps.getElement(j), time_series_values.getElement(j)); + + /// Sort time series. + sortTimeSeriesByTimestamp(time_series); + + /// Prepare a result. + auto & new_time_series = *out_time_series.Add(); + + for (const auto & [label_name, label_value] : labels) + { + auto & new_label = *new_time_series.add_labels(); + new_label.set_name(label_name); + new_label.set_value(label_value); + } + + for (const auto & [timestamp, value] : time_series) + { + auto & new_sample = *new_time_series.add_samples(); + new_sample.set_timestamp(timestamp); + new_sample.set_value(value); + } + } + } +} + + +PrometheusRemoteReadProtocol::PrometheusRemoteReadProtocol(ConstStoragePtr time_series_storage_, const ContextPtr & context_) + : WithContext{context_} + , time_series_storage(storagePtrToTimeSeries(time_series_storage_)) + , log(getLogger("PrometheusRemoteReadProtocol")) +{ +} + +PrometheusRemoteReadProtocol::~PrometheusRemoteReadProtocol() = default; + +void PrometheusRemoteReadProtocol::readTimeSeries(google::protobuf::RepeatedPtrField & out_time_series, + Int64 start_timestamp_ms, + Int64 end_timestamp_ms, + const google::protobuf::RepeatedPtrField & label_matcher, + const prometheus::ReadHints &) +{ + out_time_series.Clear(); + + auto time_series_storage_id = time_series_storage->getStorageID(); + auto time_series_settings = time_series_storage->getStorageSettingsPtr(); + auto data_table_id = time_series_storage->getTargetTableId(ViewTarget::Data); + auto tags_table_id = time_series_storage->getTargetTableId(ViewTarget::Tags); + + ASTPtr select_query = buildSelectQueryForReadingTimeSeries( + start_timestamp_ms, end_timestamp_ms, label_matcher, *time_series_settings, data_table_id, tags_table_id); + + LOG_TRACE(log, "{}: Executing query {}", + time_series_storage_id.getNameForLogs(), select_query); + + InterpreterSelectQuery interpreter(select_query, getContext(), SelectQueryOptions{}); + BlockIO io = interpreter.execute(); + PullingPipelineExecutor executor(io.pipeline); + + Block block; + while (executor.pull(block)) + { + LOG_TRACE(log, "{}: Pulled block with {} columns and {} rows", + time_series_storage_id.getNameForLogs(), block.columns(), block.rows()); + + if (block) + convertBlockToProtobuf(std::move(block), out_time_series, time_series_storage_id, *time_series_settings); + } + + LOG_TRACE(log, "{}: {} time series read", + time_series_storage_id.getNameForLogs(), out_time_series.size()); +} + +} + +#endif diff --git a/src/Storages/TimeSeries/PrometheusRemoteReadProtocol.h b/src/Storages/TimeSeries/PrometheusRemoteReadProtocol.h new file mode 100644 index 00000000000..e10e1f8c8cf --- /dev/null +++ b/src/Storages/TimeSeries/PrometheusRemoteReadProtocol.h @@ -0,0 +1,36 @@ +#pragma once + +#include "config.h" +#if USE_PROMETHEUS_PROTOBUFS + +#include +#include +#include + + +namespace DB +{ +class StorageTimeSeries; + +/// Helper class to support the prometheus remote read protocol. +class PrometheusRemoteReadProtocol : public WithContext +{ +public: + PrometheusRemoteReadProtocol(ConstStoragePtr time_series_storage_, const ContextPtr & context_); + ~PrometheusRemoteReadProtocol(); + + /// Reads time series to send to client by remote read protocol. + void readTimeSeries(google::protobuf::RepeatedPtrField & out_time_series, + Int64 start_timestamp_ms, + Int64 end_timestamp_ms, + const google::protobuf::RepeatedPtrField & label_matcher, + const prometheus::ReadHints & read_hints); + +private: + std::shared_ptr time_series_storage; + Poco::LoggerPtr log; +}; + +} + +#endif diff --git a/src/Storages/TimeSeries/PrometheusRemoteWriteProtocol.cpp b/src/Storages/TimeSeries/PrometheusRemoteWriteProtocol.cpp new file mode 100644 index 00000000000..702d058ee79 --- /dev/null +++ b/src/Storages/TimeSeries/PrometheusRemoteWriteProtocol.cpp @@ -0,0 +1,601 @@ +#include + +#include "config.h" +#if USE_PROMETHEUS_PROTOBUFS + +#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_TIME_SERIES_TAGS; + extern const int ILLEGAL_COLUMN; +} + + +namespace +{ + /// Checks that a specified set of labels is sorted and has no duplications, and there is one label named "__name__". + void checkLabels(const ::google::protobuf::RepeatedPtrField<::prometheus::Label> & labels) + { + bool metric_name_found = false; + for (size_t i = 0; i != static_cast(labels.size()); ++i) + { + const auto & label = labels[static_cast(i)]; + const auto & label_name = label.name(); + const auto & label_value = label.value(); + + if (label_name.empty()) + throw Exception(ErrorCodes::ILLEGAL_TIME_SERIES_TAGS, "Label name should not be empty"); + if (label_value.empty()) + continue; /// Empty label value is treated like the label doesn't exist. + + if (label_name == TimeSeriesTagNames::MetricName) + metric_name_found = true; + + if (i) + { + /// Check that labels are sorted. + const auto & previous_label_name = labels[static_cast(i - 1)].name(); + if (label_name <= previous_label_name) + { + if (label_name == previous_label_name) + throw Exception(ErrorCodes::ILLEGAL_TIME_SERIES_TAGS, "Found duplicate label {}", label_name); + else + throw Exception(ErrorCodes::ILLEGAL_TIME_SERIES_TAGS, "Label names are not sorted in lexicographical order ({} > {})", + previous_label_name, label_name); + } + } + } + + if (!metric_name_found) + throw Exception(ErrorCodes::ILLEGAL_TIME_SERIES_TAGS, "Metric name (label {}) not found", TimeSeriesTagNames::MetricName); + } + + /// Finds the description of an insertable column in the list. + const ColumnDescription & getInsertableColumnDescription(const ColumnsDescription & columns, const String & column_name, const StorageID & time_series_storage_id) + { + const ColumnDescription * column = columns.tryGet(column_name); + if (!column || ((column->default_desc.kind != ColumnDefaultKind::Default) && (column->default_desc.kind != ColumnDefaultKind::Ephemeral))) + { + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "{}: Column {} {}", + time_series_storage_id.getNameForLogs(), column_name, column ? "non-insertable" : "doesn't exist"); + } + return *column; + } + + /// Calculates the identifier of each time series in "tags_block" using the default expression for the "id" column, + /// and adds column "id" with the results to "tags_block". + IColumn & calculateId(const ContextPtr & context, const ColumnDescription & id_column_description, Block & tags_block) + { + auto blocks = std::make_shared(); + blocks->push_back(tags_block); + + auto header = tags_block.cloneEmpty(); + auto pipe = Pipe(std::make_shared(blocks, header)); + + Block header_with_id; + const auto & id_name = id_column_description.name; + auto id_type = id_column_description.type; + header_with_id.insert(ColumnWithTypeAndName{id_type, id_name}); + + auto adding_missing_defaults_dag = addMissingDefaults( + pipe.getHeader(), + header_with_id.getNamesAndTypesList(), + ColumnsDescription{id_column_description}, + context); + + auto adding_missing_defaults_actions = std::make_shared(std::move(adding_missing_defaults_dag)); + pipe.addSimpleTransform([&](const Block & stream_header) + { + return std::make_shared(stream_header, adding_missing_defaults_actions); + }); + + auto convert_actions_dag = ActionsDAG::makeConvertingActions( + pipe.getHeader().getColumnsWithTypeAndName(), + header_with_id.getColumnsWithTypeAndName(), + ActionsDAG::MatchColumnsMode::Position); + auto actions = std::make_shared( + std::move(convert_actions_dag), + ExpressionActionsSettings::fromContext(context, CompileExpressions::yes)); + pipe.addSimpleTransform([&](const Block & stream_header) + { + return std::make_shared(stream_header, actions); + }); + + QueryPipeline pipeline{std::move(pipe)}; + PullingPipelineExecutor executor{pipeline}; + + MutableColumnPtr id_column; + + Block block_from_executor; + while (executor.pull(block_from_executor)) + { + if (block_from_executor) + { + MutableColumnPtr id_column_part = block_from_executor.getByName(id_name).column->assumeMutable(); + if (id_column) + id_column->insertRangeFrom(*id_column_part, 0, id_column_part->size()); + else + id_column = std::move(id_column_part); + } + } + + if (!id_column) + id_column = id_type->createColumn(); + + IColumn & id_column_ref = *id_column; + tags_block.insert(0, ColumnWithTypeAndName{std::move(id_column), id_type, id_name}); + return id_column_ref; + } + + /// Converts a timestamp in milliseconds to a DateTime64 with a specified scale. + DateTime64 scaleTimestamp(Int64 timestamp_ms, UInt32 scale) + { + if (scale == 3) + return timestamp_ms; + else if (scale > 3) + return timestamp_ms * DecimalUtils::scaleMultiplier(scale - 3); + else + return timestamp_ms / DecimalUtils::scaleMultiplier(3 - scale); + } + + /// Finds min time and max time in a time series. + std::pair findMinTimeAndMaxTime(const google::protobuf::RepeatedPtrField & samples) + { + chassert(!samples.empty()); + Int64 min_time = std::numeric_limits::max(); + Int64 max_time = std::numeric_limits::min(); + for (const auto & sample : samples) + { + Int64 timestamp = sample.timestamp(); + if (timestamp < min_time) + min_time = timestamp; + if (timestamp > max_time) + max_time = timestamp; + } + return {min_time, max_time}; + } + + struct BlocksToInsert + { + std::vector> blocks; + }; + + /// Converts time series from the protobuf format to prepared blocks for inserting into target tables. + BlocksToInsert toBlocks(const google::protobuf::RepeatedPtrField & time_series, + const ContextPtr & context, + const StorageID & time_series_storage_id, + const StorageInMemoryMetadata & time_series_storage_metadata, + const TimeSeriesSettings & time_series_settings) + { + size_t num_tags_rows = time_series.size(); + + size_t num_data_rows = 0; + for (const auto & element : time_series) + num_data_rows += element.samples_size(); + + if (!num_data_rows) + return {}; /// Nothing to insert into target tables. + + /// Column types must be extracted from the target tables' metadata. + const auto & columns_description = time_series_storage_metadata.columns; + + auto get_column_description = [&](const String & column_name) -> const ColumnDescription & + { + return getInsertableColumnDescription(columns_description, column_name, time_series_storage_id); + }; + + /// We're going to prepare two blocks - one for the "data" table, and one for the "tags" table. + Block data_block, tags_block; + + auto make_column_for_data_block = [&](const ColumnDescription & column_description) -> IColumn & + { + auto column = column_description.type->createColumn(); + column->reserve(num_data_rows); + auto * column_ptr = column.get(); + data_block.insert(ColumnWithTypeAndName{std::move(column), column_description.type, column_description.name}); + return *column_ptr; + }; + + auto make_column_for_tags_block = [&](const ColumnDescription & column_description) -> IColumn & + { + auto column = column_description.type->createColumn(); + column->reserve(num_tags_rows); + auto * column_ptr = column.get(); + tags_block.insert(ColumnWithTypeAndName{std::move(column), column_description.type, column_description.name}); + return *column_ptr; + }; + + /// Create columns. + + /// Column "id". + const auto & id_description = get_column_description(TimeSeriesColumnNames::ID); + TimeSeriesColumnsValidator validator{time_series_storage_id, time_series_settings}; + validator.validateColumnForID(id_description); + auto & id_column_in_data_table = make_column_for_data_block(id_description); + + /// Column "timestamp". + const auto & timestamp_description = get_column_description(TimeSeriesColumnNames::Timestamp); + UInt32 timestamp_scale; + validator.validateColumnForTimestamp(timestamp_description, timestamp_scale); + auto & timestamp_column = make_column_for_data_block(timestamp_description); + + /// Column "value". + const auto & value_description = get_column_description(TimeSeriesColumnNames::Value); + validator.validateColumnForValue(value_description); + auto & value_column = make_column_for_data_block(value_description); + + /// Column "metric_name". + const auto & metric_name_description = get_column_description(TimeSeriesColumnNames::MetricName); + validator.validateColumnForMetricName(metric_name_description); + auto & metric_name_column = make_column_for_tags_block(metric_name_description); + + /// Columns we should check explicitly that they're filled after filling each row. + std::vector columns_to_fill_in_tags_table; + + /// Columns corresponding to specific tags specified in the "tags_to_columns" setting. + std::unordered_map columns_by_tag_name; + const Map & tags_to_columns = time_series_settings.tags_to_columns; + for (const auto & tag_name_and_column_name : tags_to_columns) + { + const auto & tuple = tag_name_and_column_name.safeGet(); + const auto & tag_name = tuple.at(0).safeGet(); + const auto & column_name = tuple.at(1).safeGet(); + const auto & column_description = get_column_description(column_name); + validator.validateColumnForTagValue(column_description); + auto & column = make_column_for_tags_block(column_description); + columns_by_tag_name[tag_name] = &column; + columns_to_fill_in_tags_table.emplace_back(&column); + } + + /// Column "tags". + const auto & tags_description = get_column_description(TimeSeriesColumnNames::Tags); + validator.validateColumnForTagsMap(tags_description); + auto & tags_column = typeid_cast(make_column_for_tags_block(tags_description)); + IColumn & tags_names = tags_column.getNestedData().getColumn(0); + IColumn & tags_values = tags_column.getNestedData().getColumn(1); + auto & tags_offsets = tags_column.getNestedColumn().getOffsets(); + + /// Column "all_tags". + IColumn * all_tags_names = nullptr; + IColumn * all_tags_values = nullptr; + IColumn::Offsets * all_tags_offsets = nullptr; + if (time_series_settings.use_all_tags_column_to_generate_id) + { + const auto & all_tags_description = get_column_description(TimeSeriesColumnNames::AllTags); + validator.validateColumnForTagsMap(all_tags_description); + auto & all_tags_column = typeid_cast(make_column_for_tags_block(all_tags_description)); + all_tags_names = &all_tags_column.getNestedData().getColumn(0); + all_tags_values = &all_tags_column.getNestedData().getColumn(1); + all_tags_offsets = &all_tags_column.getNestedColumn().getOffsets(); + } + + /// Columns "min_time" and "max_time". + IColumn * min_time_column = nullptr; + IColumn * max_time_column = nullptr; + UInt32 min_time_scale = 0; + UInt32 max_time_scale = 0; + if (time_series_settings.store_min_time_and_max_time) + { + const auto & min_time_description = get_column_description(TimeSeriesColumnNames::MinTime); + const auto & max_time_description = get_column_description(TimeSeriesColumnNames::MaxTime); + validator.validateColumnForTimestamp(min_time_description, min_time_scale); + validator.validateColumnForTimestamp(max_time_description, max_time_scale); + min_time_column = &make_column_for_tags_block(min_time_description); + max_time_column = &make_column_for_tags_block(max_time_description); + columns_to_fill_in_tags_table.emplace_back(min_time_column); + columns_to_fill_in_tags_table.emplace_back(max_time_column); + } + + /// Prepare a block for inserting into the "tags" table. + size_t current_row_in_tags = 0; + for (size_t i = 0; i != static_cast(time_series.size()); ++i) + { + const auto & element = time_series[static_cast(i)]; + if (!element.samples_size()) + continue; + + const auto & labels = element.labels(); + checkLabels(labels); + + for (size_t j = 0; j != static_cast(labels.size()); ++j) + { + const auto & label = labels[static_cast(j)]; + const auto & tag_name = label.name(); + const auto & tag_value = label.value(); + + if (tag_name == TimeSeriesTagNames::MetricName) + { + metric_name_column.insertData(tag_value.data(), tag_value.length()); + } + else + { + if (time_series_settings.use_all_tags_column_to_generate_id) + { + all_tags_names->insertData(tag_name.data(), tag_name.length()); + all_tags_values->insertData(tag_value.data(), tag_value.length()); + } + + auto it = columns_by_tag_name.find(tag_name); + bool has_column_for_tag_value = (it != columns_by_tag_name.end()); + if (has_column_for_tag_value) + { + auto * column = it->second; + column->insertData(tag_value.data(), tag_value.length()); + } + else + { + tags_names.insertData(tag_name.data(), tag_name.length()); + tags_values.insertData(tag_value.data(), tag_value.length()); + } + } + } + + tags_offsets.push_back(tags_names.size()); + + if (time_series_settings.use_all_tags_column_to_generate_id) + all_tags_offsets->push_back(all_tags_names->size()); + + if (time_series_settings.store_min_time_and_max_time) + { + auto [min_time, max_time] = findMinTimeAndMaxTime(element.samples()); + min_time_column->insert(scaleTimestamp(min_time, min_time_scale)); + max_time_column->insert(scaleTimestamp(max_time, max_time_scale)); + } + + for (auto * column : columns_to_fill_in_tags_table) + { + if (column->size() == current_row_in_tags) + column->insertDefault(); + } + + ++current_row_in_tags; + } + + /// Calculate an identifier for each time series, make a new column from those identifiers, and add it to "tags_block". + auto & id_column_in_tags_table = calculateId(context, columns_description.get(TimeSeriesColumnNames::ID), tags_block); + + /// Prepare a block for inserting to the "data" table. + current_row_in_tags = 0; + for (size_t i = 0; i != static_cast(time_series.size()); ++i) + { + const auto & element = time_series[static_cast(i)]; + if (!element.samples_size()) + continue; + + id_column_in_data_table.insertManyFrom(id_column_in_tags_table, current_row_in_tags, element.samples_size()); + for (const auto & sample : element.samples()) + { + timestamp_column.insert(scaleTimestamp(sample.timestamp(), timestamp_scale)); + value_column.insert(sample.value()); + } + + ++current_row_in_tags; + } + + /// The "all_tags" column in the "tags" table is either ephemeral or doesn't exists. + /// We've used the "all_tags" column to calculate the "id" column already, + /// and now we don't need it to insert to the "tags" table. + tags_block.erase(TimeSeriesColumnNames::AllTags); + + BlocksToInsert res; + + /// A block to the "tags" table should be inserted first. + /// (Because any INSERT can fail and we don't want to have rows in the data table with no corresponding "id" written to the "tags" table.) + res.blocks.emplace_back(ViewTarget::Tags, std::move(tags_block)); + res.blocks.emplace_back(ViewTarget::Data, std::move(data_block)); + + return res; + } + + std::string_view metricTypeToString(prometheus::MetricMetadata::MetricType metric_type) + { + using namespace std::literals; + switch (metric_type) + { + case prometheus::MetricMetadata::UNKNOWN: return "unknown"sv; + case prometheus::MetricMetadata::COUNTER: return "counter"sv; + case prometheus::MetricMetadata::GAUGE: return "gauge"sv; + case prometheus::MetricMetadata::HISTOGRAM: return "histogram"sv; + case prometheus::MetricMetadata::GAUGEHISTOGRAM: return "gaugehistogram"sv; + case prometheus::MetricMetadata::SUMMARY: return "summary"sv; + case prometheus::MetricMetadata::INFO: return "info"sv; + case prometheus::MetricMetadata::STATESET: return "stateset"sv; + default: break; + } + return ""; + } + + /// Converts metrics metadata from the protobuf format to prepared blocks for inserting into target tables. + BlocksToInsert toBlocks(const google::protobuf::RepeatedPtrField & metrics_metadata, + const StorageID & time_series_storage_id, + const StorageInMemoryMetadata & time_series_storage_metadata, + const TimeSeriesSettings & time_series_settings) + { + size_t num_rows = metrics_metadata.size(); + + if (!num_rows) + return {}; /// Nothing to insert into target tables. + + /// Column types must be extracted from the target tables' metadata. + const auto & columns_description = time_series_storage_metadata.columns; + + auto get_column_description = [&](const String & column_name) -> const ColumnDescription & + { + return getInsertableColumnDescription(columns_description, column_name, time_series_storage_id); + }; + + /// We're going to prepare one blocks for the "metrics" table. + Block block; + + auto make_column = [&](const ColumnDescription & column_description) -> IColumn & + { + auto column = column_description.type->createColumn(); + column->reserve(num_rows); + auto * column_ptr = column.get(); + block.insert(ColumnWithTypeAndName{std::move(column), column_description.type, column_description.name}); + return *column_ptr; + }; + + /// Create columns. + + /// Column "metric_family_name". + const auto & metric_family_name_description = get_column_description(TimeSeriesColumnNames::MetricFamilyName); + TimeSeriesColumnsValidator validator{time_series_storage_id, time_series_settings}; + validator.validateColumnForMetricFamilyName(metric_family_name_description); + auto & metric_family_name_column = make_column(metric_family_name_description); + + /// Column "type". + const auto & type_description = get_column_description(TimeSeriesColumnNames::Type); + validator.validateColumnForType(type_description); + auto & type_column = make_column(type_description); + + /// Column "unit". + const auto & unit_description = get_column_description(TimeSeriesColumnNames::Unit); + validator.validateColumnForUnit(unit_description); + auto & unit_column = make_column(unit_description); + + /// Column "help". + const auto & help_description = get_column_description(TimeSeriesColumnNames::Help); + validator.validateColumnForHelp(help_description); + auto & help_column = make_column(help_description); + + /// Fill those columns. + for (const auto & element : metrics_metadata) + { + const auto & metric_family_name = element.metric_family_name(); + const auto & type_str = metricTypeToString(element.type()); + const auto & help = element.help(); + const auto & unit = element.unit(); + + metric_family_name_column.insertData(metric_family_name.data(), metric_family_name.length()); + type_column.insertData(type_str.data(), type_str.length()); + unit_column.insertData(unit.data(), unit.length()); + help_column.insertData(help.data(), help.length()); + } + + /// Prepare a result. + BlocksToInsert res; + res.blocks.emplace_back(ViewTarget::Metrics, std::move(block)); + return res; + } + + /// Inserts blocks to target tables. + void insertToTargetTables(BlocksToInsert && blocks, StorageTimeSeries & time_series_storage, ContextPtr context, Poco::Logger * log) + { + auto time_series_storage_id = time_series_storage.getStorageID(); + + for (auto & [table_kind, block] : blocks.blocks) + { + if (block) + { + const auto & target_table_id = time_series_storage.getTargetTableId(table_kind); + + LOG_INFO(log, "{}: Inserting {} rows to the {} table", + time_series_storage_id.getNameForLogs(), block.rows(), toString(table_kind)); + + auto insert_query = std::make_shared(); + insert_query->table_id = target_table_id; + + auto columns_ast = std::make_shared(); + for (const auto & name : block.getNames()) + columns_ast->children.emplace_back(std::make_shared(name)); + insert_query->columns = columns_ast; + + ContextMutablePtr insert_context = Context::createCopy(context); + insert_context->setCurrentQueryId(context->getCurrentQueryId() + ":" + String{toString(table_kind)}); + + LOG_TEST(log, "{}: Executing query: {}", time_series_storage_id.getNameForLogs(), queryToString(insert_query)); + + InterpreterInsertQuery interpreter( + insert_query, + insert_context, + /* allow_materialized= */ false, + /* no_squash= */ false, + /* no_destination= */ false, + /* async_insert= */ false); + + BlockIO io = interpreter.execute(); + PushingPipelineExecutor executor(io.pipeline); + + executor.start(); + executor.push(std::move(block)); + executor.finish(); + } + } + } +} + + +PrometheusRemoteWriteProtocol::PrometheusRemoteWriteProtocol(StoragePtr time_series_storage_, const ContextPtr & context_) + : WithContext(context_) + , time_series_storage(storagePtrToTimeSeries(time_series_storage_)) + , log(getLogger("PrometheusRemoteWriteProtocol")) +{ +} + +PrometheusRemoteWriteProtocol::~PrometheusRemoteWriteProtocol() = default; + + +void PrometheusRemoteWriteProtocol::writeTimeSeries(const google::protobuf::RepeatedPtrField & time_series) +{ + auto time_series_storage_id = time_series_storage->getStorageID(); + + LOG_TRACE(log, "{}: Writing {} time series", + time_series_storage_id.getNameForLogs(), time_series.size()); + + auto time_series_storage_metadata = time_series_storage->getInMemoryMetadataPtr(); + auto time_series_settings = time_series_storage->getStorageSettingsPtr(); + + auto blocks = toBlocks(time_series, getContext(), time_series_storage_id, *time_series_storage_metadata, *time_series_settings); + insertToTargetTables(std::move(blocks), *time_series_storage, getContext(), log.get()); + + LOG_TRACE(log, "{}: {} time series written", + time_series_storage_id.getNameForLogs(), time_series.size()); +} + +void PrometheusRemoteWriteProtocol::writeMetricsMetadata(const google::protobuf::RepeatedPtrField & metrics_metadata) +{ + auto time_series_storage_id = time_series_storage->getStorageID(); + + LOG_TRACE(log, "{}: Writing {} metrics metadata", + time_series_storage_id.getNameForLogs(), metrics_metadata.size()); + + auto time_series_storage_metadata = time_series_storage->getInMemoryMetadataPtr(); + auto time_series_settings = time_series_storage->getStorageSettingsPtr(); + + auto blocks = toBlocks(metrics_metadata, time_series_storage_id, *time_series_storage_metadata, *time_series_settings); + insertToTargetTables(std::move(blocks), *time_series_storage, getContext(), log.get()); + + LOG_TRACE(log, "{}: {} metrics metadata written", + time_series_storage_id.getNameForLogs(), metrics_metadata.size()); +} + +} + +#endif diff --git a/src/Storages/TimeSeries/PrometheusRemoteWriteProtocol.h b/src/Storages/TimeSeries/PrometheusRemoteWriteProtocol.h new file mode 100644 index 00000000000..24c65e96cbe --- /dev/null +++ b/src/Storages/TimeSeries/PrometheusRemoteWriteProtocol.h @@ -0,0 +1,35 @@ +#pragma once + +#include "config.h" +#if USE_PROMETHEUS_PROTOBUFS + +#include +#include +#include + + +namespace DB +{ +class StorageTimeSeries; + +/// Helper class to support the prometheus remote write protocol. +class PrometheusRemoteWriteProtocol : WithContext +{ +public: + PrometheusRemoteWriteProtocol(StoragePtr time_series_storage_, const ContextPtr & context_); + ~PrometheusRemoteWriteProtocol(); + + /// Insert time series received by remote write protocol to our table. + void writeTimeSeries(const google::protobuf::RepeatedPtrField & time_series); + + /// Insert metrics metadata received by remote write protocol to our table. + void writeMetricsMetadata(const google::protobuf::RepeatedPtrField & metrics_metadata); + +private: + std::shared_ptr time_series_storage; + Poco::LoggerPtr log; +}; + +} + +#endif diff --git a/src/Storages/TimeSeries/TimeSeriesColumnNames.h b/src/Storages/TimeSeries/TimeSeriesColumnNames.h new file mode 100644 index 00000000000..d7b12fdeea8 --- /dev/null +++ b/src/Storages/TimeSeries/TimeSeriesColumnNames.h @@ -0,0 +1,38 @@ +#pragma once + + +namespace DB +{ + +struct TimeSeriesColumnNames +{ + /// The "data" table contains time series: + static constexpr const char * ID = "id"; + static constexpr const char * Timestamp = "timestamp"; + static constexpr const char * Value = "value"; + + /// The "tags" table contains identifiers for each combination of a metric name with corresponding tags (labels): + + /// The default expression specified for the "id" column contains an expression for calculating an identifier of a time series by a metric name and tags. + //static constexpr const char * kID = "id"; + static constexpr const char * MetricName = "metric_name"; + + /// Contains tags which have no corresponding columns specified in the "tags_to_columns" setting. + static constexpr const char * Tags = "tags"; + + /// Contains all tags, including those ones which have corresponding columns specified in the "tags_to_columns" setting. + /// This is a generated column, it's not stored anywhere, it's generated on the fly. + static constexpr const char * AllTags = "all_tags"; + + /// Contains the time range of a time series. + static constexpr const char * MinTime = "min_time"; + static constexpr const char * MaxTime = "max_time"; + + /// The "metrics" table contains general information (metadata) about metrics: + static constexpr const char * MetricFamilyName = "metric_family_name"; + static constexpr const char * Type = "type"; + static constexpr const char * Unit = "unit"; + static constexpr const char * Help = "help"; +}; + +} diff --git a/src/Storages/TimeSeries/TimeSeriesColumnsValidator.cpp b/src/Storages/TimeSeries/TimeSeriesColumnsValidator.cpp new file mode 100644 index 00000000000..a2308857e2e --- /dev/null +++ b/src/Storages/TimeSeries/TimeSeriesColumnsValidator.cpp @@ -0,0 +1,272 @@ +#include + +#include +#include +#include +#include +#include +#include +#include + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int ILLEGAL_COLUMN; + extern const int INCOMPATIBLE_COLUMNS; + extern const int THERE_IS_NO_COLUMN; +} + + +TimeSeriesColumnsValidator::TimeSeriesColumnsValidator(StorageID time_series_storage_id_, + std::reference_wrapper time_series_settings_) + : time_series_storage_id(std::move(time_series_storage_id_)) + , time_series_settings(time_series_settings_) +{ +} + + +void TimeSeriesColumnsValidator::validateColumns(const ColumnsDescription & columns) const +{ + try + { + validateColumnsImpl(columns); + } + catch (Exception & e) + { + e.addMessage("While checking columns of TimeSeries table {}", time_series_storage_id.getNameForLogs()); + throw; + } +} + + +void TimeSeriesColumnsValidator::validateColumnsImpl(const ColumnsDescription & columns) const +{ + + auto get_column_description = [&](const String & column_name) -> const ColumnDescription & + { + const auto * column = columns.tryGet(column_name); + if (!column) + { + throw Exception(ErrorCodes::THERE_IS_NO_COLUMN, "Column {} is required for the TimeSeries table engine", column_name); + } + return *column; + }; + + /// Validate columns for the "data" table. + validateColumnForID(get_column_description(TimeSeriesColumnNames::ID)); + validateColumnForTimestamp(get_column_description(TimeSeriesColumnNames::Timestamp)); + validateColumnForValue(get_column_description(TimeSeriesColumnNames::Value)); + + /// Validate columns for the "tags" table. + validateColumnForMetricName(get_column_description(TimeSeriesColumnNames::MetricName)); + + const Map & tags_to_columns = time_series_settings.tags_to_columns; + for (const auto & tag_name_and_column_name : tags_to_columns) + { + const auto & tuple = tag_name_and_column_name.safeGet(); + const auto & column_name = tuple.at(1).safeGet(); + validateColumnForTagValue(get_column_description(column_name)); + } + + validateColumnForTagsMap(get_column_description(TimeSeriesColumnNames::Tags)); + validateColumnForTagsMap(get_column_description(TimeSeriesColumnNames::AllTags)); + + /// Validate columns for the "metrics" table. + validateColumnForMetricFamilyName(get_column_description(TimeSeriesColumnNames::MetricFamilyName)); + validateColumnForType(get_column_description(TimeSeriesColumnNames::Type)); + validateColumnForUnit(get_column_description(TimeSeriesColumnNames::Unit)); + validateColumnForHelp(get_column_description(TimeSeriesColumnNames::Help)); +} + + +void TimeSeriesColumnsValidator::validateTargetColumns(ViewTarget::Kind target_kind, const StorageID & target_table_id, const ColumnsDescription & target_columns) const +{ + try + { + validateTargetColumnsImpl(target_kind, target_columns); + } + catch (Exception & e) + { + e.addMessage("While checking columns of table {} which is the {} target of TimeSeries table {}", target_table_id.getNameForLogs(), + toString(target_kind), time_series_storage_id.getNameForLogs()); + throw; + } +} + + +void TimeSeriesColumnsValidator::validateTargetColumnsImpl(ViewTarget::Kind target_kind, const ColumnsDescription & target_columns) const +{ + auto get_column_description = [&](const String & column_name) -> const ColumnDescription & + { + const auto * column = target_columns.tryGet(column_name); + if (!column) + { + throw Exception(ErrorCodes::THERE_IS_NO_COLUMN, "Column {} is required for the TimeSeries table engine", column_name); + } + return *column; + }; + + switch (target_kind) + { + case ViewTarget::Data: + { + /// Here "check_default = false" because it's ok for the "id" column in the target table not to contain + /// an expression for calculating the identifier of a time series. + validateColumnForID(get_column_description(TimeSeriesColumnNames::ID), /* check_default= */ false); + + validateColumnForTimestamp(get_column_description(TimeSeriesColumnNames::Timestamp)); + validateColumnForValue(get_column_description(TimeSeriesColumnNames::Value)); + + break; + } + + case ViewTarget::Tags: + { + validateColumnForMetricName(get_column_description(TimeSeriesColumnNames::MetricName)); + + const Map & tags_to_columns = time_series_settings.tags_to_columns; + for (const auto & tag_name_and_column_name : tags_to_columns) + { + const auto & tuple = tag_name_and_column_name.safeGet(); + const auto & column_name = tuple.at(1).safeGet(); + validateColumnForTagValue(get_column_description(column_name)); + } + + validateColumnForTagsMap(get_column_description(TimeSeriesColumnNames::Tags)); + + break; + } + + case ViewTarget::Metrics: + { + validateColumnForMetricFamilyName(get_column_description(TimeSeriesColumnNames::MetricFamilyName)); + validateColumnForType(get_column_description(TimeSeriesColumnNames::Type)); + validateColumnForUnit(get_column_description(TimeSeriesColumnNames::Unit)); + validateColumnForHelp(get_column_description(TimeSeriesColumnNames::Help)); + break; + } + + default: + UNREACHABLE(); + } +} + + +void TimeSeriesColumnsValidator::validateColumnForID(const ColumnDescription & column, bool check_default) const +{ + if (check_default && !column.default_desc.expression) + { + throw Exception(ErrorCodes::INCOMPATIBLE_COLUMNS, "The DEFAULT expression for column {} must contain an expression " + "which will be used to calculate the identifier of each time series: {} {} DEFAULT ...", + column.name, column.name, column.type->getName()); + } +} + +void TimeSeriesColumnsValidator::validateColumnForTimestamp(const ColumnDescription & column) const +{ + if (!isDateTime64(removeNullable(column.type))) + { + throw Exception(ErrorCodes::INCOMPATIBLE_COLUMNS, "Column {} has illegal data type {}, expected DateTime64", + column.name, column.type->getName()); + } +} + +void TimeSeriesColumnsValidator::validateColumnForTimestamp(const ColumnDescription & column, UInt32 & out_scale) const +{ + auto maybe_datetime64_type = removeNullable(column.type); + if (!isDateTime64(maybe_datetime64_type)) + { + throw Exception(ErrorCodes::INCOMPATIBLE_COLUMNS, "Column {} has illegal data type {}, expected DateTime64", + column.name, column.type->getName()); + } + const auto & datetime64_type = typeid_cast(*maybe_datetime64_type); + out_scale = datetime64_type.getScale(); +} + +void TimeSeriesColumnsValidator::validateColumnForValue(const ColumnDescription & column) const +{ + if (!isFloat(removeNullable(column.type))) + { + throw Exception(ErrorCodes::INCOMPATIBLE_COLUMNS, "Column {} has illegal data type {}, expected Float32 or Float64", + column.name, column.type->getName()); + } +} + +void TimeSeriesColumnsValidator::validateColumnForMetricName(const ColumnDescription & column) const +{ + validateColumnForTagValue(column); +} + +void TimeSeriesColumnsValidator::validateColumnForMetricName(const ColumnWithTypeAndName & column) const +{ + validateColumnForTagValue(column); +} + +void TimeSeriesColumnsValidator::validateColumnForTagValue(const ColumnDescription & column) const +{ + if (!isString(removeLowCardinalityAndNullable(column.type))) + { + throw Exception(ErrorCodes::INCOMPATIBLE_COLUMNS, "Column {} has illegal data type {}, expected String or LowCardinality(String)", + column.name, column.type->getName()); + } +} + +void TimeSeriesColumnsValidator::validateColumnForTagValue(const ColumnWithTypeAndName & column) const +{ + if (!isString(removeLowCardinalityAndNullable(column.type))) + { + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Column {} has illegal data type {}, expected String or LowCardinality(String)", + column.name, column.type->getName()); + } +} + +void TimeSeriesColumnsValidator::validateColumnForTagsMap(const ColumnDescription & column) const +{ + if (!isMap(column.type) + || !isString(removeLowCardinality(typeid_cast(*column.type).getKeyType())) + || !isString(removeLowCardinality(typeid_cast(*column.type).getValueType()))) + { + throw Exception(ErrorCodes::INCOMPATIBLE_COLUMNS, "Column {} has illegal data type {}, expected Map(String, String) or Map(LowCardinality(String), String)", + column.name, column.type->getName()); + } +} + +void TimeSeriesColumnsValidator::validateColumnForTagsMap(const ColumnWithTypeAndName & column) const +{ + if (!isMap(column.type) + || !isString(removeLowCardinality(typeid_cast(*column.type).getKeyType())) + || !isString(removeLowCardinality(typeid_cast(*column.type).getValueType()))) + { + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Column {} has illegal data type {}, expected Map(String, String) or Map(LowCardinality(String), String)", + column.name, column.type->getName()); + } +} + +void TimeSeriesColumnsValidator::validateColumnForMetricFamilyName(const ColumnDescription & column) const +{ + if (!isString(removeLowCardinalityAndNullable(column.type))) + { + throw Exception(ErrorCodes::INCOMPATIBLE_COLUMNS, "Column {} has illegal data type {}, expected String or LowCardinality(String)", + column.name, column.type->getName()); + } +} + +void TimeSeriesColumnsValidator::validateColumnForType(const ColumnDescription & column) const +{ + validateColumnForMetricFamilyName(column); +} + +void TimeSeriesColumnsValidator::validateColumnForUnit(const ColumnDescription & column) const +{ + validateColumnForMetricFamilyName(column); +} + +void TimeSeriesColumnsValidator::validateColumnForHelp(const ColumnDescription & column) const +{ + validateColumnForMetricFamilyName(column); +} + +} diff --git a/src/Storages/TimeSeries/TimeSeriesColumnsValidator.h b/src/Storages/TimeSeries/TimeSeriesColumnsValidator.h new file mode 100644 index 00000000000..43a54bf2ad6 --- /dev/null +++ b/src/Storages/TimeSeries/TimeSeriesColumnsValidator.h @@ -0,0 +1,55 @@ +#pragma once + +#include +#include + + +namespace DB +{ +class ColumnsDescription; +struct ColumnDescription; +struct ColumnWithTypeAndName; +struct TimeSeriesSettings; + +/// Checks the types of columns of a TimeSeries table. +class TimeSeriesColumnsValidator +{ +public: + /// Constructor stores a reference to argument `time_series_settings_` (it's unnecessary to copy it). + TimeSeriesColumnsValidator(StorageID time_series_storage_id_, + std::reference_wrapper time_series_settings_); + + /// Checks the columns of a TimeSeries table and throws an exception if some of the required columns don't exist or have illegal types. + void validateColumns(const ColumnsDescription & columns) const; + + /// Checks columns of a target table that a TimeSeries table is going to use. + /// Throws an exception if some of the required columns don't exist or have illegal types. + void validateTargetColumns(ViewTarget::Kind target_kind, const StorageID & target_table_id, const ColumnsDescription & target_columns) const; + + /// Each of the following functions validates a specific column type. + void validateColumnForID(const ColumnDescription & column, bool check_default = true) const; + void validateColumnForTimestamp(const ColumnDescription & column) const; + void validateColumnForTimestamp(const ColumnDescription & column, UInt32 & out_scale) const; + void validateColumnForValue(const ColumnDescription & column) const; + + void validateColumnForMetricName(const ColumnDescription & column) const; + void validateColumnForMetricName(const ColumnWithTypeAndName & column) const; + void validateColumnForTagValue(const ColumnDescription & column) const; + void validateColumnForTagValue(const ColumnWithTypeAndName & column) const; + void validateColumnForTagsMap(const ColumnDescription & column) const; + void validateColumnForTagsMap(const ColumnWithTypeAndName & column) const; + + void validateColumnForMetricFamilyName(const ColumnDescription & column) const; + void validateColumnForType(const ColumnDescription & column) const; + void validateColumnForUnit(const ColumnDescription & column) const; + void validateColumnForHelp(const ColumnDescription & column) const; + +private: + void validateColumnsImpl(const ColumnsDescription & columns) const; + void validateTargetColumnsImpl(ViewTarget::Kind target_kind, const ColumnsDescription & target_columns) const; + + const StorageID time_series_storage_id; + const TimeSeriesSettings & time_series_settings; +}; + +} diff --git a/src/Storages/TimeSeries/TimeSeriesDefinitionNormalizer.cpp b/src/Storages/TimeSeries/TimeSeriesDefinitionNormalizer.cpp new file mode 100644 index 00000000000..746a6a28274 --- /dev/null +++ b/src/Storages/TimeSeries/TimeSeriesDefinitionNormalizer.cpp @@ -0,0 +1,473 @@ +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int INCOMPATIBLE_COLUMNS; + extern const int INCORRECT_QUERY; +} + + +TimeSeriesDefinitionNormalizer::TimeSeriesDefinitionNormalizer(StorageID time_series_storage_id_, + std::reference_wrapper time_series_settings_, + const ASTCreateQuery * as_create_query_) + : time_series_storage_id(std::move(time_series_storage_id_)) + , time_series_settings(time_series_settings_) + , as_create_query(as_create_query_) +{ +} + + +void TimeSeriesDefinitionNormalizer::normalize(ASTCreateQuery & create_query) const +{ + reorderColumns(create_query); + addMissingColumns(create_query); + addMissingDefaultForIDColumn(create_query); + + if (as_create_query) + addMissingInnerEnginesFromAsTable(create_query); + + addMissingInnerEngines(create_query); +} + + +void TimeSeriesDefinitionNormalizer::reorderColumns(ASTCreateQuery & create) const +{ + if (!create.columns_list || !create.columns_list->columns) + return; + + auto & columns = create.columns_list->columns->children; + + /// Build a map "column_name -> column_declaration". + std::unordered_map> columns_by_name; + for (const auto & column : columns) + { + auto column_declaration = typeid_cast>(column); + columns_by_name[column_declaration->name] = column_declaration; + } + + /// Remove all columns and then add them again in the canonical order. + columns.clear(); + + auto add_column_in_correct_order = [&](std::string_view column_name) + { + auto it = columns_by_name.find(column_name); + if (it != columns_by_name.end()) + { + /// Add the column back to the list. + columns.push_back(it->second); + + /// Remove the column from the map to allow the check at the end of this function + /// that all columns from the original list are added back to the list. + columns_by_name.erase(it); + } + }; + + /// Reorder columns for the "data" table. + add_column_in_correct_order(TimeSeriesColumnNames::ID); + add_column_in_correct_order(TimeSeriesColumnNames::Timestamp); + add_column_in_correct_order(TimeSeriesColumnNames::Value); + + /// Reorder columns for the "tags" table. + add_column_in_correct_order(TimeSeriesColumnNames::MetricName); + + const Map & tags_to_columns = time_series_settings.tags_to_columns; + for (const auto & tag_name_and_column_name : tags_to_columns) + { + const auto & tuple = tag_name_and_column_name.safeGet(); + const auto & column_name = tuple.at(1).safeGet(); + add_column_in_correct_order(column_name); + } + + add_column_in_correct_order(TimeSeriesColumnNames::Tags); + add_column_in_correct_order(TimeSeriesColumnNames::AllTags); + + if (time_series_settings.store_min_time_and_max_time) + { + add_column_in_correct_order(TimeSeriesColumnNames::MinTime); + add_column_in_correct_order(TimeSeriesColumnNames::MaxTime); + } + + /// Reorder columns for the "metrics" table. + add_column_in_correct_order(TimeSeriesColumnNames::MetricFamilyName); + add_column_in_correct_order(TimeSeriesColumnNames::Type); + add_column_in_correct_order(TimeSeriesColumnNames::Unit); + add_column_in_correct_order(TimeSeriesColumnNames::Help); + + /// All columns from the original list must be added back to the list. + if (!columns_by_name.empty()) + { + throw Exception( + ErrorCodes::INCOMPATIBLE_COLUMNS, + "{}: Column {} can't be used in this table. " + "The TimeSeries table engine supports only a limited set of columns (id, timestamp, value, metric_name, tags, metric_family_name, type, unit, help). " + "Extra columns representing tags must be specified in the 'tags_to_columns' setting.", + time_series_storage_id.getNameForLogs(), columns_by_name.begin()->first); + } +} + + +void TimeSeriesDefinitionNormalizer::addMissingColumns(ASTCreateQuery & create) const +{ + if (!create.as_table.empty()) + { + /// If the create query has the "AS other_table" clause ("CREATE TABLE table AS other_table") + /// then all columns must be extracted from that "other_table". + /// Function InterpreterCreateQuery::getTablePropertiesAndNormalizeCreateQuery() will do that for us, + /// we don't need to fill missing columns by default in that case. + return; + } + + if (!create.columns_list) + create.set(create.columns_list, std::make_shared()); + + if (!create.columns_list->columns) + create.columns_list->set(create.columns_list->columns, std::make_shared()); + auto & columns = create.columns_list->columns->children; + + /// Here in this function we rely on that the columns are already sorted in the canonical order (see the reorderColumns() function). + /// NOTE: The order in which this function processes columns MUST be exactly the same as the order in reorderColumns(). + size_t position = 0; + + auto is_next_column_named = [&](std::string_view column_name) + { + if (position < columns.size() && (typeid_cast(*columns[position]).name == column_name)) + { + ++position; + return true; + } + return false; + }; + + auto make_new_column = [&](const String & column_name, ASTPtr type) + { + auto new_column = std::make_shared(); + new_column->name = column_name; + new_column->type = type; + columns.insert(columns.begin() + position, new_column); + ++position; + }; + + auto get_uuid_type = [] { return makeASTDataType("UUID"); }; + auto get_datetime_type = [] { return makeASTDataType("DateTime64", std::make_shared(3ul)); }; + auto get_float_type = [] { return makeASTDataType("Float64"); }; + auto get_string_type = [] { return makeASTDataType("String"); }; + auto get_lc_string_type = [&] { return makeASTDataType("LowCardinality", get_string_type()); }; + auto get_string_to_string_map_type = [&] { return makeASTDataType("Map", get_string_type(), get_string_type()); }; + auto get_lc_string_to_string_map_type = [&] { return makeASTDataType("Map", get_lc_string_type(), get_string_type()); }; + + auto make_nullable = [&](std::shared_ptr type) + { + if (type->name == "Nullable") + return type; + else + return makeASTDataType("Nullable", type); + }; + + /// Add missing columns for the "data" table. + if (!is_next_column_named(TimeSeriesColumnNames::ID)) + make_new_column(TimeSeriesColumnNames::ID, get_uuid_type()); + + if (!is_next_column_named(TimeSeriesColumnNames::Timestamp)) + make_new_column(TimeSeriesColumnNames::Timestamp, get_datetime_type()); + + auto timestamp_column = typeid_cast>(columns[position - 1]); + auto timestamp_type = typeid_cast>(timestamp_column->type->ptr()); + + if (!is_next_column_named(TimeSeriesColumnNames::Value)) + make_new_column(TimeSeriesColumnNames::Value, get_float_type()); + + /// Add missing columns for the "tags" table. + if (!is_next_column_named(TimeSeriesColumnNames::MetricName)) + { + /// We use 'LowCardinality(String)' as the default type of the `metric_name` column: + /// it looks like a correct optimization because there are shouldn't be too many different metrics. + make_new_column(TimeSeriesColumnNames::MetricName, get_lc_string_type()); + } + + const Map & tags_to_columns = time_series_settings.tags_to_columns; + for (const auto & tag_name_and_column_name : tags_to_columns) + { + const auto & tuple = tag_name_and_column_name.safeGet(); + const auto & column_name = tuple.at(1).safeGet(); + if (!is_next_column_named(column_name)) + make_new_column(column_name, get_string_type()); + } + + if (!is_next_column_named(TimeSeriesColumnNames::Tags)) + { + /// We use 'Map(LowCardinality(String), String)' as the default type of the `tags` column: + /// it looks like a correct optimization because there are shouldn't be too many different tag names. + make_new_column(TimeSeriesColumnNames::Tags, get_lc_string_to_string_map_type()); + } + + if (!is_next_column_named(TimeSeriesColumnNames::AllTags)) + { + /// The `all_tags` column is virtual (it's calculated on the fly and never stored anywhere) + /// so here we don't need to use the LowCardinality optimization as for the `tags` column. + make_new_column(TimeSeriesColumnNames::AllTags, get_string_to_string_map_type()); + } + + if (time_series_settings.store_min_time_and_max_time) + { + /// We use Nullable(DateTime64(3)) as the default type of the `min_time` and `max_time` columns. + /// It's nullable because it allows the aggregation (see aggregate_min_time_and_max_time) work correctly even + /// for rows in the "tags" table which doesn't have `min_time` and `max_time` (because they have no matching rows in the "data" table). + + if (!is_next_column_named(TimeSeriesColumnNames::MinTime)) + make_new_column(TimeSeriesColumnNames::MinTime, make_nullable(timestamp_type)); + if (!is_next_column_named(TimeSeriesColumnNames::MaxTime)) + make_new_column(TimeSeriesColumnNames::MaxTime, make_nullable(timestamp_type)); + } + + /// Add missing columns for the "metrics" table. + if (!is_next_column_named(TimeSeriesColumnNames::MetricFamilyName)) + make_new_column(TimeSeriesColumnNames::MetricFamilyName, get_string_type()); + + if (!is_next_column_named(TimeSeriesColumnNames::Type)) + make_new_column(TimeSeriesColumnNames::Type, get_string_type()); + + if (!is_next_column_named(TimeSeriesColumnNames::Unit)) + make_new_column(TimeSeriesColumnNames::Unit, get_string_type()); + + if (!is_next_column_named(TimeSeriesColumnNames::Help)) + make_new_column(TimeSeriesColumnNames::Help, get_string_type()); + + /// If the following fails that means the order in which columns are processed in this function doesn't match the order of columns in reorderColumns(). + chassert(position == columns.size()); +} + + +void TimeSeriesDefinitionNormalizer::addMissingDefaultForIDColumn(ASTCreateQuery & create) const +{ + /// Find the 'id' column and make a default expression for it. + if (!create.columns_list || !create.columns_list->columns) + return; + + auto & columns = create.columns_list->columns->children; + auto * it = std::find_if(columns.begin(), columns.end(), [](const ASTPtr & column) + { + return typeid_cast(*column).name == TimeSeriesColumnNames::ID; + }); + + if (it == columns.end()) + return; + + auto & column_declaration = typeid_cast(**it); + + /// We add a DEFAULT for the 'id' column only if it's not specified yet. + if (column_declaration.default_specifier.empty() && !column_declaration.default_expression) + { + column_declaration.default_specifier = "DEFAULT"; + column_declaration.default_expression = chooseIDAlgorithm(column_declaration); + } +} + + +ASTPtr TimeSeriesDefinitionNormalizer::chooseIDAlgorithm(const ASTColumnDeclaration & id_column) const +{ + /// Build a list of arguments for a hash function. + /// All hash functions below allow multiple arguments, so we use two arguments: metric_name, all_tags. + ASTs arguments_for_hash_function; + arguments_for_hash_function.push_back(std::make_shared(TimeSeriesColumnNames::MetricName)); + + if (time_series_settings.use_all_tags_column_to_generate_id) + { + arguments_for_hash_function.push_back(std::make_shared(TimeSeriesColumnNames::AllTags)); + } + else + { + const Map & tags_to_columns = time_series_settings.tags_to_columns; + for (const auto & tag_name_and_column_name : tags_to_columns) + { + const auto & tuple = tag_name_and_column_name.safeGet(); + const auto & column_name = tuple.at(1).safeGet(); + arguments_for_hash_function.push_back(std::make_shared(column_name)); + } + arguments_for_hash_function.push_back(std::make_shared(TimeSeriesColumnNames::Tags)); + } + + auto make_hash_function = [&](const String & function_name) + { + auto function = std::make_shared(); + function->name = function_name; + auto arguments_list = std::make_shared(); + arguments_list->children = std::move(arguments_for_hash_function); + function->arguments = arguments_list; + return function; + }; + + /// The type of a hash function depends on the type of the 'id' column. + auto id_type = DataTypeFactory::instance().get(id_column.type); + WhichDataType id_type_which(*id_type); + + if (id_type_which.isUInt64()) + { + return make_hash_function("sipHash64"); + } + else if (id_type_which.isFixedString() && typeid_cast(*id_type).getN() == 16) + { + return make_hash_function("sipHash128"); + } + else if (id_type_which.isUUID()) + { + return makeASTFunction("reinterpretAsUUID", make_hash_function("sipHash128")); + } + else if (id_type_which.isUInt128()) + { + return makeASTFunction("reinterpretAsUInt128", make_hash_function("sipHash128")); + } + else + { + throw Exception(ErrorCodes::INCOMPATIBLE_COLUMNS, "{}: The DEFAULT expression for column {} must contain an expression " + "which will be used to calculate the identifier of each time series: {} {} DEFAULT ... " + "If the DEFAULT expression is not specified then it can be chosen implicitly but only if the column type is one of these: UInt64, UInt128, UUID. " + "For type {} the DEFAULT expression can't be chosen automatically, so please specify it explicitly", + time_series_storage_id.getNameForLogs(), id_column.name, id_column.name, id_type->getName(), id_type->getName()); + } +} + + +void TimeSeriesDefinitionNormalizer::addMissingInnerEnginesFromAsTable(ASTCreateQuery & create) const +{ + if (!as_create_query) + return; + + for (auto target_kind : {ViewTarget::Data, ViewTarget::Tags, ViewTarget::Metrics}) + { + if (as_create_query->hasTargetTableID(target_kind)) + { + /// It's unlikely correct to use "CREATE table AS other_table" when "other_table" has external tables like this: + /// CREATE TABLE other_table ENGINE=TimeSeries data mydata + /// (because `table` would use the same table "mydata"). + /// Thus we just prohibit that. + QualifiedTableName as_table{as_create_query->getDatabase(), as_create_query->getTable()}; + throw Exception( + ErrorCodes::INCORRECT_QUERY, + "Cannot CREATE a table AS {}.{} because it has external tables", + backQuoteIfNeed(as_table.database), backQuoteIfNeed(as_table.table)); + } + + auto inner_table_engine = create.getTargetInnerEngine(target_kind); + if (!inner_table_engine) + { + /// Copy an inner engine's definition from the other table. + inner_table_engine = as_create_query->getTargetInnerEngine(target_kind); + if (inner_table_engine) + create.setTargetInnerEngine(target_kind, typeid_cast>(inner_table_engine->clone())); + } + } +} + + +void TimeSeriesDefinitionNormalizer::addMissingInnerEngines(ASTCreateQuery & create) const +{ + for (auto target_kind : {ViewTarget::Data, ViewTarget::Tags, ViewTarget::Metrics}) + { + if (create.hasTargetTableID(target_kind)) + continue; /// External target is set, inner engine is not needed. + + auto inner_table_engine = create.getTargetInnerEngine(target_kind); + if (inner_table_engine && inner_table_engine->engine) + continue; /// Engine is set already, skip it. + + if (!inner_table_engine) + { + /// Some part of storage definition (such as PARTITION BY) is specified, but the inner ENGINE is not: just set default one. + inner_table_engine = std::make_shared(); + create.setTargetInnerEngine(target_kind, inner_table_engine); + } + + /// Set engine by default. + setInnerEngineByDefault(target_kind, *inner_table_engine); + } +} + + +void TimeSeriesDefinitionNormalizer::setInnerEngineByDefault(ViewTarget::Kind inner_table_kind, ASTStorage & inner_storage_def) const +{ + switch (inner_table_kind) + { + case ViewTarget::Data: + { + inner_storage_def.set(inner_storage_def.engine, makeASTFunction("MergeTree")); + inner_storage_def.engine->no_empty_args = false; + + if (!inner_storage_def.order_by && !inner_storage_def.primary_key && inner_storage_def.engine->name.ends_with("MergeTree")) + { + inner_storage_def.set(inner_storage_def.order_by, + makeASTFunction("tuple", + std::make_shared(TimeSeriesColumnNames::ID), + std::make_shared(TimeSeriesColumnNames::Timestamp))); + } + break; + } + + case ViewTarget::Tags: + { + String engine_name; + if (time_series_settings.aggregate_min_time_and_max_time) + engine_name = "AggregatingMergeTree"; + else + engine_name = "ReplacingMergeTree"; + + inner_storage_def.set(inner_storage_def.engine, makeASTFunction(engine_name)); + inner_storage_def.engine->no_empty_args = false; + + if (!inner_storage_def.order_by && !inner_storage_def.primary_key && inner_storage_def.engine->name.ends_with("MergeTree")) + { + inner_storage_def.set(inner_storage_def.primary_key, + std::make_shared(TimeSeriesColumnNames::MetricName)); + + ASTs order_by_list; + order_by_list.push_back(std::make_shared(TimeSeriesColumnNames::MetricName)); + order_by_list.push_back(std::make_shared(TimeSeriesColumnNames::ID)); + + if (time_series_settings.store_min_time_and_max_time && !time_series_settings.aggregate_min_time_and_max_time) + { + order_by_list.push_back(std::make_shared(TimeSeriesColumnNames::MinTime)); + order_by_list.push_back(std::make_shared(TimeSeriesColumnNames::MaxTime)); + } + + auto order_by_tuple = std::make_shared(); + order_by_tuple->name = "tuple"; + auto arguments_list = std::make_shared(); + arguments_list->children = std::move(order_by_list); + order_by_tuple->arguments = arguments_list; + inner_storage_def.set(inner_storage_def.order_by, order_by_tuple); + } + break; + } + + case ViewTarget::Metrics: + { + inner_storage_def.set(inner_storage_def.engine, makeASTFunction("ReplacingMergeTree")); + inner_storage_def.engine->no_empty_args = false; + + if (!inner_storage_def.order_by && !inner_storage_def.primary_key && inner_storage_def.engine->name.ends_with("MergeTree")) + { + inner_storage_def.set(inner_storage_def.order_by, std::make_shared(TimeSeriesColumnNames::MetricFamilyName)); + } + break; + } + + default: + UNREACHABLE(); /// This function must not be called with any other `kind`. + } +} + +} diff --git a/src/Storages/TimeSeries/TimeSeriesDefinitionNormalizer.h b/src/Storages/TimeSeries/TimeSeriesDefinitionNormalizer.h new file mode 100644 index 00000000000..1f959eb3ce0 --- /dev/null +++ b/src/Storages/TimeSeries/TimeSeriesDefinitionNormalizer.h @@ -0,0 +1,55 @@ +#pragma once + +#include +#include + + +namespace DB +{ +class ASTColumnDeclaration; +class ASTCreateQuery; +struct ColumnDescription; +struct TimeSeriesSettings; + +/// Normalizes a TimeSeries table definition. +class TimeSeriesDefinitionNormalizer +{ +public: + /// Constructor stores a reference to argument `time_series_settings_` (it's unnecessary to copy it). + TimeSeriesDefinitionNormalizer(StorageID time_series_storage_id_, + std::reference_wrapper time_series_settings_, + const ASTCreateQuery * as_create_query_); + + /// Adds missing columns to the definition and reorders all the columns in the canonical way. + /// Also adds engines of inner tables to the definition if they aren't specified yet. + /// The `as_table_create_query` parameter must be nullptr if it isn't a "CREATE AS query". + void normalize(ASTCreateQuery & create_query) const; + +private: + /// Reorders existing columns in the canonical way. + void reorderColumns(ASTCreateQuery & create) const; + + /// Adds missing columns with data types set by default.. + void addMissingColumns(ASTCreateQuery & create) const; + + /// Adds the DEFAULT expression for the 'id' column if it isn't specified yet. + void addMissingDefaultForIDColumn(ASTCreateQuery & create) const; + + /// Generates a formulae for calculating the identifier of a time series from the metric name and all the tags. + ASTPtr chooseIDAlgorithm(const ASTColumnDeclaration & id_column) const; + + /// Copies the definitions of inner engines from "CREATE AS
" if this is that kind of query. + void addMissingInnerEnginesFromAsTable(ASTCreateQuery & create) const; + + /// Adds engines of inner tables to the definition if they aren't specified yet. + void addMissingInnerEngines(ASTCreateQuery & create) const; + + /// Sets the engine of an inner table by default. + void setInnerEngineByDefault(ViewTarget::Kind inner_table_kind, ASTStorage & inner_storage_def) const; + + const StorageID time_series_storage_id; + const TimeSeriesSettings & time_series_settings; + const ASTCreateQuery * as_create_query = nullptr; +}; + +} diff --git a/src/Storages/TimeSeries/TimeSeriesInnerTablesCreator.cpp b/src/Storages/TimeSeries/TimeSeriesInnerTablesCreator.cpp new file mode 100644 index 00000000000..5f616982a6f --- /dev/null +++ b/src/Storages/TimeSeries/TimeSeriesInnerTablesCreator.cpp @@ -0,0 +1,191 @@ +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace DB +{ + +TimeSeriesInnerTablesCreator::TimeSeriesInnerTablesCreator(ContextPtr context_, + StorageID time_series_storage_id_, + std::reference_wrapper time_series_columns_, + std::reference_wrapper time_series_settings_) + : WithContext(context_) + , time_series_storage_id(std::move(time_series_storage_id_)) + , time_series_columns(time_series_columns_) + , time_series_settings(time_series_settings_) +{ +} + +TimeSeriesInnerTablesCreator::~TimeSeriesInnerTablesCreator() = default; + + +ColumnsDescription TimeSeriesInnerTablesCreator::getInnerTableColumnsDescription(ViewTarget::Kind inner_table_kind) const +{ + ColumnsDescription columns; + + switch (inner_table_kind) + { + case ViewTarget::Data: + { + /// Column "id". + { + auto id_column = time_series_columns.get(TimeSeriesColumnNames::ID); + /// The expression for calculating the identifier of a time series can be transferred only to the "tags" inner table + /// (because it usually depends on columns like "metric_name" or "all_tags"). + id_column.default_desc = {}; + columns.add(std::move(id_column)); + } + + /// Column "timestamp". + columns.add(time_series_columns.get(TimeSeriesColumnNames::Timestamp)); + + /// Column "value". + columns.add(time_series_columns.get(TimeSeriesColumnNames::Value)); + break; + } + + case ViewTarget::Tags: + { + /// Column "id". + columns.add(time_series_columns.get(TimeSeriesColumnNames::ID)); + + /// Column "metric_name". + columns.add(time_series_columns.get(TimeSeriesColumnNames::MetricName)); + + /// Columns corresponding to specific tags specified in the "tags_to_columns" setting. + const Map & tags_to_columns = time_series_settings.tags_to_columns; + for (const auto & tag_name_and_column_name : tags_to_columns) + { + const auto & tuple = tag_name_and_column_name.safeGet(); + const auto & column_name = tuple.at(1).safeGet(); + columns.add(time_series_columns.get(column_name)); + } + + /// Column "tags". + columns.add(time_series_columns.get(TimeSeriesColumnNames::Tags)); + + /// Column "all_tags". + if (time_series_settings.use_all_tags_column_to_generate_id) + { + ColumnDescription all_tags_column = time_series_columns.get(TimeSeriesColumnNames::AllTags); + /// Column "all_tags" is here only to calculate the identifier of a time series for the "id" column, so it can be ephemeral. + all_tags_column.default_desc.kind = ColumnDefaultKind::Ephemeral; + if (!all_tags_column.default_desc.expression) + { + all_tags_column.default_desc.ephemeral_default = true; + all_tags_column.default_desc.expression = makeASTFunction("defaultValueOfTypeName", std::make_shared(all_tags_column.type->getName())); + } + columns.add(std::move(all_tags_column)); + } + + /// Columns "min_time" and "max_time". + if (time_series_settings.store_min_time_and_max_time) + { + auto min_time_column = time_series_columns.get(TimeSeriesColumnNames::MinTime); + auto max_time_column = time_series_columns.get(TimeSeriesColumnNames::MaxTime); + if (time_series_settings.aggregate_min_time_and_max_time) + { + AggregateFunctionProperties properties; + auto min_function = AggregateFunctionFactory::instance().get("min", NullsAction::EMPTY, {min_time_column.type}, {}, properties); + auto custom_name = std::make_unique(min_function, DataTypes{min_time_column.type}, Array{}); + min_time_column.type = DataTypeFactory::instance().getCustom(std::make_unique(std::move(custom_name))); + + auto max_function = AggregateFunctionFactory::instance().get("max", NullsAction::EMPTY, {max_time_column.type}, {}, properties); + custom_name = std::make_unique(max_function, DataTypes{max_time_column.type}, Array{}); + max_time_column.type = DataTypeFactory::instance().getCustom(std::make_unique(std::move(custom_name))); + } + columns.add(std::move(min_time_column)); + columns.add(std::move(max_time_column)); + } + + break; + } + + case ViewTarget::Metrics: + { + columns.add(time_series_columns.get(TimeSeriesColumnNames::MetricFamilyName)); + columns.add(time_series_columns.get(TimeSeriesColumnNames::Type)); + columns.add(time_series_columns.get(TimeSeriesColumnNames::Unit)); + columns.add(time_series_columns.get(TimeSeriesColumnNames::Help)); + break; + } + + default: + UNREACHABLE(); + } + + return columns; +} + + +StorageID TimeSeriesInnerTablesCreator::getInnerTableID(ViewTarget::Kind inner_table_kind, const UUID & inner_table_uuid) const +{ + StorageID res = time_series_storage_id; + if (time_series_storage_id.hasUUID()) + res.table_name = fmt::format(".inner_id.{}.{}", toString(inner_table_kind), time_series_storage_id.uuid); + else + res.table_name = fmt::format(".inner.{}.{}", toString(inner_table_kind), time_series_storage_id.table_name); + res.uuid = inner_table_uuid; + return res; +} + + +std::shared_ptr TimeSeriesInnerTablesCreator::getInnerTableCreateQuery( + ViewTarget::Kind inner_table_kind, + const UUID & inner_table_uuid, + const std::shared_ptr & inner_storage_def) const +{ + auto manual_create_query = std::make_shared(); + + auto inner_table_id = getInnerTableID(inner_table_kind, inner_table_uuid); + manual_create_query->setDatabase(inner_table_id.database_name); + manual_create_query->setTable(inner_table_id.table_name); + manual_create_query->uuid = inner_table_id.uuid; + manual_create_query->has_uuid = inner_table_id.uuid != UUIDHelpers::Nil; + + auto new_columns_list = std::make_shared(); + new_columns_list->set(new_columns_list->columns, InterpreterCreateQuery::formatColumns(getInnerTableColumnsDescription(inner_table_kind))); + manual_create_query->set(manual_create_query->columns_list, new_columns_list); + + if (inner_storage_def) + manual_create_query->set(manual_create_query->storage, inner_storage_def->clone()); + + return manual_create_query; +} + +StorageID TimeSeriesInnerTablesCreator::createInnerTable( + ViewTarget::Kind inner_table_kind, + const UUID & inner_table_uuid, + const std::shared_ptr & inner_storage_def) const +{ + /// We will make a query to create the inner target table. + auto create_context = Context::createCopy(getContext()); + + auto manual_create_query = getInnerTableCreateQuery(inner_table_kind, inner_table_uuid, inner_storage_def); + + /// Create the inner target table. + InterpreterCreateQuery create_interpreter(manual_create_query, create_context); + create_interpreter.setInternal(true); + create_interpreter.execute(); + + return DatabaseCatalog::instance().getTable({manual_create_query->getDatabase(), manual_create_query->getTable()}, getContext())->getStorageID(); +} + +} diff --git a/src/Storages/TimeSeries/TimeSeriesInnerTablesCreator.h b/src/Storages/TimeSeries/TimeSeriesInnerTablesCreator.h new file mode 100644 index 00000000000..5778dd77398 --- /dev/null +++ b/src/Storages/TimeSeries/TimeSeriesInnerTablesCreator.h @@ -0,0 +1,47 @@ +#pragma once + +#include +#include + + +namespace DB +{ +class ASTCreateQuery; +class ColumnsDescription; +struct TimeSeriesSettings; + +/// Generates inner tables for the TimeSeries table engine. +class TimeSeriesInnerTablesCreator : public WithContext +{ +public: + /// Constructor stores references to arguments `time_series_columns_` and `time_series_settings_` (it's unnecessary to copy them). + TimeSeriesInnerTablesCreator(ContextPtr context_, + StorageID time_series_storage_id_, + std::reference_wrapper time_series_columns_, + std::reference_wrapper time_series_settings_); + + ~TimeSeriesInnerTablesCreator(); + + /// Returns a column description of an inner table. + ColumnsDescription getInnerTableColumnsDescription(ViewTarget::Kind inner_table_kind) const; + + /// Returns a StorageID of an inner table. + StorageID getInnerTableID(ViewTarget::Kind inner_table_kind, const UUID & inner_table_uuid) const; + + /// Generates a CREATE TABLE query for an inner table. + std::shared_ptr getInnerTableCreateQuery(ViewTarget::Kind inner_table_kind, + const UUID & inner_table_uuid, + const std::shared_ptr & inner_storage_def) const; + + /// Creates an inner table. + StorageID createInnerTable(ViewTarget::Kind inner_table_kind, + const UUID & inner_table_uuid, + const std::shared_ptr & inner_storage_def) const; + +private: + const StorageID time_series_storage_id; + const ColumnsDescription & time_series_columns; + const TimeSeriesSettings & time_series_settings; +}; + +} diff --git a/src/Storages/TimeSeries/TimeSeriesSettings.cpp b/src/Storages/TimeSeries/TimeSeriesSettings.cpp new file mode 100644 index 00000000000..3a15be59191 --- /dev/null +++ b/src/Storages/TimeSeries/TimeSeriesSettings.cpp @@ -0,0 +1,34 @@ +#include + +#include +#include + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int UNKNOWN_SETTING; +} + +IMPLEMENT_SETTINGS_TRAITS(TimeSeriesSettingsTraits, LIST_OF_TIME_SERIES_SETTINGS) + +void TimeSeriesSettings::loadFromQuery(ASTStorage & storage_def) +{ + if (storage_def.settings) + { + try + { + applyChanges(storage_def.settings->changes); + } + catch (Exception & e) + { + if (e.code() == ErrorCodes::UNKNOWN_SETTING) + e.addMessage("for storage " + storage_def.engine->name); + throw; + } + } +} + +} diff --git a/src/Storages/TimeSeries/TimeSeriesSettings.h b/src/Storages/TimeSeries/TimeSeriesSettings.h new file mode 100644 index 00000000000..4dc6a436cd0 --- /dev/null +++ b/src/Storages/TimeSeries/TimeSeriesSettings.h @@ -0,0 +1,29 @@ +#pragma once + +#include + + +namespace DB +{ +class ASTStorage; + +#define LIST_OF_TIME_SERIES_SETTINGS(M, ALIAS) \ + M(Map, tags_to_columns, Map{}, "Map specifying which tags should be put to separate columns of the 'tags' table. Syntax: {'tag1': 'column1', 'tag2' : column2, ...}", 0) \ + M(Bool, use_all_tags_column_to_generate_id, true, "When generating an expression to calculate an identifier of a time series, this flag enables using the 'all_tags' column in that calculation. The 'all_tags' is a virtual column containing all tags except the metric name", 0) \ + M(Bool, store_min_time_and_max_time, true, "If set to true then the table will store 'min_time' and 'max_time' for each time series", 0) \ + M(Bool, aggregate_min_time_and_max_time, true, "When creating an inner target 'tags' table, this flag enables using 'SimpleAggregateFunction(min, Nullable(DateTime64(3)))' instead of just 'Nullable(DateTime64(3))' as the type of the 'min_time' column, and the same for the 'max_time' column", 0) \ + M(Bool, filter_by_min_time_and_max_time, true, "If set to true then the table will use the 'min_time' and 'max_time' columns for filtering time series", 0) \ + +DECLARE_SETTINGS_TRAITS(TimeSeriesSettingsTraits, LIST_OF_TIME_SERIES_SETTINGS) + +/// Settings for the TimeSeries table engine. +/// Could be loaded from a CREATE TABLE query (SETTINGS clause). For example: +/// CREATE TABLE mytable ENGINE = TimeSeries() SETTINGS tags_to_columns = {'job':'job', 'instance':'instance'} DATA ENGINE = ReplicatedMergeTree('zkpath', 'replica'), ... +struct TimeSeriesSettings : public BaseSettings +{ + void loadFromQuery(ASTStorage & storage_def); +}; + +using TimeSeriesSettingsPtr = std::shared_ptr; + +} diff --git a/src/Storages/TimeSeries/TimeSeriesTagNames.h b/src/Storages/TimeSeries/TimeSeriesTagNames.h new file mode 100644 index 00000000000..23b005ed414 --- /dev/null +++ b/src/Storages/TimeSeries/TimeSeriesTagNames.h @@ -0,0 +1,13 @@ +#pragma once + + +namespace DB +{ + +/// Label names with special meaning. +struct TimeSeriesTagNames +{ + static constexpr const char * MetricName = "__name__"; +}; + +} diff --git a/src/Storages/VirtualColumnUtils.cpp b/src/Storages/VirtualColumnUtils.cpp index 90c2c7f93c1..d932f5cc469 100644 --- a/src/Storages/VirtualColumnUtils.cpp +++ b/src/Storages/VirtualColumnUtils.cpp @@ -1,4 +1,3 @@ -#include #include #include #include @@ -37,11 +36,17 @@ #include #include +#include #include +#include +#include +#include +#include #include "Functions/FunctionsLogical.h" #include "Functions/IFunction.h" #include "Functions/IFunctionAdaptors.h" #include "Functions/indexHint.h" +#include #include #include #include @@ -51,6 +56,11 @@ namespace DB { +namespace ErrorCodes +{ + extern const int INCORRECT_DATA; +} + namespace VirtualColumnUtils { @@ -116,10 +126,32 @@ void filterBlockWithExpression(const ExpressionActionsPtr & actions, Block & blo NameSet getVirtualNamesForFileLikeStorage() { - return {"_path", "_file", "_size", "_time"}; + return {"_path", "_file", "_size", "_time", "_etag"}; } -VirtualColumnsDescription getVirtualsForFileLikeStorage(const ColumnsDescription & storage_columns) +std::unordered_map parseHivePartitioningKeysAndValues(const String & path, const ColumnsDescription & storage_columns) +{ + std::string pattern = "([^/]+)=([^/]+)/"; + re2::StringPiece input_piece(path); + + std::unordered_map key_values; + std::string key, value; + std::unordered_set used_keys; + while (RE2::FindAndConsume(&input_piece, pattern, &key, &value)) + { + if (used_keys.contains(key)) + throw Exception(ErrorCodes::INCORRECT_DATA, "Path '{}' to file with enabled hive-style partitioning contains duplicated partition key {}, only unique keys are allowed", path, key); + used_keys.insert(key); + + auto col_name = "_" + key; + while (storage_columns.has(col_name)) + col_name = "_" + col_name; + key_values[col_name] = value; + } + return key_values; +} + +VirtualColumnsDescription getVirtualsForFileLikeStorage(const ColumnsDescription & storage_columns, const ContextPtr & context, const std::string & path, std::optional format_settings_) { VirtualColumnsDescription desc; @@ -135,6 +167,23 @@ VirtualColumnsDescription getVirtualsForFileLikeStorage(const ColumnsDescription add_virtual("_file", std::make_shared(std::make_shared())); add_virtual("_size", makeNullable(std::make_shared())); add_virtual("_time", makeNullable(std::make_shared())); + add_virtual("_etag", std::make_shared(std::make_shared())); + + if (context->getSettingsRef().use_hive_partitioning) + { + auto map = parseHivePartitioningKeysAndValues(path, storage_columns); + auto format_settings = format_settings_ ? *format_settings_ : getFormatSettings(context); + for (auto & item : map) + { + auto type = tryInferDataTypeByEscapingRule(item.second, format_settings, FormatSettings::EscapingRule::Raw); + if (type == nullptr) + type = std::make_shared(); + if (type->canBeInsideLowCardinality()) + add_virtual(item.first, std::make_shared(type)); + else + add_virtual(item.first, type); + } + } return desc; } @@ -195,8 +244,12 @@ ColumnPtr getFilterByPathAndFileIndexes(const std::vector & paths, const void addRequestedFileLikeStorageVirtualsToChunk( Chunk & chunk, const NamesAndTypesList & requested_virtual_columns, - VirtualsForFileLikeStorage virtual_values) + VirtualsForFileLikeStorage virtual_values, ContextPtr context, const ColumnsDescription & columns) { + std::unordered_map hive_map; + if (context->getSettingsRef().use_hive_partitioning) + hive_map = parseHivePartitioningKeysAndValues(virtual_values.path, columns); + for (const auto & virtual_column : requested_virtual_columns) { if (virtual_column.name == "_path") @@ -230,6 +283,17 @@ void addRequestedFileLikeStorageVirtualsToChunk( else chunk.addColumn(virtual_column.type->createColumnConstWithDefaultValue(chunk.getNumRows())->convertToFullColumnIfConst()); } + else if (auto it = hive_map.find(virtual_column.getNameInStorage()); it != hive_map.end()) + { + chunk.addColumn(virtual_column.type->createColumnConst(chunk.getNumRows(), convertFieldToType(Field(it->second), *virtual_column.type))->convertToFullColumnIfConst()); + } + else if (virtual_column.name == "_etag") + { + if (virtual_values.etag) + chunk.addColumn(virtual_column.type->createColumnConst(chunk.getNumRows(), (*virtual_values.etag))->convertToFullColumnIfConst()); + else + chunk.addColumn(virtual_column.type->createColumnConstWithDefaultValue(chunk.getNumRows())->convertToFullColumnIfConst()); + } } } @@ -273,10 +337,7 @@ bool isDeterministicInScopeOfQuery(const ActionsDAG::Node * node) } static const ActionsDAG::Node * splitFilterNodeForAllowedInputs( - const ActionsDAG::Node * node, - const Block * allowed_inputs, - ActionsDAG::Nodes & additional_nodes, - bool allow_non_deterministic_functions) + const ActionsDAG::Node * node, const Block * allowed_inputs, ActionsDAG::Nodes & additional_nodes, bool allow_partial_result) { if (node->type == ActionsDAG::ActionType::FUNCTION) { @@ -285,13 +346,14 @@ 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, allow_non_deterministic_functions)) + if (const auto * child_copy + = splitFilterNodeForAllowedInputs(child, allowed_inputs, additional_nodes, allow_partial_result)) 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 + /// Expression like (now_allowed AND allowed) is not allowed if allow_partial_result = 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) + else if (!allow_partial_result) return nullptr; if (node_copy.children.empty()) @@ -300,7 +362,7 @@ static const ActionsDAG::Node * splitFilterNodeForAllowedInputs( if (node_copy.children.size() == 1) { const ActionsDAG::Node * res = node_copy.children.front(); - /// Expression like (not_allowed AND 256) can't be resuced to (and(256)) because AND requires + /// Expression like (not_allowed AND 256) can't be reduced to (and(256)) because AND requires /// at least two arguments; also it can't be reduced to (256) because result type is different. if (!res->result_type->equals(*node->result_type)) { @@ -318,7 +380,7 @@ 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, allow_non_deterministic_functions); !child) + if (child = splitFilterNodeForAllowedInputs(child, allowed_inputs, additional_nodes, allow_partial_result); !child) return nullptr; return &node_copy; @@ -332,7 +394,8 @@ 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, allow_non_deterministic_functions)) + if (const auto * child_copy + = splitFilterNodeForAllowedInputs(output, allowed_inputs, additional_nodes, allow_partial_result)) atoms.push_back(child_copy); if (!atoms.empty()) @@ -366,22 +429,24 @@ static const ActionsDAG::Node * splitFilterNodeForAllowedInputs( return node; } -std::optional splitFilterDagForAllowedInputs(const ActionsDAG::Node * predicate, const Block * allowed_inputs, bool allow_non_deterministic_functions) +std::optional +splitFilterDagForAllowedInputs(const ActionsDAG::Node * predicate, const Block * allowed_inputs, bool allow_partial_result) { if (!predicate) return {}; ActionsDAG::Nodes additional_nodes; - const auto * res = splitFilterNodeForAllowedInputs(predicate, allowed_inputs, additional_nodes, allow_non_deterministic_functions); + const auto * res = splitFilterNodeForAllowedInputs(predicate, allowed_inputs, additional_nodes, allow_partial_result); if (!res) return {}; return ActionsDAG::cloneSubDAG({res}, true); } -void filterBlockWithPredicate(const ActionsDAG::Node * predicate, Block & block, ContextPtr context) +void filterBlockWithPredicate( + const ActionsDAG::Node * predicate, Block & block, ContextPtr context, bool allow_filtering_with_partial_predicate) { - auto dag = splitFilterDagForAllowedInputs(predicate, &block, /*allow_non_deterministic_functions=*/ false); + auto dag = splitFilterDagForAllowedInputs(predicate, &block, /*allow_partial_result=*/allow_filtering_with_partial_predicate); if (dag) filterBlockWithExpression(buildFilterExpression(std::move(*dag), context), block); } diff --git a/src/Storages/VirtualColumnUtils.h b/src/Storages/VirtualColumnUtils.h index 73b7908b75c..23e16871798 100644 --- a/src/Storages/VirtualColumnUtils.h +++ b/src/Storages/VirtualColumnUtils.h @@ -5,6 +5,7 @@ #include #include #include +#include #include @@ -18,9 +19,21 @@ class NamesAndTypesList; namespace VirtualColumnUtils { -/// Similar to filterBlockWithQuery, but uses ActionsDAG as a predicate. +/// The filtering functions are tricky to use correctly. +/// There are 2 ways: +/// 1. Call filterBlockWithPredicate() or filterBlockWithExpression() inside SourceStepWithFilter::applyFilters(). +/// 2. Call splitFilterDagForAllowedInputs() and buildSetsForDAG() inside SourceStepWithFilter::applyFilters(). +/// Then call filterBlockWithPredicate() or filterBlockWithExpression() in initializePipeline(). +/// +/// Otherwise calling filter*() outside applyFilters() will throw "Not-ready Set is passed" +/// if there are subqueries. +/// +/// Similar to filterBlockWithExpression(buildFilterExpression(splitFilterDagForAllowedInputs(...)))./// Similar to filterBlockWithQuery, but uses ActionsDAG as a predicate. /// Basically it is filterBlockWithDAG(splitFilterDagForAllowedInputs). -void filterBlockWithPredicate(const ActionsDAG::Node * predicate, Block & block, ContextPtr context); +/// If allow_filtering_with_partial_predicate is true, then the filtering will be done even if some part of the predicate +/// cannot be evaluated using the columns from the block. +void filterBlockWithPredicate(const ActionsDAG::Node * predicate, Block & block, ContextPtr context, bool allow_filtering_with_partial_predicate = true); + /// Just filters block. Block should contain all the required columns. ExpressionActionsPtr buildFilterExpression(ActionsDAG dag, ContextPtr context); @@ -33,15 +46,15 @@ void buildSetsForDAG(const ActionsDAG & dag, const ContextPtr & context); bool isDeterministicInScopeOfQuery(const ActionsDAG::Node * node); /// Extract a part of predicate that can be evaluated using only columns from input_names. -/// 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 +/// When allow_partial_result is false, then the result will be empty if any part of if cannot be evaluated deterministically +/// on the given inputs. +/// allow_partial_result 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. -std::optional splitFilterDagForAllowedInputs(const ActionsDAG::Node * predicate, const Block * allowed_inputs, bool allow_non_deterministic_functions = true); +std::optional splitFilterDagForAllowedInputs(const ActionsDAG::Node * predicate, const Block * allowed_inputs, bool allow_partial_result = true); /// Extract from the input stream a set of `name` column values template @@ -51,12 +64,16 @@ auto extractSingleValueFromBlock(const Block & block, const String & name) const ColumnWithTypeAndName & data = block.getByName(name); size_t rows = block.rows(); for (size_t i = 0; i < rows; ++i) - res.insert((*data.column)[i].get()); + res.insert((*data.column)[i].safeGet()); return res; } NameSet getVirtualNamesForFileLikeStorage(); -VirtualColumnsDescription getVirtualsForFileLikeStorage(const ColumnsDescription & storage_columns); +VirtualColumnsDescription getVirtualsForFileLikeStorage( + const ColumnsDescription & storage_columns, + const ContextPtr & context, + const std::string & sample_path = "", + std::optional format_settings_ = std::nullopt); std::optional createPathAndFileFilterDAG(const ActionsDAG::Node * predicate, const NamesAndTypesList & virtual_columns); @@ -83,12 +100,12 @@ struct VirtualsForFileLikeStorage std::optional size { std::nullopt }; const String * filename { nullptr }; std::optional last_modified { std::nullopt }; - + const String * etag { nullptr }; }; void addRequestedFileLikeStorageVirtualsToChunk( Chunk & chunk, const NamesAndTypesList & requested_virtual_columns, - VirtualsForFileLikeStorage virtual_values); + VirtualsForFileLikeStorage virtual_values, ContextPtr context, const ColumnsDescription & columns); } } diff --git a/src/Storages/WindowView/StorageWindowView.cpp b/src/Storages/WindowView/StorageWindowView.cpp index 65bf6768b1b..5830c844582 100644 --- a/src/Storages/WindowView/StorageWindowView.cpp +++ b/src/Storages/WindowView/StorageWindowView.cpp @@ -1055,6 +1055,8 @@ void StorageWindowView::threadFuncFireProc() /// TODO: consider using time_t instead (for every timestamp in this class) UInt32 timestamp_now = now(); + LOG_TRACE(log, "Now: {}, next fire signal: {}, max watermark: {}", timestamp_now, next_fire_signal, max_watermark); + while (next_fire_signal <= timestamp_now) { try @@ -1072,6 +1074,9 @@ void StorageWindowView::threadFuncFireProc() if (slide_kind > IntervalKind::Kind::Day) slide_interval *= 86400; next_fire_signal += slide_interval; + + LOG_TRACE(log, "Now: {}, next fire signal: {}, max watermark: {}, max fired watermark: {}, slide interval: {}", + timestamp_now, next_fire_signal, max_watermark, max_fired_watermark, slide_interval); } if (max_watermark >= timestamp_now) @@ -1433,13 +1438,16 @@ void StorageWindowView::writeIntoWindowView( while (window_view.modifying_query) std::this_thread::sleep_for(std::chrono::milliseconds(100)); - if (!window_view.is_proctime && window_view.max_watermark == 0 && block.rows() > 0) + const size_t block_rows = block.rows(); + if (!window_view.is_proctime && window_view.max_watermark == 0 && block_rows > 0) { std::lock_guard lock(window_view.fire_signal_mutex); const auto & window_column = block.getByName(window_view.timestamp_column_name); const ColumnUInt32::Container & window_end_data = static_cast(*window_column.column).getData(); UInt32 first_record_timestamp = window_end_data[0]; window_view.max_watermark = window_view.getWindowUpperBound(first_record_timestamp); + + LOG_TRACE(window_view.log, "New max watermark: {}", window_view.max_watermark); } Pipe pipe(std::make_shared(block)); @@ -1649,6 +1657,8 @@ void StorageWindowView::writeIntoWindowView( auto executor = builder.execute(); executor->execute(builder.getNumThreads(), local_context->getSettingsRef().use_concurrency_control); + + LOG_TRACE(window_view.log, "Wrote {} rows into inner table ({})", block_rows, inner_table->getStorageID().getFullTableName()); } void StorageWindowView::startup() diff --git a/src/Storages/fuzzers/CMakeLists.txt b/src/Storages/fuzzers/CMakeLists.txt index e36fccec8df..f67552716a2 100644 --- a/src/Storages/fuzzers/CMakeLists.txt +++ b/src/Storages/fuzzers/CMakeLists.txt @@ -1,7 +1,7 @@ clickhouse_add_executable (mergetree_checksum_fuzzer mergetree_checksum_fuzzer.cpp) # Look at comment around fuzz_compression target declaration -target_link_libraries (mergetree_checksum_fuzzer PRIVATE dbms) +target_link_libraries (mergetree_checksum_fuzzer PRIVATE dbms clickhouse_functions) clickhouse_add_executable (columns_description_fuzzer columns_description_fuzzer.cpp) -target_link_libraries (columns_description_fuzzer PRIVATE dbms clickhouse_functions) +target_link_libraries (columns_description_fuzzer PRIVATE clickhouse_functions) diff --git a/src/Storages/fuzzers/columns_description_fuzzer.cpp b/src/Storages/fuzzers/columns_description_fuzzer.cpp index 469bfe2fb38..e39afccd1f9 100644 --- a/src/Storages/fuzzers/columns_description_fuzzer.cpp +++ b/src/Storages/fuzzers/columns_description_fuzzer.cpp @@ -1,4 +1,5 @@ #include +#include #include diff --git a/src/Storages/getStructureOfRemoteTable.cpp b/src/Storages/getStructureOfRemoteTable.cpp index 9d23f132759..1408e120bc5 100644 --- a/src/Storages/getStructureOfRemoteTable.cpp +++ b/src/Storages/getStructureOfRemoteTable.cpp @@ -102,16 +102,16 @@ ColumnsDescription getStructureOfRemoteTableInShard( { ColumnDescription column; - column.name = (*name)[i].get(); + column.name = (*name)[i].safeGet(); - String data_type_name = (*type)[i].get(); + String data_type_name = (*type)[i].safeGet(); column.type = data_type_factory.get(data_type_name); - String kind_name = (*default_kind)[i].get(); + String kind_name = (*default_kind)[i].safeGet(); if (!kind_name.empty()) { column.default_desc.kind = columnDefaultKindFromString(kind_name); - String expr_str = (*default_expr)[i].get(); + String expr_str = (*default_expr)[i].safeGet(); column.default_desc.expression = parseQuery( expr_parser, expr_str.data(), expr_str.data() + expr_str.size(), "default expression", 0, settings.max_parser_depth, settings.max_parser_backtracks); @@ -207,8 +207,8 @@ ColumnsDescriptionByShardNum getExtendedObjectsOfRemoteTables( size_t size = name_col.size(); for (size_t i = 0; i < size; ++i) { - auto name = name_col[i].get(); - auto type_name = type_col[i].get(); + auto name = name_col[i].safeGet(); + auto type_name = type_col[i].safeGet(); auto storage_column = storage_columns.tryGetPhysical(name); if (storage_column && storage_column->type->hasDynamicSubcolumnsDeprecated()) diff --git a/src/Storages/registerStorages.cpp b/src/Storages/registerStorages.cpp index 8f33314397c..4ed74763810 100644 --- a/src/Storages/registerStorages.cpp +++ b/src/Storages/registerStorages.cpp @@ -26,6 +26,9 @@ void registerStorageGenerateRandom(StorageFactory & factory); void registerStorageExecutable(StorageFactory & factory); void registerStorageWindowView(StorageFactory & factory); void registerStorageLoop(StorageFactory & factory); +void registerStorageFuzzQuery(StorageFactory & factory); +void registerStorageTimeSeries(StorageFactory & factory); + #if USE_RAPIDJSON || USE_SIMDJSON void registerStorageFuzzJSON(StorageFactory & factory); #endif @@ -126,6 +129,9 @@ void registerStorages() registerStorageExecutable(factory); registerStorageWindowView(factory); registerStorageLoop(factory); + registerStorageFuzzQuery(factory); + registerStorageTimeSeries(factory); + #if USE_RAPIDJSON || USE_SIMDJSON registerStorageFuzzJSON(factory); #endif diff --git a/src/Storages/tests/gtest_transform_query_for_external_database.cpp b/src/Storages/tests/gtest_transform_query_for_external_database.cpp index 6765e112bb9..5a63c118e2d 100644 --- a/src/Storages/tests/gtest_transform_query_for_external_database.cpp +++ b/src/Storages/tests/gtest_transform_query_for_external_database.cpp @@ -7,6 +7,7 @@ #include #include #include +#include #include #include #include @@ -74,6 +75,7 @@ private: {"a", std::make_shared()}, {"b", std::make_shared()}, {"foo", std::make_shared()}, + {"is_value", DataTypeFactory::instance().get("Bool")}, }), TableWithColumnNamesAndTypes( createDBAndTable("table2"), @@ -411,6 +413,14 @@ TEST(TransformQueryForExternalDatabase, Analyzer) R"(SELECT "column" FROM "test"."table")"); check(state, 1, {"column", "apply_id", "apply_type", "apply_status", "create_time", "field", "value", "a", "b", "foo"}, - "SELECT * FROM table WHERE (column) IN (1)", + "SELECT * EXCEPT (is_value) FROM table WHERE (column) IN (1)", R"(SELECT "column", "apply_id", "apply_type", "apply_status", "create_time", "field", "value", "a", "b", "foo" FROM "test"."table" WHERE "column" IN (1))"); + + check(state, 1, {"is_value"}, + "SELECT is_value FROM table WHERE is_value = true", + R"(SELECT "is_value" FROM "test"."table" WHERE "is_value" = true)"); + + check(state, 1, {"is_value"}, + "SELECT is_value FROM table WHERE is_value = 1", + R"(SELECT "is_value" FROM "test"."table" WHERE "is_value" = 1)"); } diff --git a/src/TableFunctions/TableFunctionExplain.cpp b/src/TableFunctions/TableFunctionExplain.cpp index 552b9fde986..69d24c879bd 100644 --- a/src/TableFunctions/TableFunctionExplain.cpp +++ b/src/TableFunctions/TableFunctionExplain.cpp @@ -83,7 +83,7 @@ void TableFunctionExplain::parseArguments(const ASTPtr & ast_function, ContextPt "Table function '{}' requires a String argument for EXPLAIN kind, got '{}'", getName(), queryToString(kind_arg)); - ASTExplainQuery::ExplainKind kind = ASTExplainQuery::fromString(kind_literal->value.get()); + ASTExplainQuery::ExplainKind kind = ASTExplainQuery::fromString(kind_literal->value.safeGet()); auto explain_query = std::make_shared(kind); const auto * settings_arg = function->arguments->children[1]->as(); @@ -92,7 +92,7 @@ void TableFunctionExplain::parseArguments(const ASTPtr & ast_function, ContextPt "Table function '{}' requires a serialized string settings argument, got '{}'", getName(), queryToString(function->arguments->children[1])); - const auto & settings_str = settings_arg->value.get(); + const auto & settings_str = settings_arg->value.safeGet(); if (!settings_str.empty()) { const Settings & settings = context->getSettingsRef(); diff --git a/src/TableFunctions/TableFunctionFile.cpp b/src/TableFunctions/TableFunctionFile.cpp index af327cfe54e..5cd249f000d 100644 --- a/src/TableFunctions/TableFunctionFile.cpp +++ b/src/TableFunctions/TableFunctionFile.cpp @@ -48,7 +48,7 @@ void TableFunctionFile::parseFirstArguments(const ASTPtr & arg, const ContextPtr else if (type == Field::Types::Int64 || type == Field::Types::UInt64) { fd = static_cast( - (type == Field::Types::Int64) ? literal->value.get() : literal->value.get()); + (type == Field::Types::Int64) ? literal->value.safeGet() : literal->value.safeGet()); if (fd < 0) throw Exception(ErrorCodes::BAD_ARGUMENTS, "File descriptor must be non-negative"); } diff --git a/src/TableFunctions/TableFunctionFuzzQuery.cpp b/src/TableFunctions/TableFunctionFuzzQuery.cpp new file mode 100644 index 00000000000..224f6666556 --- /dev/null +++ b/src/TableFunctions/TableFunctionFuzzQuery.cpp @@ -0,0 +1,54 @@ +#include + +#include +#include +#include +#include + +namespace DB +{ + + +namespace ErrorCodes +{ + extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; +} + +void TableFunctionFuzzQuery::parseArguments(const ASTPtr & ast_function, ContextPtr context) +{ + ASTs & args_func = ast_function->children; + + if (args_func.size() != 1) + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Table function '{}' must have arguments", getName()); + + auto args = args_func.at(0)->children; + configuration = StorageFuzzQuery::getConfiguration(args, context); +} + +StoragePtr TableFunctionFuzzQuery::executeImpl( + const ASTPtr & /*ast_function*/, + ContextPtr context, + const std::string & table_name, + ColumnsDescription /*cached_columns*/, + bool is_insert_query) const +{ + ColumnsDescription columns = getActualTableStructure(context, is_insert_query); + auto res = std::make_shared( + StorageID(getDatabaseName(), table_name), + columns, + /* comment */ String{}, + configuration); + res->startup(); + return res; +} + +void registerTableFunctionFuzzQuery(TableFunctionFactory & factory) +{ + factory.registerFunction( + {.documentation + = {.description = "Perturbs a query string with random variations.", + .returned_value = "A table object with a single column containing perturbed query strings."}, + .allow_readonly = true}); +} + +} diff --git a/src/TableFunctions/TableFunctionFuzzQuery.h b/src/TableFunctions/TableFunctionFuzzQuery.h new file mode 100644 index 00000000000..22d10341c4d --- /dev/null +++ b/src/TableFunctions/TableFunctionFuzzQuery.h @@ -0,0 +1,42 @@ +#pragma once + +#include + +#include +#include +#include + +#include "config.h" + +namespace DB +{ + +class TableFunctionFuzzQuery : public ITableFunction +{ +public: + static constexpr auto name = "fuzzQuery"; + std::string getName() const override { return name; } + + void parseArguments(const ASTPtr & ast_function, ContextPtr context) override; + + ColumnsDescription getActualTableStructure(ContextPtr /* context */, bool /* is_insert_query */) const override + { + return ColumnsDescription{{"query", std::make_shared()}}; + } + +private: + StoragePtr executeImpl( + const ASTPtr & ast_function, + ContextPtr context, + const std::string & table_name, + ColumnsDescription cached_columns, + bool is_insert_query) const override; + + const char * getStorageTypeName() const override { return "fuzzQuery"; } + + String source; + std::optional random_seed; + StorageFuzzQuery::Configuration configuration; +}; + +} diff --git a/src/TableFunctions/TableFunctionMergeTreeIndex.cpp b/src/TableFunctions/TableFunctionMergeTreeIndex.cpp index 06a48f0e25f..27ed50fb711 100644 --- a/src/TableFunctions/TableFunctionMergeTreeIndex.cpp +++ b/src/TableFunctions/TableFunctionMergeTreeIndex.cpp @@ -76,9 +76,9 @@ void TableFunctionMergeTreeIndex::parseArguments(const ASTPtr & ast_function, Co "Table function '{}' expected bool flag for 'with_marks' argument", getName()); if (value.getType() == Field::Types::Bool) - with_marks = value.get(); + with_marks = value.safeGet(); else - with_marks = value.get(); + with_marks = value.safeGet(); } if (!params.empty()) diff --git a/src/TableFunctions/TableFunctionObjectStorage.cpp b/src/TableFunctions/TableFunctionObjectStorage.cpp index 550d9cc799b..39392a4c44c 100644 --- a/src/TableFunctions/TableFunctionObjectStorage.cpp +++ b/src/TableFunctions/TableFunctionObjectStorage.cpp @@ -84,7 +84,8 @@ ColumnsDescription TableFunctionObjectStorage< context->checkAccess(getSourceAccessType()); ColumnsDescription columns; auto storage = getObjectStorage(context, !is_insert_query); - resolveSchemaAndFormat(columns, configuration->format, storage, configuration, std::nullopt, context); + std::string sample_path; + resolveSchemaAndFormat(columns, configuration->format, storage, configuration, std::nullopt, sample_path, context); return columns; } else diff --git a/src/TableFunctions/TableFunctionRemote.cpp b/src/TableFunctions/TableFunctionRemote.cpp index e60c31b2d77..8a877ff0802 100644 --- a/src/TableFunctions/TableFunctionRemote.cpp +++ b/src/TableFunctions/TableFunctionRemote.cpp @@ -306,21 +306,7 @@ StoragePtr TableFunctionRemote::executeImpl(const ASTPtr & /*ast_function*/, Con cached_columns = getActualTableStructure(context, is_insert_query); assert(cluster); - StoragePtr res = remote_table_function_ptr - ? std::make_shared( - StorageID(getDatabaseName(), table_name), - cached_columns, - ConstraintsDescription{}, - remote_table_function_ptr, - String{}, - context, - sharding_key, - String{}, - String{}, - DistributedSettings{}, - LoadingStrictnessLevel::CREATE, - cluster) - : std::make_shared( + StoragePtr res = std::make_shared( StorageID(getDatabaseName(), table_name), cached_columns, ConstraintsDescription{}, @@ -334,7 +320,9 @@ StoragePtr TableFunctionRemote::executeImpl(const ASTPtr & /*ast_function*/, Con String{}, DistributedSettings{}, LoadingStrictnessLevel::CREATE, - cluster); + cluster, + remote_table_function_ptr, + !is_cluster_function); res->startup(); return res; diff --git a/src/TableFunctions/TableFunctionTimeSeries.cpp b/src/TableFunctions/TableFunctionTimeSeries.cpp new file mode 100644 index 00000000000..62ea088eba0 --- /dev/null +++ b/src/TableFunctions/TableFunctionTimeSeries.cpp @@ -0,0 +1,128 @@ +#include + +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int BAD_ARGUMENTS; + extern const int LOGICAL_ERROR; + extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH; +} + + +template +void TableFunctionTimeSeriesTarget::parseArguments(const ASTPtr & ast_function, ContextPtr context) +{ + const auto & args_func = ast_function->as(); + + if (!args_func.arguments) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Table function '{}' must have arguments.", name); + + auto & args = args_func.arguments->children; + + if ((args.size() != 1) && (args.size() != 2)) + throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, + "Table function '{}' requires one or two arguments: {}([database, ] time_series_table)", name, name); + + if (args.size() == 1) + { + /// timeSeriesMetrics( [my_db.]my_time_series_table ) + if (const auto * id = args[0]->as()) + { + if (auto table_id = id->createTable()) + time_series_storage_id = table_id->getTableId(); + } + } + + if (time_series_storage_id.empty()) + { + for (auto & arg : args) + arg = evaluateConstantExpressionOrIdentifierAsLiteral(arg, context); + + if (args.size() == 1) + { + /// timeSeriesMetrics( 'my_time_series_table' ) + time_series_storage_id.table_name = checkAndGetLiteralArgument(args[0], "table_name"); + } + else + { + /// timeSeriesMetrics( 'mydb', 'my_time_series_table' ) + time_series_storage_id.database_name = checkAndGetLiteralArgument(args[0], "database_name"); + time_series_storage_id.table_name = checkAndGetLiteralArgument(args[1], "table_name"); + } + } + + if (time_series_storage_id.empty()) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Couldn't get a table name from the arguments of the {} table function", name); + + time_series_storage_id = context->resolveStorageID(time_series_storage_id); + target_table_type_name = getTargetTable(context)->getName(); +} + + +template +StoragePtr TableFunctionTimeSeriesTarget::getTargetTable(const ContextPtr & context) const +{ + auto time_series_storage = storagePtrToTimeSeries(DatabaseCatalog::instance().getTable(time_series_storage_id, context)); + return time_series_storage->getTargetTable(target_kind, context); +} + + +template +StoragePtr TableFunctionTimeSeriesTarget::executeImpl( + const ASTPtr & /* ast_function */, + ContextPtr context, + const String & /* table_name */, + ColumnsDescription /* cached_columns */, + bool /* is_insert_query */) const +{ + return getTargetTable(context); +} + +template +ColumnsDescription TableFunctionTimeSeriesTarget::getActualTableStructure(ContextPtr context, bool /* is_insert_query */) const +{ + return getTargetTable(context)->getInMemoryMetadataPtr()->columns; +} + +template +const char * TableFunctionTimeSeriesTarget::getStorageTypeName() const +{ + return target_table_type_name.c_str(); +} + + +void registerTableFunctionTimeSeries(TableFunctionFactory & factory) +{ + factory.registerFunction>( + {.documentation = { + .description=R"(Provides direct access to the 'data' target table for a specified TimeSeries table.)", + .examples{{"timeSeriesData", "SELECT * from timeSeriesData('mydb', 'time_series_table');", ""}}, + .categories{"Time Series"}} + }); + factory.registerFunction>( + {.documentation = { + .description=R"(Provides direct access to the 'tags' target table for a specified TimeSeries table.)", + .examples{{"timeSeriesTags", "SELECT * from timeSeriesTags('mydb', 'time_series_table');", ""}}, + .categories{"Time Series"}} + }); + factory.registerFunction>( + {.documentation = { + .description=R"(Provides direct access to the 'metrics' target table for a specified TimeSeries table.)", + .examples{{"timeSeriesMetrics", "SELECT * from timeSeriesMetrics('mydb', 'time_series_table');", ""}}, + .categories{"Time Series"}} + }); +} + +} diff --git a/src/TableFunctions/TableFunctionTimeSeries.h b/src/TableFunctions/TableFunctionTimeSeries.h new file mode 100644 index 00000000000..57654413fe4 --- /dev/null +++ b/src/TableFunctions/TableFunctionTimeSeries.h @@ -0,0 +1,42 @@ +#pragma once + +#include +#include +#include + + +namespace DB +{ + +/// Table functions timeSeriesData('mydb', 'my_ts_table'), timeSeriesTags('mydb', 'my_ts_table'), timeSeriesMetrics('mydb', 'my_ts_table') +/// return the data table, the tags table, and the metrics table respectively associated with any TimeSeries table mydb.my_ts_table +template +class TableFunctionTimeSeriesTarget : public ITableFunction +{ +public: + static constexpr auto name = (target_kind == ViewTarget::Data) + ? "timeSeriesData" + : ((target_kind == ViewTarget::Tags) ? "timeSeriesTags" : "timeSeriesMetrics"); + + String getName() const override { return name; } + +private: + void parseArguments(const ASTPtr & ast_function, ContextPtr context) override; + + StoragePtr executeImpl( + const ASTPtr & ast_function, + ContextPtr context, + const std::string & table_name, + ColumnsDescription cached_columns, + bool is_insert_query) const override; + + ColumnsDescription getActualTableStructure(ContextPtr context, bool is_insert_query) const override; + const char * getStorageTypeName() const override; + + StoragePtr getTargetTable(const ContextPtr & context) const; + + StorageID time_series_storage_id = StorageID::createEmpty(); + String target_table_type_name; +}; + +} diff --git a/src/TableFunctions/registerTableFunctions.cpp b/src/TableFunctions/registerTableFunctions.cpp index ca4913898f9..0b21de00f86 100644 --- a/src/TableFunctions/registerTableFunctions.cpp +++ b/src/TableFunctions/registerTableFunctions.cpp @@ -26,6 +26,7 @@ void registerTableFunctions() registerTableFunctionMongoDB(factory); registerTableFunctionRedis(factory); registerTableFunctionMergeTreeIndex(factory); + registerTableFunctionFuzzQuery(factory); #if USE_RAPIDJSON || USE_SIMDJSON registerTableFunctionFuzzJSON(factory); #endif @@ -56,6 +57,7 @@ void registerTableFunctions() registerTableFunctionFormat(factory); registerTableFunctionExplain(factory); + registerTableFunctionTimeSeries(factory); registerTableFunctionObjectStorage(factory); registerTableFunctionObjectStorageCluster(factory); diff --git a/src/TableFunctions/registerTableFunctions.h b/src/TableFunctions/registerTableFunctions.h index efde4d6dcdc..1dd6341b67e 100644 --- a/src/TableFunctions/registerTableFunctions.h +++ b/src/TableFunctions/registerTableFunctions.h @@ -23,6 +23,7 @@ void registerTableFunctionGenerate(TableFunctionFactory & factory); void registerTableFunctionMongoDB(TableFunctionFactory & factory); void registerTableFunctionRedis(TableFunctionFactory & factory); void registerTableFunctionMergeTreeIndex(TableFunctionFactory & factory); +void registerTableFunctionFuzzQuery(TableFunctionFactory & factory); #if USE_RAPIDJSON || USE_SIMDJSON void registerTableFunctionFuzzJSON(TableFunctionFactory & factory); #endif @@ -67,6 +68,8 @@ void registerTableFunctionObjectStorage(TableFunctionFactory & factory); void registerTableFunctionObjectStorageCluster(TableFunctionFactory & factory); void registerDataLakeTableFunctions(TableFunctionFactory & factory); +void registerTableFunctionTimeSeries(TableFunctionFactory & factory); + void registerTableFunctions(); } diff --git a/src/configure_config.cmake b/src/configure_config.cmake index d22bf674df4..5a1aa179e01 100644 --- a/src/configure_config.cmake +++ b/src/configure_config.cmake @@ -135,6 +135,12 @@ endif() if (TARGET ch_contrib::vectorscan) set(USE_VECTORSCAN 1) endif() +if (TARGET ch_contrib::qpl) + set(USE_QPL 1) +endif() +if (TARGET ch_contrib::qatlib) + set(USE_QATLIB 1) +endif() if (TARGET ch_contrib::avrocpp) set(USE_AVRO 1) endif() @@ -158,11 +164,14 @@ endif() if (TARGET ch_contrib::bcrypt) set(USE_BCRYPT 1) endif() +if (TARGET ch_contrib::usearch) + set(USE_USEARCH 1) +endif() if (TARGET ch_contrib::ssh) set(USE_SSH 1) endif() -if (TARGET ch_contrib::fiu) - set(FIU_ENABLE 1) +if (TARGET ch_contrib::libfiu) + set(USE_LIBFIU 1) endif() if (TARGET ch_contrib::libarchive) set(USE_LIBARCHIVE 1) diff --git a/tests/ci/changelog.py b/tests/ci/changelog.py index 929f0f3523a..554ba339892 100755 --- a/tests/ci/changelog.py +++ b/tests/ci/changelog.py @@ -19,7 +19,6 @@ from env_helper import TEMP_PATH from git_helper import git_runner, is_shallow from github_helper import GitHub, PullRequest, PullRequests, Repository from s3_helper import S3Helper -from get_robot_token import get_best_robot_token from ci_utils import Shell from version_helper import ( FILE_WITH_VERSION_PATH, @@ -38,7 +37,7 @@ categories_preferred_order = ( "Experimental Feature", "Performance Improvement", "Improvement", - "Critical Bug Fix (crash, LOGICAL_ERROR, data loss, RBAC)", + # "Critical Bug Fix (crash, LOGICAL_ERROR, data loss, RBAC)", "Bug Fix (user-visible misbehavior in an official stable release)", "Build/Testing/Packaging Improvement", "Other", @@ -115,7 +114,6 @@ def get_descriptions(prs: PullRequests) -> Dict[str, List[Description]]: # pylint: enable=protected-access if repo_name not in repos: repos[repo_name] = pr.base.repo - in_changelog = False merge_commit = pr.merge_commit_sha if merge_commit is None: logging.warning("PR %s does not have merge-commit, skipping", pr.number) @@ -173,7 +171,6 @@ def parse_args() -> argparse.Namespace: parser.add_argument( "--gh-user-or-token", help="user name or GH token to authenticate", - default=get_best_robot_token(), ) parser.add_argument( "--gh-password", @@ -291,10 +288,11 @@ def generate_description(item: PullRequest, repo: Repository) -> Optional[Descri # Normalize bug fixes if ( re.match( - r"(?i)bug\Wfix", + r".*(?i)bug\Wfix", category, ) - and "Critical Bug Fix" not in category + # Map "Critical Bug Fix" to "Bug fix" category for changelog + # and "Critical Bug Fix" not in category ): category = "Bug Fix (user-visible misbehavior in an official stable release)" diff --git a/tests/ci/ci.py b/tests/ci/ci.py index 8d0414ce7a8..49b597333dc 100644 --- a/tests/ci/ci.py +++ b/tests/ci/ci.py @@ -985,6 +985,7 @@ def _run_test(job_name: str, run_command: str) -> int: else: print("Use run command from the workflow") env["CHECK_NAME"] = job_name + env["MAX_RUN_TIME"] = str(timeout or 0) print(f"Going to start run command [{run_command}]") stopwatch = Stopwatch() job_log = Path(TEMP_PATH) / "job_log.txt" diff --git a/tests/ci/ci_buddy.py b/tests/ci/ci_buddy.py index f0e73e925fe..07f318207a4 100644 --- a/tests/ci/ci_buddy.py +++ b/tests/ci/ci_buddy.py @@ -1,7 +1,7 @@ import argparse import json import os -from typing import Union, Dict +from typing import Union, Dict, List import boto3 import requests @@ -9,20 +9,44 @@ from botocore.exceptions import ClientError from pr_info import PRInfo from ci_config import CI +from ci_utils import WithIter + + +class Channels(metaclass=WithIter): + # Channel names must match json keys in ParameterStore + ALERTS = "alerts-channel" + INFO = "info-channel" + DRY_RUN = "dry-ryn-channel" + DEFAULT = "default" class CIBuddy: + Channels = Channels _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 = "" + self.channels = {} if res: - self.test_channel = json.loads(res)["test_channel"] - self.dev_ci_channel = json.loads(res)["ci_channel"] + channels = json.loads(res) + for channel in Channels: + if channel in channels: + self.channels[channel] = channels[channel] + + for channel in Channels: + if channel not in self.channels: + if Channels.DEFAULT in self.channels: + print( + f"ERROR: missing config for channel [{channel}] - will use default channel instead" + ) + self.channels[channel] = self.channels[Channels.DEFAULT] + else: + print( + f"ERROR: missing config for channel [{channel}] - will disable notification" + ) + self.channels[channel] = "" self.job_name = os.getenv("CHECK_NAME", "unknown") pr_info = PRInfo() self.pr_number = pr_info.number @@ -63,22 +87,33 @@ class CIBuddy: 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 + def post(self, message: str, channels: List[str]) -> None: + print(f"Posting slack message, dry_run [{self.dry_run}]") + if self.dry_run: + urls = [self.channels[Channels.DRY_RUN]] else: - url = self.dev_ci_channel + urls = [] + for channel in channels: + url = self.channels[channel] + if url: + urls.append(url) + else: + print(f"WARNING: no channel config for [{channel}] - skip") data = {"text": message} try: - requests.post(url, headers=self._HEADERS, data=json.dumps(data), timeout=10) + for url in urls: + 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 + self, + title: str, + body: Union[Dict, str], + with_wf_link: bool, + channels: Union[List[str], str], ) -> None: message = title if isinstance(body, dict): @@ -96,31 +131,49 @@ class CIBuddy: run_id = os.getenv("GITHUB_RUN_ID", "") if with_wf_link and run_id: message += f" *workflow*: \n" - self.post(message) + self.post( + message, channels=[channels] if isinstance(channels, str) else channels + ) def post_info( - self, title: str, body: Union[Dict, str], with_wf_link: bool = True + self, + title: str, + body: Union[Dict, str], + with_wf_link: bool = True, + channels: Union[List[str], str] = Channels.INFO, ) -> None: title_extended = f":white_circle: *{title}*\n\n" - self._post_formatted(title_extended, body, with_wf_link) + self._post_formatted(title_extended, body, with_wf_link, channels=channels) def post_done( - self, title: str, body: Union[Dict, str], with_wf_link: bool = True + self, + title: str, + body: Union[Dict, str], + with_wf_link: bool = True, + channels: Union[List[str], str] = Channels.INFO, ) -> None: title_extended = f":white_check_mark: *{title}*\n\n" - self._post_formatted(title_extended, body, with_wf_link) + self._post_formatted(title_extended, body, with_wf_link, channels=channels) def post_warning( - self, title: str, body: Union[Dict, str], with_wf_link: bool = True + self, + title: str, + body: Union[Dict, str], + with_wf_link: bool = True, + channels: Union[List[str], str] = Channels.ALERTS, ) -> None: title_extended = f":warning: *{title}*\n\n" - self._post_formatted(title_extended, body, with_wf_link) + self._post_formatted(title_extended, body, with_wf_link, channels=channels) def post_critical( - self, title: str, body: Union[Dict, str], with_wf_link: bool = True + self, + title: str, + body: Union[Dict, str], + with_wf_link: bool = True, + channels: Union[List[str], str] = Channels.ALERTS, ) -> None: title_extended = f":black_circle: *{title}*\n\n" - self._post_formatted(title_extended, body, with_wf_link) + self._post_formatted(title_extended, body, with_wf_link, channels=channels) def post_job_error( self, @@ -129,6 +182,7 @@ class CIBuddy: with_instance_info: bool = True, with_wf_link: bool = True, critical: bool = False, + channel: Union[List[str], str] = Channels.ALERTS, ) -> None: instance_id, instance_type = "unknown", "unknown" if with_instance_info: @@ -159,7 +213,7 @@ class CIBuddy: run_id = os.getenv("GITHUB_RUN_ID", "") if with_wf_link and run_id: message += f" *workflow*: \n" - self.post(message) + self.post(message, channels=[channel] if isinstance(channel, str) else channel) def parse_args(): diff --git a/tests/ci/ci_cache.py b/tests/ci/ci_cache.py index a59fd3e5a29..9eeda7161ee 100644 --- a/tests/ci/ci_cache.py +++ b/tests/ci/ci_cache.py @@ -731,7 +731,8 @@ class CiCache: job_config=reference_config, ): remove_from_workflow.append(job_name) - has_test_jobs_to_skip = True + if job_name != CI.JobNames.DOCS_CHECK: + has_test_jobs_to_skip = True else: required_builds += ( job_config.required_builds if job_config.required_builds else [] diff --git a/tests/ci/ci_config.py b/tests/ci/ci_config.py index 8cb587a1062..7a19eb6f827 100644 --- a/tests/ci/ci_config.py +++ b/tests/ci/ci_config.py @@ -682,6 +682,34 @@ class CI: assert res, f"not a build [{build_name}] or invalid JobConfig" return res + @classmethod + def is_workflow_ok(cls) -> bool: + # TODO: temporary method to make Mergeable check working + res = cls.GH.get_workflow_results() + if not res: + print("ERROR: no workflow results found") + return False + for workflow_job, workflow_data in res.items(): + status = workflow_data["result"] + if status in ( + cls.GH.ActionStatuses.SUCCESS, + cls.GH.ActionStatuses.SKIPPED, + ): + print(f"Workflow status for [{workflow_job}] is [{status}] - continue") + elif status in (cls.GH.ActionStatuses.FAILURE,): + if workflow_job in ( + WorkflowStages.TESTS_2, + WorkflowStages.TESTS_2_WW, + ): + print( + f"Failed Workflow status for [{workflow_job}], it's not required - continue" + ) + continue + + print(f"Failed Workflow status for [{workflow_job}]") + return False + return True + if __name__ == "__main__": parser = ArgumentParser( diff --git a/tests/ci/ci_utils.py b/tests/ci/ci_utils.py index 067bedb19c3..b8778e0cc50 100644 --- a/tests/ci/ci_utils.py +++ b/tests/ci/ci_utils.py @@ -44,9 +44,10 @@ class GH: FAILURE = "failure" PENDING = "pending" SUCCESS = "success" + SKIPPED = "skipped" @classmethod - def _get_workflow_results(cls): + 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}]" @@ -65,13 +66,13 @@ class GH: @classmethod def print_workflow_results(cls): - res = cls._get_workflow_results() + 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 is_workflow_ok(cls) -> bool: - res = cls._get_workflow_results() + res = cls.get_workflow_results() for _job, data in res.items(): if data["result"] == "failure": return False @@ -79,7 +80,7 @@ class GH: @classmethod def get_workflow_job_result(cls, wf_job_name: str) -> Optional[str]: - res = cls._get_workflow_results() + res = cls.get_workflow_results() if wf_job_name in res: return res[wf_job_name]["result"] # type: ignore else: @@ -166,6 +167,11 @@ class GH: latest_branch = Shell.get_output( 'gh pr list --label release --repo ClickHouse/ClickHouse --search "sort:created" -L1 --json headRefName' ) + if latest_branch: + latest_branch = json.loads(latest_branch)[0]["headRefName"] + print( + f"Latest branch [{latest_branch}], release branch [{branch}], release latest [{latest_branch == branch}]" + ) return latest_branch == branch diff --git a/tests/ci/commit_status_helper.py b/tests/ci/commit_status_helper.py index dcca96370bd..a6287d5629b 100644 --- a/tests/ci/commit_status_helper.py +++ b/tests/ci/commit_status_helper.py @@ -497,9 +497,9 @@ def trigger_mergeable_check( description = format_description(description) if set_from_sync: - # update Mergeable Check from sync WF only if its status already present or its new status is not SUCCESS + # update Mergeable Check from sync WF only if its status already present or its new status is FAILURE # to avoid false-positives - if mergeable_status or state != SUCCESS: + if mergeable_status or state == FAILURE: set_mergeable_check(commit, description, state) elif mergeable_status is None or mergeable_status.description != description: set_mergeable_check(commit, description, state) diff --git a/tests/ci/create_release.py b/tests/ci/create_release.py index b4e08f29dbe..68268b033fe 100755 --- a/tests/ci/create_release.py +++ b/tests/ci/create_release.py @@ -61,6 +61,7 @@ class ReleaseContextManager: # create initial release info self.release_info = ReleaseInfo( release_branch="NA", + release_type="NA", commit_sha=args.ref, release_tag="NA", version="NA", @@ -93,6 +94,7 @@ class ReleaseContextManager: @dataclasses.dataclass class ReleaseInfo: version: str + release_type: str release_tag: str release_branch: str commit_sha: str @@ -131,7 +133,7 @@ class ReleaseInfo: return self def prepare( - self, commit_ref: str, release_type: str, skip_tag_check: bool + self, commit_ref: str, release_type: str, _skip_tag_check: bool ) -> "ReleaseInfo": version = None release_branch = None @@ -143,17 +145,18 @@ class ReleaseInfo: assert release_type in ("patch", "new") if release_type == "new": # check commit_ref is right and on a right branch - Shell.check( - f"git merge-base --is-ancestor {commit_ref} origin/master", - strict=True, - verbose=True, - ) + if commit_ref != "master": + Shell.check( + f"git merge-base --is-ancestor {commit_ref} origin/master", + strict=True, + verbose=True, + ) with checkout(commit_ref): commit_sha = Shell.get_output_or_raise(f"git rev-list -n1 {commit_ref}") # Git() must be inside "with checkout" contextmanager git = Git() version = get_version_from_repo(git=git) - release_branch = "master" + release_branch = f"{version.major}.{version.minor}" expected_prev_tag = f"v{version.major}.{version.minor}.1.1-new" version.bump().with_description(VersionType.NEW) assert ( @@ -204,10 +207,11 @@ class ReleaseInfo: expected_tag_prefix ) and git.latest_tag.endswith(expected_tag_suffix): pass - elif not skip_tag_check: - assert ( - False - ), f"BUG: Unexpected latest tag [{git.latest_tag}] expected [{expected_tag_prefix}*{expected_tag_suffix}]. Already Released?" + # TODO: uncomment and check with dry-run + # elif not skip_tag_check: + # assert ( + # False + # ), f"BUG: Unexpected latest tag [{git.latest_tag}] expected [{expected_tag_prefix}*{expected_tag_suffix}]. Already Released?" previous_release_sha = Shell.get_output_or_raise( f"git rev-list -n1 {previous_release_tag}" @@ -238,6 +242,7 @@ class ReleaseInfo: self.release_progress = ReleaseProgress.STARTED self.progress_status = ReleaseProgressDescription.OK self.latest = latest_release + self.release_type = release_type return self def push_release_tag(self, dry_run: bool) -> None: @@ -262,16 +267,15 @@ class ReleaseInfo: @staticmethod def _create_gh_label(label: str, color_hex: str, dry_run: bool) -> None: cmd = f"gh api repos/{CI.Envs.GITHUB_REPOSITORY}/labels -f name={label} -f color={color_hex}" - Shell.check(cmd, dry_run=dry_run, strict=True) + res = Shell.check(cmd, dry_run=dry_run, verbose=True) + if not res: + # not a critical error - do not fail. branch might be created already (recovery case) + print("WARNING: failed to create backport labels for the new branch") 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() + new_release_branch = self.release_branch version_after_release = copy(version) version_after_release.bump() assert ( @@ -285,11 +289,8 @@ class ReleaseInfo: print( f"Create and push new release branch [{new_release_branch}], commit [{self.commit_sha}]" ) - with checkout(self.release_branch): + with checkout("master"): 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}" ) @@ -302,67 +303,108 @@ class ReleaseInfo: ReleaseInfo._create_gh_label( f"v{new_release_branch}-affected", "c2bfff", dry_run=dry_run ) - Shell.check( - f"""gh pr create --repo {CI.Envs.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, - strict=True, - verbose=True, - ) def get_version_bump_branch(self): return f"bump_version_{self.version}" def update_version_and_contributors_list(self, dry_run: bool) -> None: - # Bump version, update contributors list, create PR - branch_upd_version_contributors = self.get_version_bump_branch() + # Bump version, update contributors list, create on release branch with checkout(self.commit_sha): git = Git() version = get_version_from_repo(git=git) - if self.release_branch == "master": + if self.release_type == "patch": + assert ( + version.string == self.version + ), f"BUG: version in release info does not match version in git commit, expected [{self.version}], got [{version.string}]" + version.bump_patch() + else: + version.reset_tweak() + version.with_description(version.get_stable_release_type()) + + with checkout(self.release_branch): + 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" + Shell.check( + cmd_commit_version_upd, strict=True, dry_run=dry_run, verbose=True + ) + Shell.check(cmd_push_branch, strict=True, dry_run=dry_run, verbose=True) + if dry_run: + Shell.check( + f"{GIT_PREFIX} diff '{CMAKE_PATH}' '{CONTRIBUTORS_PATH}'", + verbose=True, + ) + Shell.check( + f"{GIT_PREFIX} checkout '{CMAKE_PATH}' '{CONTRIBUTORS_PATH}'", + verbose=True, + ) + + # TODO: move to new GH step? + if self.release_type == "new": + print("Update version on master branch") + branch_upd_version_contributors = self.get_version_bump_branch() + with checkout(self.commit_sha): + git = Git() + version = get_version_from_repo(git=git) 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}" - actor = os.getenv("GITHUB_ACTOR", "") or "me" - body = f"Automatic version bump after release {self.release_tag}\n### Changelog category (leave one):\n- Not for changelog (changelog entry is not required)\n" - cmd_create_pr = f"gh pr create --repo {CI.Envs.GITHUB_REPOSITORY} --title 'Update version after release' --head {branch_upd_version_contributors} --base {self.release_branch} --body \"{body}\" --assignee {actor}" + with checkout("master"): + 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}" + actor = os.getenv("GITHUB_ACTOR", "") or "me" + body = f"Automatic version bump after release {self.release_tag}\n### Changelog category (leave one):\n- Not for changelog (changelog entry is not required)\n" + cmd_create_pr = f"gh pr create --repo {CI.Envs.GITHUB_REPOSITORY} --title 'Update version after release' --head {branch_upd_version_contributors} --base master --body \"{body}\" --assignee {actor}" + Shell.check( + cmd_commit_version_upd, + strict=True, + dry_run=dry_run, + verbose=True, + ) + Shell.check( + cmd_push_branch, strict=True, dry_run=dry_run, verbose=True + ) + Shell.check( + cmd_create_pr, strict=True, dry_run=dry_run, verbose=True + ) + if dry_run: + Shell.check( + f"{GIT_PREFIX} diff '{CMAKE_PATH}' '{CONTRIBUTORS_PATH}'", + verbose=True, + ) + Shell.check( + f"{GIT_PREFIX} checkout '{CMAKE_PATH}' '{CONTRIBUTORS_PATH}'", + verbose=True, + ) + self.version_bump_pr = "dry-run" + else: + self.version_bump_pr = GH.get_pr_url_by_branch( + branch=branch_upd_version_contributors + ) + + # TODO: move to new GH step? + print("Create Release PR") + with checkout(self.release_branch): + pr_labels = f"--label {CI.Labels.RELEASE}" + if version.get_stable_release_type() == VersionType.LTS: + pr_labels += f" --label {CI.Labels.RELEASE_LTS}" Shell.check( - cmd_commit_version_upd, strict=True, dry_run=dry_run, verbose=True + f"""gh pr create --repo {CI.Envs.GITHUB_REPOSITORY} --title 'Release pull request for branch {self.release_branch}' \ + --head {self.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, + strict=True, + verbose=True, ) - Shell.check(cmd_push_branch, strict=True, dry_run=dry_run, verbose=True) - Shell.check(cmd_create_pr, strict=True, dry_run=dry_run, verbose=True) - if dry_run: - Shell.check( - f"{GIT_PREFIX} diff '{CMAKE_PATH}' '{CONTRIBUTORS_PATH}'", - verbose=True, - ) - Shell.check( - f"{GIT_PREFIX} checkout '{CMAKE_PATH}' '{CONTRIBUTORS_PATH}'", - verbose=True, - ) - self.version_bump_pr = "dry-run" - else: - self.version_bump_pr = GH.get_pr_url_by_branch( - branch=branch_upd_version_contributors - ) def get_change_log_branch(self): return f"auto/{self.release_tag}" def update_release_info(self, dry_run: bool) -> "ReleaseInfo": - if self.release_branch != "master": + if self.release_type == "patch": if not self.changelog_pr: branch = self.get_change_log_branch() if not dry_run: @@ -371,21 +413,22 @@ class ReleaseInfo: url = "dry-run" print(f"ChangeLog PR url [{url}]") self.changelog_pr = url - - if not self.version_bump_pr: - branch = self.get_version_bump_branch() - if not dry_run: - url = GH.get_pr_url_by_branch(branch=branch) - else: - url = "dry-run" - print(f"Version bump PR url [{url}]") - self.version_bump_pr = url - - self.release_url = f"https://github.com/{CI.Envs.GITHUB_REPOSITORY}/releases/tag/{self.release_tag}" - print(f"Release url [{self.release_url}]") - self.docker = f"docker run --rm clickhouse/clickhouse:{self.version} clickhouse --version" + else: + # new release branch - find version bump pr on a master branch + branch = self.get_version_bump_branch() + if not dry_run: + url = GH.get_pr_url_by_branch(branch=branch) + else: + url = "dry-run" + print(f"Version bump PR url [{url}]") + self.version_bump_pr = url + + self.release_url = f"https://github.com/{CI.Envs.GITHUB_REPOSITORY}/releases/tag/{self.release_tag}" + print(f"Release url [{self.release_url}]") + self.dump() + return self def create_gh_release(self, packages_files: List[str], dry_run: bool) -> None: @@ -410,35 +453,40 @@ class ReleaseInfo: def merge_prs(self, dry_run: bool) -> None: repo = CI.Envs.GITHUB_REPOSITORY - assert self.version_bump_pr - if dry_run: - version_bump_pr_num = 12345 - else: - version_bump_pr_num = int(self.version_bump_pr.split("/")[-1]) - print("Merging Version bump PR") - res_1 = Shell.check( - f"gh pr merge {version_bump_pr_num} --repo {repo} --merge --auto", - verbose=True, - dry_run=dry_run, - ) - - res_2 = True - if not self.release_tag.endswith("-new"): + if self.release_type == "patch": assert self.changelog_pr print("Merging ChangeLog PR") if dry_run: changelog_pr_num = 23456 else: changelog_pr_num = int(self.changelog_pr.split("/")[-1]) - res_2 = Shell.check( + res = Shell.check( f"gh pr merge {changelog_pr_num} --repo {repo} --merge --auto", verbose=True, dry_run=dry_run, ) else: - assert not self.changelog_pr + if not dry_run: + assert not self.changelog_pr + res = True - self.prs_merged = res_1 and res_2 + if self.release_type == "new": + assert self.version_bump_pr + print("Merging Version Bump PR") + if dry_run: + version_bump_pr = 23456 + else: + version_bump_pr = int(self.version_bump_pr.split("/")[-1]) + res = res and Shell.check( + f"gh pr merge {version_bump_pr} --repo {repo} --merge --auto", + verbose=True, + dry_run=dry_run, + ) + else: + if not dry_run: + assert not self.version_bump_pr + + self.prs_merged = res class RepoTypes: @@ -759,7 +807,7 @@ if __name__ == "__main__": release_info.prepare( commit_ref=args.ref, release_type=args.release_type, - skip_tag_check=args.skip_tag_check, + _skip_tag_check=args.skip_tag_check, ) if args.download_packages: @@ -821,7 +869,9 @@ if __name__ == "__main__": else: title = "Failed: " + title CIBuddy(dry_run=args.dry_run).post_critical( - title, dataclasses.asdict(release_info) + title, + dataclasses.asdict(release_info), + channels=[CIBuddy.Channels.ALERTS, CIBuddy.Channels.INFO], ) if args.set_progress_started: diff --git a/tests/ci/docker_server.py b/tests/ci/docker_server.py index 3251ec5644e..34439c19f0a 100644 --- a/tests/ci/docker_server.py +++ b/tests/ci/docker_server.py @@ -70,7 +70,7 @@ def parse_args() -> argparse.Namespace: parser.add_argument( "--tag-type", type=str, - choices=("head", "release", "latest-release"), + choices=("head", "release", "release-latest"), default="head", help="defines required tags for resulting docker image. " "head - for master image (tag: head) " diff --git a/tests/ci/functional_test_check.py b/tests/ci/functional_test_check.py index 52970404d2d..b7391eff01b 100644 --- a/tests/ci/functional_test_check.py +++ b/tests/ci/functional_test_check.py @@ -114,6 +114,9 @@ def get_run_command( if flaky_check: envs.append("-e NUM_TRIES=50") envs.append("-e MAX_RUN_TIME=2800") + else: + max_run_time = os.getenv("MAX_RUN_TIME", "0") + envs.append(f"-e MAX_RUN_TIME={max_run_time}") envs += [f"-e {e}" for e in additional_envs] diff --git a/tests/ci/integration_tests_runner.py b/tests/ci/integration_tests_runner.py index 2b348be8b51..f5dbef4f6db 100755 --- a/tests/ci/integration_tests_runner.py +++ b/tests/ci/integration_tests_runner.py @@ -20,6 +20,7 @@ from typing import Any, Dict from env_helper import IS_CI from integration_test_images import IMAGES +from tee_popen import TeePopen MAX_RETRY = 1 NUM_WORKERS = 5 @@ -68,9 +69,9 @@ def get_changed_tests_to_run(pr_info, repo_path): return [] for fpath in changed_files: - if "tests/integration/test_" in fpath: + if re.search(r"tests/integration/test_.*/test.*\.py", fpath) is not None: logging.info("File %s changed and seems like integration test", fpath) - result.add(fpath.split("/")[2]) + result.add("/".join(fpath.split("/")[2:])) return filter_existing_tests(result, repo_path) @@ -356,20 +357,13 @@ class ClickhouseIntegrationTestsRunner: logging.info("Package found in %s", full_path) log_name = "install_" + f + ".log" log_path = os.path.join(str(self.path()), log_name) - with open(log_path, "w", encoding="utf-8") as log: - cmd = f"dpkg -x {full_path} ." - logging.info("Executing installation cmd %s", cmd) - with subprocess.Popen( - cmd, shell=True, stderr=log, stdout=log - ) as proc: - if proc.wait() == 0: - logging.info( - "Installation of %s successfull", full_path - ) - else: - raise RuntimeError( - f"Installation of {full_path} failed" - ) + cmd = f"dpkg -x {full_path} ." + logging.info("Executing installation cmd %s", cmd) + with TeePopen(cmd, log_file=log_path) as proc: + if proc.wait() == 0: + logging.info("Installation of %s successfull", full_path) + else: + raise RuntimeError(f"Installation of {full_path} failed") break else: raise FileNotFoundError(f"Package with {package} not found") @@ -784,8 +778,8 @@ class ClickhouseIntegrationTestsRunner: logging.info("Starting check with retries") final_retry = 0 logs = [] - tires_num = 1 if should_fail else FLAKY_TRIES_COUNT - for i in range(tires_num): + tries_num = 1 if should_fail else FLAKY_TRIES_COUNT + for i in range(tries_num): final_retry += 1 logging.info("Running tests for the %s time", i) counters, tests_times, log_paths = self.try_run_test_group( diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index d9e33229932..8f19dd7d023 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -75,7 +75,7 @@ def get_run_command( 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} " - "python3 ./utils/runner.py" + "python3 /usr/share/clickhouse-test/fuzz/runner.py" ) diff --git a/tests/ci/merge_pr.py b/tests/ci/merge_pr.py index 13c7537a84b..ddeb76adf7e 100644 --- a/tests/ci/merge_pr.py +++ b/tests/ci/merge_pr.py @@ -23,7 +23,7 @@ from commit_status_helper import ( from get_robot_token import get_best_robot_token from github_helper import GitHub, NamedUser, PullRequest, Repository from pr_info import PRInfo -from report import SUCCESS, FAILURE +from report import SUCCESS from env_helper import GITHUB_UPSTREAM_REPOSITORY, GITHUB_REPOSITORY from synchronizer_utils import SYNC_BRANCH_PREFIX from ci_config import CI @@ -248,23 +248,27 @@ def main(): repo = gh.get_repo(args.repo) if args.set_ci_status: + CI.GH.print_workflow_results() # set Mergeable check status and exit - assert args.wf_status in (FAILURE, SUCCESS) commit = get_commit(gh, args.pr_info.sha) statuses = get_commit_filtered_statuses(commit) 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}]") + if ( + CI.is_required(status.context) + and status.state != SUCCESS + and status.context != CI.StatusNames.SYNC + ): + print( + f"WARNING: Not success 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 or has_failed_statuses: - # set Mergeable check if workflow is successful (green) + workflow_ok = CI.is_workflow_ok() + if workflow_ok 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( @@ -283,7 +287,7 @@ def main(): 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: + if workflow_ok and not has_failed_statuses: sys.exit(0) else: sys.exit(1) diff --git a/tests/ci/report.py b/tests/ci/report.py index 0b6c818aed0..6779a6dae96 100644 --- a/tests/ci/report.py +++ b/tests/ci/report.py @@ -738,7 +738,7 @@ def create_test_html_report( if test_results: rows_part = [] num_fails = 0 - has_test_time = False + has_test_time = any(tr.time is not None for tr in test_results) has_log_urls = False # Display entires with logs at the top (they correspond to failed tests) @@ -770,9 +770,11 @@ def create_test_html_report( row.append(f'') colspan += 1 - if test_result.time is not None: - has_test_time = True - row.append(f"") + if has_test_time: + if test_result.time is not None: + row.append(f"") + else: + row.append("") colspan += 1 if test_result.log_urls is not None: diff --git a/tests/ci/stress.py b/tests/ci/stress.py index 8b8b17dd2a9..a3cdd8b3d6f 100755 --- a/tests/ci/stress.py +++ b/tests/ci/stress.py @@ -47,6 +47,8 @@ def get_options(i: int, upgrade_check: bool) -> str: if i > 0 and random.random() < 1 / 3: client_options.append("use_query_cache=1") + client_options.append("query_cache_nondeterministic_function_handling='ignore'") + client_options.append("query_cache_system_table_handling='ignore'") if i % 5 == 1: client_options.append("memory_tracker_fault_probability=0.001") diff --git a/tests/ci/test_ci_config.py b/tests/ci/test_ci_config.py index 6ffedfdecd4..525b3bf367b 100644 --- a/tests/ci/test_ci_config.py +++ b/tests/ci/test_ci_config.py @@ -642,7 +642,7 @@ class TestCIConfig(unittest.TestCase): release_branch=True, ) for record_t_, records_ in ci_cache.records.items(): - if record_t_.value == CiCache.RecordType.FAILED.value: + if record_t_.value == record.record_type.value: records_[record.to_str_key()] = record ci_cache.filter_out_not_affected_jobs() @@ -716,7 +716,7 @@ class TestCIConfig(unittest.TestCase): release_branch=True, ) for record_t_, records_ in ci_cache.records.items(): - if record_t_.value == CiCache.RecordType.FAILED.value: + if record_t_.value == record.record_type.value: records_[record.to_str_key()] = record ci_cache.filter_out_not_affected_jobs() @@ -726,3 +726,42 @@ class TestCIConfig(unittest.TestCase): MOCK_REQUIRED_BUILDS, ) self.assertCountEqual(list(ci_cache.jobs_to_do), expected_to_do) + + def test_ci_py_filters_not_affected_jobs_in_prs_docs_check(self): + """ + checks ci.py filters not affected jobs in PRs, + Docs Check is special from ci_cache perspective - + check it ci pr pipline is filtered properly when only docs check is to be skipped + """ + 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 + + job_config = ci_cache.jobs_to_do[CI.JobNames.DOCS_CHECK] + for batch in range(job_config.num_batches): + # add any record into cache + record = CiCache.Record( + record_type=CiCache.RecordType.PENDING, + job_name=CI.JobNames.DOCS_CHECK, + job_digest=ci_cache.job_digests[CI.JobNames.DOCS_CHECK], + batch=batch, + num_batches=job_config.num_batches, + release_branch=True, + ) + for record_t_, records_ in ci_cache.records.items(): + if record_t_.value == record.record_type.value: + records_[record.to_str_key()] = record + + expected_jobs = list(ci_cache.jobs_to_do) + expected_jobs.remove(CI.JobNames.DOCS_CHECK) + ci_cache.filter_out_not_affected_jobs() + self.assertCountEqual(list(ci_cache.jobs_to_do), expected_jobs) diff --git a/tests/ci/unit_tests_check.py b/tests/ci/unit_tests_check.py index b66a4312657..9cc8ec379bf 100644 --- a/tests/ci/unit_tests_check.py +++ b/tests/ci/unit_tests_check.py @@ -174,10 +174,13 @@ def main(): test_output = temp_path / "test_output" test_output.mkdir(parents=True, exist_ok=True) + # Don't run ASAN under gdb since that breaks leak detection + gdb_enabled = "NO_GDB" if "asan" in check_name else "GDB" + run_command = ( f"docker run --cap-add=SYS_PTRACE --volume={tests_binary}:/unit_tests_dbms " "--security-opt seccomp=unconfined " # required to issue io_uring sys-calls - f"--volume={test_output}:/test_output {docker_image}" + f"--volume={test_output}:/test_output {docker_image} {gdb_enabled}" ) run_log_path = test_output / "run.log" @@ -194,6 +197,11 @@ def main(): subprocess.check_call(f"sudo chown -R ubuntu:ubuntu {TEMP_PATH}", shell=True) state, description, test_results = process_results(test_output) + if retcode != 0 and state == SUCCESS: + # The process might have failed without reporting it in the test_output (e.g. LeakSanitizer) + state = FAILURE + description = "Invalid return code. Check run.log" + additional_files = [run_log_path] + [ p for p in test_output.iterdir() if not p.is_dir() ] diff --git a/tests/ci/version_helper.py b/tests/ci/version_helper.py index 07a7a9601c0..b20b2bb25cf 100755 --- a/tests/ci/version_helper.py +++ b/tests/ci/version_helper.py @@ -85,6 +85,16 @@ class ClickHouseVersion: self._tweak = 1 return self + def bump_patch(self) -> "ClickHouseVersion": + self._revision += 1 + self._patch += 1 + self._tweak = 1 + return self + + def reset_tweak(self) -> "ClickHouseVersion": + self._tweak = 1 + return self + def major_update(self) -> "ClickHouseVersion": if self._git is not None: self._git.update() @@ -104,13 +114,6 @@ class ClickHouseVersion: self.major, self.minor, self.patch + 1, self.revision, self._git ) - def reset_tweak(self) -> "ClickHouseVersion": - if self._git is not None: - self._git.update() - return ClickHouseVersion( - self.major, self.minor, self.patch, self.revision, self._git, 1 - ) - @property def major(self) -> int: return self._major diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 907d773337a..a3d7e0e922d 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -267,7 +267,7 @@ def clickhouse_execute_http( max_http_retries=5, retry_error_codes=False, ): - if args.secure: + if base_args.secure: client = http.client.HTTPSConnection( host=base_args.tcp_host, port=base_args.http_port, timeout=timeout ) @@ -358,14 +358,89 @@ def clickhouse_execute_json( return rows +# Should we capture client's stacktraces via SIGTSTP +CAPTURE_CLIENT_STACKTRACE = False + + +def kill_process_group(pgid): + print(f"Killing process group {pgid}") + print(f"Processes in process group {pgid}:") + print( + subprocess.check_output( + f"pgrep --pgroup {pgid} -a", shell=True, stderr=subprocess.STDOUT + ).decode("utf-8"), + end="", + ) + try: + if CAPTURE_CLIENT_STACKTRACE: + # Let's try to dump stacktrace in client (useful to catch issues there) + os.killpg(pgid, signal.SIGTSTP) + # Wait some time for clickhouse utilities to gather stacktrace + if RELEASE_NON_SANITIZED: + sleep(0.5) + else: + sleep(10) + # NOTE: this still may leave some processes, that had been + # created by timeout(1), since it also creates new process + # group. But this should not be a problem with default + # options, since the default time for each test is 10min, + # and this is way more bigger then the timeout for each + # timeout(1) invocation. + # + # But as a workaround we are sending SIGTERM first, and + # only after SIGKILL, that way timeout(1) will have an + # ability to terminate childrens (though not always since + # signals are asynchronous). + os.killpg(pgid, signal.SIGTERM) + # We need minimal delay to let processes handle SIGTERM - 0.1 (this may + # not be enough, but at least something) + sleep(0.1) + os.killpg(pgid, signal.SIGKILL) + except OSError as e: + if e.errno == ESRCH: + print(f"Got ESRCH while killing {pgid}. Ignoring.") + else: + raise + print(f"Process group {pgid} should be killed") + + +def cleanup_child_processes(pid): + pgid = os.getpgid(os.getpid()) + print(f"Child processes of {pid}:") + print( + subprocess.check_output( + f"pgrep --parent {pid} -a", shell=True, stderr=subprocess.STDOUT + ).decode("utf-8"), + end="", + ) + # Due to start_new_session=True, it is not enough to kill by PGID, we need + # to look at children processes as well. + # But we are hoping that nobody creates session in the tests (though it is + # possible via timeout(), but we are assuming that they will be killed by + # timeout). + processes = subprocess.check_output( + f"pgrep --parent {pid}", shell=True, stderr=subprocess.STDOUT + ) + processes = processes.decode("utf-8") + processes = processes.strip() + processes = processes.split("\n") + processes = map(lambda x: int(x.strip()), processes) + processes = list(processes) + for child in processes: + child_pgid = os.getpgid(child) + if child_pgid != pgid: + kill_process_group(child_pgid) + + # SIGKILL should not be sent, since this will kill the script itself + os.killpg(pgid, signal.SIGTERM) + + +# send signal to all processes in group to avoid hung check triggering +# (to avoid terminating clickhouse-test itself, the signal should be ignored) def stop_tests(): - # 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") + cleanup_child_processes(os.getpid()) + signal.signal(signal.SIGTERM, signal_handler) def get_db_engine(args, database_name): @@ -1248,39 +1323,35 @@ class TestCase: return None - def process_result_impl( - self, proc, stdout: str, stderr: str, debug_log: str, total_time: float - ): + def process_result_impl(self, proc, total_time: float): + if proc: + if proc.returncode is None: + kill_process_group(os.getpgid(proc.pid)) + description = "" + debug_log = "" + if os.path.exists(self.testcase_args.debug_log_file): + with open(self.testcase_args.debug_log_file, "rb") as stream: + debug_log += self.testcase_args.debug_log_file + ":\n" + debug_log += str(stream.read(), errors="replace", encoding="utf-8") + debug_log += "\n" + + stdout = "" + if os.path.exists(self.stdout_file): + with open(self.stdout_file, "rb") as stdfd: + stdout = str(stdfd.read(), errors="replace", encoding="utf-8") + + stderr = "" + if os.path.exists(self.stderr_file): + with open(self.stderr_file, "rb") as stdfd: + stderr += str(stdfd.read(), errors="replace", encoding="utf-8") + if debug_log: debug_log = "\n".join(debug_log.splitlines()[:100]) if proc: if proc.returncode is None: - try: - pgid = os.getpgid(proc.pid) - # NOTE: this still may leave some processes, that had been - # created by timeout(1), since it also creates new process - # group. But this should not be a problem with default - # options, since the default time for each test is 10min, - # and this is way more bigger then the timeout for each - # timeout(1) invocation. - # - # But as a workaround we are sending SIGTERM first, and - # only after SIGKILL, that way timeout(1) will have an - # ability to terminate childrens (though not always since - # signals are asynchronous). - os.killpg(pgid, signal.SIGTERM) - # This may not be enough, but this is at least something - # (and anyway it is OK to spend 0.1 second more in case of - # test timeout). - sleep(0.1) - os.killpg(pgid, signal.SIGKILL) - except OSError as e: - if e.errno != ESRCH: - raise - if stderr: description += stderr if debug_log: @@ -1532,7 +1603,7 @@ class TestCase: def run_single_test( self, server_logs_level, client_options - ) -> Tuple[Optional[Popen], str, str, str, float]: + ) -> Tuple[Optional[Popen], float]: args = self.testcase_args client = args.testcase_client start_time = args.testcase_start_time @@ -1609,13 +1680,6 @@ class TestCase: # Whether the test timed out will be decided later pass - debug_log = "" - if os.path.exists(self.testcase_args.debug_log_file): - with open(self.testcase_args.debug_log_file, "rb") as stream: - debug_log += self.testcase_args.debug_log_file + ":\n" - debug_log += str(stream.read(), errors="replace", encoding="utf-8") - debug_log += "\n" - total_time = (datetime.now() - start_time).total_seconds() # Normalize randomized database names in stdout, stderr files. @@ -1667,17 +1731,7 @@ class TestCase: "https://localhost:8443/", ) - stdout = "" - if os.path.exists(self.stdout_file): - with open(self.stdout_file, "rb") as stdfd: - stdout = str(stdfd.read(), errors="replace", encoding="utf-8") - - stderr = "" - if os.path.exists(self.stderr_file): - with open(self.stderr_file, "rb") as stdfd: - stderr += str(stdfd.read(), errors="replace", encoding="utf-8") - - return proc, stdout, stderr, debug_log, total_time + return proc, total_time def run(self, args, suite, client_options, server_logs_level): start_time = datetime.now() @@ -1709,14 +1763,14 @@ class TestCase: if not is_valid_utf_8(self.case_file) or ( self.reference_file and not is_valid_utf_8(self.reference_file) ): - proc, stdout, stderr, debug_log, total_time = self.run_single_test( + proc, total_time = self.run_single_test( server_logs_level, client_options ) - result = self.process_result_impl( - proc, stdout, stderr, debug_log, total_time + result = self.process_result_impl(proc, total_time) + result.check_if_need_retry( + args, result.description, result.description, self.runs_count ) - result.check_if_need_retry(args, stdout, stderr, self.runs_count) # to avoid breaking CSV parser result.description = result.description.replace("\0", "") else: @@ -1734,17 +1788,16 @@ class TestCase: ): ( proc, - stdout, - stderr, - debug_log, total_time, ) = self.run_single_test(server_logs_level, client_options) - result = self.process_result_impl( - proc, stdout, stderr, debug_log, total_time - ) + result = self.process_result_impl(proc, total_time) + result.check_if_need_retry( - args, stdout, stderr, self.runs_count + args, + result.description, + result.description, + self.runs_count, ) # to avoid breaking CSV parser result.description = result.description.replace("\0", "") @@ -2353,7 +2406,13 @@ class BuildFlags: POLYMORPHIC_PARTS = "polymorphic-parts" +# Release and non-sanitizer build +RELEASE_NON_SANITIZED = False + + def collect_build_flags(args): + global RELEASE_NON_SANITIZED + result = [] value = clickhouse_execute( @@ -2378,6 +2437,8 @@ def collect_build_flags(args): elif b"RelWithDebInfo" in value or b"Release" in value: result.append(BuildFlags.RELEASE) + RELEASE_NON_SANITIZED = result == [BuildFlags.RELEASE] + value = clickhouse_execute( args, "SELECT value FROM system.settings WHERE name = 'allow_deprecated_database_ordinary'", @@ -3390,29 +3451,36 @@ def parse_args(): default="./client.fatal.log", help="Path to file for fatal logs from client", ) + parser.add_argument( + "--capture-client-stacktrace", + action="store_true", + help="Capture stacktraces from clickhouse-client/local on errors", + ) return parser.parse_args() class Terminated(KeyboardInterrupt): - pass + def __init__(self, signal): + self.signal = signal -def signal_handler(sig, frame): - raise Terminated(f"Terminated with {sig} signal") +def signal_handler(signal, frame): + raise Terminated(signal) if __name__ == "__main__": + # Move to a new process group and kill it at exit so that we don't have any + # infinite tests processes left + # (new process group is required to avoid killing some parent processes) + os.setpgid(0, 0) + stop_time = None exit_code = multiprocessing.Value("i", 0) server_died = multiprocessing.Event() multiprocessing_manager = multiprocessing.Manager() restarted_tests = multiprocessing_manager.list() - # Move to a new process group and kill it at exit so that we don't have any - # infinite tests processes left - # (new process group is required to avoid killing some parent processes) - os.setpgid(0, 0) signal.signal(signal.SIGTERM, signal_handler) signal.signal(signal.SIGINT, signal_handler) signal.signal(signal.SIGHUP, signal_handler) @@ -3430,6 +3498,8 @@ if __name__ == "__main__": ) sys.exit(1) + CAPTURE_CLIENT_STACKTRACE = args.capture_client_stacktrace + # Autodetect the directory with queries if not specified if args.queries is None: args.queries = "queries" @@ -3551,4 +3621,14 @@ if __name__ == "__main__": if args.replace_replicated_with_shared: args.s3_storage = True - main(args) + try: + main(args) + except ServerDied as e: + print(f"{e}", file=sys.stderr) + sys.exit(1) + except Terminated as e: + print(f"Terminated with {e.signal} signal", file=sys.stderr) + sys.exit(128 + e.signal) + except KeyboardInterrupt: + print("Interrupted") + sys.exit(128 + signal.SIGINT) diff --git a/tests/config/config.d/rocksdb.xml b/tests/config/config.d/rocksdb.xml index a3790a3dc1d..3002e008a2d 100644 --- a/tests/config/config.d/rocksdb.xml +++ b/tests/config/config.d/rocksdb.xml @@ -1,7 +1,8 @@ - DEBUG_LEVEL + + ERROR_LEVEL diff --git a/tests/config/config.d/transactions.xml b/tests/config/config.d/transactions.xml index 9948b1f1865..64e166b81b5 100644 --- a/tests/config/config.d/transactions.xml +++ b/tests/config/config.d/transactions.xml @@ -1,4 +1,4 @@ - + 42 @@ -18,4 +18,4 @@ 0.01 - + diff --git a/tests/fuzz/README.md b/tests/fuzz/README.md deleted file mode 100644 index 6b5b161b2d5..00000000000 --- a/tests/fuzz/README.md +++ /dev/null @@ -1,23 +0,0 @@ -The list of functions generated via the following query - -``` - clickhouse client -q "SELECT * FROM (SELECT DISTINCT concat('\"', name, '\"') as res FROM system.functions ORDER BY name UNION ALL SELECT concat('\"', a.name, b.name, '\"') as res FROM system.functions as a CROSS JOIN system.aggregate_function_combinators as b WHERE a.is_aggregate = 1) ORDER BY res" > functions.dict -``` - -The list of datatypes generated via the following query: - -``` - clickhouse client -q "SELECT DISTINCT concat('\"', name, '\"') as res FROM system.data_type_families ORDER BY name" > datatypes.dict -``` - -The list of keywords generated via the following query: - -``` - clickhouse client -q "SELECT DISTINCT concat('\"', keyword, '\"') as res FROM system.keywords ORDER BY keyword" > key_words.dict -``` - -Then merge all dictionaries into one (all.dict) - -``` - cat ./dictionaries/* | sort | uniq > all.dict -``` \ No newline at end of file diff --git a/tests/fuzz/all.dict b/tests/fuzz/all.dict index f08e319f0d4..30af3746fca 100644 --- a/tests/fuzz/all.dict +++ b/tests/fuzz/all.dict @@ -1,38 +1,991 @@ +"ADD COLUMN" +"ADD CONSTRAINT" +"ADD INDEX" +"ADD PROJECTION" +"ADD STATISTICS" +"ADD" +"ADMIN OPTION FOR" +"AFTER" +"ALGORITHM" +"ALIAS" +"ALL" +"ALLOWED_LATENESS" +"ALTER COLUMN" +"ALTER DATABASE" +"ALTER LIVE VIEW" +"ALTER POLICY" +"ALTER PROFILE" +"ALTER QUOTA" +"ALTER ROLE" +"ALTER ROW POLICY" +"ALTER SETTINGS PROFILE" +"ALTER TABLE" +"ALTER TEMPORARY TABLE" +"ALTER USER" +"ALTER" +"AND STDOUT" +"AND" +"ANTI" +"ANY" +"APPEND" +"APPLY DELETED MASK" +"APPLY" +"ARRAY JOIN" +"AS" +"ASC" +"ASCENDING" +"ASOF" +"ASSUME" +"AST" +"ASYNC" +"ATTACH PART" +"ATTACH PARTITION" +"ATTACH POLICY" +"ATTACH PROFILE" +"ATTACH QUOTA" +"ATTACH ROLE" +"ATTACH ROW POLICY" +"ATTACH SETTINGS PROFILE" +"ATTACH USER" +"ATTACH" +"AUTO_INCREMENT" +"AZURE" +"AggregateFunction" +"Array" +"BACKUP" +"BCRYPT_HASH" +"BCRYPT_PASSWORD" +"BEGIN TRANSACTION" +"BETWEEN" +"BIDIRECTIONAL" +"BIGINT SIGNED" +"BIGINT UNSIGNED" +"BIGINT" +"BINARY LARGE OBJECT" +"BINARY VARYING" +"BINARY" +"BIT" +"BIT_AND" +"BIT_ANDArgMax" +"BIT_ANDArgMin" +"BIT_ANDArray" +"BIT_ANDDistinct" +"BIT_ANDForEach" +"BIT_ANDIf" +"BIT_ANDMap" +"BIT_ANDMerge" +"BIT_ANDNull" +"BIT_ANDOrDefault" +"BIT_ANDOrNull" +"BIT_ANDResample" +"BIT_ANDSimpleState" +"BIT_ANDState" +"BIT_OR" +"BIT_ORArgMax" +"BIT_ORArgMin" +"BIT_ORArray" +"BIT_ORDistinct" +"BIT_ORForEach" +"BIT_ORIf" +"BIT_ORMap" +"BIT_ORMerge" +"BIT_ORNull" +"BIT_OROrDefault" +"BIT_OROrNull" +"BIT_ORResample" +"BIT_ORSimpleState" +"BIT_ORState" +"BIT_XOR" +"BIT_XORArgMax" +"BIT_XORArgMin" +"BIT_XORArray" +"BIT_XORDistinct" +"BIT_XORForEach" +"BIT_XORIf" +"BIT_XORMap" +"BIT_XORMerge" +"BIT_XORNull" +"BIT_XOROrDefault" +"BIT_XOROrNull" +"BIT_XORResample" +"BIT_XORSimpleState" +"BIT_XORState" +"BLAKE3" +"BLOB" +"BOTH" +"BY" +"BYTE" +"BYTEA" +"Bool" +"CASCADE" +"CASE" +"CAST" +"CHANGE" +"CHANGEABLE_IN_READONLY" +"CHANGED" +"CHAR LARGE OBJECT" +"CHAR VARYING" +"CHAR" +"CHARACTER LARGE OBJECT" +"CHARACTER VARYING" +"CHARACTER" +"CHARACTER_LENGTH" +"CHAR_LENGTH" +"CHECK ALL TABLES" +"CHECK TABLE" +"CHECK" +"CLEANUP" +"CLEAR COLUMN" +"CLEAR INDEX" +"CLEAR PROJECTION" +"CLEAR STATISTICS" +"CLOB" +"CLUSTER" +"CLUSTERS" +"CN" +"CODEC" +"COLLATE" +"COLUMN" +"COLUMNS" +"COMMENT COLUMN" +"COMMENT" +"COMMIT" +"COMPRESSION" +"CONST" +"CONSTRAINT" +"COVAR_POP" +"COVAR_POPArgMax" +"COVAR_POPArgMin" +"COVAR_POPArray" +"COVAR_POPDistinct" +"COVAR_POPForEach" +"COVAR_POPIf" +"COVAR_POPMap" +"COVAR_POPMerge" +"COVAR_POPNull" +"COVAR_POPOrDefault" +"COVAR_POPOrNull" +"COVAR_POPResample" +"COVAR_POPSimpleState" +"COVAR_POPState" +"COVAR_SAMP" +"COVAR_SAMPArgMax" +"COVAR_SAMPArgMin" +"COVAR_SAMPArray" +"COVAR_SAMPDistinct" +"COVAR_SAMPForEach" +"COVAR_SAMPIf" +"COVAR_SAMPMap" +"COVAR_SAMPMerge" +"COVAR_SAMPNull" +"COVAR_SAMPOrDefault" +"COVAR_SAMPOrNull" +"COVAR_SAMPResample" +"COVAR_SAMPSimpleState" +"COVAR_SAMPState" +"CRC32" +"CRC32IEEE" +"CRC64" +"CREATE POLICY" +"CREATE PROFILE" +"CREATE QUOTA" +"CREATE ROLE" +"CREATE ROW POLICY" +"CREATE SETTINGS PROFILE" +"CREATE TABLE" +"CREATE TEMPORARY TABLE" +"CREATE USER" +"CREATE" +"CROSS" +"CUBE" +"CURRENT GRANTS" +"CURRENT QUOTA" +"CURRENT ROLES" +"CURRENT ROW" +"CURRENT TRANSACTION" +"CURRENTUSER" +"CURRENT_USER" +"D" +"DATA INNER UUID" +"DATA" +"DATABASE" +"DATABASES" +"DATE" +"DATEADD" +"DATEDIFF" +"DATESUB" +"DATE_ADD" +"DATE_DIFF" +"DATE_FORMAT" +"DATE_SUB" +"DATE_TRUNC" +"DAY" +"DAYOFMONTH" +"DAYOFWEEK" +"DAYOFYEAR" +"DAYS" +"DD" +"DEC" +"DEDUPLICATE" +"DEFAULT DATABASE" +"DEFAULT ROLE" +"DEFAULT" +"DEFINER" +"DELETE WHERE" +"DELETE" +"DEPENDS ON" +"DESC" +"DESCENDING" +"DESCRIBE" +"DETACH PART" +"DETACH PARTITION" +"DETACH" +"DICTIONARIES" +"DICTIONARY" +"DISK" +"DISTINCT ON" +"DISTINCT" +"DIV" +"DOUBLE PRECISION" +"DOUBLE" +"DOUBLE_SHA1_HASH" +"DOUBLE_SHA1_PASSWORD" +"DROP COLUMN" +"DROP CONSTRAINT" +"DROP DEFAULT" +"DROP DETACHED PART" +"DROP DETACHED PARTITION" +"DROP INDEX" +"DROP PART" +"DROP PARTITION" +"DROP PROJECTION" +"DROP STATISTICS" +"DROP TABLE" +"DROP TEMPORARY TABLE" +"DROP" +"Date" +"Date32" +"DateTime" +"DateTime32" +"DateTime64" +"Decimal" +"Decimal128" +"Decimal256" +"Decimal32" +"Decimal64" +"Dynamic" +"ELSE" +"EMPTY AS" +"EMPTY" +"ENABLED ROLES" +"END" +"ENFORCED" +"ENGINE" +"ENUM" +"EPHEMERAL SEQUENTIAL" +"EPHEMERAL" +"ESTIMATE" +"EVENT" +"EVENTS" +"EVERY" +"EXCEPT DATABASE" +"EXCEPT DATABASES" +"EXCEPT TABLE" +"EXCEPT TABLES" +"EXCEPT" +"EXCHANGE DICTIONARIES" +"EXCHANGE TABLES" +"EXISTS" +"EXPLAIN" +"EXPRESSION" +"EXTENDED" +"EXTERNAL DDL FROM" +"EXTRACT" +"Enum" +"Enum16" +"Enum8" +"FALSE" +"FETCH PART" +"FETCH PARTITION" +"FETCH" +"FIELDS" +"FILE" +"FILESYSTEM CACHE" +"FILESYSTEM CACHES" +"FILTER" +"FINAL" +"FIRST" +"FIXED" +"FLOAT" +"FOLLOWING" +"FOR" +"FOREIGN KEY" +"FOREIGN" +"FORGET PARTITION" +"FORMAT" +"FORMAT_BYTES" +"FQDN" +"FREEZE" +"FROM INFILE" +"FROM SHARD" +"FROM" +"FROM_BASE64" +"FROM_DAYS" +"FROM_UNIXTIME" +"FULL" +"FULLTEXT" +"FUNCTION" +"FixedString" +"Float32" +"Float64" +"ForEach" +"GEOMETRY" +"GLOBAL IN" +"GLOBAL NOT IN" +"GLOBAL" +"GRANT OPTION FOR" +"GRANT" +"GRANTEES" +"GRANULARITY" +"GROUP BY" +"GROUPING SETS" +"GROUPS" +"H" +"HASH" +"HAVING" +"HDFS" +"HH" +"HIERARCHICAL" +"HOST" +"HOUR" +"HOURS" +"HTTP" +"ID" +"IDENTIFIED" +"IF EMPTY" +"IF EXISTS" +"IF NOT EXISTS" +"IGNORE NULLS" +"ILIKE" +"IN PARTITION" +"IN" +"INDEX" +"INDEXES" +"INDICES" +"INET4" +"INET6" +"INET6_ATON" +"INET6_NTOA" +"INET_ATON" +"INET_NTOA" +"INHERIT" +"INJECTIVE" +"INNER" +"INSERT INTO" +"INT SIGNED" +"INT UNSIGNED" +"INT" +"INT1 SIGNED" +"INT1 UNSIGNED" +"INT1" +"INTEGER SIGNED" +"INTEGER UNSIGNED" +"INTEGER" +"INTERPOLATE" +"INTERSECT" +"INTERVAL" +"INTO OUTFILE" +"INVISIBLE" +"INVOKER" +"IP" +"IPv4" +"IPv4CIDRToRange" +"IPv4NumToString" +"IPv4NumToStringClassC" +"IPv4StringToNum" +"IPv4StringToNumOrDefault" +"IPv4StringToNumOrNull" +"IPv4ToIPv6" +"IPv6" +"IPv6CIDRToRange" +"IPv6NumToString" +"IPv6StringToNum" +"IPv6StringToNumOrDefault" +"IPv6StringToNumOrNull" +"IS NOT DISTINCT FROM" +"IS NOT NULL" +"IS NULL" +"IS_OBJECT_ID" +"Int128" +"Int16" +"Int256" +"Int32" +"Int64" +"Int8" +"IntervalDay" +"IntervalHour" +"IntervalMicrosecond" +"IntervalMillisecond" +"IntervalMinute" +"IntervalMonth" +"IntervalNanosecond" +"IntervalQuarter" +"IntervalSecond" +"IntervalWeek" +"IntervalYear" +"JOIN" +"JSON" +"JSONArrayLength" +"JSONExtract" +"JSONExtractArrayRaw" +"JSONExtractBool" +"JSONExtractFloat" +"JSONExtractInt" +"JSONExtractKeys" +"JSONExtractKeysAndValues" +"JSONExtractKeysAndValuesRaw" +"JSONExtractRaw" +"JSONExtractString" +"JSONExtractUInt" +"JSONHas" +"JSONKey" +"JSONLength" +"JSONMergePatch" +"JSONType" +"JSON_ARRAY_LENGTH" +"JSON_EXISTS" +"JSON_QUERY" +"JSON_VALUE" +"JWT" +"KERBEROS" +"KEY BY" +"KEY" +"KEYED BY" +"KEYS" +"KILL" +"KIND" +"L1Distance" +"L1Norm" +"L1Normalize" +"L2Distance" +"L2Norm" +"L2Normalize" +"L2SquaredDistance" +"L2SquaredNorm" +"LARGE OBJECT" +"LAST" +"LAST_DAY" +"LAYOUT" +"LDAP" +"LEADING" +"LEFT ARRAY JOIN" +"LEFT" +"LESS THAN" +"LEVEL" +"LIFETIME" +"LIGHTWEIGHT" +"LIKE" +"LIMIT" +"LINEAR" +"LIST" +"LIVE" +"LOCAL" +"LONGBLOB" +"LONGTEXT" +"LTRIM" +"LineString" +"LinfDistance" +"LinfNorm" +"LinfNormalize" +"LowCardinality" +"LpDistance" +"LpNorm" +"LpNormalize" +"M" +"MACNumToString" +"MACStringToNum" +"MACStringToOUI" +"MAP_FROM_ARRAYS" +"MATCH" +"MATERIALIZE COLUMN" +"MATERIALIZE INDEX" +"MATERIALIZE PROJECTION" +"MATERIALIZE STATISTICS" +"MATERIALIZE TTL" +"MATERIALIZE" +"MATERIALIZED" +"MAX" +"MCS" +"MD4" +"MD5" +"MEDIUMBLOB" +"MEDIUMINT SIGNED" +"MEDIUMINT UNSIGNED" +"MEDIUMINT" +"MEDIUMTEXT" +"MEMORY" +"MERGES" +"METRICS INNER UUID" +"METRICS" +"MI" +"MICROSECOND" +"MICROSECONDS" +"MILLISECOND" +"MILLISECONDS" +"MIN" +"MINUTE" +"MINUTES" +"MM" +"MOD" +"MODIFY COLUMN" +"MODIFY COMMENT" +"MODIFY DEFINER" +"MODIFY ORDER BY" +"MODIFY QUERY" +"MODIFY REFRESH" +"MODIFY SAMPLE BY" +"MODIFY SETTING" +"MODIFY SQL SECURITY" +"MODIFY STATISTICS" +"MODIFY TTL" +"MODIFY" +"MONTH" +"MONTHS" +"MOVE PART" +"MOVE PARTITION" +"MOVE" +"MS" +"MUTATION" +"Map" +"Merge" +"MultiLineString" +"MultiPolygon" +"N" +"NAME" +"NAMED COLLECTION" +"NANOSECOND" +"NANOSECONDS" +"NATIONAL CHAR VARYING" +"NATIONAL CHAR" +"NATIONAL CHARACTER LARGE OBJECT" +"NATIONAL CHARACTER VARYING" +"NATIONAL CHARACTER" +"NCHAR LARGE OBJECT" +"NCHAR VARYING" +"NCHAR" +"NEXT" +"NO ACTION" +"NO DELAY" +"NO LIMITS" +"NONE" +"NOT BETWEEN" +"NOT IDENTIFIED" +"NOT ILIKE" +"NOT IN" +"NOT KEYED" +"NOT LIKE" +"NOT OVERRIDABLE" +"NOT" +"NO_PASSWORD" +"NS" +"NULL" +"NULLS" +"NUMERIC" +"NVARCHAR" +"Nested" +"Nothing" +"Null" +"Nullable" +"OCTET_LENGTH" +"OFFSET" +"ON DELETE" +"ON UPDATE" +"ON VOLUME" +"ON" +"ONLY" +"OPTIMIZE TABLE" +"OR REPLACE" +"OR" +"ORDER BY" +"OUTER" +"OVER" +"OVERRIDABLE" +"Object" +"PART" +"PARTIAL" +"PARTITION BY" +"PARTITION" +"PARTITIONS" +"PART_MOVE_TO_SHARD" +"PASTE" +"PERIODIC REFRESH" +"PERMANENTLY" +"PERMISSIVE" +"PERSISTENT SEQUENTIAL" +"PERSISTENT" +"PIPELINE" +"PLAINTEXT_PASSWORD" +"PLAN" +"POPULATE" +"PRECEDING" +"PRECISION" +"PREWHERE" +"PRIMARY KEY" +"PRIMARY" +"PROFILE" +"PROJECTION" +"PULL" +"Point" +"Polygon" +"Protobuf" +"Q" +"QQ" +"QUALIFY" +"QUARTER" +"QUARTERS" +"QUERY TREE" +"QUERY" +"QUOTA" +"RANDOMIZE FOR" +"RANDOMIZED" +"RANGE" +"READONLY" +"REAL" +"REALM" +"RECOMPRESS" +"RECURSIVE" +"REFERENCES" +"REFRESH" +"REGEXP" +"REGEXP_EXTRACT" +"REGEXP_MATCHES" +"REGEXP_REPLACE" +"REMOVE SAMPLE BY" +"REMOVE TTL" +"REMOVE" +"RENAME COLUMN" +"RENAME DATABASE" +"RENAME DICTIONARY" +"RENAME TABLE" +"RENAME TO" +"RENAME" +"REPLACE PARTITION" +"REPLACE" +"RESET SETTING" +"RESPECT NULLS" +"RESTORE" +"RESTRICT" +"RESTRICTIVE" +"RESUME" +"REVOKE" +"RIGHT" +"ROLLBACK" +"ROLLUP" +"ROW" +"ROWS" +"RTRIM" +"Resample" +"Ring" +"S" +"S3" +"SALT" +"SAMPLE BY" +"SAMPLE" +"SAN" +"SCHEMA" +"SCHEME" +"SECOND" +"SECONDS" +"SELECT" +"SEMI" +"SERVER" +"SET DEFAULT ROLE" +"SET DEFAULT" +"SET FAKE TIME" +"SET NULL" +"SET ROLE DEFAULT" +"SET ROLE" +"SET TRANSACTION SNAPSHOT" +"SET" +"SETTINGS" +"SHA1" +"SHA224" +"SHA256" +"SHA256_HASH" +"SHA256_PASSWORD" +"SHA384" +"SHA512" +"SHA512_256" +"SHOW ACCESS" +"SHOW CREATE" +"SHOW ENGINES" +"SHOW FUNCTIONS" +"SHOW GRANTS" +"SHOW PRIVILEGES" +"SHOW PROCESSLIST" +"SHOW SETTING" +"SHOW" +"SIGNED" +"SIMPLE" +"SINGLE" +"SMALLINT SIGNED" +"SMALLINT UNSIGNED" +"SMALLINT" +"SOURCE" +"SPATIAL" +"SQL SECURITY" +"SQL_TSI_DAY" +"SQL_TSI_HOUR" +"SQL_TSI_MICROSECOND" +"SQL_TSI_MILLISECOND" +"SQL_TSI_MINUTE" +"SQL_TSI_MONTH" +"SQL_TSI_NANOSECOND" +"SQL_TSI_QUARTER" +"SQL_TSI_SECOND" +"SQL_TSI_WEEK" +"SQL_TSI_YEAR" +"SS" +"SSH_KEY" +"SSL_CERTIFICATE" +"START TRANSACTION" +"STATISTICS" +"STD" +"STDArgMax" +"STDArgMin" +"STDArray" +"STDDEV_POP" +"STDDEV_POPArgMax" +"STDDEV_POPArgMin" +"STDDEV_POPArray" +"STDDEV_POPDistinct" +"STDDEV_POPForEach" +"STDDEV_POPIf" +"STDDEV_POPMap" +"STDDEV_POPMerge" +"STDDEV_POPNull" +"STDDEV_POPOrDefault" +"STDDEV_POPOrNull" +"STDDEV_POPResample" +"STDDEV_POPSimpleState" +"STDDEV_POPState" +"STDDEV_SAMP" +"STDDEV_SAMPArgMax" +"STDDEV_SAMPArgMin" +"STDDEV_SAMPArray" +"STDDEV_SAMPDistinct" +"STDDEV_SAMPForEach" +"STDDEV_SAMPIf" +"STDDEV_SAMPMap" +"STDDEV_SAMPMerge" +"STDDEV_SAMPNull" +"STDDEV_SAMPOrDefault" +"STDDEV_SAMPOrNull" +"STDDEV_SAMPResample" +"STDDEV_SAMPSimpleState" +"STDDEV_SAMPState" +"STDDistinct" +"STDForEach" +"STDIf" +"STDMap" +"STDMerge" +"STDNull" +"STDOrDefault" +"STDOrNull" +"STDResample" +"STDSimpleState" +"STDState" +"STEP" +"STORAGE" +"STRICT" +"STRICTLY_ASCENDING" +"SUBPARTITION BY" +"SUBPARTITION" +"SUBPARTITIONS" +"SUBSTRING" +"SUBSTRING_INDEX" +"SUSPEND" +"SVG" +"SYNC" +"SYNTAX" +"SYSTEM" +"SimpleAggregateFunction" +"State" +"String" +"TABLE OVERRIDE" +"TABLE" +"TABLES" +"TAGS INNER UUID" +"TAGS" +"TEMPORARY TABLE" +"TEMPORARY" +"TEST" +"TEXT" +"THEN" +"TIME" +"TIMESTAMP" +"TIMESTAMPADD" +"TIMESTAMPDIFF" +"TIMESTAMPSUB" +"TIMESTAMP_ADD" +"TIMESTAMP_DIFF" +"TIMESTAMP_SUB" +"TINYBLOB" +"TINYINT SIGNED" +"TINYINT UNSIGNED" +"TINYINT" +"TINYTEXT" +"TO DISK" +"TO INNER UUID" +"TO SHARD" +"TO TABLE" +"TO VOLUME" +"TO" +"TOP" +"TOTALS" +"TO_BASE64" +"TO_DAYS" +"TO_UNIXTIME" +"TRACKING ONLY" +"TRAILING" +"TRANSACTION" +"TRIGGER" +"TRIM" +"TRUE" +"TRUNCATE" +"TTL" +"TYPE" +"TYPEOF" +"Tuple" +"UInt128" +"UInt16" +"UInt256" +"UInt32" +"UInt64" +"UInt8" +"ULIDStringToDateTime" +"UNBOUNDED" +"UNDROP" +"UNFREEZE" +"UNION" +"UNIQUE" +"UNSET FAKE TIME" +"UNSIGNED" +"UPDATE" +"URL" +"URLHash" +"URLHierarchy" +"URLPathHierarchy" +"USE" +"USING" +"UTCTimestamp" +"UTC_timestamp" +"UUID" +"UUIDNumToString" +"UUIDStringToNum" +"UUIDToNum" +"UUIDv7ToDateTime" +"VALID UNTIL" +"VALUES" +"VARBINARY" +"VARCHAR" +"VARCHAR2" +"VARYING" +"VAR_POP" +"VAR_POPArgMax" +"VAR_POPArgMin" +"VAR_POPArray" +"VAR_POPDistinct" +"VAR_POPForEach" +"VAR_POPIf" +"VAR_POPMap" +"VAR_POPMerge" +"VAR_POPNull" +"VAR_POPOrDefault" +"VAR_POPOrNull" +"VAR_POPResample" +"VAR_POPSimpleState" +"VAR_POPState" +"VAR_SAMP" +"VAR_SAMPArgMax" +"VAR_SAMPArgMin" +"VAR_SAMPArray" +"VAR_SAMPDistinct" +"VAR_SAMPForEach" +"VAR_SAMPIf" +"VAR_SAMPMap" +"VAR_SAMPMerge" +"VAR_SAMPNull" +"VAR_SAMPOrDefault" +"VAR_SAMPOrNull" +"VAR_SAMPResample" +"VAR_SAMPSimpleState" +"VAR_SAMPState" +"VIEW" +"VISIBLE" +"Variant" +"WATCH" +"WATERMARK" +"WEEK" +"WEEKS" +"WHEN" +"WHERE" +"WINDOW" +"WITH ADMIN OPTION" +"WITH CHECK" +"WITH FILL" +"WITH GRANT OPTION" +"WITH NAME" +"WITH REPLACE OPTION" +"WITH TIES" +"WITH" +"WITH_ITEMINDEX" +"WK" +"WRITABLE" +"WW" +"YEAR" +"YEARS" +"YY" +"YYYY" +"YYYYMMDDToDate" +"YYYYMMDDToDate32" +"YYYYMMDDhhmmssToDateTime" +"YYYYMMDDhhmmssToDateTime64" +"ZKPATH" +"_CAST" +"__actionName" +"__bitBoolMaskAnd" +"__bitBoolMaskOr" +"__bitSwapLastTwo" +"__bitWrapperFunc" +"__getScalar" +"__scalarSubqueryResult" "abs" "accurateCast" "accurateCastOrDefault" "accurateCastOrNull" "acos" "acosh" -"ADD" -"ADD COLUMN" -"ADD CONSTRAINT" "addDate" "addDays" "addHours" -"ADD INDEX" "addInterval" "addMicroseconds" "addMilliseconds" "addMinutes" "addMonths" "addNanoseconds" -"ADD PROJECTION" "addQuarters" -"addressToLine" -"addressToLineWithInlines" -"addressToSymbol" "addSeconds" -"ADD STATISTIC" "addTupleOfIntervals" "addWeeks" "addYears" -"ADMIN OPTION FOR" +"addressToLine" +"addressToLineWithInlines" +"addressToSymbol" "aes_decrypt_mysql" "aes_encrypt_mysql" -"AFTER" "age" -"AggregateFunction" "aggThrow" "aggThrowArgMax" "aggThrowArgMin" @@ -48,24 +1001,7 @@ "aggThrowResample" "aggThrowSimpleState" "aggThrowState" -"ALGORITHM" -"ALIAS" -"ALL" -"ALLOWED_LATENESS" "alphaTokens" -"ALTER" -"ALTER COLUMN" -"ALTER DATABASE" -"ALTER LIVE VIEW" -"ALTER POLICY" -"ALTER PROFILE" -"ALTER QUOTA" -"ALTER ROLE" -"ALTER ROW POLICY" -"ALTER SETTINGS PROFILE" -"ALTER TABLE" -"ALTER TEMPORARY TABLE" -"ALTER USER" "analysisOfVariance" "analysisOfVarianceArgMax" "analysisOfVarianceArgMin" @@ -82,8 +1018,6 @@ "analysisOfVarianceSimpleState" "analysisOfVarianceState" "and" -"AND" -"AND STDOUT" "anova" "anovaArgMax" "anovaArgMin" @@ -99,9 +1033,7 @@ "anovaResample" "anovaSimpleState" "anovaState" -"ANTI" "any" -"ANY" "anyArgMax" "anyArgMin" "anyArray" @@ -136,6 +1068,8 @@ "anyLastOrDefault" "anyLastOrNull" "anyLastResample" +"anyLastSimpleState" +"anyLastState" "anyLast_respect_nulls" "anyLast_respect_nullsArgMax" "anyLast_respect_nullsArgMin" @@ -151,14 +1085,14 @@ "anyLast_respect_nullsResample" "anyLast_respect_nullsSimpleState" "anyLast_respect_nullsState" -"anyLastSimpleState" -"anyLastState" "anyMap" "anyMerge" "anyNull" "anyOrDefault" "anyOrNull" "anyResample" +"anySimpleState" +"anyState" "any_respect_nulls" "any_respect_nullsArgMax" "any_respect_nullsArgMin" @@ -174,8 +1108,6 @@ "any_respect_nullsResample" "any_respect_nullsSimpleState" "any_respect_nullsState" -"anySimpleState" -"anyState" "any_value" "any_valueArgMax" "any_valueArgMin" @@ -189,6 +1121,8 @@ "any_valueOrDefault" "any_valueOrNull" "any_valueResample" +"any_valueSimpleState" +"any_valueState" "any_value_respect_nulls" "any_value_respect_nullsArgMax" "any_value_respect_nullsArgMin" @@ -204,12 +1138,7 @@ "any_value_respect_nullsResample" "any_value_respect_nullsSimpleState" "any_value_respect_nullsState" -"any_valueSimpleState" -"any_valueState" -"APPEND" "appendTrailingCharIfAbsent" -"APPLY" -"APPLY DELETED MASK" "approx_top_count" "approx_top_countArgMax" "approx_top_countArgMin" @@ -286,42 +1215,11 @@ "argMinSimpleState" "argMinState" "array" -"Array" -"array_agg" -"array_aggArgMax" -"array_aggArgMin" -"array_aggArray" -"array_aggDistinct" -"array_aggForEach" -"array_aggIf" -"array_aggMap" -"array_aggMerge" -"array_aggNull" -"array_aggOrDefault" -"array_aggOrNull" -"array_aggResample" -"array_aggSimpleState" -"array_aggState" -"arrayAll" "arrayAUC" +"arrayAll" "arrayAvg" "arrayCompact" "arrayConcat" -"array_concat_agg" -"array_concat_aggArgMax" -"array_concat_aggArgMin" -"array_concat_aggArray" -"array_concat_aggDistinct" -"array_concat_aggForEach" -"array_concat_aggIf" -"array_concat_aggMap" -"array_concat_aggMerge" -"array_concat_aggNull" -"array_concat_aggOrDefault" -"array_concat_aggOrNull" -"array_concat_aggResample" -"array_concat_aggSimpleState" -"array_concat_aggState" "arrayCount" "arrayCumSum" "arrayCumSumNonNegative" @@ -345,7 +1243,6 @@ "arrayIntersect" "arrayJaccardIndex" "arrayJoin" -"ARRAY JOIN" "arrayLast" "arrayLastIndex" "arrayLastOrNull" @@ -382,31 +1279,43 @@ "arrayUniq" "arrayWithConstant" "arrayZip" -"AS" -"ASC" -"ASCENDING" +"array_agg" +"array_aggArgMax" +"array_aggArgMin" +"array_aggArray" +"array_aggDistinct" +"array_aggForEach" +"array_aggIf" +"array_aggMap" +"array_aggMerge" +"array_aggNull" +"array_aggOrDefault" +"array_aggOrNull" +"array_aggResample" +"array_aggSimpleState" +"array_aggState" +"array_concat_agg" +"array_concat_aggArgMax" +"array_concat_aggArgMin" +"array_concat_aggArray" +"array_concat_aggDistinct" +"array_concat_aggForEach" +"array_concat_aggIf" +"array_concat_aggMap" +"array_concat_aggMerge" +"array_concat_aggNull" +"array_concat_aggOrDefault" +"array_concat_aggOrNull" +"array_concat_aggResample" +"array_concat_aggSimpleState" +"array_concat_aggState" "ascii" "asin" "asinh" -"ASOF" -"ASSUME" "assumeNotNull" -"AST" -"ASYNC" "atan" "atan2" "atanh" -"ATTACH" -"ATTACH PART" -"ATTACH PARTITION" -"ATTACH POLICY" -"ATTACH PROFILE" -"ATTACH QUOTA" -"ATTACH ROLE" -"ATTACH ROW POLICY" -"ATTACH SETTINGS PROFILE" -"ATTACH USER" -"AUTO_INCREMENT" "avg" "avgArgMax" "avgArgMin" @@ -437,49 +1346,32 @@ "avgWeightedResample" "avgWeightedSimpleState" "avgWeightedState" -"AZURE" -"BACKUP" "bagexpansion" "bar" "base58Decode" "base58Encode" "base64Decode" "base64Encode" +"base64URLDecode" +"base64URLEncode" "base_backup" "basename" -"BCRYPT_HASH" -"BCRYPT_PASSWORD" -"BEGIN TRANSACTION" -"BETWEEN" -"BIDIRECTIONAL" -"BIGINT" -"BIGINT SIGNED" -"BIGINT UNSIGNED" "bin" -"BINARY" -"BINARY LARGE OBJECT" -"BINARY VARYING" -"BIT" "bitAnd" -"BIT_AND" -"BIT_ANDArgMax" -"BIT_ANDArgMin" -"BIT_ANDArray" -"BIT_ANDDistinct" -"BIT_ANDForEach" -"BIT_ANDIf" -"BIT_ANDMap" -"BIT_ANDMerge" -"BIT_ANDNull" -"BIT_ANDOrDefault" -"BIT_ANDOrNull" -"BIT_ANDResample" -"BIT_ANDSimpleState" -"BIT_ANDState" -"__bitBoolMaskAnd" -"__bitBoolMaskOr" "bitCount" "bitHammingDistance" +"bitNot" +"bitOr" +"bitPositionsToArray" +"bitRotateLeft" +"bitRotateRight" +"bitShiftLeft" +"bitShiftRight" +"bitSlice" +"bitTest" +"bitTestAll" +"bitTestAny" +"bitXor" "bitmapAnd" "bitmapAndCardinality" "bitmapAndnot" @@ -501,59 +1393,11 @@ "bitmapXorCardinality" "bitmaskToArray" "bitmaskToList" -"bitNot" -"bitOr" -"BIT_OR" -"BIT_ORArgMax" -"BIT_ORArgMin" -"BIT_ORArray" -"BIT_ORDistinct" -"BIT_ORForEach" -"BIT_ORIf" -"BIT_ORMap" -"BIT_ORMerge" -"BIT_ORNull" -"BIT_OROrDefault" -"BIT_OROrNull" -"BIT_ORResample" -"BIT_ORSimpleState" -"BIT_ORState" -"bitPositionsToArray" -"bitRotateLeft" -"bitRotateRight" -"bitShiftLeft" -"bitShiftRight" -"bitSlice" -"__bitSwapLastTwo" -"bitTest" -"bitTestAll" -"bitTestAny" -"__bitWrapperFunc" -"bitXor" -"BIT_XOR" -"BIT_XORArgMax" -"BIT_XORArgMin" -"BIT_XORArray" -"BIT_XORDistinct" -"BIT_XORForEach" -"BIT_XORIf" -"BIT_XORMap" -"BIT_XORMerge" -"BIT_XORNull" -"BIT_XOROrDefault" -"BIT_XOROrNull" -"BIT_XORResample" -"BIT_XORSimpleState" -"BIT_XORState" -"BLAKE3" -"BLOB" "blockNumber" "blockSerializedSize" "blockSize" "bool" -"Bool" "boolean" -"BOTH" "boundingRatio" "boundingRatioArgMax" "boundingRatioArgMin" @@ -570,21 +1414,14 @@ "boundingRatioSimpleState" "boundingRatioState" "buildId" -"BY" -"BYTE" -"BYTEA" "byteHammingDistance" "byteSize" "byteSlice" "byteSwap" -"CASCADE" -"CASE" "caseWithExpr" "caseWithExpression" "caseWithoutExpr" "caseWithoutExpression" -"_CAST" -"CAST" "catboostEvaluate" "categoricalInformationValue" "categoricalInformationValueArgMax" @@ -604,50 +1441,24 @@ "cbrt" "ceil" "ceiling" -"CHANGE" -"CHANGEABLE_IN_READONLY" -"CHANGED" +"changeDay" +"changeHour" +"changeMinute" +"changeMonth" +"changeSecond" +"changeYear" "char" -"CHAR" -"CHARACTER" -"CHARACTER LARGE OBJECT" -"CHARACTER_LENGTH" -"CHARACTER VARYING" -"CHAR LARGE OBJECT" -"CHAR_LENGTH" -"CHAR VARYING" -"CHECK" -"CHECK ALL TABLES" -"CHECK TABLE" "cityHash64" -"CLEANUP" -"CLEAR COLUMN" -"CLEAR INDEX" -"CLEAR PROJECTION" -"CLEAR STATISTIC" -"CLOB" -"CLUSTER" +"clamp" "cluster_host_ids" -"CLUSTERS" -"CN" "coalesce" -"CODEC" -"COLLATE" -"COLUMN" -"COLUMNS" -"COMMENT" -"COMMENT COLUMN" -"COMMIT" -"COMPRESSION" "concat" "concatAssumeInjective" "concatWithSeparator" "concatWithSeparatorAssumeInjective" "concat_ws" -"connection_id" "connectionId" -"CONST" -"CONSTRAINT" +"connection_id" "contingency" "contingencyArgMax" "contingencyArgMin" @@ -735,21 +1546,13 @@ "countSubstringsCaseInsensitive" "countSubstringsCaseInsensitiveUTF8" "covarPop" -"COVAR_POP" "covarPopArgMax" -"COVAR_POPArgMax" "covarPopArgMin" -"COVAR_POPArgMin" "covarPopArray" -"COVAR_POPArray" "covarPopDistinct" -"COVAR_POPDistinct" "covarPopForEach" -"COVAR_POPForEach" "covarPopIf" -"COVAR_POPIf" "covarPopMap" -"COVAR_POPMap" "covarPopMatrix" "covarPopMatrixArgMax" "covarPopMatrixArgMin" @@ -766,17 +1569,11 @@ "covarPopMatrixSimpleState" "covarPopMatrixState" "covarPopMerge" -"COVAR_POPMerge" "covarPopNull" -"COVAR_POPNull" "covarPopOrDefault" -"COVAR_POPOrDefault" "covarPopOrNull" -"COVAR_POPOrNull" "covarPopResample" -"COVAR_POPResample" "covarPopSimpleState" -"COVAR_POPSimpleState" "covarPopStable" "covarPopStableArgMax" "covarPopStableArgMin" @@ -793,23 +1590,14 @@ "covarPopStableSimpleState" "covarPopStableState" "covarPopState" -"COVAR_POPState" "covarSamp" -"COVAR_SAMP" "covarSampArgMax" -"COVAR_SAMPArgMax" "covarSampArgMin" -"COVAR_SAMPArgMin" "covarSampArray" -"COVAR_SAMPArray" "covarSampDistinct" -"COVAR_SAMPDistinct" "covarSampForEach" -"COVAR_SAMPForEach" "covarSampIf" -"COVAR_SAMPIf" "covarSampMap" -"COVAR_SAMPMap" "covarSampMatrix" "covarSampMatrixArgMax" "covarSampMatrixArgMin" @@ -826,17 +1614,11 @@ "covarSampMatrixSimpleState" "covarSampMatrixState" "covarSampMerge" -"COVAR_SAMPMerge" "covarSampNull" -"COVAR_SAMPNull" "covarSampOrDefault" -"COVAR_SAMPOrDefault" "covarSampOrNull" -"COVAR_SAMPOrNull" "covarSampResample" -"COVAR_SAMPResample" "covarSampSimpleState" -"COVAR_SAMPSimpleState" "covarSampStable" "covarSampStableArgMax" "covarSampStableArgMin" @@ -853,7 +1635,6 @@ "covarSampStableSimpleState" "covarSampStableState" "covarSampState" -"COVAR_SAMPState" "cramersV" "cramersVArgMax" "cramersVArgMin" @@ -884,38 +1665,17 @@ "cramersVResample" "cramersVSimpleState" "cramersVState" -"CRC32" -"CRC32IEEE" -"CRC64" -"CREATE" -"CREATE POLICY" -"CREATE PROFILE" -"CREATE QUOTA" -"CREATE ROLE" -"CREATE ROW POLICY" -"CREATE SETTINGS PROFILE" -"CREATE TABLE" -"CREATE TEMPORARY TABLE" -"CREATE USER" -"CROSS" -"CUBE" "curdate" -"current_database" "currentDatabase" -"current_date" -"CURRENT GRANTS" "currentProfiles" -"CURRENT QUOTA" "currentRoles" -"CURRENT ROLES" -"CURRENT ROW" -"current_schemas" "currentSchemas" -"current_timestamp" -"CURRENT TRANSACTION" "currentUser" -"CURRENT_USER" -"CURRENTUSER" +"current_database" +"current_date" +"current_schemas" +"current_timestamp" +"current_user" "cutFragment" "cutIPv6" "cutQueryString" @@ -930,59 +1690,25 @@ "cutToFirstSignificantSubdomainWithWWWRFC" "cutURLParameter" "cutWWW" -"D" "damerauLevenshteinDistance" -"DATABASE" -"DATABASES" -"Date" -"DATE" -"Date32" -"DATE_ADD" -"DATEADD" -"date_diff" "dateDiff" -"DATE_DIFF" -"DATEDIFF" -"DATE_FORMAT" "dateName" -"DATE_SUB" -"DATESUB" -"DateTime" -"DateTime32" -"DateTime64" "dateTime64ToSnowflake" +"dateTime64ToSnowflakeID" "dateTimeToSnowflake" +"dateTimeToSnowflakeID" "dateTrunc" -"DATE_TRUNC" -"DAY" -"DAYOFMONTH" -"DAYOFWEEK" -"DAYOFYEAR" -"DAYS" -"DD" -"DEC" -"Decimal" -"Decimal128" -"Decimal256" -"Decimal32" -"Decimal64" +"date_diff" "decodeHTMLComponent" "decodeURLComponent" "decodeURLFormComponent" "decodeXMLComponent" "decrypt" -"DEDUPLICATE" -"DEFAULT" -"DEFAULT DATABASE" "defaultProfiles" -"DEFAULT ROLE" "defaultRoles" "defaultValueOfArgumentType" "defaultValueOfTypeName" -"DEFINER" "degrees" -"DELETE" -"DELETE WHERE" "deltaSum" "deltaSumArgMax" "deltaSumArgMin" @@ -1014,6 +1740,21 @@ "deltaSumTimestampSimpleState" "deltaSumTimestampState" "demangle" +"denseRank" +"denseRankArgMax" +"denseRankArgMin" +"denseRankArray" +"denseRankDistinct" +"denseRankForEach" +"denseRankIf" +"denseRankMap" +"denseRankMerge" +"denseRankNull" +"denseRankOrDefault" +"denseRankOrNull" +"denseRankResample" +"denseRankSimpleState" +"denseRankState" "dense_rank" "dense_rankArgMax" "dense_rankArgMin" @@ -1029,13 +1770,6 @@ "dense_rankResample" "dense_rankSimpleState" "dense_rankState" -"DEPENDS ON" -"DESC" -"DESCENDING" -"DESCRIBE" -"DETACH" -"DETACH PART" -"DETACH PARTITION" "detectCharset" "detectLanguage" "detectLanguageMixed" @@ -1055,6 +1789,10 @@ "dictGetFloat64" "dictGetFloat64OrDefault" "dictGetHierarchy" +"dictGetIPv4" +"dictGetIPv4OrDefault" +"dictGetIPv6" +"dictGetIPv6OrDefault" "dictGetInt16" "dictGetInt16OrDefault" "dictGetInt32" @@ -1063,10 +1801,6 @@ "dictGetInt64OrDefault" "dictGetInt8" "dictGetInt8OrDefault" -"dictGetIPv4" -"dictGetIPv4OrDefault" -"dictGetIPv6" -"dictGetIPv6OrDefault" "dictGetOrDefault" "dictGetOrNull" "dictGetString" @@ -1082,19 +1816,13 @@ "dictGetUUID" "dictGetUUIDOrDefault" "dictHas" -"DICTIONARIES" -"DICTIONARY" "dictIsIn" -"DISK" "displayName" "distanceL1" "distanceL2" "distanceL2Squared" "distanceLinf" "distanceLp" -"DISTINCT" -"DISTINCT ON" -"DIV" "divide" "divideDecimal" "domain" @@ -1102,29 +1830,13 @@ "domainWithoutWWW" "domainWithoutWWWRFC" "dotProduct" -"DOUBLE" -"DOUBLE PRECISION" -"DOUBLE_SHA1_HASH" -"DOUBLE_SHA1_PASSWORD" -"DROP" -"DROP COLUMN" -"DROP CONSTRAINT" -"DROP DEFAULT" -"DROP DETACHED PART" -"DROP DETACHED PARTITION" -"DROP INDEX" -"DROP PART" -"DROP PARTITION" -"DROP PROJECTION" -"DROP STATISTIC" -"DROP TABLE" -"DROP TEMPORARY TABLE" "dumpColumnStructure" +"dynamicElement" +"dynamicType" "e" "editDistance" -"ELSE" +"editDistanceUTF8" "empty" -"EMPTY" "emptyArrayDate" "emptyArrayDateTime" "emptyArrayFloat32" @@ -1139,19 +1851,14 @@ "emptyArrayUInt32" "emptyArrayUInt64" "emptyArrayUInt8" -"EMPTY AS" "enabledProfiles" "enabledRoles" -"ENABLED ROLES" "encodeURLComponent" "encodeURLFormComponent" "encodeXMLComponent" "encrypt" -"END" "endsWith" "endsWithUTF8" -"ENFORCED" -"ENGINE" "entropy" "entropyArgMax" "entropyArgMin" @@ -1167,33 +1874,14 @@ "entropyResample" "entropySimpleState" "entropyState" -"Enum" -"ENUM" -"Enum16" -"Enum8" -"EPHEMERAL" -"EPHEMERAL SEQUENTIAL" "equals" "erf" "erfc" "errorCodeToName" -"ESTIMATE" "evalMLMethod" -"EVENT" -"EVENTS" -"EVERY" -"EXCEPT" -"EXCEPT DATABASE" -"EXCEPT DATABASES" -"EXCEPT TABLE" -"EXCEPT TABLES" -"EXCHANGE DICTIONARIES" -"EXCHANGE TABLES" -"EXISTS" "exp" "exp10" "exp2" -"EXPLAIN" "exponentialMovingAverage" "exponentialMovingAverageArgMax" "exponentialMovingAverageArgMin" @@ -1269,11 +1957,7 @@ "exponentialTimeDecayedSumResample" "exponentialTimeDecayedSumSimpleState" "exponentialTimeDecayedSumState" -"EXPRESSION" -"EXTENDED" -"EXTERNAL DDL FROM" "extract" -"EXTRACT" "extractAll" "extractAllGroups" "extractAllGroupsHorizontal" @@ -1286,24 +1970,13 @@ "extractURLParameterNames" "extractURLParameters" "factorial" -"FALSE" "farmFingerprint64" "farmHash64" -"FETCH" -"FETCH PART" -"FETCH PARTITION" -"FIELDS" "file" -"FILE" "filesystemAvailable" -"FILESYSTEM CACHE" -"FILESYSTEM CACHES" "filesystemCapacity" "filesystemUnreserved" -"FILTER" -"FINAL" "finalizeAggregation" -"FIRST" "firstLine" "firstSignificantSubdomain" "firstSignificantSubdomainCustom" @@ -1322,6 +1995,8 @@ "first_valueOrDefault" "first_valueOrNull" "first_valueResample" +"first_valueSimpleState" +"first_valueState" "first_value_respect_nulls" "first_value_respect_nullsArgMax" "first_value_respect_nullsArgMin" @@ -1337,10 +2012,6 @@ "first_value_respect_nullsResample" "first_value_respect_nullsSimpleState" "first_value_respect_nullsState" -"first_valueSimpleState" -"first_valueState" -"FIXED" -"FixedString" "flameGraph" "flameGraphArgMax" "flameGraphArgMin" @@ -1358,19 +2029,8 @@ "flameGraphState" "flatten" "flattenTuple" -"FLOAT" -"Float32" -"Float64" "floor" -"FOLLOWING" -"FOR" -"ForEach" -"FOREIGN" -"FOREIGN KEY" -"FORGET PARTITION" "format" -"FORMAT" -"FORMAT_BYTES" "formatDateTime" "formatDateTimeInJodaSyntax" "formatQuery" @@ -1383,67 +2043,50 @@ "formatReadableTimeDelta" "formatRow" "formatRowNoNewline" -"FQDN" "fragment" -"FREEZE" -"FROM" -"FROM_BASE64" -"FROM_DAYS" "fromDaysSinceYearZero" "fromDaysSinceYearZero32" -"FROM INFILE" "fromModifiedJulianDay" "fromModifiedJulianDayOrNull" -"FROM SHARD" -"FROM_UNIXTIME" +"fromUTCTimestamp" "fromUnixTimestamp" "fromUnixTimestamp64Micro" "fromUnixTimestamp64Milli" "fromUnixTimestamp64Nano" "fromUnixTimestampInJodaSyntax" "from_utc_timestamp" -"fromUTCTimestamp" -"FULL" "fullHostName" -"FULLTEXT" -"FUNCTION" "fuzzBits" "gccMurmurHash" "gcd" "generateRandomStructure" +"generateSnowflakeID" "generateULID" "generateUUIDv4" +"generateUUIDv7" "geoDistance" +"geoToH3" +"geoToS2" "geohashDecode" "geohashEncode" "geohashesInBox" -"GEOMETRY" -"geoToH3" -"geoToS2" +"getClientHTTPHeader" "getMacro" "getOSKernelVersion" -"__getScalar" "getServerPort" "getSetting" "getSizeOfEnumType" "getSubcolumn" "getTypeSerializationStreams" -"GLOBAL" "globalIn" -"GLOBAL IN" "globalInIgnoreSet" "globalNotIn" -"GLOBAL NOT IN" "globalNotInIgnoreSet" "globalNotNullIn" "globalNotNullInIgnoreSet" "globalNullIn" "globalNullInIgnoreSet" "globalVariable" -"GRANT" -"GRANTEES" -"GRANT OPTION FOR" -"GRANULARITY" "greatCircleAngle" "greatCircleDistance" "greater" @@ -1584,6 +2227,36 @@ "groupBitAndResample" "groupBitAndSimpleState" "groupBitAndState" +"groupBitOr" +"groupBitOrArgMax" +"groupBitOrArgMin" +"groupBitOrArray" +"groupBitOrDistinct" +"groupBitOrForEach" +"groupBitOrIf" +"groupBitOrMap" +"groupBitOrMerge" +"groupBitOrNull" +"groupBitOrOrDefault" +"groupBitOrOrNull" +"groupBitOrResample" +"groupBitOrSimpleState" +"groupBitOrState" +"groupBitXor" +"groupBitXorArgMax" +"groupBitXorArgMin" +"groupBitXorArray" +"groupBitXorDistinct" +"groupBitXorForEach" +"groupBitXorIf" +"groupBitXorMap" +"groupBitXorMerge" +"groupBitXorNull" +"groupBitXorOrDefault" +"groupBitXorOrNull" +"groupBitXorResample" +"groupBitXorSimpleState" +"groupBitXorState" "groupBitmap" "groupBitmapAnd" "groupBitmapAndArgMax" @@ -1643,39 +2316,21 @@ "groupBitmapXorResample" "groupBitmapXorSimpleState" "groupBitmapXorState" -"groupBitOr" -"groupBitOrArgMax" -"groupBitOrArgMin" -"groupBitOrArray" -"groupBitOrDistinct" -"groupBitOrForEach" -"groupBitOrIf" -"groupBitOrMap" -"groupBitOrMerge" -"groupBitOrNull" -"groupBitOrOrDefault" -"groupBitOrOrNull" -"groupBitOrResample" -"groupBitOrSimpleState" -"groupBitOrState" -"groupBitXor" -"groupBitXorArgMax" -"groupBitXorArgMin" -"groupBitXorArray" -"groupBitXorDistinct" -"groupBitXorForEach" -"groupBitXorIf" -"groupBitXorMap" -"groupBitXorMerge" -"groupBitXorNull" -"groupBitXorOrDefault" -"groupBitXorOrNull" -"groupBitXorResample" -"groupBitXorSimpleState" -"groupBitXorState" -"GROUP BY" -"GROUPING SETS" -"GROUPS" +"groupConcat" +"groupConcatArgMax" +"groupConcatArgMin" +"groupConcatArray" +"groupConcatDistinct" +"groupConcatForEach" +"groupConcatIf" +"groupConcatMap" +"groupConcatMerge" +"groupConcatNull" +"groupConcatOrDefault" +"groupConcatOrNull" +"groupConcatResample" +"groupConcatSimpleState" +"groupConcatState" "groupUniqArray" "groupUniqArrayArgMax" "groupUniqArrayArgMin" @@ -1691,7 +2346,21 @@ "groupUniqArrayResample" "groupUniqArraySimpleState" "groupUniqArrayState" -"H" +"group_concat" +"group_concatArgMax" +"group_concatArgMin" +"group_concatArray" +"group_concatDistinct" +"group_concatForEach" +"group_concatIf" +"group_concatMap" +"group_concatMerge" +"group_concatNull" +"group_concatOrDefault" +"group_concatOrNull" +"group_concatResample" +"group_concatSimpleState" +"group_concatState" "h3CellAreaM2" "h3CellAreaRads2" "h3Distance" @@ -1719,7 +2388,6 @@ "h3IsPentagon" "h3IsResClassIII" "h3IsValid" -"h3kRing" "h3Line" "h3NumHexagons" "h3PointDistKm" @@ -1732,12 +2400,12 @@ "h3ToParent" "h3ToString" "h3UnidirectionalEdgeIsValid" +"h3kRing" "halfMD5" "has" "hasAll" "hasAny" "hasColumnInTable" -"HASH" "hasSubsequence" "hasSubsequenceCaseInsensitive" "hasSubsequenceCaseInsensitiveUTF8" @@ -1748,11 +2416,9 @@ "hasTokenCaseInsensitive" "hasTokenCaseInsensitiveOrNull" "hasTokenOrNull" -"HAVING" -"HDFS" "hex" -"HH" -"HIERARCHICAL" +"hilbertDecode" +"hilbertEncode" "histogram" "histogramArgMax" "histogramArgMin" @@ -1772,73 +2438,33 @@ "hop" "hopEnd" "hopStart" -"HOST" -"hostname" "hostName" -"HOUR" -"HOURS" -"HTTP" +"hostname" "hypot" -"ID" -"IDENTIFIED" "identity" "idnaDecode" "idnaEncode" "if" -"IF EMPTY" -"IF EXISTS" -"IF NOT EXISTS" "ifNotFinite" "ifNull" "ignore" -"IGNORE NULLS" "ilike" -"ILIKE" "in" -"IN" -"INDEX" -"INDEXES" +"inIgnoreSet" "indexHint" "indexOf" -"INDICES" -"INET4" -"INET6" -"INET6_ATON" -"INET6_NTOA" -"INET_ATON" -"INET_NTOA" -"INHERIT" -"inIgnoreSet" "initcap" "initcapUTF8" -"initializeAggregation" -"initial_query_id" "initialQueryID" -"INJECTIVE" -"INNER" -"IN PARTITION" -"INSERT INTO" +"initial_query_id" +"initializeAggregation" "instr" -"INT" -"INT1" -"Int128" -"Int16" -"INT1 SIGNED" -"INT1 UNSIGNED" -"Int256" -"Int32" -"Int64" -"Int8" "intDiv" "intDivOrZero" -"INTEGER" -"INTEGER SIGNED" -"INTEGER UNSIGNED" -"INTERPOLATE" -"INTERSECT" -"INTERVAL" -"IntervalDay" -"IntervalHour" +"intExp10" +"intExp2" +"intHash32" +"intHash64" "intervalLengthSum" "intervalLengthSumArgMax" "intervalLengthSumArgMin" @@ -1854,55 +2480,18 @@ "intervalLengthSumResample" "intervalLengthSumSimpleState" "intervalLengthSumState" -"IntervalMicrosecond" -"IntervalMillisecond" -"IntervalMinute" -"IntervalMonth" -"IntervalNanosecond" -"IntervalQuarter" -"IntervalSecond" -"IntervalWeek" -"IntervalYear" -"intExp10" -"intExp2" -"intHash32" -"intHash64" -"INTO OUTFILE" -"INT SIGNED" -"INT UNSIGNED" -"INVISIBLE" -"INVOKER" -"IP" -"IPv4" -"IPv4CIDRToRange" -"IPv4NumToString" -"IPv4NumToStringClassC" -"IPv4StringToNum" -"IPv4StringToNumOrDefault" -"IPv4StringToNumOrNull" -"IPv4ToIPv6" -"IPv6" -"IPv6CIDRToRange" -"IPv6NumToString" -"IPv6StringToNum" -"IPv6StringToNumOrDefault" -"IPv6StringToNumOrNull" "isConstant" "isDecimalOverflow" "isFinite" -"isInfinite" "isIPAddressInRange" "isIPv4String" "isIPv6String" +"isInfinite" "isNaN" "isNotDistinctFrom" -"IS NOT DISTINCT FROM" "isNotNull" -"IS NOT NULL" "isNull" -"IS NULL" "isNullable" -"IS_OBJECT_ID" "isValidJSON" "isValidUTF8" "isZeroOrNull" @@ -1910,40 +2499,11 @@ "jaroWinklerSimilarity" "javaHash" "javaHashUTF16LE" -"JOIN" "joinGet" "joinGetOrNull" -"JSON" -"JSONArrayLength" -"JSON_ARRAY_LENGTH" -"JSON_EXISTS" -"JSONExtract" -"JSONExtractArrayRaw" -"JSONExtractBool" -"JSONExtractFloat" -"JSONExtractInt" -"JSONExtractKeys" -"JSONExtractKeysAndValues" -"JSONExtractKeysAndValuesRaw" -"JSONExtractRaw" -"JSONExtractString" -"JSONExtractUInt" -"JSONHas" -"JSONKey" -"JSONLength" "jsonMergePatch" -"JSON_QUERY" -"JSONType" -"JSON_VALUE" "jumpConsistentHash" "kafkaMurmurHash" -"KERBEROS" -"KEY" -"KEY BY" -"KEYED BY" -"KEYS" -"KILL" -"KIND" "kolmogorovSmirnovTest" "kolmogorovSmirnovTestArgMax" "kolmogorovSmirnovTestArgMin" @@ -1992,14 +2552,6 @@ "kurtSampResample" "kurtSampSimpleState" "kurtSampState" -"L1Distance" -"L1Norm" -"L1Normalize" -"L2Distance" -"L2Norm" -"L2Normalize" -"L2SquaredDistance" -"L2SquaredNorm" "lagInFrame" "lagInFrameArgMax" "lagInFrameArgMin" @@ -2015,7 +2567,6 @@ "lagInFrameResample" "lagInFrameSimpleState" "lagInFrameState" -"LARGE OBJECT" "largestTriangleThreeBuckets" "largestTriangleThreeBucketsArgMax" "largestTriangleThreeBucketsArgMin" @@ -2031,8 +2582,6 @@ "largestTriangleThreeBucketsResample" "largestTriangleThreeBucketsSimpleState" "largestTriangleThreeBucketsState" -"LAST" -"LAST_DAY" "last_value" "last_valueArgMax" "last_valueArgMin" @@ -2046,6 +2595,8 @@ "last_valueOrDefault" "last_valueOrNull" "last_valueResample" +"last_valueSimpleState" +"last_valueState" "last_value_respect_nulls" "last_value_respect_nullsArgMax" "last_value_respect_nullsArgMin" @@ -2061,12 +2612,8 @@ "last_value_respect_nullsResample" "last_value_respect_nullsSimpleState" "last_value_respect_nullsState" -"last_valueSimpleState" -"last_valueState" -"LAYOUT" "lcase" "lcm" -"LDAP" "leadInFrame" "leadInFrameArgMax" "leadInFrameArgMin" @@ -2082,11 +2629,8 @@ "leadInFrameResample" "leadInFrameSimpleState" "leadInFrameState" -"LEADING" "least" "left" -"LEFT" -"LEFT ARRAY JOIN" "leftPad" "leftPadUTF8" "leftUTF8" @@ -2095,42 +2639,23 @@ "lengthUTF8" "less" "lessOrEquals" -"LESS THAN" -"LEVEL" "levenshteinDistance" +"levenshteinDistanceUTF8" "lgamma" -"LIFETIME" -"LIGHTWEIGHT" "like" -"LIKE" -"LIMIT" -"LINEAR" -"LinfDistance" -"LinfNorm" -"LinfNormalize" -"LIST" -"LIVE" "ln" -"LOCAL" "locate" "log" "log10" "log1p" "log2" "logTrace" -"LONGBLOB" -"LONGTEXT" -"LowCardinality" "lowCardinalityIndices" "lowCardinalityKeys" "lower" "lowerUTF8" "lpad" -"LpDistance" -"LpNorm" -"LpNormalize" "ltrim" -"LTRIM" "lttb" "lttbArgMax" "lttbArgMin" @@ -2146,10 +2671,6 @@ "lttbResample" "lttbSimpleState" "lttbState" -"M" -"MACNumToString" -"MACStringToNum" -"MACStringToOUI" "makeDate" "makeDate32" "makeDateTime" @@ -2170,7 +2691,6 @@ "mannWhitneyUTestSimpleState" "mannWhitneyUTestState" "map" -"Map" "mapAdd" "mapAll" "mapApply" @@ -2181,7 +2701,6 @@ "mapExtractKeyLike" "mapFilter" "mapFromArrays" -"MAP_FROM_ARRAYS" "mapFromString" "mapKeys" "mapPartialReverseSort" @@ -2193,17 +2712,8 @@ "mapUpdate" "mapValues" "match" -"MATCH" "materialize" -"MATERIALIZE" -"MATERIALIZE COLUMN" -"MATERIALIZED" -"MATERIALIZE INDEX" -"MATERIALIZE PROJECTION" -"MATERIALIZE STATISTIC" -"MATERIALIZE TTL" "max" -"MAX" "max2" "maxArgMax" "maxArgMin" @@ -2264,9 +2774,6 @@ "maxResample" "maxSimpleState" "maxState" -"MCS" -"MD4" -"MD5" "meanZTest" "meanZTestArgMax" "meanZTestArgMin" @@ -2507,23 +3014,9 @@ "medianTimingWeightedResample" "medianTimingWeightedSimpleState" "medianTimingWeightedState" -"MEDIUMBLOB" -"MEDIUMINT" -"MEDIUMINT SIGNED" -"MEDIUMINT UNSIGNED" -"MEDIUMTEXT" -"MEMORY" -"Merge" -"MERGES" "metroHash64" -"MI" -"MICROSECOND" -"MICROSECONDS" "mid" -"MILLISECOND" -"MILLISECONDS" "min" -"MIN" "min2" "minArgMax" "minArgMin" @@ -2558,35 +3051,15 @@ "minSimpleState" "minState" "minus" -"MINUTE" -"MINUTES" "mismatches" -"MM" "mod" -"MOD" -"MODIFY" -"MODIFY COLUMN" -"MODIFY COMMENT" -"MODIFY ORDER BY" -"MODIFY QUERY" -"MODIFY REFRESH" -"MODIFY SAMPLE BY" -"MODIFY SETTING" -"MODIFY SQL SECURITY" -"MODIFY TTL" "modulo" "moduloLegacy" "moduloOrZero" -"MONTH" "monthName" -"MONTHS" "mortonDecode" "mortonEncode" -"MOVE" -"MOVE PART" -"MOVE PARTITION" "movingXXX" -"MS" "multiFuzzyMatchAllIndices" "multiFuzzyMatchAny" "multiFuzzyMatchAnyIndex" @@ -2594,9 +3067,6 @@ "multiMatchAllIndices" "multiMatchAny" "multiMatchAnyIndex" -"multiply" -"multiplyDecimal" -"MultiPolygon" "multiSearchAllPositions" "multiSearchAllPositionsCaseInsensitive" "multiSearchAllPositionsCaseInsensitiveUTF8" @@ -2613,31 +3083,17 @@ "multiSearchFirstPositionCaseInsensitive" "multiSearchFirstPositionCaseInsensitiveUTF8" "multiSearchFirstPositionUTF8" +"multiply" +"multiplyDecimal" "murmurHash2_32" "murmurHash2_64" "murmurHash3_128" "murmurHash3_32" "murmurHash3_64" -"MUTATION" -"N" -"NAME" -"NAMED COLLECTION" -"NANOSECOND" -"NANOSECONDS" -"NATIONAL CHAR" -"NATIONAL CHARACTER" -"NATIONAL CHARACTER LARGE OBJECT" -"NATIONAL CHARACTER VARYING" -"NATIONAL CHAR VARYING" -"NCHAR" -"NCHAR LARGE OBJECT" -"NCHAR VARYING" "negate" "neighbor" "nested" -"Nested" "netloc" -"NEXT" "ngramDistance" "ngramDistanceCaseInsensitive" "ngramDistanceCaseInsensitiveUTF8" @@ -2650,7 +3106,6 @@ "ngramMinHashCaseInsensitive" "ngramMinHashCaseInsensitiveUTF8" "ngramMinHashUTF8" -"ngrams" "ngramSearch" "ngramSearchCaseInsensitive" "ngramSearchCaseInsensitiveUTF8" @@ -2659,10 +3114,7 @@ "ngramSimHashCaseInsensitive" "ngramSimHashCaseInsensitiveUTF8" "ngramSimHashUTF8" -"NO ACTION" -"NO DELAY" -"NO LIMITS" -"NONE" +"ngrams" "nonNegativeDerivative" "nonNegativeDerivativeArgMax" "nonNegativeDerivativeArgMin" @@ -2678,9 +3130,11 @@ "nonNegativeDerivativeResample" "nonNegativeDerivativeSimpleState" "nonNegativeDerivativeState" -"NO_PASSWORD" -"normalizedQueryHash" -"normalizedQueryHashKeepNames" +"normL1" +"normL2" +"normL2Squared" +"normLinf" +"normLp" "normalizeL1" "normalizeL2" "normalizeLinf" @@ -2691,18 +3145,18 @@ "normalizeUTF8NFD" "normalizeUTF8NFKC" "normalizeUTF8NFKD" -"normL1" -"normL2" -"normL2Squared" -"normLinf" -"normLp" +"normalizedQueryHash" +"normalizedQueryHashKeepNames" "not" -"NOT" -"NOT BETWEEN" "notEmpty" "notEquals" +"notILike" +"notIn" +"notInIgnoreSet" +"notLike" +"notNullIn" +"notNullInIgnoreSet" "nothing" -"Nothing" "nothingArgMax" "nothingArgMin" "nothingArray" @@ -2746,22 +3200,9 @@ "nothingUInt64Resample" "nothingUInt64SimpleState" "nothingUInt64State" -"NOT IDENTIFIED" -"notILike" -"NOT ILIKE" -"notIn" -"NOT IN" -"notInIgnoreSet" -"NOT KEYED" -"notLike" -"NOT LIKE" -"notNullIn" -"notNullInIgnoreSet" -"NOT OVERRIDABLE" "now" "now64" "nowInBlock" -"NS" "nth_value" "nth_valueArgMax" "nth_valueArgMin" @@ -2792,31 +3233,10 @@ "ntileResample" "ntileSimpleState" "ntileState" -"Null" -"NULL" -"Nullable" "nullIf" "nullIn" "nullInIgnoreSet" -"NULLS" -"NUMERIC" -"NVARCHAR" -"Object" -"OCTET_LENGTH" -"OFFSET" -"ON" -"ON DELETE" -"ONLY" -"ON UPDATE" -"ON VOLUME" -"OPTIMIZE TABLE" "or" -"OR" -"ORDER BY" -"OR REPLACE" -"OUTER" -"OVER" -"OVERRIDABLE" "parseDateTime" "parseDateTime32BestEffort" "parseDateTime32BestEffortOrNull" @@ -2838,32 +3258,49 @@ "parseDateTimeInJodaSyntaxOrZero" "parseDateTimeOrNull" "parseDateTimeOrZero" +"parseReadableSize" +"parseReadableSizeOrNull" +"parseReadableSizeOrZero" "parseTimeDelta" -"PART" -"PARTIAL" -"PARTITION" -"PARTITION BY" +"partitionID" "partitionId" -"PARTITIONS" -"PART_MOVE_TO_SHARD" -"PASTE" "path" "pathFull" -"PERIODIC REFRESH" -"PERMANENTLY" -"PERMISSIVE" -"PERSISTENT" -"PERSISTENT SEQUENTIAL" +"percentRank" +"percentRankArgMax" +"percentRankArgMin" +"percentRankArray" +"percentRankDistinct" +"percentRankForEach" +"percentRankIf" +"percentRankMap" +"percentRankMerge" +"percentRankNull" +"percentRankOrDefault" +"percentRankOrNull" +"percentRankResample" +"percentRankSimpleState" +"percentRankState" +"percent_rank" +"percent_rankArgMax" +"percent_rankArgMin" +"percent_rankArray" +"percent_rankDistinct" +"percent_rankForEach" +"percent_rankIf" +"percent_rankMap" +"percent_rankMerge" +"percent_rankNull" +"percent_rankOrDefault" +"percent_rankOrNull" +"percent_rankResample" +"percent_rankSimpleState" +"percent_rankState" "pi" -"PIPELINE" -"PLAINTEXT_PASSWORD" -"PLAN" "plus" "pmod" -"Point" "pointInEllipses" "pointInPolygon" -"Polygon" "polygonAreaCartesian" "polygonAreaSpherical" "polygonConvexHullCartesian" @@ -2880,32 +3317,21 @@ "polygonsUnionSpherical" "polygonsWithinCartesian" "polygonsWithinSpherical" -"POPULATE" "port" "portRFC" "position" "positionCaseInsensitive" "positionCaseInsensitiveUTF8" "positionUTF8" -"positive_modulo" "positiveModulo" +"positive_modulo" "pow" "power" -"PRECEDING" -"PRECISION" -"PREWHERE" -"PRIMARY" -"PRIMARY KEY" -"PROFILE" -"PROJECTION" +"printf" "proportionsZTest" -"Protobuf" "protocol" -"PULL" "punycodeDecode" "punycodeEncode" -"Q" -"QQ" "quantile" "quantileArgMax" "quantileArgMin" @@ -3099,6 +3525,68 @@ "quantileOrDefault" "quantileOrNull" "quantileResample" +"quantileSimpleState" +"quantileState" +"quantileTDigest" +"quantileTDigestArgMax" +"quantileTDigestArgMin" +"quantileTDigestArray" +"quantileTDigestDistinct" +"quantileTDigestForEach" +"quantileTDigestIf" +"quantileTDigestMap" +"quantileTDigestMerge" +"quantileTDigestNull" +"quantileTDigestOrDefault" +"quantileTDigestOrNull" +"quantileTDigestResample" +"quantileTDigestSimpleState" +"quantileTDigestState" +"quantileTDigestWeighted" +"quantileTDigestWeightedArgMax" +"quantileTDigestWeightedArgMin" +"quantileTDigestWeightedArray" +"quantileTDigestWeightedDistinct" +"quantileTDigestWeightedForEach" +"quantileTDigestWeightedIf" +"quantileTDigestWeightedMap" +"quantileTDigestWeightedMerge" +"quantileTDigestWeightedNull" +"quantileTDigestWeightedOrDefault" +"quantileTDigestWeightedOrNull" +"quantileTDigestWeightedResample" +"quantileTDigestWeightedSimpleState" +"quantileTDigestWeightedState" +"quantileTiming" +"quantileTimingArgMax" +"quantileTimingArgMin" +"quantileTimingArray" +"quantileTimingDistinct" +"quantileTimingForEach" +"quantileTimingIf" +"quantileTimingMap" +"quantileTimingMerge" +"quantileTimingNull" +"quantileTimingOrDefault" +"quantileTimingOrNull" +"quantileTimingResample" +"quantileTimingSimpleState" +"quantileTimingState" +"quantileTimingWeighted" +"quantileTimingWeightedArgMax" +"quantileTimingWeightedArgMin" +"quantileTimingWeightedArray" +"quantileTimingWeightedDistinct" +"quantileTimingWeightedForEach" +"quantileTimingWeightedIf" +"quantileTimingWeightedMap" +"quantileTimingWeightedMerge" +"quantileTimingWeightedNull" +"quantileTimingWeightedOrDefault" +"quantileTimingWeightedOrNull" +"quantileTimingWeightedResample" +"quantileTimingWeightedSimpleState" +"quantileTimingWeightedState" "quantiles" "quantilesArgMax" "quantilesArgMin" @@ -3271,7 +3759,6 @@ "quantilesGKSimpleState" "quantilesGKState" "quantilesIf" -"quantileSimpleState" "quantilesInterpolatedWeighted" "quantilesInterpolatedWeightedArgMax" "quantilesInterpolatedWeightedArgMin" @@ -3295,7 +3782,6 @@ "quantilesResample" "quantilesSimpleState" "quantilesState" -"quantileState" "quantilesTDigest" "quantilesTDigestArgMax" "quantilesTDigestArgMin" @@ -3356,75 +3842,10 @@ "quantilesTimingWeightedResample" "quantilesTimingWeightedSimpleState" "quantilesTimingWeightedState" -"quantileTDigest" -"quantileTDigestArgMax" -"quantileTDigestArgMin" -"quantileTDigestArray" -"quantileTDigestDistinct" -"quantileTDigestForEach" -"quantileTDigestIf" -"quantileTDigestMap" -"quantileTDigestMerge" -"quantileTDigestNull" -"quantileTDigestOrDefault" -"quantileTDigestOrNull" -"quantileTDigestResample" -"quantileTDigestSimpleState" -"quantileTDigestState" -"quantileTDigestWeighted" -"quantileTDigestWeightedArgMax" -"quantileTDigestWeightedArgMin" -"quantileTDigestWeightedArray" -"quantileTDigestWeightedDistinct" -"quantileTDigestWeightedForEach" -"quantileTDigestWeightedIf" -"quantileTDigestWeightedMap" -"quantileTDigestWeightedMerge" -"quantileTDigestWeightedNull" -"quantileTDigestWeightedOrDefault" -"quantileTDigestWeightedOrNull" -"quantileTDigestWeightedResample" -"quantileTDigestWeightedSimpleState" -"quantileTDigestWeightedState" -"quantileTiming" -"quantileTimingArgMax" -"quantileTimingArgMin" -"quantileTimingArray" -"quantileTimingDistinct" -"quantileTimingForEach" -"quantileTimingIf" -"quantileTimingMap" -"quantileTimingMerge" -"quantileTimingNull" -"quantileTimingOrDefault" -"quantileTimingOrNull" -"quantileTimingResample" -"quantileTimingSimpleState" -"quantileTimingState" -"quantileTimingWeighted" -"quantileTimingWeightedArgMax" -"quantileTimingWeightedArgMin" -"quantileTimingWeightedArray" -"quantileTimingWeightedDistinct" -"quantileTimingWeightedForEach" -"quantileTimingWeightedIf" -"quantileTimingWeightedMap" -"quantileTimingWeightedMerge" -"quantileTimingWeightedNull" -"quantileTimingWeightedOrDefault" -"quantileTimingWeightedOrNull" -"quantileTimingWeightedResample" -"quantileTimingWeightedSimpleState" -"quantileTimingWeightedState" -"QUARTER" -"QUARTERS" -"QUERY" -"query_id" "queryID" "queryString" "queryStringAndFragment" -"QUERY TREE" -"QUOTA" +"query_id" "radians" "rand" "rand32" @@ -3439,17 +3860,14 @@ "randLogNormal" "randNegativeBinomial" "randNormal" -"randomFixedString" -"RANDOMIZED" -"RANDOMIZE FOR" -"randomPrintableASCII" -"randomString" -"randomStringUTF8" "randPoisson" "randStudentT" "randUniform" +"randomFixedString" +"randomPrintableASCII" +"randomString" +"randomStringUTF8" "range" -"RANGE" "rank" "rankArgMax" "rankArgMin" @@ -3480,22 +3898,14 @@ "rankResample" "rankSimpleState" "rankState" -"READONLY" +"readWKTLineString" +"readWKTMultiLineString" "readWKTMultiPolygon" "readWKTPoint" "readWKTPolygon" "readWKTRing" -"REAL" -"REALM" -"RECOMPRESS" -"REFERENCES" -"REFRESH" -"REGEXP" "regexpExtract" -"REGEXP_EXTRACT" -"REGEXP_MATCHES" "regexpQuoteMeta" -"REGEXP_REPLACE" "regionHierarchy" "regionIn" "regionToArea" @@ -3526,31 +3936,13 @@ "reinterpretAsUInt64" "reinterpretAsUInt8" "reinterpretAsUUID" -"REMOVE" -"REMOVE SAMPLE BY" -"REMOVE TTL" -"RENAME" -"RENAME COLUMN" -"RENAME DATABASE" -"RENAME DICTIONARY" -"RENAME TABLE" -"RENAME TO" "repeat" "replace" -"REPLACE" "replaceAll" "replaceOne" -"REPLACE PARTITION" "replaceRegexpAll" "replaceRegexpOne" "replicate" -"Resample" -"RESET SETTING" -"RESPECT NULLS" -"RESTORE" -"RESTRICT" -"RESTRICTIVE" -"RESUME" "retention" "retentionArgMax" "retentionArgMin" @@ -3569,22 +3961,18 @@ "reverse" "reverseUTF8" "revision" -"REVOKE" "right" -"RIGHT" "rightPad" "rightPadUTF8" "rightUTF8" -"Ring" -"ROLLBACK" -"ROLLUP" "round" "roundAge" "roundBankers" "roundDown" "roundDuration" "roundToExp2" -"ROW" +"rowNumberInAllBlocks" +"rowNumberInBlock" "row_number" "row_numberArgMax" "row_numberArgMin" @@ -3592,8 +3980,6 @@ "row_numberDistinct" "row_numberForEach" "row_numberIf" -"rowNumberInAllBlocks" -"rowNumberInBlock" "row_numberMap" "row_numberMerge" "row_numberNull" @@ -3602,15 +3988,12 @@ "row_numberResample" "row_numberSimpleState" "row_numberState" -"ROWS" "rpad" "rtrim" -"RTRIM" "runningAccumulate" "runningConcurrency" "runningDifference" "runningDifferenceStartingWithFirstValue" -"S" "s2CapContains" "s2CapUnion" "s2CellsIntersect" @@ -3620,18 +4003,7 @@ "s2RectIntersection" "s2RectUnion" "s2ToGeo" -"S3" -"SALT" -"SAMPLE" -"SAMPLE BY" "scalarProduct" -"__scalarSubqueryResult" -"SCHEMA" -"SCHEME" -"SECOND" -"SECONDS" -"SELECT" -"SEMI" "sequenceCount" "sequenceCountArgMax" "sequenceCountArgMin" @@ -3680,44 +4052,14 @@ "seriesDecomposeSTL" "seriesOutliersDetectTukey" "seriesPeriodDetectFFT" -"SERVER" -"serverTimezone" "serverTimeZone" +"serverTimezone" "serverUUID" -"SET" -"SET DEFAULT" -"SET DEFAULT ROLE" -"SET FAKE TIME" -"SET NULL" -"SET ROLE" -"SET ROLE DEFAULT" -"SETTINGS" -"SET TRANSACTION SNAPSHOT" -"SHA1" -"SHA224" -"SHA256" -"SHA256_HASH" -"SHA256_PASSWORD" -"SHA384" -"SHA512" -"SHA512_256" "shardCount" "shardNum" -"SHOW" -"SHOW ACCESS" "showCertificate" -"SHOW CREATE" -"SHOW ENGINES" -"SHOW FUNCTIONS" -"SHOW GRANTS" -"SHOW PRIVILEGES" -"SHOW PROCESSLIST" -"SHOW SETTING" "sigmoid" "sign" -"SIGNED" -"SIMPLE" -"SimpleAggregateFunction" "simpleJSONExtractBool" "simpleJSONExtractFloat" "simpleJSONExtractInt" @@ -3741,7 +4083,6 @@ "simpleLinearRegressionSimpleState" "simpleLinearRegressionState" "sin" -"SINGLE" "singleValueOrNull" "singleValueOrNullArgMax" "singleValueOrNullArgMin" @@ -3796,45 +4137,42 @@ "skewSampState" "sleep" "sleepEachRow" -"SMALLINT" -"SMALLINT SIGNED" -"SMALLINT UNSIGNED" +"snowflakeIDToDateTime" +"snowflakeIDToDateTime64" "snowflakeToDateTime" "snowflakeToDateTime64" "soundex" -"SOURCE" "space" -"sparkbar" "sparkBar" -"sparkbarArgMax" "sparkBarArgMax" -"sparkbarArgMin" "sparkBarArgMin" -"sparkbarArray" "sparkBarArray" -"sparkbarDistinct" "sparkBarDistinct" -"sparkbarForEach" "sparkBarForEach" -"sparkbarIf" "sparkBarIf" -"sparkbarMap" "sparkBarMap" -"sparkbarMerge" "sparkBarMerge" -"sparkbarNull" "sparkBarNull" -"sparkbarOrDefault" "sparkBarOrDefault" -"sparkbarOrNull" "sparkBarOrNull" -"sparkbarResample" "sparkBarResample" -"sparkbarSimpleState" "sparkBarSimpleState" -"sparkbarState" "sparkBarState" -"SPATIAL" +"sparkbar" +"sparkbarArgMax" +"sparkbarArgMin" +"sparkbarArray" +"sparkbarDistinct" +"sparkbarForEach" +"sparkbarIf" +"sparkbarMap" +"sparkbarMerge" +"sparkbarNull" +"sparkbarOrDefault" +"sparkbarOrNull" +"sparkbarResample" +"sparkbarSimpleState" +"sparkbarState" "splitByAlpha" "splitByChar" "splitByNonAlpha" @@ -3844,58 +4182,23 @@ "sqid" "sqidDecode" "sqidEncode" -"SQL SECURITY" -"SQL_TSI_DAY" -"SQL_TSI_HOUR" -"SQL_TSI_MICROSECOND" -"SQL_TSI_MILLISECOND" -"SQL_TSI_MINUTE" -"SQL_TSI_MONTH" -"SQL_TSI_NANOSECOND" -"SQL_TSI_QUARTER" -"SQL_TSI_SECOND" -"SQL_TSI_WEEK" -"SQL_TSI_YEAR" "sqrt" -"SS" -"SSH_KEY" -"SSL_CERTIFICATE" "startsWith" "startsWithUTF8" -"State" -"STATISTIC" -"STD" -"STDArgMax" -"STDArgMin" -"STDArray" "stddevPop" -"STDDEV_POP" "stddevPopArgMax" -"STDDEV_POPArgMax" "stddevPopArgMin" -"STDDEV_POPArgMin" "stddevPopArray" -"STDDEV_POPArray" "stddevPopDistinct" -"STDDEV_POPDistinct" "stddevPopForEach" -"STDDEV_POPForEach" "stddevPopIf" -"STDDEV_POPIf" "stddevPopMap" -"STDDEV_POPMap" "stddevPopMerge" -"STDDEV_POPMerge" "stddevPopNull" -"STDDEV_POPNull" "stddevPopOrDefault" -"STDDEV_POPOrDefault" "stddevPopOrNull" -"STDDEV_POPOrNull" "stddevPopResample" -"STDDEV_POPResample" "stddevPopSimpleState" -"STDDEV_POPSimpleState" "stddevPopStable" "stddevPopStableArgMax" "stddevPopStableArgMin" @@ -3912,35 +4215,20 @@ "stddevPopStableSimpleState" "stddevPopStableState" "stddevPopState" -"STDDEV_POPState" "stddevSamp" -"STDDEV_SAMP" "stddevSampArgMax" -"STDDEV_SAMPArgMax" "stddevSampArgMin" -"STDDEV_SAMPArgMin" "stddevSampArray" -"STDDEV_SAMPArray" "stddevSampDistinct" -"STDDEV_SAMPDistinct" "stddevSampForEach" -"STDDEV_SAMPForEach" "stddevSampIf" -"STDDEV_SAMPIf" "stddevSampMap" -"STDDEV_SAMPMap" "stddevSampMerge" -"STDDEV_SAMPMerge" "stddevSampNull" -"STDDEV_SAMPNull" "stddevSampOrDefault" -"STDDEV_SAMPOrDefault" "stddevSampOrNull" -"STDDEV_SAMPOrNull" "stddevSampResample" -"STDDEV_SAMPResample" "stddevSampSimpleState" -"STDDEV_SAMPSimpleState" "stddevSampStable" "stddevSampStableArgMax" "stddevSampStableArgMin" @@ -3957,20 +4245,7 @@ "stddevSampStableSimpleState" "stddevSampStableState" "stddevSampState" -"STDDEV_SAMPState" -"STDDistinct" -"STDForEach" -"STDIf" -"STDMap" -"STDMerge" -"STDNull" -"STDOrDefault" -"STDOrNull" -"STDResample" -"STDSimpleState" -"STDState" "stem" -"STEP" "stochasticLinearRegression" "stochasticLinearRegressionArgMax" "stochasticLinearRegressionArgMin" @@ -4001,15 +4276,11 @@ "stochasticLogisticRegressionResample" "stochasticLogisticRegressionSimpleState" "stochasticLogisticRegressionState" -"STORAGE" -"STRICT" -"STRICTLY_ASCENDING" -"String" +"str_to_date" +"str_to_map" "stringJaccardIndex" "stringJaccardIndexUTF8" "stringToH3" -"str_to_date" -"str_to_map" "structureToCapnProtoSchema" "structureToProtobufSchema" "studentTTest" @@ -4029,14 +4300,9 @@ "studentTTestState" "subBitmap" "subDate" -"SUBPARTITION" -"SUBPARTITION BY" -"SUBPARTITIONS" "substr" "substring" -"SUBSTRING" "substringIndex" -"SUBSTRING_INDEX" "substringIndexUTF8" "substringUTF8" "subtractDays" @@ -4120,21 +4386,6 @@ "sumMapFilteredWithOverflowResample" "sumMapFilteredWithOverflowSimpleState" "sumMapFilteredWithOverflowState" -"sumMappedArrays" -"sumMappedArraysArgMax" -"sumMappedArraysArgMin" -"sumMappedArraysArray" -"sumMappedArraysDistinct" -"sumMappedArraysForEach" -"sumMappedArraysIf" -"sumMappedArraysMap" -"sumMappedArraysMerge" -"sumMappedArraysNull" -"sumMappedArraysOrDefault" -"sumMappedArraysOrNull" -"sumMappedArraysResample" -"sumMappedArraysSimpleState" -"sumMappedArraysState" "sumMapWithOverflow" "sumMapWithOverflowArgMax" "sumMapWithOverflowArgMin" @@ -4150,6 +4401,21 @@ "sumMapWithOverflowResample" "sumMapWithOverflowSimpleState" "sumMapWithOverflowState" +"sumMappedArrays" +"sumMappedArraysArgMax" +"sumMappedArraysArgMin" +"sumMappedArraysArray" +"sumMappedArraysDistinct" +"sumMappedArraysForEach" +"sumMappedArraysIf" +"sumMappedArraysMap" +"sumMappedArraysMerge" +"sumMappedArraysNull" +"sumMappedArraysOrDefault" +"sumMappedArraysOrNull" +"sumMappedArraysResample" +"sumMappedArraysSimpleState" +"sumMappedArraysState" "sumMerge" "sumNull" "sumOrDefault" @@ -4172,23 +4438,11 @@ "sumWithOverflowResample" "sumWithOverflowSimpleState" "sumWithOverflowState" -"SUSPEND" "svg" -"SVG" -"SYNC" "synonyms" -"SYNTAX" -"SYSTEM" -"TABLE" -"TABLE OVERRIDE" -"TABLES" "tan" "tanh" "tcpPort" -"TEMPORARY" -"TEMPORARY TABLE" -"TEST" -"TEXT" "tgamma" "theilsU" "theilsUArgMax" @@ -4205,36 +4459,20 @@ "theilsUResample" "theilsUSimpleState" "theilsUState" -"THEN" "throwIf" "tid" -"TIME" "timeDiff" "timeSlot" "timeSlots" -"timestamp" -"TIMESTAMP" -"TIMESTAMP_ADD" -"TIMESTAMPADD" -"timestamp_diff" -"timestampDiff" -"TIMESTAMP_DIFF" -"TIMESTAMPDIFF" -"TIMESTAMP_SUB" -"TIMESTAMPSUB" -"timezone" "timeZone" -"timezoneOf" "timeZoneOf" -"timezoneOffset" "timeZoneOffset" -"TINYBLOB" -"TINYINT" -"TINYINT SIGNED" -"TINYINT UNSIGNED" -"TINYTEXT" -"TO" -"TO_BASE64" +"timestamp" +"timestampDiff" +"timestamp_diff" +"timezone" +"timezoneOf" +"timezoneOffset" "toBool" "toColumnTypeName" "toDate" @@ -4254,11 +4492,9 @@ "toDateTimeOrDefault" "toDateTimeOrNull" "toDateTimeOrZero" -"today" "toDayOfMonth" "toDayOfWeek" "toDayOfYear" -"TO_DAYS" "toDaysSinceYearZero" "toDecimal128" "toDecimal128OrDefault" @@ -4277,7 +4513,6 @@ "toDecimal64OrNull" "toDecimal64OrZero" "toDecimalString" -"TO DISK" "toFixedString" "toFloat32" "toFloat32OrDefault" @@ -4288,7 +4523,16 @@ "toFloat64OrNull" "toFloat64OrZero" "toHour" -"TO INNER UUID" +"toIPv4" +"toIPv4OrDefault" +"toIPv4OrNull" +"toIPv4OrZero" +"toIPv6" +"toIPv6OrDefault" +"toIPv6OrNull" +"toIPv6OrZero" +"toISOWeek" +"toISOYear" "toInt128" "toInt128OrDefault" "toInt128OrNull" @@ -4324,18 +4568,7 @@ "toIntervalSecond" "toIntervalWeek" "toIntervalYear" -"toIPv4" -"toIPv4OrDefault" -"toIPv4OrNull" -"toIPv4OrZero" -"toIPv6" -"toIPv6OrDefault" -"toIPv6OrNull" -"toIPv6OrZero" -"toISOWeek" -"toISOYear" "toJSONString" -"tokens" "toLastDayOfMonth" "toLastDayOfWeek" "toLowCardinality" @@ -4346,7 +4579,82 @@ "toMonday" "toMonth" "toNullable" -"TOP" +"toQuarter" +"toRelativeDayNum" +"toRelativeHourNum" +"toRelativeMinuteNum" +"toRelativeMonthNum" +"toRelativeQuarterNum" +"toRelativeSecondNum" +"toRelativeWeekNum" +"toRelativeYearNum" +"toSecond" +"toStartOfDay" +"toStartOfFifteenMinutes" +"toStartOfFiveMinute" +"toStartOfFiveMinutes" +"toStartOfHour" +"toStartOfISOYear" +"toStartOfInterval" +"toStartOfMicrosecond" +"toStartOfMillisecond" +"toStartOfMinute" +"toStartOfMonth" +"toStartOfNanosecond" +"toStartOfQuarter" +"toStartOfSecond" +"toStartOfTenMinutes" +"toStartOfWeek" +"toStartOfYear" +"toString" +"toStringCutToZero" +"toTime" +"toTimeZone" +"toTimezone" +"toTypeName" +"toUInt128" +"toUInt128OrDefault" +"toUInt128OrNull" +"toUInt128OrZero" +"toUInt16" +"toUInt16OrDefault" +"toUInt16OrNull" +"toUInt16OrZero" +"toUInt256" +"toUInt256OrDefault" +"toUInt256OrNull" +"toUInt256OrZero" +"toUInt32" +"toUInt32OrDefault" +"toUInt32OrNull" +"toUInt32OrZero" +"toUInt64" +"toUInt64OrDefault" +"toUInt64OrNull" +"toUInt64OrZero" +"toUInt8" +"toUInt8OrDefault" +"toUInt8OrNull" +"toUInt8OrZero" +"toUTCTimestamp" +"toUUID" +"toUUIDOrDefault" +"toUUIDOrNull" +"toUUIDOrZero" +"toUnixTimestamp" +"toUnixTimestamp64Micro" +"toUnixTimestamp64Milli" +"toUnixTimestamp64Nano" +"toValidUTF8" +"toWeek" +"toYYYYMM" +"toYYYYMMDD" +"toYYYYMMDDhhmmss" +"toYear" +"toYearWeek" +"to_utc_timestamp" +"today" +"tokens" "topK" "topKArgMax" "topKArgMin" @@ -4379,115 +4687,28 @@ "topKWeightedState" "topLevelDomain" "topLevelDomainRFC" -"toQuarter" -"toRelativeDayNum" -"toRelativeHourNum" -"toRelativeMinuteNum" -"toRelativeMonthNum" -"toRelativeQuarterNum" -"toRelativeSecondNum" -"toRelativeWeekNum" -"toRelativeYearNum" -"toSecond" -"TO SHARD" -"toStartOfDay" -"toStartOfFifteenMinutes" -"toStartOfFiveMinute" -"toStartOfFiveMinutes" -"toStartOfHour" -"toStartOfInterval" -"toStartOfISOYear" -"toStartOfMicrosecond" -"toStartOfMillisecond" -"toStartOfMinute" -"toStartOfMonth" -"toStartOfNanosecond" -"toStartOfQuarter" -"toStartOfSecond" -"toStartOfTenMinutes" -"toStartOfWeek" -"toStartOfYear" -"toString" -"toStringCutToZero" -"TO TABLE" -"TOTALS" -"toTime" -"toTimezone" -"toTimeZone" -"toTypeName" -"toUInt128" -"toUInt128OrDefault" -"toUInt128OrNull" -"toUInt128OrZero" -"toUInt16" -"toUInt16OrDefault" -"toUInt16OrNull" -"toUInt16OrZero" -"toUInt256" -"toUInt256OrDefault" -"toUInt256OrNull" -"toUInt256OrZero" -"toUInt32" -"toUInt32OrDefault" -"toUInt32OrNull" -"toUInt32OrZero" -"toUInt64" -"toUInt64OrDefault" -"toUInt64OrNull" -"toUInt64OrZero" -"toUInt8" -"toUInt8OrDefault" -"toUInt8OrNull" -"toUInt8OrZero" -"TO_UNIXTIME" -"toUnixTimestamp" -"toUnixTimestamp64Micro" -"toUnixTimestamp64Milli" -"toUnixTimestamp64Nano" -"to_utc_timestamp" -"toUTCTimestamp" -"toUUID" -"toUUIDOrDefault" -"toUUIDOrNull" -"toUUIDOrZero" -"toValidUTF8" -"TO VOLUME" -"toWeek" -"toYear" -"toYearWeek" -"toYYYYMM" -"toYYYYMMDD" -"toYYYYMMDDhhmmss" -"TRACKING ONLY" -"TRAILING" -"TRANSACTION" "transactionID" "transactionLatestSnapshot" "transactionOldestSnapshot" "transform" "translate" "translateUTF8" -"TRIGGER" "trim" -"TRIM" "trimBoth" "trimLeft" "trimRight" -"TRUE" "trunc" "truncate" -"TRUNCATE" "tryBase58Decode" "tryBase64Decode" +"tryBase64URLDecode" "tryDecrypt" "tryIdnaEncode" "tryPunycodeDecode" -"TTL" "tumble" "tumbleEnd" "tumbleStart" "tuple" -"Tuple" "tupleConcat" "tupleDivide" "tupleDivideByNumber" @@ -4502,25 +4723,13 @@ "tupleModuloByNumber" "tupleMultiply" "tupleMultiplyByNumber" +"tupleNames" "tupleNegate" "tuplePlus" "tupleToNameValuePairs" -"TYPE" -"TYPEOF" "ucase" -"UInt128" -"UInt16" -"UInt256" -"UInt32" -"UInt64" -"UInt8" -"ULIDStringToDateTime" "unbin" -"UNBOUNDED" -"UNDROP" -"UNFREEZE" "unhex" -"UNION" "uniq" "uniqArgMax" "uniqArgMin" @@ -4614,7 +4823,6 @@ "uniqThetaSimpleState" "uniqThetaState" "uniqThetaUnion" -"UNIQUE" "uniqUpTo" "uniqUpToArgMax" "uniqUpToArgMin" @@ -4630,61 +4838,25 @@ "uniqUpToResample" "uniqUpToSimpleState" "uniqUpToState" -"UNSET FAKE TIME" -"UNSIGNED" -"UPDATE" "upper" "upperUTF8" "uptime" -"URL" -"URLHash" -"URLHierarchy" -"URLPathHierarchy" -"USE" "user" -"USING" -"UTC_timestamp" -"UTCTimestamp" -"UUID" -"UUIDNumToString" -"UUIDStringToNum" "validateNestedArraySizes" -"VALID UNTIL" -"VALUES" -"VARBINARY" -"VARCHAR" -"VARCHAR2" -"Variant" -"variantElement" -"variantType" "varPop" -"VAR_POP" "varPopArgMax" -"VAR_POPArgMax" "varPopArgMin" -"VAR_POPArgMin" "varPopArray" -"VAR_POPArray" "varPopDistinct" -"VAR_POPDistinct" "varPopForEach" -"VAR_POPForEach" "varPopIf" -"VAR_POPIf" "varPopMap" -"VAR_POPMap" "varPopMerge" -"VAR_POPMerge" "varPopNull" -"VAR_POPNull" "varPopOrDefault" -"VAR_POPOrDefault" "varPopOrNull" -"VAR_POPOrNull" "varPopResample" -"VAR_POPResample" "varPopSimpleState" -"VAR_POPSimpleState" "varPopStable" "varPopStableArgMax" "varPopStableArgMin" @@ -4701,35 +4873,20 @@ "varPopStableSimpleState" "varPopStableState" "varPopState" -"VAR_POPState" "varSamp" -"VAR_SAMP" "varSampArgMax" -"VAR_SAMPArgMax" "varSampArgMin" -"VAR_SAMPArgMin" "varSampArray" -"VAR_SAMPArray" "varSampDistinct" -"VAR_SAMPDistinct" "varSampForEach" -"VAR_SAMPForEach" "varSampIf" -"VAR_SAMPIf" "varSampMap" -"VAR_SAMPMap" "varSampMerge" -"VAR_SAMPMerge" "varSampNull" -"VAR_SAMPNull" "varSampOrDefault" -"VAR_SAMPOrDefault" "varSampOrNull" -"VAR_SAMPOrNull" "varSampResample" -"VAR_SAMPResample" "varSampSimpleState" -"VAR_SAMPSimpleState" "varSampStable" "varSampStableArgMax" "varSampStableArgMin" @@ -4746,13 +4903,11 @@ "varSampStableSimpleState" "varSampStableState" "varSampState" -"VAR_SAMPState" -"VARYING" +"variantElement" +"variantType" "vectorDifference" "vectorSum" "version" -"VIEW" -"VISIBLE" "visibleWidth" "visitParamExtractBool" "visitParamExtractFloat" @@ -4761,11 +4916,7 @@ "visitParamExtractString" "visitParamExtractUInt" "visitParamHas" -"WATCH" -"WATERMARK" "week" -"WEEK" -"WEEKS" "welchTTest" "welchTTestArgMax" "welchTTestArgMin" @@ -4781,11 +4932,8 @@ "welchTTestResample" "welchTTestSimpleState" "welchTTestState" -"WHEN" -"WHERE" -"width_bucket" "widthBucket" -"WINDOW" +"width_bucket" "windowFunnel" "windowFunnelArgMax" "windowFunnelArgMin" @@ -4802,16 +4950,6 @@ "windowFunnelSimpleState" "windowFunnelState" "windowID" -"WITH" -"WITH ADMIN OPTION" -"WITH CHECK" -"WITH FILL" -"WITH GRANT OPTION" -"with_itemindex" -"WITH NAME" -"WITH REPLACE OPTION" -"WITH TIES" -"WK" "wkt" "wordShingleMinHash" "wordShingleMinHashArg" @@ -4825,23 +4963,12 @@ "wordShingleSimHashCaseInsensitive" "wordShingleSimHashCaseInsensitiveUTF8" "wordShingleSimHashUTF8" -"WRITABLE" -"WW" "wyHash64" "xor" -"xxh3" "xxHash32" "xxHash64" +"xxh3" "yandexConsistentHash" -"YEAR" -"YEARS" "yearweek" "yesterday" -"YY" -"YYYY" -"YYYYMMDDhhmmssToDateTime" -"YYYYMMDDhhmmssToDateTime64" -"YYYYMMDDToDate" -"YYYYMMDDToDate32" -"ZKPATH" "zookeeperSessionUptime" diff --git a/tests/fuzz/dictionaries/datatypes.dict b/tests/fuzz/dictionaries/datatypes.dict index a01a94fd3e3..797905203b2 100644 --- a/tests/fuzz/dictionaries/datatypes.dict +++ b/tests/fuzz/dictionaries/datatypes.dict @@ -31,6 +31,7 @@ "Decimal256" "Decimal32" "Decimal64" +"Dynamic" "ENUM" "Enum" "Enum16" @@ -74,6 +75,7 @@ "JSON" "LONGBLOB" "LONGTEXT" +"LineString" "LowCardinality" "MEDIUMBLOB" "MEDIUMINT" @@ -81,6 +83,7 @@ "MEDIUMINT UNSIGNED" "MEDIUMTEXT" "Map" +"MultiLineString" "MultiPolygon" "NATIONAL CHAR" "NATIONAL CHAR VARYING" @@ -132,4 +135,3 @@ "YEAR" "bool" "boolean" -"Dynamic" diff --git a/tests/fuzz/dictionaries/functions.dict b/tests/fuzz/dictionaries/functions.dict index 6f2a88c22fa..e562595fb67 100644 --- a/tests/fuzz/dictionaries/functions.dict +++ b/tests/fuzz/dictionaries/functions.dict @@ -126,6 +126,7 @@ "JSONHas" "JSONKey" "JSONLength" +"JSONMergePatch" "JSONType" "JSON_ARRAY_LENGTH" "JSON_EXISTS" @@ -227,6 +228,8 @@ "UTC_timestamp" "UUIDNumToString" "UUIDStringToNum" +"UUIDToNum" +"UUIDv7ToDateTime" "VAR_POP" "VAR_POPArgMax" "VAR_POPArgMin" @@ -263,6 +266,7 @@ "YYYYMMDDhhmmssToDateTime" "YYYYMMDDhhmmssToDateTime64" "_CAST" +"__actionName" "__bitBoolMaskAnd" "__bitBoolMaskOr" "__bitSwapLastTwo" @@ -660,6 +664,8 @@ "base58Encode" "base64Decode" "base64Encode" +"base64URLDecode" +"base64URLEncode" "basename" "bin" "bitAnd" @@ -744,8 +750,15 @@ "cbrt" "ceil" "ceiling" +"changeDay" +"changeHour" +"changeMinute" +"changeMonth" +"changeSecond" +"changeYear" "char" "cityHash64" +"clamp" "coalesce" "concat" "concatAssumeInjective" @@ -970,6 +983,7 @@ "current_date" "current_schemas" "current_timestamp" +"current_user" "cutFragment" "cutIPv6" "cutQueryString" @@ -988,7 +1002,9 @@ "dateDiff" "dateName" "dateTime64ToSnowflake" +"dateTime64ToSnowflakeID" "dateTimeToSnowflake" +"dateTimeToSnowflakeID" "dateTrunc" "date_diff" "decodeHTMLComponent" @@ -1032,6 +1048,21 @@ "deltaSumTimestampSimpleState" "deltaSumTimestampState" "demangle" +"denseRank" +"denseRankArgMax" +"denseRankArgMin" +"denseRankArray" +"denseRankDistinct" +"denseRankForEach" +"denseRankIf" +"denseRankMap" +"denseRankMerge" +"denseRankNull" +"denseRankOrDefault" +"denseRankOrNull" +"denseRankResample" +"denseRankSimpleState" +"denseRankState" "dense_rank" "dense_rankArgMax" "dense_rankArgMin" @@ -1108,8 +1139,11 @@ "domainWithoutWWWRFC" "dotProduct" "dumpColumnStructure" +"dynamicElement" +"dynamicType" "e" "editDistance" +"editDistanceUTF8" "empty" "emptyArrayDate" "emptyArrayDateTime" @@ -1334,14 +1368,17 @@ "gccMurmurHash" "gcd" "generateRandomStructure" +"generateSnowflakeID" "generateULID" "generateUUIDv4" +"generateUUIDv7" "geoDistance" "geoToH3" "geoToS2" "geohashDecode" "geohashEncode" "geohashesInBox" +"getClientHTTPHeader" "getMacro" "getOSKernelVersion" "getServerPort" @@ -1589,6 +1626,20 @@ "groupBitmapXorSimpleState" "groupBitmapXorState" "groupConcat" +"groupConcatArgMax" +"groupConcatArgMin" +"groupConcatArray" +"groupConcatDistinct" +"groupConcatForEach" +"groupConcatIf" +"groupConcatMap" +"groupConcatMerge" +"groupConcatNull" +"groupConcatOrDefault" +"groupConcatOrNull" +"groupConcatResample" +"groupConcatSimpleState" +"groupConcatState" "groupUniqArray" "groupUniqArrayArgMax" "groupUniqArrayArgMin" @@ -1604,6 +1655,21 @@ "groupUniqArrayResample" "groupUniqArraySimpleState" "groupUniqArrayState" +"group_concat" +"group_concatArgMax" +"group_concatArgMin" +"group_concatArray" +"group_concatDistinct" +"group_concatForEach" +"group_concatIf" +"group_concatMap" +"group_concatMerge" +"group_concatNull" +"group_concatOrDefault" +"group_concatOrNull" +"group_concatResample" +"group_concatSimpleState" +"group_concatState" "h3CellAreaM2" "h3CellAreaRads2" "h3Distance" @@ -1660,6 +1726,8 @@ "hasTokenCaseInsensitiveOrNull" "hasTokenOrNull" "hex" +"hilbertDecode" +"hilbertEncode" "histogram" "histogramArgMax" "histogramArgMin" @@ -1881,6 +1949,7 @@ "less" "lessOrEquals" "levenshteinDistance" +"levenshteinDistanceUTF8" "lgamma" "like" "ln" @@ -2498,10 +2567,44 @@ "parseDateTimeInJodaSyntaxOrZero" "parseDateTimeOrNull" "parseDateTimeOrZero" +"parseReadableSize" +"parseReadableSizeOrNull" +"parseReadableSizeOrZero" "parseTimeDelta" +"partitionID" "partitionId" "path" "pathFull" +"percentRank" +"percentRankArgMax" +"percentRankArgMin" +"percentRankArray" +"percentRankDistinct" +"percentRankForEach" +"percentRankIf" +"percentRankMap" +"percentRankMerge" +"percentRankNull" +"percentRankOrDefault" +"percentRankOrNull" +"percentRankResample" +"percentRankSimpleState" +"percentRankState" +"percent_rank" +"percent_rankArgMax" +"percent_rankArgMin" +"percent_rankArray" +"percent_rankDistinct" +"percent_rankForEach" +"percent_rankIf" +"percent_rankMap" +"percent_rankMerge" +"percent_rankNull" +"percent_rankOrDefault" +"percent_rankOrNull" +"percent_rankResample" +"percent_rankSimpleState" +"percent_rankState" "pi" "plus" "pmod" @@ -2533,6 +2636,7 @@ "positive_modulo" "pow" "power" +"printf" "proportionsZTest" "protocol" "punycodeDecode" @@ -3103,6 +3207,8 @@ "rankResample" "rankSimpleState" "rankState" +"readWKTLineString" +"readWKTMultiLineString" "readWKTMultiPolygon" "readWKTPoint" "readWKTPolygon" @@ -3340,6 +3446,8 @@ "skewSampState" "sleep" "sleepEachRow" +"snowflakeIDToDateTime" +"snowflakeIDToDateTime64" "snowflakeToDateTime" "snowflakeToDateTime64" "soundex" @@ -3902,6 +4010,7 @@ "truncate" "tryBase58Decode" "tryBase64Decode" +"tryBase64URLDecode" "tryDecrypt" "tryIdnaEncode" "tryPunycodeDecode" @@ -3923,6 +4032,7 @@ "tupleModuloByNumber" "tupleMultiply" "tupleMultiplyByNumber" +"tupleNames" "tupleNegate" "tuplePlus" "tupleToNameValuePairs" diff --git a/tests/fuzz/dictionaries/key_words.dict b/tests/fuzz/dictionaries/keywords.dict similarity index 95% rename from tests/fuzz/dictionaries/key_words.dict rename to tests/fuzz/dictionaries/keywords.dict index db517a2382c..abaaf9e53b5 100644 --- a/tests/fuzz/dictionaries/key_words.dict +++ b/tests/fuzz/dictionaries/keywords.dict @@ -3,7 +3,7 @@ "ADD CONSTRAINT" "ADD INDEX" "ADD PROJECTION" -"ADD STATISTIC" +"ADD STATISTICS" "ADMIN OPTION FOR" "AFTER" "ALGORITHM" @@ -76,7 +76,7 @@ "CLEAR COLUMN" "CLEAR INDEX" "CLEAR PROJECTION" -"CLEAR STATISTIC" +"CLEAR STATISTICS" "CLUSTER" "CLUSTERS" "CN" @@ -110,6 +110,8 @@ "CURRENTUSER" "CURRENT_USER" "D" +"DATA" +"DATA INNER UUID" "DATABASE" "DATABASES" "DATE" @@ -147,7 +149,7 @@ "DROP PART" "DROP PARTITION" "DROP PROJECTION" -"DROP STATISTIC" +"DROP STATISTICS" "DROP TABLE" "DROP TEMPORARY TABLE" "ELSE" @@ -247,6 +249,7 @@ "IS NULL" "IS_OBJECT_ID" "JOIN" +"JWT" "KERBEROS" "KEY" "KEY BY" @@ -277,13 +280,15 @@ "MATERIALIZE COLUMN" "MATERIALIZE INDEX" "MATERIALIZE PROJECTION" -"MATERIALIZE STATISTIC" +"MATERIALIZE STATISTICS" "MATERIALIZE TTL" "MATERIALIZED" "MAX" "MCS" "MEMORY" "MERGES" +"METRICS" +"METRICS INNER UUID" "MI" "MICROSECOND" "MICROSECONDS" @@ -297,12 +302,14 @@ "MODIFY" "MODIFY COLUMN" "MODIFY COMMENT" +"MODIFY DEFINER" "MODIFY ORDER BY" "MODIFY QUERY" "MODIFY REFRESH" "MODIFY SAMPLE BY" "MODIFY SETTING" "MODIFY SQL SECURITY" +"MODIFY STATISTICS" "MODIFY TTL" "MONTH" "MONTHS" @@ -373,6 +380,7 @@ "Protobuf" "Q" "QQ" +"QUALIFY" "QUARTER" "QUARTERS" "QUERY" @@ -384,6 +392,7 @@ "READONLY" "REALM" "RECOMPRESS" +"RECURSIVE" "REFERENCES" "REFRESH" "REGEXP" @@ -415,6 +424,7 @@ "SALT" "SAMPLE" "SAMPLE BY" +"SAN" "SCHEME" "SECOND" "SECONDS" @@ -460,7 +470,8 @@ "SS" "SSH_KEY" "SSL_CERTIFICATE" -"STATISTIC" +"START TRANSACTION" +"STATISTICS" "STEP" "STORAGE" "STRICT" @@ -475,6 +486,8 @@ "TABLE" "TABLE OVERRIDE" "TABLES" +"TAGS" +"TAGS INNER UUID" "TEMPORARY" "TEMPORARY TABLE" "TEST" @@ -529,6 +542,7 @@ "WITH NAME" "WITH REPLACE OPTION" "WITH TIES" +"WITH_ITEMINDEX" "WK" "WRITABLE" "WW" @@ -540,4 +554,3 @@ "bagexpansion" "base_backup" "cluster_host_ids" -"with_itemindex" diff --git a/utils/libfuzzer/runner.py b/tests/fuzz/runner.py similarity index 75% rename from utils/libfuzzer/runner.py rename to tests/fuzz/runner.py index bbe648dbbc2..44259228f60 100644 --- a/utils/libfuzzer/runner.py +++ b/tests/fuzz/runner.py @@ -11,7 +11,7 @@ FUZZER_ARGS = os.getenv("FUZZER_ARGS", "") def run_fuzzer(fuzzer: str): - logging.info(f"Running fuzzer {fuzzer}...") + logging.info("Running fuzzer %s...", fuzzer) corpus_dir = f"{fuzzer}.in" with Path(corpus_dir) as path: @@ -29,28 +29,27 @@ def run_fuzzer(fuzzer: str): 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())}" + f"{os.environ['ASAN_OPTIONS']}:{':'.join(f'{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())}" + f"{os.environ['MSAN_OPTIONS']}:{':'.join(f'{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())}" + f"{os.environ['UBSAN_OPTIONS']}:{':'.join(f'{key}={value}' for key, value in parser['ubsan'].items())}" ) if parser.has_section("libfuzzer"): custom_libfuzzer_options = " ".join( - "-%s=%s" % (key, value) - for key, value in parser["libfuzzer"].items() + f"-{key}={value}" 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)) + (f"{key}") if value == "" else (f"{key}={value}") for key, value in parser["fuzzer_arguments"].items() ) @@ -65,7 +64,7 @@ def run_fuzzer(fuzzer: str): cmd_line += " < /dev/null" - logging.info(f"...will execute: {cmd_line}") + logging.info("...will execute: %s", cmd_line) subprocess.check_call(cmd_line, shell=True) diff --git a/tests/fuzz/tcp_protocol_fuzzer.options b/tests/fuzz/tcp_protocol_fuzzer.options new file mode 100644 index 00000000000..4885669d91d --- /dev/null +++ b/tests/fuzz/tcp_protocol_fuzzer.options @@ -0,0 +1,4 @@ +[fuzzer_arguments] +--log-file=tcp_protocol_fuzzer.log +--= +--logging.terminal=0 diff --git a/tests/fuzz/update_dict.sh b/tests/fuzz/update_dict.sh new file mode 100755 index 00000000000..a83c9167129 --- /dev/null +++ b/tests/fuzz/update_dict.sh @@ -0,0 +1,20 @@ +#!/bin/bash + +set -euo pipefail + +SCRIPT_DIR=$(dirname "$(realpath "$0")") +ROOT_PATH="$(git rev-parse --show-toplevel)" +CLICKHOUSE_BIN="${CLICKHOUSE_BIN:-$ROOT_PATH/build/programs/clickhouse}" +DICTIONARIES_DIR="$SCRIPT_DIR/dictionaries" + +echo "Generating functions dict" +$CLICKHOUSE_BIN local -q "SELECT * FROM (SELECT DISTINCT concat('\"', name, '\"') as res FROM system.functions ORDER BY name UNION ALL SELECT concat('\"', a.name, b.name, '\"') as res FROM system.functions as a CROSS JOIN system.aggregate_function_combinators as b WHERE a.is_aggregate = 1) ORDER BY res" > "$DICTIONARIES_DIR/functions.dict" + +echo "Generating data types dict" +$CLICKHOUSE_BIN local -q "SELECT DISTINCT concat('\"', name, '\"') as res FROM system.data_type_families ORDER BY name" > "$DICTIONARIES_DIR/datatypes.dict" + +echo "Generating keywords dict" +$CLICKHOUSE_BIN local -q "SELECT DISTINCT concat('\"', keyword, '\"') as res FROM system.keywords ORDER BY keyword" > "$DICTIONARIES_DIR/keywords.dict" + +echo "Merging dictionaries into all.dict" +cat "$DICTIONARIES_DIR"/* | LC_ALL=C sort | uniq > "$SCRIPT_DIR/all.dict" \ No newline at end of file diff --git a/tests/instructions/easy_tasks_sorted_ru.md b/tests/instructions/easy_tasks_sorted_ru.md index fbd86ebf08f..e0607126ecc 100644 --- a/tests/instructions/easy_tasks_sorted_ru.md +++ b/tests/instructions/easy_tasks_sorted_ru.md @@ -97,10 +97,6 @@ Upd: сŠ“ŠµŠ»Š°Š»Šø ŠæŠ¾-Š“руŠ³Š¾Š¼Ńƒ: тŠµŠæŠµŃ€ŃŒ Š²ŃŃ‘ Š±ŠµŠ·Š¾ŠæŠ°ŃŠ½Š¾. Š’Š¾Š·Š²Ń€Š°Ń‰Š°ŠµŃ‚ ŠøŠ½ŠŗрŠµŠ¼ŠµŠ½Ń‚Š°Š»ŃŒŠ½Š¾Šµ чŠøсŠ»Š¾ Š“Š»Ń ŠæŠ¾Š²Ń‚Š¾Ń€Š½Š¾ Š²ŃŃ‚Ń€ŠµŃ‡Š°ŃŽŃ‰Šøхся Š·Š½Š°Ń‡ŠµŠ½ŠøŠ¹ key. -## ŠŠ³Ń€ŠµŠ³Š°Ń‚Š½Š°Ń фуŠ½ŠŗцŠøя groupConcat. - -`groupConcat(x, ',')` - сŠ¾Š±Ń€Š°Ń‚ŃŒ ŠøŠ· ŠæŠµŃ€ŠµŠ“Š°Š½Š½Ń‹Ń… Š·Š½Š°Ń‡ŠµŠ½ŠøŠ¹ x стрŠ¾Šŗу, рŠ°Š·Š“ŠµŠ»Ń‘Š½Š½ŃƒŃŽ Š·Š°ŠæятыŠ¼Šø. - ## Š¤ŃƒŠ½ŠŗцŠøŠø DATE_ADD, DATE_SUB ŠŗŠ°Šŗ сŠøŠ½Š¾Š½ŠøŠ¼Ń‹ Š“Š»Ń сŠ¾Š²Š¼ŠµŃŃ‚ŠøŠ¼Š¾ŃŃ‚Šø с SQL. https://dev.mysql.com/doc/refman/8.0/en/date-and-time-functions.html#function_date-add diff --git a/tests/integration/compose/docker_compose_hdfs.yml b/tests/integration/compose/docker_compose_hdfs.yml index 1cae54ad9e1..40a10df01f7 100644 --- a/tests/integration/compose/docker_compose_hdfs.yml +++ b/tests/integration/compose/docker_compose_hdfs.yml @@ -1,7 +1,7 @@ version: '2.3' services: hdfs1: - image: sequenceiq/hadoop-docker:2.7.0 + image: prasanthj/docker-hadoop:2.6.0 hostname: hdfs1 restart: always expose: diff --git a/tests/integration/compose/docker_compose_prometheus.yml b/tests/integration/compose/docker_compose_prometheus.yml new file mode 100644 index 00000000000..0a1db2138ba --- /dev/null +++ b/tests/integration/compose/docker_compose_prometheus.yml @@ -0,0 +1,57 @@ +version: '2.3' +services: + prometheus_writer: + image: prom/prometheus:v2.50.1 + hostname: ${PROMETHEUS_WRITER_HOSTNAME:-prometheus_writer} + restart: always + entrypoint: | + /bin/sh -c 'truncate -s 0 /etc/prometheus/prometheus.yml + cat << EOF >> /etc/prometheus/prometheus.yml + global: + scrape_interval: 1s + scrape_configs: + - job_name: "prometheus" + static_configs: + - targets: ["localhost:${PROMETHEUS_WRITER_PORT}"] + EOF + if [ -n "${PROMETHEUS_REMOTE_WRITE_HANDLER}" ]; then + echo "remote_write:" >> /etc/prometheus/prometheus.yml + echo " - url: \"${PROMETHEUS_REMOTE_WRITE_HANDLER}\"" >> /etc/prometheus/prometheus.yml + fi + #cat "/etc/prometheus/prometheus.yml" + /bin/prometheus --config.file="/etc/prometheus/prometheus.yml" --storage.tsdb.path="/prometheus" --web.console.libraries="/usr/share/prometheus/console_libraries" --web.console.templates="/usr/share/prometheus/consoles" --web.listen-address="0.0.0.0:${PROMETHEUS_WRITER_PORT}" &> /var/log/prometheus/prometheus.log' + expose: + - ${PROMETHEUS_WRITER_PORT} + healthcheck: + test: curl -f "ttps://localhost:${PROMETHEUS_WRITER_PORT}/api/v1/status/runtimeinfo" || exit 1 + interval: 5s + timeout: 3s + retries: 30 + volumes: + - type: ${PROMETHEUS_WRITER_LOGS_FS:-tmpfs} + source: ${PROMETHEUS_WRITER_LOGS:-} + target: /var/log/prometheus + + prometheus_reader: + image: prom/prometheus:v2.50.1 + hostname: ${PROMETHEUS_READER_HOSTNAME:-prometheus_reader} + restart: always + entrypoint: | + /bin/sh -c 'truncate -s 0 /etc/prometheus/prometheus.yml + if [ -n "${PROMETHEUS_REMOTE_READ_HANDLER}" ]; then + echo "remote_read:" >> /etc/prometheus/prometheus.yml + echo " - url: \"${PROMETHEUS_REMOTE_READ_HANDLER}\"" >> /etc/prometheus/prometheus.yml + fi + #cat "/etc/prometheus/prometheus.yml" + /bin/prometheus --config.file="/etc/prometheus/prometheus.yml" --storage.tsdb.path="/prometheus" --web.console.libraries="/usr/share/prometheus/console_libraries" --web.console.templates="/usr/share/prometheus/consoles" --web.listen-address="0.0.0.0:${PROMETHEUS_READER_PORT}" &> /var/log/prometheus/prometheus.log' + expose: + - ${PROMETHEUS_READER_PORT} + healthcheck: + test: curl -f "ttps://localhost:${PROMETHEUS_READER_PORT}/api/v1/status/runtimeinfo" || exit 1 + interval: 5s + timeout: 3s + retries: 30 + volumes: + - type: ${PROMETHEUS_READER_LOGS_FS:-tmpfs} + source: ${PROMETHEUS_READER_LOGS:-} + target: /var/log/prometheus diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index 673fc07fe94..53f4f1e1f26 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -570,6 +570,8 @@ class ClickHouseCluster: self.spark_session = None self.with_azurite = False + self.azurite_container = "azurite-container" + self.blob_service_client = None self._azurite_port = 0 # available when with_hdfs == True @@ -737,6 +739,25 @@ class ClickHouseCluster: self.jdbc_driver_dir = p.abspath(p.join(self.instances_dir, "jdbc_driver")) self.jdbc_driver_logs_dir = os.path.join(self.jdbc_driver_dir, "logs") + # available when with_prometheus == True + self.with_prometheus = False + self.prometheus_writer_host = "prometheus_writer" + self.prometheus_writer_port = 9090 + self.prometheus_writer_logs_dir = p.abspath( + p.join(self.instances_dir, "prometheus_writer/logs") + ) + self.prometheus_reader_host = "prometheus_reader" + self.prometheus_reader_port = 9091 + self.prometheus_reader_logs_dir = p.abspath( + p.join(self.instances_dir, "prometheus_reader/logs") + ) + self.prometheus_remote_write_handler_host = None + self.prometheus_remote_write_handler_port = 9092 + self.prometheus_remote_write_handler_path = "/write" + self.prometheus_remote_read_handler_host = None + self.prometheus_remote_read_handler_port = 9092 + self.prometheus_remote_read_handler_path = "/read" + self.docker_client = None self.is_up = False self.env = os.environ.copy() @@ -1619,6 +1640,42 @@ class ClickHouseCluster: ] return self.base_hive_cmd + def setup_prometheus_cmd(self, instance, env_variables, docker_compose_yml_dir): + env_variables["PROMETHEUS_WRITER_HOST"] = self.prometheus_writer_host + env_variables["PROMETHEUS_WRITER_PORT"] = str(self.prometheus_writer_port) + env_variables["PROMETHEUS_WRITER_LOGS"] = self.prometheus_writer_logs_dir + env_variables["PROMETHEUS_WRITER_LOGS_FS"] = "bind" + env_variables["PROMETHEUS_READER_HOST"] = self.prometheus_reader_host + env_variables["PROMETHEUS_READER_PORT"] = str(self.prometheus_reader_port) + env_variables["PROMETHEUS_READER_LOGS"] = self.prometheus_reader_logs_dir + env_variables["PROMETHEUS_READER_LOGS_FS"] = "bind" + if self.prometheus_remote_write_handler_host: + env_variables["PROMETHEUS_REMOTE_WRITE_HANDLER"] = ( + f"http://{self.prometheus_remote_write_handler_host}:{self.prometheus_remote_write_handler_port}/{self.prometheus_remote_write_handler_path.strip('/')}" + ) + if self.prometheus_remote_read_handler_host: + env_variables["PROMETHEUS_REMOTE_READ_HANDLER"] = ( + f"http://{self.prometheus_remote_read_handler_host}:{self.prometheus_remote_read_handler_port}/{self.prometheus_remote_read_handler_path.strip('/')}" + ) + if not self.with_prometheus: + self.with_prometheus = True + self.base_cmd.extend( + [ + "--file", + p.join(docker_compose_yml_dir, "docker_compose_prometheus.yml"), + ] + ) + self.base_prometheus_cmd = [ + "docker-compose", + "--env-file", + instance.env_file, + "--project-name", + self.project_name, + "--file", + p.join(docker_compose_yml_dir, "docker_compose_prometheus.yml"), + ] + return self.base_prometheus_cmd + def add_instance( self, name, @@ -1659,6 +1716,9 @@ class ClickHouseCluster: with_jdbc_bridge=False, with_hive=False, with_coredns=False, + with_prometheus=False, + handle_prometheus_remote_write=False, + handle_prometheus_remote_read=False, use_old_analyzer=None, hostname=None, env_variables=None, @@ -2001,6 +2061,17 @@ class ClickHouseCluster: self.setup_hive(instance, env_variables, docker_compose_yml_dir) ) + if with_prometheus: + if handle_prometheus_remote_write: + self.prometheus_remote_write_handler_host = instance.hostname + if handle_prometheus_remote_read: + self.prometheus_remote_read_handler_host = instance.hostname + cmds.append( + self.setup_prometheus_cmd( + instance, env_variables, docker_compose_yml_dir + ) + ) + logging.debug( "Cluster name:{} project_name:{}. Added instance name:{} tag:{} base_cmd:{} docker_compose_yml_dir:{}".format( self.name, @@ -2623,6 +2694,32 @@ class ClickHouseCluster: connection_string ) logging.debug(blob_service_client.get_account_information()) + containers = [ + c + for c in blob_service_client.list_containers( + name_starts_with=self.azurite_container + ) + if c.name == self.azurite_container + ] + if len(containers) > 0: + for c in containers: + blob_service_client.delete_container(c) + + container_client = blob_service_client.get_container_client( + self.azurite_container + ) + if container_client.exists(): + logging.debug( + f"azurite container '{self.azurite_container}' exist, deleting all blobs" + ) + for b in container_client.list_blobs(): + container_client.delete_blob(b.name) + else: + logging.debug( + f"azurite container '{self.azurite_container}' doesn't exist, creating it" + ) + container_client.create_container() + self.blob_service_client = blob_service_client return except Exception as ex: @@ -3064,6 +3161,12 @@ class ClickHouseCluster: f"http://{self.jdbc_bridge_ip}:{self.jdbc_bridge_port}/ping" ) + if self.with_prometheus: + os.makedirs(self.prometheus_writer_logs_dir) + os.chmod(self.prometheus_writer_logs_dir, stat.S_IRWXU | stat.S_IRWXO) + os.makedirs(self.prometheus_reader_logs_dir) + os.chmod(self.prometheus_reader_logs_dir, stat.S_IRWXU | stat.S_IRWXO) + clickhouse_start_cmd = self.base_cmd + ["up", "-d", "--no-recreate"] logging.debug( ( @@ -4484,7 +4587,7 @@ class ClickHouseInstance: use_old_analyzer = os.environ.get("CLICKHOUSE_USE_OLD_ANALYZER") is not None # If specific version was used there can be no - # allow_experimental_analyzer setting, so do this only if it was + # enable_analyzer setting, so do this only if it was # explicitly requested. if self.tag: use_old_analyzer = False diff --git a/tests/integration/parallel_skip.json b/tests/integration/parallel_skip.json index 99fa626bd1e..fca2126d824 100644 --- a/tests/integration/parallel_skip.json +++ b/tests/integration/parallel_skip.json @@ -162,9 +162,13 @@ "test_storage_kafka/test.py::test_system_kafka_consumers_rebalance_mv", "test_storage_kafka/test.py::test_formats_errors", "test_storage_kafka/test.py::test_multiple_read_in_materialized_views", + "test_storage_kafka/test.py::test_kafka_null_message", + + "test_storage_kafka/test_produce_http_interface.py::test_kafka_produce_http_interface_row_based_format", "test_storage_kerberized_kafka/test.py::test_kafka_json_as_string", "test_storage_kerberized_kafka/test.py::test_kafka_json_as_string_request_new_ticket_after_expiration", "test_storage_kerberized_kafka/test.py::test_kafka_json_as_string_no_kdc", "test_storage_kerberized_kafka/test.py::test_kafka_config_from_sql_named_collection" + ] diff --git a/tests/integration/test_access_control_on_cluster/test.py b/tests/integration/test_access_control_on_cluster/test.py index b12add7ad3f..c292d0cc3a4 100644 --- a/tests/integration/test_access_control_on_cluster/test.py +++ b/tests/integration/test_access_control_on_cluster/test.py @@ -89,3 +89,5 @@ def test_grant_current_database_on_cluster(): assert ch1.query("SHOW DATABASES", user="test_user") == "user_db\n" ch1.query("GRANT SELECT ON * TO test_user ON CLUSTER 'cluster'", user="test_user") assert ch1.query("SHOW DATABASES", user="test_user") == "user_db\n" + ch1.query("DROP DATABASE user_db ON CLUSTER 'cluster'") + ch1.query("DROP USER test_user ON CLUSTER 'cluster'") diff --git a/tests/integration/test_access_for_functions/test.py b/tests/integration/test_access_for_functions/test.py index 004d39e1dea..82125b35b49 100644 --- a/tests/integration/test_access_for_functions/test.py +++ b/tests/integration/test_access_for_functions/test.py @@ -65,8 +65,8 @@ def test_ignore_obsolete_grant_on_database(): "-c", f""" cat > /var/lib/clickhouse/access/{user_id}.sql << EOF -ATTACH USER X; -ATTACH GRANT CREATE FUNCTION, SELECT ON mydb.* TO X; +ATTACH USER \`{user_id}\`; +ATTACH GRANT CREATE FUNCTION, SELECT ON mydb.* TO \`{user_id}\`; EOF""", ] ) @@ -76,4 +76,7 @@ EOF""", ) instance.start_clickhouse() - assert instance.query("SHOW GRANTS FOR X") == "GRANT SELECT ON mydb.* TO X\n" + assert ( + instance.query(f"SHOW GRANTS FOR `{user_id}`") + == f"GRANT SELECT ON mydb.* TO `{user_id}`\n" + ) diff --git a/tests/integration/test_alter_comment_on_cluster/test.py b/tests/integration/test_alter_comment_on_cluster/test.py index e6767e35c1b..4cb10bbc751 100644 --- a/tests/integration/test_alter_comment_on_cluster/test.py +++ b/tests/integration/test_alter_comment_on_cluster/test.py @@ -59,3 +59,4 @@ def test_comment(started_cluster): expected = "CREATE TABLE default.test_table (`id` Int64 COMMENT \\'column_comment_2\\') ENGINE = ReplicatedMergeTree(\\'/clickhouse/tables/{uuid}/{shard}\\', \\'{replica}\\') ORDER BY id SETTINGS index_granularity = 8192 COMMENT \\'table_comment_2\\'" assert_create_query([node_1, node_2], "default", "test_table", expected) + node_1.query("DROP TABLE test_table ON CLUSTER 'cluster'") diff --git a/tests/integration/test_alter_moving_garbage/test.py b/tests/integration/test_alter_moving_garbage/test.py index 21be46a7e1b..76b40b0df8d 100644 --- a/tests/integration/test_alter_moving_garbage/test.py +++ b/tests/integration/test_alter_moving_garbage/test.py @@ -36,6 +36,16 @@ def cluster(): cluster.shutdown() +def drop_table(node, table_name, replicated): + + create_table_statement = f"DROP TABLE {table_name} SYNC" + + if replicated: + node.query_with_retry(create_table_statement) + else: + node.query(create_table_statement) + + def create_table(node, table_name, replicated, additional_settings): settings = { "storage_policy": "two_disks", @@ -158,6 +168,9 @@ def test_alter_moving( assert data_digest == "1000\n" + for node in nodes: + drop_table(node, table_name, replicated_engine) + def test_delete_race_leftovers(cluster): """ @@ -248,3 +261,4 @@ def test_delete_race_leftovers(cluster): # Check that we have all data assert table_digest == node.query(table_digest_query) + drop_table(node, table_name, replicated=True) diff --git a/tests/integration/test_alter_on_mixed_type_cluster/test.py b/tests/integration/test_alter_on_mixed_type_cluster/test.py index f21a97d40e1..da126e307a0 100644 --- a/tests/integration/test_alter_on_mixed_type_cluster/test.py +++ b/tests/integration/test_alter_on_mixed_type_cluster/test.py @@ -88,6 +88,9 @@ def test_alter_on_cluter_non_replicated(started_cluster): assert node3.query("SELECT COUNT() FROM test_table") == "2\n" assert node4.query("SELECT COUNT() FROM test_table") == "2\n" + for node in [node1, node2, node3, node4]: + node.query("TRUNCATE TABLE test_table") + def test_alter_replicated_on_cluster(started_cluster): for node in [node1, node3]: @@ -133,3 +136,6 @@ def test_alter_replicated_on_cluster(started_cluster): assert node2.query("SELECT COUNT() FROM test_table_replicated") == "2\n" assert node3.query("SELECT COUNT() FROM test_table_replicated") == "2\n" assert node4.query("SELECT COUNT() FROM test_table_replicated") == "2\n" + + for node in [node1, node2, node3, node4]: + node.query("TRUNCATE TABLE test_table_replicated") diff --git a/tests/integration/test_analyzer_compatibility/test.py b/tests/integration/test_analyzer_compatibility/test.py index d4ded420c61..505d1629cd2 100644 --- a/tests/integration/test_analyzer_compatibility/test.py +++ b/tests/integration/test_analyzer_compatibility/test.py @@ -1,4 +1,5 @@ import uuid +import time import pytest from helpers.cluster import ClickHouseCluster @@ -58,12 +59,12 @@ ORDER BY h;""" == TSV([["backward", "true"], ["current", "true"]]) ) - # Should be enabled everywhere - analyzer_enabled = current.query( + # Should be enabled explicitly on the old instance. + analyzer_enabled = backward.query( f""" SELECT DISTINCT Settings['allow_experimental_analyzer'] -FROM clusterAllReplicas('test_cluster_mixed', system.query_log) +FROM system.query_log WHERE initial_query_id = '{query_id}';""" ) @@ -78,6 +79,8 @@ WHERE initial_query_id = '{query_id}';""" current.query("SYSTEM FLUSH LOGS") backward.query("SYSTEM FLUSH LOGS") + # The old version doesn't know about the alias. + # For this we will ask about the old experimental name. assert ( backward.query( """ @@ -98,3 +101,26 @@ WHERE initial_query_id = '{query_id}';""" ) assert TSV(analyzer_enabled) == TSV("0") + + # Only new version knows about the alias + # and it will send the old setting `allow_experimental_analyzer` + # to the remote server. + query_id = str(uuid.uuid4()) + current.query( + "SELECT * FROM clusterAllReplicas('test_cluster_mixed', system.tables) SETTINGS enable_analyzer = 1;", + query_id=query_id, + ) + + current.query("SYSTEM FLUSH LOGS") + backward.query("SYSTEM FLUSH LOGS") + + # Should be disabled explicitly everywhere. + analyzer_enabled = current.query( + f""" +SELECT +DISTINCT Settings['allow_experimental_analyzer'] +FROM system.query_log +WHERE initial_query_id = '{query_id}';""" + ) + + assert TSV(analyzer_enabled) == TSV("1") diff --git a/tests/queries/0_stateless/02597_column_update_tricy_expression_and_replication.reference b/tests/integration/test_async_metrics_in_cgroup/__init__.py similarity index 100% rename from tests/queries/0_stateless/02597_column_update_tricy_expression_and_replication.reference rename to tests/integration/test_async_metrics_in_cgroup/__init__.py diff --git a/tests/integration/test_async_metrics_in_cgroup/test.py b/tests/integration/test_async_metrics_in_cgroup/test.py new file mode 100644 index 00000000000..d9f2e3aaaed --- /dev/null +++ b/tests/integration/test_async_metrics_in_cgroup/test.py @@ -0,0 +1,69 @@ +import pytest + +from helpers.cluster import ClickHouseCluster + +cluster = ClickHouseCluster(__file__) +node1 = cluster.add_instance("node1", stay_alive=True) +node2 = cluster.add_instance("node2", stay_alive=True) + + +@pytest.fixture(scope="module") +def start_cluster(): + try: + cluster.start() + yield cluster + finally: + cluster.shutdown() + + +def run_cpu_intensive_task(node): + node.query( + "SELECT sum(*) FROM system.numbers_mt FORMAT Null SETTINGS max_execution_time=10", + ignore_error=True, + ) + + +def get_async_metric(node, metric): + node.query("SYSTEM FLUSH LOGS") + return node.query( + f""" + SELECT max(value) + FROM ( + SELECT toStartOfInterval(event_time, toIntervalSecond(1)) AS t, avg(value) AS value + FROM system.asynchronous_metric_log + WHERE event_time >= now() - 60 AND metric = '{metric}' + GROUP BY t + ) + SETTINGS max_threads = 1 + """ + ).strip("\n") + + +def test_user_cpu_accounting(start_cluster): + if node1.is_built_with_sanitizer(): + pytest.skip("Disabled for sanitizers") + + # run query on the other node, its usage shouldn't be accounted by node1 + run_cpu_intensive_task(node2) + + node1_cpu_time = get_async_metric(node1, "OSUserTime") + assert float(node1_cpu_time) < 2 + + # then let's test that we will account cpu time spent by the server itself + node2_cpu_time = get_async_metric(node2, "OSUserTime") + # this check is really weak, but CI is tough place and we cannot guarantee that test process will get many cpu time + assert float(node2_cpu_time) > 2 + + +def test_normalized_user_cpu(start_cluster): + if node1.is_built_with_sanitizer(): + pytest.skip("Disabled for sanitizers") + + # run query on the other node, its usage shouldn't be accounted by node1 + run_cpu_intensive_task(node2) + + node1_cpu_time = get_async_metric(node1, "OSUserTimeNormalized") + assert float(node1_cpu_time) < 1.01 + + node2_cpu_time = get_async_metric(node2, "OSUserTimeNormalized") + assert float(node2_cpu_time) < 1.01 diff --git a/tests/integration/test_broken_projections/config.d/dont_start_broken.xml b/tests/integration/test_broken_projections/config.d/dont_start_broken.xml new file mode 100644 index 00000000000..9603cdc7e3e --- /dev/null +++ b/tests/integration/test_broken_projections/config.d/dont_start_broken.xml @@ -0,0 +1,6 @@ + + + + 0 + + diff --git a/tests/integration/test_broken_projections/test.py b/tests/integration/test_broken_projections/test.py index 162c0dbaa2f..578ff42369c 100644 --- a/tests/integration/test_broken_projections/test.py +++ b/tests/integration/test_broken_projections/test.py @@ -4,6 +4,7 @@ import logging import string import random from helpers.cluster import ClickHouseCluster +from multiprocessing.dummy import Pool cluster = ClickHouseCluster(__file__) @@ -18,6 +19,12 @@ def cluster(): stay_alive=True, with_zookeeper=True, ) + cluster.add_instance( + "node_restart", + main_configs=["config.d/dont_start_broken.xml"], + stay_alive=True, + with_zookeeper=True, + ) logging.info("Starting cluster...") cluster.start() @@ -632,6 +639,49 @@ def test_broken_on_start(cluster): check(node, table_name, 0) +def test_disappeared_projection_on_start(cluster): + node = cluster.instances["node_restart"] + + table_name = "test_disapperead_projection" + create_table(node, table_name, 1) + + node.query(f"SYSTEM STOP MERGES {table_name}") + + insert(node, table_name, 0, 5) + insert(node, table_name, 5, 5) + insert(node, table_name, 10, 5) + insert(node, table_name, 15, 5) + + assert ["all_0_0_0", "all_1_1_0", "all_2_2_0", "all_3_3_0"] == get_parts( + node, table_name + ) + + def drop_projection(): + node.query( + f"ALTER TABLE {table_name} DROP PROJECTION proj2", + settings={"mutations_sync": "0"}, + ) + + p = Pool(2) + p.apply_async(drop_projection) + + for i in range(30): + create_query = node.query(f"SHOW CREATE TABLE {table_name}") + if "proj2" not in create_query: + break + time.sleep(0.5) + + assert "proj2" not in create_query + + # Remove 'proj2' for part all_2_2_0 + break_projection(node, table_name, "proj2", "all_2_2_0", "part") + + node.restart_clickhouse() + + # proj2 is not broken, it doesn't exist, but ok + check(node, table_name, 0, expect_broken_part="proj2", do_check_command=0) + + def test_mutation_with_broken_projection(cluster): node = cluster.instances["node"] diff --git a/tests/integration/test_cgroup_limit/test.py b/tests/integration/test_cgroup_limit/test.py index e77b0f70960..5d56135d9ff 100644 --- a/tests/integration/test_cgroup_limit/test.py +++ b/tests/integration/test_cgroup_limit/test.py @@ -46,7 +46,7 @@ def test_cgroup_cpu_limit(): "clickhouse local -q \"select value from system.settings where name='max_threads'\"", num_cpus, ) - expect_output = (r"\'auto({})\'".format(math.ceil(num_cpus))).encode() + expect_output = (r"auto({})".format(math.ceil(num_cpus))).encode() assert ( result.strip() == expect_output ), f"fail for cpu limit={num_cpus}, result={result.strip()}, expect={expect_output}" diff --git a/tests/integration/test_checking_s3_blobs_paranoid/test.py b/tests/integration/test_checking_s3_blobs_paranoid/test.py index b995b4d6461..76a0f30f82e 100644 --- a/tests/integration/test_checking_s3_blobs_paranoid/test.py +++ b/tests/integration/test_checking_s3_blobs_paranoid/test.py @@ -209,7 +209,7 @@ def test_upload_s3_fail_upload_part_when_multi_part_upload( ) assert create_multipart == 1 assert upload_parts >= 2 - assert s3_errors >= 2 + assert s3_errors == 1 @pytest.mark.parametrize( @@ -708,7 +708,7 @@ def test_no_key_found_disk(cluster, broken_s3): """ SELECT value FROM system.metrics - WHERE metric = 'S3DiskNoKeyErrors' + WHERE metric = 'DiskS3NoSuchKeyErrors' """ ).strip() ) diff --git a/tests/integration/test_cluster_all_replicas/test.py b/tests/integration/test_cluster_all_replicas/test.py index d8bad180e1b..9797db7c498 100644 --- a/tests/integration/test_cluster_all_replicas/test.py +++ b/tests/integration/test_cluster_all_replicas/test.py @@ -21,14 +21,14 @@ def start_cluster(): def test_cluster(start_cluster): assert ( node1.query( - "SELECT hostName() FROM clusterAllReplicas('one_shard_two_nodes', system.one)" + "SELECT hostName() FROM clusterAllReplicas('one_shard_two_nodes', system.one) ORDER BY ALL" ) == "node1\nnode2\n" ) assert set( node1.query( - """SELECT hostName(), * FROM clusterAllReplicas("one_shard_two_nodes", system.one) ORDER BY dummy""" + """SELECT hostName(), * FROM clusterAllReplicas("one_shard_two_nodes", system.one) ORDER BY ALL""" ).splitlines() ) == {"node1\t0", "node2\t0"} @@ -48,7 +48,7 @@ def test_global_in(start_cluster): assert set( node1.query( - """SELECT hostName(), * FROM clusterAllReplicas("one_shard_two_nodes", system.one) where dummy GLOBAL IN u""" + """SELECT hostName(), * FROM clusterAllReplicas("one_shard_two_nodes", system.one) where dummy GLOBAL IN u ORDER BY ALL""" ).splitlines() ) == {"node1\t0", "node2\t0"} @@ -63,7 +63,7 @@ def test_global_in(start_cluster): def test_skip_unavailable_replica(start_cluster, cluster): assert ( node1.query( - f"SELECT hostName() FROM clusterAllReplicas('{cluster}', system.one) settings skip_unavailable_shards=1" + f"SELECT hostName() FROM clusterAllReplicas('{cluster}', system.one) ORDER BY ALL settings skip_unavailable_shards=1" ) == "node1\nnode2\n" ) @@ -81,5 +81,5 @@ def test_error_on_unavailable_replica(start_cluster, cluster): # so when skip_unavailable_shards=0 - any unavailable replica should lead to an error with pytest.raises(QueryRuntimeException): node1.query( - f"SELECT hostName() FROM clusterAllReplicas('{cluster}', system.one) settings skip_unavailable_shards=0" + f"SELECT hostName() FROM clusterAllReplicas('{cluster}', system.one) ORDER BY ALL settings skip_unavailable_shards=0" ) diff --git a/tests/integration/test_delayed_replica_failover/test.py b/tests/integration/test_delayed_replica_failover/test.py index a480ee3f278..f1034e26b25 100644 --- a/tests/integration/test_delayed_replica_failover/test.py +++ b/tests/integration/test_delayed_replica_failover/test.py @@ -20,21 +20,30 @@ node_1_2 = cluster.add_instance("node_1_2", with_zookeeper=True) node_2_1 = cluster.add_instance("node_2_1", with_zookeeper=True) node_2_2 = cluster.add_instance("node_2_2", with_zookeeper=True) +# For test to be runnable multiple times +seqno = 0 + @pytest.fixture(scope="module") def started_cluster(): try: cluster.start() + yield cluster + finally: + cluster.shutdown() + +@pytest.fixture(scope="function", autouse=True) +def create_tables(): + global seqno + try: + seqno += 1 for shard in (1, 2): for replica in (1, 2): node = cluster.instances["node_{}_{}".format(shard, replica)] node.query( - """ -CREATE TABLE replicated (d Date, x UInt32) ENGINE = - ReplicatedMergeTree('/clickhouse/tables/{shard}/replicated', '{instance}') PARTITION BY toYYYYMM(d) ORDER BY d""".format( - shard=shard, instance=node.name - ) + f"CREATE TABLE replicated (d Date, x UInt32) ENGINE = " + f"ReplicatedMergeTree('/clickhouse/tables/{shard}/replicated_{seqno}', '{node.name}') PARTITION BY toYYYYMM(d) ORDER BY d" ) node_1_1.query( @@ -42,10 +51,15 @@ CREATE TABLE replicated (d Date, x UInt32) ENGINE = "Distributed('test_cluster', 'default', 'replicated')" ) - yield cluster + yield finally: - cluster.shutdown() + node_1_1.query("DROP TABLE distributed") + + node_1_1.query("DROP TABLE replicated") + node_1_2.query("DROP TABLE replicated") + node_2_1.query("DROP TABLE replicated") + node_2_2.query("DROP TABLE replicated") def test(started_cluster): @@ -101,7 +115,9 @@ SELECT sum(x) FROM distributed WITH TOTALS SETTINGS # allow pings to zookeeper to timeout (must be greater than ZK session timeout). for _ in range(30): try: - node_2_2.query("SELECT * FROM system.zookeeper where path = '/'") + node_2_2.query( + "SELECT * FROM system.zookeeper where path = '/' SETTINGS insert_keeper_max_retries = 0" + ) time.sleep(0.5) except: break @@ -120,7 +136,7 @@ SELECT sum(x) FROM distributed SETTINGS == "3" ) - # Regression for skip_unavailable_shards in conjunction with skip_unavailable_shards + # Prefer fallback_to_stale_replicas over skip_unavailable_shards assert ( instance_with_dist_table.query( """ diff --git a/tests/integration/test_dictionaries_update_and_reload/test.py b/tests/integration/test_dictionaries_update_and_reload/test.py index db1c8e47467..3ed854d2c9f 100644 --- a/tests/integration/test_dictionaries_update_and_reload/test.py +++ b/tests/integration/test_dictionaries_update_and_reload/test.py @@ -37,16 +37,6 @@ def get_status(dictionary_name): ).rstrip("\n") -def get_status_retry(dictionary_name, expect, retry_count=10, sleep_time=0.5): - for _ in range(retry_count): - res = get_status(dictionary_name) - if res == expect: - return res - time.sleep(sleep_time) - - raise Exception(f'Expected result "{expect}" did not occur') - - def get_last_exception(dictionary_name): return ( instance.query( @@ -263,13 +253,7 @@ def test_reload_after_fail_by_timer(started_cluster): # on sanitizers builds it can return 'FAILED_AND_RELOADING' which is not quite right # add retry for these builds - if ( - instance.is_built_with_sanitizer() - and get_status("no_file_2") == "FAILED_AND_RELOADING" - ): - get_status_retry("no_file_2", expect="FAILED") - - assert get_status("no_file_2") == "FAILED" + assert get_status("no_file_2") in ["FAILED", "FAILED_AND_RELOADING"] # Creating the file source makes the dictionary able to load. instance.copy_file_to_container( @@ -284,7 +268,7 @@ def test_reload_after_fail_by_timer(started_cluster): ) instance.query("SYSTEM RELOAD DICTIONARY no_file_2") instance.query("SELECT dictGetInt32('no_file_2', 'a', toUInt64(9))") == "10\n" - assert get_status("no_file_2") == "LOADED" + assert get_status("no_file_2") in ["LOADED", "LOADED_AND_RELOADING"] # Removing the file source should not spoil the loaded dictionary. instance.exec_in_container( @@ -292,7 +276,7 @@ def test_reload_after_fail_by_timer(started_cluster): ) time.sleep(6) instance.query("SELECT dictGetInt32('no_file_2', 'a', toUInt64(9))") == "10\n" - assert get_status("no_file_2") == "LOADED" + assert get_status("no_file_2") in ["LOADED", "LOADED_AND_RELOADING"] def test_reload_after_fail_in_cache_dictionary(started_cluster): diff --git a/tests/integration/test_disk_configuration/test.py b/tests/integration/test_disk_configuration/test.py index afc5303298c..f297c665dc5 100644 --- a/tests/integration/test_disk_configuration/test.py +++ b/tests/integration/test_disk_configuration/test.py @@ -373,7 +373,7 @@ def test_merge_tree_setting_override(start_cluster): CREATE TABLE {TABLE_NAME} (a Int32) ENGINE = MergeTree() ORDER BY tuple() - SETTINGS disk = 'kek', storage_policy = 's3'; + SETTINGS disk = 's3', storage_policy = 's3'; """ ) ) diff --git a/tests/integration/test_distributed_inter_server_secret/test.py b/tests/integration/test_distributed_inter_server_secret/test.py index 7ecb2cda257..d74cf97e5c6 100644 --- a/tests/integration/test_distributed_inter_server_secret/test.py +++ b/tests/integration/test_distributed_inter_server_secret/test.py @@ -46,6 +46,10 @@ users = pytest.mark.parametrize( ) +def generate_query_id(): + return str(uuid.uuid4()) + + def bootstrap(): for n in list(cluster.instances.values()): n.query("DROP TABLE IF EXISTS data") @@ -116,10 +120,10 @@ def start_cluster(): cluster.shutdown() -# @return -- [user, initial_user] +# @return -- [[user, initial_user]] def get_query_user_info(node, query_pattern): node.query("SYSTEM FLUSH LOGS") - return ( + lines = ( node.query( """ SELECT user, initial_user @@ -133,8 +137,10 @@ def get_query_user_info(node, query_pattern): ) ) .strip() - .split("\t") + .split("\n") ) + lines = map(lambda x: x.split("\t"), lines) + return list(lines) # @return -- [user, initial_user] @@ -266,13 +272,13 @@ def test_secure_insert_buffer_async(): n1.query("SYSTEM RELOAD CONFIG") # ensure that SELECT creates new connection (we need separate table for # this, so that separate distributed pool will be used) - query_id = uuid.uuid4().hex + query_id = generate_query_id() n1.query("SELECT * FROM dist_secure_from_buffer", user="ro", query_id=query_id) assert n1.contains_in_log( "{" + query_id + "} Connection (n2:9000): Connecting." ) - query_id = uuid.uuid4().hex + query_id = generate_query_id() n1.query( "INSERT INTO dist_secure_buffer SELECT * FROM numbers(2)", query_id=query_id ) @@ -329,26 +335,26 @@ def test_secure_disagree_insert(): @users def test_user_insecure_cluster(user, password): - id_ = "query-dist_insecure-" + user + id_ = "query-dist_insecure-" + user + "-" + generate_query_id() n1.query(f"SELECT *, '{id_}' FROM dist_insecure", user=user, password=password) - assert get_query_user_info(n1, id_) == [ + assert get_query_user_info(n1, id_)[0] == [ user, user, ] # due to prefer_localhost_replica - assert get_query_user_info(n2, id_) == ["default", user] + assert get_query_user_info(n2, id_)[0] == ["default", user] @users def test_user_secure_cluster(user, password): - id_ = "query-dist_secure-" + user + id_ = "query-dist_secure-" + user + "-" + generate_query_id() n1.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(n2, id_) == [user, user] + assert get_query_user_info(n1, id_)[0] == [user, user] + assert get_query_user_info(n2, id_)[0] == [user, user] @users def test_per_user_inline_settings_insecure_cluster(user, password): - id_ = "query-ddl-settings-dist_insecure-" + user + id_ = "query-ddl-settings-dist_insecure-" + user + "-" + generate_query_id() n1.query( f""" SELECT *, '{id_}' FROM dist_insecure @@ -365,7 +371,7 @@ def test_per_user_inline_settings_insecure_cluster(user, password): @users def test_per_user_inline_settings_secure_cluster(user, password): - id_ = "query-ddl-settings-dist_secure-" + user + id_ = "query-ddl-settings-dist_secure-" + user + "-" + generate_query_id() n1.query( f""" SELECT *, '{id_}' FROM dist_secure @@ -384,7 +390,7 @@ def test_per_user_inline_settings_secure_cluster(user, password): @users def test_per_user_protocol_settings_insecure_cluster(user, password): - id_ = "query-protocol-settings-dist_insecure-" + user + id_ = "query-protocol-settings-dist_insecure-" + user + "-" + generate_query_id() n1.query( f"SELECT *, '{id_}' FROM dist_insecure", user=user, @@ -400,7 +406,7 @@ def test_per_user_protocol_settings_insecure_cluster(user, password): @users def test_per_user_protocol_settings_secure_cluster(user, password): - id_ = "query-protocol-settings-dist_secure-" + user + id_ = "query-protocol-settings-dist_secure-" + user + "-" + generate_query_id() n1.query( f"SELECT *, '{id_}' FROM dist_secure", user=user, @@ -416,7 +422,7 @@ def test_per_user_protocol_settings_secure_cluster(user, password): ) -def test_secure_cluster_distributed_over_distributed_different_users(): +def test_secure_cluster_distributed_over_distributed_different_users_remote(): # This works because we will have initial_user='default' n1.query( "SELECT * FROM remote('n1', currentDatabase(), dist_secure)", user="new_user" @@ -431,3 +437,16 @@ def test_secure_cluster_distributed_over_distributed_different_users(): # and stuff). with pytest.raises(QueryRuntimeException): n1.query("SELECT * FROM dist_over_dist_secure", user="new_user") + + +def test_secure_cluster_distributed_over_distributed_different_users_cluster(): + id_ = "cluster-user" + "-" + generate_query_id() + n1.query( + f"SELECT *, '{id_}' FROM cluster(secure, currentDatabase(), dist_secure)", + user="nopass", + settings={ + "prefer_localhost_replica": 0, + }, + ) + assert get_query_user_info(n1, id_) == [["nopass", "nopass"]] * 4 + assert get_query_user_info(n2, id_) == [["nopass", "nopass"]] * 3 diff --git a/tests/integration/test_distributed_type_object/configs/remote_servers.xml b/tests/integration/test_distributed_type_object/configs/remote_servers.xml index ebce4697529..68b420f36b4 100644 --- a/tests/integration/test_distributed_type_object/configs/remote_servers.xml +++ b/tests/integration/test_distributed_type_object/configs/remote_servers.xml @@ -1,4 +1,4 @@ - + @@ -15,4 +15,4 @@ - + diff --git a/tests/integration/test_distributed_type_object/test.py b/tests/integration/test_distributed_type_object/test.py index 360087c9dda..e274bd6b774 100644 --- a/tests/integration/test_distributed_type_object/test.py +++ b/tests/integration/test_distributed_type_object/test.py @@ -31,6 +31,9 @@ def started_cluster(): def test_distributed_type_object(started_cluster): + node1.query("TRUNCATE TABLE local_table") + node2.query("TRUNCATE TABLE local_table") + node1.query( 'INSERT INTO local_table FORMAT JSONEachRow {"id": 1, "data": {"k1": 10}}' ) @@ -89,7 +92,7 @@ def test_distributed_type_object(started_cluster): assert ( TSV( node1.query( - "SELECT id, data.k1, data.k2.k3, data.k2.k4, data.k5 FROM dist_table ORDER BY id SETTINGS allow_experimental_analyzer = 0" + "SELECT id, data.k1, data.k2.k3, data.k2.k4, data.k5 FROM dist_table ORDER BY id SETTINGS enable_analyzer = 0" ) ) == expected diff --git a/tests/integration/test_drop_is_lock_free/test.py b/tests/integration/test_drop_is_lock_free/test.py index 1bb8767a9a0..3855bc21f90 100644 --- a/tests/integration/test_drop_is_lock_free/test.py +++ b/tests/integration/test_drop_is_lock_free/test.py @@ -176,7 +176,7 @@ def test_query_is_permanent(transaction, permanent, exclusive_table): select_handler = node.get_query_request( f""" - SELECT sleepEachRow(3) FROM {exclusive_table} SETTINGS function_sleep_max_microseconds_per_block = 0; + SELECT sleepEachRow(3) FROM {exclusive_table} SETTINGS function_sleep_max_microseconds_per_block = 0, max_threads=1; """, query_id=query_id, ) diff --git a/tests/integration/test_hedged_requests/test.py b/tests/integration/test_hedged_requests/test.py index 02ecf3c1367..0d72f7c45b1 100644 --- a/tests/integration/test_hedged_requests/test.py +++ b/tests/integration/test_hedged_requests/test.py @@ -333,7 +333,7 @@ def test_receive_timeout2(started_cluster): # in packet receiving but there are replicas in process of # connection establishing. update_configs( - node_1_sleep_in_send_data=4000, + node_1_sleep_in_send_data=5000, node_2_sleep_in_send_tables_status=2000, node_3_sleep_in_send_tables_status=2000, ) diff --git a/tests/integration/test_host_regexp_hosts_file_resolution/configs/host_regexp.xml b/tests/integration/test_host_regexp_hosts_file_resolution/configs/host_regexp.xml index 7a2141e6c7e..9329c8dbde2 100644 --- a/tests/integration/test_host_regexp_hosts_file_resolution/configs/host_regexp.xml +++ b/tests/integration/test_host_regexp_hosts_file_resolution/configs/host_regexp.xml @@ -1,4 +1,4 @@ - + @@ -8,4 +8,4 @@ default - \ No newline at end of file + diff --git a/tests/integration/test_host_regexp_hosts_file_resolution/configs/listen_host.xml b/tests/integration/test_host_regexp_hosts_file_resolution/configs/listen_host.xml index 58ef55cd3f3..9c27c612f63 100644 --- a/tests/integration/test_host_regexp_hosts_file_resolution/configs/listen_host.xml +++ b/tests/integration/test_host_regexp_hosts_file_resolution/configs/listen_host.xml @@ -1,5 +1,5 @@ - + :: 0.0.0.0 1 - + diff --git a/tests/integration/test_jbod_ha/configs/config.d/storage_configuration.xml b/tests/integration/test_jbod_ha/configs/config.d/storage_configuration.xml index b5c351d105b..fb9acc58ad6 100644 --- a/tests/integration/test_jbod_ha/configs/config.d/storage_configuration.xml +++ b/tests/integration/test_jbod_ha/configs/config.d/storage_configuration.xml @@ -1,4 +1,4 @@ - + 1000 @@ -27,4 +27,4 @@ - + diff --git a/tests/integration/test_mask_sensitive_info/test.py b/tests/integration/test_mask_sensitive_info/test.py index 6f6dc4d287f..8d5345082ff 100644 --- a/tests/integration/test_mask_sensitive_info/test.py +++ b/tests/integration/test_mask_sensitive_info/test.py @@ -202,6 +202,10 @@ def test_create_table(): f"S3Queue('http://minio1:9001/root/data/', 'CSV', 'gzip') settings mode = 'ordered'", f"S3Queue('http://minio1:9001/root/data/', 'minio', '{password}', 'CSV') settings mode = 'ordered'", f"S3Queue('http://minio1:9001/root/data/', 'minio', '{password}', 'CSV', 'gzip') settings mode = 'ordered'", + ( + f"Iceberg('http://minio1:9001/root/data/test11.csv.gz', 'minio', '{password}')", + "DNS_ERROR", + ), ] def make_test_case(i): @@ -266,6 +270,7 @@ def test_create_table(): # due to sensitive data substituion the query will be normalized, so not "settings" but "SETTINGS" "CREATE TABLE table19 (`x` int) ENGINE = S3Queue('http://minio1:9001/root/data/', 'minio', '[HIDDEN]', 'CSV') SETTINGS mode = 'ordered'", "CREATE TABLE table20 (`x` int) ENGINE = S3Queue('http://minio1:9001/root/data/', 'minio', '[HIDDEN]', 'CSV', 'gzip') SETTINGS mode = 'ordered'", + "CREATE TABLE table21 (`x` int) ENGINE = Iceberg('http://minio1:9001/root/data/test11.csv.gz', 'minio', '[HIDDEN]')", ], must_not_contain=[password], ) @@ -387,6 +392,7 @@ def test_table_functions(): f"azureBlobStorageCluster('test_shard_localhost', '{azure_storage_account_url}', 'cont', 'test_simple_15.csv', '{azure_account_name}', '{azure_account_key}', 'CSV', 'none', 'auto')", f"azureBlobStorageCluster('test_shard_localhost', named_collection_2, connection_string = '{azure_conn_string}', container = 'cont', blob_path = 'test_simple_16.csv', format = 'CSV')", f"azureBlobStorageCluster('test_shard_localhost', named_collection_2, storage_account_url = '{azure_storage_account_url}', container = 'cont', blob_path = 'test_simple_17.csv', account_name = '{azure_account_name}', account_key = '{azure_account_key}')", + f"iceberg('http://minio1:9001/root/data/test11.csv.gz', 'minio', '{password}')", ] def make_test_case(i): @@ -478,6 +484,7 @@ def test_table_functions(): f"CREATE TABLE tablefunc48 (`x` int) AS azureBlobStorageCluster('test_shard_localhost', '{azure_storage_account_url}', 'cont', 'test_simple_15.csv', '{azure_account_name}', '[HIDDEN]', 'CSV', 'none', 'auto')", f"CREATE TABLE tablefunc49 (x int) AS azureBlobStorageCluster('test_shard_localhost', named_collection_2, connection_string = '{azure_conn_string}', container = 'cont', blob_path = 'test_simple_16.csv', format = 'CSV')", f"CREATE TABLE tablefunc50 (`x` int) AS azureBlobStorageCluster('test_shard_localhost', named_collection_2, storage_account_url = '{azure_storage_account_url}', container = 'cont', blob_path = 'test_simple_17.csv', account_name = '{azure_account_name}', account_key = '[HIDDEN]')", + "CREATE TABLE tablefunc51 (`x` int) AS iceberg('http://minio1:9001/root/data/test11.csv.gz', 'minio', '[HIDDEN]')", ], must_not_contain=[password], ) diff --git a/tests/integration/test_odbc_interaction/test.py b/tests/integration/test_odbc_interaction/test.py index 0d0d7a0afb1..9d4ca5ad49f 100644 --- a/tests/integration/test_odbc_interaction/test.py +++ b/tests/integration/test_odbc_interaction/test.py @@ -51,9 +51,9 @@ create_table_sql_nullable_template = """ """ -def skip_test_msan(instance): - if instance.is_built_with_memory_sanitizer(): - pytest.skip("Memory Sanitizer cannot work with third-party shared libraries") +def skip_test_sanitizers(instance): + if instance.is_built_with_sanitizer(): + pytest.skip("Sanitizers cannot work with third-party shared libraries") def get_mysql_conn(): @@ -208,7 +208,7 @@ def started_cluster(): def test_mysql_odbc_select_nullable(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) mysql_setup = node1.odbc_drivers["MySQL"] table_name = "test_insert_nullable_select" @@ -248,7 +248,7 @@ def test_mysql_odbc_select_nullable(started_cluster): def test_mysql_simple_select_works(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) mysql_setup = node1.odbc_drivers["MySQL"] @@ -331,7 +331,7 @@ CREATE TABLE {}(id UInt32, name String, age UInt32, money UInt32, column_x Nulla def test_mysql_insert(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) mysql_setup = node1.odbc_drivers["MySQL"] table_name = "test_insert" @@ -374,7 +374,7 @@ def test_mysql_insert(started_cluster): def test_sqlite_simple_select_function_works(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) sqlite_setup = node1.odbc_drivers["SQLite3"] sqlite_db = sqlite_setup["Database"] @@ -438,7 +438,7 @@ def test_sqlite_simple_select_function_works(started_cluster): def test_sqlite_table_function(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) sqlite_setup = node1.odbc_drivers["SQLite3"] sqlite_db = sqlite_setup["Database"] @@ -470,7 +470,7 @@ def test_sqlite_table_function(started_cluster): def test_sqlite_simple_select_storage_works(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) sqlite_setup = node1.odbc_drivers["SQLite3"] sqlite_db = sqlite_setup["Database"] @@ -503,7 +503,7 @@ def test_sqlite_simple_select_storage_works(started_cluster): def test_sqlite_odbc_hashed_dictionary(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) sqlite_db = node1.odbc_drivers["SQLite3"]["Database"] node1.exec_in_container( @@ -586,7 +586,7 @@ def test_sqlite_odbc_hashed_dictionary(started_cluster): def test_sqlite_odbc_cached_dictionary(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) sqlite_db = node1.odbc_drivers["SQLite3"]["Database"] node1.exec_in_container( @@ -635,7 +635,7 @@ def test_sqlite_odbc_cached_dictionary(started_cluster): def test_postgres_odbc_hashed_dictionary_with_schema(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) try: conn = get_postgres_conn(started_cluster) @@ -663,7 +663,7 @@ def test_postgres_odbc_hashed_dictionary_with_schema(started_cluster): def test_postgres_odbc_hashed_dictionary_no_tty_pipe_overflow(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) try: conn = get_postgres_conn(started_cluster) @@ -685,7 +685,7 @@ def test_postgres_odbc_hashed_dictionary_no_tty_pipe_overflow(started_cluster): def test_no_connection_pooling(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) try: conn = get_postgres_conn(started_cluster) @@ -717,7 +717,7 @@ def test_no_connection_pooling(started_cluster): def test_postgres_insert(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) conn = get_postgres_conn(started_cluster) @@ -754,7 +754,7 @@ def test_postgres_insert(started_cluster): def test_odbc_postgres_date_data_type(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) try: conn = get_postgres_conn(started_cluster) @@ -783,7 +783,7 @@ def test_odbc_postgres_date_data_type(started_cluster): def test_odbc_postgres_conversions(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) try: conn = get_postgres_conn(started_cluster) @@ -841,7 +841,7 @@ def test_odbc_postgres_conversions(started_cluster): def test_odbc_cyrillic_with_varchar(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) conn = get_postgres_conn(started_cluster) cursor = conn.cursor() @@ -868,7 +868,7 @@ def test_odbc_cyrillic_with_varchar(started_cluster): def test_many_connections(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) conn = get_postgres_conn(started_cluster) cursor = conn.cursor() @@ -894,7 +894,7 @@ def test_many_connections(started_cluster): def test_concurrent_queries(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) conn = get_postgres_conn(started_cluster) cursor = conn.cursor() @@ -948,7 +948,7 @@ def test_concurrent_queries(started_cluster): def test_odbc_long_column_names(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) conn = get_postgres_conn(started_cluster) cursor = conn.cursor() @@ -986,7 +986,7 @@ def test_odbc_long_column_names(started_cluster): def test_odbc_long_text(started_cluster): - skip_test_msan(node1) + skip_test_sanitizers(node1) conn = get_postgres_conn(started_cluster) cursor = conn.cursor() diff --git a/tests/queries/0_stateless/02864_statistics_exception.reference b/tests/integration/test_parquet_page_index/__init__.py similarity index 100% rename from tests/queries/0_stateless/02864_statistics_exception.reference rename to tests/integration/test_parquet_page_index/__init__.py diff --git a/tests/integration/test_parquet_page_index/test.py b/tests/integration/test_parquet_page_index/test.py new file mode 100644 index 00000000000..778b6618a61 --- /dev/null +++ b/tests/integration/test_parquet_page_index/test.py @@ -0,0 +1,108 @@ +import pytest +from helpers.cluster import ClickHouseCluster +import pyarrow.parquet as pq +import os +import time + +cluster = ClickHouseCluster(__file__) +path_to_userfiles = "/var/lib/clickhouse/user_files/" +node = cluster.add_instance("node", external_dirs=[path_to_userfiles]) + + +@pytest.fixture(scope="module") +def start_cluster(): + try: + cluster.start() + yield cluster + finally: + cluster.shutdown() + + +def check_page_index(file_path): + metadata = pq.read_metadata(file_path) + assert ( + metadata + ), "pyarrow.parquet library can't read parquet file written by Clickhouse" + return metadata.row_group(0).column(0).has_offset_index + + +def delete_if_exists(file_path): + if os.path.exists(file_path): + os.remove(file_path) + + +@pytest.mark.parametrize( + "query, expected_result", + { + ( + "SELECT number, number+1 FROM system.numbers LIMIT 100 " + "INTO OUTFILE '{file_name}' FORMAT Parquet " + "SETTINGS output_format_parquet_use_custom_encoder = false, " + "output_format_parquet_write_page_index = true;", + True, + ), + ( + "SELECT number, number+1 FROM system.numbers LIMIT 100 " + "INTO OUTFILE '{file_name}' FORMAT Parquet " + "SETTINGS output_format_parquet_use_custom_encoder = false, " + "output_format_parquet_write_page_index = false;", + False, + ), + # # default settings: + # # output_format_parquet_use_custom_encoder = true + ( + "SELECT number, number+1 FROM system.numbers LIMIT 100 " + "INTO OUTFILE '{file_name}' FORMAT Parquet;", + False, + ), + }, +) +def test_parquet_page_index_select_into_outfile(query, expected_result, start_cluster): + file_name = f"export{time.time()}.parquet" + query = query.format(file_name=file_name) + delete_if_exists(file_name) + assert node.query(query) == "" + assert ( + check_page_index(file_name) == expected_result + ), "Page offset index have wrong value" + delete_if_exists(file_name) + + +@pytest.mark.parametrize( + "query, expected_result", + { + ( + "INSERT INTO TABLE FUNCTION file('{file_name}') " + "SELECT number, number+1 FROM system.numbers LIMIT 100 " + "SETTINGS output_format_parquet_use_custom_encoder=false, " + "output_format_parquet_write_page_index=true FORMAT Parquet", + True, + ), + ( + "INSERT INTO TABLE FUNCTION file('{file_name}') " + "SELECT number, number+1 FROM system.numbers LIMIT 100 " + "SETTINGS output_format_parquet_use_custom_encoder=false, " + "output_format_parquet_write_page_index=false FORMAT Parquet", + False, + ), + # # default settings: + # # output_format_parquet_use_custom_encoder = true + ( + "INSERT INTO TABLE FUNCTION file('{file_name}') " + "SELECT number, number+1 FROM system.numbers LIMIT 100 FORMAT Parquet", + False, + ), + }, +) +def test_parquet_page_index_insert_into_table_function_file( + query, expected_result, start_cluster +): + file_name = f"export{time.time()}.parquet" + query = query.format(file_name=file_name) + file_path = f"{cluster.instances_dir}{path_to_userfiles}{file_name}" + delete_if_exists(file_path) + assert node.query(query) == "" + assert ( + check_page_index(file_path) == expected_result + ), "Page offset index have wrong value" + delete_if_exists(file_path) diff --git a/tests/integration/test_postgresql_replica_database_engine_2/test.py b/tests/integration/test_postgresql_replica_database_engine_2/test.py index 406b50bc486..7fdd17625a9 100644 --- a/tests/integration/test_postgresql_replica_database_engine_2/test.py +++ b/tests/integration/test_postgresql_replica_database_engine_2/test.py @@ -953,12 +953,14 @@ def test_generated_columns(started_cluster): "", f"""CREATE TABLE {table} ( key integer PRIMARY KEY, - x integer, + x integer DEFAULT 0, + temp integer DEFAULT 0, y integer GENERATED ALWAYS AS (x*2) STORED, - z text); + z text DEFAULT 'z'); """, ) + pg_manager.execute(f"alter table {table} drop column temp;") pg_manager.execute(f"insert into {table} (key, x, z) values (1,1,'1');") pg_manager.execute(f"insert into {table} (key, x, z) values (2,2,'2');") @@ -991,6 +993,44 @@ def test_generated_columns(started_cluster): ) +def test_generated_columns_with_sequence(started_cluster): + table = "test_generated_columns_with_sequence" + + pg_manager.create_postgres_table( + table, + "", + f"""CREATE TABLE {table} ( + key integer PRIMARY KEY, + x integer, + y integer GENERATED ALWAYS AS (x*2) STORED, + z text); + """, + ) + + pg_manager.execute( + f"create sequence {table}_id_seq increment by 1 minvalue 1 start 1;" + ) + pg_manager.execute( + f"alter table {table} alter key set default nextval('{table}_id_seq');" + ) + pg_manager.execute(f"insert into {table} (key, x, z) values (1,1,'1');") + pg_manager.execute(f"insert into {table} (key, x, z) values (2,2,'2');") + + pg_manager.create_materialized_db( + ip=started_cluster.postgres_ip, + port=started_cluster.postgres_port, + settings=[ + f"materialized_postgresql_tables_list = '{table}'", + "materialized_postgresql_backoff_min_ms = 100", + "materialized_postgresql_backoff_max_ms = 100", + ], + ) + + check_tables_are_synchronized( + instance, table, postgres_database=pg_manager.get_default_database() + ) + + def test_default_columns(started_cluster): table = "test_default_columns" @@ -1087,9 +1127,13 @@ def test_dependent_loading(started_cluster): nested_time = instance.query( f"SELECT event_time_microseconds FROM system.text_log WHERE message like 'Loading table default.{uuid}_nested' and message not like '%like%'" ).strip() - time = instance.query( - f"SELECT event_time_microseconds FROM system.text_log WHERE message like 'Loading table default.{table}' and message not like '%like%'" - ).strip() + time = ( + instance.query( + f"SELECT event_time_microseconds FROM system.text_log WHERE message like 'Loading table default.{table}' and message not like '%like%'" + ) + .strip() + .split("\n")[-1] + ) instance.query( f"SELECT toDateTime64('{nested_time}', 6) < toDateTime64('{time}', 6)" ) diff --git a/tests/integration/test_prometheus_protocols/__init__.py b/tests/integration/test_prometheus_protocols/__init__.py new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/integration/test_prometheus_protocols/configs/allow_experimental_time_series_table.xml b/tests/integration/test_prometheus_protocols/configs/allow_experimental_time_series_table.xml new file mode 100644 index 00000000000..d71cfcaf2c8 --- /dev/null +++ b/tests/integration/test_prometheus_protocols/configs/allow_experimental_time_series_table.xml @@ -0,0 +1,7 @@ + + + + 1 + + + diff --git a/tests/integration/test_prometheus_protocols/configs/prometheus.xml b/tests/integration/test_prometheus_protocols/configs/prometheus.xml new file mode 100644 index 00000000000..071a29620cf --- /dev/null +++ b/tests/integration/test_prometheus_protocols/configs/prometheus.xml @@ -0,0 +1,21 @@ + + + 9092 + + + /write + + remote_write +
{test_result.status}{test_result.time}{test_result.time}
default.prometheus
+ + + + /read + + remote_read + default.prometheus
+ + + + + diff --git a/tests/integration/test_prometheus_protocols/test.py b/tests/integration/test_prometheus_protocols/test.py new file mode 100644 index 00000000000..0c75a8194c7 --- /dev/null +++ b/tests/integration/test_prometheus_protocols/test.py @@ -0,0 +1,177 @@ +import pytest +import time +import requests +from http import HTTPStatus +from helpers.cluster import ClickHouseCluster + +cluster = ClickHouseCluster(__file__) + +node = cluster.add_instance( + "node", + main_configs=["configs/prometheus.xml"], + user_configs=["configs/allow_experimental_time_series_table.xml"], + with_prometheus=True, + handle_prometheus_remote_write=True, + handle_prometheus_remote_read=True, +) + + +def execute_query_on_prometheus_writer(query, timestamp): + return execute_query_impl( + cluster.get_instance_ip(cluster.prometheus_writer_host), + cluster.prometheus_writer_port, + "/api/v1/query", + query, + timestamp, + ) + + +def execute_query_on_prometheus_reader(query, timestamp): + return execute_query_impl( + cluster.get_instance_ip(cluster.prometheus_reader_host), + cluster.prometheus_reader_port, + "/api/v1/query", + query, + timestamp, + ) + + +def execute_query_impl(host, port, path, query, timestamp): + if not path.startswith("/"): + path += "/" + url = f"http://{host}:{port}/{path.strip('/')}?query={query}&time={timestamp}" + print(f"Requesting {url}") + r = requests.get(url) + print(f"Status code: {r.status_code} {HTTPStatus(r.status_code).phrase}") + if r.status_code != requests.codes.ok: + print(f"Response: {r.text}") + raise Exception(f"Got unexpected status code {r.status_code}") + return r.json() + + +def show_query_result(query): + evaluation_time = time.time() + print(f"Evaluating query: {query}") + print(f"Evaluation time: {evaluation_time}") + result_from_writer = execute_query_on_prometheus_writer(query, evaluation_time) + print(f"Result from prometheus_writer: {result_from_writer}") + result_from_reader = execute_query_on_prometheus_reader(query, evaluation_time) + print(f"Result from prometheus_reader: {result_from_reader}") + + +def compare_query(query): + timeout = 60 + start_time = time.time() + evaluation_time = start_time + print(f"Evaluating query: {query}") + print(f"Evaluation time: {evaluation_time}") + while time.time() < start_time + timeout: + result_from_writer = execute_query_on_prometheus_writer(query, evaluation_time) + time.sleep(1) + result_from_reader = execute_query_on_prometheus_reader(query, evaluation_time) + print(f"Result from prometheus_writer: {result_from_writer}") + print(f"Result from prometheus_reader: {result_from_reader}") + if result_from_writer == result_from_reader: + return + raise Exception( + f"Got different results from prometheus_writer and prometheus_reader" + ) + + +def compare_queries(): + compare_query("up") + + +@pytest.fixture(scope="module", autouse=True) +def start_cluster(): + try: + cluster.start() + yield cluster + finally: + cluster.shutdown() + + +@pytest.fixture(autouse=True) +def cleanup_after_test(): + try: + yield + finally: + node.query("DROP TABLE IF EXISTS prometheus SYNC") + node.query("DROP TABLE IF EXISTS original SYNC") + node.query("DROP TABLE IF EXISTS mydata SYNC") + node.query("DROP TABLE IF EXISTS mytable SYNC") + node.query("DROP TABLE IF EXISTS mymetrics SYNC") + + +def test_default(): + node.query("CREATE TABLE prometheus ENGINE=TimeSeries") + compare_queries() + + +def test_tags_to_columns(): + node.query( + "CREATE TABLE prometheus ENGINE=TimeSeries SETTINGS tags_to_columns = {'job': 'job', 'instance': 'instance'}" + ) + compare_queries() + + +def test_64bit_id(): + node.query("CREATE TABLE prometheus (id UInt64) ENGINE=TimeSeries") + compare_queries() + + +def test_custom_id_algorithm(): + node.query( + "CREATE TABLE prometheus (id FixedString(16) DEFAULT murmurHash3_128(metric_name, all_tags)) ENGINE=TimeSeries" + ) + compare_queries() + + +def test_create_as_table(): + node.query("CREATE TABLE original ENGINE=TimeSeries") + node.query("CREATE TABLE prometheus AS original") + compare_queries() + + +def test_inner_engines(): + node.query( + "CREATE TABLE prometheus ENGINE=TimeSeries " + "DATA ENGINE=MergeTree ORDER BY (id, timestamp) " + "TAGS ENGINE=AggregatingMergeTree ORDER BY (metric_name, id) " + "METRICS ENGINE=ReplacingMergeTree ORDER BY metric_family_name" + ) + compare_queries() + + +def test_external_tables(): + node.query("DROP TABLE IF EXISTS mydata") + node.query("DROP TABLE IF EXISTS mytags") + node.query("DROP TABLE IF EXISTS mymetrics") + node.query("DROP TABLE IF EXISTS prometheus") + + node.query( + "CREATE TABLE mydata (id UUID, timestamp DateTime64(3), value Float64) " + "ENGINE=MergeTree ORDER BY (id, timestamp)" + ) + node.query( + "CREATE TABLE mytags (" + "id UUID, " + "metric_name LowCardinality(String), " + "tags Map(LowCardinality(String), String), " + "min_time SimpleAggregateFunction(min, Nullable(DateTime64(3))), " + "max_time SimpleAggregateFunction(max, Nullable(DateTime64(3)))) " + "ENGINE=AggregatingMergeTree ORDER BY (metric_name, id)" + ) + + # FIXME: The table structure should be: + # "CREATE TABLE mymetrics (metric_family_name String, type LowCardinality(String), unit LowCardinality(String), help String)" + # Renamed it because of the bug and potential type mismatch. + node.query( + "CREATE TABLE mymetrics (metric_family_name String, type String, unit String, help String) " + "ENGINE=ReplacingMergeTree ORDER BY metric_family_name" + ) + node.query( + "CREATE TABLE prometheus ENGINE=TimeSeries " + "DATA mydata TAGS mytags METRICS mymetrics" + ) + compare_queries() diff --git a/tests/integration/test_recovery_time_metric/__init__.py b/tests/integration/test_recovery_time_metric/__init__.py new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/integration/test_recovery_time_metric/configs/config.xml b/tests/integration/test_recovery_time_metric/configs/config.xml new file mode 100644 index 00000000000..bad9b1fa9ea --- /dev/null +++ b/tests/integration/test_recovery_time_metric/configs/config.xml @@ -0,0 +1,41 @@ + + 9000 + + + + + + + + + default + + + + + + 2181 + 1 + /var/lib/clickhouse/coordination/log + /var/lib/clickhouse/coordination/snapshots + + 20000 + + + + 1 + localhost + 9444 + + + + + + + localhost + 2181 + + 20000 + + + diff --git a/tests/integration/test_recovery_time_metric/test.py b/tests/integration/test_recovery_time_metric/test.py new file mode 100644 index 00000000000..6fcf2fad423 --- /dev/null +++ b/tests/integration/test_recovery_time_metric/test.py @@ -0,0 +1,61 @@ +import pytest +from helpers.cluster import ClickHouseCluster + +cluster = ClickHouseCluster(__file__) +node = cluster.add_instance( + "node", + main_configs=["configs/config.xml"], + stay_alive=True, +) + + +@pytest.fixture(scope="module") +def start_cluster(): + try: + cluster.start() + yield cluster + finally: + cluster.shutdown() + + +def test_recovery_time_metric(start_cluster): + node.query( + """ + DROP DATABASE IF EXISTS rdb; + CREATE DATABASE rdb + ENGINE = Replicated('/test/test_recovery_time_metric', 'shard1', 'replica1') + """ + ) + + node.query( + """ + DROP TABLE IF EXISTS rdb.t; + CREATE TABLE rdb.t + ( + `x` UInt32 + ) + ENGINE = MergeTree + ORDER BY x + """ + ) + + node.exec_in_container(["bash", "-c", "rm /var/lib/clickhouse/metadata/rdb/t.sql"]) + + node.restart_clickhouse() + + ret = int( + node.query( + """ + SELECT recovery_time + FROM system.clusters + WHERE cluster = 'rdb' + """ + ).strip() + ) + assert ret > 0 + + node.query( + """ + DROP DATABASE rdb + """ + ) diff --git a/tests/integration/test_replicated_merge_tree_replicated_db_ttl/configs/enable_parallel_replicas.xml b/tests/integration/test_replicated_merge_tree_replicated_db_ttl/configs/enable_parallel_replicas.xml index c654074740a..30a0b6276b7 100644 --- a/tests/integration/test_replicated_merge_tree_replicated_db_ttl/configs/enable_parallel_replicas.xml +++ b/tests/integration/test_replicated_merge_tree_replicated_db_ttl/configs/enable_parallel_replicas.xml @@ -1,7 +1,7 @@ - 1 + 1 1 default 100 diff --git a/tests/integration/test_restore_external_engines/__init__.py b/tests/integration/test_restore_external_engines/__init__.py new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/integration/test_restore_external_engines/configs/backups_disk.xml b/tests/integration/test_restore_external_engines/configs/backups_disk.xml new file mode 100644 index 00000000000..f7d666c6542 --- /dev/null +++ b/tests/integration/test_restore_external_engines/configs/backups_disk.xml @@ -0,0 +1,14 @@ + + + + + local + /backups/ + + + + + backups + /backups/ + + diff --git a/tests/integration/test_restore_external_engines/configs/remote_servers.xml b/tests/integration/test_restore_external_engines/configs/remote_servers.xml new file mode 100644 index 00000000000..76ad3618339 --- /dev/null +++ b/tests/integration/test_restore_external_engines/configs/remote_servers.xml @@ -0,0 +1,21 @@ + + + + + true + + replica1 + 9000 + + + replica2 + 9000 + + + replica3 + 9000 + + + + + diff --git a/tests/integration/test_restore_external_engines/test.py b/tests/integration/test_restore_external_engines/test.py new file mode 100644 index 00000000000..cf189f2a6ed --- /dev/null +++ b/tests/integration/test_restore_external_engines/test.py @@ -0,0 +1,218 @@ +import pytest + +import pymysql.cursors +import pytest +from helpers.cluster import ClickHouseCluster + +cluster = ClickHouseCluster(__file__) +configs = ["configs/remote_servers.xml", "configs/backups_disk.xml"] + +node1 = cluster.add_instance( + "replica1", + with_zookeeper=True, + with_mysql8=True, + main_configs=configs, + external_dirs=["/backups/"], +) +node2 = cluster.add_instance( + "replica2", + with_zookeeper=True, + with_mysql8=True, + main_configs=configs, + external_dirs=["/backups/"], +) +node3 = cluster.add_instance( + "replica3", + with_zookeeper=True, + with_mysql8=True, + main_configs=configs, + external_dirs=["/backups/"], +) +nodes = [node1, node2, node3] + +backup_id_counter = 0 + + +def new_backup_name(): + global backup_id_counter + backup_id_counter += 1 + return f"Disk('backups', '{backup_id_counter}/')" + + +def cleanup_nodes(nodes, dbname): + for node in nodes: + node.query(f"DROP DATABASE IF EXISTS {dbname} SYNC") + + +def fill_nodes(nodes, dbname): + cleanup_nodes(nodes, dbname) + for node in nodes: + node.query( + f"CREATE DATABASE {dbname} ENGINE = Replicated('/clickhouse/databases/{dbname}', 'default', '{node.name}')" + ) + + +def drop_mysql_table(conn, tableName): + with conn.cursor() as cursor: + cursor.execute(f"DROP TABLE IF EXISTS `clickhouse`.`{tableName}`") + + +def get_mysql_conn(cluster): + conn = pymysql.connect( + user="root", + password="clickhouse", + host=cluster.mysql8_ip, + port=cluster.mysql8_port, + ) + return conn + + +def fill_tables(cluster, dbname): + fill_nodes(nodes, dbname) + + conn = get_mysql_conn(cluster) + + with conn.cursor() as cursor: + cursor.execute("DROP DATABASE IF EXISTS clickhouse") + cursor.execute("CREATE DATABASE clickhouse") + cursor.execute("DROP TABLE IF EXISTS clickhouse.inference_table") + cursor.execute( + "CREATE TABLE clickhouse.inference_table (id INT PRIMARY KEY, data BINARY(16) NOT NULL)" + ) + cursor.execute( + "INSERT INTO clickhouse.inference_table VALUES (100, X'9fad5e9eefdfb449')" + ) + conn.commit() + + parameters = "'mysql80:3306', 'clickhouse', 'inference_table', 'root', 'clickhouse'" + + node1.query( + f"CREATE TABLE {dbname}.mysql_schema_inference_engine ENGINE=MySQL({parameters})" + ) + node1.query( + f"CREATE TABLE {dbname}.mysql_schema_inference_function AS mysql({parameters})" + ) + + node1.query(f"CREATE TABLE {dbname}.merge_tree (id UInt64, b String) ORDER BY id") + node1.query(f"INSERT INTO {dbname}.merge_tree VALUES (100, 'abc')") + + expected = "id\tInt32\t\t\t\t\t\ndata\tFixedString(16)\t\t\t\t\t\n" + assert ( + node1.query(f"DESCRIBE TABLE {dbname}.mysql_schema_inference_engine") + == expected + ) + assert ( + node1.query(f"DESCRIBE TABLE {dbname}.mysql_schema_inference_function") + == expected + ) + assert node1.query(f"SELECT id FROM mysql({parameters})") == "100\n" + assert ( + node1.query(f"SELECT id FROM {dbname}.mysql_schema_inference_engine") == "100\n" + ) + assert ( + node1.query(f"SELECT id FROM {dbname}.mysql_schema_inference_function") + == "100\n" + ) + assert node1.query(f"SELECT id FROM {dbname}.merge_tree") == "100\n" + + +@pytest.fixture(scope="module") +def start_cluster(): + try: + cluster.start() + yield cluster + + except Exception as ex: + print(ex) + + finally: + cluster.shutdown() + + +def test_restore_table(start_cluster): + fill_tables(cluster, "replicated") + backup_name = new_backup_name() + node2.query(f"SYSTEM SYNC DATABASE REPLICA replicated") + + node2.query(f"BACKUP DATABASE replicated TO {backup_name}") + + node2.query("DROP TABLE replicated.mysql_schema_inference_engine") + node2.query("DROP TABLE replicated.mysql_schema_inference_function") + + node3.query(f"SYSTEM SYNC DATABASE REPLICA replicated") + + assert node3.query("EXISTS replicated.mysql_schema_inference_engine") == "0\n" + assert node3.query("EXISTS replicated.mysql_schema_inference_function") == "0\n" + + node3.query( + f"RESTORE DATABASE replicated FROM {backup_name} SETTINGS allow_different_database_def=true" + ) + node1.query(f"SYSTEM SYNC DATABASE REPLICA replicated") + + assert ( + node1.query( + "SELECT count(), sum(id) FROM replicated.mysql_schema_inference_engine" + ) + == "1\t100\n" + ) + assert ( + node1.query( + "SELECT count(), sum(id) FROM replicated.mysql_schema_inference_function" + ) + == "1\t100\n" + ) + assert ( + node1.query("SELECT count(), sum(id) FROM replicated.merge_tree") == "1\t100\n" + ) + cleanup_nodes(nodes, "replicated") + + +def test_restore_table_null(start_cluster): + fill_tables(cluster, "replicated2") + + backup_name = new_backup_name() + node2.query(f"SYSTEM SYNC DATABASE REPLICA replicated2") + + node2.query(f"BACKUP DATABASE replicated2 TO {backup_name}") + + node2.query("DROP TABLE replicated2.mysql_schema_inference_engine") + node2.query("DROP TABLE replicated2.mysql_schema_inference_function") + + node3.query(f"SYSTEM SYNC DATABASE REPLICA replicated2") + + assert node3.query("EXISTS replicated2.mysql_schema_inference_engine") == "0\n" + assert node3.query("EXISTS replicated2.mysql_schema_inference_function") == "0\n" + + node3.query( + f"RESTORE DATABASE replicated2 FROM {backup_name} SETTINGS allow_different_database_def=1, allow_different_table_def=1 SETTINGS restore_replace_external_engines_to_null=1, restore_replace_external_table_functions_to_null=1" + ) + node1.query(f"SYSTEM SYNC DATABASE REPLICA replicated2") + + assert ( + node1.query( + "SELECT count(), sum(id) FROM replicated2.mysql_schema_inference_engine" + ) + == "0\t0\n" + ) + assert ( + node1.query( + "SELECT count(), sum(id) FROM replicated2.mysql_schema_inference_function" + ) + == "0\t0\n" + ) + assert ( + node1.query("SELECT count(), sum(id) FROM replicated2.merge_tree") == "1\t100\n" + ) + assert ( + node1.query( + "SELECT engine FROM system.tables where database = 'replicated2' and name like '%mysql%'" + ) + == "Null\nNull\n" + ) + assert ( + node1.query( + "SELECT engine FROM system.tables where database = 'replicated2' and name like '%merge_tree%'" + ) + == "MergeTree\n" + ) + cleanup_nodes(nodes, "replicated2") diff --git a/tests/integration/test_s3_imds/test_simple.py b/tests/integration/test_s3_imds/test_simple.py index 0dacac2b0b9..4884c824f99 100644 --- a/tests/integration/test_s3_imds/test_simple.py +++ b/tests/integration/test_s3_imds/test_simple.py @@ -56,7 +56,7 @@ def test_credentials_from_metadata(): ) expected_logs = [ - "Calling EC2MetadataService to get token failed, falling back to less secure way", + "Calling EC2MetadataService to get token failed, falling back to a less secure way", "Getting default credentials for ec2 instance from resolver:8080", "Calling EC2MetadataService resource, /latest/meta-data/iam/security-credentials returned credential string myrole", "Calling EC2MetadataService resource /latest/meta-data/iam/security-credentials/myrole", diff --git a/tests/integration/test_server_keep_alive/__init__.py b/tests/integration/test_server_keep_alive/__init__.py new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/integration/test_server_keep_alive/configs/keep_alive_settings.xml b/tests/integration/test_server_keep_alive/configs/keep_alive_settings.xml new file mode 100644 index 00000000000..06e68044817 --- /dev/null +++ b/tests/integration/test_server_keep_alive/configs/keep_alive_settings.xml @@ -0,0 +1,4 @@ + + 3600 + 5 + diff --git a/tests/integration/test_server_keep_alive/test.py b/tests/integration/test_server_keep_alive/test.py new file mode 100644 index 00000000000..e550319b6df --- /dev/null +++ b/tests/integration/test_server_keep_alive/test.py @@ -0,0 +1,55 @@ +import logging +import pytest +import random +import requests + +from helpers.cluster import ClickHouseCluster + +cluster = ClickHouseCluster(__file__) +node = cluster.add_instance("node", main_configs=["configs/keep_alive_settings.xml"]) + + +@pytest.fixture(scope="module") +def start_cluster(): + try: + logging.info("Starting cluster...") + cluster.start() + logging.info("Cluster started") + + yield cluster + finally: + cluster.shutdown() + + +def test_max_keep_alive_requests_on_user_side(start_cluster): + # In this test we have `keep_alive_timeout` set to one hour to never trigger connection reset by timeout, `max_keep_alive_requests` is set to 5. + # We expect server to close connection after each 5 requests. We detect connection reset by change in src port. + # So the first 5 requests should come from the same port, the following 5 requests should come from another port. + + log_comments = [] + for _ in range(10): + rand_id = random.randint(0, 1000000) + log_comment = f"test_requests_with_keep_alive_{rand_id}" + log_comments.append(log_comment) + log_comments = sorted(log_comments) + + session = requests.Session() + for i in range(10): + session.get( + f"http://{node.ip_address}:8123/?query=select%201&log_comment={log_comments[i]}" + ) + + ports = node.query( + f""" + SYSTEM FLUSH LOGS; + + SELECT port + FROM system.query_log + WHERE log_comment IN ({", ".join(f"'{comment}'" for comment in log_comments)}) AND type = 'QueryFinish' + ORDER BY log_comment + """ + ).split("\n")[:-1] + + expected = 5 * [ports[0]] + [ports[5]] * 5 + + assert ports == expected diff --git a/tests/integration/test_server_reload/configs/default_passwd.xml b/tests/integration/test_server_reload/configs/default_passwd.xml index f79149e7e23..9d664cbf9c4 100644 --- a/tests/integration/test_server_reload/configs/default_passwd.xml +++ b/tests/integration/test_server_reload/configs/default_passwd.xml @@ -1,4 +1,4 @@ - + @@ -9,4 +9,4 @@ 123 - + diff --git a/tests/integration/test_server_reload/configs/overrides_from_zk.xml b/tests/integration/test_server_reload/configs/overrides_from_zk.xml index d420faa88a2..aa6105f6ebe 100644 --- a/tests/integration/test_server_reload/configs/overrides_from_zk.xml +++ b/tests/integration/test_server_reload/configs/overrides_from_zk.xml @@ -1,4 +1,4 @@ - + @@ -7,4 +7,4 @@ - + diff --git a/tests/integration/test_session_log/configs/users.xml b/tests/integration/test_session_log/configs/users.xml index 0416dfadc8a..766fdbcf00f 100644 --- a/tests/integration/test_session_log/configs/users.xml +++ b/tests/integration/test_session_log/configs/users.xml @@ -7,17 +7,5 @@ - - pass - - - pass - - - pass - - - pass - \ No newline at end of file diff --git a/tests/integration/test_session_log/test.py b/tests/integration/test_session_log/test.py index 922e2557c50..0eb614f7aa9 100644 --- a/tests/integration/test_session_log/test.py +++ b/tests/integration/test_session_log/test.py @@ -5,6 +5,7 @@ import pytest import random import sys import threading +import time from helpers.cluster import ClickHouseCluster, run_and_check @@ -59,6 +60,19 @@ def next_session_id(): return str(session_id) +user_counter = 0 + + +def create_unique_user(prefix): + global user_counter + user_counter += 1 + user_name = f"{prefix}_{os.getppid()}_{user_counter}" + instance.query( + f"CREATE USER {user_name} IDENTIFIED WITH plaintext_password BY 'pass'" + ) + return user_name + + def grpc_query(query, user_, pass_, raise_exception): try: query_info = clickhouse_grpc_pb2.QueryInfo( @@ -117,6 +131,50 @@ def mysql_query(query, user_, pass_, raise_exception): assert raise_exception +def wait_for_corresponding_login_success_and_logout(user, expected_login_count): + # The client can exit sooner than the server records its disconnection and closes the session. + # When the client disconnects, two processes happen at the same time and are in the race condition: + # - the client application exits and returns control to the shell; + # - the server closes the session and records the logout event to the session log. + # We cannot expect that after the control is returned to the shell, the server records the logout event. + sql = f"SELECT COUNT(*) FROM (SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '{user}' AND type = 'LoginSuccess' INTERSECT SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '{user}' AND type = 'Logout')" + logins_and_logouts = instance.query(sql) + while int(logins_and_logouts) != expected_login_count: + time.sleep(0.1) + logins_and_logouts = instance.query(sql) + + +def check_session_log(user): + instance.query("SYSTEM FLUSH LOGS") + login_success_records = instance.query( + f"SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='{user}' AND type = 'LoginSuccess'" + ) + assert login_success_records == f"{user}\t1\t1\n" + logout_records = instance.query( + f"SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='{user}' AND type = 'Logout'" + ) + assert logout_records == f"{user}\t1\t1\n" + login_failure_records = instance.query( + f"SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='{user}' AND type = 'LoginFailure'" + ) + assert login_failure_records == f"{user}\t1\t1\n" + + wait_for_corresponding_login_success_and_logout(user, 1) + + +def session_log_test(prefix, query_function): + user = create_unique_user(prefix) + wrong_user = "wrong_" + user + + query_function("SELECT 1", user, "pass", False) + query_function("SELECT 2", user, "wrong_pass", True) + query_function("SELECT 3", wrong_user, "pass", True) + + check_session_log(user) + + instance.query(f"DROP USER {user}") + + @pytest.fixture(scope="module") def started_cluster(): try: @@ -131,78 +189,21 @@ def started_cluster(): def test_grpc_session(started_cluster): - grpc_query("SELECT 1", "grpc_user", "pass", False) - grpc_query("SELECT 2", "grpc_user", "wrong_pass", True) - grpc_query("SELECT 3", "wrong_grpc_user", "pass", True) - - instance.query("SYSTEM FLUSH LOGS") - login_success_records = instance.query( - "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='grpc_user' AND type = 'LoginSuccess'" - ) - assert login_success_records == "grpc_user\t1\t1\n" - logout_records = instance.query( - "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='grpc_user' AND type = 'Logout'" - ) - assert logout_records == "grpc_user\t1\t1\n" - login_failure_records = instance.query( - "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='grpc_user' AND type = 'LoginFailure'" - ) - assert login_failure_records == "grpc_user\t1\t1\n" - logins_and_logouts = instance.query( - f"SELECT COUNT(*) FROM (SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'grpc_user' AND type = 'LoginSuccess' INTERSECT SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'grpc_user' AND type = 'Logout')" - ) - assert logins_and_logouts == "1\n" + session_log_test("grpc", grpc_query) def test_mysql_session(started_cluster): - mysql_query("SELECT 1", "mysql_user", "pass", False) - mysql_query("SELECT 2", "mysql_user", "wrong_pass", True) - mysql_query("SELECT 3", "wrong_mysql_user", "pass", True) - - instance.query("SYSTEM FLUSH LOGS") - login_success_records = instance.query( - "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='mysql_user' AND type = 'LoginSuccess'" - ) - assert login_success_records == "mysql_user\t1\t1\n" - logout_records = instance.query( - "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='mysql_user' AND type = 'Logout'" - ) - assert logout_records == "mysql_user\t1\t1\n" - login_failure_records = instance.query( - "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='mysql_user' AND type = 'LoginFailure'" - ) - assert login_failure_records == "mysql_user\t1\t1\n" - logins_and_logouts = instance.query( - f"SELECT COUNT(*) FROM (SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'mysql_user' AND type = 'LoginSuccess' INTERSECT SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'mysql_user' AND type = 'Logout')" - ) - assert logins_and_logouts == "1\n" + session_log_test("mysql", mysql_query) def test_postgres_session(started_cluster): - postgres_query("SELECT 1", "postgres_user", "pass", False) - postgres_query("SELECT 2", "postgres_user", "wrong_pass", True) - postgres_query("SELECT 3", "wrong_postgres_user", "pass", True) - - instance.query("SYSTEM FLUSH LOGS") - login_success_records = instance.query( - "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='postgres_user' AND type = 'LoginSuccess'" - ) - assert login_success_records == "postgres_user\t1\t1\n" - logout_records = instance.query( - "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='postgres_user' AND type = 'Logout'" - ) - assert logout_records == "postgres_user\t1\t1\n" - login_failure_records = instance.query( - "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='postgres_user' AND type = 'LoginFailure'" - ) - assert login_failure_records == "postgres_user\t1\t1\n" - logins_and_logouts = instance.query( - f"SELECT COUNT(*) FROM (SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'postgres_user' AND type = 'LoginSuccess' INTERSECT SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'postgres_user' AND type = 'Logout')" - ) - assert logins_and_logouts == "1\n" + session_log_test("postgres", postgres_query) def test_parallel_sessions(started_cluster): + user = create_unique_user("parallel") + wrong_user = "wrong_" + user + thread_list = [] for _ in range(10): # Sleep time does not significantly matter here, @@ -212,7 +213,7 @@ def test_parallel_sessions(started_cluster): target=function, args=( f"SELECT sleep({random.uniform(0.03, 0.04)})", - "parallel_user", + user, "pass", False, ), @@ -223,7 +224,7 @@ def test_parallel_sessions(started_cluster): target=function, args=( f"SELECT sleep({random.uniform(0.03, 0.04)})", - "parallel_user", + user, "wrong_pass", True, ), @@ -234,7 +235,7 @@ def test_parallel_sessions(started_cluster): target=function, args=( f"SELECT sleep({random.uniform(0.03, 0.04)})", - "wrong_parallel_user", + wrong_user, "pass", True, ), @@ -247,41 +248,38 @@ def test_parallel_sessions(started_cluster): instance.query("SYSTEM FLUSH LOGS") port_0_sessions = instance.query( - f"SELECT COUNT(*) FROM system.session_log WHERE user = 'parallel_user'" + f"SELECT COUNT(*) FROM system.session_log WHERE user = '{user}'" ) assert port_0_sessions == "90\n" port_0_sessions = instance.query( - f"SELECT COUNT(*) FROM system.session_log WHERE user = 'parallel_user' AND client_port = 0" + f"SELECT COUNT(*) FROM system.session_log WHERE user = '{user}' AND client_port = 0" ) assert port_0_sessions == "0\n" address_0_sessions = instance.query( - f"SELECT COUNT(*) FROM system.session_log WHERE user = 'parallel_user' AND client_address = toIPv6('::')" + f"SELECT COUNT(*) FROM system.session_log WHERE user = '{user}' AND client_address = toIPv6('::')" ) assert address_0_sessions == "0\n" grpc_sessions = instance.query( - f"SELECT COUNT(*) FROM system.session_log WHERE user = 'parallel_user' AND interface = 'gRPC'" + f"SELECT COUNT(*) FROM system.session_log WHERE user = '{user}' AND interface = 'gRPC'" ) assert grpc_sessions == "30\n" mysql_sessions = instance.query( - f"SELECT COUNT(*) FROM system.session_log WHERE user = 'parallel_user' AND interface = 'MySQL'" + f"SELECT COUNT(*) FROM system.session_log WHERE user = '{user}' AND interface = 'MySQL'" ) assert mysql_sessions == "30\n" postgres_sessions = instance.query( - f"SELECT COUNT(*) FROM system.session_log WHERE user = 'parallel_user' AND interface = 'PostgreSQL'" + f"SELECT COUNT(*) FROM system.session_log WHERE user = '{user}' AND interface = 'PostgreSQL'" ) assert postgres_sessions == "30\n" - logins_and_logouts = instance.query( - f"SELECT COUNT(*) FROM (SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'parallel_user' AND type = 'LoginSuccess' INTERSECT SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'parallel_user' AND type = 'Logout')" - ) - assert logins_and_logouts == "30\n" + wait_for_corresponding_login_success_and_logout(user, 30) logout_failure_sessions = instance.query( - f"SELECT COUNT(*) FROM system.session_log WHERE user = 'parallel_user' AND type = 'LoginFailure'" + f"SELECT COUNT(*) FROM system.session_log WHERE user = '{user}' AND type = 'LoginFailure'" ) assert logout_failure_sessions == "30\n" diff --git a/tests/integration/test_settings_profile/test.py b/tests/integration/test_settings_profile/test.py index e5c0a072ff9..4800ab798bf 100644 --- a/tests/integration/test_settings_profile/test.py +++ b/tests/integration/test_settings_profile/test.py @@ -459,7 +459,7 @@ def test_show_profiles(): query_possible_response = [ "CREATE SETTINGS PROFILE `default`\n", - "CREATE SETTINGS PROFILE `default` SETTINGS allow_experimental_analyzer = true\n", + "CREATE SETTINGS PROFILE `default` SETTINGS enable_analyzer = true\n", ] assert ( instance.query("SHOW CREATE SETTINGS PROFILE default") @@ -470,7 +470,7 @@ def test_show_profiles(): "CREATE SETTINGS PROFILE `default`\n" "CREATE SETTINGS PROFILE `readonly` SETTINGS readonly = 1\n" "CREATE SETTINGS PROFILE `xyz`\n", - "CREATE SETTINGS PROFILE `default` SETTINGS allow_experimental_analyzer = true\n" + "CREATE SETTINGS PROFILE `default` SETTINGS enable_analyzer = true\n" "CREATE SETTINGS PROFILE `readonly` SETTINGS readonly = 1\n" "CREATE SETTINGS PROFILE `xyz`\n", ] @@ -482,7 +482,7 @@ def test_show_profiles(): "CREATE SETTINGS PROFILE `xyz`\n" ) expected_access_analyzer = ( - "CREATE SETTINGS PROFILE `default` SETTINGS allow_experimental_analyzer = true\n" + "CREATE SETTINGS PROFILE `default` SETTINGS enable_analyzer = true\n" "CREATE SETTINGS PROFILE `readonly` SETTINGS readonly = 1\n" "CREATE SETTINGS PROFILE `xyz`\n" ) diff --git a/tests/integration/test_storage_azure_blob_storage/test.py b/tests/integration/test_storage_azure_blob_storage/test.py index 6fbe7634642..fbdc7f29f98 100644 --- a/tests/integration/test_storage_azure_blob_storage/test.py +++ b/tests/integration/test_storage_azure_blob_storage/test.py @@ -5,6 +5,7 @@ import json import logging import os import io +import re import random import threading import time @@ -134,6 +135,7 @@ def test_create_table_connection_string(cluster): Engine = AzureBlobStorage('{cluster.env_variables['AZURITE_CONNECTION_STRING']}', 'cont', 'test_create_connection_string', 'CSV') """, ) + azure_query(node, "DROP TABLE IF EXISTS test_create_table_conn_string") def test_create_table_account_string(cluster): @@ -143,6 +145,7 @@ def test_create_table_account_string(cluster): f"CREATE TABLE test_create_table_account_url (key UInt64, data String) Engine = AzureBlobStorage('{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}'," f"'cont', 'test_create_connection_string', 'devstoreaccount1', 'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'CSV')", ) + azure_query(node, "DROP TABLE IF EXISTS test_create_table_account_url") def test_simple_write_account_string(cluster): @@ -156,6 +159,7 @@ def test_simple_write_account_string(cluster): azure_query(node, "INSERT INTO test_simple_write VALUES (1, 'a')") print(get_azure_file_content("test_simple_write.csv", port)) assert get_azure_file_content("test_simple_write.csv", port) == '1,"a"\n' + azure_query(node, "DROP TABLE test_simple_write") def test_simple_write_connection_string(cluster): @@ -169,6 +173,7 @@ def test_simple_write_connection_string(cluster): azure_query(node, "INSERT INTO test_simple_write_connection_string VALUES (1, 'a')") print(get_azure_file_content("test_simple_write_c.csv", port)) assert get_azure_file_content("test_simple_write_c.csv", port) == '1,"a"\n' + azure_query(node, "DROP TABLE test_simple_write_connection_string") def test_simple_write_named_collection_1(cluster): @@ -184,7 +189,7 @@ def test_simple_write_named_collection_1(cluster): ) print(get_azure_file_content("test_simple_write_named.csv", port)) assert get_azure_file_content("test_simple_write_named.csv", port) == '1,"a"\n' - azure_query(node, "TRUNCATE TABLE test_simple_write_named_collection_1") + azure_query(node, "DROP TABLE test_simple_write_named_collection_1") def test_simple_write_named_collection_2(cluster): @@ -201,6 +206,7 @@ def test_simple_write_named_collection_2(cluster): ) print(get_azure_file_content("test_simple_write_named_2.csv", port)) assert get_azure_file_content("test_simple_write_named_2.csv", port) == '1,"a"\n' + azure_query(node, "DROP TABLE test_simple_write_named_collection_2") def test_partition_by(cluster): @@ -222,6 +228,7 @@ def test_partition_by(cluster): assert "1,2,3\n" == get_azure_file_content("test_3.csv", port) assert "3,2,1\n" == get_azure_file_content("test_1.csv", port) assert "78,43,45\n" == get_azure_file_content("test_45.csv", port) + azure_query(node, "DROP TABLE test_partitioned_write") def test_partition_by_string_column(cluster): @@ -242,6 +249,7 @@ def test_partition_by_string_column(cluster): assert '1,"foo/bar"\n' == get_azure_file_content("test_foo/bar.csv", port) assert '3,"Š¹Ń†ŃƒŠŗ"\n' == get_azure_file_content("test_Š¹Ń†ŃƒŠŗ.csv", port) assert '78,"你儽"\n' == get_azure_file_content("test_你儽.csv", port) + azure_query(node, "DROP TABLE test_partitioned_string_write") def test_partition_by_const_column(cluster): @@ -260,6 +268,7 @@ def test_partition_by_const_column(cluster): ) azure_query(node, f"INSERT INTO test_partitioned_const_write VALUES {values}") assert values_csv == get_azure_file_content("test_88.csv", port) + azure_query(node, "DROP TABLE test_partitioned_const_write") def test_truncate(cluster): @@ -275,6 +284,7 @@ def test_truncate(cluster): azure_query(node, "TRUNCATE TABLE test_truncate") with pytest.raises(Exception): print(get_azure_file_content("test_truncate.csv", port)) + azure_query(node, "DROP TABLE test_truncate") def test_simple_read_write(cluster): @@ -291,6 +301,7 @@ def test_simple_read_write(cluster): assert get_azure_file_content("test_simple_read_write.csv", port) == '1,"a"\n' print(azure_query(node, "SELECT * FROM test_simple_read_write")) assert azure_query(node, "SELECT * FROM test_simple_read_write") == "1\ta\n" + azure_query(node, "DROP TABLE test_simple_read_write") def test_create_new_files_on_insert(cluster): @@ -343,6 +354,7 @@ def test_overwrite(cluster): result = azure_query(node, f"select count() from test_overwrite") assert int(result) == 200 + azure_query(node, f"DROP TABLE test_overwrite") def test_insert_with_path_with_globs(cluster): @@ -355,6 +367,7 @@ def test_insert_with_path_with_globs(cluster): node.query_and_get_error( f"insert into table function test_insert_globs SELECT number, randomString(100) FROM numbers(500)" ) + azure_query(node, f"DROP TABLE test_insert_globs") def test_put_get_with_globs(cluster): @@ -363,6 +376,7 @@ def test_put_get_with_globs(cluster): node = cluster.instances["node"] # type: ClickHouseInstance table_format = "column1 UInt32, column2 UInt32, column3 UInt32" max_path = "" + used_names = [] for i in range(10): for j in range(10): path = "{}/{}_{}/{}.csv".format( @@ -371,6 +385,8 @@ def test_put_get_with_globs(cluster): max_path = max(path, max_path) values = f"({i},{j},{i + j})" + used_names.append(f"test_put_{i}_{j}") + azure_query( node, f"CREATE TABLE test_put_{i}_{j} ({table_format}) Engine = AzureBlobStorage(azure_conf2, " @@ -391,6 +407,9 @@ def test_put_get_with_globs(cluster): bucket="cont", max_path=max_path ) ] + azure_query(node, "DROP TABLE test_glob_select") + for name in used_names: + azure_query(node, f"DROP TABLE {name}") def test_azure_glob_scheherazade(cluster): @@ -399,12 +418,14 @@ def test_azure_glob_scheherazade(cluster): values = "(1, 1, 1)" nights_per_job = 1001 // 30 jobs = [] + used_names = [] for night in range(0, 1001, nights_per_job): def add_tales(start, end): for i in range(start, end): path = "night_{}/tale.csv".format(i) unique_num = random.randint(1, 10000) + used_names.append(f"test_scheherazade_{i}_{unique_num}") azure_query( node, f"CREATE TABLE test_scheherazade_{i}_{unique_num} ({table_format}) Engine = AzureBlobStorage(azure_conf2, " @@ -432,6 +453,9 @@ def test_azure_glob_scheherazade(cluster): ) query = "select count(), sum(column1), sum(column2), sum(column3) from test_glob_select_scheherazade" assert azure_query(node, query).splitlines() == ["1001\t1001\t1001\t1001"] + azure_query(node, "DROP TABLE test_glob_select_scheherazade") + for name in used_names: + azure_query(node, f"DROP TABLE {name}") @pytest.mark.parametrize( @@ -505,6 +529,8 @@ def test_schema_inference_no_globs(cluster): assert azure_query(node, query).splitlines() == [ "499500\t2890\t332833500\ttest_schema_inference_no_globs.csv\tcont/test_schema_inference_no_globs.csv" ] + azure_query(node, f"DROP TABLE test_schema_inference_src") + azure_query(node, f"DROP TABLE test_select_inference") def test_schema_inference_from_globs(cluster): @@ -513,6 +539,7 @@ def test_schema_inference_from_globs(cluster): node = cluster.instances["node"] # type: ClickHouseInstance table_format = "column1 UInt32, column2 UInt32, column3 UInt32" max_path = "" + used_names = [] for i in range(10): for j in range(10): path = "{}/{}_{}/{}.csv".format( @@ -520,6 +547,7 @@ def test_schema_inference_from_globs(cluster): ) max_path = max(path, max_path) values = f"({i},{j},{i + j})" + used_names.append(f"test_schema_{i}_{j}") azure_query( node, @@ -545,6 +573,9 @@ def test_schema_inference_from_globs(cluster): bucket="cont", max_path=max_path ) ] + azure_query(node, "DROP TABLE test_glob_select_inference") + for name in used_names: + azure_query(node, f"DROP TABLE {name}") def test_simple_write_account_string_table_function(cluster): @@ -594,7 +625,7 @@ def test_simple_write_named_collection_1_table_function(cluster): azure_query( node, - "TRUNCATE TABLE drop_table", + "DROP TABLE drop_table", ) @@ -605,6 +636,7 @@ def test_simple_write_named_collection_2_table_function(cluster): node, f"INSERT INTO TABLE FUNCTION azureBlobStorage(azure_conf2, storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}'," f" container='cont', blob_path='test_simple_write_named_2_tf.csv', format='CSV', structure='key UInt64, data String') VALUES (1, 'a')", + settings={"azure_truncate_on_insert": 1}, ) print(get_azure_file_content("test_simple_write_named_2_tf.csv", port)) assert get_azure_file_content("test_simple_write_named_2_tf.csv", port) == '1,"a"\n' @@ -628,6 +660,7 @@ def test_put_get_with_globs_tf(cluster): node, f"INSERT INTO TABLE FUNCTION azureBlobStorage(azure_conf2, storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}'," f" container='cont', blob_path='{path}', format='CSV', compression='auto', structure='{table_format}') VALUES {values}", + settings={"azure_truncate_on_insert": 1}, ) query = ( f"select sum(column1), sum(column2), sum(column3), min(_file), max(_path) from azureBlobStorage(azure_conf2, " @@ -648,7 +681,7 @@ def test_schema_inference_no_globs_tf(cluster): query = ( f"insert into table function azureBlobStorage(azure_conf2, storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', " f"container='cont', blob_path='test_schema_inference_no_globs_tf.csv', format='CSVWithNames', structure='{table_format}') " - f"SELECT number, toString(number), number * number FROM numbers(1000)" + f"SELECT number, toString(number), number * number FROM numbers(1000) SETTINGS azure_truncate_on_insert=1" ) azure_query(node, query) @@ -681,7 +714,7 @@ def test_schema_inference_from_globs_tf(cluster): f"insert into table function azureBlobStorage(azure_conf2, storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', " f"container='cont', blob_path='{path}', format='CSVWithNames', structure='{table_format}') VALUES {values}" ) - azure_query(node, query) + azure_query(node, query, settings={"azure_truncate_on_insert": 1}) query = ( f"select sum(column1), sum(column2), sum(column3), min(_file), max(_path) from azureBlobStorage(azure_conf2, " @@ -708,6 +741,7 @@ def test_partition_by_tf(cluster): f"INSERT INTO TABLE FUNCTION azureBlobStorage('{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', " f"'cont', '{filename}', 'devstoreaccount1', 'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', " f"'CSV', 'auto', '{table_format}') PARTITION BY {partition_by} VALUES {values}", + settings={"azure_truncate_on_insert": 1}, ) assert "1,2,3\n" == get_azure_file_content("test_partition_tf_3.csv", port) @@ -727,6 +761,7 @@ def test_filter_using_file(cluster): f"INSERT INTO TABLE FUNCTION azureBlobStorage('{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', 'cont', '{filename}', " f"'devstoreaccount1', 'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'CSV', 'auto', " f"'{table_format}') PARTITION BY {partition_by} VALUES {values}", + settings={"azure_truncate_on_insert": 1}, ) query = ( @@ -744,7 +779,7 @@ def test_read_subcolumns(cluster): node, f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_subcolumns.tsv', " f"'devstoreaccount1', 'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', 'auto'," - f" 'a Tuple(b Tuple(c UInt32, d UInt32), e UInt32)') select ((1, 2), 3)", + f" 'a Tuple(b Tuple(c UInt32, d UInt32), e UInt32)') select ((1, 2), 3) SETTINGS azure_truncate_on_insert=1", ) azure_query( @@ -794,7 +829,7 @@ def test_read_subcolumn_time(cluster): node, f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_subcolumn_time.tsv', " f"'devstoreaccount1', 'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', 'auto'," - f" 'a UInt32') select (42)", + f" 'a UInt32') select (42) SETTINGS azure_truncate_on_insert=1", ) res = node.query( @@ -825,6 +860,7 @@ def test_function_signatures(cluster): azure_query( node, f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_signature.csv', '{account_name}', '{account_key}', 'CSV', 'auto', 'column1 UInt32') VALUES (1),(2),(3)", + settings={"azure_truncate_on_insert": 1}, ) # " - connection_string, container_name, blobpath\n" @@ -939,11 +975,13 @@ def test_union_schema_inference_mode(cluster): azure_query( node, f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_union_schema_inference1.jsonl', '{account_name}', '{account_key}', 'JSONEachRow', 'auto', 'a UInt32') VALUES (1)", + settings={"azure_truncate_on_insert": 1}, ) azure_query( node, f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_union_schema_inference2.jsonl', '{account_name}', '{account_key}', 'JSONEachRow', 'auto', 'b UInt32') VALUES (2)", + settings={"azure_truncate_on_insert": 1}, ) node.query("system drop schema cache for azure") @@ -981,6 +1019,7 @@ def test_union_schema_inference_mode(cluster): azure_query( node, f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_union_schema_inference3.jsonl', '{account_name}', '{account_key}', 'CSV', 'auto', 's String') VALUES ('Error')", + settings={"azure_truncate_on_insert": 1}, ) error = azure_query( @@ -1002,7 +1041,7 @@ def test_schema_inference_cache(cluster): azure_query( node, f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_cache0.jsonl', '{account_name}', '{account_key}') " - f"select * from numbers(100)", + f"select * from numbers(100) SETTINGS azure_truncate_on_insert=1", ) time.sleep(1) @@ -1209,19 +1248,19 @@ def test_filtering_by_file_or_path(cluster): azure_query( node, f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}','cont', 'test_filter1.tsv', 'devstoreaccount1', " - f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', 'auto', 'x UInt64') select 1", + f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', 'auto', 'x UInt64') select 1 SETTINGS azure_truncate_on_insert=1", ) azure_query( node, f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}','cont', 'test_filter2.tsv', 'devstoreaccount1', " - f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', 'auto', 'x UInt64') select 2", + f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', 'auto', 'x UInt64') select 2 SETTINGS azure_truncate_on_insert=1", ) azure_query( node, f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_filter3.tsv', 'devstoreaccount1', " - f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', 'auto', 'x UInt64') select 3", + f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', 'auto', 'x UInt64') select 3 SETTINGS azure_truncate_on_insert=1", ) node.query( @@ -1233,7 +1272,7 @@ def test_filtering_by_file_or_path(cluster): node.query("SYSTEM FLUSH LOGS") result = node.query( - f"SELECT ProfileEvents['EngineFileLikeReadFiles'] FROM system.query_log WHERE query ilike '%select%azure%test_filter%' AND type='QueryFinish'" + f"SELECT ProfileEvents['EngineFileLikeReadFiles'] FROM system.query_log WHERE query ilike '%select%azure%test_filter%' AND type='QueryFinish' ORDER BY event_time_microseconds DESC LIMIT 1" ) assert int(result) == 1 @@ -1245,19 +1284,19 @@ def test_size_virtual_column(cluster): azure_query( node, f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}','cont', 'test_size_virtual_column1.tsv', 'devstoreaccount1', " - f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', 'auto', 'x UInt64') select 1", + f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', 'auto', 'x UInt64') select 1 SETTINGS azure_truncate_on_insert=1", ) azure_query( node, f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}','cont', 'test_size_virtual_column2.tsv', 'devstoreaccount1', " - f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', 'auto', 'x UInt64') select 11", + f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', 'auto', 'x UInt64') select 11 SETTINGS azure_truncate_on_insert=1", ) azure_query( node, f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_size_virtual_column3.tsv', 'devstoreaccount1', " - f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', 'auto', 'x UInt64') select 111", + f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', 'auto', 'x UInt64') select 111 SETTINGS azure_truncate_on_insert=1", ) result = azure_query( @@ -1280,7 +1319,7 @@ def test_format_detection(cluster): account_key = "Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==" azure_query( node, - f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_format_detection0', '{account_name}', '{account_key}', 'JSONEachRow', 'auto', 'x UInt64, y String') select number as x, 'str_' || toString(number) from numbers(0)", + f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_format_detection0', '{account_name}', '{account_key}', 'JSONEachRow', 'auto', 'x UInt64, y String') select number as x, 'str_' || toString(number) from numbers(0) SETTINGS azure_truncate_on_insert=1", ) azure_query( @@ -1350,7 +1389,7 @@ def test_write_to_globbed_partitioned_path(cluster): account_key = "Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==" error = azure_query( node, - f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_data_*_{{_partition_id}}', '{account_name}', '{account_key}', 'CSV', 'auto', 'x UInt64') partition by 42 select 42", + f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_data_*_{{_partition_id}}', '{account_name}', '{account_key}', 'CSV', 'auto', 'x UInt64') partition by 42 select 42 SETTINGS azure_truncate_on_insert=1", expect_error="true", ) @@ -1462,3 +1501,115 @@ def test_insert_create_new_file(cluster): assert TSV(res) == TSV( "test_create_new_file.csv\t1\ntest_create_new_file.1.csv\t2\n" ) + + +def test_hive_partitioning_with_one_parameter(cluster): + # type: (ClickHouseCluster) -> None + node = cluster.instances["node"] # type: ClickHouseInstance + table_format = "column1 String, column2 String" + values = f"('Elizabeth', 'Gordon')" + path = "a/column1=Elizabeth/sample.csv" + + azure_query( + node, + f"INSERT INTO TABLE FUNCTION azureBlobStorage(azure_conf2, storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}'," + f" container='cont', blob_path='{path}', format='CSV', compression='auto', structure='{table_format}') VALUES {values}", + settings={"azure_truncate_on_insert": 1}, + ) + + query = ( + f"SELECT column1, column2, _file, _path, _column1 FROM azureBlobStorage(azure_conf2, " + f"storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', container='cont', " + f"blob_path='{path}', format='CSV', structure='{table_format}')" + ) + assert azure_query( + node, query, settings={"use_hive_partitioning": 1} + ).splitlines() == [ + "Elizabeth\tGordon\tsample.csv\t{bucket}/{max_path}\tElizabeth".format( + bucket="cont", max_path=path + ) + ] + + query = ( + f"SELECT column2 FROM azureBlobStorage(azure_conf2, " + f"storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', container='cont', " + f"blob_path='{path}', format='CSV', structure='{table_format}') WHERE column1=_column1;" + ) + assert azure_query( + node, query, settings={"use_hive_partitioning": 1} + ).splitlines() == ["Gordon"] + + +def test_hive_partitioning_with_two_parameters(cluster): + # type: (ClickHouseCluster) -> None + node = cluster.instances["node"] # type: ClickHouseInstance + table_format = "column1 String, column2 String" + values_1 = f"('Elizabeth', 'Gordon')" + values_2 = f"('Emilia', 'Gregor')" + path = "a/column1=Elizabeth/column2=Gordon/sample.csv" + + azure_query( + node, + f"INSERT INTO TABLE FUNCTION azureBlobStorage(azure_conf2, storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}'," + f" container='cont', blob_path='{path}', format='CSV', compression='auto', structure='{table_format}') VALUES {values_1}, {values_2}", + settings={"azure_truncate_on_insert": 1}, + ) + + query = ( + f"SELECT column1, column2, _file, _path, _column1, _column2 FROM azureBlobStorage(azure_conf2, " + f"storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', container='cont', " + f"blob_path='{path}', format='CSV', structure='{table_format}') WHERE column1=_column1;" + ) + assert azure_query( + node, query, settings={"use_hive_partitioning": 1} + ).splitlines() == [ + "Elizabeth\tGordon\tsample.csv\t{bucket}/{max_path}\tElizabeth\tGordon".format( + bucket="cont", max_path=path + ) + ] + + query = ( + f"SELECT column1 FROM azureBlobStorage(azure_conf2, " + f"storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', container='cont', " + f"blob_path='{path}', format='CSV', structure='{table_format}') WHERE column2=_column2;" + ) + assert azure_query( + node, query, settings={"use_hive_partitioning": 1} + ).splitlines() == ["Elizabeth"] + + query = ( + f"SELECT column1 FROM azureBlobStorage(azure_conf2, " + f"storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', container='cont', " + f"blob_path='{path}', format='CSV', structure='{table_format}') WHERE column2=_column2 AND column1=_column1;" + ) + assert azure_query( + node, query, settings={"use_hive_partitioning": 1} + ).splitlines() == ["Elizabeth"] + + +def test_hive_partitioning_without_setting(cluster): + # type: (ClickHouseCluster) -> None + node = cluster.instances["node"] # type: ClickHouseInstance + table_format = "column1 String, column2 String" + values_1 = f"('Elizabeth', 'Gordon')" + values_2 = f"('Emilia', 'Gregor')" + path = "a/column1=Elizabeth/column2=Gordon/sample.csv" + + azure_query( + node, + f"INSERT INTO TABLE FUNCTION azureBlobStorage(azure_conf2, storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}'," + f" container='cont', blob_path='{path}', format='CSV', compression='auto', structure='{table_format}') VALUES {values_1}, {values_2}", + settings={"azure_truncate_on_insert": 1}, + ) + + query = ( + f"SELECT column1, column2, _file, _path, _column1, _column2 FROM azureBlobStorage(azure_conf2, " + f"storage_account_url = '{cluster.env_variables['AZURITE_STORAGE_ACCOUNT_URL']}', container='cont', " + f"blob_path='{path}', format='CSV', structure='{table_format}') WHERE column1=_column1;" + ) + pattern = re.compile( + r"DB::Exception: Unknown expression identifier '.*' in scope.*", re.DOTALL + ) + + with pytest.raises(Exception, match=pattern): + azure_query(node, query, settings={"use_hive_partitioning": 0}) diff --git a/tests/integration/test_storage_azure_blob_storage/test_cluster.py b/tests/integration/test_storage_azure_blob_storage/test_cluster.py index 6c5e2d20ca5..04baf007c69 100644 --- a/tests/integration/test_storage_azure_blob_storage/test_cluster.py +++ b/tests/integration/test_storage_azure_blob_storage/test_cluster.py @@ -72,6 +72,7 @@ def test_select_all(cluster): f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_cluster_select_all.csv', 'devstoreaccount1'," f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'CSV', 'auto', 'key UInt64, data String') " f"VALUES (1, 'a'), (2, 'b')", + settings={"azure_truncate_on_insert": 1}, ) print(get_azure_file_content("test_cluster_select_all.csv", port)) @@ -101,6 +102,7 @@ def test_count(cluster): f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_cluster_count.csv', 'devstoreaccount1', " f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'CSV', " f"'auto', 'key UInt64') VALUES (1), (2)", + settings={"azure_truncate_on_insert": 1}, ) print(get_azure_file_content("test_cluster_count.csv", port)) @@ -129,6 +131,7 @@ def test_union_all(cluster): f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_parquet_union_all', 'devstoreaccount1', " f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'Parquet', " f"'auto', 'a Int32, b String') VALUES (1, 'a'), (2, 'b'), (3, 'c'), (4, 'd')", + settings={"azure_truncate_on_insert": 1}, ) pure_azure = azure_query( @@ -180,6 +183,7 @@ def test_skip_unavailable_shards(cluster): f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_skip_unavailable.csv', 'devstoreaccount1', " f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', " f"'auto', 'a UInt64') VALUES (1), (2)", + settings={"azure_truncate_on_insert": 1}, ) result = azure_query( node, @@ -200,6 +204,7 @@ def test_unset_skip_unavailable_shards(cluster): f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_unset_skip_unavailable.csv', 'devstoreaccount1', " f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', " f"'auto', 'a UInt64') VALUES (1), (2)", + settings={"azure_truncate_on_insert": 1}, ) result = azure_query( node, @@ -218,6 +223,7 @@ def test_cluster_with_named_collection(cluster): f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_cluster_with_named_collection.csv', 'devstoreaccount1', " f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'auto', " f"'auto', 'a UInt64') VALUES (1), (2)", + settings={"azure_truncate_on_insert": 1}, ) pure_azure = azure_query( @@ -249,6 +255,7 @@ def test_partition_parallel_reading_with_cluster(cluster): f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', '{filename}', 'devstoreaccount1', " f"'Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==', 'CSV', 'auto', '{table_format}') " f"PARTITION BY {partition_by} VALUES {values}", + settings={"azure_truncate_on_insert": 1}, ) assert "1,2,3\n" == get_azure_file_content("test_tf_3.csv", port) @@ -272,12 +279,12 @@ def test_format_detection(cluster): azure_query( node, - f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_format_detection0', '{account_name}', '{account_key}', 'JSONEachRow', 'auto', 'x UInt32, y String') select number as x, 'str_' || toString(number) from numbers(10)", + f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_format_detection0', '{account_name}', '{account_key}', 'JSONEachRow', 'auto', 'x UInt32, y String') select number as x, 'str_' || toString(number) from numbers(10) SETTINGS azure_truncate_on_insert=1", ) azure_query( node, - f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_format_detection1', '{account_name}', '{account_key}', 'JSONEachRow', 'auto', 'x UInt32, y String') select number as x, 'str_' || toString(number) from numbers(10, 10)", + f"INSERT INTO TABLE FUNCTION azureBlobStorage('{storage_account_url}', 'cont', 'test_format_detection1', '{account_name}', '{account_key}', 'JSONEachRow', 'auto', 'x UInt32, y String') select number as x, 'str_' || toString(number) from numbers(10, 10) SETTINGS azure_truncate_on_insert=1", ) expected_desc_result = azure_query( diff --git a/tests/integration/test_storage_delta/test.py b/tests/integration/test_storage_delta/test.py index 054b79ff6fe..a595d01e6b3 100644 --- a/tests/integration/test_storage_delta/test.py +++ b/tests/integration/test_storage_delta/test.py @@ -464,7 +464,7 @@ def test_restart_broken(started_cluster): """ SELECT value FROM system.metrics - WHERE metric = 'S3DiskNoKeyErrors' + WHERE metric = 'DiskS3NoSuchKeyErrors' """ ).strip() ) diff --git a/tests/integration/test_storage_hdfs/test.py b/tests/integration/test_storage_hdfs/test.py index ccd2c7eaf11..fdbf7c5bacb 100644 --- a/tests/integration/test_storage_hdfs/test.py +++ b/tests/integration/test_storage_hdfs/test.py @@ -1,8 +1,11 @@ import os import pytest +import uuid import time +import re from helpers.cluster import ClickHouseCluster, is_arm +from helpers.client import QueryRuntimeException from helpers.test_tools import TSV from pyhdfs import HdfsClient @@ -31,13 +34,15 @@ def started_cluster(): def test_read_write_storage(started_cluster): + id = uuid.uuid4() hdfs_api = started_cluster.hdfs_api + filename = f"simple_storage_{id}" node1.query("drop table if exists SimpleHDFSStorage SYNC") node1.query( - "create table SimpleHDFSStorage (id UInt32, name String, weight Float64) ENGINE = HDFS('hdfs://hdfs1:9000/simple_storage', 'TSV')" + f"create table SimpleHDFSStorage (id UInt32, name String, weight Float64) ENGINE = HDFS('hdfs://hdfs1:9000/{filename}', 'TSV')" ) node1.query("insert into SimpleHDFSStorage values (1, 'Mark', 72.53)") - assert hdfs_api.read_data("/simple_storage") == "1\tMark\t72.53\n" + assert hdfs_api.read_data(f"/{filename}") == "1\tMark\t72.53\n" assert node1.query("select * from SimpleHDFSStorage") == "1\tMark\t72.53\n" @@ -61,6 +66,17 @@ def test_read_write_storage_with_globs(started_cluster): hdfs_api.write_data("/storage" + i, i + "\tMark\t72.53\n") assert hdfs_api.read_data("/storage" + i) == i + "\tMark\t72.53\n" + node1.query( + "create table HDFSStorageWithDoubleAsterisk (id UInt32, name String, weight Float64) ENGINE = HDFS('hdfs://hdfs1:9000/**.doublestar.tsv', 'TSV')" + ) + + for i in ["1", "2", "3"]: + hdfs_api.write_data(f"/subdir{i}/file{i}.doublestar.tsv", f"{i}\tMark\t72.53\n") + assert ( + hdfs_api.read_data(f"/subdir{i}/file{i}.doublestar.tsv") + == f"{i}\tMark\t72.53\n" + ) + assert ( node1.query( "select count(*) from HDFSStorageWithRange settings s3_throw_on_zero_files_match=1" @@ -70,6 +86,7 @@ def test_read_write_storage_with_globs(started_cluster): assert node1.query("select count(*) from HDFSStorageWithEnum") == "3\n" assert node1.query("select count(*) from HDFSStorageWithQuestionMark") == "3\n" assert node1.query("select count(*) from HDFSStorageWithAsterisk") == "3\n" + assert node1.query("select count(*) from HDFSStorageWithDoubleAsterisk") == "3\n" try: node1.query("insert into HDFSStorageWithEnum values (1, 'NEW', 4.2)") @@ -92,6 +109,19 @@ def test_read_write_storage_with_globs(started_cluster): print(ex) assert "in readonly mode" in str(ex) + try: + node1.query("insert into HDFSStorageWithDoubleAsterisk values (1, 'NEW', 4.2)") + assert False, "Exception have to be thrown" + except Exception as ex: + print(ex) + assert "in readonly mode" in str(ex) + + node1.query("drop table HDFSStorageWithRange") + node1.query("drop table HDFSStorageWithEnum") + node1.query("drop table HDFSStorageWithQuestionMark") + node1.query("drop table HDFSStorageWithAsterisk") + node1.query("drop table HDFSStorageWithDoubleAsterisk") + def test_storage_with_multidirectory_glob(started_cluster): hdfs_api = started_cluster.hdfs_api @@ -137,7 +167,6 @@ def test_read_write_table(started_cluster): def test_write_table(started_cluster): hdfs_api = started_cluster.hdfs_api - node1.query( "create table OtherHDFSStorage (id UInt32, name String, weight Float64) ENGINE = HDFS('hdfs://hdfs1:9000/other_storage', 'TSV')" ) @@ -148,6 +177,8 @@ def test_write_table(started_cluster): result = "10\ttomas\t55.55\n11\tjack\t32.54\n" assert hdfs_api.read_data("/other_storage") == result assert node1.query("select * from OtherHDFSStorage order by id") == result + node1.query("truncate table OtherHDFSStorage") + node1.query("drop table OtherHDFSStorage") def test_bad_hdfs_uri(started_cluster): @@ -166,6 +197,7 @@ def test_bad_hdfs_uri(started_cluster): print(ex) assert "Unable to connect to HDFS" in str(ex) + node1.query("drop table BadStorage2") try: node1.query( "create table BadStorage3 (id UInt32, name String, weight Float64) ENGINE = HDFS('hdfs://hdfs1:9000/<>', 'TSV')" @@ -173,6 +205,7 @@ def test_bad_hdfs_uri(started_cluster): except Exception as ex: print(ex) assert "Unable to open HDFS file" in str(ex) + node1.query("drop table BadStorage3") @pytest.mark.timeout(800) @@ -304,6 +337,8 @@ def test_write_gz_storage(started_cluster): node1.query("insert into GZHDFSStorage values (1, 'Mark', 72.53)") assert hdfs_api.read_gzip_data("/storage.gz") == "1\tMark\t72.53\n" assert node1.query("select * from GZHDFSStorage") == "1\tMark\t72.53\n" + node1.query("truncate table GZHDFSStorage") + node1.query("drop table GZHDFSStorage") def test_write_gzip_storage(started_cluster): @@ -315,6 +350,8 @@ def test_write_gzip_storage(started_cluster): node1.query("insert into GZIPHDFSStorage values (1, 'Mark', 72.53)") assert hdfs_api.read_gzip_data("/gzip_storage") == "1\tMark\t72.53\n" assert node1.query("select * from GZIPHDFSStorage") == "1\tMark\t72.53\n" + node1.query("truncate table GZIPHDFSStorage") + node1.query("drop table GZIPHDFSStorage") def test_virtual_columns(started_cluster): @@ -333,6 +370,7 @@ def test_virtual_columns(started_cluster): ) == expected ) + node1.query("drop table virtual_cols") def test_read_files_with_spaces(started_cluster): @@ -354,6 +392,7 @@ def test_read_files_with_spaces(started_cluster): ) assert node1.query("select * from test order by id") == "1\n2\n3\n" fs.delete(dir, recursive=True) + node1.query(f"drop table test") def test_truncate_table(started_cluster): @@ -375,47 +414,54 @@ def test_truncate_table(started_cluster): def test_partition_by(started_cluster): - hdfs_api = started_cluster.hdfs_api - + fs = HdfsClient(hosts=started_cluster.hdfs_ip) + id = uuid.uuid4() table_format = "column1 UInt32, column2 UInt32, column3 UInt32" + dir = f"partition_{id}" + fs.mkdirs(f"/{dir}/", permission=777) + file_name = "test_{_partition_id}" partition_by = "column3" values = "(1, 2, 3), (3, 2, 1), (1, 3, 2)" - table_function = f"hdfs('hdfs://hdfs1:9000/{file_name}', 'TSV', '{table_format}')" + table_function = ( + f"hdfs('hdfs://hdfs1:9000/{dir}/{file_name}', 'TSV', '{table_format}')" + ) node1.query( f"insert into table function {table_function} PARTITION BY {partition_by} values {values}" ) result = node1.query( - f"select * from hdfs('hdfs://hdfs1:9000/test_1', 'TSV', '{table_format}')" + f"select * from hdfs('hdfs://hdfs1:9000/{dir}/test_1', 'TSV', '{table_format}')" ) assert result.strip() == "3\t2\t1" result = node1.query( - f"select * from hdfs('hdfs://hdfs1:9000/test_2', 'TSV', '{table_format}')" + f"select * from hdfs('hdfs://hdfs1:9000/{dir}/test_2', 'TSV', '{table_format}')" ) assert result.strip() == "1\t3\t2" result = node1.query( - f"select * from hdfs('hdfs://hdfs1:9000/test_3', 'TSV', '{table_format}')" + f"select * from hdfs('hdfs://hdfs1:9000/{dir}/test_3', 'TSV', '{table_format}')" ) assert result.strip() == "1\t2\t3" file_name = "test2_{_partition_id}" node1.query( - f"create table p(column1 UInt32, column2 UInt32, column3 UInt32) engine = HDFS('hdfs://hdfs1:9000/{file_name}', 'TSV') partition by column3" + f"create table p(column1 UInt32, column2 UInt32, column3 UInt32) engine = HDFS('hdfs://hdfs1:9000/{dir}/{file_name}', 'TSV') partition by column3" ) node1.query(f"insert into p values {values}") result = node1.query( - f"select * from hdfs('hdfs://hdfs1:9000/test2_1', 'TSV', '{table_format}')" + f"select * from hdfs('hdfs://hdfs1:9000/{dir}/test2_1', 'TSV', '{table_format}')" ) assert result.strip() == "3\t2\t1" result = node1.query( - f"select * from hdfs('hdfs://hdfs1:9000/test2_2', 'TSV', '{table_format}')" + f"select * from hdfs('hdfs://hdfs1:9000/{dir}/test2_2', 'TSV', '{table_format}')" ) assert result.strip() == "1\t3\t2" result = node1.query( - f"select * from hdfs('hdfs://hdfs1:9000/test2_3', 'TSV', '{table_format}')" + f"select * from hdfs('hdfs://hdfs1:9000/{dir}/test2_3', 'TSV', '{table_format}')" ) assert result.strip() == "1\t2\t3" + node1.query(f"drop table p") + fs.delete("/{dir}", recursive=True) def test_seekable_formats(started_cluster): @@ -425,7 +471,7 @@ def test_seekable_formats(started_cluster): f"hdfs('hdfs://hdfs1:9000/parquet', 'Parquet', 'a Int32, b String')" ) node1.query( - f"insert into table function {table_function} SELECT number, randomString(100) FROM numbers(5000000)" + f"insert into table function {table_function} SELECT number, randomString(100) FROM numbers(5000000) SETTINGS hdfs_truncate_on_insert=1" ) result = node1.query(f"SELECT count() FROM {table_function}") @@ -433,7 +479,7 @@ def test_seekable_formats(started_cluster): table_function = f"hdfs('hdfs://hdfs1:9000/orc', 'ORC', 'a Int32, b String')" node1.query( - f"insert into table function {table_function} SELECT number, randomString(100) FROM numbers(5000000)" + f"insert into table function {table_function} SELECT number, randomString(100) FROM numbers(5000000) SETTINGS hdfs_truncate_on_insert=1" ) result = node1.query(f"SELECT count() FROM {table_function}") assert int(result) == 5000000 @@ -457,7 +503,7 @@ def test_read_table_with_default(started_cluster): def test_schema_inference(started_cluster): node1.query( - f"insert into table function hdfs('hdfs://hdfs1:9000/native', 'Native', 'a Int32, b String') SELECT number, randomString(100) FROM numbers(5000000)" + f"insert into table function hdfs('hdfs://hdfs1:9000/native', 'Native', 'a Int32, b String') SELECT number, randomString(100) FROM numbers(5000000) SETTINGS hdfs_truncate_on_insert=1" ) result = node1.query(f"desc hdfs('hdfs://hdfs1:9000/native', 'Native')") @@ -476,6 +522,7 @@ def test_schema_inference(started_cluster): result = node1.query(f"select count(*) from schema_inference") assert int(result) == 5000000 + node1.query(f"drop table schema_inference") def test_hdfsCluster(started_cluster): @@ -510,6 +557,7 @@ def test_hdfs_directory_not_exist(started_cluster): assert "" == node1.query( "select * from HDFSStorageWithNotExistDir settings hdfs_ignore_file_doesnt_exist=1" ) + node1.query("drop table HDFSStorageWithNotExistDir") def test_overwrite(started_cluster): @@ -529,12 +577,16 @@ def test_overwrite(started_cluster): result = node1.query(f"select count() from test_overwrite") assert int(result) == 10 + node1.query(f"truncate table test_overwrite") + node1.query(f"drop table test_overwrite") def test_multiple_inserts(started_cluster): - hdfs_api = started_cluster.hdfs_api + fs = HdfsClient(hosts=started_cluster.hdfs_ip) + id = uuid.uuid4() + fs.mkdirs(f"/{id}/", permission=777) - table_function = f"hdfs('hdfs://hdfs1:9000/data_multiple_inserts', 'Parquet', 'a Int32, b String')" + table_function = f"hdfs('hdfs://hdfs1:9000/{id}/data_multiple_inserts', 'Parquet', 'a Int32, b String')" node1.query(f"create table test_multiple_inserts as {table_function}") node1.query( f"insert into test_multiple_inserts select number, randomString(100) from numbers(10)" @@ -551,7 +603,7 @@ def test_multiple_inserts(started_cluster): result = node1.query(f"drop table test_multiple_inserts") - table_function = f"hdfs('hdfs://hdfs1:9000/data_multiple_inserts.gz', 'Parquet', 'a Int32, b String')" + table_function = f"hdfs('hdfs://hdfs1:9000/{id}/data_multiple_inserts.gz', 'Parquet', 'a Int32, b String')" node1.query(f"create table test_multiple_inserts as {table_function}") node1.query( f"insert into test_multiple_inserts select number, randomString(100) FROM numbers(10)" @@ -565,6 +617,7 @@ def test_multiple_inserts(started_cluster): result = node1.query(f"select count() from test_multiple_inserts") assert int(result) == 60 + node1.query(f"drop table test_multiple_inserts") def test_format_detection(started_cluster): @@ -574,50 +627,58 @@ def test_format_detection(started_cluster): node1.query(f"insert into arrow_table select 1") result = node1.query(f"select * from hdfs('hdfs://hdfs1:9000/data.arrow')") assert int(result) == 1 + node1.query(f"truncate table arrow_table") + node1.query(f"drop table arrow_table") def test_schema_inference_with_globs(started_cluster): + fs = HdfsClient(hosts=started_cluster.hdfs_ip) + dir = "/test_schema_inference_with_globs" + fs.mkdirs(dir) node1.query( - f"insert into table function hdfs('hdfs://hdfs1:9000/data1.jsoncompacteachrow', 'JSONCompactEachRow', 'x Nullable(UInt32)') select NULL" + f"insert into table function hdfs('hdfs://hdfs1:9000{dir}/data1.jsoncompacteachrow', 'JSONCompactEachRow', 'x Nullable(UInt32)') select NULL" ) node1.query( - f"insert into table function hdfs('hdfs://hdfs1:9000/data2.jsoncompacteachrow', 'JSONCompactEachRow', 'x Nullable(UInt32)') select 0" + f"insert into table function hdfs('hdfs://hdfs1:9000{dir}/data2.jsoncompacteachrow', 'JSONCompactEachRow', 'x Nullable(UInt32)') select 0" ) result = node1.query( - f"desc hdfs('hdfs://hdfs1:9000/data*.jsoncompacteachrow') settings input_format_json_infer_incomplete_types_as_strings=0" + f"desc hdfs('hdfs://hdfs1:9000{dir}/data*.jsoncompacteachrow') settings input_format_json_infer_incomplete_types_as_strings=0" ) assert result.strip() == "c1\tNullable(Int64)" result = node1.query( - f"select * from hdfs('hdfs://hdfs1:9000/data*.jsoncompacteachrow') settings input_format_json_infer_incomplete_types_as_strings=0" + f"select * from hdfs('hdfs://hdfs1:9000{dir}/data*.jsoncompacteachrow') settings input_format_json_infer_incomplete_types_as_strings=0" ) assert sorted(result.split()) == ["0", "\\N"] node1.query( - f"insert into table function hdfs('hdfs://hdfs1:9000/data3.jsoncompacteachrow', 'JSONCompactEachRow', 'x Nullable(UInt32)') select NULL" + f"insert into table function hdfs('hdfs://hdfs1:9000{dir}/data3.jsoncompacteachrow', 'JSONCompactEachRow', 'x Nullable(UInt32)') select NULL" ) filename = "data{1,3}.jsoncompacteachrow" result = node1.query_and_get_error( - f"desc hdfs('hdfs://hdfs1:9000/{filename}') settings schema_inference_use_cache_for_hdfs=0, input_format_json_infer_incomplete_types_as_strings=0" + f"desc hdfs('hdfs://hdfs1:9000{dir}/{filename}') settings schema_inference_use_cache_for_hdfs=0, input_format_json_infer_incomplete_types_as_strings=0" ) assert "All attempts to extract table structure from files failed" in result node1.query( - f"insert into table function hdfs('hdfs://hdfs1:9000/data0.jsoncompacteachrow', 'TSV', 'x String') select '[123;]'" + f"insert into table function hdfs('hdfs://hdfs1:9000{dir}/data0.jsoncompacteachrow', 'TSV', 'x String') select '[123;]'" ) result = node1.query_and_get_error( - f"desc hdfs('hdfs://hdfs1:9000/data*.jsoncompacteachrow') settings schema_inference_use_cache_for_hdfs=0, input_format_json_infer_incomplete_types_as_strings=0" + f"desc hdfs('hdfs://hdfs1:9000{dir}/data*.jsoncompacteachrow') settings schema_inference_use_cache_for_hdfs=0, input_format_json_infer_incomplete_types_as_strings=0" ) assert "CANNOT_EXTRACT_TABLE_STRUCTURE" in result + fs.delete(dir, recursive=True) def test_insert_select_schema_inference(started_cluster): + fs = HdfsClient(hosts=started_cluster.hdfs_ip) + node1.query( f"insert into table function hdfs('hdfs://hdfs1:9000/test.native.zst') select toUInt64(1) as x" ) @@ -627,6 +688,7 @@ def test_insert_select_schema_inference(started_cluster): result = node1.query(f"select * from hdfs('hdfs://hdfs1:9000/test.native.zst')") assert int(result) == 1 + fs.delete("/test.native.zst") def test_cluster_join(started_cluster): @@ -658,6 +720,7 @@ def test_cluster_macro(started_cluster): def test_virtual_columns_2(started_cluster): hdfs_api = started_cluster.hdfs_api + fs = HdfsClient(hosts=started_cluster.hdfs_ip) table_function = ( f"hdfs('hdfs://hdfs1:9000/parquet_2', 'Parquet', 'a Int32, b String')" @@ -674,6 +737,8 @@ def test_virtual_columns_2(started_cluster): result = node1.query(f"SELECT _path FROM {table_function}") assert result.strip() == "kek" + fs.delete("/parquet_2") + fs.delete("/parquet_3") def check_profile_event_for_query(node, file, profile_event, amount=1): @@ -967,11 +1032,11 @@ def test_read_subcolumns(started_cluster): node = started_cluster.instances["node1"] node.query( - f"insert into function hdfs('hdfs://hdfs1:9000/test_subcolumns.tsv', auto, 'a Tuple(b Tuple(c UInt32, d UInt32), e UInt32)') select ((1, 2), 3)" + f"insert into function hdfs('hdfs://hdfs1:9000/test_subcolumns.tsv', auto, 'a Tuple(b Tuple(c UInt32, d UInt32), e UInt32)') select ((1, 2), 3) settings hdfs_truncate_on_insert=1" ) node.query( - f"insert into function hdfs('hdfs://hdfs1:9000/test_subcolumns.jsonl', auto, 'a Tuple(b Tuple(c UInt32, d UInt32), e UInt32)') select ((1, 2), 3)" + f"insert into function hdfs('hdfs://hdfs1:9000/test_subcolumns.jsonl', auto, 'a Tuple(b Tuple(c UInt32, d UInt32), e UInt32)') select ((1, 2), 3) settings hdfs_truncate_on_insert=1" ) res = node.query( @@ -1003,7 +1068,7 @@ def test_read_subcolumn_time(started_cluster): node = started_cluster.instances["node1"] node.query( - f"insert into function hdfs('hdfs://hdfs1:9000/test_subcolumn_time.tsv', auto, 'a UInt32') select (42)" + f"insert into function hdfs('hdfs://hdfs1:9000/test_subcolumn_time.tsv', auto, 'a UInt32') select (42) settings hdfs_truncate_on_insert=1" ) res = node.query( @@ -1014,91 +1079,103 @@ def test_read_subcolumn_time(started_cluster): def test_union_schema_inference_mode(started_cluster): + id = uuid.uuid4() + fs = HdfsClient(hosts=started_cluster.hdfs_ip) + + dir = f"union_{id}" + fs.mkdirs(f"/{dir}/", permission=777) + node = started_cluster.instances["node1"] node.query( - "insert into function hdfs('hdfs://hdfs1:9000/test_union_schema_inference1.jsonl') select 1 as a" + f"insert into function hdfs('hdfs://hdfs1:9000/{dir}/test_union_schema_inference1.jsonl') select 1 as a" ) node.query( - "insert into function hdfs('hdfs://hdfs1:9000/test_union_schema_inference2.jsonl') select 2 as b" + f"insert into function hdfs('hdfs://hdfs1:9000/{dir}/test_union_schema_inference2.jsonl') select 2 as b" ) node.query("system drop schema cache for hdfs") result = node.query( - "desc hdfs('hdfs://hdfs1:9000/test_union_schema_inference*.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" + f"desc hdfs('hdfs://hdfs1:9000/{dir}/test_union_schema_inference*.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" ) assert result == "a\tNullable(Int64)\nb\tNullable(Int64)\n" result = node.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 '%test_union_schema_inference%' 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" ) result = node.query( - "select * from hdfs('hdfs://hdfs1:9000/test_union_schema_inference*.jsonl') order by tuple(*) settings schema_inference_mode='union', describe_compact_output=1 format TSV" + f"select * from hdfs('hdfs://hdfs1:9000/{dir}/test_union_schema_inference*.jsonl') order by tuple(*) settings schema_inference_mode='union', describe_compact_output=1 format TSV" ) assert result == "1\t\\N\n" "\\N\t2\n" node.query(f"system drop schema cache for hdfs") result = node.query( - "desc hdfs('hdfs://hdfs1:9000/test_union_schema_inference2.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" + f"desc hdfs('hdfs://hdfs1:9000/{dir}/test_union_schema_inference2.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" ) assert result == "b\tNullable(Int64)\n" result = node.query( - "desc hdfs('hdfs://hdfs1:9000/test_union_schema_inference*.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" + f"desc hdfs('hdfs://hdfs1:9000/{dir}/test_union_schema_inference*.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" ) assert result == "a\tNullable(Int64)\n" "b\tNullable(Int64)\n" node.query( - f"insert into function hdfs('hdfs://hdfs1:9000/test_union_schema_inference3.jsonl', TSV) select 'Error'" + f"insert into function hdfs('hdfs://hdfs1:9000/{dir}/test_union_schema_inference3.jsonl', TSV) select 'Error'" ) error = node.query_and_get_error( - "desc hdfs('hdfs://hdfs1:9000/test_union_schema_inference*.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" + f"desc hdfs('hdfs://hdfs1:9000/{dir}/test_union_schema_inference*.jsonl') settings schema_inference_mode='union', describe_compact_output=1 format TSV" ) assert "CANNOT_EXTRACT_TABLE_STRUCTURE" in error def test_format_detection(started_cluster): node = started_cluster.instances["node1"] + fs = HdfsClient(hosts=started_cluster.hdfs_ip) + id = uuid.uuid4() + dir = f"{id}" + fs.mkdirs(f"/{dir}/", permission=777) node.query( - "insert into function hdfs('hdfs://hdfs1:9000/test_format_detection0', JSONEachRow) select number as x, 'str_' || toString(number) as y from numbers(0)" + f"insert into function hdfs('hdfs://hdfs1:9000/{dir}/test_format_detection0', JSONEachRow) select number as x, 'str_' || toString(number) as y from numbers(0)" ) node.query( - "insert into function hdfs('hdfs://hdfs1:9000/test_format_detection1', JSONEachRow) select number as x, 'str_' || toString(number) as y from numbers(10)" + f"insert into function hdfs('hdfs://hdfs1:9000/{dir}/test_format_detection1', JSONEachRow) select number as x, 'str_' || toString(number) as y from numbers(10)" ) expected_desc_result = node.query( - "desc hdfs('hdfs://hdfs1:9000/test_format_detection1', JSONEachRow)" + f"desc hdfs('hdfs://hdfs1:9000/{dir}/test_format_detection1', JSONEachRow)" ) - desc_result = node.query("desc hdfs('hdfs://hdfs1:9000/test_format_detection1')") + desc_result = node.query( + f"desc hdfs('hdfs://hdfs1:9000/{dir}/test_format_detection1')" + ) assert expected_desc_result == desc_result expected_result = node.query( - "select * from hdfs('hdfs://hdfs1:9000/test_format_detection1', JSONEachRow, 'x UInt64, y String') order by x, y" + f"select * from hdfs('hdfs://hdfs1:9000/{dir}/test_format_detection1', JSONEachRow, 'x UInt64, y String') order by x, y" ) result = node.query( - "select * from hdfs('hdfs://hdfs1:9000/test_format_detection1') order by x, y" + f"select * from hdfs('hdfs://hdfs1:9000/{dir}/test_format_detection1') order by x, y" ) assert expected_result == result result = node.query( - "select * from hdfs('hdfs://hdfs1:9000/test_format_detection1', auto, 'x UInt64, y String') order by x, y" + f"select * from hdfs('hdfs://hdfs1:9000/{dir}/test_format_detection1', auto, 'x UInt64, y String') order by x, y" ) assert expected_result == result result = node.query( - "select * from hdfs('hdfs://hdfs1:9000/test_format_detection{0,1}') order by x, y" + f"select * from hdfs('hdfs://hdfs1:9000/{dir}/test_format_detection{{0,1}}') order by x, y" ) assert expected_result == result @@ -1106,25 +1183,25 @@ def test_format_detection(started_cluster): node.query("system drop schema cache for hdfs") result = node.query( - "select * from hdfs('hdfs://hdfs1:9000/test_format_detection{0,1}') order by x, y" + f"select * from hdfs('hdfs://hdfs1:9000/{dir}/test_format_detection{{0,1}}') order by x, y" ) assert expected_result == result result = node.query( - "select * from hdfsCluster(test_cluster_two_shards, 'hdfs://hdfs1:9000/test_format_detection{0,1}') order by x, y" + f"select * from hdfsCluster(test_cluster_two_shards, 'hdfs://hdfs1:9000/{dir}/test_format_detection{{0,1}}') order by x, y" ) assert expected_result == result result = node.query( - "select * from hdfsCluster(test_cluster_two_shards, 'hdfs://hdfs1:9000/test_format_detection{0,1}', auto, auto) order by x, y" + f"select * from hdfsCluster(test_cluster_two_shards, 'hdfs://hdfs1:9000/{dir}/test_format_detection{{0,1}}', auto, auto) order by x, y" ) assert expected_result == result result = node.query( - "select * from hdfsCluster(test_cluster_two_shards, 'hdfs://hdfs1:9000/test_format_detection{0,1}', auto, 'x UInt64, y String') order by x, y" + f"select * from hdfsCluster(test_cluster_two_shards, 'hdfs://hdfs1:9000/{dir}/test_format_detection{{0,1}}', auto, 'x UInt64, y String') order by x, y" ) assert expected_result == result @@ -1180,6 +1257,55 @@ def test_respect_object_existence_on_partitioned_write(started_cluster): assert int(result) == 44 +def test_hive_partitioning_with_one_parameter(started_cluster): + hdfs_api = started_cluster.hdfs_api + hdfs_api.write_data(f"/column0=Elizabeth/parquet_1", f"Elizabeth\tGordon\n") + assert hdfs_api.read_data(f"/column0=Elizabeth/parquet_1") == f"Elizabeth\tGordon\n" + + r = node1.query( + "SELECT _column0 FROM hdfs('hdfs://hdfs1:9000/column0=Elizabeth/parquet_1', 'TSV')", + settings={"use_hive_partitioning": 1}, + ) + assert r == f"Elizabeth\n" + + +def test_hive_partitioning_with_two_parameters(started_cluster): + hdfs_api = started_cluster.hdfs_api + hdfs_api.write_data( + f"/column0=Elizabeth/column1=Gordon/parquet_2", f"Elizabeth\tGordon\n" + ) + assert ( + hdfs_api.read_data(f"/column0=Elizabeth/column1=Gordon/parquet_2") + == f"Elizabeth\tGordon\n" + ) + + r = node1.query( + "SELECT _column1 FROM hdfs('hdfs://hdfs1:9000/column0=Elizabeth/column1=Gordon/parquet_2', 'TSV');", + settings={"use_hive_partitioning": 1}, + ) + assert r == f"Gordon\n" + + +def test_hive_partitioning_without_setting(started_cluster): + hdfs_api = started_cluster.hdfs_api + hdfs_api.write_data( + f"/column0=Elizabeth/column1=Gordon/parquet_2", f"Elizabeth\tGordon\n" + ) + assert ( + hdfs_api.read_data(f"/column0=Elizabeth/column1=Gordon/parquet_2") + == f"Elizabeth\tGordon\n" + ) + pattern = re.compile( + r"DB::Exception: Unknown expression identifier '.*' in scope.*", re.DOTALL + ) + + with pytest.raises(QueryRuntimeException, match=pattern): + node1.query( + f"SELECT _column1 FROM hdfs('hdfs://hdfs1:9000/column0=Elizabeth/column1=Gordon/parquet_2', 'TSV');", + settings={"use_hive_partitioning": 0}, + ) + + if __name__ == "__main__": cluster.start() input("Cluster created, press any key to destroy...") diff --git a/tests/integration/test_storage_kafka/clickhouse_path/format_schemas/string_key_value.capnp b/tests/integration/test_storage_kafka/clickhouse_path/format_schemas/string_key_value.capnp new file mode 100644 index 00000000000..4f3eabe22f0 --- /dev/null +++ b/tests/integration/test_storage_kafka/clickhouse_path/format_schemas/string_key_value.capnp @@ -0,0 +1,6 @@ +@0x99f75f775fe63dae; + +struct StringKeyValuePair { + key@0 : Text; + value@1 : Text; +} diff --git a/tests/integration/test_storage_kafka/clickhouse_path/format_schemas/string_key_value.format b/tests/integration/test_storage_kafka/clickhouse_path/format_schemas/string_key_value.format new file mode 100644 index 00000000000..83dff6ce401 --- /dev/null +++ b/tests/integration/test_storage_kafka/clickhouse_path/format_schemas/string_key_value.format @@ -0,0 +1 @@ +(key = ${key:CSV}, value = ${value:CSV}) diff --git a/tests/integration/test_storage_kafka/clickhouse_path/format_schemas/string_key_value.proto b/tests/integration/test_storage_kafka/clickhouse_path/format_schemas/string_key_value.proto new file mode 100644 index 00000000000..71905c63bdf --- /dev/null +++ b/tests/integration/test_storage_kafka/clickhouse_path/format_schemas/string_key_value.proto @@ -0,0 +1,6 @@ +syntax = "proto3"; + +message StringKeyValuePair { + string key = 1; + string value = 2; +} diff --git a/tests/integration/test_storage_kafka/configs/kafka.xml b/tests/integration/test_storage_kafka/configs/kafka.xml index b10db879b72..a846fdbb295 100644 --- a/tests/integration/test_storage_kafka/configs/kafka.xml +++ b/tests/integration/test_storage_kafka/configs/kafka.xml @@ -48,7 +48,7 @@ - 30001 + 30001 60001 diff --git a/tests/integration/test_storage_kafka/test.py b/tests/integration/test_storage_kafka/test.py index 37457e00701..bef90e1b9d3 100644 --- a/tests/integration/test_storage_kafka/test.py +++ b/tests/integration/test_storage_kafka/test.py @@ -29,6 +29,7 @@ from kafka import KafkaAdminClient, KafkaProducer, KafkaConsumer, BrokerConnecti from kafka.protocol.admin import DescribeGroupsRequest_v1 from kafka.protocol.group import MemberAssignment from kafka.admin import NewTopic +from contextlib import contextmanager # protoc --version @@ -46,6 +47,13 @@ if is_arm(): # TODO: add test for run-time offset update in CH, if we manually update it on Kafka side. # TODO: add test for SELECT LIMIT is working. + +KAFKA_TOPIC_OLD = "old_t" +KAFKA_CONSUMER_GROUP_OLD = "old_cg" +KAFKA_TOPIC_NEW = "new_t" +KAFKA_CONSUMER_GROUP_NEW = "new_cg" + + cluster = ClickHouseCluster(__file__) instance = cluster.add_instance( "instance", @@ -55,10 +63,10 @@ instance = cluster.add_instance( with_zookeeper=True, # For Replicated Table macros={ "kafka_broker": "kafka1", - "kafka_topic_old": "old", - "kafka_group_name_old": "old", - "kafka_topic_new": "new", - "kafka_group_name_new": "new", + "kafka_topic_old": KAFKA_TOPIC_OLD, + "kafka_group_name_old": KAFKA_CONSUMER_GROUP_OLD, + "kafka_topic_new": KAFKA_TOPIC_NEW, + "kafka_group_name_new": KAFKA_CONSUMER_GROUP_NEW, "kafka_client_id": "instance", "kafka_format_json_each_row": "JSONEachRow", }, @@ -142,6 +150,44 @@ def kafka_delete_topic(admin_client, topic, max_retries=50): raise Exception(f"Failed to delete topics {topic}, {result}") +@contextmanager +def kafka_topic( + admin_client, + topic_name, + num_partitions=1, + replication_factor=1, + max_retries=50, + config=None, +): + kafka_create_topic( + admin_client, + topic_name, + num_partitions, + replication_factor, + max_retries, + config, + ) + try: + yield None + finally: + # Code to release resource, e.g.: + kafka_delete_topic(admin_client, topic_name, max_retries) + + +@contextmanager +def existing_kafka_topic(admin_client, topic_name, max_retries=50): + try: + yield None + finally: + kafka_delete_topic(admin_client, topic_name, max_retries) + + +def get_admin_client(kafka_cluster): + return KafkaAdminClient( + bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) + ) + + def kafka_produce(kafka_cluster, topic, messages, timestamp=None, retries=15): logging.debug( "kafka_produce server:{}:{} topic:{}".format( @@ -161,7 +207,7 @@ def kafka_producer_send_heartbeat_msg(max_retries=50): kafka_produce(kafka_cluster, "test_heartbeat_topic", ["test"], retries=max_retries) -def kafka_consume(kafka_cluster, topic, needDecode=True, timestamp=0): +def kafka_consume(kafka_cluster, topic, need_decode=True, timestamp=0): consumer = KafkaConsumer( bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port), auto_offset_reset="earliest", @@ -171,7 +217,7 @@ def kafka_consume(kafka_cluster, topic, needDecode=True, timestamp=0): if toppar.topic == topic: for message in messages: assert timestamp == 0 or message.timestamp / 1000 == timestamp - if needDecode: + if need_decode: yield message.value.decode() else: yield message.value @@ -196,7 +242,33 @@ def kafka_produce_protobuf_messages(kafka_cluster, topic, start_index, num_messa logging.debug(("Produced {} messages for topic {}".format(num_messages, topic))) -def kafka_produce_protobuf_messages_no_delimeters( +def kafka_consume_with_retry( + kafka_cluster, + topic, + expected_messages, + need_decode=True, + timestamp=0, + retry_count=20, + sleep_time=0.1, +): + messages = [] + try_count = 0 + while try_count < retry_count: + try_count += 1 + messages.extend( + kafka_consume( + kafka_cluster, topic, need_decode=need_decode, timestamp=timestamp + ) + ) + if len(messages) == expected_messages: + break + time.sleep(sleep_time) + if len(messages) != expected_messages: + raise Exception(f"Got only {len(messages)} messages") + return messages + + +def kafka_produce_protobuf_messages_no_delimiters( kafka_cluster, topic, start_index, num_messages ): data = "" @@ -284,10 +356,99 @@ def avro_confluent_message(schema_registry_client, value): return serializer.encode_record_with_schema("test_subject", schema, value) +def create_settings_string(settings): + if settings is None: + return "" + + def format_value(value): + if isinstance(value, str): + return f"'{value}'" + elif isinstance(value, bool): + return str(int(value)) + return str(value) + + settings_string = "SETTINGS " + keys = settings.keys() + first_key = next(iter(settings)) + settings_string += str(first_key) + " = " + format_value(settings[first_key]) + for key in keys: + if key == first_key: + continue + settings_string += ", " + str(key) + " = " + format_value(settings[key]) + return settings_string + + +def generate_old_create_table_query( + table_name, + columns_def, + database="test", + brokers="{kafka_broker}:19092", + topic_list="{kafka_topic_new}", + consumer_group="{kafka_group_name_new}", + format="{kafka_format_json_each_row}", + row_delimiter="\\n", + keeper_path=None, # it is not used, but it is easier to handle keeper_path and replica_name like this + replica_name=None, + settings=None, +): + settings_string = create_settings_string(settings) + query = f"""CREATE TABLE {database}.{table_name} ({columns_def}) ENGINE = Kafka('{brokers}', '{topic_list}', '{consumer_group}', '{format}', '{row_delimiter}') +{settings_string}""" + logging.debug(f"Generated old create query: {query}") + return query + + +def generate_new_create_table_query( + table_name, + columns_def, + database="test", + brokers="{kafka_broker}:19092", + topic_list="{kafka_topic_new}", + consumer_group="{kafka_group_name_new}", + format="{kafka_format_json_each_row}", + row_delimiter="\\n", + keeper_path=None, + replica_name=None, + settings=None, +): + if settings is None: + settings = {} + if keeper_path is None: + keeper_path = f"/clickhouse/{{database}}/{table_name}" + if replica_name is None: + replica_name = "r1" + settings["kafka_keeper_path"] = keeper_path + settings["kafka_replica_name"] = replica_name + settings_string = create_settings_string(settings) + query = f"""CREATE TABLE {database}.{table_name} ({columns_def}) ENGINE = Kafka('{brokers}', '{topic_list}', '{consumer_group}', '{format}', '{row_delimiter}') +{settings_string} +SETTINGS allow_experimental_kafka_offsets_storage_in_keeper=1""" + logging.debug(f"Generated new create query: {query}") + return query + + +def must_use_thread_per_consumer(generator): + if generator == generate_old_create_table_query: + return False + if generator == generate_new_create_table_query: + return True + raise Exception("Unexpected generator") + + +def get_topic_postfix(generator): + if generator == generate_old_create_table_query: + return "old" + if generator == generate_new_create_table_query: + return "new" + raise Exception("Unexpected generator") + + # Tests - - -def test_kafka_column_types(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator, do_direct_read", + [(generate_old_create_table_query, True), (generate_new_create_table_query, False)], +) +def test_kafka_column_types(kafka_cluster, create_query_generator, do_direct_read): def assert_returned_exception(e): assert e.value.returncode == 36 assert ( @@ -297,57 +458,14 @@ def test_kafka_column_types(kafka_cluster): # check column with DEFAULT expression with pytest.raises(QueryRuntimeException) as exception: - instance.query( - """ - CREATE TABLE test.kafka (a Int, b Int DEFAULT 0) - ENGINE = Kafka('{kafka_broker}:19092', '{kafka_topic_new}', '{kafka_group_name_new}', '{kafka_format_json_each_row}', '\\n') - """ - ) + instance.query(create_query_generator("kafka", "a Int, b Int DEFAULT 0")) assert_returned_exception(exception) # check EPHEMERAL with pytest.raises(QueryRuntimeException) as exception: - instance.query( - """ - CREATE TABLE test.kafka (a Int, b Int EPHEMERAL) - ENGINE = Kafka('{kafka_broker}:19092', '{kafka_topic_new}', '{kafka_group_name_new}', '{kafka_format_json_each_row}', '\\n') - """ - ) + instance.query(create_query_generator("kafka", "a Int, b Int EPHEMERAL")) assert_returned_exception(exception) - # check ALIAS - instance.query( - """ - CREATE TABLE test.kafka (a Int, b String Alias toString(a)) - ENGINE = Kafka('{kafka_broker}:19092', '{kafka_topic_new}', '{kafka_group_name_new}', '{kafka_format_json_each_row}', '\\n') - SETTINGS kafka_commit_on_select = 1; - """ - ) - messages = [] - for i in range(5): - messages.append(json.dumps({"a": i})) - kafka_produce(kafka_cluster, "new", messages) - result = "" - expected = TSV( - """ -0\t0 -1\t1 -2\t2 -3\t3 -4\t4 - """ - ) - retries = 50 - while retries > 0: - result += instance.query("SELECT a, b FROM test.kafka", ignore_error=True) - if TSV(result) == expected: - break - retries -= 1 - - assert TSV(result) == expected - - instance.query("DROP TABLE test.kafka SYNC") - # check MATERIALIZED with pytest.raises(QueryRuntimeException) as exception: instance.query( @@ -358,6 +476,41 @@ def test_kafka_column_types(kafka_cluster): ) assert_returned_exception(exception) + if do_direct_read: + # check ALIAS + instance.query( + create_query_generator( + "kafka", + "a Int, b String Alias toString(a)", + settings={"kafka_commit_on_select": True}, + ) + ) + messages = [] + for i in range(5): + messages.append(json.dumps({"a": i})) + kafka_produce(kafka_cluster, KAFKA_TOPIC_NEW, messages) + result = "" + expected = TSV( + """ + 0\t0 + 1\t1 + 2\t2 + 3\t3 + 4\t4 + """ + ) + retries = 50 + while retries > 0: + result += instance.query("SELECT a, b FROM test.kafka", ignore_error=True) + if TSV(result) == expected: + break + retries -= 1 + time.sleep(0.5) + + assert TSV(result) == expected + + instance.query("DROP TABLE test.kafka SYNC") + def test_kafka_settings_old_syntax(kafka_cluster): assert TSV( @@ -366,13 +519,13 @@ def test_kafka_settings_old_syntax(kafka_cluster): ignore_error=True, ) ) == TSV( - """kafka_broker kafka1 + f"""kafka_broker kafka1 kafka_client_id instance kafka_format_json_each_row JSONEachRow -kafka_group_name_new new -kafka_group_name_old old -kafka_topic_new new -kafka_topic_old old +kafka_group_name_new {KAFKA_CONSUMER_GROUP_NEW} +kafka_group_name_old {KAFKA_CONSUMER_GROUP_OLD} +kafka_topic_new new_t +kafka_topic_old old_t """ ) @@ -389,7 +542,7 @@ kafka_topic_old old messages = [] for i in range(50): messages.append(json.dumps({"key": i, "value": i})) - kafka_produce(kafka_cluster, "old", messages) + kafka_produce(kafka_cluster, KAFKA_TOPIC_OLD, messages) result = "" while True: @@ -399,7 +552,7 @@ kafka_topic_old old kafka_check_result(result, True) - members = describe_consumer_group(kafka_cluster, "old") + members = describe_consumer_group(kafka_cluster, KAFKA_CONSUMER_GROUP_OLD) assert members[0]["client_id"] == "ClickHouse-instance-test-kafka" # text_desc = kafka_cluster.exec_in_container(kafka_cluster.get_container_id('kafka1'),"kafka-consumer-groups --bootstrap-server localhost:9092 --describe --members --group old --verbose")) @@ -423,16 +576,16 @@ def test_kafka_settings_new_syntax(kafka_cluster): messages = [] for i in range(25): messages.append(json.dumps({"key": i, "value": i})) - kafka_produce(kafka_cluster, "new", messages) + kafka_produce(kafka_cluster, KAFKA_TOPIC_NEW, messages) # Insert couple of malformed messages. - kafka_produce(kafka_cluster, "new", ["}{very_broken_message,"]) - kafka_produce(kafka_cluster, "new", ["}another{very_broken_message,"]) + kafka_produce(kafka_cluster, KAFKA_TOPIC_NEW, ["}{very_broken_message,"]) + kafka_produce(kafka_cluster, KAFKA_TOPIC_NEW, ["}another{very_broken_message,"]) messages = [] for i in range(25, 50): messages.append(json.dumps({"key": i, "value": i})) - kafka_produce(kafka_cluster, "new", messages) + kafka_produce(kafka_cluster, KAFKA_TOPIC_NEW, messages) result = "" while True: @@ -442,7 +595,7 @@ def test_kafka_settings_new_syntax(kafka_cluster): kafka_check_result(result, True) - members = describe_consumer_group(kafka_cluster, "new") + members = describe_consumer_group(kafka_cluster, KAFKA_CONSUMER_GROUP_NEW) assert members[0]["client_id"] == "instance test 1234" @@ -520,12 +673,13 @@ def test_kafka_json_as_string(kafka_cluster): ) -def test_kafka_formats(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_formats(kafka_cluster, create_query_generator): schema_registry_client = CachedSchemaRegistryClient( - "http://localhost:{}".format(kafka_cluster.schema_registry_port) - ) - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) + {"url": f"http://localhost:{kafka_cluster.schema_registry_port}"} ) # data was dumped from clickhouse itself in a following manner @@ -649,7 +803,7 @@ def test_kafka_formats(kafka_cluster): '(id = 1, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)\n(id = 2, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)\n(id = 3, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)\n(id = 4, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)\n(id = 5, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)\n(id = 6, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)\n(id = 7, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)\n(id = 8, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)\n(id = 9, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)\n(id = 10, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)\n(id = 11, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)\n(id = 12, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)\n(id = 13, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)\n(id = 14, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)\n(id = 15, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)', '(id = 0, blockNo = 0, val1 = "AM", val2 = 0.5, val3 = 1)', ], - "extra_settings": ", format_template_row='template_row.format'", + "extra_settings": {"format_template_row": "template_row.format"}, }, "Regexp": { "data_sample": [ @@ -660,7 +814,10 @@ def test_kafka_formats(kafka_cluster): # On empty message exception happens: Line "" doesn't match the regexp.: (at row 1) # /src/Processors/Formats/Impl/RegexpRowInputFormat.cpp:140: DB::RegexpRowInputFormat::readRow(std::__1::vector::mutable_ptr, std::__1::allocator::mutable_ptr > >&, DB::RowReadExtension&) @ 0x1df82fcb in /usr/bin/clickhouse ], - "extra_settings": r", format_regexp='\(id = (.+?), blockNo = (.+?), val1 = \"(.+?)\", val2 = (.+?), val3 = (.+?)\)', format_regexp_escaping_rule='Escaped'", + "extra_settings": { + "format_regexp": r"\(id = (.+?), blockNo = (.+?), val1 = \"(.+?)\", val2 = (.+?), val3 = (.+?)\)", + "format_regexp_escaping_rule": "Escaped", + }, }, ## BINARY FORMATS # dumped with @@ -732,7 +889,7 @@ def test_kafka_formats(kafka_cluster): # /src/Processors/Formats/Impl/ProtobufRowInputFormat.cpp:25: DB::ProtobufRowInputFormat::readRow(std::__1::vector::mutable_ptr, std::__1::allocator::mutable_ptr > >&, DB::RowReadExtension&) @ 0x1df4cc71 in /usr/bin/clickhouse # /src/Processors/Formats/IRowInputFormat.cpp:64: DB::IRowInputFormat::generate() @ 0x1de727cf in /usr/bin/clickhouse ], - "extra_settings": ", kafka_schema='test:TestMessage'", + "extra_settings": {"kafka_schema": "test:TestMessage"}, }, "ORC": { "data_sample": [ @@ -756,7 +913,7 @@ def test_kafka_formats(kafka_cluster): # /src/Processors/Formats/Impl/CapnProtoRowInputFormat.cpp:212: DB::CapnProtoRowInputFormat::readMessage() @ 0x1ded1cab in /usr/bin/clickhouse # /src/Processors/Formats/Impl/CapnProtoRowInputFormat.cpp:241: DB::CapnProtoRowInputFormat::readRow(std::__1::vector::mutable_ptr, std::__1::allocator::mutable_ptr > >&, DB::RowReadExtension&) @ 0x1ded205d in /usr/bin/clickhouse ], - "extra_settings": ", kafka_schema='test:TestRecordStruct'", + "extra_settings": {"kafka_schema": "test:TestRecordStruct"}, }, "Parquet": { "data_sample": [ @@ -791,9 +948,12 @@ def test_kafka_formats(kafka_cluster): {"id": 0, "blockNo": 0, "val1": str("AM"), "val2": 0.5, "val3": 1}, ), ], - "extra_settings": ", format_avro_schema_registry_url='http://{}:{}'".format( - kafka_cluster.schema_registry_host, kafka_cluster.schema_registry_port - ), + "extra_settings": { + "format_avro_schema_registry_url": "http://{}:{}".format( + kafka_cluster.schema_registry_host, + kafka_cluster.schema_registry_port, + ) + }, "supports_empty_value": True, }, "Avro": { @@ -837,40 +997,41 @@ def test_kafka_formats(kafka_cluster): }, } + topic_postfix = str(hash(create_query_generator)) for format_name, format_opts in list(all_formats.items()): - logging.debug(("Set up {}".format(format_name))) - topic_name = "format_tests_{}".format(format_name) + logging.debug(f"Set up {format_name}") + topic_name = f"format_tests_{format_name}-{topic_postfix}" data_sample = format_opts["data_sample"] data_prefix = [] # prepend empty value when supported if format_opts.get("supports_empty_value", False): data_prefix = data_prefix + [""] kafka_produce(kafka_cluster, topic_name, data_prefix + data_sample) + + extra_settings = format_opts.get("extra_settings") or {} + extra_settings["kafka_flush_interval_ms"] = 1000 + instance.query( """ DROP TABLE IF EXISTS test.kafka_{format_name}; - CREATE TABLE test.kafka_{format_name} ( - id Int64, - blockNo UInt16, - val1 String, - val2 Float32, - val3 UInt8 - ) ENGINE = Kafka() - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = '{topic_name}', - kafka_group_name = '{topic_name}_group', - kafka_format = '{format_name}', - kafka_flush_interval_ms = 1000 {extra_settings}; + {create_query}; DROP TABLE IF EXISTS test.kafka_{format_name}_mv; - CREATE MATERIALIZED VIEW test.kafka_{format_name}_mv Engine=Log AS + CREATE MATERIALIZED VIEW test.kafka_{format_name}_mv ENGINE=MergeTree ORDER BY tuple() AS SELECT *, _topic, _partition, _offset FROM test.kafka_{format_name}; """.format( topic_name=topic_name, format_name=format_name, - extra_settings=format_opts.get("extra_settings") or "", + create_query=create_query_generator( + f"kafka_{format_name}", + "id Int64, blockNo UInt16, val1 String, val2 Float32, val3 UInt8", + topic_list=f"{topic_name}", + consumer_group=f"{topic_name}_group", + format=format_name, + settings=extra_settings, + ), ) ) raw_expected = """\ @@ -905,13 +1066,16 @@ def test_kafka_formats(kafka_cluster): for format_name, format_opts in list(all_formats.items()): logging.debug(("Checking {}".format(format_name))) - topic_name = f"format_tests_{format_name}" + topic_name = f"format_tests_{format_name}-{topic_postfix}" # shift offsets by 1 if format supports empty value offsets = ( [1, 2, 3] if format_opts.get("supports_empty_value", False) else [0, 1, 2] ) - result = instance.query( - "SELECT * FROM test.kafka_{format_name}_mv;".format(format_name=format_name) + result = instance.query_with_retry( + "SELECT * FROM test.kafka_{format_name}_mv;".format( + format_name=format_name + ), + check_callback=lambda x: x.count("\n") == raw_expected.count("\n"), ) expected = raw_expected.format( topic_name=topic_name, @@ -922,7 +1086,7 @@ def test_kafka_formats(kafka_cluster): assert TSV(result) == TSV(expected), "Proper result for format: {}".format( format_name ) - kafka_delete_topic(admin_client, topic_name) + kafka_delete_topic(get_admin_client(kafka_cluster), topic_name) # Since everything is async and shaky when receiving messages from Kafka, @@ -997,8 +1161,29 @@ def kafka_cluster(): @pytest.fixture(autouse=True) def kafka_setup_teardown(): - instance.query("DROP DATABASE IF EXISTS test; CREATE DATABASE test;") - # logging.debug("kafka is available - running test") + instance.query("DROP DATABASE IF EXISTS test SYNC; CREATE DATABASE test;") + admin_client = get_admin_client(cluster) + + def get_topics_to_delete(): + return [t for t in admin_client.list_topics() if not t.startswith("_")] + + topics = get_topics_to_delete() + logging.debug(f"Deleting topics: {topics}") + result = admin_client.delete_topics(topics) + for topic, error in result.topic_error_codes: + if error != 0: + logging.warning(f"Received error {error} while deleting topic {topic}") + else: + logging.info(f"Deleted topic {topic}") + + retries = 0 + topics = get_topics_to_delete() + while len(topics) != 0: + logging.info(f"Existing topics: {topics}") + if retries >= 5: + raise Exception(f"Failed to delete topics {topics}") + retries += 1 + time.sleep(0.5) yield # run test @@ -1098,9 +1283,7 @@ def test_kafka_issue4116(kafka_cluster): def test_kafka_consumer_hang(kafka_cluster): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) + admin_client = get_admin_client(kafka_cluster) topic_name = "consumer_hang" kafka_create_topic(admin_client, topic_name, num_partitions=8) @@ -1180,9 +1363,7 @@ def test_kafka_consumer_hang(kafka_cluster): def test_kafka_consumer_hang2(kafka_cluster): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) + admin_client = get_admin_client(kafka_cluster) topic_name = "consumer_hang2" kafka_create_topic(admin_client, topic_name) @@ -1243,9 +1424,7 @@ def test_kafka_consumer_hang2(kafka_cluster): # sequential read from different consumers leads to breaking lot of kafka invariants # (first consumer will get all partitions initially, and may have problems in doing polls every 60 sec) def test_kafka_read_consumers_in_parallel(kafka_cluster): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) + admin_client = get_admin_client(kafka_cluster) topic_name = "read_consumers_in_parallel" kafka_create_topic(admin_client, topic_name, num_partitions=8) @@ -1365,9 +1544,7 @@ def test_kafka_tsv_with_delimiter(kafka_cluster): def test_kafka_select_empty(kafka_cluster): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) + admin_client = get_admin_client(kafka_cluster) topic_name = "empty" kafka_create_topic(admin_client, topic_name) @@ -1545,13 +1722,13 @@ def test_kafka_protobuf_no_delimiter(kafka_cluster): """ ) - kafka_produce_protobuf_messages_no_delimeters( + kafka_produce_protobuf_messages_no_delimiters( kafka_cluster, "pb_no_delimiter", 0, 20 ) - kafka_produce_protobuf_messages_no_delimeters( + kafka_produce_protobuf_messages_no_delimiters( kafka_cluster, "pb_no_delimiter", 20, 1 ) - kafka_produce_protobuf_messages_no_delimeters( + kafka_produce_protobuf_messages_no_delimiters( kafka_cluster, "pb_no_delimiter", 21, 29 ) @@ -1592,18 +1769,20 @@ def test_kafka_protobuf_no_delimiter(kafka_cluster): assert TSV(result) == TSV(expected) -def test_kafka_materialized_view(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_materialized_view(kafka_cluster, create_query_generator): + topic_name = "mv" + instance.query( - """ + f""" DROP TABLE IF EXISTS test.view; DROP TABLE IF EXISTS test.consumer; - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'mv', - kafka_group_name = 'mv', - kafka_format = 'JSONEachRow', - kafka_row_delimiter = '\\n'; + DROP TABLE IF EXISTS test.kafka; + + {create_query_generator("kafka", "key UInt64, value UInt64", topic_list=topic_name, consumer_group="mv")}; CREATE TABLE test.view (key UInt64, value UInt64) ENGINE = MergeTree() ORDER BY key; @@ -1615,110 +1794,124 @@ def test_kafka_materialized_view(kafka_cluster): messages = [] for i in range(50): messages.append(json.dumps({"key": i, "value": i})) - kafka_produce(kafka_cluster, "mv", messages) + kafka_produce(kafka_cluster, topic_name, messages) - while True: - result = instance.query("SELECT * FROM test.view") - if kafka_check_result(result): - break + with existing_kafka_topic(get_admin_client(kafka_cluster), topic_name): + result = instance.query_with_retry( + "SELECT * FROM test.view", check_callback=kafka_check_result + ) - instance.query( + kafka_check_result(result, True) + + instance.query( + """ + DROP TABLE test.consumer; + DROP TABLE test.view; + DROP TABLE test.kafka; """ - DROP TABLE test.consumer; - DROP TABLE test.view; - """ - ) - - kafka_check_result(result, True) + ) -def test_kafka_recreate_kafka_table(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator, log_line", + [ + ( + generate_new_create_table_query, + r"kafka.*Saved offset [0-9]+ for topic-partition \[recreate_kafka_table:[0-9]+", + ), + ( + generate_old_create_table_query, + "kafka.*Committed offset [0-9]+.*recreate_kafka_table", + ), + ], +) +def test_kafka_recreate_kafka_table(kafka_cluster, create_query_generator, log_line): """ Checks that materialized view work properly after dropping and recreating the Kafka table. """ - # line for backporting: - # admin_client = KafkaAdminClient(bootstrap_servers="localhost:9092") - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) - topic_name = "recreate_kafka_table" - kafka_create_topic(admin_client, topic_name, num_partitions=6) + thread_per_consumer = must_use_thread_per_consumer(create_query_generator) - instance.query( + with kafka_topic(get_admin_client(kafka_cluster), topic_name, num_partitions=6): + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group="recreate_kafka_table_group", + settings={ + "kafka_num_consumers": 4, + "kafka_flush_interval_ms": 1000, + "kafka_skip_broken_messages": 1048577, + "kafka_thread_per_consumer": thread_per_consumer, + }, + ) + + instance.query( + f""" + DROP TABLE IF EXISTS test.view; + DROP TABLE IF EXISTS test.consumer; + {create_query}; + + CREATE TABLE test.view (key UInt64, value UInt64) + ENGINE = MergeTree() + ORDER BY key; + CREATE MATERIALIZED VIEW test.consumer TO test.view AS + SELECT * FROM test.kafka; """ - DROP TABLE IF EXISTS test.view; - DROP TABLE IF EXISTS test.consumer; - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'recreate_kafka_table', - kafka_group_name = 'recreate_kafka_table_group', - kafka_format = 'JSONEachRow', - kafka_num_consumers = 6, - kafka_flush_interval_ms = 1000, - kafka_skip_broken_messages = 1048577; + ) - CREATE TABLE test.view (key UInt64, value UInt64) - ENGINE = MergeTree() - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.kafka; - """ - ) + messages = [] + for i in range(120): + messages.append(json.dumps({"key": i, "value": i})) + kafka_produce(kafka_cluster, "recreate_kafka_table", messages) - messages = [] - for i in range(120): - messages.append(json.dumps({"key": i, "value": i})) - kafka_produce(kafka_cluster, "recreate_kafka_table", messages) + instance.wait_for_log_line( + log_line, + repetitions=6, + look_behind_lines=100, + ) - instance.wait_for_log_line( - "kafka.*Committed offset [0-9]+.*recreate_kafka_table", - repetitions=6, - look_behind_lines=100, - ) - - instance.query( + instance.query( + """ + DROP TABLE test.kafka; """ - DROP TABLE test.kafka; - """ - ) + ) - kafka_produce(kafka_cluster, "recreate_kafka_table", messages) + instance.rotate_logs() - instance.query( + kafka_produce(kafka_cluster, "recreate_kafka_table", messages) + + instance.query(create_query) + + instance.wait_for_log_line( + log_line, + repetitions=6, + look_behind_lines=100, + ) + + # data was not flushed yet (it will be flushed 7.5 sec after creating MV) + assert int(instance.query("SELECT count() FROM test.view")) == 240 + + instance.query( + """ + DROP TABLE test.consumer; + DROP TABLE test.kafka; + DROP TABLE test.view; """ - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'recreate_kafka_table', - kafka_group_name = 'recreate_kafka_table_group', - kafka_format = 'JSONEachRow', - kafka_num_consumers = 6, - kafka_flush_interval_ms = 1000, - kafka_skip_broken_messages = 1048577; - """ - ) - - instance.wait_for_log_line( - "kafka.*Committed offset [0-9]+.*recreate_kafka_table", - repetitions=6, - look_behind_lines=100, - ) - - # data was not flushed yet (it will be flushed 7.5 sec after creating MV) - assert int(instance.query("SELECT count() FROM test.view")) == 240 - - instance.query( - """ - DROP TABLE test.consumer; - DROP TABLE test.view; - """ - ) - kafka_delete_topic(admin_client, topic_name) + ) -def test_librdkafka_compression(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator, log_line", + [ + (generate_old_create_table_query, "Committed offset {offset}"), + ( + generate_new_create_table_query, + r"kafka.*Saved offset [0-9]+ for topic-partition \[{topic}:[0-9]+\]", + ), + ], +) +def test_librdkafka_compression(kafka_cluster, create_query_generator, log_line): """ Regression for UB in snappy-c (that is used in librdkafka), backport pr is [1]. @@ -1754,58 +1947,68 @@ def test_librdkafka_compression(kafka_cluster): expected = "\n".join(expected) + admin_client = get_admin_client(kafka_cluster) + for compression_type in supported_compression_types: logging.debug(("Check compression {}".format(compression_type))) topic_name = "test_librdkafka_compression_{}".format(compression_type) - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) + topic_config = {"compression.type": compression_type} + with kafka_topic(admin_client, topic_name, config=topic_config): + instance.query( + """{create_query}; - kafka_create_topic( - admin_client, topic_name, config={"compression.type": compression_type} - ) + CREATE TABLE test.view (key UInt64, value String) + ENGINE = MergeTree() + ORDER BY key; - instance.query( - """ - CREATE TABLE test.kafka (key UInt64, value String) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = '{topic_name}', - kafka_group_name = '{topic_name}_group', - kafka_format = 'JSONEachRow', - kafka_flush_interval_ms = 1000; - CREATE MATERIALIZED VIEW test.consumer Engine=Log AS - SELECT * FROM test.kafka; - """.format( - topic_name=topic_name + CREATE MATERIALIZED VIEW test.consumer TO test.view AS + SELECT * FROM test.kafka; + """.format( + create_query=create_query_generator( + "kafka", + "key UInt64, value String", + topic_list=topic_name, + format="JSONEachRow", + settings={"kafka_flush_interval_ms": 1000}, + ), + ) ) - ) - kafka_produce(kafka_cluster, topic_name, messages) + kafka_produce(kafka_cluster, topic_name, messages) - instance.wait_for_log_line("Committed offset {}".format(number_of_messages)) + instance.wait_for_log_line( + log_line.format(offset=number_of_messages, topic=topic_name) + ) + result = instance.query("SELECT * FROM test.view") + assert TSV(result) == TSV(expected) - result = instance.query("SELECT * FROM test.consumer") - assert TSV(result) == TSV(expected) - - instance.query("DROP TABLE test.kafka SYNC") - instance.query("DROP TABLE test.consumer SYNC") - kafka_delete_topic(admin_client, topic_name) + instance.query("DROP TABLE test.kafka SYNC") + instance.query("DROP TABLE test.consumer SYNC") + instance.query("DROP TABLE test.view SYNC") -def test_kafka_materialized_view_with_subquery(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator", + [generate_new_create_table_query, generate_old_create_table_query], +) +def test_kafka_materialized_view_with_subquery(kafka_cluster, create_query_generator): + topic_name = "mysq" + logging.debug(f"Using topic {topic_name}") + + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + ) instance.query( - """ + f""" + DROP TABLE IF EXISTS test.kafka; DROP TABLE IF EXISTS test.view; DROP TABLE IF EXISTS test.consumer; - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'mvsq', - kafka_group_name = 'mvsq', - kafka_format = 'JSONEachRow', - kafka_row_delimiter = '\\n'; + + {create_query}; CREATE TABLE test.view (key UInt64, value UInt64) ENGINE = MergeTree() ORDER BY key; @@ -1817,37 +2020,46 @@ def test_kafka_materialized_view_with_subquery(kafka_cluster): messages = [] for i in range(50): messages.append(json.dumps({"key": i, "value": i})) - kafka_produce(kafka_cluster, "mvsq", messages) + kafka_produce(kafka_cluster, topic_name, messages) - while True: - result = instance.query("SELECT * FROM test.view") - if kafka_check_result(result): - break + with existing_kafka_topic(get_admin_client(kafka_cluster), topic_name): + result = instance.query_with_retry( + "SELECT * FROM test.view", + check_callback=kafka_check_result, + retry_count=40, + sleep_time=0.75, + ) - instance.query( + instance.query( + """ + DROP TABLE test.consumer; + DROP TABLE test.view; """ - DROP TABLE test.consumer; - DROP TABLE test.view; - """ + ) + + kafka_check_result(result, True) + + +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_many_materialized_views(kafka_cluster, create_query_generator): + topic_name = f"mmv-{get_topic_postfix(create_query_generator)}" + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=f"{topic_name}-group", ) - kafka_check_result(result, True) - - -def test_kafka_many_materialized_views(kafka_cluster): instance.query( - """ + f""" DROP TABLE IF EXISTS test.view1; DROP TABLE IF EXISTS test.view2; DROP TABLE IF EXISTS test.consumer1; DROP TABLE IF EXISTS test.consumer2; - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'mmv', - kafka_group_name = 'mmv', - kafka_format = 'JSONEachRow', - kafka_row_delimiter = '\\n'; + {create_query}; CREATE TABLE test.view1 (key UInt64, value UInt64) ENGINE = MergeTree() ORDER BY key; @@ -1864,237 +2076,258 @@ def test_kafka_many_materialized_views(kafka_cluster): messages = [] for i in range(50): messages.append(json.dumps({"key": i, "value": i})) - kafka_produce(kafka_cluster, "mmv", messages) + kafka_produce(kafka_cluster, topic_name, messages) - while True: - result1 = instance.query("SELECT * FROM test.view1") - result2 = instance.query("SELECT * FROM test.view2") - if kafka_check_result(result1) and kafka_check_result(result2): - break + with existing_kafka_topic(get_admin_client(kafka_cluster), topic_name): + result1 = instance.query_with_retry( + "SELECT * FROM test.view1", check_callback=kafka_check_result + ) + result2 = instance.query_with_retry( + "SELECT * FROM test.view2", check_callback=kafka_check_result + ) - instance.query( + instance.query( + """ + DROP TABLE test.consumer1; + DROP TABLE test.consumer2; + DROP TABLE test.view1; + DROP TABLE test.view2; """ - DROP TABLE test.consumer1; - DROP TABLE test.consumer2; - DROP TABLE test.view1; - DROP TABLE test.view2; - """ - ) + ) - kafka_check_result(result1, True) - kafka_check_result(result2, True) + kafka_check_result(result1, True) + kafka_check_result(result2, True) -def test_kafka_flush_on_big_message(kafka_cluster): - # Create batchs of messages of size ~100Kb +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_flush_on_big_message(kafka_cluster, create_query_generator): + # Create batches of messages of size ~100Kb kafka_messages = 1000 batch_messages = 1000 + topic_name = "flush" + get_topic_postfix(create_query_generator) messages = [ json.dumps({"key": i, "value": "x" * 100}) * batch_messages for i in range(kafka_messages) ] - kafka_produce(kafka_cluster, "flush", messages) + kafka_produce(kafka_cluster, topic_name, messages) - instance.query( + admin_client = get_admin_client(kafka_cluster) + + with existing_kafka_topic(admin_client, topic_name): + create_query = create_query_generator( + "kafka", + "key UInt64, value String", + topic_list=topic_name, + consumer_group=topic_name, + settings={"kafka_max_block_size": 10}, + ) + + instance.query( + f""" + DROP TABLE IF EXISTS test.view; + DROP TABLE IF EXISTS test.consumer; + {create_query}; + CREATE TABLE test.view (key UInt64, value String) + ENGINE = MergeTree + ORDER BY key; + CREATE MATERIALIZED VIEW test.consumer TO test.view AS + SELECT * FROM test.kafka; """ - DROP TABLE IF EXISTS test.view; - DROP TABLE IF EXISTS test.consumer; - CREATE TABLE test.kafka (key UInt64, value String) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'flush', - kafka_group_name = 'flush', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 10; - CREATE TABLE test.view (key UInt64, value String) - ENGINE = MergeTree - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.kafka; - """ - ) + ) - client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) - received = False - while not received: - try: - offsets = client.list_consumer_group_offsets("flush") - for topic, offset in list(offsets.items()): - if topic.topic == "flush" and offset.offset == kafka_messages: - received = True - break - except kafka.errors.GroupCoordinatorNotAvailableError: - continue + received = False + while not received: + try: + offsets = admin_client.list_consumer_group_offsets(topic_name) + for topic, offset in list(offsets.items()): + if topic.topic == topic_name and offset.offset == kafka_messages: + received = True + break + except kafka.errors.GroupCoordinatorNotAvailableError: + continue - while True: - result = instance.query("SELECT count() FROM test.view") - if int(result) == kafka_messages * batch_messages: - break + while True: + result = instance.query("SELECT count() FROM test.view") + if int(result) == kafka_messages * batch_messages: + break - instance.query( + instance.query( + """ + DROP TABLE test.consumer; + DROP TABLE test.view; """ - DROP TABLE test.consumer; - DROP TABLE test.view; - """ - ) + ) - assert ( - int(result) == kafka_messages * batch_messages - ), "ClickHouse lost some messages: {}".format(result) + assert ( + int(result) == kafka_messages * batch_messages + ), "ClickHouse lost some messages: {}".format(result) def test_kafka_virtual_columns(kafka_cluster): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) topic_config = { # default retention, since predefined timestamp_ms is used. "retention.ms": "-1", } - kafka_create_topic(admin_client, "virt1", config=topic_config) - - instance.query( - """ - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'virt1', - kafka_group_name = 'virt1', - kafka_commit_on_select = 1, - kafka_format = 'JSONEachRow'; - """ - ) - - messages = "" - for i in range(25): - messages += json.dumps({"key": i, "value": i}) + "\n" - kafka_produce(kafka_cluster, "virt1", [messages], 0) - - messages = "" - for i in range(25, 50): - messages += json.dumps({"key": i, "value": i}) + "\n" - kafka_produce(kafka_cluster, "virt1", [messages], 0) - - result = "" - while True: - result += instance.query( - """SELECT _key, key, _topic, value, _offset, _partition, _timestamp = 0 ? '0000-00-00 00:00:00' : toString(_timestamp) AS _timestamp FROM test.kafka""", - ignore_error=True, + with kafka_topic(get_admin_client(kafka_cluster), "virt1", config=topic_config): + instance.query( + """ + CREATE TABLE test.kafka (key UInt64, value UInt64) + ENGINE = Kafka + SETTINGS kafka_broker_list = 'kafka1:19092', + kafka_topic_list = 'virt1', + kafka_group_name = 'virt1', + kafka_commit_on_select = 1, + kafka_format = 'JSONEachRow'; + """ ) - if kafka_check_result(result, False, "test_kafka_virtual1.reference"): - break - kafka_check_result(result, True, "test_kafka_virtual1.reference") + messages = "" + for i in range(25): + messages += json.dumps({"key": i, "value": i}) + "\n" + kafka_produce(kafka_cluster, "virt1", [messages], 0) + + messages = "" + for i in range(25, 50): + messages += json.dumps({"key": i, "value": i}) + "\n" + kafka_produce(kafka_cluster, "virt1", [messages], 0) + + result = "" + while True: + result += instance.query( + """SELECT _key, key, _topic, value, _offset, _partition, _timestamp = 0 ? '0000-00-00 00:00:00' : toString(_timestamp) AS _timestamp FROM test.kafka""", + ignore_error=True, + ) + if kafka_check_result(result, False, "test_kafka_virtual1.reference"): + break + + kafka_check_result(result, True, "test_kafka_virtual1.reference") -def test_kafka_virtual_columns_with_materialized_view(kafka_cluster): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_virtual_columns_with_materialized_view( + kafka_cluster, create_query_generator +): topic_config = { # default retention, since predefined timestamp_ms is used. "retention.ms": "-1", } - kafka_create_topic(admin_client, "virt2", config=topic_config) - - instance.query( - """ - DROP TABLE IF EXISTS test.view; - DROP TABLE IF EXISTS test.consumer; - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'virt2', - kafka_group_name = 'virt2', - kafka_format = 'JSONEachRow', - kafka_row_delimiter = '\\n'; - CREATE TABLE test.view (key UInt64, value UInt64, kafka_key String, topic String, offset UInt64, partition UInt64, timestamp Nullable(DateTime('UTC'))) - ENGINE = MergeTree() - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT *, _key as kafka_key, _topic as topic, _offset as offset, _partition as partition, _timestamp = 0 ? '0000-00-00 00:00:00' : toString(_timestamp) as timestamp FROM test.kafka; - """ + # the topic name is hardcoded in reference, it doesn't worth to create two reference files to have separate topics, + # as the context manager will always clean up the topic + topic_name = "virt2" + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=f"{topic_name}-group", ) - - messages = [] - for i in range(50): - messages.append(json.dumps({"key": i, "value": i})) - kafka_produce(kafka_cluster, "virt2", messages, 0) - - sql = "SELECT kafka_key, key, topic, value, offset, partition, timestamp FROM test.view ORDER BY kafka_key, key" - result = instance.query(sql) - iterations = 0 - while ( - not kafka_check_result(result, False, "test_kafka_virtual2.reference") - and iterations < 10 - ): - time.sleep(3) - iterations += 1 - result = instance.query(sql) - - kafka_check_result(result, True, "test_kafka_virtual2.reference") - - instance.query( + with kafka_topic(get_admin_client(kafka_cluster), topic_name, config=topic_config): + instance.query( + f""" + DROP TABLE IF EXISTS test.view; + DROP TABLE IF EXISTS test.consumer; + {create_query}; + CREATE TABLE test.view (key UInt64, value UInt64, kafka_key String, topic String, offset UInt64, partition UInt64, timestamp Nullable(DateTime('UTC'))) + ENGINE = MergeTree() + ORDER BY key; + CREATE MATERIALIZED VIEW test.consumer TO test.view AS + SELECT *, _key as kafka_key, _topic as topic, _offset as offset, _partition as partition, _timestamp = 0 ? '0000-00-00 00:00:00' : toString(_timestamp) as timestamp FROM test.kafka; """ - DROP TABLE test.consumer; - DROP TABLE test.view; - """ - ) + ) + messages = [] + for i in range(50): + messages.append(json.dumps({"key": i, "value": i})) + kafka_produce(kafka_cluster, topic_name, messages, 0) -def test_kafka_insert(kafka_cluster): - instance.query( + def check_callback(result): + return kafka_check_result(result, False, "test_kafka_virtual2.reference") + + result = instance.query_with_retry( + "SELECT kafka_key, key, topic, value, offset, partition, timestamp FROM test.view ORDER BY kafka_key, key", + check_callback=check_callback, + ) + + kafka_check_result(result, True, "test_kafka_virtual2.reference") + + instance.query( + """ + DROP TABLE test.consumer; + DROP TABLE test.view; """ - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'insert1', - kafka_group_name = 'insert1', - kafka_format = 'TSV', - kafka_commit_on_select = 1, - kafka_row_delimiter = '\\n'; - """ - ) + ) - values = [] - for i in range(50): - values.append("({i}, {i})".format(i=i)) - values = ",".join(values) +def insert_with_retry(instance, values, table_name="kafka", max_try_count=5): + try_count = 0 while True: + logging.debug(f"Inserting, try_count is {try_count}") try: - instance.query("INSERT INTO test.kafka VALUES {}".format(values)) + try_count += 1 + instance.query(f"INSERT INTO test.{table_name} VALUES {values}") break except QueryRuntimeException as e: - if "Local: Timed out." in str(e): + if "Local: Timed out." in str(e) and try_count < max_try_count: continue else: raise - messages = [] - while True: - messages.extend(kafka_consume(kafka_cluster, "insert1")) - if len(messages) == 50: - break - result = "\n".join(messages) - kafka_check_result(result, True) +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_insert(kafka_cluster, create_query_generator): + topic_name = "insert1" + get_topic_postfix(create_query_generator) - -def test_kafka_produce_consume(kafka_cluster): instance.query( - """ + create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + format="TSV", + ) + ) + + message_count = 50 + values = [] + for i in range(message_count): + values.append("({i}, {i})".format(i=i)) + values = ",".join(values) + + with existing_kafka_topic(get_admin_client(kafka_cluster), topic_name): + insert_with_retry(instance, values) + + messages = kafka_consume_with_retry(kafka_cluster, topic_name, message_count) + result = "\n".join(messages) + kafka_check_result(result, True) + + +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_produce_consume(kafka_cluster, create_query_generator): + topic_name = "insert2" + get_topic_postfix(create_query_generator) + + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + format="TSV", + ) + instance.query( + f""" DROP TABLE IF EXISTS test.view; DROP TABLE IF EXISTS test.consumer; - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'insert2', - kafka_group_name = 'insert2', - kafka_format = 'TSV', - kafka_row_delimiter = '\\n'; + {create_query}; CREATE TABLE test.view (key UInt64, value UInt64) ENGINE = MergeTree ORDER BY key; @@ -2111,15 +2344,7 @@ def test_kafka_produce_consume(kafka_cluster): values.append("({i}, {i})".format(i=i)) values = ",".join(values) - while True: - try: - instance.query("INSERT INTO test.kafka VALUES {}".format(values)) - break - except QueryRuntimeException as e: - if "Local: Timed out." in str(e): - continue - else: - raise + insert_with_retry(instance, values) threads = [] threads_num = 16 @@ -2129,40 +2354,48 @@ def test_kafka_produce_consume(kafka_cluster): time.sleep(random.uniform(0, 1)) thread.start() - while True: - result = instance.query("SELECT count() FROM test.view") - time.sleep(1) - if int(result) == messages_num * threads_num: - break + with existing_kafka_topic(get_admin_client(kafka_cluster), topic_name): + expected_row_count = messages_num * threads_num + result = instance.query_with_retry( + "SELECT count() FROM test.view", + sleep_time=1, + retry_count=20, + check_callback=lambda result: int(result) == expected_row_count, + ) - instance.query( + instance.query( + """ + DROP TABLE test.consumer; + DROP TABLE test.view; """ - DROP TABLE test.consumer; - DROP TABLE test.view; - """ + ) + + for thread in threads: + thread.join() + + assert ( + int(result) == expected_row_count + ), "ClickHouse lost some messages: {}".format(result) + + +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_commit_on_block_write(kafka_cluster, create_query_generator): + topic_name = "block" + get_topic_postfix(create_query_generator) + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + settings={"kafka_max_block_size": 100}, ) - - for thread in threads: - thread.join() - - assert ( - int(result) == messages_num * threads_num - ), "ClickHouse lost some messages: {}".format(result) - - -def test_kafka_commit_on_block_write(kafka_cluster): instance.query( - """ + f""" DROP TABLE IF EXISTS test.view; DROP TABLE IF EXISTS test.consumer; - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'block', - kafka_group_name = 'block', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 100, - kafka_row_delimiter = '\\n'; + {create_query}; CREATE TABLE test.view (key UInt64, value UInt64) ENGINE = MergeTree() ORDER BY key; @@ -2173,45 +2406,39 @@ def test_kafka_commit_on_block_write(kafka_cluster): cancel = threading.Event() + # We need to pass i as a reference. Simple integers are passed by value. + # Making an array is probably the easiest way to "force pass by reference". i = [0] - def produce(): + def produce(i): while not cancel.is_set(): messages = [] for _ in range(101): messages.append(json.dumps({"key": i[0], "value": i[0]})) i[0] += 1 - kafka_produce(kafka_cluster, "block", messages) + kafka_produce(kafka_cluster, topic_name, messages) - kafka_thread = threading.Thread(target=produce) + kafka_thread = threading.Thread(target=produce, args=[i]) kafka_thread.start() - while int(instance.query("SELECT count() FROM test.view")) == 0: - time.sleep(1) + instance.query_with_retry( + "SELECT count() FROM test.view", + sleep_time=1, + check_callback=lambda res: int(res) >= 100, + ) cancel.set() - instance.query( - """ - DROP TABLE test.kafka; - """ - ) + instance.query("DROP TABLE test.kafka SYNC") - instance.query( - """ - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'block', - kafka_group_name = 'block', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 100, - kafka_row_delimiter = '\\n'; - """ - ) + instance.query(create_query) + kafka_thread.join() - while int(instance.query("SELECT uniqExact(key) FROM test.view")) < i[0]: - time.sleep(1) + instance.query_with_retry( + "SELECT uniqExact(key) FROM test.view", + sleep_time=1, + check_callback=lambda res: int(res) >= i[0], + ) result = int(instance.query("SELECT count() == uniqExact(key) FROM test.view")) @@ -2227,163 +2454,183 @@ def test_kafka_commit_on_block_write(kafka_cluster): assert result == 1, "Messages from kafka get duplicated!" -def test_kafka_virtual_columns2(kafka_cluster): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) +@pytest.mark.parametrize( + "create_query_generator, log_line", + [ + (generate_old_create_table_query, "kafka.*Committed offset 2.*virt2_[01]"), + ( + generate_new_create_table_query, + r"kafka.*Saved offset 2 for topic-partition \[virt2_[01]:[0-9]+", + ), + ], +) +def test_kafka_virtual_columns2(kafka_cluster, create_query_generator, log_line): + admin_client = get_admin_client(kafka_cluster) topic_config = { # default retention, since predefined timestamp_ms is used. "retention.ms": "-1", } - kafka_create_topic(admin_client, "virt2_0", num_partitions=2, config=topic_config) - kafka_create_topic(admin_client, "virt2_1", num_partitions=2, config=topic_config) + thread_per_consumer = must_use_thread_per_consumer(create_query_generator) + topic_name_0 = "virt2_0" + topic_name_1 = "virt2_1" + consumer_group = "virt2" + get_topic_postfix(create_query_generator) + with kafka_topic(admin_client, topic_name_0, num_partitions=2, config=topic_config): + with kafka_topic( + admin_client, topic_name_1, num_partitions=2, config=topic_config + ): + create_query = create_query_generator( + "kafka", + "value UInt64", + topic_list=f"{topic_name_0},{topic_name_1}", + consumer_group=consumer_group, + settings={ + "kafka_num_consumers": 2, + "kafka_thread_per_consumer": thread_per_consumer, + }, + ) - instance.query( + instance.query( + f""" + {create_query}; + + CREATE MATERIALIZED VIEW test.view ENGINE=MergeTree ORDER BY tuple() AS + SELECT value, _key, _topic, _partition, _offset, toUnixTimestamp(_timestamp), toUnixTimestamp64Milli(_timestamp_ms), _headers.name, _headers.value FROM test.kafka; + """ + ) + + producer = KafkaProducer( + bootstrap_servers="localhost:{}".format(cluster.kafka_port), + value_serializer=producer_serializer, + key_serializer=producer_serializer, + ) + + producer.send( + topic=topic_name_0, + value=json.dumps({"value": 1}), + partition=0, + key="k1", + timestamp_ms=1577836801001, + headers=[("content-encoding", b"base64")], + ) + producer.send( + topic=topic_name_0, + value=json.dumps({"value": 2}), + partition=0, + key="k2", + timestamp_ms=1577836802002, + headers=[ + ("empty_value", b""), + ("", b"empty name"), + ("", b""), + ("repetition", b"1"), + ("repetition", b"2"), + ], + ) + producer.flush() + + producer.send( + topic=topic_name_0, + value=json.dumps({"value": 3}), + partition=1, + key="k3", + timestamp_ms=1577836803003, + headers=[("b", b"b"), ("a", b"a")], + ) + producer.send( + topic=topic_name_0, + value=json.dumps({"value": 4}), + partition=1, + key="k4", + timestamp_ms=1577836804004, + headers=[("a", b"a"), ("b", b"b")], + ) + producer.flush() + + producer.send( + topic=topic_name_1, + value=json.dumps({"value": 5}), + partition=0, + key="k5", + timestamp_ms=1577836805005, + ) + producer.send( + topic=topic_name_1, + value=json.dumps({"value": 6}), + partition=0, + key="k6", + timestamp_ms=1577836806006, + ) + producer.flush() + + producer.send( + topic=topic_name_1, + value=json.dumps({"value": 7}), + partition=1, + key="k7", + timestamp_ms=1577836807007, + ) + producer.send( + topic=topic_name_1, + value=json.dumps({"value": 8}), + partition=1, + key="k8", + timestamp_ms=1577836808008, + ) + producer.flush() + + instance.wait_for_log_line(log_line, repetitions=4, look_behind_lines=6000) + + members = describe_consumer_group(kafka_cluster, consumer_group) + # pprint.pprint(members) + # members[0]['client_id'] = 'ClickHouse-instance-test-kafka-0' + # members[1]['client_id'] = 'ClickHouse-instance-test-kafka-1' + + result = instance.query( + "SELECT * FROM test.view ORDER BY value", ignore_error=True + ) + + expected = f"""\ + 1 k1 {topic_name_0} 0 0 1577836801 1577836801001 ['content-encoding'] ['base64'] + 2 k2 {topic_name_0} 0 1 1577836802 1577836802002 ['empty_value','','','repetition','repetition'] ['','empty name','','1','2'] + 3 k3 {topic_name_0} 1 0 1577836803 1577836803003 ['b','a'] ['b','a'] + 4 k4 {topic_name_0} 1 1 1577836804 1577836804004 ['a','b'] ['a','b'] + 5 k5 {topic_name_1} 0 0 1577836805 1577836805005 [] [] + 6 k6 {topic_name_1} 0 1 1577836806 1577836806006 [] [] + 7 k7 {topic_name_1} 1 0 1577836807 1577836807007 [] [] + 8 k8 {topic_name_1} 1 1 1577836808 1577836808008 [] [] """ - CREATE TABLE test.kafka (value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'virt2_0,virt2_1', - kafka_group_name = 'virt2', - kafka_num_consumers = 2, - kafka_format = 'JSONEachRow'; - CREATE MATERIALIZED VIEW test.view Engine=Log AS - SELECT value, _key, _topic, _partition, _offset, toUnixTimestamp(_timestamp), toUnixTimestamp64Milli(_timestamp_ms), _headers.name, _headers.value FROM test.kafka; - """ - ) + assert TSV(result) == TSV(expected) - producer = KafkaProducer( - bootstrap_servers="localhost:{}".format(cluster.kafka_port), - value_serializer=producer_serializer, - key_serializer=producer_serializer, - ) + instance.query( + """ + DROP TABLE test.kafka; + DROP TABLE test.view; + """ + ) + instance.rotate_logs() - producer.send( - topic="virt2_0", - value=json.dumps({"value": 1}), - partition=0, - key="k1", - timestamp_ms=1577836801001, - headers=[("content-encoding", b"base64")], - ) - producer.send( - topic="virt2_0", - value=json.dumps({"value": 2}), - partition=0, - key="k2", - timestamp_ms=1577836802002, - headers=[ - ("empty_value", b""), - ("", b"empty name"), - ("", b""), - ("repetition", b"1"), - ("repetition", b"2"), - ], - ) - producer.flush() - producer.send( - topic="virt2_0", - value=json.dumps({"value": 3}), - partition=1, - key="k3", - timestamp_ms=1577836803003, - headers=[("b", b"b"), ("a", b"a")], - ) - producer.send( - topic="virt2_0", - value=json.dumps({"value": 4}), - partition=1, - key="k4", - timestamp_ms=1577836804004, - headers=[("a", b"a"), ("b", b"b")], - ) - producer.flush() - - producer.send( - topic="virt2_1", - value=json.dumps({"value": 5}), - partition=0, - key="k5", - timestamp_ms=1577836805005, - ) - producer.send( - topic="virt2_1", - value=json.dumps({"value": 6}), - partition=0, - key="k6", - timestamp_ms=1577836806006, - ) - producer.flush() - - producer.send( - topic="virt2_1", - value=json.dumps({"value": 7}), - partition=1, - key="k7", - timestamp_ms=1577836807007, - ) - producer.send( - topic="virt2_1", - value=json.dumps({"value": 8}), - partition=1, - key="k8", - timestamp_ms=1577836808008, - ) - producer.flush() - - instance.wait_for_log_line( - "kafka.*Committed offset 2.*virt2_[01]", repetitions=4, look_behind_lines=6000 - ) - - members = describe_consumer_group(kafka_cluster, "virt2") - # pprint.pprint(members) - # members[0]['client_id'] = 'ClickHouse-instance-test-kafka-0' - # members[1]['client_id'] = 'ClickHouse-instance-test-kafka-1' - - result = instance.query("SELECT * FROM test.view ORDER BY value", ignore_error=True) - - expected = """\ -1 k1 virt2_0 0 0 1577836801 1577836801001 ['content-encoding'] ['base64'] -2 k2 virt2_0 0 1 1577836802 1577836802002 ['empty_value','','','repetition','repetition'] ['','empty name','','1','2'] -3 k3 virt2_0 1 0 1577836803 1577836803003 ['b','a'] ['b','a'] -4 k4 virt2_0 1 1 1577836804 1577836804004 ['a','b'] ['a','b'] -5 k5 virt2_1 0 0 1577836805 1577836805005 [] [] -6 k6 virt2_1 0 1 1577836806 1577836806006 [] [] -7 k7 virt2_1 1 0 1577836807 1577836807007 [] [] -8 k8 virt2_1 1 1 1577836808 1577836808008 [] [] -""" - - assert TSV(result) == TSV(expected) - - instance.query( - """ - DROP TABLE test.kafka; - DROP TABLE test.view; - """ - ) - kafka_delete_topic(admin_client, "virt2_0") - kafka_delete_topic(admin_client, "virt2_1") +@pytest.mark.parametrize( + "create_query_generator, do_direct_read", + [(generate_old_create_table_query, True), (generate_new_create_table_query, False)], +) +def test_kafka_producer_consumer_separate_settings( + kafka_cluster, create_query_generator, do_direct_read +): instance.rotate_logs() - - -def test_kafka_producer_consumer_separate_settings(kafka_cluster): instance.query( - """ - DROP TABLE IF EXISTS test.test_kafka; - CREATE TABLE test.test_kafka (key UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'separate_settings', - kafka_group_name = 'test', - kafka_format = 'JSONEachRow', - kafka_row_delimiter = '\\n'; - """ + create_query_generator( + "test_kafka", + "key UInt64", + topic_list="separate_settings", + consumer_group="test", + ) ) - instance.query("SELECT * FROM test.test_kafka") + if do_direct_read: + instance.query("SELECT * FROM test.test_kafka") instance.query("INSERT INTO test.test_kafka VALUES (1)") assert instance.contains_in_log("Kafka producer created") @@ -2398,11 +2645,11 @@ def test_kafka_producer_consumer_separate_settings(kafka_cluster): # and producer configurations assert "heartbeat.interval.ms" in warn - kafka_consumer_applyed_properties = instance.grep_in_log("Consumer set property") - kafka_producer_applyed_properties = instance.grep_in_log("Producer set property") + kafka_consumer_applied_properties = instance.grep_in_log("Consumer set property") + kafka_producer_applied_properties = instance.grep_in_log("Producer set property") - assert kafka_consumer_applyed_properties is not None - assert kafka_producer_applyed_properties is not None + assert kafka_consumer_applied_properties is not None + assert kafka_producer_applied_properties is not None # global settings should be applied for consumer and producer global_settings = { @@ -2412,118 +2659,128 @@ def test_kafka_producer_consumer_separate_settings(kafka_cluster): for name, value in global_settings.items(): property_in_log = f"{name}:{value}" - assert property_in_log in kafka_consumer_applyed_properties - assert property_in_log in kafka_producer_applyed_properties + assert property_in_log in kafka_consumer_applied_properties + assert property_in_log in kafka_producer_applied_properties settings_topic__separate_settings__consumer = {"session.timeout.ms": "6001"} for name, value in settings_topic__separate_settings__consumer.items(): property_in_log = f"{name}:{value}" - assert property_in_log in kafka_consumer_applyed_properties - assert property_in_log not in kafka_producer_applyed_properties + assert property_in_log in kafka_consumer_applied_properties + assert property_in_log not in kafka_producer_applied_properties producer_settings = {"transaction.timeout.ms": "60001"} for name, value in producer_settings.items(): property_in_log = f"{name}:{value}" - assert property_in_log not in kafka_consumer_applyed_properties - assert property_in_log in kafka_producer_applyed_properties + assert property_in_log not in kafka_consumer_applied_properties + assert property_in_log in kafka_producer_applied_properties # Should be ignored, because it is inside producer tag producer_legacy_syntax__topic_separate_settings = {"message.timeout.ms": "300001"} for name, value in producer_legacy_syntax__topic_separate_settings.items(): property_in_log = f"{name}:{value}" - assert property_in_log not in kafka_consumer_applyed_properties - assert property_in_log not in kafka_producer_applyed_properties + assert property_in_log not in kafka_consumer_applied_properties + assert property_in_log not in kafka_producer_applied_properties # Old syntax, applied on consumer and producer legacy_syntax__topic_separated_settings = {"heartbeat.interval.ms": "302"} for name, value in legacy_syntax__topic_separated_settings.items(): property_in_log = f"{name}:{value}" - assert property_in_log in kafka_consumer_applyed_properties - assert property_in_log in kafka_producer_applyed_properties + assert property_in_log in kafka_consumer_applied_properties + assert property_in_log in kafka_producer_applied_properties -def test_kafka_produce_key_timestamp(kafka_cluster): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) - +@pytest.mark.parametrize( + "create_query_generator, log_line", + [ + (generate_new_create_table_query, "Saved offset 5"), + (generate_old_create_table_query, "Committed offset 5"), + ], +) +def test_kafka_produce_key_timestamp(kafka_cluster, create_query_generator, log_line): topic_name = "insert3" topic_config = { # default retention, since predefined timestamp_ms is used. "retention.ms": "-1", } - kafka_create_topic(admin_client, topic_name, config=topic_config) - instance.query( + with kafka_topic(get_admin_client(kafka_cluster), topic_name, config=topic_config): + writer_create_query = create_query_generator( + "kafka_writer", + "key UInt64, value UInt64, _key String, _timestamp DateTime('UTC')", + topic_list=topic_name, + consumer_group=topic_name, + format="TSV", + ) + reader_create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64, inserted_key String, inserted_timestamp DateTime('UTC')", + topic_list=topic_name, + consumer_group=topic_name, + format="TSV", + ) + + instance.query( + f""" + DROP TABLE IF EXISTS test.view; + DROP TABLE IF EXISTS test.consumer; + {writer_create_query}; + {reader_create_query}; + CREATE MATERIALIZED VIEW test.view ENGINE=MergeTree ORDER BY tuple() AS + SELECT key, value, inserted_key, toUnixTimestamp(inserted_timestamp), _key, _topic, _partition, _offset, toUnixTimestamp(_timestamp) FROM test.kafka; """ - DROP TABLE IF EXISTS test.view; - DROP TABLE IF EXISTS test.consumer; - CREATE TABLE test.kafka_writer (key UInt64, value UInt64, _key String, _timestamp DateTime('UTC')) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'insert3', - kafka_group_name = 'insert3', - kafka_format = 'TSV', - kafka_row_delimiter = '\\n'; + ) - CREATE TABLE test.kafka (key UInt64, value UInt64, inserted_key String, inserted_timestamp DateTime('UTC')) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'insert3', - kafka_group_name = 'insert3', - kafka_format = 'TSV', - kafka_row_delimiter = '\\n'; + instance.query( + "INSERT INTO test.kafka_writer VALUES ({},{},'{}',toDateTime({}))".format( + 1, 1, "k1", 1577836801 + ) + ) + instance.query( + "INSERT INTO test.kafka_writer VALUES ({},{},'{}',toDateTime({}))".format( + 2, 2, "k2", 1577836802 + ) + ) + instance.query( + "INSERT INTO test.kafka_writer VALUES ({},{},'{}',toDateTime({})),({},{},'{}',toDateTime({}))".format( + 3, 3, "k3", 1577836803, 4, 4, "k4", 1577836804 + ) + ) + instance.query( + "INSERT INTO test.kafka_writer VALUES ({},{},'{}',toDateTime({}))".format( + 5, 5, "k5", 1577836805 + ) + ) - CREATE MATERIALIZED VIEW test.view Engine=Log AS - SELECT key, value, inserted_key, toUnixTimestamp(inserted_timestamp), _key, _topic, _partition, _offset, toUnixTimestamp(_timestamp) FROM test.kafka; + # instance.wait_for_log_line(log_line) + + expected = """\ + 1 1 k1 1577836801 k1 insert3 0 0 1577836801 + 2 2 k2 1577836802 k2 insert3 0 1 1577836802 + 3 3 k3 1577836803 k3 insert3 0 2 1577836803 + 4 4 k4 1577836804 k4 insert3 0 3 1577836804 + 5 5 k5 1577836805 k5 insert3 0 4 1577836805 """ - ) - instance.query( - "INSERT INTO test.kafka_writer VALUES ({},{},'{}',toDateTime({}))".format( - 1, 1, "k1", 1577836801 + result = instance.query_with_retry( + "SELECT * FROM test.view ORDER BY value", + ignore_error=True, + retry_count=5, + sleep_time=1, + check_callback=lambda res: TSV(res) == TSV(expected), ) - ) - instance.query( - "INSERT INTO test.kafka_writer VALUES ({},{},'{}',toDateTime({}))".format( - 2, 2, "k2", 1577836802 - ) - ) - instance.query( - "INSERT INTO test.kafka_writer VALUES ({},{},'{}',toDateTime({})),({},{},'{}',toDateTime({}))".format( - 3, 3, "k3", 1577836803, 4, 4, "k4", 1577836804 - ) - ) - instance.query( - "INSERT INTO test.kafka_writer VALUES ({},{},'{}',toDateTime({}))".format( - 5, 5, "k5", 1577836805 - ) - ) - instance.wait_for_log_line("Committed offset 5") - - result = instance.query("SELECT * FROM test.view ORDER BY value", ignore_error=True) - - # logging.debug(result) - - expected = """\ -1 1 k1 1577836801 k1 insert3 0 0 1577836801 -2 2 k2 1577836802 k2 insert3 0 1 1577836802 -3 3 k3 1577836803 k3 insert3 0 2 1577836803 -4 4 k4 1577836804 k4 insert3 0 3 1577836804 -5 5 k5 1577836805 k5 insert3 0 4 1577836805 -""" - - assert TSV(result) == TSV(expected) - - kafka_delete_topic(admin_client, topic_name) + assert TSV(result) == TSV(expected) -def test_kafka_insert_avro(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_insert_avro(kafka_cluster, create_query_generator): admin_client = KafkaAdminClient( bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) ) @@ -2531,296 +2788,327 @@ def test_kafka_insert_avro(kafka_cluster): # default retention, since predefined timestamp_ms is used. "retention.ms": "-1", } - kafka_create_topic(admin_client, "avro1", config=topic_config) - - instance.query( - """ - DROP TABLE IF EXISTS test.kafka; - CREATE TABLE test.kafka (key UInt64, value UInt64, _timestamp DateTime('UTC')) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'avro1', - kafka_group_name = 'avro1', - kafka_commit_on_select = 1, - kafka_format = 'Avro'; - """ - ) - - instance.query( - "INSERT INTO test.kafka select number*10 as key, number*100 as value, 1636505534 as _timestamp from numbers(4) SETTINGS output_format_avro_rows_in_file = 2, output_format_avro_codec = 'deflate'" - ) - - messages = [] - while True: - messages.extend( - kafka_consume( - kafka_cluster, "avro1", needDecode=False, timestamp=1636505534 - ) + topic_name = "avro1" + get_topic_postfix(create_query_generator) + with kafka_topic(admin_client, topic_name, config=topic_config): + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64, _timestamp DateTime('UTC')", + topic_list=topic_name, + consumer_group=topic_name, + format="Avro", + ) + instance.query( + f""" + DROP TABLE IF EXISTS test.kafka; + {create_query} + """ ) - if len(messages) == 2: - break - result = "" - for a_message in messages: - result += decode_avro(a_message) + "\n" + instance.query( + "INSERT INTO test.kafka select number*10 as key, number*100 as value, 1636505534 as _timestamp from numbers(4) SETTINGS output_format_avro_rows_in_file = 2, output_format_avro_codec = 'deflate'" + ) - expected_result = """{'key': 0, 'value': 0, '_timestamp': 1636505534} + message_count = 2 + messages = kafka_consume_with_retry( + kafka_cluster, + topic_name, + message_count, + need_decode=False, + timestamp=1636505534, + ) + + result = "" + for a_message in messages: + result += decode_avro(a_message) + "\n" + + expected_result = """{'key': 0, 'value': 0, '_timestamp': 1636505534} {'key': 10, 'value': 100, '_timestamp': 1636505534} {'key': 20, 'value': 200, '_timestamp': 1636505534} {'key': 30, 'value': 300, '_timestamp': 1636505534} """ - assert result == expected_result + assert result == expected_result -def test_kafka_produce_consume_avro(kafka_cluster): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_produce_consume_avro(kafka_cluster, create_query_generator): + topic_name = "insert_avro" + get_topic_postfix(create_query_generator) + with kafka_topic(get_admin_client(kafka_cluster), topic_name): + num_rows = 75 - topic_name = "insert_avro" - kafka_create_topic(admin_client, topic_name) - - num_rows = 75 - - instance.query( - """ - DROP TABLE IF EXISTS test.view; - DROP TABLE IF EXISTS test.kafka; - DROP TABLE IF EXISTS test.kafka_writer; - - CREATE TABLE test.kafka_writer (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'avro', - kafka_group_name = 'avro', - kafka_format = 'Avro'; - - - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'avro', - kafka_group_name = 'avro', - kafka_format = 'Avro'; - - CREATE MATERIALIZED VIEW test.view Engine=Log AS - SELECT key, value FROM test.kafka; - """ - ) - - instance.query( - "INSERT INTO test.kafka_writer select number*10 as key, number*100 as value from numbers({num_rows}) SETTINGS output_format_avro_rows_in_file = 7".format( - num_rows=num_rows + writer_create_query = create_query_generator( + "kafka_writer", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + format="Avro", ) - ) - instance.wait_for_log_line( - "Committed offset {offset}".format(offset=math.ceil(num_rows / 7)) - ) + reader_create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + format="Avro", + ) - expected_num_rows = instance.query( - "SELECT COUNT(1) FROM test.view", ignore_error=True - ) - assert int(expected_num_rows) == num_rows - - expected_max_key = instance.query( - "SELECT max(key) FROM test.view", ignore_error=True - ) - assert int(expected_max_key) == (num_rows - 1) * 10 - - kafka_delete_topic(admin_client, topic_name) - - -def test_kafka_flush_by_time(kafka_cluster): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) - topic_name = "flush_by_time" - kafka_create_topic(admin_client, topic_name) - - instance.query( - """ - DROP TABLE IF EXISTS test.view; - DROP TABLE IF EXISTS test.consumer; - - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'flush_by_time', - kafka_group_name = 'flush_by_time', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 100, - kafka_row_delimiter = '\\n'; - - SELECT * FROM test.kafka; - - CREATE TABLE test.view (key UInt64, value UInt64, ts DateTime64(3) MATERIALIZED now64(3)) - ENGINE = MergeTree() - ORDER BY key; - """ - ) - - cancel = threading.Event() - - def produce(): - while not cancel.is_set(): - messages = [] - messages.append(json.dumps({"key": 0, "value": 0})) - kafka_produce(kafka_cluster, "flush_by_time", messages) - time.sleep(0.8) - - kafka_thread = threading.Thread(target=produce) - kafka_thread.start() - - instance.query( - """ - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.kafka; - """ - ) - - time.sleep(18) - - result = instance.query("SELECT uniqExact(ts) = 2, count() >= 15 FROM test.view") - - cancel.set() - kafka_thread.join() - - # kafka_cluster.open_bash_shell('instance') - - instance.query( - """ - DROP TABLE test.consumer; - DROP TABLE test.view; - """ - ) - - assert TSV(result) == TSV("1 1") - kafka_delete_topic(admin_client, topic_name) - - -def test_kafka_flush_by_block_size(kafka_cluster): - cancel = threading.Event() - - def produce(): - while not cancel.is_set(): - messages = [] - messages.append(json.dumps({"key": 0, "value": 0})) - kafka_produce(kafka_cluster, "flush_by_block_size", messages) - - kafka_thread = threading.Thread(target=produce) - kafka_thread.start() - - instance.query( - """ - DROP TABLE IF EXISTS test.view; - DROP TABLE IF EXISTS test.consumer; - - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'flush_by_block_size', - kafka_group_name = 'flush_by_block_size', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 100, - kafka_poll_max_batch_size = 1, - kafka_flush_interval_ms = 120000, /* should not flush by time during test */ - kafka_row_delimiter = '\\n'; - - CREATE TABLE test.view (key UInt64, value UInt64) - ENGINE = MergeTree() - ORDER BY key; - - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.kafka; - """ - ) - - # Wait for Kafka engine to consume this data - while 1 != int( instance.query( - "SELECT count() FROM system.parts WHERE database = 'test' AND table = 'view' AND name = 'all_1_1_0'" + f""" + DROP TABLE IF EXISTS test.view; + DROP TABLE IF EXISTS test.kafka; + DROP TABLE IF EXISTS test.kafka_writer; + + {writer_create_query}; + {reader_create_query}; + + CREATE MATERIALIZED VIEW test.view ENGINE=MergeTree ORDER BY tuple() AS + SELECT key, value FROM test.kafka; + """ ) - ): - time.sleep(0.5) - cancel.set() - kafka_thread.join() + instance.query( + "INSERT INTO test.kafka_writer select number*10 as key, number*100 as value from numbers({num_rows}) SETTINGS output_format_avro_rows_in_file = 7".format( + num_rows=num_rows + ) + ) - # more flushes can happens during test, we need to check only result of first flush (part named all_1_1_0). - result = instance.query("SELECT count() FROM test.view WHERE _part='all_1_1_0'") - # logging.debug(result) + instance.wait_for_log_line( + "Committed offset {offset}".format(offset=math.ceil(num_rows / 7)) + ) - instance.query( - """ - DROP TABLE test.consumer; - DROP TABLE test.view; - """ + expected_num_rows = instance.query( + "SELECT COUNT(1) FROM test.view", ignore_error=True + ) + assert int(expected_num_rows) == num_rows + + expected_max_key = instance.query( + "SELECT max(key) FROM test.view", ignore_error=True + ) + assert int(expected_max_key) == (num_rows - 1) * 10 + + +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_flush_by_time(kafka_cluster, create_query_generator): + admin_client = KafkaAdminClient( + bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) ) + topic_name = "flush_by_time" + get_topic_postfix(create_query_generator) - # 100 = first poll should return 100 messages (and rows) - # not waiting for stream_flush_interval_ms - assert ( - int(result) == 100 - ), "Messages from kafka should be flushed when block of size kafka_max_block_size is formed!" + with kafka_topic(admin_client, topic_name): + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + settings={ + "kafka_max_block_size": 100, + }, + ) + instance.query( + f""" + DROP TABLE IF EXISTS test.view; + DROP TABLE IF EXISTS test.consumer; + + {create_query}; + + CREATE TABLE test.view (key UInt64, value UInt64, ts DateTime64(3) MATERIALIZED now64(3)) + ENGINE = MergeTree() + ORDER BY key; + """ + ) + + cancel = threading.Event() + + def produce(): + while not cancel.is_set(): + messages = [json.dumps({"key": 0, "value": 0})] + kafka_produce(kafka_cluster, topic_name, messages) + time.sleep(0.8) + + kafka_thread = threading.Thread(target=produce) + kafka_thread.start() + + instance.query( + """ + CREATE MATERIALIZED VIEW test.consumer TO test.view AS + SELECT * FROM test.kafka; + """ + ) + + # By default the flush timeout should be 7.5 seconds => 18 seconds should be enough for 2 flushes, but not for 3 + time.sleep(18) + + result = instance.query("SELECT uniqExact(ts), count() >= 15 FROM test.view") + + cancel.set() + kafka_thread.join() + + instance.query( + """ + DROP TABLE test.consumer; + DROP TABLE test.view; + """ + ) + + assert TSV(result) == TSV("2 1") -def test_kafka_lot_of_partitions_partial_commit_of_bulk(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_flush_by_block_size(kafka_cluster, create_query_generator): + topic_name = "flush_by_block_size" + get_topic_postfix(create_query_generator) + + cancel = threading.Event() + + def produce(): + while not cancel.is_set(): + messages = [] + messages.append(json.dumps({"key": 0, "value": 0})) + kafka_produce(kafka_cluster, topic_name, messages) + + kafka_thread = threading.Thread(target=produce) + + with kafka_topic(get_admin_client(kafka_cluster), topic_name): + kafka_thread.start() + + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + settings={ + "kafka_max_block_size": 100, + "kafka_poll_max_batch_size": 1, + "kafka_flush_interval_ms": 120000, + }, + ) + + instance.query( + f""" + DROP TABLE IF EXISTS test.view; + DROP TABLE IF EXISTS test.consumer; + + {create_query}; + + CREATE TABLE test.view (key UInt64, value UInt64) + ENGINE = MergeTree() + ORDER BY key; + + CREATE MATERIALIZED VIEW test.consumer TO test.view AS + SELECT * FROM test.kafka; + """ + ) + + # Wait for Kafka engine to consume this data + while 1 != int( + instance.query( + "SELECT count() FROM system.parts WHERE database = 'test' AND table = 'view' AND name = 'all_1_1_0'" + ) + ): + time.sleep(0.5) + + cancel.set() + kafka_thread.join() + + # more flushes can happens during test, we need to check only result of first flush (part named all_1_1_0). + result = instance.query("SELECT count() FROM test.view WHERE _part='all_1_1_0'") + # logging.debug(result) + + instance.query( + """ + DROP TABLE test.consumer; + DROP TABLE test.view; + """ + ) + + # 100 = first poll should return 100 messages (and rows) + # not waiting for stream_flush_interval_ms + assert ( + int(result) == 100 + ), "Messages from kafka should be flushed when block of size kafka_max_block_size is formed!" + + +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_lot_of_partitions_partial_commit_of_bulk( + kafka_cluster, create_query_generator +): admin_client = KafkaAdminClient( bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) ) - topic_name = "topic_with_multiple_partitions2" - kafka_create_topic(admin_client, topic_name, num_partitions=10) - - instance.query( - """ - DROP TABLE IF EXISTS test.view; - DROP TABLE IF EXISTS test.consumer; - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'topic_with_multiple_partitions2', - kafka_group_name = 'topic_with_multiple_partitions2', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 211, - kafka_flush_interval_ms = 500; - CREATE TABLE test.view (key UInt64, value UInt64) - ENGINE = MergeTree() - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.kafka; - """ + topic_name = "topic_with_multiple_partitions2" + get_topic_postfix( + create_query_generator ) - - messages = [] - count = 0 - for dummy_msg in range(1000): - rows = [] - for dummy_row in range(random.randrange(3, 10)): - count = count + 1 - rows.append(json.dumps({"key": count, "value": count})) - messages.append("\n".join(rows)) - kafka_produce(kafka_cluster, "topic_with_multiple_partitions2", messages) - - instance.wait_for_log_line("kafka.*Stalled", repetitions=5) - - result = instance.query("SELECT count(), uniqExact(key), max(key) FROM test.view") - logging.debug(result) - assert TSV(result) == TSV("{0}\t{0}\t{0}".format(count)) - - instance.query( + with kafka_topic(admin_client, topic_name): + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + settings={ + "kafka_max_block_size": 211, + "kafka_flush_interval_ms": 500, + }, + ) + instance.query( + f""" + DROP TABLE IF EXISTS test.view; + DROP TABLE IF EXISTS test.consumer; + {create_query}; + CREATE TABLE test.view (key UInt64, value UInt64) + ENGINE = MergeTree() + ORDER BY key; + CREATE MATERIALIZED VIEW test.consumer TO test.view AS + SELECT * FROM test.kafka; """ - DROP TABLE test.consumer; - DROP TABLE test.view; - """ - ) - kafka_delete_topic(admin_client, topic_name) + ) + + messages = [] + count = 0 + for dummy_msg in range(1000): + rows = [] + for dummy_row in range(random.randrange(3, 10)): + count = count + 1 + rows.append(json.dumps({"key": count, "value": count})) + messages.append("\n".join(rows)) + kafka_produce(kafka_cluster, topic_name, messages) + + instance.wait_for_log_line("kafka.*Stalled", repetitions=5) + + result = instance.query( + "SELECT count(), uniqExact(key), max(key) FROM test.view" + ) + logging.debug(result) + assert TSV(result) == TSV("{0}\t{0}\t{0}".format(count)) + + instance.query( + """ + DROP TABLE test.consumer; + DROP TABLE test.view; + """ + ) -def test_kafka_rebalance(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator, log_line", + [ + (generate_old_create_table_query, "{}.*Polled offset [0-9]+"), + (generate_new_create_table_query, "{}.*Saved offset"), + ], +) +def test_kafka_rebalance(kafka_cluster, create_query_generator, log_line): NUMBER_OF_CONSURRENT_CONSUMERS = 11 instance.query( @@ -2841,212 +3129,232 @@ def test_kafka_rebalance(kafka_cluster): """ ) - # kafka_cluster.open_bash_shell('instance') - - # time.sleep(2) - admin_client = KafkaAdminClient( bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) ) - topic_name = "topic_with_multiple_partitions" - kafka_create_topic(admin_client, topic_name, num_partitions=11) + topic_name = "topic_with_multiple_partitions" + get_topic_postfix( + create_query_generator + ) + table_name_prefix = "kafka_consumer" + keeper_path = f"/clickhouse/{{database}}/{table_name_prefix}" + with kafka_topic(admin_client, topic_name, num_partitions=11): + cancel = threading.Event() - cancel = threading.Event() + msg_index = [0] - msg_index = [0] + def produce(): + while not cancel.is_set(): + messages = [] + for _ in range(59): + messages.append( + json.dumps({"key": msg_index[0], "value": msg_index[0]}) + ) + msg_index[0] += 1 + kafka_produce(kafka_cluster, topic_name, messages) - def produce(): - while not cancel.is_set(): - messages = [] - for _ in range(59): - messages.append( - json.dumps({"key": msg_index[0], "value": msg_index[0]}) - ) - msg_index[0] += 1 - kafka_produce(kafka_cluster, "topic_with_multiple_partitions", messages) + kafka_thread = threading.Thread(target=produce) + kafka_thread.start() - kafka_thread = threading.Thread(target=produce) - kafka_thread.start() + for consumer_index in range(NUMBER_OF_CONSURRENT_CONSUMERS): + table_name = f"{table_name_prefix}{consumer_index}" + replica_name = f"r{consumer_index}" + logging.debug(f"Setting up {consumer_index}") - for consumer_index in range(NUMBER_OF_CONSURRENT_CONSUMERS): - table_name = "kafka_consumer{}".format(consumer_index) - logging.debug(("Setting up {}".format(table_name))) - - instance.query( - """ - DROP TABLE IF EXISTS test.{0}; - DROP TABLE IF EXISTS test.{0}_mv; - CREATE TABLE test.{0} (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'topic_with_multiple_partitions', - kafka_group_name = 'rebalance_test_group', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 33, - kafka_flush_interval_ms = 500; - CREATE MATERIALIZED VIEW test.{0}_mv TO test.destination AS - SELECT - key, - value, - _topic, - _key, - _offset, - _partition, - _timestamp, - '{0}' as _consumed_by - FROM test.{0}; - """.format( - table_name + create_query = create_query_generator( + table_name, + "key UInt64, value UInt64", + topic_list=topic_name, + keeper_path=keeper_path, + replica_name=replica_name, + settings={ + "kafka_max_block_size": 33, + "kafka_flush_interval_ms": 500, + }, ) - ) + instance.query( + f""" + DROP TABLE IF EXISTS test.{table_name}; + DROP TABLE IF EXISTS test.{table_name}_mv; + {create_query}; + CREATE MATERIALIZED VIEW test.{table_name}_mv TO test.destination AS + SELECT + key, + value, + _topic, + _key, + _offset, + _partition, + _timestamp, + '{table_name}' as _consumed_by + FROM test.{table_name}; + """ + ) + # kafka_cluster.open_bash_shell('instance') + # Waiting for test.kafka_consumerX to start consume ... + instance.wait_for_log_line(log_line.format(table_name)) + + cancel.set() + + # I leave last one working by intent (to finish consuming after all rebalances) + for consumer_index in range(NUMBER_OF_CONSURRENT_CONSUMERS - 1): + logging.debug(("Dropping test.kafka_consumer{}".format(consumer_index))) + instance.query( + "DROP TABLE IF EXISTS test.kafka_consumer{} SYNC".format(consumer_index) + ) + + # logging.debug(instance.query('SELECT count(), uniqExact(key), max(key) + 1 FROM test.destination')) # kafka_cluster.open_bash_shell('instance') - # Waiting for test.kafka_consumerX to start consume ... - instance.wait_for_log_line( - "kafka_consumer{}.*Polled offset [0-9]+".format(consumer_index) - ) - cancel.set() + while 1: + messages_consumed = int( + instance.query("SELECT uniqExact(key) FROM test.destination") + ) + if messages_consumed >= msg_index[0]: + break + time.sleep(1) + logging.debug( + ( + "Waiting for finishing consuming (have {}, should be {})".format( + messages_consumed, msg_index[0] + ) + ) + ) - # I leave last one working by intent (to finish consuming after all rebalances) - for consumer_index in range(NUMBER_OF_CONSURRENT_CONSUMERS - 1): - logging.debug(("Dropping test.kafka_consumer{}".format(consumer_index))) - instance.query( - "DROP TABLE IF EXISTS test.kafka_consumer{} SYNC".format(consumer_index) - ) - - # logging.debug(instance.query('SELECT count(), uniqExact(key), max(key) + 1 FROM test.destination')) - # kafka_cluster.open_bash_shell('instance') - - while 1: - messages_consumed = int( - instance.query("SELECT uniqExact(key) FROM test.destination") - ) - if messages_consumed >= msg_index[0]: - break - time.sleep(1) logging.debug( ( - "Waiting for finishing consuming (have {}, should be {})".format( - messages_consumed, msg_index[0] + instance.query( + "SELECT count(), uniqExact(key), max(key) + 1 FROM test.destination" ) ) ) - logging.debug( - ( - instance.query( - "SELECT count(), uniqExact(key), max(key) + 1 FROM test.destination" - ) + # Some queries to debug... + # SELECT * FROM test.destination where key in (SELECT key FROM test.destination group by key having count() <> 1) + # select number + 1 as key from numbers(4141) x left join test.destination using (key) where test.destination.key = 0; + # SELECT * FROM test.destination WHERE key between 2360 and 2370 order by key; + # select _partition from test.destination group by _partition having count() <> max(_offset) + 1; + # select toUInt64(0) as _partition, number + 1 as _offset from numbers(400) x left join test.destination using (_partition,_offset) where test.destination.key = 0 order by _offset; + # SELECT * FROM test.destination WHERE _partition = 0 and _offset between 220 and 240 order by _offset; + + # CREATE TABLE test.reference (key UInt64, value UInt64) ENGINE = Kafka SETTINGS kafka_broker_list = 'kafka1:19092', + # kafka_topic_list = 'topic_with_multiple_partitions', + # kafka_group_name = 'rebalance_test_group_reference', + # kafka_format = 'JSONEachRow', + # kafka_max_block_size = 100000; + # + # CREATE MATERIALIZED VIEW test.reference_mv Engine=Log AS + # SELECT key, value, _topic,_key,_offset, _partition, _timestamp, 'reference' as _consumed_by + # FROM test.reference; + # + # select * from test.reference_mv left join test.destination using (key,_topic,_offset,_partition) where test.destination._consumed_by = ''; + + result = int( + instance.query("SELECT count() == uniqExact(key) FROM test.destination") ) - ) - # Some queries to debug... - # SELECT * FROM test.destination where key in (SELECT key FROM test.destination group by key having count() <> 1) - # select number + 1 as key from numbers(4141) x left join test.destination using (key) where test.destination.key = 0; - # SELECT * FROM test.destination WHERE key between 2360 and 2370 order by key; - # select _partition from test.destination group by _partition having count() <> max(_offset) + 1; - # select toUInt64(0) as _partition, number + 1 as _offset from numbers(400) x left join test.destination using (_partition,_offset) where test.destination.key = 0 order by _offset; - # SELECT * FROM test.destination WHERE _partition = 0 and _offset between 220 and 240 order by _offset; + for consumer_index in range(NUMBER_OF_CONSURRENT_CONSUMERS): + logging.debug(("kafka_consumer{}".format(consumer_index))) + table_name = "kafka_consumer{}".format(consumer_index) + instance.query( + """ + DROP TABLE IF EXISTS test.{0}; + DROP TABLE IF EXISTS test.{0}_mv; + """.format( + table_name + ) + ) - # CREATE TABLE test.reference (key UInt64, value UInt64) ENGINE = Kafka SETTINGS kafka_broker_list = 'kafka1:19092', - # kafka_topic_list = 'topic_with_multiple_partitions', - # kafka_group_name = 'rebalance_test_group_reference', - # kafka_format = 'JSONEachRow', - # kafka_max_block_size = 100000; - # - # CREATE MATERIALIZED VIEW test.reference_mv Engine=Log AS - # SELECT key, value, _topic,_key,_offset, _partition, _timestamp, 'reference' as _consumed_by - # FROM test.reference; - # - # select * from test.reference_mv left join test.destination using (key,_topic,_offset,_partition) where test.destination._consumed_by = ''; - - result = int( - instance.query("SELECT count() == uniqExact(key) FROM test.destination") - ) - - for consumer_index in range(NUMBER_OF_CONSURRENT_CONSUMERS): - logging.debug(("kafka_consumer{}".format(consumer_index))) - table_name = "kafka_consumer{}".format(consumer_index) instance.query( """ - DROP TABLE IF EXISTS test.{0}; - DROP TABLE IF EXISTS test.{0}_mv; - """.format( - table_name - ) + DROP TABLE IF EXISTS test.destination; + """ ) - instance.query( + kafka_thread.join() + + assert result == 1, "Messages from kafka get duplicated!" + + +# TODO(antaljanosbenjamin): find another way to make insertion fail +@pytest.mark.parametrize( + "create_query_generator", + [ + generate_old_create_table_query, + # generate_new_create_table_query, + ], +) +def test_kafka_no_holes_when_write_suffix_failed(kafka_cluster, create_query_generator): + admin_client = KafkaAdminClient( + bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) + ) + topic_name = "no_holes_when_write_suffix_failed" + get_topic_postfix( + create_query_generator + ) + + with kafka_topic(admin_client, topic_name): + messages = [json.dumps({"key": j + 1, "value": "x" * 300}) for j in range(22)] + kafka_produce(kafka_cluster, topic_name, messages) + + create_query = create_query_generator( + "kafka", + "key UInt64, value String", + topic_list=topic_name, + consumer_group=topic_name, + settings={ + "kafka_max_block_size": 20, + "kafka_flush_interval_ms": 2000, + }, + ) + instance.query( + f""" + DROP TABLE IF EXISTS test.view SYNC; + DROP TABLE IF EXISTS test.consumer; + + {create_query}; + + CREATE TABLE test.view (key UInt64, value String) + ENGINE = ReplicatedMergeTree('/clickhouse/kafkatest/tables/{topic_name}', 'node1') + ORDER BY key; """ - DROP TABLE IF EXISTS test.destination; - """ - ) + ) - kafka_thread.join() + # init PartitionManager (it starts container) earlier + pm = PartitionManager() - assert result == 1, "Messages from kafka get duplicated!" - kafka_delete_topic(admin_client, topic_name) - - -def test_kafka_no_holes_when_write_suffix_failed(kafka_cluster): - messages = [json.dumps({"key": j + 1, "value": "x" * 300}) for j in range(22)] - kafka_produce(kafka_cluster, "no_holes_when_write_suffix_failed", messages) - - instance.query( + instance.query( + """ + CREATE MATERIALIZED VIEW test.consumer TO test.view AS + SELECT * FROM test.kafka + WHERE NOT sleepEachRow(0.25); """ - DROP TABLE IF EXISTS test.view; - DROP TABLE IF EXISTS test.consumer; + ) - CREATE TABLE test.kafka (key UInt64, value String) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'no_holes_when_write_suffix_failed', - kafka_group_name = 'no_holes_when_write_suffix_failed', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 20, - kafka_flush_interval_ms = 2000; + instance.wait_for_log_line("Polled batch of 20 messages") + # the tricky part here is that disconnect should happen after write prefix, but before write suffix + # we have 0.25 (sleepEachRow) * 20 ( Rows ) = 5 sec window after "Polled batch of 20 messages" + # while materialized view is working to inject zookeeper failure + pm.drop_instance_zk_connections(instance) + instance.wait_for_log_line( + "Error.*(Connection loss|Coordination::Exception).*while pushing to view" + ) + pm.heal_all() + instance.wait_for_log_line("Committed offset 22") - CREATE TABLE test.view (key UInt64, value String) - ENGINE = ReplicatedMergeTree('/clickhouse/kafkatest/tables/no_holes_when_write_suffix_failed', 'node1') - ORDER BY key; - """ - ) + result = instance.query( + "SELECT count(), uniqExact(key), max(key) FROM test.view" + ) + logging.debug(result) - # init PartitionManager (it starts container) earlier - pm = PartitionManager() + # kafka_cluster.open_bash_shell('instance') - instance.query( + instance.query( + """ + DROP TABLE test.consumer; + DROP TABLE test.view SYNC; """ - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.kafka - WHERE NOT sleepEachRow(0.25); - """ - ) + ) - instance.wait_for_log_line("Polled batch of 20 messages") - # the tricky part here is that disconnect should happen after write prefix, but before write suffix - # we have 0.25 (sleepEachRow) * 20 ( Rows ) = 5 sec window after "Polled batch of 20 messages" - # while materialized view is working to inject zookeeper failure - pm.drop_instance_zk_connections(instance) - instance.wait_for_log_line( - "Error.*(Connection loss|Coordination::Exception).*while pushing to view" - ) - pm.heal_all() - instance.wait_for_log_line("Committed offset 22") - - result = instance.query("SELECT count(), uniqExact(key), max(key) FROM test.view") - logging.debug(result) - - # kafka_cluster.open_bash_shell('instance') - - instance.query( - """ - DROP TABLE test.consumer; - DROP TABLE test.view; - """ - ) - - assert TSV(result) == TSV("22\t22\t22") + assert TSV(result) == TSV("22\t22\t22") def test_exception_from_destructor(kafka_cluster): @@ -3092,12 +3400,30 @@ def test_exception_from_destructor(kafka_cluster): assert TSV(instance.query("SELECT 1")) == TSV("1") -def test_commits_of_unprocessed_messages_on_drop(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_commits_of_unprocessed_messages_on_drop(kafka_cluster, create_query_generator): + topic_name = "commits_of_unprocessed_messages_on_drop" + get_topic_postfix( + create_query_generator + ) messages = [json.dumps({"key": j + 1, "value": j + 1}) for j in range(1)] - kafka_produce(kafka_cluster, "commits_of_unprocessed_messages_on_drop", messages) + kafka_produce(kafka_cluster, topic_name, messages) + + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=f"{topic_name}_test_group", + settings={ + "kafka_max_block_size": 1000, + "kafka_flush_interval_ms": 1000, + }, + ) instance.query( - """ + f""" DROP TABLE IF EXISTS test.destination SYNC; CREATE TABLE test.destination ( key UInt64, @@ -3112,14 +3438,7 @@ def test_commits_of_unprocessed_messages_on_drop(kafka_cluster): ENGINE = MergeTree() ORDER BY key; - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'commits_of_unprocessed_messages_on_drop', - kafka_group_name = 'commits_of_unprocessed_messages_on_drop_test_group', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 1000, - kafka_flush_interval_ms = 1000; + {create_query}; CREATE MATERIALIZED VIEW test.kafka_consumer TO test.destination AS SELECT @@ -3147,9 +3466,7 @@ def test_commits_of_unprocessed_messages_on_drop(kafka_cluster): for _ in range(113): messages.append(json.dumps({"key": i[0], "value": i[0]})) i[0] += 1 - kafka_produce( - kafka_cluster, "commits_of_unprocessed_messages_on_drop", messages - ) + kafka_produce(kafka_cluster, topic_name, messages) time.sleep(0.5) kafka_thread = threading.Thread(target=produce) @@ -3162,18 +3479,17 @@ def test_commits_of_unprocessed_messages_on_drop(kafka_cluster): """ ) - instance.query( - """ - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'commits_of_unprocessed_messages_on_drop', - kafka_group_name = 'commits_of_unprocessed_messages_on_drop_test_group', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 10000, - kafka_flush_interval_ms = 1000; - """ + new_create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=f"{topic_name}_test_group", + settings={ + "kafka_max_block_size": 10000, + "kafka_flush_interval_ms": 1000, + }, ) + instance.query(new_create_query) cancel.set() instance.wait_for_log_line("kafka.*Stalled", repetitions=5) @@ -3197,22 +3513,31 @@ def test_commits_of_unprocessed_messages_on_drop(kafka_cluster): assert TSV(result) == TSV("{0}\t{0}\t{0}".format(i[0] - 1)), "Missing data!" -def test_bad_reschedule(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_bad_reschedule(kafka_cluster, create_query_generator): + topic_name = "test_bad_reschedule" + get_topic_postfix(create_query_generator) + messages = [json.dumps({"key": j + 1, "value": j + 1}) for j in range(20000)] - kafka_produce(kafka_cluster, "test_bad_reschedule", messages) + kafka_produce(kafka_cluster, topic_name, messages) + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + settings={ + "kafka_max_block_size": 1000, + "kafka_flush_interval_ms": 1000, + }, + ) instance.query( - """ - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'test_bad_reschedule', - kafka_group_name = 'test_bad_reschedule', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 1000, - kafka_flush_interval_ms = 1000; + f""" + {create_query}; - CREATE MATERIALIZED VIEW test.destination Engine=Log AS + CREATE MATERIALIZED VIEW test.destination ENGINE=MergeTree ORDER BY tuple() AS SELECT key, now() as consume_ts, @@ -3308,21 +3633,26 @@ def test_kafka_duplicates_when_commit_failed(kafka_cluster): # if we came to partition end we will repeat polling until reaching kafka_max_block_size or flush_interval -# that behavior is a bit quesionable - we can just take a bigger pauses between polls instead - +# that behavior is a bit questionable - we can just take a bigger pauses between polls instead - # to do more job in a single pass, and give more rest for a thread. # But in cases of some peaky loads in kafka topic the current contract sounds more predictable and # easier to understand, so let's keep it as is for now. # also we can came to eof because we drained librdkafka internal queue too fast -def test_premature_flush_on_eof(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_premature_flush_on_eof(kafka_cluster, create_query_generator): + topic_name = "premature_flush_on_eof" + get_topic_postfix(create_query_generator) + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + ) instance.query( - """ - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'premature_flush_on_eof', - kafka_group_name = 'premature_flush_on_eof', - kafka_format = 'JSONEachRow'; - SELECT * FROM test.kafka LIMIT 1; + f""" + {create_query}; CREATE TABLE test.destination ( key UInt64, value UInt64, @@ -3338,13 +3668,13 @@ def test_premature_flush_on_eof(kafka_cluster): """ ) - # messages created here will be consumed immedeately after MV creation + # messages created here will be consumed immediately after MV creation # reaching topic EOF. - # But we should not do flush immedeately after reaching EOF, because + # But we should not do flush immediately after reaching EOF, because # next poll can return more data, and we should respect kafka_flush_interval_ms # and try to form bigger block - messages = [json.dumps({"key": j + 1, "value": j + 1}) for j in range(1)] - kafka_produce(kafka_cluster, "premature_flush_on_eof", messages) + messages = [json.dumps({"key": 1, "value": 1})] + kafka_produce(kafka_cluster, topic_name, messages) instance.query( """ @@ -3368,7 +3698,7 @@ def test_premature_flush_on_eof(kafka_cluster): instance.wait_for_log_line("Stalled") # produce more messages after delay - kafka_produce(kafka_cluster, "premature_flush_on_eof", messages) + kafka_produce(kafka_cluster, topic_name, messages) # data was not flushed yet (it will be flushed 7.5 sec after creating MV) assert int(instance.query("SELECT count() FROM test.destination")) == 0 @@ -3389,58 +3719,81 @@ def test_premature_flush_on_eof(kafka_cluster): ) -def test_kafka_unavailable(kafka_cluster): - messages = [json.dumps({"key": j + 1, "value": j + 1}) for j in range(20000)] - kafka_produce(kafka_cluster, "test_bad_reschedule", messages) +@pytest.mark.parametrize( + "create_query_generator, do_direct_read", + [(generate_old_create_table_query, True), (generate_new_create_table_query, False)], +) +def test_kafka_unavailable(kafka_cluster, create_query_generator, do_direct_read): + number_of_messages = 20000 + topic_name = "test_bad_reschedule" + get_topic_postfix(create_query_generator) + messages = [ + json.dumps({"key": j + 1, "value": j + 1}) for j in range(number_of_messages) + ] + kafka_produce(kafka_cluster, topic_name, messages) - kafka_cluster.pause_container("kafka1") + with existing_kafka_topic(get_admin_client(kafka_cluster), topic_name): + kafka_cluster.pause_container("kafka1") - instance.query( + create_query = create_query_generator( + "test_bad_reschedule", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + settings={"kafka_max_block_size": 1000}, + ) + instance.query( + f""" + {create_query}; + + CREATE MATERIALIZED VIEW test.destination_unavailable ENGINE=MergeTree ORDER BY tuple() AS + SELECT + key, + now() as consume_ts, + value, + _topic, + _key, + _offset, + _partition, + _timestamp + FROM test.test_bad_reschedule; """ - CREATE TABLE test.test_bad_reschedule (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'test_bad_reschedule', - kafka_group_name = 'test_bad_reschedule', - kafka_format = 'JSONEachRow', - kafka_commit_on_select = 1, - kafka_max_block_size = 1000; + ) - CREATE MATERIALIZED VIEW test.destination_unavailable Engine=Log AS - SELECT - key, - now() as consume_ts, - value, - _topic, - _key, - _offset, - _partition, - _timestamp - FROM test.test_bad_reschedule; - """ - ) + if do_direct_read: + instance.query("SELECT * FROM test.test_bad_reschedule") + instance.query("SELECT count() FROM test.destination_unavailable") - instance.query("SELECT * FROM test.test_bad_reschedule") - instance.query("SELECT count() FROM test.destination_unavailable") + # enough to trigger issue + time.sleep(30) + kafka_cluster.unpause_container("kafka1") - # enough to trigger issue - time.sleep(30) - kafka_cluster.unpause_container("kafka1") + result = instance.query_with_retry( + "SELECT count() FROM test.destination_unavailable", + sleep_time=1, + check_callback=lambda res: int(res) == number_of_messages, + ) - while ( - int(instance.query("SELECT count() FROM test.destination_unavailable")) < 20000 - ): - print("Waiting for consume") - time.sleep(1) + assert int(result) == number_of_messages -def test_kafka_issue14202(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_issue14202(kafka_cluster, create_query_generator): """ INSERT INTO Kafka Engine from an empty SELECT sub query was leading to failure """ + topic_name = "issue14202" + get_topic_postfix(create_query_generator) + create_query = create_query_generator( + "kafka_q", + "t UInt64, some_string String", + topic_list=topic_name, + consumer_group=topic_name, + ) instance.query( - """ + f""" CREATE TABLE test.empty_table ( dt Date, some_string String @@ -3449,12 +3802,7 @@ def test_kafka_issue14202(kafka_cluster): PARTITION BY toYYYYMM(dt) ORDER BY some_string; - CREATE TABLE test.kafka_q (t UInt64, `some_string` String) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'issue14202', - kafka_group_name = 'issue14202', - kafka_format = 'JSONEachRow'; + {create_query}; """ ) @@ -3505,20 +3853,30 @@ def random_string(size=8): return "".join(random.choices(string.ascii_uppercase + string.digits, k=size)) -def test_kafka_engine_put_errors_to_stream(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_engine_put_errors_to_stream(kafka_cluster, create_query_generator): + topic_name = "kafka_engine_put_errors_to_stream" + get_topic_postfix( + create_query_generator + ) + create_query = create_query_generator( + "kafka", + "i Int64, s String", + topic_list=topic_name, + consumer_group=topic_name, + settings={ + "kafka_max_block_size": 128, + "kafka_handle_error_mode": "stream", + }, + ) instance.query( - """ + f""" DROP TABLE IF EXISTS test.kafka; DROP TABLE IF EXISTS test.kafka_data; DROP TABLE IF EXISTS test.kafka_errors; - CREATE TABLE test.kafka (i Int64, s String) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'kafka_engine_put_errors_to_stream', - kafka_group_name = 'kafka_engine_put_errors_to_stream', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 128, - kafka_handle_error_mode = 'stream'; + {create_query}; CREATE MATERIALIZED VIEW test.kafka_data (i Int64, s String) ENGINE = MergeTree ORDER BY i @@ -3546,19 +3904,20 @@ def test_kafka_engine_put_errors_to_stream(kafka_cluster): json.dumps({"i": "n_" + random_string(4), "s": random_string(8)}) ) - kafka_produce(kafka_cluster, "kafka_engine_put_errors_to_stream", messages) - instance.wait_for_log_line("Committed offset 128") + kafka_produce(kafka_cluster, topic_name, messages) + with existing_kafka_topic(get_admin_client(kafka_cluster), topic_name): + instance.wait_for_log_line("Committed offset 128") - assert TSV(instance.query("SELECT count() FROM test.kafka_data")) == TSV("64") - assert TSV(instance.query("SELECT count() FROM test.kafka_errors")) == TSV("64") + assert TSV(instance.query("SELECT count() FROM test.kafka_data")) == TSV("64") + assert TSV(instance.query("SELECT count() FROM test.kafka_errors")) == TSV("64") - instance.query( + instance.query( + """ + DROP TABLE test.kafka; + DROP TABLE test.kafka_data; + DROP TABLE test.kafka_errors; """ - DROP TABLE test.kafka; - DROP TABLE test.kafka_data; - DROP TABLE test.kafka_errors; - """ - ) + ) def gen_normal_json(): @@ -3587,21 +3946,35 @@ def gen_message_with_jsons(jsons=10, malformed=0): return s.getvalue() -def test_kafka_engine_put_errors_to_stream_with_random_malformed_json(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_engine_put_errors_to_stream_with_random_malformed_json( + kafka_cluster, create_query_generator +): + topic_name = ( + "kafka_engine_put_errors_to_stream_with_random_malformed_json" + + get_topic_postfix(create_query_generator) + ) + create_query = create_query_generator( + "kafka", + "i Int64, s String", + topic_list=topic_name, + consumer_group=topic_name, + settings={ + "kafka_max_block_size": 100, + "kafka_poll_max_batch_size": 1, + "kafka_handle_error_mode": "stream", + }, + ) + instance.query( - """ + f""" DROP TABLE IF EXISTS test.kafka; DROP TABLE IF EXISTS test.kafka_data; DROP TABLE IF EXISTS test.kafka_errors; - CREATE TABLE test.kafka (i Int64, s String) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'kafka_engine_put_errors_to_stream_with_random_malformed_json', - kafka_group_name = 'kafka_engine_put_errors_to_stream_with_random_malformed_json', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 100, - kafka_poll_max_batch_size = 1, - kafka_handle_error_mode = 'stream'; + {create_query}; CREATE MATERIALIZED VIEW test.kafka_data (i Int64, s String) ENGINE = MergeTree ORDER BY i @@ -3626,28 +3999,28 @@ def test_kafka_engine_put_errors_to_stream_with_random_malformed_json(kafka_clus else: messages.append(gen_message_with_jsons(10, 0)) - kafka_produce( - kafka_cluster, - "kafka_engine_put_errors_to_stream_with_random_malformed_json", - messages, - ) + kafka_produce(kafka_cluster, topic_name, messages) + with existing_kafka_topic(get_admin_client(kafka_cluster), topic_name): + instance.wait_for_log_line("Committed offset 128") + # 64 good messages, each containing 10 rows + assert TSV(instance.query("SELECT count() FROM test.kafka_data")) == TSV("640") + # 64 bad messages, each containing some broken row + assert TSV(instance.query("SELECT count() FROM test.kafka_errors")) == TSV("64") - instance.wait_for_log_line("Committed offset 128") - # 64 good messages, each containing 10 rows - assert TSV(instance.query("SELECT count() FROM test.kafka_data")) == TSV("640") - # 64 bad messages, each containing some broken row - assert TSV(instance.query("SELECT count() FROM test.kafka_errors")) == TSV("64") - - instance.query( + instance.query( + """ + DROP TABLE test.kafka; + DROP TABLE test.kafka_data; + DROP TABLE test.kafka_errors; """ - DROP TABLE test.kafka; - DROP TABLE test.kafka_data; - DROP TABLE test.kafka_errors; - """ - ) + ) -def test_kafka_formats_with_broken_message(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_formats_with_broken_message(kafka_cluster, create_query_generator): # data was dumped from clickhouse itself in a following manner # clickhouse-client --format=Native --query='SELECT toInt64(number) as id, toUInt16( intDiv( id, 65536 ) ) as blockNo, reinterpretAsString(19777) as val1, toFloat32(0.5) as val2, toUInt8(1) as val3 from numbers(100) ORDER BY id' | xxd -ps | tr -d '\n' | sed 's/\(..\)/\\x\1/g' admin_client = KafkaAdminClient( @@ -3665,7 +4038,10 @@ def test_kafka_formats_with_broken_message(kafka_cluster): # broken message '{"id":"0","blockNo":"BAD","val1":"AM","val2":0.5,"val3":1}', ], - "expected": """{"raw_message":"{\\"id\\":\\"0\\",\\"blockNo\\":\\"BAD\\",\\"val1\\":\\"AM\\",\\"val2\\":0.5,\\"val3\\":1}","error":"Cannot parse input: expected '\\"' before: 'BAD\\",\\"val1\\":\\"AM\\",\\"val2\\":0.5,\\"val3\\":1}': (while reading the value of key blockNo)"}""", + "expected": { + "raw_message": '{"id":"0","blockNo":"BAD","val1":"AM","val2":0.5,"val3":1}', + "error": 'Cannot parse input: expected \'"\' before: \'BAD","val1":"AM","val2":0.5,"val3":1}\': (while reading the value of key blockNo)', + }, "supports_empty_value": True, "printable": True, }, @@ -3678,7 +4054,10 @@ def test_kafka_formats_with_broken_message(kafka_cluster): # broken message '["0", "BAD", "AM", 0.5, 1]', ], - "expected": """{"raw_message":"[\\"0\\", \\"BAD\\", \\"AM\\", 0.5, 1]","error":"Cannot parse input: expected '\\"' before: 'BAD\\", \\"AM\\", 0.5, 1]': (while reading the value of key blockNo)"}""", + "expected": { + "raw_message": '["0", "BAD", "AM", 0.5, 1]', + "error": "Cannot parse input: expected '\"' before: 'BAD\", \"AM\", 0.5, 1]': (while reading the value of key blockNo)", + }, "supports_empty_value": True, "printable": True, }, @@ -3690,7 +4069,10 @@ def test_kafka_formats_with_broken_message(kafka_cluster): # broken message '["0", "BAD", "AM", 0.5, 1]', ], - "expected": """{"raw_message":"[\\"0\\", \\"BAD\\", \\"AM\\", 0.5, 1]","error":"Cannot parse JSON string: expected opening quote"}""", + "expected": { + "raw_message": '["0", "BAD", "AM", 0.5, 1]', + "error": "Cannot parse JSON string: expected opening quote", + }, "printable": True, }, "TSKV": { @@ -3701,7 +4083,10 @@ def test_kafka_formats_with_broken_message(kafka_cluster): # broken message "id=0\tblockNo=BAD\tval1=AM\tval2=0.5\tval3=1\n", ], - "expected": '{"raw_message":"id=0\\tblockNo=BAD\\tval1=AM\\tval2=0.5\\tval3=1\\n","error":"Found garbage after field in TSKV format: blockNo: (at row 1)\\n"}', + "expected": { + "raw_message": "id=0\tblockNo=BAD\tval1=AM\tval2=0.5\tval3=1\n", + "error": "Found garbage after field in TSKV format: blockNo: (at row 1)\n", + }, "printable": True, }, "CSV": { @@ -3712,7 +4097,10 @@ def test_kafka_formats_with_broken_message(kafka_cluster): # broken message '0,"BAD","AM",0.5,1\n', ], - "expected": """{"raw_message":"0,\\"BAD\\",\\"AM\\",0.5,1\\n","error":"Cannot parse input: expected '\\"' before: 'BAD\\",\\"AM\\",0.5,1\\\\n'"}""", + "expected": { + "raw_message": '0,"BAD","AM",0.5,1\n', + "error": "Cannot parse input: expected '\"' before: 'BAD\",\"AM\",0.5,1\\n'", + }, "printable": True, "supports_empty_value": True, }, @@ -3724,7 +4112,10 @@ def test_kafka_formats_with_broken_message(kafka_cluster): # broken message "0\tBAD\tAM\t0.5\t1\n", ], - "expected": """{"raw_message":"0\\tBAD\\tAM\\t0.5\\t1\\n","error":"Cannot parse input: expected '\\\\t' before: 'BAD\\\\tAM\\\\t0.5\\\\t1\\\\n'"}""", + "expected": { + "raw_message": "0\tBAD\tAM\t0.5\t1\n", + "error": "Cannot parse input: expected '\\t' before: 'BAD\\tAM\\t0.5\\t1\\n'", + }, "supports_empty_value": True, "printable": True, }, @@ -3736,7 +4127,10 @@ def test_kafka_formats_with_broken_message(kafka_cluster): # broken message '"id","blockNo","val1","val2","val3"\n0,"BAD","AM",0.5,1\n', ], - "expected": """{"raw_message":"\\"id\\",\\"blockNo\\",\\"val1\\",\\"val2\\",\\"val3\\"\\n0,\\"BAD\\",\\"AM\\",0.5,1\\n","error":"Cannot parse input: expected '\\"' before: 'BAD\\",\\"AM\\",0.5,1\\\\n'"}""", + "expected": { + "raw_message": '"id","blockNo","val1","val2","val3"\n0,"BAD","AM",0.5,1\n', + "error": "Cannot parse input: expected '\"' before: 'BAD\",\"AM\",0.5,1\\n'", + }, "printable": True, }, "Values": { @@ -3747,7 +4141,10 @@ def test_kafka_formats_with_broken_message(kafka_cluster): # broken message "(0,'BAD','AM',0.5,1)", ], - "expected": r"""{"raw_message":"(0,'BAD','AM',0.5,1)","error":"Cannot parse string 'BAD' as UInt16: syntax error at begin of string. Note: there are toUInt16OrZero and toUInt16OrNull functions, which returns zero\/NULL instead of throwing exception"}""", + "expected": { + "raw_message": "(0,'BAD','AM',0.5,1)", + "error": "Cannot parse string 'BAD' as UInt16: syntax error at begin of string. Note: there are toUInt16OrZero and toUInt16OrNull functions, which returns zero/NULL instead of throwing exception", + }, "supports_empty_value": True, "printable": True, }, @@ -3759,7 +4156,10 @@ def test_kafka_formats_with_broken_message(kafka_cluster): # broken message "id\tblockNo\tval1\tval2\tval3\n0\tBAD\tAM\t0.5\t1\n", ], - "expected": """{"raw_message":"id\\tblockNo\\tval1\\tval2\\tval3\\n0\\tBAD\\tAM\\t0.5\\t1\\n","error":"Cannot parse input: expected '\\\\t' before: 'BAD\\\\tAM\\\\t0.5\\\\t1\\\\n"}""", + "expected": { + "raw_message": "id\tblockNo\tval1\tval2\tval3\n0\tBAD\tAM\t0.5\t1\n", + "error": "Cannot parse input: expected '\\t' before: 'BAD\\tAM\\t0.5\\t1\\n", + }, "supports_empty_value": True, "printable": True, }, @@ -3771,7 +4171,10 @@ def test_kafka_formats_with_broken_message(kafka_cluster): # broken message "id\tblockNo\tval1\tval2\tval3\nInt64\tUInt16\tString\tFloat32\tUInt8\n0\tBAD\tAM\t0.5\t1\n", ], - "expected": """{"raw_message":"id\\tblockNo\\tval1\\tval2\\tval3\\nInt64\\tUInt16\\tString\\tFloat32\\tUInt8\\n0\\tBAD\\tAM\\t0.5\\t1\\n","error":"Cannot parse input: expected '\\\\t' before: 'BAD\\\\tAM\\\\t0.5\\\\t1\\\\n'"}""", + "expected": { + "raw_message": "id\tblockNo\tval1\tval2\tval3\nInt64\tUInt16\tString\tFloat32\tUInt8\n0\tBAD\tAM\t0.5\t1\n", + "error": "Cannot parse input: expected '\\t' before: 'BAD\\tAM\\t0.5\\t1\\n'", + }, "printable": True, }, "Native": { @@ -3782,7 +4185,10 @@ def test_kafka_formats_with_broken_message(kafka_cluster): # broken message b"\x05\x01\x02\x69\x64\x05\x49\x6e\x74\x36\x34\x00\x00\x00\x00\x00\x00\x00\x00\x07\x62\x6c\x6f\x63\x6b\x4e\x6f\x06\x53\x74\x72\x69\x6e\x67\x03\x42\x41\x44\x04\x76\x61\x6c\x31\x06\x53\x74\x72\x69\x6e\x67\x02\x41\x4d\x04\x76\x61\x6c\x32\x07\x46\x6c\x6f\x61\x74\x33\x32\x00\x00\x00\x3f\x04\x76\x61\x6c\x33\x05\x55\x49\x6e\x74\x38\x01", ], - "expected": """{"raw_message":"050102696405496E743634000000000000000007626C6F636B4E6F06537472696E67034241440476616C3106537472696E6702414D0476616C3207466C6F617433320000003F0476616C330555496E743801","error":"Cannot convert: String to UInt16"}""", + "expected": { + "raw_message": "050102696405496E743634000000000000000007626C6F636B4E6F06537472696E67034241440476616C3106537472696E6702414D0476616C3207466C6F617433320000003F0476616C330555496E743801", + "error": "Cannot convert: String to UInt16", + }, "printable": False, }, "RowBinary": { @@ -3793,7 +4199,10 @@ def test_kafka_formats_with_broken_message(kafka_cluster): # broken message b"\x00\x00\x00\x00\x00\x00\x00\x00\x03\x42\x41\x44\x02\x41\x4d\x00\x00\x00\x3f\x01", ], - "expected": '{"raw_message":"00000000000000000342414402414D0000003F01","error":"Cannot read all data. Bytes read: 9. Bytes expected: 65.: (at row 1)\\n"}', + "expected": { + "raw_message": "00000000000000000342414402414D0000003F01", + "error": "Cannot read all data. Bytes read: 9. Bytes expected: 65.: (at row 1)\n", + }, "printable": False, }, "RowBinaryWithNamesAndTypes": { @@ -3804,7 +4213,10 @@ def test_kafka_formats_with_broken_message(kafka_cluster): # broken message b"\x05\x02\x69\x64\x07\x62\x6c\x6f\x63\x6b\x4e\x6f\x04\x76\x61\x6c\x31\x04\x76\x61\x6c\x32\x04\x76\x61\x6c\x33\x05\x49\x6e\x74\x36\x34\x06\x53\x74\x72\x69\x6e\x67\x06\x53\x74\x72\x69\x6e\x67\x07\x46\x6c\x6f\x61\x74\x33\x32\x05\x55\x49\x6e\x74\x38\x00\x00\x00\x00\x00\x00\x00\x00\x03\x42\x41\x44\x02\x41\x4d\x00\x00\x00\x3f\x01", ], - "expected": '{"raw_message":"0502696407626C6F636B4E6F0476616C310476616C320476616C3305496E74363406537472696E6706537472696E6707466C6F617433320555496E743800000000000000000342414402414D0000003F01","error":"Type of \'blockNo\' must be UInt16, not String"}', + "expected": { + "raw_message": "0502696407626C6F636B4E6F0476616C310476616C320476616C3305496E74363406537472696E6706537472696E6707466C6F617433320555496E743800000000000000000342414402414D0000003F01", + "error": "Type of 'blockNo' must be UInt16, not String", + }, "printable": False, }, "ORC": { @@ -3815,15 +4227,19 @@ def test_kafka_formats_with_broken_message(kafka_cluster): # broken message b"\x4f\x52\x43\x0a\x0b\x0a\x03\x00\x00\x00\x12\x04\x08\x01\x50\x00\x0a\x15\x0a\x05\x00\x00\x00\x00\x00\x12\x0c\x08\x01\x12\x06\x08\x00\x10\x00\x18\x00\x50\x00\x0a\x12\x0a\x06\x00\x00\x00\x00\x00\x00\x12\x08\x08\x01\x42\x02\x08\x06\x50\x00\x0a\x12\x0a\x06\x00\x00\x00\x00\x00\x00\x12\x08\x08\x01\x42\x02\x08\x04\x50\x00\x0a\x29\x0a\x04\x00\x00\x00\x00\x12\x21\x08\x01\x1a\x1b\x09\x00\x00\x00\x00\x00\x00\xe0\x3f\x11\x00\x00\x00\x00\x00\x00\xe0\x3f\x19\x00\x00\x00\x00\x00\x00\xe0\x3f\x50\x00\x0a\x15\x0a\x05\x00\x00\x00\x00\x00\x12\x0c\x08\x01\x12\x06\x08\x02\x10\x02\x18\x02\x50\x00\xff\x80\xff\x80\xff\x00\xff\x80\xff\x03\x42\x41\x44\xff\x80\xff\x02\x41\x4d\xff\x80\x00\x00\x00\x3f\xff\x80\xff\x01\x0a\x06\x08\x06\x10\x00\x18\x0d\x0a\x06\x08\x06\x10\x01\x18\x17\x0a\x06\x08\x06\x10\x02\x18\x14\x0a\x06\x08\x06\x10\x03\x18\x14\x0a\x06\x08\x06\x10\x04\x18\x2b\x0a\x06\x08\x06\x10\x05\x18\x17\x0a\x06\x08\x00\x10\x00\x18\x02\x0a\x06\x08\x00\x10\x01\x18\x02\x0a\x06\x08\x01\x10\x01\x18\x02\x0a\x06\x08\x00\x10\x02\x18\x02\x0a\x06\x08\x02\x10\x02\x18\x02\x0a\x06\x08\x01\x10\x02\x18\x03\x0a\x06\x08\x00\x10\x03\x18\x02\x0a\x06\x08\x02\x10\x03\x18\x02\x0a\x06\x08\x01\x10\x03\x18\x02\x0a\x06\x08\x00\x10\x04\x18\x02\x0a\x06\x08\x01\x10\x04\x18\x04\x0a\x06\x08\x00\x10\x05\x18\x02\x0a\x06\x08\x01\x10\x05\x18\x02\x12\x04\x08\x00\x10\x00\x12\x04\x08\x00\x10\x00\x12\x04\x08\x00\x10\x00\x12\x04\x08\x00\x10\x00\x12\x04\x08\x00\x10\x00\x12\x04\x08\x00\x10\x00\x1a\x03\x47\x4d\x54\x0a\x59\x0a\x04\x08\x01\x50\x00\x0a\x0c\x08\x01\x12\x06\x08\x00\x10\x00\x18\x00\x50\x00\x0a\x08\x08\x01\x42\x02\x08\x06\x50\x00\x0a\x08\x08\x01\x42\x02\x08\x04\x50\x00\x0a\x21\x08\x01\x1a\x1b\x09\x00\x00\x00\x00\x00\x00\xe0\x3f\x11\x00\x00\x00\x00\x00\x00\xe0\x3f\x19\x00\x00\x00\x00\x00\x00\xe0\x3f\x50\x00\x0a\x0c\x08\x01\x12\x06\x08\x02\x10\x02\x18\x02\x50\x00\x08\x03\x10\xec\x02\x1a\x0c\x08\x03\x10\x8e\x01\x18\x1d\x20\xc1\x01\x28\x01\x22\x2e\x08\x0c\x12\x05\x01\x02\x03\x04\x05\x1a\x02\x69\x64\x1a\x07\x62\x6c\x6f\x63\x6b\x4e\x6f\x1a\x04\x76\x61\x6c\x31\x1a\x04\x76\x61\x6c\x32\x1a\x04\x76\x61\x6c\x33\x20\x00\x28\x00\x30\x00\x22\x08\x08\x04\x20\x00\x28\x00\x30\x00\x22\x08\x08\x08\x20\x00\x28\x00\x30\x00\x22\x08\x08\x08\x20\x00\x28\x00\x30\x00\x22\x08\x08\x05\x20\x00\x28\x00\x30\x00\x22\x08\x08\x01\x20\x00\x28\x00\x30\x00\x30\x01\x3a\x04\x08\x01\x50\x00\x3a\x0c\x08\x01\x12\x06\x08\x00\x10\x00\x18\x00\x50\x00\x3a\x08\x08\x01\x42\x02\x08\x06\x50\x00\x3a\x08\x08\x01\x42\x02\x08\x04\x50\x00\x3a\x21\x08\x01\x1a\x1b\x09\x00\x00\x00\x00\x00\x00\xe0\x3f\x11\x00\x00\x00\x00\x00\x00\xe0\x3f\x19\x00\x00\x00\x00\x00\x00\xe0\x3f\x50\x00\x3a\x0c\x08\x01\x12\x06\x08\x02\x10\x02\x18\x02\x50\x00\x40\x90\x4e\x48\x01\x08\xd5\x01\x10\x00\x18\x80\x80\x04\x22\x02\x00\x0b\x28\x5b\x30\x06\x82\xf4\x03\x03\x4f\x52\x43\x18", ], - "expected": r"""{"raw_message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error":"Cannot parse string 'BAD' as UInt16: syntax error at begin of string. Note: there are toUInt16OrZero and toUInt16OrNull functions, which returns zero\/NULL instead of throwing exception."}""", + "expected": { + "raw_message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error": "Cannot parse string 'BAD' as UInt16: syntax error at begin of string. Note: there are toUInt16OrZero and toUInt16OrNull functions, which returns zero/NULL instead of throwing exception.", + }, "printable": False, }, } topic_name_prefix = "format_tests_4_stream_" + topic_name_postfix = get_topic_postfix(create_query_generator) for format_name, format_opts in list(all_formats.items()): logging.debug(f"Set up {format_name}") - topic_name = f"{topic_name_prefix}{format_name}" + topic_name = f"{topic_name_prefix}{format_name}{topic_name_postfix}" data_sample = format_opts["data_sample"] data_prefix = [] raw_message = "_raw_message" @@ -3833,39 +4249,33 @@ def test_kafka_formats_with_broken_message(kafka_cluster): if format_opts.get("printable", False) == False: raw_message = "hex(_raw_message)" kafka_produce(kafka_cluster, topic_name, data_prefix + data_sample) + create_query = create_query_generator( + f"kafka_{format_name}", + "id Int64, blockNo UInt16, val1 String, val2 Float32, val3 UInt8", + topic_list=topic_name, + consumer_group=topic_name, + format=format_name, + settings={ + "kafka_handle_error_mode": "stream", + "kafka_flush_interval_ms": 1000, + }, + ) instance.query( - """ + f""" DROP TABLE IF EXISTS test.kafka_{format_name}; - CREATE TABLE test.kafka_{format_name} ( - id Int64, - blockNo UInt16, - val1 String, - val2 Float32, - val3 UInt8 - ) ENGINE = Kafka() - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = '{topic_name}', - kafka_group_name = '{topic_name}', - kafka_format = '{format_name}', - kafka_handle_error_mode = 'stream', - kafka_flush_interval_ms = 1000 {extra_settings}; + {create_query}; DROP TABLE IF EXISTS test.kafka_data_{format_name}_mv; - CREATE MATERIALIZED VIEW test.kafka_data_{format_name}_mv Engine=Log AS + CREATE MATERIALIZED VIEW test.kafka_data_{format_name}_mv ENGINE=MergeTree ORDER BY tuple() AS SELECT *, _topic, _partition, _offset FROM test.kafka_{format_name} WHERE length(_error) = 0; DROP TABLE IF EXISTS test.kafka_errors_{format_name}_mv; - CREATE MATERIALIZED VIEW test.kafka_errors_{format_name}_mv Engine=Log AS + CREATE MATERIALIZED VIEW test.kafka_errors_{format_name}_mv ENGINE=MergeTree ORDER BY tuple() AS SELECT {raw_message} as raw_message, _error as error, _topic as topic, _partition as partition, _offset as offset FROM test.kafka_{format_name} WHERE length(_error) > 0; - """.format( - topic_name=topic_name, - format_name=format_name, - raw_message=raw_message, - extra_settings=format_opts.get("extra_settings") or "", - ) + """ ) raw_expected = """\ @@ -3900,7 +4310,7 @@ def test_kafka_formats_with_broken_message(kafka_cluster): for format_name, format_opts in list(all_formats.items()): logging.debug(f"Checking {format_name}") - topic_name = f"{topic_name_prefix}{format_name}" + topic_name = f"{topic_name_prefix}{format_name}{topic_name_postfix}" # shift offsets by 1 if format supports empty value offsets = ( [1, 2, 3] if format_opts.get("supports_empty_value", False) else [0, 1, 2] @@ -3920,229 +4330,222 @@ def test_kafka_formats_with_broken_message(kafka_cluster): assert TSV(result) == TSV(expected), "Proper result for format: {}".format( format_name ) - errors_result = ast.literal_eval( + errors_result = json.loads( instance.query( "SELECT raw_message, error FROM test.kafka_errors_{format_name}_mv format JSONEachRow".format( format_name=format_name ) ) ) - errors_expected = ast.literal_eval(format_opts["expected"]) # print(errors_result.strip()) # print(errors_expected.strip()) assert ( - errors_result["raw_message"] == errors_expected["raw_message"] + errors_result["raw_message"] == format_opts["expected"]["raw_message"] ), "Proper raw_message for format: {}".format(format_name) # Errors text can change, just checking prefixes assert ( - errors_expected["error"] in errors_result["error"] + format_opts["expected"]["error"] in errors_result["error"] ), "Proper error for format: {}".format(format_name) kafka_delete_topic(admin_client, topic_name) -def wait_for_new_data(table_name, prev_count=0, max_retries=120): - retries = 0 - while True: - new_count = int(instance.query("SELECT count() FROM {}".format(table_name))) - print(new_count) - if new_count > prev_count: - return new_count - else: - retries += 1 - time.sleep(0.5) - if retries > max_retries: - raise Exception("No new data :(") +@pytest.mark.parametrize( + "create_query_generator", + [ + generate_old_create_table_query, + generate_new_create_table_query, + ], +) +def test_kafka_consumer_failover(kafka_cluster, create_query_generator): + topic_name = "kafka_consumer_failover" + get_topic_postfix(create_query_generator) + with kafka_topic(get_admin_client(kafka_cluster), topic_name, num_partitions=2): + consumer_group = f"{topic_name}_group" + create_queries = [] + for counter in range(3): + create_queries.append( + create_query_generator( + f"kafka{counter+1}", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=consumer_group, + settings={ + "kafka_max_block_size": 1, + "kafka_poll_timeout_ms": 200, + }, + ) + ) -def test_kafka_consumer_failover(kafka_cluster): - # for backporting: - # admin_client = KafkaAdminClient(bootstrap_servers="localhost:9092") - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) + instance.query( + f""" + {create_queries[0]}; + {create_queries[1]}; + {create_queries[2]}; - topic_name = "kafka_consumer_failover" - kafka_create_topic(admin_client, topic_name, num_partitions=2) + CREATE TABLE test.destination ( + key UInt64, + value UInt64, + _consumed_by LowCardinality(String) + ) + ENGINE = MergeTree() + ORDER BY key; - instance.query( - """ - DROP TABLE IF EXISTS test.kafka; - DROP TABLE IF EXISTS test.kafka2; + CREATE MATERIALIZED VIEW test.kafka1_mv TO test.destination AS + SELECT key, value, 'kafka1' as _consumed_by + FROM test.kafka1; - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'kafka_consumer_failover', - kafka_group_name = 'kafka_consumer_failover_group', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 1, - kafka_poll_timeout_ms = 200; + CREATE MATERIALIZED VIEW test.kafka2_mv TO test.destination AS + SELECT key, value, 'kafka2' as _consumed_by + FROM test.kafka2; - CREATE TABLE test.kafka2 (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'kafka_consumer_failover', - kafka_group_name = 'kafka_consumer_failover_group', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 1, - kafka_poll_timeout_ms = 200; - - CREATE TABLE test.kafka3 (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'kafka_consumer_failover', - kafka_group_name = 'kafka_consumer_failover_group', - kafka_format = 'JSONEachRow', - kafka_max_block_size = 1, - kafka_poll_timeout_ms = 200; - - CREATE TABLE test.destination ( - key UInt64, - value UInt64, - _consumed_by LowCardinality(String) + CREATE MATERIALIZED VIEW test.kafka3_mv TO test.destination AS + SELECT key, value, 'kafka3' as _consumed_by + FROM test.kafka3; + """ ) - ENGINE = MergeTree() - ORDER BY key; - CREATE MATERIALIZED VIEW test.kafka_mv TO test.destination AS - SELECT key, value, 'kafka' as _consumed_by - FROM test.kafka; + producer = KafkaProducer( + bootstrap_servers="localhost:{}".format(cluster.kafka_port), + value_serializer=producer_serializer, + key_serializer=producer_serializer, + ) - CREATE MATERIALIZED VIEW test.kafka2_mv TO test.destination AS - SELECT key, value, 'kafka2' as _consumed_by - FROM test.kafka2; + ## all 3 attached, 2 working + producer.send( + topic=topic_name, + value=json.dumps({"key": 1, "value": 1}), + partition=0, + ) + producer.send( + topic=topic_name, + value=json.dumps({"key": 1, "value": 1}), + partition=1, + ) + producer.flush() - CREATE MATERIALIZED VIEW test.kafka3_mv TO test.destination AS - SELECT key, value, 'kafka3' as _consumed_by - FROM test.kafka3; - """ - ) + count_query = "SELECT count() FROM test.destination" + prev_count = instance.query_with_retry( + count_query, check_callback=lambda res: int(res) > 0 + ) - producer = KafkaProducer( - bootstrap_servers="localhost:{}".format(cluster.kafka_port), - value_serializer=producer_serializer, - key_serializer=producer_serializer, - ) + ## 2 attached, 2 working + instance.query("DETACH TABLE test.kafka1") + producer.send( + topic=topic_name, + value=json.dumps({"key": 2, "value": 2}), + partition=0, + ) + producer.send( + topic=topic_name, + value=json.dumps({"key": 2, "value": 2}), + partition=1, + ) + producer.flush() + prev_count = instance.query_with_retry( + count_query, check_callback=lambda res: int(res) > prev_count + ) - ## all 3 attached, 2 working - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 1, "value": 1}), - partition=0, - ) - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 1, "value": 1}), - partition=1, - ) - producer.flush() - prev_count = wait_for_new_data("test.destination") + ## 1 attached, 1 working + instance.query("DETACH TABLE test.kafka2") + producer.send( + topic=topic_name, + value=json.dumps({"key": 3, "value": 3}), + partition=0, + ) + producer.send( + topic=topic_name, + value=json.dumps({"key": 3, "value": 3}), + partition=1, + ) + producer.flush() + prev_count = instance.query_with_retry( + count_query, check_callback=lambda res: int(res) > prev_count + ) - ## 2 attached, 2 working - instance.query("DETACH TABLE test.kafka") - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 2, "value": 2}), - partition=0, - ) - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 2, "value": 2}), - partition=1, - ) - producer.flush() - prev_count = wait_for_new_data("test.destination", prev_count) + ## 2 attached, 2 working + instance.query("ATTACH TABLE test.kafka1") + producer.send( + topic=topic_name, + value=json.dumps({"key": 4, "value": 4}), + partition=0, + ) + producer.send( + topic=topic_name, + value=json.dumps({"key": 4, "value": 4}), + partition=1, + ) + producer.flush() + prev_count = instance.query_with_retry( + count_query, check_callback=lambda res: int(res) > prev_count + ) - ## 1 attached, 1 working - instance.query("DETACH TABLE test.kafka2") - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 3, "value": 3}), - partition=0, - ) - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 3, "value": 3}), - partition=1, - ) - producer.flush() - prev_count = wait_for_new_data("test.destination", prev_count) + ## 1 attached, 1 working + instance.query("DETACH TABLE test.kafka3") + producer.send( + topic=topic_name, + value=json.dumps({"key": 5, "value": 5}), + partition=0, + ) + producer.send( + topic=topic_name, + value=json.dumps({"key": 5, "value": 5}), + partition=1, + ) + producer.flush() + prev_count = instance.query_with_retry( + count_query, check_callback=lambda res: int(res) > prev_count + ) - ## 2 attached, 2 working - instance.query("ATTACH TABLE test.kafka") - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 4, "value": 4}), - partition=0, - ) - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 4, "value": 4}), - partition=1, - ) - producer.flush() - prev_count = wait_for_new_data("test.destination", prev_count) + ## 2 attached, 2 working + instance.query("ATTACH TABLE test.kafka2") + producer.send( + topic=topic_name, + value=json.dumps({"key": 6, "value": 6}), + partition=0, + ) + producer.send( + topic=topic_name, + value=json.dumps({"key": 6, "value": 6}), + partition=1, + ) + producer.flush() + prev_count = instance.query_with_retry( + count_query, check_callback=lambda res: int(res) > prev_count + ) - ## 1 attached, 1 working - instance.query("DETACH TABLE test.kafka3") - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 5, "value": 5}), - partition=0, - ) - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 5, "value": 5}), - partition=1, - ) - producer.flush() - prev_count = wait_for_new_data("test.destination", prev_count) + ## 3 attached, 2 working + instance.query("ATTACH TABLE test.kafka3") + producer.send( + topic=topic_name, + value=json.dumps({"key": 7, "value": 7}), + partition=0, + ) + producer.send( + topic=topic_name, + value=json.dumps({"key": 7, "value": 7}), + partition=1, + ) + producer.flush() + prev_count = instance.query_with_retry( + count_query, check_callback=lambda res: int(res) > prev_count + ) - ## 2 attached, 2 working - instance.query("ATTACH TABLE test.kafka2") - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 6, "value": 6}), - partition=0, - ) - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 6, "value": 6}), - partition=1, - ) - producer.flush() - prev_count = wait_for_new_data("test.destination", prev_count) - - ## 3 attached, 2 working - instance.query("ATTACH TABLE test.kafka3") - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 7, "value": 7}), - partition=0, - ) - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 7, "value": 7}), - partition=1, - ) - producer.flush() - prev_count = wait_for_new_data("test.destination", prev_count) - - ## 2 attached, same 2 working - instance.query("DETACH TABLE test.kafka3") - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 8, "value": 8}), - partition=0, - ) - producer.send( - topic="kafka_consumer_failover", - value=json.dumps({"key": 8, "value": 8}), - partition=1, - ) - producer.flush() - prev_count = wait_for_new_data("test.destination", prev_count) - kafka_delete_topic(admin_client, topic_name) + ## 2 attached, same 2 working + instance.query("DETACH TABLE test.kafka3") + producer.send( + topic=topic_name, + value=json.dumps({"key": 8, "value": 8}), + partition=0, + ) + producer.send( + topic=topic_name, + value=json.dumps({"key": 8, "value": 8}), + partition=1, + ) + producer.flush() + prev_count = instance.query_with_retry( + count_query, check_callback=lambda res: int(res) > prev_count + ) def test_kafka_predefined_configuration(kafka_cluster): @@ -4172,269 +4575,261 @@ def test_kafka_predefined_configuration(kafka_cluster): # https://github.com/ClickHouse/ClickHouse/issues/26643 -def test_issue26643(kafka_cluster): - # for backporting: - # admin_client = KafkaAdminClient(bootstrap_servers="localhost:9092") - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_issue26643(kafka_cluster, create_query_generator): producer = KafkaProducer( bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port), value_serializer=producer_serializer, ) + topic_name = "test_issue26643" + get_topic_postfix(create_query_generator) + thread_per_consumer = must_use_thread_per_consumer(create_query_generator) - topic_list = [] - topic_list.append( - NewTopic(name="test_issue26643", num_partitions=4, replication_factor=1) - ) - admin_client.create_topics(new_topics=topic_list, validate_only=False) - - msg = message_with_repeated_pb2.Message( - tnow=1629000000, - server="server1", - clien="host1", - sPort=443, - cPort=50000, - r=[ - message_with_repeated_pb2.dd( - name="1", type=444, ttl=123123, data=b"adsfasd" - ), - message_with_repeated_pb2.dd(name="2"), - ], - method="GET", - ) - - data = b"" - serialized_msg = msg.SerializeToString() - data = data + _VarintBytes(len(serialized_msg)) + serialized_msg - - msg = message_with_repeated_pb2.Message(tnow=1629000002) - - serialized_msg = msg.SerializeToString() - data = data + _VarintBytes(len(serialized_msg)) + serialized_msg - - producer.send(topic="test_issue26643", value=data) - - data = _VarintBytes(len(serialized_msg)) + serialized_msg - producer.send(topic="test_issue26643", value=data) - producer.flush() - - instance.query( - """ - CREATE TABLE IF NOT EXISTS test.test_queue - ( - `tnow` UInt32, - `server` String, - `client` String, - `sPort` UInt16, - `cPort` UInt16, - `r.name` Array(String), - `r.class` Array(UInt16), - `r.type` Array(UInt16), - `r.ttl` Array(UInt32), - `r.data` Array(String), - `method` String + with kafka_topic(get_admin_client(kafka_cluster), topic_name): + msg = message_with_repeated_pb2.Message( + tnow=1629000000, + server="server1", + clien="host1", + sPort=443, + cPort=50000, + r=[ + message_with_repeated_pb2.dd( + name="1", type=444, ttl=123123, data=b"adsfasd" + ), + message_with_repeated_pb2.dd(name="2"), + ], + method="GET", ) - ENGINE = Kafka - SETTINGS - kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'test_issue26643', - kafka_group_name = 'test_issue26643_group', - kafka_format = 'Protobuf', - kafka_schema = 'message_with_repeated.proto:Message', - kafka_num_consumers = 4, - kafka_skip_broken_messages = 10000; - SET allow_suspicious_low_cardinality_types=1; + data = b"" + serialized_msg = msg.SerializeToString() + data = data + _VarintBytes(len(serialized_msg)) + serialized_msg - CREATE TABLE test.log - ( - `tnow` DateTime('Asia/Istanbul') CODEC(DoubleDelta, LZ4), - `server` LowCardinality(String), - `client` LowCardinality(String), - `sPort` LowCardinality(UInt16), - `cPort` UInt16 CODEC(T64, LZ4), - `r.name` Array(String), - `r.class` Array(LowCardinality(UInt16)), - `r.type` Array(LowCardinality(UInt16)), - `r.ttl` Array(LowCardinality(UInt32)), - `r.data` Array(String), - `method` LowCardinality(String) + msg = message_with_repeated_pb2.Message(tnow=1629000002) + + serialized_msg = msg.SerializeToString() + data = data + _VarintBytes(len(serialized_msg)) + serialized_msg + + producer.send(topic_name, value=data) + + data = _VarintBytes(len(serialized_msg)) + serialized_msg + producer.send(topic_name, value=data) + producer.flush() + + create_query = create_query_generator( + "test_queue", + """`tnow` UInt32, + `server` String, + `client` String, + `sPort` UInt16, + `cPort` UInt16, + `r.name` Array(String), + `r.class` Array(UInt16), + `r.type` Array(UInt16), + `r.ttl` Array(UInt32), + `r.data` Array(String), + `method` String""", + topic_list=topic_name, + consumer_group=f"{topic_name}_group", + format="Protobuf", + settings={ + "kafka_schema": "message_with_repeated.proto:Message", + "kafka_skip_broken_messages": 10000, + "kafka_thread_per_consumer": thread_per_consumer, + }, ) - ENGINE = MergeTree - PARTITION BY toYYYYMMDD(tnow) - ORDER BY (tnow, server) - TTL toDate(tnow) + toIntervalMonth(1000) - SETTINGS index_granularity = 16384, merge_with_ttl_timeout = 7200; - CREATE MATERIALIZED VIEW test.test_consumer TO test.log AS - SELECT - toDateTime(a.tnow) AS tnow, - a.server AS server, - a.client AS client, - a.sPort AS sPort, - a.cPort AS cPort, - a.`r.name` AS `r.name`, - a.`r.class` AS `r.class`, - a.`r.type` AS `r.type`, - a.`r.ttl` AS `r.ttl`, - a.`r.data` AS `r.data`, - a.method AS method - FROM test.test_queue AS a; - """ - ) + instance.query( + f""" + {create_query}; - instance.wait_for_log_line("Committed offset") - result = instance.query("SELECT * FROM test.log") + SET allow_suspicious_low_cardinality_types=1; - expected = """\ -2021-08-15 07:00:00 server1 443 50000 ['1','2'] [0,0] [444,0] [123123,0] ['adsfasd',''] GET -2021-08-15 07:00:02 0 0 [] [] [] [] [] -2021-08-15 07:00:02 0 0 [] [] [] [] [] -""" - assert TSV(result) == TSV(expected) + CREATE TABLE test.log + ( + `tnow` DateTime('Asia/Istanbul') CODEC(DoubleDelta, LZ4), + `server` LowCardinality(String), + `client` LowCardinality(String), + `sPort` LowCardinality(UInt16), + `cPort` UInt16 CODEC(T64, LZ4), + `r.name` Array(String), + `r.class` Array(LowCardinality(UInt16)), + `r.type` Array(LowCardinality(UInt16)), + `r.ttl` Array(LowCardinality(UInt32)), + `r.data` Array(String), + `method` LowCardinality(String) + ) + ENGINE = MergeTree + PARTITION BY toYYYYMMDD(tnow) + ORDER BY (tnow, server) + TTL toDate(tnow) + toIntervalMonth(1000) + SETTINGS index_granularity = 16384, merge_with_ttl_timeout = 7200; - # kafka_cluster.open_bash_shell('instance') + CREATE MATERIALIZED VIEW test.test_consumer TO test.log AS + SELECT + toDateTime(a.tnow) AS tnow, + a.server AS server, + a.client AS client, + a.sPort AS sPort, + a.cPort AS cPort, + a.`r.name` AS `r.name`, + a.`r.class` AS `r.class`, + a.`r.type` AS `r.type`, + a.`r.ttl` AS `r.ttl`, + a.`r.data` AS `r.data`, + a.method AS method + FROM test.test_queue AS a; + """ + ) + + instance.wait_for_log_line("Committed offset") + result = instance.query("SELECT * FROM test.log") + + expected = """\ + 2021-08-15 07:00:00 server1 443 50000 ['1','2'] [0,0] [444,0] [123123,0] ['adsfasd',''] GET + 2021-08-15 07:00:02 0 0 [] [] [] [] [] + 2021-08-15 07:00:02 0 0 [] [] [] [] [] + """ + assert TSV(result) == TSV(expected) -def test_num_consumers_limit(kafka_cluster): +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_num_consumers_limit(kafka_cluster, create_query_generator): instance.query("DROP TABLE IF EXISTS test.kafka") - error = instance.query_and_get_error( - """ - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka('{kafka_broker}:19092', '{kafka_topic_old}', '{kafka_group_name_old}', '{kafka_format_json_each_row}', '\\n', '', 100) - SETTINGS kafka_commit_on_select = 1; - """ + thread_per_consumer = must_use_thread_per_consumer(create_query_generator) + + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + settings={ + "kafka_num_consumers": 100, + "kafka_thread_per_consumer": thread_per_consumer, + }, + ) + error = instance.query_and_get_error(create_query) + + assert ( + "BAD_ARGUMENTS" in error + and "The number of consumers can not be bigger than" in error ) - assert "BAD_ARGUMENTS" in error - instance.query( - """ + f""" SET kafka_disable_num_consumers_limit = 1; - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka('{kafka_broker}:19092', '{kafka_topic_old}', '{kafka_group_name_old}', '{kafka_format_json_each_row}', '\\n', '', 100) - SETTINGS kafka_commit_on_select = 1; + {create_query}; """ ) instance.query("DROP TABLE test.kafka") -def test_format_with_prefix_and_suffix(kafka_cluster): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_format_with_prefix_and_suffix(kafka_cluster, create_query_generator): + topic_name = "custom" + get_topic_postfix(create_query_generator) + + with kafka_topic(get_admin_client(kafka_cluster), topic_name): + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + format="CustomSeparated", + ) + instance.query( + f""" + DROP TABLE IF EXISTS test.kafka; + {create_query}; + """ + ) + + instance.query( + "INSERT INTO test.kafka select number*10 as key, number*100 as value from numbers(2) settings format_custom_result_before_delimiter='\n', format_custom_result_after_delimiter='\n'" + ) + + message_count = 2 + messages = kafka_consume_with_retry(kafka_cluster, topic_name, message_count) + + assert len(messages) == 2 + + assert ( + "".join(messages) + == "\n0\t0\n\n\n10\t100\n\n" + ) + + +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_max_rows_per_message(kafka_cluster, create_query_generator): + topic_name = "custom_max_rows_per_message" + get_topic_postfix( + create_query_generator ) - kafka_create_topic(admin_client, "custom") + with kafka_topic(get_admin_client(kafka_cluster), topic_name): + num_rows = 5 - instance.query( + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + format="CustomSeparated", + settings={ + "format_custom_result_before_delimiter": "\n", + "format_custom_result_after_delimiter": "\n", + "kafka_max_rows_per_message": 3, + }, + ) + instance.query( + f""" + DROP TABLE IF EXISTS test.view; + DROP TABLE IF EXISTS test.kafka; + {create_query}; + + CREATE MATERIALIZED VIEW test.view ENGINE=MergeTree ORDER BY (key, value) AS + SELECT key, value FROM test.kafka; """ - DROP TABLE IF EXISTS test.kafka; + ) - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = 'custom', - kafka_group_name = 'custom', - kafka_format = 'CustomSeparated'; - """ - ) + instance.query( + f"INSERT INTO test.kafka select number*10 as key, number*100 as value from numbers({num_rows}) settings format_custom_result_before_delimiter='\n', format_custom_result_after_delimiter='\n'" + ) - instance.query( - "INSERT INTO test.kafka select number*10 as key, number*100 as value from numbers(2) settings format_custom_result_before_delimiter='\n', format_custom_result_after_delimiter='\n'" - ) + message_count = 2 + messages = kafka_consume_with_retry(kafka_cluster, topic_name, message_count) - messages = [] + assert len(messages) == message_count - attempt = 0 - while attempt < 100: - messages.extend(kafka_consume(kafka_cluster, "custom")) - if len(messages) == 2: - break - attempt += 1 + assert ( + "".join(messages) + == "\n0\t0\n10\t100\n20\t200\n\n\n30\t300\n40\t400\n\n" + ) - assert len(messages) == 2 + instance.query_with_retry( + "SELECT count() FROM test.view", + check_callback=lambda res: int(res) == num_rows, + ) - assert ( - "".join(messages) == "\n0\t0\n\n\n10\t100\n\n" - ) - - kafka_delete_topic(admin_client, "custom") + result = instance.query("SELECT * FROM test.view") + assert result == "0\t0\n10\t100\n20\t200\n30\t300\n40\t400\n" -def test_max_rows_per_message(kafka_cluster): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) - - topic = "custom_max_rows_per_message" - - kafka_create_topic(admin_client, topic) - - num_rows = 5 - - instance.query( - f""" - DROP TABLE IF EXISTS test.view; - DROP TABLE IF EXISTS test.kafka; - - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = '{topic}', - kafka_group_name = '{topic}', - kafka_format = 'CustomSeparated', - format_custom_result_before_delimiter = '\n', - format_custom_result_after_delimiter = '\n', - kafka_max_rows_per_message = 3; - - CREATE MATERIALIZED VIEW test.view Engine=Log AS - SELECT key, value FROM test.kafka; - """ - ) - - instance.query( - f"INSERT INTO test.kafka select number*10 as key, number*100 as value from numbers({num_rows}) settings format_custom_result_before_delimiter='\n', format_custom_result_after_delimiter='\n'" - ) - - messages = [] - - attempt = 0 - while attempt < 500: - messages.extend(kafka_consume(kafka_cluster, topic)) - if len(messages) == 2: - break - attempt += 1 - - assert len(messages) == 2 - - assert ( - "".join(messages) - == "\n0\t0\n10\t100\n20\t200\n\n\n30\t300\n40\t400\n\n" - ) - - attempt = 0 - rows = 0 - while attempt < 500: - rows = int(instance.query("SELECT count() FROM test.view")) - if rows == num_rows: - break - attempt += 1 - - assert rows == num_rows - - result = instance.query("SELECT * FROM test.view") - assert result == "0\t0\n10\t100\n20\t200\n30\t300\n40\t400\n" - - kafka_delete_topic(admin_client, topic) - - -def test_row_based_formats(kafka_cluster): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_row_based_formats(kafka_cluster, create_query_generator): + admin_client = get_admin_client(kafka_cluster) for format_name in [ "TSV", @@ -4454,121 +4849,111 @@ def test_row_based_formats(kafka_cluster): "RowBinaryWithNamesAndTypes", "MsgPack", ]: - print(format_name) + logging.debug("Checking {format_name}") - kafka_create_topic(admin_client, format_name) + topic_name = format_name + get_topic_postfix(create_query_generator) + table_name = f"kafka_{format_name}" - num_rows = 10 + with kafka_topic(admin_client, topic_name): + num_rows = 10 + max_rows_per_message = 5 + message_count = num_rows / max_rows_per_message + create_query = create_query_generator( + table_name, + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + format=format_name, + settings={"kafka_max_rows_per_message": max_rows_per_message}, + ) + + instance.query( + f""" + DROP TABLE IF EXISTS test.view; + DROP TABLE IF EXISTS test.{table_name}; + + {create_query}; + + CREATE MATERIALIZED VIEW test.view ENGINE=MergeTree ORDER BY (key, value) AS + SELECT key, value FROM test.{table_name}; + + INSERT INTO test.{table_name} SELECT number * 10 as key, number * 100 as value FROM numbers({num_rows}); + """ + ) + + messages = kafka_consume_with_retry( + kafka_cluster, topic_name, message_count, need_decode=False + ) + + assert len(messages) == message_count + + instance.query_with_retry( + "SELECT count() FROM test.view", + check_callback=lambda res: int(res) == num_rows, + ) + + result = instance.query("SELECT * FROM test.view") + expected = "" + for i in range(num_rows): + expected += str(i * 10) + "\t" + str(i * 100) + "\n" + assert result == expected + + +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_block_based_formats_1(kafka_cluster, create_query_generator): + topic_name = "pretty_space" + get_topic_postfix(create_query_generator) + + with kafka_topic(get_admin_client(kafka_cluster), topic_name): + create_query = create_query_generator( + "kafka", + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + format="PrettySpace", + ) instance.query( f""" - DROP TABLE IF EXISTS test.view; DROP TABLE IF EXISTS test.kafka; - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = '{format_name}', - kafka_group_name = '{format_name}', - kafka_format = '{format_name}', - kafka_max_rows_per_message = 5; + {create_query}; - CREATE MATERIALIZED VIEW test.view Engine=Log AS - SELECT key, value FROM test.kafka; - - INSERT INTO test.kafka SELECT number * 10 as key, number * 100 as value FROM numbers({num_rows}); + INSERT INTO test.kafka SELECT number * 10 as key, number * 100 as value FROM numbers(5) settings max_block_size=2, optimize_trivial_insert_select=0, output_format_pretty_color=1, output_format_pretty_row_numbers=0; """ ) - messages = [] + message_count = 3 + messages = kafka_consume_with_retry(kafka_cluster, topic_name, message_count) + assert len(messages) == 3 - attempt = 0 - while attempt < 500: - messages.extend(kafka_consume(kafka_cluster, format_name, needDecode=False)) - if len(messages) == 2: - break - attempt += 1 + data = [] + for message in messages: + splitted = message.split("\n") + assert splitted[0] == " \x1b[1mkey\x1b[0m \x1b[1mvalue\x1b[0m" + assert splitted[1] == "" + assert splitted[-1] == "" + data += [line.split() for line in splitted[2:-1]] - assert len(messages) == 2 - - attempt = 0 - rows = 0 - while attempt < 500: - rows = int(instance.query("SELECT count() FROM test.view")) - if rows == num_rows: - break - attempt += 1 - - assert rows == num_rows - - result = instance.query("SELECT * FROM test.view") - expected = "" - for i in range(num_rows): - expected += str(i * 10) + "\t" + str(i * 100) + "\n" - assert result == expected - - kafka_delete_topic(admin_client, format_name) + assert data == [ + ["0", "0"], + ["10", "100"], + ["20", "200"], + ["30", "300"], + ["40", "400"], + ] -def test_block_based_formats_1(kafka_cluster): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) - - topic = "pretty_space" - kafka_create_topic(admin_client, topic) - - instance.query( - f""" - DROP TABLE IF EXISTS test.kafka; - - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = '{topic}', - kafka_group_name = '{topic}', - kafka_format = 'PrettySpace'; - - INSERT INTO test.kafka SELECT number * 10 as key, number * 100 as value FROM numbers(5) settings max_block_size=2, optimize_trivial_insert_select=0, output_format_pretty_color=1, output_format_pretty_row_numbers=0; - """ - ) - - messages = [] - - attempt = 0 - while attempt < 500: - messages.extend(kafka_consume(kafka_cluster, topic)) - if len(messages) == 3: - break - attempt += 1 - - assert len(messages) == 3 - - data = [] - for message in messages: - splitted = message.split("\n") - assert splitted[0] == " \x1b[1mkey\x1b[0m \x1b[1mvalue\x1b[0m" - assert splitted[1] == "" - assert splitted[-1] == "" - data += [line.split() for line in splitted[2:-1]] - - assert data == [ - ["0", "0"], - ["10", "100"], - ["20", "200"], - ["30", "300"], - ["40", "400"], - ] - - kafka_delete_topic(admin_client, topic) - - -def test_block_based_formats_2(kafka_cluster): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) - +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_block_based_formats_2(kafka_cluster, create_query_generator): + admin_client = get_admin_client(kafka_cluster) num_rows = 100 + message_count = 9 for format_name in [ "JSONColumns", @@ -4578,55 +4963,50 @@ def test_block_based_formats_2(kafka_cluster): "ORC", "JSONCompactColumns", ]: - kafka_create_topic(admin_client, format_name) + topic_name = format_name + get_topic_postfix(create_query_generator) + table_name = f"kafka_{format_name}" + logging.debug(f"Checking format {format_name}") + with kafka_topic(admin_client, topic_name): + create_query = create_query_generator( + table_name, + "key UInt64, value UInt64", + topic_list=topic_name, + consumer_group=topic_name, + format=format_name, + ) - instance.query( - f""" - DROP TABLE IF EXISTS test.view; - DROP TABLE IF EXISTS test.kafka; + instance.query( + f""" + DROP TABLE IF EXISTS test.view; + DROP TABLE IF EXISTS test.{table_name}; - CREATE TABLE test.kafka (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = '{format_name}', - kafka_group_name = '{format_name}', - kafka_format = '{format_name}'; + {create_query}; - CREATE MATERIALIZED VIEW test.view Engine=Log AS - SELECT key, value FROM test.kafka; + CREATE MATERIALIZED VIEW test.view ENGINE=MergeTree ORDER BY (key, value) AS + SELECT key, value FROM test.{table_name}; - INSERT INTO test.kafka SELECT number * 10 as key, number * 100 as value FROM numbers({num_rows}) settings max_block_size=12, optimize_trivial_insert_select=0; - """ - ) + INSERT INTO test.{table_name} SELECT number * 10 as key, number * 100 as value FROM numbers({num_rows}) settings max_block_size=12, optimize_trivial_insert_select=0; + """ + ) + messages = kafka_consume_with_retry( + kafka_cluster, topic_name, message_count, need_decode=False + ) + assert len(messages) == message_count - messages = [] + rows = int( + instance.query_with_retry( + "SELECT count() FROM test.view", + check_callback=lambda res: int(res) == num_rows, + ) + ) - attempt = 0 - while attempt < 500: - messages.extend(kafka_consume(kafka_cluster, format_name, needDecode=False)) - if len(messages) == 9: - break - attempt += 1 + assert rows == num_rows - assert len(messages) == 9 - - attempt = 0 - rows = 0 - while attempt < 500: - rows = int(instance.query("SELECT count() FROM test.view")) - if rows == num_rows: - break - attempt += 1 - - assert rows == num_rows - - result = instance.query("SELECT * FROM test.view ORDER by key") - expected = "" - for i in range(num_rows): - expected += str(i * 10) + "\t" + str(i * 100) + "\n" - assert result == expected - - kafka_delete_topic(admin_client, format_name) + result = instance.query("SELECT * FROM test.view ORDER by key") + expected = "" + for i in range(num_rows): + expected += str(i * 10) + "\t" + str(i * 100) + "\n" + assert result == expected def test_system_kafka_consumers(kafka_cluster): @@ -4962,137 +5342,186 @@ def test_formats_errors(kafka_cluster): "HiveText", "MySQLDump", ]: - kafka_create_topic(admin_client, format_name) - table_name = f"kafka_{format_name}" + with kafka_topic(admin_client, format_name): + table_name = f"kafka_{format_name}" + instance.query( + f""" + DROP TABLE IF EXISTS test.view; + DROP TABLE IF EXISTS test.{table_name}; + + CREATE TABLE test.{table_name} (key UInt64, value UInt64) + ENGINE = Kafka + SETTINGS kafka_broker_list = 'kafka1:19092', + kafka_topic_list = '{format_name}', + kafka_group_name = '{format_name}', + kafka_format = '{format_name}', + kafka_max_rows_per_message = 5, + format_template_row='template_row.format', + format_regexp='id: (.+?)', + input_format_with_names_use_header=0, + format_schema='key_value_message:Message'; + + CREATE MATERIALIZED VIEW test.view ENGINE=MergeTree ORDER BY (key, value) AS + SELECT key, value FROM test.{table_name}; + """ + ) + + kafka_produce( + kafka_cluster, + format_name, + ["Broken message\nBroken message\nBroken message\n"], + ) + + num_errors = int( + instance.query_with_retry( + f"SELECT length(exceptions.text) from system.kafka_consumers where database = 'test' and table = '{table_name}'", + check_callback=lambda res: int(res) > 0, + ) + ) + + assert num_errors > 0 + + instance.query(f"DROP TABLE test.{table_name}") + instance.query("DROP TABLE test.view") + + +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_multiple_read_in_materialized_views(kafka_cluster, create_query_generator): + topic_name = "multiple_read_from_mv" + get_topic_postfix(create_query_generator) + + with kafka_topic(get_admin_client(kafka_cluster), topic_name): + create_query = create_query_generator( + "kafka_multiple_read_input", + "id Int64", + topic_list=topic_name, + consumer_group=topic_name, + ) instance.query( f""" - DROP TABLE IF EXISTS test.view; - DROP TABLE IF EXISTS test.{table_name}; + DROP TABLE IF EXISTS test.kafka_multiple_read_input SYNC; + DROP TABLE IF EXISTS test.kafka_multiple_read_table; + DROP TABLE IF EXISTS test.kafka_multiple_read_mv; - CREATE TABLE test.{table_name} (key UInt64, value UInt64) - ENGINE = Kafka - SETTINGS kafka_broker_list = 'kafka1:19092', - kafka_topic_list = '{format_name}', - kafka_group_name = '{format_name}', - kafka_format = '{format_name}', - kafka_max_rows_per_message = 5, - format_template_row='template_row.format', - format_regexp='id: (.+?)', - input_format_with_names_use_header=0, - format_schema='key_value_message:Message'; + {create_query}; - CREATE MATERIALIZED VIEW test.view Engine=Log AS - SELECT key, value FROM test.{table_name}; - """ + CREATE TABLE test.kafka_multiple_read_table (id Int64) + ENGINE = MergeTree + ORDER BY id; + + + CREATE MATERIALIZED VIEW test.kafka_multiple_read_mv TO test.kafka_multiple_read_table AS + SELECT id + FROM test.kafka_multiple_read_input + WHERE id NOT IN ( + SELECT id + FROM test.kafka_multiple_read_table + WHERE id IN ( + SELECT id + FROM test.kafka_multiple_read_input + ) + ); + """ ) kafka_produce( - kafka_cluster, - format_name, - ["Broken message\nBroken message\nBroken message\n"], + kafka_cluster, topic_name, [json.dumps({"id": 42}), json.dumps({"id": 43})] ) - attempt = 0 - num_errors = 0 - while attempt < 200: - num_errors = int( - instance.query( - f"SELECT length(exceptions.text) from system.kafka_consumers where database = 'test' and table = '{table_name}'" - ) - ) - if num_errors > 0: - break - attempt += 1 + expected_result = "42\n43\n" + res = instance.query_with_retry( + f"SELECT id FROM test.kafka_multiple_read_table ORDER BY id", + check_callback=lambda res: res == expected_result, + ) + assert res == expected_result - assert num_errors > 0 + # Verify that the query deduplicates the records as it meant to be + messages = [] + for _ in range(0, 10): + messages.append(json.dumps({"id": 42})) + messages.append(json.dumps({"id": 43})) - kafka_delete_topic(admin_client, format_name) - instance.query(f"DROP TABLE test.{table_name}") - instance.query("DROP TABLE test.view") + messages.append(json.dumps({"id": 44})) + + kafka_produce(kafka_cluster, topic_name, messages) + + expected_result = "42\n43\n44\n" + res = instance.query_with_retry( + f"SELECT id FROM test.kafka_multiple_read_table ORDER BY id", + check_callback=lambda res: res == expected_result, + ) + assert res == expected_result + + instance.query( + f""" + DROP TABLE test.kafka_multiple_read_input; + DROP TABLE test.kafka_multiple_read_table; + DROP TABLE test.kafka_multiple_read_mv; + """ + ) -def test_multiple_read_in_materialized_views(kafka_cluster, max_retries=15): - admin_client = KafkaAdminClient( - bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) - ) - - topic = "multiple_read_from_mv" - kafka_create_topic(admin_client, topic) +@pytest.mark.parametrize( + "create_query_generator", + [generate_old_create_table_query, generate_new_create_table_query], +) +def test_kafka_null_message(kafka_cluster, create_query_generator): + topic_name = "null_message" instance.query( f""" - DROP TABLE IF EXISTS test.kafka_multiple_read_input; - DROP TABLE IF EXISTS test.kafka_multiple_read_table; - DROP TABLE IF EXISTS test.kafka_multiple_read_mv; + DROP TABLE IF EXISTS test.null_message_view; + DROP TABLE IF EXISTS test.null_message_consumer; + DROP TABLE IF EXISTS test.null_message_kafka; - CREATE TABLE test.kafka_multiple_read_input (id Int64) - ENGINE = Kafka - SETTINGS - kafka_broker_list = 'kafka1:19092', - kafka_topic_list = '{topic}', - kafka_group_name = '{topic}', - kafka_format = 'JSONEachRow'; + {create_query_generator("null_message_kafka", "value UInt64", topic_list=topic_name, consumer_group="mv")}; + CREATE TABLE test.null_message_view (value UInt64) + ENGINE = MergeTree() + ORDER BY value; + CREATE MATERIALIZED VIEW test.null_message_consumer TO test.null_message_view AS + SELECT * FROM test.null_message_kafka; + """ + ) - CREATE TABLE test.kafka_multiple_read_table (id Int64) - ENGINE = MergeTree - ORDER BY id; + message_key_values = [] + for i in range(5): + # Here the key is key for Kafka message + message = json.dumps({"value": i}) if i != 3 else None + message_key_values.append({"key": f"{i}".encode(), "message": message}) + producer = get_kafka_producer(kafka_cluster.kafka_port, producer_serializer, 15) + for message_kv in message_key_values: + producer.send( + topic=topic_name, key=message_kv["key"], value=message_kv["message"] + ) + producer.flush() - CREATE MATERIALIZED VIEW IF NOT EXISTS test.kafka_multiple_read_mv TO test.kafka_multiple_read_table AS - SELECT id - FROM test.kafka_multiple_read_input - WHERE id NOT IN ( - SELECT id - FROM test.kafka_multiple_read_table - WHERE id IN ( - SELECT id - FROM test.kafka_multiple_read_input - ) - ); + expected = TSV( """ +0 +1 +2 +4 +""" ) + with existing_kafka_topic(get_admin_client(kafka_cluster), topic_name): + result = instance.query_with_retry( + "SELECT * FROM test.null_message_view", + check_callback=lambda res: TSV(res) == expected, + ) - kafka_produce( - kafka_cluster, topic, [json.dumps({"id": 42}), json.dumps({"id": 43})] - ) + assert expected == TSV(result) - expected_result = "42\n43\n" - res = instance.query_with_retry( - f"SELECT id FROM test.kafka_multiple_read_table ORDER BY id", - retry_count=30, - sleep_time=0.5, - check_callback=lambda res: res == expected_result, - ) - assert res == expected_result - - # Verify that the query deduplicates the records as it meant to be - messages = [] - for i in range(0, 10): - messages.append(json.dumps({"id": 42})) - messages.append(json.dumps({"id": 43})) - - messages.append(json.dumps({"id": 44})) - - kafka_produce(kafka_cluster, topic, messages) - - expected_result = "42\n43\n44\n" - res = instance.query_with_retry( - f"SELECT id FROM test.kafka_multiple_read_table ORDER BY id", - retry_count=30, - sleep_time=0.5, - check_callback=lambda res: res == expected_result, - ) - assert res == expected_result - - kafka_delete_topic(admin_client, topic) - instance.query( - f""" - DROP TABLE test.kafka_multiple_read_input; - DROP TABLE test.kafka_multiple_read_table; - DROP TABLE test.kafka_multiple_read_mv; + instance.query( + """ + DROP TABLE test.null_message_consumer SYNC; + DROP TABLE test.null_message_view; + DROP TABLE test.null_message_kafka SYNC; """ - ) + ) if __name__ == "__main__": diff --git a/tests/integration/test_storage_kafka/test_produce_http_interface.py b/tests/integration/test_storage_kafka/test_produce_http_interface.py new file mode 100644 index 00000000000..fc10a07f239 --- /dev/null +++ b/tests/integration/test_storage_kafka/test_produce_http_interface.py @@ -0,0 +1,243 @@ +import time +import logging + +import pytest +from helpers.cluster import ClickHouseCluster, is_arm +from helpers.test_tools import TSV +from kafka import KafkaAdminClient +from kafka.admin import NewTopic + +if is_arm(): + pytestmark = pytest.mark.skip + +cluster = ClickHouseCluster(__file__) +instance = cluster.add_instance( + "instance", + main_configs=["configs/kafka.xml", "configs/named_collection.xml"], + user_configs=["configs/users.xml"], + with_kafka=True, + with_zookeeper=True, # For Replicated Table + macros={ + "kafka_broker": "kafka1", + "kafka_topic_old": "old", + "kafka_group_name_old": "old", + "kafka_topic_new": "new", + "kafka_group_name_new": "new", + "kafka_client_id": "instance", + "kafka_format_json_each_row": "JSONEachRow", + }, + clickhouse_path_dir="clickhouse_path", +) + + +@pytest.fixture(scope="module") +def kafka_cluster(): + try: + cluster.start() + kafka_id = instance.cluster.kafka_docker_id + print(("kafka_id is {}".format(kafka_id))) + yield cluster + finally: + cluster.shutdown() + + +@pytest.fixture(autouse=True) +def kafka_setup_teardown(): + instance.query("DROP DATABASE IF EXISTS test; CREATE DATABASE test;") + # logging.debug("kafka is available - running test") + yield # run test + + +def kafka_create_topic( + admin_client, + topic_name, + num_partitions=1, + replication_factor=1, + max_retries=50, + config=None, +): + logging.debug( + f"Kafka create topic={topic_name}, num_partitions={num_partitions}, replication_factor={replication_factor}" + ) + topics_list = [ + NewTopic( + name=topic_name, + num_partitions=num_partitions, + replication_factor=replication_factor, + topic_configs=config, + ) + ] + retries = 0 + while True: + try: + admin_client.create_topics(new_topics=topics_list, validate_only=False) + logging.debug("Admin client succeed") + return + except Exception as e: + retries += 1 + time.sleep(0.5) + if retries < max_retries: + logging.warning(f"Failed to create topic {e}") + else: + raise + + +def kafka_delete_topic(admin_client, topic, max_retries=50): + result = admin_client.delete_topics([topic]) + for topic, e in result.topic_error_codes: + if e == 0: + logging.debug(f"Topic {topic} deleted") + else: + logging.error(f"Failed to delete topic {topic}: {e}") + + retries = 0 + while True: + topics_listed = admin_client.list_topics() + logging.debug(f"TOPICS LISTED: {topics_listed}") + if topic not in topics_listed: + return + else: + retries += 1 + time.sleep(0.5) + if retries > max_retries: + raise Exception(f"Failed to delete topics {topic}, {result}") + + +def test_kafka_produce_http_interface_row_based_format(kafka_cluster): + # reproduction of #61060 with validating the written messages + admin_client = KafkaAdminClient( + bootstrap_servers="localhost:{}".format(kafka_cluster.kafka_port) + ) + + topic_prefix = "http_row_" + + # It is important to have: + # - long enough messages + # - enough messages + # I don't know the exact requirement for message sizes, but it doesn't reproduce with short messages + # For the number of messages it seems like at least 3 messages is necessary + expected_key = "01234567890123456789" + expected_value = "aaaaabbbbbccccc" + + insert_query_end = f"(key, value) VALUES ('{expected_key}', '{expected_value}'), ('{expected_key}', '{expected_value}'), ('{expected_key}', '{expected_value}')" + insert_query_template = "INSERT INTO {table_name} " + insert_query_end + + extra_settings = { + "Protobuf": ", kafka_schema = 'string_key_value.proto:StringKeyValuePair'", + "CapnProto": ", kafka_schema='string_key_value:StringKeyValuePair'", + "Template": ", format_template_row='string_key_value.format'", + } + + # Only the formats that can be used both and input and output format are tested + # Reasons to exclude following formats: + # - JSONStrings: not actually an input format + # - ProtobufSingle: I cannot make it work to parse the messages. Probably something is broken, + # because the producer can write multiple rows into a same message, which makes them impossible to parse properly. Should added after #67549 is fixed. + # - ProtobufList: I didn't want to deal with the envelope and stuff + # - Npy: supports only single column + # - LineAsString: supports only single column + # - RawBLOB: supports only single column + formats_to_test = [ + "TabSeparated", + "TabSeparatedRaw", + "TabSeparatedWithNames", + "TabSeparatedWithNamesAndTypes", + "TabSeparatedRawWithNames", + "TabSeparatedRawWithNamesAndTypes", + "Template", + "CSV", + "CSVWithNames", + "CSVWithNamesAndTypes", + "CustomSeparated", + "CustomSeparatedWithNames", + "CustomSeparatedWithNamesAndTypes", + "Values", + "JSON", + "JSONColumns", + "JSONColumnsWithMetadata", + "JSONCompact", + "JSONCompactColumns", + "JSONEachRow", + "JSONStringsEachRow", + "JSONCompactEachRow", + "JSONCompactEachRowWithNames", + "JSONCompactEachRowWithNamesAndTypes", + "JSONCompactStringsEachRow", + "JSONCompactStringsEachRowWithNames", + "JSONCompactStringsEachRowWithNamesAndTypes", + "JSONObjectEachRow", + "BSONEachRow", + "TSKV", + "Protobuf", + "Avro", + "Parquet", + "Arrow", + "ArrowStream", + "ORC", + "RowBinary", + "RowBinaryWithNames", + "RowBinaryWithNamesAndTypes", + "Native", + "CapnProto", + "MsgPack", + ] + for format in formats_to_test: + logging.debug(f"Creating tables and writing messages to {format}") + topic = topic_prefix + format + kafka_create_topic(admin_client, topic) + + extra_setting = extra_settings.get(format, "") + + # kafka_max_rows_per_message is set to 2 to make sure every format produces at least 2 messages, thus increasing the chance of catching a bug + instance.query( + f""" + DROP TABLE IF EXISTS test.view_{topic}; + DROP TABLE IF EXISTS test.consumer_{topic}; + CREATE TABLE test.kafka_writer_{topic} (key String, value String) + ENGINE = Kafka + SETTINGS kafka_broker_list = 'kafka1:19092', + kafka_topic_list = '{topic}', + kafka_group_name = '{topic}', + kafka_format = '{format}', + kafka_max_rows_per_message = 2 {extra_setting}; + + CREATE TABLE test.kafka_{topic} (key String, value String) + ENGINE = Kafka + SETTINGS kafka_broker_list = 'kafka1:19092', + kafka_topic_list = '{topic}', + kafka_group_name = '{topic}', + kafka_format = '{format}' {extra_setting}; + + CREATE MATERIALIZED VIEW test.view_{topic} Engine=Log AS + SELECT key, value FROM test.kafka_{topic}; + """ + ) + instance.http_query( + insert_query_template.format(table_name="test.kafka_writer_" + topic), + method="POST", + ) + + expected = f"""\ +{expected_key}\t{expected_value} +{expected_key}\t{expected_value} +{expected_key}\t{expected_value} +""" + # give some times for the readers to read the messages + for format in formats_to_test: + logging.debug(f"Checking result for {format}") + topic = topic_prefix + format + + result = instance.query_with_retry( + f"SELECT * FROM test.view_{topic}", + check_callback=lambda res: res.count("\n") == 3, + ) + + assert TSV(result) == TSV(expected) + + kafka_delete_topic(admin_client, topic) + + +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 5948954ff5f..c724c5bb498 100644 --- a/tests/integration/test_storage_mysql/test.py +++ b/tests/integration/test_storage_mysql/test.py @@ -445,7 +445,7 @@ def test_mysql_distributed(started_cluster): query = "SELECT * FROM (" for i in range(3): query += "SELECT name FROM test_replicas UNION DISTINCT " - query += "SELECT name FROM test_replicas)" + query += "SELECT name FROM test_replicas) ORDER BY name" result = node2.query(query) assert result == "host2\nhost3\nhost4\n" @@ -827,6 +827,9 @@ def test_settings(started_cluster): f"with settings: connect_timeout={connect_timeout}, read_write_timeout={rw_timeout}" ) + node1.query("DROP DATABASE IF EXISTS m") + node1.query("DROP DATABASE IF EXISTS mm") + rw_timeout = 40123001 connect_timeout = 40123002 node1.query( @@ -855,6 +858,9 @@ def test_settings(started_cluster): f"with settings: connect_timeout={connect_timeout}, read_write_timeout={rw_timeout}" ) + node1.query("DROP DATABASE m") + node1.query("DROP DATABASE mm") + drop_mysql_table(conn, table_name) conn.close() @@ -930,6 +936,9 @@ def test_joins(started_cluster): conn.commit() + node1.query("DROP TABLE IF EXISTS test_joins_table_users") + node1.query("DROP TABLE IF EXISTS test_joins_table_tickets") + node1.query( """ CREATE TABLE test_joins_table_users @@ -964,6 +973,9 @@ def test_joins(started_cluster): """ ) == "281607\tFeedback\t2024-06-25 12:09:41\tuser@example.com\n" + node1.query("DROP TABLE test_joins_table_users") + node1.query("DROP TABLE test_joins_table_tickets") + if __name__ == "__main__": with contextmanager(started_cluster)() as cluster: diff --git a/tests/integration/test_storage_s3_queue/test.py b/tests/integration/test_storage_s3_queue/test.py index 92d6f181464..9e3ee19179a 100644 --- a/tests/integration/test_storage_s3_queue/test.py +++ b/tests/integration/test_storage_s3_queue/test.py @@ -1,6 +1,7 @@ import io import logging import random +import string import time import pytest @@ -13,7 +14,6 @@ from uuid import uuid4 AVAILABLE_MODES = ["unordered", "ordered"] DEFAULT_AUTH = ["'minio'", "'minio123'"] NO_AUTH = ["NOSIGN"] -AZURE_CONTAINER_NAME = "cont" def prepare_public_s3_bucket(started_cluster): @@ -68,13 +68,24 @@ def s3_queue_setup_teardown(started_cluster): instance = started_cluster.instances["instance"] instance_2 = started_cluster.instances["instance2"] - instance.query("DROP DATABASE IF EXISTS test; CREATE DATABASE test;") - instance_2.query("DROP DATABASE IF EXISTS test; CREATE DATABASE test;") + instance.query("DROP DATABASE IF EXISTS default; CREATE DATABASE default;") + instance_2.query("DROP DATABASE IF EXISTS default; CREATE DATABASE default;") minio = started_cluster.minio_client objects = list(minio.list_objects(started_cluster.minio_bucket, recursive=True)) for obj in objects: minio.remove_object(started_cluster.minio_bucket, obj.object_name) + + container_client = started_cluster.blob_service_client.get_container_client( + started_cluster.azurite_container + ) + + if container_client.exists(): + blob_names = [b.name for b in container_client.list_blobs()] + logging.debug(f"Deleting blobs: {blob_names}") + for b in blob_names: + container_client.delete_blob(b) + yield # run test @@ -129,11 +140,6 @@ 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() @@ -190,7 +196,7 @@ def put_s3_file_content(started_cluster, filename, data, bucket=None): def put_azure_file_content(started_cluster, filename, data, bucket=None): client = started_cluster.blob_service_client.get_blob_client( - AZURE_CONTAINER_NAME, filename + started_cluster.azurite_container, filename ) buf = io.BytesIO(data) client.upload_blob(buf, "BlockBlob", len(data)) @@ -226,7 +232,7 @@ def create_table( 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')" + engine_def = f"{engine_name}('{started_cluster.env_variables['AZURITE_CONNECTION_STRING']}', '{started_cluster.azurite_container}', '{files_path}/', 'CSV')" node.query(f"DROP TABLE IF EXISTS {table_name}") create_query = f""" @@ -262,15 +268,21 @@ def create_mv( ) +def generate_random_string(length=6): + return "".join(random.choice(string.ascii_lowercase) for i in range(length)) + + @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}_{engine_name}" + table_name = f"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 + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" if engine_name == "S3Queue": storage = "s3" else: @@ -285,7 +297,7 @@ def test_delete_after_processing(started_cluster, mode, engine_name): table_name, mode, files_path, - additional_settings={"after_processing": "delete"}, + additional_settings={"after_processing": "delete", "keeper_path": keeper_path}, engine_name=engine_name, ) create_mv(node, table_name, dst_table_name) @@ -313,7 +325,7 @@ def test_delete_after_processing(started_cluster, mode, engine_name): assert len(objects) == 0 else: client = started_cluster.blob_service_client.get_container_client( - AZURE_CONTAINER_NAME + started_cluster.azurite_container ) objects_iterator = client.list_blobs(files_path) for objects in objects_iterator: @@ -324,11 +336,12 @@ def test_delete_after_processing(started_cluster, mode, engine_name): @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}_{engine_name}" + table_name = f"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" - keeper_path = f"/clickhouse/test_{table_name}" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" retries_num = 3 values = [ @@ -385,8 +398,9 @@ def test_failed_retry(started_cluster, mode, engine_name): @pytest.mark.parametrize("mode", AVAILABLE_MODES) def test_direct_select_file(started_cluster, mode): node = started_cluster.instances["instance"] - table_name = f"test.direct_select_file_{mode}" - keeper_path = f"/clickhouse/test_{table_name}" + table_name = f"direct_select_file_{mode}" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{mode}_{generate_random_string()}" files_path = f"{table_name}_data" file_path = f"{files_path}/test.csv" @@ -447,7 +461,7 @@ def test_direct_select_file(started_cluster, mode): ] == [] # New table with different zookeeper path - keeper_path = f"/clickhouse/test_{table_name}_{mode}_2" + keeper_path = f"{keeper_path}_2" create_table( started_cluster, node, @@ -491,8 +505,17 @@ def test_direct_select_multiple_files(started_cluster, mode): node = started_cluster.instances["instance"] table_name = f"direct_select_multiple_files_{mode}" files_path = f"{table_name}_data" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" - create_table(started_cluster, node, table_name, mode, files_path) + create_table( + started_cluster, + node, + table_name, + mode, + files_path, + additional_settings={"keeper_path": keeper_path}, + ) for i in range(5): rand_values = [[random.randint(0, 50) for _ in range(3)] for _ in range(10)] values_csv = ( @@ -515,14 +538,23 @@ def test_direct_select_multiple_files(started_cluster, mode): @pytest.mark.parametrize("mode", AVAILABLE_MODES) -def test_streaming_to_view_(started_cluster, mode): +def test_streaming_to_view(started_cluster, mode): node = started_cluster.instances["instance"] table_name = f"streaming_to_view_{mode}" dst_table_name = f"{table_name}_dst" files_path = f"{table_name}_data" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" total_values = generate_random_files(started_cluster, files_path, 10) - create_table(started_cluster, node, table_name, mode, files_path) + create_table( + started_cluster, + node, + table_name, + mode, + files_path, + additional_settings={"keeper_path": keeper_path}, + ) create_mv(node, table_name, dst_table_name) expected_values = set([tuple(i) for i in total_values]) @@ -544,7 +576,8 @@ def test_streaming_to_many_views(started_cluster, mode): node = started_cluster.instances["instance"] table_name = f"streaming_to_many_views_{mode}" dst_table_name = f"{table_name}_dst" - keeper_path = f"/clickhouse/test_{table_name}" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" files_path = f"{table_name}_data" for i in range(3): @@ -582,7 +615,8 @@ def test_streaming_to_many_views(started_cluster, mode): def test_multiple_tables_meta_mismatch(started_cluster): node = started_cluster.instances["instance"] table_name = f"multiple_tables_meta_mismatch" - keeper_path = f"/clickhouse/test_{table_name}" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" files_path = f"{table_name}_data" create_table( @@ -675,7 +709,8 @@ def test_multiple_tables_streaming_sync(started_cluster, mode): node = started_cluster.instances["instance"] table_name = f"multiple_tables_streaming_sync_{mode}" dst_table_name = f"{table_name}_dst" - keeper_path = f"/clickhouse/test_{table_name}" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" files_path = f"{table_name}_data" files_to_generate = 300 @@ -756,7 +791,10 @@ def test_multiple_tables_streaming_sync(started_cluster, mode): def test_multiple_tables_streaming_sync_distributed(started_cluster, mode): node = started_cluster.instances["instance"] node_2 = started_cluster.instances["instance2"] - table_name = f"multiple_tables_streaming_sync_distributed_{mode}" + # A unique table name is necessary for repeatable tests + table_name = ( + f"multiple_tables_streaming_sync_distributed_{mode}_{generate_random_string()}" + ) dst_table_name = f"{table_name}_dst" keeper_path = f"/clickhouse/test_{table_name}" files_path = f"{table_name}_data" @@ -771,7 +809,11 @@ def test_multiple_tables_streaming_sync_distributed(started_cluster, mode): table_name, mode, files_path, - additional_settings={"keeper_path": keeper_path, "s3queue_buckets": 2}, + additional_settings={ + "keeper_path": keeper_path, + "s3queue_buckets": 2, + **({"s3queue_processing_threads_num": 1} if mode == "ordered" else {}), + }, ) for instance in [node, node_2]: @@ -806,6 +848,10 @@ def test_multiple_tables_streaming_sync_distributed(started_cluster, mode): list(map(int, l.split())) for l in run_query(node_2, get_query).splitlines() ] + logging.debug( + f"res1 size: {len(res1)}, res2 size: {len(res2)}, total_rows: {total_rows}" + ) + assert len(res1) + len(res2) == total_rows # Checking that all engines have made progress @@ -825,7 +871,8 @@ def test_max_set_age(started_cluster): node = started_cluster.instances["instance"] table_name = "max_set_age" dst_table_name = f"{table_name}_dst" - keeper_path = f"/clickhouse/test_{table_name}" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" files_path = f"{table_name}_data" max_age = 20 files_to_generate = 10 @@ -936,10 +983,9 @@ def test_max_set_age(started_cluster): def test_max_set_size(started_cluster): node = started_cluster.instances["instance"] table_name = f"max_set_size" - dst_table_name = f"{table_name}_dst" - keeper_path = f"/clickhouse/test_{table_name}" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" files_path = f"{table_name}_data" - max_age = 10 files_to_generate = 10 create_table( @@ -983,7 +1029,8 @@ def test_drop_table(started_cluster): node = started_cluster.instances["instance"] table_name = f"test_drop" dst_table_name = f"{table_name}_dst" - keeper_path = f"/clickhouse/test_{table_name}" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" files_path = f"{table_name}_data" files_to_generate = 300 @@ -1013,9 +1060,11 @@ def test_drop_table(started_cluster): def test_s3_client_reused(started_cluster): node = started_cluster.instances["instance"] - table_name = f"test.test_s3_client_reused" + table_name = f"test_s3_client_reused" dst_table_name = f"{table_name}_dst" files_path = f"{table_name}_data" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" row_num = 10 def get_created_s3_clients_count(): @@ -1049,6 +1098,7 @@ def test_s3_client_reused(started_cluster): additional_settings={ "after_processing": "delete", "s3queue_processing_threads_num": 1, + "keeper_path": keeper_path, }, auth=NO_AUTH, bucket=started_cluster.minio_public_bucket, @@ -1106,7 +1156,8 @@ def test_processing_threads(started_cluster, mode): node = started_cluster.instances["instance"] table_name = f"processing_threads_{mode}" dst_table_name = f"{table_name}_dst" - keeper_path = f"/clickhouse/test_{table_name}" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" files_path = f"{table_name}_data" files_to_generate = 300 processing_threads = 32 @@ -1173,7 +1224,8 @@ def test_shards(started_cluster, mode, processing_threads): node = started_cluster.instances["instance"] table_name = f"test_shards_{mode}_{processing_threads}" dst_table_name = f"{table_name}_dst" - keeper_path = f"/clickhouse/test_{table_name}" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" files_path = f"{table_name}_data" files_to_generate = 300 shards_num = 3 @@ -1292,7 +1344,7 @@ where zookeeper_path ilike '%{table_name}%' and status = 'Processed' and rows_pr pytest.param("unordered", 1), pytest.param("unordered", 8), pytest.param("ordered", 1), - pytest.param("ordered", 8), + pytest.param("ordered", 2), ], ) def test_shards_distributed(started_cluster, mode, processing_threads): @@ -1300,10 +1352,11 @@ def test_shards_distributed(started_cluster, mode, processing_threads): node_2 = started_cluster.instances["instance2"] table_name = f"test_shards_distributed_{mode}_{processing_threads}" dst_table_name = f"{table_name}_dst" - keeper_path = f"/clickhouse/test_{table_name}" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" files_path = f"{table_name}_data" files_to_generate = 300 - row_num = 50 + row_num = 300 total_rows = row_num * files_to_generate shards_num = 2 @@ -1453,8 +1506,8 @@ def test_settings_check(started_cluster): node = started_cluster.instances["instance"] node_2 = started_cluster.instances["instance2"] table_name = f"test_settings_check" - dst_table_name = f"{table_name}_dst" - keeper_path = f"/clickhouse/test_{table_name}" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" files_path = f"{table_name}_data" mode = "ordered" @@ -1496,7 +1549,10 @@ def test_processed_file_setting(started_cluster, processing_threads): node = started_cluster.instances["instance"] table_name = f"test_processed_file_setting_{processing_threads}" dst_table_name = f"{table_name}_dst" - keeper_path = f"/clickhouse/test_{table_name}_{processing_threads}" + # A unique path is necessary for repeatable tests + keeper_path = ( + f"/clickhouse/test_{table_name}_{processing_threads}_{generate_random_string()}" + ) files_path = f"{table_name}_data" files_to_generate = 10 @@ -1547,7 +1603,10 @@ def test_processed_file_setting_distributed(started_cluster, processing_threads) node_2 = started_cluster.instances["instance2"] table_name = f"test_processed_file_setting_distributed_{processing_threads}" dst_table_name = f"{table_name}_dst" - keeper_path = f"/clickhouse/test_{table_name}" + # A unique path is necessary for repeatable tests + keeper_path = ( + f"/clickhouse/test_{table_name}_{processing_threads}_{generate_random_string()}" + ) files_path = f"{table_name}_data" files_to_generate = 10 @@ -1601,7 +1660,8 @@ def test_upgrade(started_cluster): table_name = f"test_upgrade" dst_table_name = f"{table_name}_dst" - keeper_path = f"/clickhouse/test_{table_name}" + # A unique path is necessary for repeatable tests + keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" files_path = f"{table_name}_data" files_to_generate = 10 @@ -1640,7 +1700,8 @@ def test_upgrade(started_cluster): def test_exception_during_insert(started_cluster): node = started_cluster.instances["instance_too_many_parts"] - table_name = f"test_exception_during_insert" + # A unique table name is necessary for repeatable tests + table_name = f"test_exception_during_insert_{generate_random_string()}" dst_table_name = f"{table_name}_dst" keeper_path = f"/clickhouse/test_{table_name}" files_path = f"{table_name}_data" @@ -1656,6 +1717,7 @@ def test_exception_during_insert(started_cluster): "keeper_path": keeper_path, }, ) + node.rotate_logs() total_values = generate_random_files( started_cluster, files_path, files_to_generate, start_ind=0, row_num=1 ) @@ -1672,33 +1734,49 @@ def test_exception_during_insert(started_cluster): ) assert "Too many parts" in exception + original_parts_to_throw_insert = 0 + modified_parts_to_throw_insert = 10 node.replace_in_config( "/etc/clickhouse-server/config.d/merge_tree.xml", - "parts_to_throw_insert>0", - "parts_to_throw_insert>10", + f"parts_to_throw_insert>{original_parts_to_throw_insert}", + f"parts_to_throw_insert>{modified_parts_to_throw_insert}", ) - node.restart_clickhouse() + try: + node.restart_clickhouse() - def get_count(): - return int(node.query(f"SELECT count() FROM {dst_table_name}")) + def get_count(): + return int(node.query(f"SELECT count() FROM {dst_table_name}")) - expected_rows = 10 - for _ in range(20): - if expected_rows == get_count(): - break - time.sleep(1) - assert expected_rows == get_count() + expected_rows = 10 + for _ in range(20): + if expected_rows == get_count(): + break + time.sleep(1) + assert expected_rows == get_count() + finally: + node.replace_in_config( + "/etc/clickhouse-server/config.d/merge_tree.xml", + f"parts_to_throw_insert>{modified_parts_to_throw_insert}", + f"parts_to_throw_insert>{original_parts_to_throw_insert}", + ) + node.restart_clickhouse() def test_commit_on_limit(started_cluster): node = started_cluster.instances["instance"] - table_name = f"test_commit_on_limit" + # A unique table name is necessary for repeatable tests + table_name = f"test_commit_on_limit_{generate_random_string()}" dst_table_name = f"{table_name}_dst" keeper_path = f"/clickhouse/test_{table_name}" files_path = f"{table_name}_data" files_to_generate = 10 + failed_files_event_before = int( + node.query( + "SELECT value FROM system.events WHERE name = 'ObjectStorageQueueFailedFiles' SETTINGS system_events_show_zero_values=1" + ) + ) create_table( started_cluster, node, @@ -1774,6 +1852,9 @@ def test_commit_on_limit(started_cluster): assert "test_999999.csv" in get_processed_files() assert 1 == int( + node.count_in_log(f"Setting file {files_path}/test_9999.csv as failed") + ) + assert failed_files_event_before + 1 == int( node.query( "SELECT value FROM system.events WHERE name = 'ObjectStorageQueueFailedFiles' SETTINGS system_events_show_zero_values=1" ) diff --git a/tests/integration/test_system_flush_logs/test.py b/tests/integration/test_system_flush_logs/test.py index 713b327eb76..cfecea5b3d6 100644 --- a/tests/integration/test_system_flush_logs/test.py +++ b/tests/integration/test_system_flush_logs/test.py @@ -4,7 +4,7 @@ import pytest from helpers.cluster import ClickHouseCluster -from helpers.test_tools import assert_eq_with_retry +from helpers.test_tools import assert_eq_with_retry, assert_logs_contain_with_retry, TSV cluster = ClickHouseCluster(__file__) node = cluster.add_instance( @@ -12,17 +12,6 @@ node = cluster.add_instance( stay_alive=True, ) -system_logs = [ - # enabled by default - ("system.text_log", 1), - ("system.query_log", 1), - ("system.query_thread_log", 1), - ("system.part_log", 1), - ("system.trace_log", 1), - ("system.metric_log", 1), - ("system.error_log", 1), -] - @pytest.fixture(scope="module", autouse=True) def start_cluster(): @@ -33,22 +22,28 @@ def start_cluster(): cluster.shutdown() -@pytest.fixture(scope="function") -def flush_logs(): +def test_system_logs_exists(): + system_logs = [ + ("system.text_log", 1), + ("system.query_log", 1), + ("system.query_thread_log", 1), + ("system.part_log", 1), + ("system.trace_log", 1), + ("system.metric_log", 1), + ("system.error_log", 1), + ] + node.query("SYSTEM FLUSH LOGS") - - -@pytest.mark.parametrize("table,exists", system_logs) -def test_system_logs(flush_logs, table, exists): - q = "SELECT * FROM {}".format(table) - if exists: - node.query(q) - else: - response = node.query_and_get_error(q) - assert ( - "Table {} does not exist".format(table) in response - or "Unknown table expression identifier '{}'".format(table) in response - ) + for table, exists in system_logs: + q = "SELECT * FROM {}".format(table) + if exists: + node.query(q) + else: + response = node.query_and_get_error(q) + assert ( + "Table {} does not exist".format(table) in response + or "Unknown table expression identifier '{}'".format(table) in response + ) # Logic is tricky, let's check that there is no hang in case of message queue @@ -67,14 +62,19 @@ def test_system_logs_non_empty_queue(): def test_system_suspend(): - node.query("CREATE TABLE t (x DateTime) ENGINE=Memory;") - node.query("INSERT INTO t VALUES (now());") - node.query("SYSTEM SUSPEND FOR 1 SECOND;") - node.query("INSERT INTO t VALUES (now());") - assert "1\n" == node.query("SELECT max(x) - min(x) >= 1 FROM t;") + try: + node.query("CREATE TABLE t (x DateTime) ENGINE=Memory;") + node.query("INSERT INTO t VALUES (now());") + node.query("SYSTEM SUSPEND FOR 1 SECOND;") + node.query("INSERT INTO t VALUES (now());") + assert "1\n" == node.query("SELECT max(x) - min(x) >= 1 FROM t;") + finally: + node.query("DROP TABLE IF EXISTS t;") def test_log_max_size(start_cluster): + # we do misconfiguration here: buffer_size_rows_flush_threshold > max_size_rows, flush_interval_milliseconds is huge + # no auto flush by size not by time has a chance node.exec_in_container( [ "bash", @@ -83,6 +83,7 @@ def test_log_max_size(start_cluster): 1000000 + 1000000 10 10 @@ -91,11 +92,24 @@ def test_log_max_size(start_cluster): """, ] ) - node.restart_clickhouse() - for i in range(10): - node.query(f"select {i}") - assert node.query("select count() >= 10 from system.query_log") == "1\n" + node.query("SYSTEM FLUSH LOGS") + node.query(f"TRUNCATE TABLE IF EXISTS system.query_log") + node.restart_clickhouse() + + # all logs records above max_size_rows are lost + # The accepted logs records are never flushed until system flush logs is called by us + for i in range(21): + node.query(f"select {i}") + node.query("system flush logs") + + assert_logs_contain_with_retry( + node, "Queue had been full at 0, accepted 10 logs, ignored 34 logs." + ) + assert node.query( + "select count() >= 10, count() < 20 from system.query_log" + ) == TSV([[1, 1]]) + node.exec_in_container( ["rm", f"/etc/clickhouse-server/config.d/yyy-override-query_log.xml"] ) diff --git a/tests/integration/test_system_logs_recreate/test.py b/tests/integration/test_system_logs_recreate/test.py index 1bdb1fe3261..8b84734ed02 100644 --- a/tests/integration/test_system_logs_recreate/test.py +++ b/tests/integration/test_system_logs_recreate/test.py @@ -33,124 +33,139 @@ def test_system_logs_recreate(): "error_log", ] - node.query("SYSTEM FLUSH LOGS") - for table in system_logs: - assert "ENGINE = MergeTree" in node.query(f"SHOW CREATE TABLE system.{table}") - assert "ENGINE = Null" not in node.query(f"SHOW CREATE TABLE system.{table}") - assert ( - len( - node.query(f"SHOW TABLES FROM system LIKE '{table}%'") - .strip() - .split("\n") + try: + node.query("SYSTEM FLUSH LOGS") + for table in system_logs: + assert "ENGINE = MergeTree" in node.query( + f"SHOW CREATE TABLE system.{table}" ) - == 1 - ) - - # NOTE: we use zzz- prefix to make it the last file, - # so that it will be applied last. - for table in system_logs: - node.exec_in_container( - [ - "bash", - "-c", - f"""echo " - - <{table}> - ENGINE = Null - - - - " > /etc/clickhouse-server/config.d/zzz-override-{table}.xml - """, - ] - ) - - node.restart_clickhouse() - node.query("SYSTEM FLUSH LOGS") - for table in system_logs: - assert "ENGINE = MergeTree" not in node.query( - f"SHOW CREATE TABLE system.{table}" - ) - assert "ENGINE = Null" in node.query(f"SHOW CREATE TABLE system.{table}") - assert ( - len( - node.query(f"SHOW TABLES FROM system LIKE '{table}%'") - .strip() - .split("\n") + assert "ENGINE = Null" not in node.query( + f"SHOW CREATE TABLE system.{table}" ) - == 2 - ) - - # apply only storage_policy for all system tables - for table in system_logs: - node.exec_in_container( - [ - "bash", - "-c", - f"""echo " - - <{table}> - system_tables - - - " > /etc/clickhouse-server/config.d/zzz-override-{table}.xml - """, - ] - ) - node.restart_clickhouse() - node.query("SYSTEM FLUSH LOGS") - import logging - - for table in system_logs: - create_table_sql = node.query(f"SHOW CREATE TABLE system.{table} FORMAT TSVRaw") - logging.debug( - "With storage policy, SHOW CREATE TABLE system.%s is: %s", - table, - create_table_sql, - ) - assert "ENGINE = MergeTree" in create_table_sql - assert "ENGINE = Null" not in create_table_sql - assert "SETTINGS storage_policy = 'system_tables'" in create_table_sql - assert ( - len( - node.query(f"SHOW TABLES FROM system LIKE '{table}%'") - .strip() - .split("\n") + assert ( + len( + node.query(f"SHOW TABLES FROM system LIKE '{table}%'") + .strip() + .split("\n") + ) + == 1 ) - == 3 - ) - for table in system_logs: - node.exec_in_container( - ["rm", f"/etc/clickhouse-server/config.d/zzz-override-{table}.xml"] - ) - - node.restart_clickhouse() - node.query("SYSTEM FLUSH LOGS") - for table in system_logs: - assert "ENGINE = MergeTree" in node.query(f"SHOW CREATE TABLE system.{table}") - assert "ENGINE = Null" not in node.query(f"SHOW CREATE TABLE system.{table}") - assert ( - len( - node.query(f"SHOW TABLES FROM system LIKE '{table}%'") - .strip() - .split("\n") + # NOTE: we use zzz- prefix to make it the last file, + # so that it will be applied last. + for table in system_logs: + node.exec_in_container( + [ + "bash", + "-c", + f"""echo " + + <{table}> + ENGINE = Null + + + + " > /etc/clickhouse-server/config.d/zzz-override-{table}.xml + """, + ] ) - == 4 - ) - node.query("SYSTEM FLUSH LOGS") - # Ensure that there was no superfluous RENAME's - # IOW that the table created only when the structure is indeed different. - for table in system_logs: - assert ( - len( - node.query(f"SHOW TABLES FROM system LIKE '{table}%'") - .strip() - .split("\n") + node.restart_clickhouse() + node.query("SYSTEM FLUSH LOGS") + for table in system_logs: + assert "ENGINE = MergeTree" not in node.query( + f"SHOW CREATE TABLE system.{table}" ) - == 4 - ) + assert "ENGINE = Null" in node.query(f"SHOW CREATE TABLE system.{table}") + assert ( + len( + node.query(f"SHOW TABLES FROM system LIKE '{table}%'") + .strip() + .split("\n") + ) + == 2 + ) + + # apply only storage_policy for all system tables + for table in system_logs: + node.exec_in_container( + [ + "bash", + "-c", + f"""echo " + + <{table}> + system_tables + + + " > /etc/clickhouse-server/config.d/zzz-override-{table}.xml + """, + ] + ) + node.restart_clickhouse() + node.query("SYSTEM FLUSH LOGS") + import logging + + for table in system_logs: + create_table_sql = node.query( + f"SHOW CREATE TABLE system.{table} FORMAT TSVRaw" + ) + logging.debug( + "With storage policy, SHOW CREATE TABLE system.%s is: %s", + table, + create_table_sql, + ) + assert "ENGINE = MergeTree" in create_table_sql + assert "ENGINE = Null" not in create_table_sql + assert "SETTINGS storage_policy = 'system_tables'" in create_table_sql + assert ( + len( + node.query(f"SHOW TABLES FROM system LIKE '{table}%'") + .strip() + .split("\n") + ) + == 3 + ) + + for table in system_logs: + node.exec_in_container( + ["rm", f"/etc/clickhouse-server/config.d/zzz-override-{table}.xml"] + ) + + node.restart_clickhouse() + node.query("SYSTEM FLUSH LOGS") + for table in system_logs: + assert "ENGINE = MergeTree" in node.query( + f"SHOW CREATE TABLE system.{table}" + ) + assert "ENGINE = Null" not in node.query( + f"SHOW CREATE TABLE system.{table}" + ) + assert ( + len( + node.query(f"SHOW TABLES FROM system LIKE '{table}%'") + .strip() + .split("\n") + ) + == 4 + ) + + node.query("SYSTEM FLUSH LOGS") + # Ensure that there was no superfluous RENAME's + # IOW that the table created only when the structure is indeed different. + for table in system_logs: + assert ( + len( + node.query(f"SHOW TABLES FROM system LIKE '{table}%'") + .strip() + .split("\n") + ) + == 4 + ) + finally: + for table in system_logs: + for syffix in range(3): + node.query(f"DROP TABLE IF EXISTS system.{table}_{syffix} sync") def test_drop_system_log(): @@ -173,11 +188,20 @@ def test_drop_system_log(): node.query("system flush logs") node.query("select 2") node.query("system flush logs") - assert node.query("select count() > 0 from system.query_log") == "1\n" + assert node.query("select count() >= 2 from system.query_log") == "1\n" + node.query("drop table system.query_log sync") node.query("select 3") node.query("system flush logs") - assert node.query("select count() > 0 from system.query_log") == "1\n" + assert node.query("select count() >= 1 from system.query_log") == "1\n" + + node.query("drop table system.query_log sync") + node.restart_clickhouse() + node.query("system flush logs") + assert ( + node.query("select count() >= 0 from system.query_log") == "1\n" + ) # we check that query_log just exists + node.exec_in_container( ["rm", f"/etc/clickhouse-server/config.d/yyy-override-query_log.xml"] ) diff --git a/tests/integration/test_throttling/test.py b/tests/integration/test_throttling/test.py index c53c2bb1ddf..4bd96e2756d 100644 --- a/tests/integration/test_throttling/test.py +++ b/tests/integration/test_throttling/test.py @@ -121,21 +121,15 @@ def node_update_config(mode, setting, value=None): node.restart_clickhouse() -def assert_took(took, should_took): +def assert_took(took, should_take): # we need to decrease the lower limit because the server limits could # be enforced by throttling some server background IO instead of query IO # and we have no control over it - # - # and the same for upper limit, it can be slightly larger, due to for - # instance network latencies or CPU starvation - if should_took > 0: - assert took >= should_took * 0.85 and took <= should_took * 1.8 - else: - assert took >= should_took * 0.85 + assert took >= should_take * 0.85 @pytest.mark.parametrize( - "policy,backup_name,mode,setting,value,should_took", + "policy,backup_name,mode,setting,value,should_take", [ # # Local -> Local @@ -149,7 +143,7 @@ def assert_took(took, should_took): 0, id="no_local_throttling", ), - # reading 1e6*8 bytes with 1M default bandwith should take (8-1)/1=7 seconds + # reading 1e6*8 bytes with 1M default bandwidth should take (8-1)/1=7 seconds pytest.param( "default", next_backup_name("local"), @@ -159,7 +153,7 @@ def assert_took(took, should_took): 7, id="user_local_throttling", ), - # reading 1e6*8 bytes with 2M default bandwith should take (8-2)/2=3 seconds + # reading 1e6*8 bytes with 2M default bandwidth should take (8-2)/2=3 seconds pytest.param( "default", next_backup_name("local"), @@ -181,7 +175,7 @@ def assert_took(took, should_took): 0, id="no_remote_to_local_throttling", ), - # reading 1e6*8 bytes with 1M default bandwith should take (8-1)/1=7 seconds + # reading 1e6*8 bytes with 1M default bandwidth should take (8-1)/1=7 seconds pytest.param( "s3", next_backup_name("local"), @@ -191,7 +185,7 @@ def assert_took(took, should_took): 7, id="user_remote_to_local_throttling", ), - # reading 1e6*8 bytes with 2M default bandwith should take (8-2)/2=3 seconds + # reading 1e6*8 bytes with 2M default bandwidth should take (8-2)/2=3 seconds pytest.param( "s3", next_backup_name("local"), @@ -252,7 +246,7 @@ def assert_took(took, should_took): 0, id="no_local_to_remote_throttling", ), - # reading 1e6*8 bytes with 1M default bandwith should take (8-1)/1=7 seconds + # reading 1e6*8 bytes with 1M default bandwidth should take (8-1)/1=7 seconds pytest.param( "default", next_backup_name("remote"), @@ -262,7 +256,7 @@ def assert_took(took, should_took): 7, id="user_local_to_remote_throttling", ), - # reading 1e6*8 bytes with 2M default bandwith should take (8-2)/2=3 seconds + # reading 1e6*8 bytes with 2M default bandwidth should take (8-2)/2=3 seconds pytest.param( "default", next_backup_name("remote"), @@ -274,7 +268,7 @@ def assert_took(took, should_took): ), ], ) -def test_backup_throttling(policy, backup_name, mode, setting, value, should_took): +def test_backup_throttling(policy, backup_name, mode, setting, value, should_take): node_update_config(mode, setting, value) node.query( f""" @@ -284,7 +278,7 @@ def test_backup_throttling(policy, backup_name, mode, setting, value, should_too """ ) _, took = elapsed(node.query, f"backup table data to {backup_name}") - assert_took(took, should_took) + assert_took(took, should_take) def test_backup_throttling_override(): @@ -305,18 +299,18 @@ def test_backup_throttling_override(): "max_backup_bandwidth": "500K", }, ) - # reading 1e6*8 bytes with 500Ki default bandwith should take (8-0.5)/0.5=15 seconds + # reading 1e6*8 bytes with 500Ki default bandwidth should take (8-0.5)/0.5=15 seconds assert_took(took, 15) @pytest.mark.parametrize( - "policy,mode,setting,value,should_took", + "policy,mode,setting,value,should_take", [ # # Local # pytest.param("default", None, None, None, 0, id="no_local_throttling"), - # reading 1e6*8 bytes with 1M default bandwith should take (8-1)/1=7 seconds + # reading 1e6*8 bytes with 1M default bandwidth should take (8-1)/1=7 seconds pytest.param( "default", "user", @@ -325,7 +319,7 @@ def test_backup_throttling_override(): 7, id="user_local_throttling", ), - # reading 1e6*8 bytes with 2M default bandwith should take (8-2)/2=3 seconds + # reading 1e6*8 bytes with 2M default bandwidth should take (8-2)/2=3 seconds pytest.param( "default", "server", @@ -338,7 +332,7 @@ def test_backup_throttling_override(): # Remote # pytest.param("s3", None, None, None, 0, id="no_remote_throttling"), - # reading 1e6*8 bytes with 1M default bandwith should take (8-1)/1=7 seconds + # reading 1e6*8 bytes with 1M default bandwidth should take (8-1)/1=7 seconds pytest.param( "s3", "user", @@ -347,7 +341,7 @@ def test_backup_throttling_override(): 7, id="user_remote_throttling", ), - # reading 1e6*8 bytes with 2M default bandwith should take (8-2)/2=3 seconds + # reading 1e6*8 bytes with 2M default bandwidth should take (8-2)/2=3 seconds pytest.param( "s3", "server", @@ -358,7 +352,7 @@ def test_backup_throttling_override(): ), ], ) -def test_read_throttling(policy, mode, setting, value, should_took): +def test_read_throttling(policy, mode, setting, value, should_take): node_update_config(mode, setting, value) node.query( f""" @@ -368,17 +362,17 @@ def test_read_throttling(policy, mode, setting, value, should_took): """ ) _, took = elapsed(node.query, f"select * from data") - assert_took(took, should_took) + assert_took(took, should_take) @pytest.mark.parametrize( - "policy,mode,setting,value,should_took", + "policy,mode,setting,value,should_take", [ # # Local # pytest.param("default", None, None, None, 0, id="no_local_throttling"), - # reading 1e6*8 bytes with 1M default bandwith should take (8-1)/1=7 seconds + # reading 1e6*8 bytes with 1M default bandwidth should take (8-1)/1=7 seconds pytest.param( "default", "user", @@ -387,7 +381,7 @@ def test_read_throttling(policy, mode, setting, value, should_took): 7, id="local_user_throttling", ), - # reading 1e6*8 bytes with 2M default bandwith should take (8-2)/2=3 seconds + # reading 1e6*8 bytes with 2M default bandwidth should take (8-2)/2=3 seconds pytest.param( "default", "server", @@ -400,7 +394,7 @@ def test_read_throttling(policy, mode, setting, value, should_took): # Remote # pytest.param("s3", None, None, None, 0, id="no_remote_throttling"), - # writing 1e6*8 bytes with 1M default bandwith should take (8-1)/1=7 seconds + # writing 1e6*8 bytes with 1M default bandwidth should take (8-1)/1=7 seconds pytest.param( "s3", "user", @@ -409,7 +403,7 @@ def test_read_throttling(policy, mode, setting, value, should_took): 7, id="user_remote_throttling", ), - # writing 1e6*8 bytes with 2M default bandwith should take (8-2)/2=3 seconds + # writing 1e6*8 bytes with 2M default bandwidth should take (8-2)/2=3 seconds pytest.param( "s3", "server", @@ -420,7 +414,7 @@ def test_read_throttling(policy, mode, setting, value, should_took): ), ], ) -def test_write_throttling(policy, mode, setting, value, should_took): +def test_write_throttling(policy, mode, setting, value, should_take): node_update_config(mode, setting, value) node.query( f""" @@ -429,7 +423,7 @@ def test_write_throttling(policy, mode, setting, value, should_took): """ ) _, took = elapsed(node.query, f"insert into data select * from numbers(1e6)") - assert_took(took, should_took) + assert_took(took, should_take) def test_max_mutations_bandwidth_for_server(): @@ -444,7 +438,7 @@ def test_max_mutations_bandwidth_for_server(): node.query, "alter table data update key = -key where 1 settings mutations_sync = 1", ) - # reading 1e6*8 bytes with 1M/s bandwith should take (8-1)/1=7 seconds + # reading 1e6*8 bytes with 1M/s bandwidth should take (8-1)/1=7 seconds assert_took(took, 7) @@ -457,5 +451,5 @@ def test_max_merges_bandwidth_for_server(): ) node.query("insert into data select * from numbers(1e6)") _, took = elapsed(node.query, "optimize table data final") - # reading 1e6*8 bytes with 1M/s bandwith should take (8-1)/1=7 seconds + # reading 1e6*8 bytes with 1M/s bandwidth should take (8-1)/1=7 seconds assert_took(took, 7) diff --git a/tests/integration/test_version_update/configs/log_conf.xml b/tests/integration/test_version_update/configs/log_conf.xml index f9d15e572aa..27c7107ce5e 100644 --- a/tests/integration/test_version_update/configs/log_conf.xml +++ b/tests/integration/test_version_update/configs/log_conf.xml @@ -1,4 +1,4 @@ - + trace /var/log/clickhouse-server/log.log @@ -8,4 +8,4 @@ /var/log/clickhouse-server/stderr.log /var/log/clickhouse-server/stdout.log - + diff --git a/tests/performance/all_join_opt.xml b/tests/performance/all_join_opt.xml new file mode 100644 index 00000000000..0ab9c39f67c --- /dev/null +++ b/tests/performance/all_join_opt.xml @@ -0,0 +1,15 @@ + + CREATE TABLE test (a Int64, b String, c LowCardinality(String)) ENGINE = MergeTree() ORDER BY a + CREATE TABLE test1 (a Int64, b String, c LowCardinality(String)) ENGINE = MergeTree() ORDER BY a + + INSERT INTO test SELECT number % 10000, number % 10000, number % 10000 FROM numbers(10000000) + INSERT INTO test1 SELECT number % 1000 , number % 1000, number % 1000 FROM numbers(100000) + + SELECT MAX(test1.a) FROM test INNER JOIN test1 on test.b = test1.b + SELECT MAX(test1.a) FROM test LEFT JOIN test1 on test.b = test1.b + SELECT MAX(test1.a) FROM test RIGHT JOIN test1 on test.b = test1.b + SELECT MAX(test1.a) FROM test FULL JOIN test1 on test.b = test1.b + + DROP TABLE IF EXISTS test + DROP TABLE IF EXISTS test1 + \ No newline at end of file diff --git a/tests/performance/insert_select_squashing.xml b/tests/performance/insert_select_squashing.xml new file mode 100644 index 00000000000..4c2c88f3d22 --- /dev/null +++ b/tests/performance/insert_select_squashing.xml @@ -0,0 +1,23 @@ + + + 1000 + + + +CREATE TABLE squash_performance +( + s1 String, + s2 Nullable(String), + a1 Array(Array(String)), + a2 Array(Array(UInt32)), + m1 Map(String, Array(String)), + m2 Map(String, Array(UInt64)), + t Tuple(String, Array(String), Map(String, String)) +) +ENGINE = Null; + + + INSERT INTO squash_performance SELECT * FROM generateRandom(42) LIMIT 500000 + + DROP TABLE IF EXISTS squash_performance + diff --git a/tests/performance/insert_select_squashing_dynamic.xml b/tests/performance/insert_select_squashing_dynamic.xml new file mode 100644 index 00000000000..f7f600fd8bd --- /dev/null +++ b/tests/performance/insert_select_squashing_dynamic.xml @@ -0,0 +1,59 @@ + + + 1000 + 0 + 1 + + + +CREATE TABLE dynamic_squash_performance_1 +( + d Dynamic +) +ENGINE = Null; + + + +CREATE TABLE dynamic_squash_performance_2 +( + d Dynamic(max_types=6) +) +ENGINE = Null; + + + +CREATE TABLE src_dynamic_squash_performance_1 +( + d Dynamic +) +ENGINE = Memory; + + + +CREATE TABLE src_dynamic_squash_performance_2 +( + d Dynamic(max_types=6) +) +ENGINE = Memory; + + + + + + + + + + + + INSERT INTO dynamic_squash_performance_1 SELECT number::Dynamic FROM numbers(10000000) + INSERT INTO dynamic_squash_performance_1 SELECT range(number % 100)::Dynamic FROM numbers(2000000) + INSERT INTO dynamic_squash_performance_1 SELECT * FROM src_dynamic_squash_performance_1 + INSERT INTO dynamic_squash_performance_2 SELECT * FROM src_dynamic_squash_performance_2 + + DROP TABLE IF EXISTS dynamic_squash_performance_1 + DROP TABLE IF EXISTS dynamic_squash_performance_2 + DROP TABLE IF EXISTS src_dynamic_squash_performance_1 + DROP TABLE IF EXISTS src_dynamic_squash_performance_2 + + diff --git a/tests/performance/insert_select_squashing_variant.xml b/tests/performance/insert_select_squashing_variant.xml new file mode 100644 index 00000000000..5c59fc7b50f --- /dev/null +++ b/tests/performance/insert_select_squashing_variant.xml @@ -0,0 +1,34 @@ + + + 1000 + 0 + 1 + 1 + + + +CREATE TABLE variant_squash_performance +( + v Variant(Tuple(v1 Array(UInt64)), Tuple(v2 Array(UInt64)), Tuple(v3 Array(UInt64)), Tuple(v4 Array(UInt64)), Tuple(v5 Array(UInt64))) +) +ENGINE = Null; + + + +CREATE TABLE src_variant_squash_performance +( + v Variant(Tuple(v1 Array(UInt64)), Tuple(v2 Array(UInt64)), Tuple(v3 Array(UInt64)), Tuple(v4 Array(UInt64)), Tuple(v5 Array(UInt64))) +) +ENGINE = Memory; + + + + + + + INSERT INTO variant_squash_performance SELECT * FROM src_variant_squash_performance + + DROP TABLE IF EXISTS variant_squash_performance + DROP TABLE IF EXISTS src_variant_squash_performance + + diff --git a/tests/performance/materialized_view_parallel_insert.xml b/tests/performance/materialized_view_parallel_insert.xml index df0a23182c3..c3e7cdc50c0 100644 --- a/tests/performance/materialized_view_parallel_insert.xml +++ b/tests/performance/materialized_view_parallel_insert.xml @@ -1,4 +1,8 @@ + + 1 + + CREATE MATERIALIZED VIEW hits_mv ENGINE MergeTree PARTITION BY toYYYYMM(EventDate) diff --git a/tests/performance/optimize_functions_to_subcolumns.xml b/tests/performance/optimize_functions_to_subcolumns.xml new file mode 100644 index 00000000000..146af1605c4 --- /dev/null +++ b/tests/performance/optimize_functions_to_subcolumns.xml @@ -0,0 +1,26 @@ + + + 4 + + + + CREATE TABLE t_subcolumns (a Array(UInt64), s Nullable(String), m Map(String, UInt64)) ENGINE = MergeTree ORDER BY tuple() + + + + INSERT INTO t_subcolumns SELECT range(number % 20), toString(number), mapFromArrays(range(number % 20), range(number % 20)) FROM numbers_mt(50000000) + + + + OPTIMIZE TABLE t_subcolumns FINAL + + + SELECT count() FROM t_subcolumns WHERE NOT ignore(length(a)) + SELECT count() FROM t_subcolumns WHERE notEmpty(a) + SELECT count() FROM t_subcolumns WHERE NOT ignore(length(m)) + SELECT count() FROM t_subcolumns WHERE notEmpty(m) + SELECT count() FROM t_subcolumns WHERE isNotNull(s) + SELECT count(s) FROM t_subcolumns + + DROP TABLE t_subcolumns + diff --git a/tests/performance/parquet_read_with_index.xml b/tests/performance/parquet_read_with_index.xml new file mode 100644 index 00000000000..1bb2d8eb4a2 --- /dev/null +++ b/tests/performance/parquet_read_with_index.xml @@ -0,0 +1,30 @@ + + + INSERT INTO FUNCTION file('test_pq_index', Parquet) SELECT * FROM generateRandom('int64_column Nullable(Int64), tuple_column Tuple(a Nullable(String), b Nullable(Float64), c Tuple(i UInt32, j UInt32)),array_tuple_column Array(Tuple(a Nullable(String), b Nullable(Float64), c Nullable(Int64))), map_tuple_column Map(String, Tuple(a Nullable(String), b Nullable(Float64), c Nullable(Int64)))') limit 1000000 SETTINGS output_format_parquet_use_custom_encoder=false, output_format_parquet_write_page_index=true + + + + SELECT * FROM file('test_pq_index', Parquet, 'tuple_column Tuple(a Nullable(String))') Format Null + + + + SELECT tuple_column.a FROM file('test_pq_index', Parquet) Format Null + + + + SELECT tuple_column.a FROM file('test_pq_index', Parquet, 'tuple_column Tuple(a Nullable(String))') Format Null + + + + SELECT tuple_column.c.i FROM file('test_pq_index', Parquet) Format Null + + + + SELECT * FROM file('test_pq_index', Parquet, 'array_tuple_column Array (Tuple(a Nullable(String)))') Format Null + + + + SELECT * FROM file('test_pq_index', Parquet, 'map_tuple_column Map(String, Tuple(a Nullable(String)))') Format Null + + + diff --git a/tests/performance/storage_join_direct_join.xml b/tests/performance/storage_join_direct_join.xml index 2fc63c2c926..0e67abb275e 100644 --- a/tests/performance/storage_join_direct_join.xml +++ b/tests/performance/storage_join_direct_join.xml @@ -16,4 +16,7 @@ SELECT keys.key, value1 FROM keys ANY LEFT JOIN dict AS d ON (keys.key = d.key) FORMAT Null; SELECT keys.key, value1 FROM keys ANY LEFT JOIN dict AS d ON (keys.key = d.key) FORMAT Null SETTINGS allow_experimental_analyzer=1 - \ No newline at end of file + + DROP TABLE IF EXISTS keys + DROP TABLE IF EXISTS dict + diff --git a/tests/performance/views_max_insert_threads.xml b/tests/performance/views_max_insert_threads.xml new file mode 100644 index 00000000000..c16fb330b35 --- /dev/null +++ b/tests/performance/views_max_insert_threads.xml @@ -0,0 +1,15 @@ + + + + 1 + + + create table views_max_insert_threads_null (a UInt64) Engine = Null + create materialized view views_max_insert_threads_mv Engine = Null AS select now() as ts, max(a) from views_max_insert_threads_null group by ts + + insert into views_max_insert_threads_null select * from numbers_mt(3000000000) settings max_threads = 16, max_insert_threads=16 + + drop table if exists views_max_insert_threads_null + drop table if exists views_max_insert_threads_mv + + 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 b806aff3a00..53321afc94c 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 @@ -1,5 +1,7 @@ -- Tags: no-parallel, no-fasttest, no-ubsan, no-batch, no-flaky-check -- no-parallel because we want to run this test when most of the other tests already passed +-- This is not a regular test. It is intended to run once after other tests to validate certain statistics about the whole test runs. +-- TODO: I advise to put in inside clickhouse-test instead. -- 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/00116_storage_set.sql b/tests/queries/0_stateless/00116_storage_set.sql index c156b387c8f..36ad015c69e 100644 --- a/tests/queries/0_stateless/00116_storage_set.sql +++ b/tests/queries/0_stateless/00116_storage_set.sql @@ -28,8 +28,8 @@ RENAME TABLE set2 TO set; SELECT arrayJoin(['Hello', 'test', 'World', 'world', 'abc', 'xyz']) AS s WHERE s IN set; create table tab (x String) engine = MergeTree order by x as select 'Hello'; -SELECT * FROM tab PREWHERE x IN (set) WHERE x IN (set) LIMIT 1 settings allow_experimental_analyzer=0; -SELECT * FROM tab PREWHERE x IN (set) WHERE x IN (set) LIMIT 1 settings allow_experimental_analyzer=1; +SELECT * FROM tab PREWHERE x IN (set) WHERE x IN (set) LIMIT 1 settings enable_analyzer=0; +SELECT * FROM tab PREWHERE x IN (set) WHERE x IN (set) LIMIT 1 settings enable_analyzer=1; DROP TABLE tab; DROP TABLE set; diff --git a/tests/queries/0_stateless/00202_cross_join.sql b/tests/queries/0_stateless/00202_cross_join.sql index 8d62c56b3f1..301161b6583 100644 --- a/tests/queries/0_stateless/00202_cross_join.sql +++ b/tests/queries/0_stateless/00202_cross_join.sql @@ -3,5 +3,6 @@ SELECT x, y FROM (SELECT number AS x FROM system.numbers LIMIT 3) js1 CROSS JOIN SET join_algorithm = 'auto'; SELECT x, y FROM (SELECT number AS x FROM system.numbers LIMIT 3) js1 CROSS JOIN (SELECT number AS y FROM system.numbers LIMIT 5) js2; +-- Just to test that we preserved old setting name this we use `enable_analyzer` instead of `enable_analyzer` here. SET allow_experimental_analyzer = 1; SELECT x, y FROM (SELECT number AS x FROM system.numbers LIMIT 3) js1 CROSS JOIN (SELECT number AS y FROM system.numbers LIMIT 5) js2; diff --git a/tests/queries/0_stateless/00257_shard_no_aggregates_and_constant_keys.sql b/tests/queries/0_stateless/00257_shard_no_aggregates_and_constant_keys.sql index e3634141613..b72162be49a 100644 --- a/tests/queries/0_stateless/00257_shard_no_aggregates_and_constant_keys.sql +++ b/tests/queries/0_stateless/00257_shard_no_aggregates_and_constant_keys.sql @@ -1,6 +1,6 @@ -- Tags: shard -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; set enable_positional_arguments = 0; select 40 as z from (select * from system.numbers limit 3) group by z; diff --git a/tests/queries/0_stateless/00284_external_aggregation.reference b/tests/queries/0_stateless/00284_external_aggregation.reference index be0db217a97..48e30e781e0 100644 --- a/tests/queries/0_stateless/00284_external_aggregation.reference +++ b/tests/queries/0_stateless/00284_external_aggregation.reference @@ -1,22 +1,2 @@ 49999995000000 10000000 499999500000 1000000 15 -100033 2 -100034 2 -100035 2 -100036 2 -100037 2 -100038 2 -100039 2 -10004 2 -100040 2 -100041 2 -100033 2 -100034 2 -100035 2 -100036 2 -100037 2 -100038 2 -100039 2 -10004 2 -100040 2 -100041 2 diff --git a/tests/queries/0_stateless/00284_external_aggregation.sql b/tests/queries/0_stateless/00284_external_aggregation.sql index c1140faaa28..cdc31ff68c8 100644 --- a/tests/queries/0_stateless/00284_external_aggregation.sql +++ b/tests/queries/0_stateless/00284_external_aggregation.sql @@ -1,5 +1,8 @@ -- Tags: long +-- This test was split in two due to long runtimes in sanitizers. +-- The other part is 00284_external_aggregation_2. + SET max_bytes_before_external_group_by = 100000000; SET max_memory_usage = 410000000; SET group_by_two_level_threshold = 100000; @@ -7,19 +10,3 @@ SET group_by_two_level_threshold_bytes = 50000000; SELECT sum(k), sum(c) FROM (SELECT number AS k, count() AS c FROM (SELECT * FROM system.numbers LIMIT 10000000) GROUP BY k); SELECT sum(k), sum(c), max(u) FROM (SELECT number AS k, count() AS c, uniqArray(range(number % 16)) AS u FROM (SELECT * FROM system.numbers LIMIT 1000000) GROUP BY k); - -SET max_memory_usage = 0; -SET group_by_two_level_threshold = 100000; -SET max_bytes_before_external_group_by = '1Mi'; - --- method: key_string & key_string_two_level -CREATE TABLE t_00284_str(s String) ENGINE = MergeTree() ORDER BY tuple() SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; -INSERT INTO t_00284_str SELECT toString(number) FROM numbers_mt(1e6); -INSERT INTO t_00284_str SELECT toString(number) FROM numbers_mt(1e6); -SELECT s, count() FROM t_00284_str GROUP BY s ORDER BY s LIMIT 10 OFFSET 42; - --- method: low_cardinality_key_string & low_cardinality_key_string_two_level -CREATE TABLE t_00284_lc_str(s LowCardinality(String)) ENGINE = MergeTree() ORDER BY tuple() SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; -INSERT INTO t_00284_lc_str SELECT toString(number) FROM numbers_mt(1e6); -INSERT INTO t_00284_lc_str SELECT toString(number) FROM numbers_mt(1e6); -SELECT s, count() FROM t_00284_lc_str GROUP BY s ORDER BY s LIMIT 10 OFFSET 42; diff --git a/tests/queries/0_stateless/00284_external_aggregation_2.reference b/tests/queries/0_stateless/00284_external_aggregation_2.reference new file mode 100644 index 00000000000..71d2e96d4b0 --- /dev/null +++ b/tests/queries/0_stateless/00284_external_aggregation_2.reference @@ -0,0 +1,20 @@ +100033 2 +100034 2 +100035 2 +100036 2 +100037 2 +100038 2 +100039 2 +10004 2 +100040 2 +100041 2 +100033 2 +100034 2 +100035 2 +100036 2 +100037 2 +100038 2 +100039 2 +10004 2 +100040 2 +100041 2 diff --git a/tests/queries/0_stateless/00284_external_aggregation_2.sql b/tests/queries/0_stateless/00284_external_aggregation_2.sql new file mode 100644 index 00000000000..7960e3894d0 --- /dev/null +++ b/tests/queries/0_stateless/00284_external_aggregation_2.sql @@ -0,0 +1,22 @@ +-- Tags: long + +-- This test was split in two due to long runtimes in sanitizers. +-- The other part is 00284_external_aggregation. + +SET group_by_two_level_threshold_bytes = 50000000; +SET max_memory_usage = 0; +SET group_by_two_level_threshold = 100000; +SET max_bytes_before_external_group_by = '1Mi'; + +-- method: key_string & key_string_two_level +CREATE TABLE t_00284_str(s String) ENGINE = MergeTree() ORDER BY tuple() SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; +INSERT INTO t_00284_str SELECT toString(number) FROM numbers_mt(1e6); +INSERT INTO t_00284_str SELECT toString(number) FROM numbers_mt(1e6); +SELECT s, count() FROM t_00284_str GROUP BY s ORDER BY s LIMIT 10 OFFSET 42; + +-- method: low_cardinality_key_string & low_cardinality_key_string_two_level +CREATE TABLE t_00284_lc_str(s LowCardinality(String)) ENGINE = MergeTree() ORDER BY tuple() SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; +INSERT INTO t_00284_lc_str SELECT toString(number) FROM numbers_mt(1e6); +INSERT INTO t_00284_lc_str SELECT toString(number) FROM numbers_mt(1e6); +SELECT s, count() FROM t_00284_lc_str GROUP BY s ORDER BY s LIMIT 10 OFFSET 42; + diff --git a/tests/queries/0_stateless/00313_const_totals_extremes.sh b/tests/queries/0_stateless/00313_const_totals_extremes.sh index 539a19817e2..6267d5066dd 100755 --- a/tests/queries/0_stateless/00313_const_totals_extremes.sh +++ b/tests/queries/0_stateless/00313_const_totals_extremes.sh @@ -4,10 +4,10 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&extremes=1&output_format_write_statistics=0" -d "SELECT 1 AS k, count() GROUP BY k WITH TOTALS SETTINGS allow_experimental_analyzer = 1"; -${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&extremes=1&output_format_write_statistics=0" -d "SELECT 1234567890123 AS k, count() GROUP BY k WITH TOTALS SETTINGS allow_experimental_analyzer = 1 FORMAT JSON"; -${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&extremes=1&output_format_write_statistics=0" -d "SELECT toFloat32(1.23) AS k, count() GROUP BY k WITH TOTALS SETTINGS allow_experimental_analyzer = 1 FORMAT JSONCompact"; +${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&extremes=1&output_format_write_statistics=0" -d "SELECT 1 AS k, count() GROUP BY k WITH TOTALS SETTINGS enable_analyzer = 1"; +${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&extremes=1&output_format_write_statistics=0" -d "SELECT 1234567890123 AS k, count() GROUP BY k WITH TOTALS SETTINGS enable_analyzer = 1 FORMAT JSON"; +${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&extremes=1&output_format_write_statistics=0" -d "SELECT toFloat32(1.23) AS k, count() GROUP BY k WITH TOTALS SETTINGS enable_analyzer = 1 FORMAT JSONCompact"; -${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&extremes=1&output_format_write_statistics=0" -d "SELECT toDate('2010-01-01') AS k, count() GROUP BY k WITH TOTALS SETTINGS allow_experimental_analyzer = 1"; -${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&extremes=1&output_format_write_statistics=0" -d "SELECT toDateTime('2010-01-01 01:02:03', 'UTC') AS k, count() GROUP BY k WITH TOTALS SETTINGS allow_experimental_analyzer = 1 FORMAT JSON"; -${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&extremes=1&output_format_write_statistics=0" -d "SELECT 1.1 AS k, count() GROUP BY k WITH TOTALS SETTINGS allow_experimental_analyzer = 1 FORMAT JSONCompact"; +${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&extremes=1&output_format_write_statistics=0" -d "SELECT toDate('2010-01-01') AS k, count() GROUP BY k WITH TOTALS SETTINGS enable_analyzer = 1"; +${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&extremes=1&output_format_write_statistics=0" -d "SELECT toDateTime('2010-01-01 01:02:03', 'UTC') AS k, count() GROUP BY k WITH TOTALS SETTINGS enable_analyzer = 1 FORMAT JSON"; +${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&extremes=1&output_format_write_statistics=0" -d "SELECT 1.1 AS k, count() GROUP BY k WITH TOTALS SETTINGS enable_analyzer = 1 FORMAT JSONCompact"; diff --git a/tests/queries/0_stateless/00314_sample_factor_virtual_column.sql b/tests/queries/0_stateless/00314_sample_factor_virtual_column.sql index 6e3dc019069..b8ac5e733ed 100644 --- a/tests/queries/0_stateless/00314_sample_factor_virtual_column.sql +++ b/tests/queries/0_stateless/00314_sample_factor_virtual_column.sql @@ -1,3 +1,7 @@ +-- Tags: no-msan +-- ^ +-- makes SELECTs extremely slow sometimes for some reason: "Aggregated. 1000000 to 1 rows (from 7.63 MiB) in 242.829221645 sec." + DROP TABLE IF EXISTS sample_00314_1; DROP TABLE IF EXISTS sample_00314_2; DROP TABLE IF EXISTS sample_merge_00314; diff --git a/tests/queries/0_stateless/00331_final_and_prewhere_condition_ver_column.sql b/tests/queries/0_stateless/00331_final_and_prewhere_condition_ver_column.sql index 78a58a979d1..a3c499f1688 100644 --- a/tests/queries/0_stateless/00331_final_and_prewhere_condition_ver_column.sql +++ b/tests/queries/0_stateless/00331_final_and_prewhere_condition_ver_column.sql @@ -1,8 +1,8 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- https://github.com/ClickHouse/ClickHouse/issues/45804 -CREATE TABLE myRMT( +CREATE TABLE myRMT( key Int64, someCol String, ver DateTime diff --git a/tests/queries/0_stateless/00366_multi_statements.sh b/tests/queries/0_stateless/00366_multi_statements.sh index 0b2e80fe457..8546e547581 100755 --- a/tests/queries/0_stateless/00366_multi_statements.sh +++ b/tests/queries/0_stateless/00366_multi_statements.sh @@ -14,22 +14,22 @@ $CLICKHOUSE_CLIENT --query="SELECT 1; SELECT 2" $CLICKHOUSE_CLIENT --query="SELECT 1; SELECT 2;" $CLICKHOUSE_CLIENT --query="SELECT 1; SELECT 2; SELECT" 2>&1 | grep -o 'Syntax error' -$CLICKHOUSE_CLIENT -n --query="SELECT 1; S" 2>&1 | grep -o 'Syntax error' -$CLICKHOUSE_CLIENT -n --query="SELECT 1; SELECT 2" -$CLICKHOUSE_CLIENT -n --query="SELECT 1; SELECT 2;" -$CLICKHOUSE_CLIENT -n --query="SELECT 1; SELECT 2; SELECT" 2>&1 | grep -o 'Syntax error' +$CLICKHOUSE_CLIENT --query="SELECT 1; S" 2>&1 | grep -o 'Syntax error' +$CLICKHOUSE_CLIENT --query="SELECT 1; SELECT 2" +$CLICKHOUSE_CLIENT --query="SELECT 1; SELECT 2;" +$CLICKHOUSE_CLIENT --query="SELECT 1; SELECT 2; SELECT" 2>&1 | grep -o 'Syntax error' -$CLICKHOUSE_CLIENT -n --query="DROP TABLE IF EXISTS t_00366; CREATE TABLE t_00366 (x UInt64) ENGINE = TinyLog;" +$CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS t_00366; CREATE TABLE t_00366 (x UInt64) ENGINE = TinyLog;" $CLICKHOUSE_CLIENT --query="INSERT INTO t_00366 VALUES (1),(2),(3);" $CLICKHOUSE_CLIENT --query="SELECT * FROM t_00366" $CLICKHOUSE_CLIENT --query="INSERT INTO t_00366 VALUES" <<< "(4),(5),(6)" $CLICKHOUSE_CLIENT --query="SELECT * FROM t_00366" -$CLICKHOUSE_CLIENT -n --query="INSERT INTO t_00366 VALUES (1),(2),(3);" -$CLICKHOUSE_CLIENT -n --query="SELECT * FROM t_00366" -$CLICKHOUSE_CLIENT -n --query="INSERT INTO t_00366 VALUES" <<< "(4),(5),(6)" -$CLICKHOUSE_CLIENT -n --query="SELECT * FROM t_00366" +$CLICKHOUSE_CLIENT --query="INSERT INTO t_00366 VALUES (1),(2),(3);" +$CLICKHOUSE_CLIENT --query="SELECT * FROM t_00366" +$CLICKHOUSE_CLIENT --query="INSERT INTO t_00366 VALUES" <<< "(4),(5),(6)" +$CLICKHOUSE_CLIENT --query="SELECT * FROM t_00366" ${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}" -d "SELECT 1" ${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}" -d "SELECT 1;" @@ -48,4 +48,4 @@ $CLICKHOUSE_CLIENT --query="SELECT * FROM t_00366" ${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&query=INSERT+INTO+t_00366+VALUES" -d "(7),(8),(9)" $CLICKHOUSE_CLIENT --query="SELECT * FROM t_00366" -$CLICKHOUSE_CLIENT -n --query="DROP TABLE t_00366;" +$CLICKHOUSE_CLIENT --query="DROP TABLE t_00366;" diff --git a/tests/queries/0_stateless/00370_duplicate_columns_in_subqueries.sql b/tests/queries/0_stateless/00370_duplicate_columns_in_subqueries.sql index 118e50c35e0..8a93c0a9d26 100644 --- a/tests/queries/0_stateless/00370_duplicate_columns_in_subqueries.sql +++ b/tests/queries/0_stateless/00370_duplicate_columns_in_subqueries.sql @@ -1,6 +1,6 @@ SET any_join_distinct_right_table_keys = 1; SET joined_subquery_requires_alias = 0; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; select x, y from (select 1 as x, 2 as y, x, y); select x, y from (select 1 as x, 1 as y, x, y); diff --git a/tests/queries/0_stateless/00378_json_quote_64bit_integers.sql b/tests/queries/0_stateless/00378_json_quote_64bit_integers.sql index e7b59bc3f7f..941d5ecd135 100644 --- a/tests/queries/0_stateless/00378_json_quote_64bit_integers.sql +++ b/tests/queries/0_stateless/00378_json_quote_64bit_integers.sql @@ -2,7 +2,7 @@ SET output_format_write_statistics = 0; SET extremes = 1; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET output_format_json_quote_64bit_integers = 1; SELECT toInt64(0) as i0, toUInt64(0) as u0, toInt64(9223372036854775807) as ip, toInt64(-9223372036854775808) as in, toUInt64(18446744073709551615) as up, [toInt64(0)] as arr, (toUInt64(0), toUInt64(0)) as tuple GROUP BY i0, u0, ip, in, up, arr, tuple WITH TOTALS FORMAT JSON; diff --git a/tests/queries/0_stateless/00408_http_keep_alive.reference b/tests/queries/0_stateless/00408_http_keep_alive.reference index 17a7fd690a8..5402036bfd7 100644 --- a/tests/queries/0_stateless/00408_http_keep_alive.reference +++ b/tests/queries/0_stateless/00408_http_keep_alive.reference @@ -1,6 +1,6 @@ < Connection: Keep-Alive -< Keep-Alive: timeout=10 +< Keep-Alive: timeout=10, max=? < Connection: Keep-Alive -< Keep-Alive: timeout=10 +< Keep-Alive: timeout=10, max=? < Connection: Keep-Alive -< Keep-Alive: timeout=10 +< Keep-Alive: timeout=10, max=? diff --git a/tests/queries/0_stateless/00408_http_keep_alive.sh b/tests/queries/0_stateless/00408_http_keep_alive.sh index 4bd0e494eb8..4a1cb4ed712 100755 --- a/tests/queries/0_stateless/00408_http_keep_alive.sh +++ b/tests/queries/0_stateless/00408_http_keep_alive.sh @@ -6,9 +6,10 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) URL="${CLICKHOUSE_PORT_HTTP_PROTO}://${CLICKHOUSE_HOST}:${CLICKHOUSE_PORT_HTTP}/" -${CLICKHOUSE_CURL} -vsS "${URL}" --data-binary @- <<< "SELECT 1" 2>&1 | perl -lnE 'print if /Keep-Alive/'; -${CLICKHOUSE_CURL} -vsS "${URL}" --data-binary @- <<< " error here " 2>&1 | perl -lnE 'print if /Keep-Alive/'; -${CLICKHOUSE_CURL} -vsS "${URL}"ping 2>&1 | perl -lnE 'print if /Keep-Alive/'; +# the sed command here replaces the real number of left requests with a question mark, because it can vary and we don't really have control over it +${CLICKHOUSE_CURL} -vsS "${URL}" --data-binary @- <<< "SELECT 1" 2>&1 | sed -r 's/(keep-alive: timeout=10, max=)[0-9]+/\1?/I' | grep -i 'keep-alive'; +${CLICKHOUSE_CURL} -vsS "${URL}" --data-binary @- <<< " error here " 2>&1 | sed -r 's/(keep-alive: timeout=10, max=)[0-9]+/\1?/I' | grep -i 'keep-alive'; +${CLICKHOUSE_CURL} -vsS "${URL}"ping 2>&1 | perl -lnE 'print if /Keep-Alive/' | sed -r 's/(keep-alive: timeout=10, max=)[0-9]+/\1?/I' | grep -i 'keep-alive'; # no keep-alive: ${CLICKHOUSE_CURL} -vsS "${URL}"404/not/found/ 2>&1 | perl -lnE 'print if /Keep-Alive/'; diff --git a/tests/queries/0_stateless/00443_preferred_block_size_bytes.sh b/tests/queries/0_stateless/00443_preferred_block_size_bytes.sh index 27b9f5c00c7..0635fbc2a57 100755 --- a/tests/queries/0_stateless/00443_preferred_block_size_bytes.sh +++ b/tests/queries/0_stateless/00443_preferred_block_size_bytes.sh @@ -43,10 +43,10 @@ popd > /dev/null #SCRIPTDIR=`dirname "$SCRIPTPATH"` SCRIPTDIR=$SCRIPTPATH -cat "$SCRIPTDIR"/00282_merging.sql | $CLICKHOUSE_CLIENT --preferred_block_size_bytes=10 -n > "${CLICKHOUSE_TMP}"/preferred_block_size_bytes.stdout +cat "$SCRIPTDIR"/00282_merging.sql | $CLICKHOUSE_CLIENT --preferred_block_size_bytes=10 > "${CLICKHOUSE_TMP}"/preferred_block_size_bytes.stdout cmp "$SCRIPTDIR"/00282_merging.reference "${CLICKHOUSE_TMP}"/preferred_block_size_bytes.stdout && echo PASSED || echo FAILED -cat "$SCRIPTDIR"/00282_merging.sql | $CLICKHOUSE_CLIENT --preferred_block_size_bytes=20 -n > "${CLICKHOUSE_TMP}"/preferred_block_size_bytes.stdout +cat "$SCRIPTDIR"/00282_merging.sql | $CLICKHOUSE_CLIENT --preferred_block_size_bytes=20 > "${CLICKHOUSE_TMP}"/preferred_block_size_bytes.stdout cmp "$SCRIPTDIR"/00282_merging.reference "${CLICKHOUSE_TMP}"/preferred_block_size_bytes.stdout && echo PASSED || echo FAILED rm "${CLICKHOUSE_TMP}"/preferred_block_size_bytes.stdout diff --git a/tests/queries/0_stateless/00445_join_nullable_keys.sql b/tests/queries/0_stateless/00445_join_nullable_keys.sql index 774594f90f3..bec0c76eb5f 100644 --- a/tests/queries/0_stateless/00445_join_nullable_keys.sql +++ b/tests/queries/0_stateless/00445_join_nullable_keys.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET join_use_nulls = 0; SET any_join_distinct_right_table_keys = 1; diff --git a/tests/queries/0_stateless/00490_with_select.sql b/tests/queries/0_stateless/00490_with_select.sql index c803cf1d3ad..d63f0ca4f31 100644 --- a/tests/queries/0_stateless/00490_with_select.sql +++ b/tests/queries/0_stateless/00490_with_select.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; with pow(2,2) as four select pow(four, 2), 2 as two, pow(two, 2); select `pow(four, 2)`, `pow(two, 2)` from (with pow(2,2) as four select pow(four, 2), 2 as two, pow(two, 2)); diff --git a/tests/queries/0_stateless/00501_http_head.reference b/tests/queries/0_stateless/00501_http_head.reference index 8351327b356..807bcd4922e 100644 --- a/tests/queries/0_stateless/00501_http_head.reference +++ b/tests/queries/0_stateless/00501_http_head.reference @@ -2,11 +2,11 @@ HTTP/1.1 200 OK Connection: Keep-Alive Content-Type: text/tab-separated-values; charset=UTF-8 Transfer-Encoding: chunked -Keep-Alive: timeout=10 +Keep-Alive: timeout=10, max=? HTTP/1.1 200 OK Connection: Keep-Alive Content-Type: text/tab-separated-values; charset=UTF-8 Transfer-Encoding: chunked -Keep-Alive: timeout=10 +Keep-Alive: timeout=10, max=? diff --git a/tests/queries/0_stateless/00501_http_head.sh b/tests/queries/0_stateless/00501_http_head.sh index 60283f26833..30da64c31f0 100755 --- a/tests/queries/0_stateless/00501_http_head.sh +++ b/tests/queries/0_stateless/00501_http_head.sh @@ -4,8 +4,9 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -( ${CLICKHOUSE_CURL} -s --head "${CLICKHOUSE_URL}&query=SELECT%201"; - ${CLICKHOUSE_CURL} -s --head "${CLICKHOUSE_URL}&query=select+*+from+system.numbers+limit+1000000" ) | grep -v "Date:" | grep -v "X-ClickHouse-Server-Display-Name:" | grep -v "X-ClickHouse-Query-Id:" | grep -v "X-ClickHouse-Format:" | grep -v "X-ClickHouse-Timezone:" +# the sed command here replaces the real number of left requests with a question mark, because it can vary and we don't really have control over it +( ${CLICKHOUSE_CURL} -s --head "${CLICKHOUSE_URL}&query=SELECT%201" | sed -r 's/(keep-alive: timeout=10, max=)[0-9]+/\1?/I'; + ${CLICKHOUSE_CURL} -s --head "${CLICKHOUSE_URL}&query=select+*+from+system.numbers+limit+1000000" ) | sed -r 's/(keep-alive: timeout=10, max=)[0-9]+/\1?/I' | grep -v "Date:" | grep -v "X-ClickHouse-Server-Display-Name:" | grep -v "X-ClickHouse-Query-Id:" | grep -v "X-ClickHouse-Format:" | grep -v "X-ClickHouse-Timezone:" if [[ $(${CLICKHOUSE_CURL} -sS -X POST -I "${CLICKHOUSE_URL}&query=SELECT+1" | grep -c '411 Length Required') -ne 1 ]]; then echo FAIL diff --git a/tests/queries/0_stateless/00543_access_to_temporary_table_in_readonly_mode.sh b/tests/queries/0_stateless/00543_access_to_temporary_table_in_readonly_mode.sh index 560b97a1d1b..5550fa69d3d 100755 --- a/tests/queries/0_stateless/00543_access_to_temporary_table_in_readonly_mode.sh +++ b/tests/queries/0_stateless/00543_access_to_temporary_table_in_readonly_mode.sh @@ -4,7 +4,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" DROP TABLE IF EXISTS test_readonly; CREATE TABLE test_readonly ( ID Int @@ -16,7 +16,7 @@ $CLICKHOUSE_CLIENT -n --query=" ################ # Try to create temporary table -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" SET readonly = 1; CREATE TEMPORARY TABLE readonly ( ID Int @@ -26,7 +26,7 @@ CODE=$?; [ "$CODE" -ne "164" ] && [ "$CODE" -ne "0" ] && echo "Fail" && exit $CODE; # Try to insert into exists (non temporary) table -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" SET readonly = 1; INSERT INTO test_readonly (ID) VALUES (1); " 2> /dev/null; @@ -34,7 +34,7 @@ CODE=$?; [ "$CODE" -ne "164" ] && [ "$CODE" -ne "0" ] && echo "Fail" && exit $CODE; # Try to drop exists (non temporary) table -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" SET readonly = 1; DROP TABLE test_readonly; " 2> /dev/null; @@ -46,7 +46,7 @@ CODE=$?; ################ # Try to create temporary table -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" SET readonly = 2; CREATE TEMPORARY TABLE readonly ( ID Int @@ -58,7 +58,7 @@ CODE=$?; [ "$CODE" -ne "0" ] && echo "Fail" && exit $CODE; # Try to insert into exists (non temporary) table -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" SET readonly = 2; INSERT INTO test_readonly (ID) VALUES (1); " 2> /dev/null; @@ -66,7 +66,7 @@ CODE=$?; [ "$CODE" -ne "164" ] && [ "$CODE" -ne "0" ] && echo "Fail" && exit $CODE; # Try to drop exists (non temporary) table -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" SET readonly = 2; DROP TABLE test_readonly; " 2> /dev/null; @@ -78,7 +78,7 @@ CODE=$?; ################ # Try to create temporary table -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" SET readonly = 0; CREATE TEMPORARY TABLE readonly ( ID Int @@ -90,7 +90,7 @@ CODE=$?; [ "$CODE" -ne "0" ] && echo "Fail" && exit $CODE; # Try to insert into exists (non temporary) table -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" SET readonly = 0; INSERT INTO test_readonly (ID) VALUES (1); " 2> /dev/null; @@ -98,7 +98,7 @@ CODE=$?; [ "$CODE" -ne "0" ] && echo "Fail" && exit $CODE; # Try to drop exists (non temporary) table -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" SET readonly = 0; DROP TABLE test_readonly; " 2> /dev/null; diff --git a/tests/queries/0_stateless/00597_push_down_predicate_long.reference b/tests/queries/0_stateless/00597_push_down_predicate_long.reference index 2c46edc98bf..55b7cdd3c64 100644 --- a/tests/queries/0_stateless/00597_push_down_predicate_long.reference +++ b/tests/queries/0_stateless/00597_push_down_predicate_long.reference @@ -390,7 +390,7 @@ ANY LEFT JOIN ) USING (id) WHERE id = 1 2000-01-01 1 test string 1 1 2000-01-01 test string 1 1 -------- allow_experimental_analyzer=1 ------- +------- enable_analyzer=1 ------- 1 2000-01-01 test string 1 1 2000-01-01 test string 1 1 SELECT id, @@ -454,7 +454,7 @@ FROM ) WHERE id = 1 2000-01-01 1 test string 1 1 -------- allow_experimental_analyzer=1 ------- +------- enable_analyzer=1 ------- 1 2000-01-01 test string 1 1 2000-01-01 test string 1 1 SELECT date, @@ -484,7 +484,7 @@ ANY LEFT JOIN ) AS b USING (id) WHERE b.id = 1 2000-01-01 1 test string 1 1 2000-01-01 test string 1 1 -------- allow_experimental_analyzer=1 ------- +------- enable_analyzer=1 ------- 1 2000-01-01 test string 1 1 2000-01-01 test string 1 1 SELECT id, @@ -510,7 +510,7 @@ ANY LEFT JOIN ) AS b USING (date, id) WHERE b.date = toDate(\'2000-01-01\') 1 2000-01-01 test string 1 1 -------- allow_experimental_analyzer=1 ------- +------- enable_analyzer=1 ------- 2000-01-01 1 test string 1 1 SELECT date, @@ -593,7 +593,7 @@ SEMI LEFT JOIN ) AS r USING (id) WHERE r.id = 1 2000-01-01 1 test string 1 1 2000-01-01 test string 1 1 -------- allow_experimental_analyzer=1 ------- +------- enable_analyzer=1 ------- 1 2000-01-01 test string 1 1 2000-01-01 test string 1 1 SELECT value + t1.value AS expr FROM diff --git a/tests/queries/0_stateless/00597_push_down_predicate_long.sql b/tests/queries/0_stateless/00597_push_down_predicate_long.sql index caf6edd7372..f79b24abe56 100644 --- a/tests/queries/0_stateless/00597_push_down_predicate_long.sql +++ b/tests/queries/0_stateless/00597_push_down_predicate_long.sql @@ -110,9 +110,9 @@ SELECT * FROM (SELECT * FROM test_00597 UNION ALL SELECT * FROM test_00597) WHER -- Optimize predicate expression with join query EXPLAIN SYNTAX SELECT * FROM (SELECT * FROM test_00597) ANY LEFT JOIN (SELECT * FROM test_00597) USING id WHERE id = 1; -SELECT * FROM (SELECT * FROM test_00597) ANY LEFT JOIN (SELECT * FROM test_00597) USING id WHERE id = 1 SETTINGS allow_experimental_analyzer=0; -SELECT '------- allow_experimental_analyzer=1 -------'; -SELECT * FROM (SELECT * FROM test_00597) ANY LEFT JOIN (SELECT * FROM test_00597) USING id WHERE id = 1 SETTINGS allow_experimental_analyzer=1; +SELECT * FROM (SELECT * FROM test_00597) ANY LEFT JOIN (SELECT * FROM test_00597) USING id WHERE id = 1 SETTINGS enable_analyzer=0; +SELECT '------- enable_analyzer=1 -------'; +SELECT * FROM (SELECT * FROM test_00597) ANY LEFT JOIN (SELECT * FROM test_00597) USING id WHERE id = 1 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT * FROM (SELECT toInt8(1) AS id) ANY LEFT JOIN test_00597 USING id WHERE value = 1; SELECT * FROM (SELECT toInt8(1) AS id) ANY LEFT JOIN test_00597 USING id WHERE value = 1; @@ -123,30 +123,30 @@ SELECT b.value FROM (SELECT toInt8(1) AS id) ANY LEFT JOIN test_00597 AS b USING -- Optimize predicate expression with join and nested subquery EXPLAIN SYNTAX SELECT * FROM (SELECT * FROM (SELECT * FROM test_00597) ANY LEFT JOIN (SELECT * FROM test_00597) USING id) WHERE id = 1; -SELECT * FROM (SELECT * FROM (SELECT * FROM test_00597) ANY LEFT JOIN (SELECT * FROM test_00597) USING id) WHERE id = 1 SETTINGS allow_experimental_analyzer=0; -SELECT '------- allow_experimental_analyzer=1 -------'; -SELECT * FROM (SELECT * FROM (SELECT * FROM test_00597) ANY LEFT JOIN (SELECT * FROM test_00597) USING id) WHERE id = 1 SETTINGS allow_experimental_analyzer=1; +SELECT * FROM (SELECT * FROM (SELECT * FROM test_00597) ANY LEFT JOIN (SELECT * FROM test_00597) USING id) WHERE id = 1 SETTINGS enable_analyzer=0; +SELECT '------- enable_analyzer=1 -------'; +SELECT * FROM (SELECT * FROM (SELECT * FROM test_00597) ANY LEFT JOIN (SELECT * FROM test_00597) USING id) WHERE id = 1 SETTINGS enable_analyzer=1; -- Optimize predicate expression with join query and qualified EXPLAIN SYNTAX SELECT * FROM (SELECT * FROM test_00597) ANY LEFT JOIN (SELECT * FROM test_00597) AS b USING id WHERE b.id = 1; -SELECT * FROM (SELECT * FROM test_00597) ANY LEFT JOIN (SELECT * FROM test_00597) AS b USING id WHERE b.id = 1 SETTINGS allow_experimental_analyzer=0; -SELECT '------- allow_experimental_analyzer=1 -------'; -SELECT * FROM (SELECT * FROM test_00597) ANY LEFT JOIN (SELECT * FROM test_00597) AS b USING id WHERE b.id = 1 SETTINGS allow_experimental_analyzer=1; +SELECT * FROM (SELECT * FROM test_00597) ANY LEFT JOIN (SELECT * FROM test_00597) AS b USING id WHERE b.id = 1 SETTINGS enable_analyzer=0; +SELECT '------- enable_analyzer=1 -------'; +SELECT * FROM (SELECT * FROM test_00597) ANY LEFT JOIN (SELECT * FROM test_00597) AS b USING id WHERE b.id = 1 SETTINGS enable_analyzer=1; -- Compatibility test EXPLAIN SYNTAX SELECT * FROM (SELECT toInt8(1) AS id, toDate('2000-01-01') AS date FROM system.numbers LIMIT 1) ANY LEFT JOIN (SELECT * FROM test_00597) AS b USING date, id WHERE b.date = toDate('2000-01-01'); -SELECT * FROM (SELECT toInt8(1) AS id, toDate('2000-01-01') AS date FROM system.numbers LIMIT 1) ANY LEFT JOIN (SELECT * FROM test_00597) AS b USING date, id WHERE b.date = toDate('2000-01-01') SETTINGS allow_experimental_analyzer=0; -SELECT '------- allow_experimental_analyzer=1 -------'; -SELECT * FROM (SELECT toInt8(1) AS id, toDate('2000-01-01') AS date FROM system.numbers LIMIT 1) ANY LEFT JOIN (SELECT * FROM test_00597) AS b USING date, id WHERE b.date = toDate('2000-01-01') SETTINGS allow_experimental_analyzer=1; +SELECT * FROM (SELECT toInt8(1) AS id, toDate('2000-01-01') AS date FROM system.numbers LIMIT 1) ANY LEFT JOIN (SELECT * FROM test_00597) AS b USING date, id WHERE b.date = toDate('2000-01-01') SETTINGS enable_analyzer=0; +SELECT '------- enable_analyzer=1 -------'; +SELECT * FROM (SELECT toInt8(1) AS id, toDate('2000-01-01') AS date FROM system.numbers LIMIT 1) ANY LEFT JOIN (SELECT * FROM test_00597) AS b USING date, id WHERE b.date = toDate('2000-01-01') SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT * FROM (SELECT * FROM (SELECT * FROM test_00597) AS a ANY LEFT JOIN (SELECT * FROM test_00597) AS b ON a.id = b.id) WHERE id = 1; SELECT * FROM (SELECT * FROM (SELECT * FROM test_00597) AS a ANY LEFT JOIN (SELECT * FROM test_00597) AS b ON a.id = b.id) WHERE id = 1; -- Explain with join subquery EXPLAIN SYNTAX SELECT * FROM (SELECT * FROM test_00597) ANY INNER JOIN (SELECT * FROM (SELECT * FROM test_00597)) as r USING id WHERE r.id = 1; -SELECT * FROM (SELECT * FROM test_00597) ANY INNER JOIN (SELECT * FROM (SELECT * FROM test_00597)) as r USING id WHERE r.id = 1 SETTINGS allow_experimental_analyzer=0; -SELECT '------- allow_experimental_analyzer=1 -------'; -SELECT * FROM (SELECT * FROM test_00597) ANY INNER JOIN (SELECT * FROM (SELECT * FROM test_00597)) as r USING id WHERE r.id = 1 SETTINGS allow_experimental_analyzer=1; +SELECT * FROM (SELECT * FROM test_00597) ANY INNER JOIN (SELECT * FROM (SELECT * FROM test_00597)) as r USING id WHERE r.id = 1 SETTINGS enable_analyzer=0; +SELECT '------- enable_analyzer=1 -------'; +SELECT * FROM (SELECT * FROM test_00597) ANY INNER JOIN (SELECT * FROM (SELECT * FROM test_00597)) as r USING id WHERE r.id = 1 SETTINGS enable_analyzer=1; -- issue 20497 EXPLAIN SYNTAX SELECT value + t1.value AS expr FROM (SELECT t0.value, t1.value FROM test_00597 AS t0 FULL JOIN test_00597 AS t1 USING date) WHERE expr < 3; diff --git a/tests/queries/0_stateless/00600_replace_running_query.sh b/tests/queries/0_stateless/00600_replace_running_query.sh index 7a71d17f19b..80e2ecf5d5b 100755 --- a/tests/queries/0_stateless/00600_replace_running_query.sh +++ b/tests/queries/0_stateless/00600_replace_running_query.sh @@ -6,41 +6,56 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -TEST_PREFIX=$RANDOM +TEST_PREFIX="${CLICKHOUSE_DATABASE}" ${CLICKHOUSE_CLIENT} -q "drop user if exists u_00600${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() { - while [[ $($CLICKHOUSE_CURL -sS "$CLICKHOUSE_URL" -d "SELECT count() FROM system.processes WHERE query_id = '$1'") == 0 ]]; do sleep 0.1; done + while [[ 0 -eq $($CLICKHOUSE_CURL -sS "$CLICKHOUSE_URL" -d "SELECT count() FROM system.processes WHERE query_id = '$1'") ]] + do + sleep 0.1 + done +} + +function wait_for_queries_to_finish() +{ + while [[ 0 -ne $($CLICKHOUSE_CURL -sS "$CLICKHOUSE_URL" -d "SELECT count() FROM system.processes WHERE current_database = '${CLICKHOUSE_DATABASE}' AND query NOT LIKE '%this query%'") ]] + do + sleep 0.1 + done } -$CLICKHOUSE_CURL -sS "$CLICKHOUSE_URL&query_id=hello&replace_running_query=1" -d 'SELECT 1, count() FROM system.numbers' > /dev/null 2>&1 & -wait_for_query_to_start 'hello' +$CLICKHOUSE_CURL -sS "$CLICKHOUSE_URL&query_id=${CLICKHOUSE_DATABASE}hello&replace_running_query=1" -d 'SELECT 1, count() FROM system.numbers' > /dev/null 2>&1 & +wait_for_query_to_start "${CLICKHOUSE_DATABASE}hello" # Replace it -$CLICKHOUSE_CURL -sS "$CLICKHOUSE_URL&query_id=hello&replace_running_query=1" -d 'SELECT 0' +$CLICKHOUSE_CURL -sS "$CLICKHOUSE_URL&query_id=${CLICKHOUSE_DATABASE}hello&replace_running_query=1" -d 'SELECT 0' # Wait for it to be replaced wait +wait_for_queries_to_finish -${CLICKHOUSE_CLIENT_BINARY} --user=u_00600${TEST_PREFIX} --query_id=42 --query='SELECT 2, count() FROM system.numbers' 2>&1 | grep -cF 'was cancelled' & -wait_for_query_to_start '42' +${CLICKHOUSE_CLIENT_BINARY} --user=u_00600${TEST_PREFIX} --query_id="${CLICKHOUSE_DATABASE}42" --query='SELECT 2, count() FROM system.numbers' 2>&1 | grep -cF 'QUERY_WAS_CANCELLED' & +wait_for_query_to_start "${CLICKHOUSE_DATABASE}42" # Trying to run another query with the same query_id -${CLICKHOUSE_CLIENT} --query_id=42 --query='SELECT 43' 2>&1 | grep -cF 'is already running by user' +${CLICKHOUSE_CLIENT} --query_id="${CLICKHOUSE_DATABASE}42" --query='SELECT 43' 2>&1 | grep -cF 'is already running by user' # Trying to replace query of a different user -$CLICKHOUSE_CURL -sS "$CLICKHOUSE_URL&query_id=42&replace_running_query=1" -d 'SELECT 1' | grep -cF 'is already running by user' +$CLICKHOUSE_CURL -sS "$CLICKHOUSE_URL&query_id=${CLICKHOUSE_DATABASE}42&replace_running_query=1" -d 'SELECT 1' | grep -cF 'is already running by user' -$CLICKHOUSE_CURL -sS "$CLICKHOUSE_URL" -d "KILL QUERY WHERE query_id = '42' SYNC" > /dev/null +$CLICKHOUSE_CURL -sS "$CLICKHOUSE_URL" -d "KILL QUERY WHERE query_id = '${CLICKHOUSE_DATABASE}42' SYNC" > /dev/null wait +wait_for_queries_to_finish -${CLICKHOUSE_CLIENT} --query_id=42 --query='SELECT 3, count() FROM system.numbers' 2>&1 | grep -cF 'was cancelled' & -wait_for_query_to_start '42' -${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 +${CLICKHOUSE_CLIENT} --query_id="${CLICKHOUSE_DATABASE}42" --query='SELECT 3, count() FROM system.numbers' 2>&1 | grep -cF 'QUERY_WAS_CANCELLED' & +wait_for_query_to_start "${CLICKHOUSE_DATABASE}42" +${CLICKHOUSE_CLIENT} --query_id="${CLICKHOUSE_DATABASE}42" --replace_running_query=1 --replace_running_query_max_wait_ms=500 --query='SELECT 43' 2>&1 | grep -F "can't be stopped" > /dev/null wait -${CLICKHOUSE_CLIENT} --query_id=42 --replace_running_query=1 --query='SELECT 44' +wait_for_queries_to_finish + +${CLICKHOUSE_CLIENT} --query_id="${CLICKHOUSE_DATABASE}42" --replace_running_query=1 --query='SELECT 44' ${CLICKHOUSE_CLIENT} -q "drop user u_00600${TEST_PREFIX}" diff --git a/tests/queries/0_stateless/00621_regression_for_in_operator.reference b/tests/queries/0_stateless/00621_regression_for_in_operator.reference index b68f550a742..ab8a1499f6d 100644 --- a/tests/queries/0_stateless/00621_regression_for_in_operator.reference +++ b/tests/queries/0_stateless/00621_regression_for_in_operator.reference @@ -17,7 +17,7 @@ QUERY id: 0 LIST id: 5, nodes: 2 COLUMN id: 6, column_name: g, result_type: String, source_id: 3 CONSTANT id: 7, constant_value: Tuple_(\'5\', \'6\'), constant_value_type: Tuple(String, String) - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 2 2 QUERY id: 0 @@ -42,4 +42,4 @@ QUERY id: 0 LIST id: 11, nodes: 2 COLUMN id: 8, column_name: g, result_type: String, source_id: 3 CONSTANT id: 12, constant_value: \'6\', constant_value_type: String - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 diff --git a/tests/queries/0_stateless/00621_regression_for_in_operator.sql b/tests/queries/0_stateless/00621_regression_for_in_operator.sql index db1bcb4a39a..0d8c4933c65 100644 --- a/tests/queries/0_stateless/00621_regression_for_in_operator.sql +++ b/tests/queries/0_stateless/00621_regression_for_in_operator.sql @@ -12,13 +12,13 @@ SELECT count() FROM regression_for_in_operator_view WHERE g IN ('5','6'); SET optimize_min_equality_disjunction_chain_length = 1; SELECT count() FROM regression_for_in_operator_view WHERE g = '5' OR g = '6'; -SELECT count() FROM regression_for_in_operator_view WHERE g = '5' OR g = '6' SETTINGS allow_experimental_analyzer = 1; -EXPLAIN QUERY TREE SELECT count() FROM regression_for_in_operator_view WHERE g = '5' OR g = '6' SETTINGS allow_experimental_analyzer = 1; +SELECT count() FROM regression_for_in_operator_view WHERE g = '5' OR g = '6' SETTINGS enable_analyzer = 1; +EXPLAIN QUERY TREE SELECT count() FROM regression_for_in_operator_view WHERE g = '5' OR g = '6' SETTINGS enable_analyzer = 1; SET optimize_min_equality_disjunction_chain_length = 3; SELECT count() FROM regression_for_in_operator_view WHERE g = '5' OR g = '6'; -SELECT count() FROM regression_for_in_operator_view WHERE g = '5' OR g = '6' SETTINGS allow_experimental_analyzer = 1; -EXPLAIN QUERY TREE SELECT count() FROM regression_for_in_operator_view WHERE g = '5' OR g = '6' SETTINGS allow_experimental_analyzer = 1; +SELECT count() FROM regression_for_in_operator_view WHERE g = '5' OR g = '6' SETTINGS enable_analyzer = 1; +EXPLAIN QUERY TREE SELECT count() FROM regression_for_in_operator_view WHERE g = '5' OR g = '6' SETTINGS enable_analyzer = 1; DROP TABLE regression_for_in_operator_view; DROP TABLE regression_for_in_operator; diff --git a/tests/queries/0_stateless/00634_performance_introspection_and_logging.sh b/tests/queries/0_stateless/00634_performance_introspection_and_logging.sh index 93fd0c4a977..e9a4369a5bf 100755 --- a/tests/queries/0_stateless/00634_performance_introspection_and_logging.sh +++ b/tests/queries/0_stateless/00634_performance_introspection_and_logging.sh @@ -19,13 +19,13 @@ settings="$server_logs --log_queries=1 --log_query_threads=1 --log_profile_event # Test insert logging on each block and checkPacket() method -$CLICKHOUSE_CLIENT $settings -n -q " +$CLICKHOUSE_CLIENT $settings -q " DROP TABLE IF EXISTS null_00634; CREATE TABLE null_00634 (i UInt8) ENGINE = MergeTree PARTITION BY tuple() ORDER BY tuple();" head -c 1000 /dev/zero | $CLICKHOUSE_CLIENT $settings --max_insert_block_size=10 --min_insert_block_size_rows=1 --min_insert_block_size_bytes=1 -q "INSERT INTO null_00634 FORMAT RowBinary" -$CLICKHOUSE_CLIENT $settings -n -q " +$CLICKHOUSE_CLIENT $settings -q " SELECT count() FROM null_00634; DROP TABLE null_00634;" diff --git a/tests/queries/0_stateless/00652_mergetree_mutations.sh b/tests/queries/0_stateless/00652_mergetree_mutations.sh index a9d7908a1af..edb306d3883 100755 --- a/tests/queries/0_stateless/00652_mergetree_mutations.sh +++ b/tests/queries/0_stateless/00652_mergetree_mutations.sh @@ -70,6 +70,21 @@ sleep 1 ${CLICKHOUSE_CLIENT} --query="INSERT INTO mutations_cleaner(x) VALUES (4)" sleep 0.1 +for i in {1..10} +do + + if [ "$(${CLICKHOUSE_CLIENT} --query="SELECT count() FROM system.mutations WHERE database = '$CLICKHOUSE_DATABASE' and table = 'mutations_cleaner'")" -eq 2 ]; then + break + fi + + if [[ $i -eq 100 ]]; then + echo "Timed out while waiting for outdated mutation record to be deleted!" + fi + + sleep 1 + ${CLICKHOUSE_CLIENT} --query="INSERT INTO mutations_cleaner(x) VALUES (4)" +done + # Check that the first mutation is cleaned ${CLICKHOUSE_CLIENT} --query="SELECT mutation_id, command, is_done FROM system.mutations WHERE database = '$CLICKHOUSE_DATABASE' and table = 'mutations_cleaner' ORDER BY mutation_id" diff --git a/tests/queries/0_stateless/00674_join_on_syntax.sql b/tests/queries/0_stateless/00674_join_on_syntax.sql index 9ff26db1536..584e43b88bf 100644 --- a/tests/queries/0_stateless/00674_join_on_syntax.sql +++ b/tests/queries/0_stateless/00674_join_on_syntax.sql @@ -1,5 +1,5 @@ SET joined_subquery_requires_alias = 0; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; drop table if exists tab1; drop table if exists tab2; diff --git a/tests/queries/0_stateless/00700_decimal_compare.sql b/tests/queries/0_stateless/00700_decimal_compare.sql index beadbdade16..1b901e04c28 100644 --- a/tests/queries/0_stateless/00700_decimal_compare.sql +++ b/tests/queries/0_stateless/00700_decimal_compare.sql @@ -27,7 +27,7 @@ SELECT a > 0, b > 0, g > 0 FROM decimal ORDER BY a DESC; SELECT a, g > toInt8(0), g > toInt16(0), g > toInt32(0), g > toInt64(0) FROM decimal ORDER BY a; SELECT a, g > toUInt8(0), g > toUInt16(0), g > toUInt32(0), g > toUInt64(0) FROM decimal ORDER BY a; SELECT a, b, g FROM decimal WHERE a IN(42) AND b IN(42) AND g IN(42); -SELECT a, b, g FROM decimal WHERE a IN(42) AND b IN(42) AND g IN(42) SETTINGS allow_experimental_analyzer = 1; +SELECT a, b, g FROM decimal WHERE a IN(42) AND b IN(42) AND g IN(42) SETTINGS enable_analyzer = 1; SELECT a, b, g FROM decimal WHERE a > 0 AND a <= 42 AND b <= 42 AND g <= 42; SELECT d, e, f from decimal WHERE d > 0 AND d < 1 AND e > 0 AND e < 1 AND f > 0 AND f < 1; diff --git a/tests/queries/0_stateless/00715_fetch_merged_or_mutated_part_zookeeper.sh b/tests/queries/0_stateless/00715_fetch_merged_or_mutated_part_zookeeper.sh index 96d5764780f..d69e14bdbb9 100755 --- a/tests/queries/0_stateless/00715_fetch_merged_or_mutated_part_zookeeper.sh +++ b/tests/queries/0_stateless/00715_fetch_merged_or_mutated_part_zookeeper.sh @@ -8,7 +8,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/mergetree_mutations.lib -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" DROP TABLE IF EXISTS fetches_r1 SYNC; DROP TABLE IF EXISTS fetches_r2 SYNC" @@ -17,7 +17,7 @@ ${CLICKHOUSE_CLIENT} --query="CREATE TABLE fetches_r2(x UInt32) ENGINE Replicate SETTINGS prefer_fetch_merged_part_time_threshold=0, \ prefer_fetch_merged_part_size_threshold=0" -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET insert_keeper_fault_injection_probability=0; INSERT INTO fetches_r1 VALUES (1); INSERT INTO fetches_r1 VALUES (2); @@ -51,6 +51,6 @@ ${CLICKHOUSE_CLIENT} --query="SYSTEM SYNC REPLICA fetches_r2" ${CLICKHOUSE_CLIENT} --query="SELECT '*** Check data after fetch/clone of mutated part ***'" ${CLICKHOUSE_CLIENT} --query="SELECT _part, * FROM fetches_r2 ORDER BY x" -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" DROP TABLE fetches_r1 SYNC; DROP TABLE fetches_r2 SYNC" diff --git a/tests/queries/0_stateless/00722_inner_join.sql b/tests/queries/0_stateless/00722_inner_join.sql index 0d5a543b99d..aa590f470ae 100644 --- a/tests/queries/0_stateless/00722_inner_join.sql +++ b/tests/queries/0_stateless/00722_inner_join.sql @@ -1,6 +1,6 @@ -- Tags: no-parallel -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS one; CREATE TABLE one(dummy UInt8) ENGINE = Memory; diff --git a/tests/queries/0_stateless/00736_disjunction_optimisation.reference b/tests/queries/0_stateless/00736_disjunction_optimisation.reference index f28dcacef0e..4e7f2e09dea 100644 --- a/tests/queries/0_stateless/00736_disjunction_optimisation.reference +++ b/tests/queries/0_stateless/00736_disjunction_optimisation.reference @@ -49,7 +49,7 @@ QUERY id: 0 LIST id: 12, nodes: 2 COLUMN id: 13, column_name: s, result_type: UInt64, source_id: 3 CONSTANT id: 14, constant_value: Tuple_(UInt64_21, UInt64_22, UInt64_23), constant_value_type: Tuple(UInt8, UInt8, UInt8) - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 1 21 1 22 1 23 @@ -99,7 +99,7 @@ QUERY id: 0 LIST id: 14, nodes: 2 COLUMN id: 15, column_name: s, result_type: UInt64, source_id: 3 CONSTANT id: 16, constant_value: Tuple_(UInt64_21, UInt64_22, UInt64_23), constant_value_type: Tuple(UInt8, UInt8, UInt8) - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 1 1 21 1 1 1 1 1 22 0 1 1 1 1 23 0 0 1 @@ -152,7 +152,7 @@ QUERY id: 0 CONSTANT id: 16, constant_value: Tuple_(UInt64_21, UInt64_22, UInt64_23), constant_value_type: Tuple(UInt8, UInt8, UInt8) JOIN TREE TABLE id: 3, alias: __table1, table_name: default.bug - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 21 1 22 1 23 1 @@ -185,7 +185,7 @@ QUERY id: 0 CONSTANT id: 6, constant_value: Tuple_(UInt64_21, UInt64_22, UInt64_23), constant_value_type: Tuple(UInt8, UInt8, UInt8) JOIN TREE TABLE id: 3, alias: __table1, table_name: default.bug - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 1 21 1 22 1 23 @@ -237,7 +237,7 @@ QUERY id: 0 LIST id: 12, nodes: 2 COLUMN id: 13, column_name: s, result_type: UInt64, source_id: 3 CONSTANT id: 14, constant_value: Tuple_(UInt64_21, UInt64_22, UInt64_23), constant_value_type: Tuple(UInt8, UInt8, UInt8) - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 1 21 1 22 1 23 @@ -287,7 +287,7 @@ QUERY id: 0 LIST id: 14, nodes: 2 COLUMN id: 15, column_name: s, result_type: UInt64, source_id: 3 CONSTANT id: 16, constant_value: Tuple_(UInt64_21, UInt64_22, UInt64_23), constant_value_type: Tuple(UInt8, UInt8, UInt8) - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 1 1 21 1 1 1 1 1 22 0 1 1 1 1 23 0 0 1 @@ -348,7 +348,7 @@ QUERY id: 0 CONSTANT id: 21, constant_value: Tuple_(UInt64_21, UInt64_22, UInt64_23), constant_value_type: Tuple(UInt8, UInt8, UInt8) JOIN TREE TABLE id: 3, alias: __table1, table_name: default.bug - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 21 1 22 1 23 1 @@ -381,7 +381,7 @@ QUERY id: 0 CONSTANT id: 6, constant_value: Tuple_(UInt64_21, UInt64_22, UInt64_23), constant_value_type: Tuple(UInt8, UInt8, UInt8) JOIN TREE TABLE id: 3, alias: __table1, table_name: default.bug - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 21 1 22 1 23 1 @@ -414,4 +414,4 @@ QUERY id: 0 CONSTANT id: 6, constant_value: Tuple_(UInt64_21, UInt64_22, UInt64_23), constant_value_type: Tuple(UInt8, UInt8, UInt8) JOIN TREE TABLE id: 3, alias: __table1, table_name: default.bug - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 diff --git a/tests/queries/0_stateless/00736_disjunction_optimisation.sql b/tests/queries/0_stateless/00736_disjunction_optimisation.sql index e5bfc81f7ae..38f77622bd4 100644 --- a/tests/queries/0_stateless/00736_disjunction_optimisation.sql +++ b/tests/queries/0_stateless/00736_disjunction_optimisation.sql @@ -7,43 +7,43 @@ set optimize_min_equality_disjunction_chain_length = 2; select * from bug; select * from bug where (k =1 or k=2 or k =3) and (s=21 or s=22 or s=23); -select * from bug where (k =1 or k=2 or k =3) and (s=21 or s=22 or s=23) SETTINGS allow_experimental_analyzer = 1; -explain query tree select * from bug where (k =1 or k=2 or k =3) and (s=21 or s=22 or s=23) SETTINGS allow_experimental_analyzer = 1;; +select * from bug where (k =1 or k=2 or k =3) and (s=21 or s=22 or s=23) SETTINGS enable_analyzer = 1; +explain query tree select * from bug where (k =1 or k=2 or k =3) and (s=21 or s=22 or s=23) SETTINGS enable_analyzer = 1;; select * from (select * from bug where k=1 or k=2 or k=3) where (s=21 or s=22 or s=23); -select * from (select * from bug where k=1 or k=2 or k=3) where (s=21 or s=22 or s=23) SETTINGS allow_experimental_analyzer = 1;; -explain query tree select * from (select * from bug where k=1 or k=2 or k=3) where (s=21 or s=22 or s=23) SETTINGS allow_experimental_analyzer = 1;; +select * from (select * from bug where k=1 or k=2 or k=3) where (s=21 or s=22 or s=23) SETTINGS enable_analyzer = 1;; +explain query tree select * from (select * from bug where k=1 or k=2 or k=3) where (s=21 or s=22 or s=23) SETTINGS enable_analyzer = 1;; select k, (k=1 or k=2 or k=3), s, (s=21), (s=21 or s=22), (s=21 or s=22 or s=23) from bug; -select k, (k=1 or k=2 or k=3), s, (s=21), (s=21 or s=22), (s=21 or s=22 or s=23) from bug SETTINGS allow_experimental_analyzer = 1;; -explain query tree select k, (k=1 or k=2 or k=3), s, (s=21), (s=21 or s=22), (s=21 or s=22 or s=23) from bug SETTINGS allow_experimental_analyzer = 1;; +select k, (k=1 or k=2 or k=3), s, (s=21), (s=21 or s=22), (s=21 or s=22 or s=23) from bug SETTINGS enable_analyzer = 1;; +explain query tree select k, (k=1 or k=2 or k=3), s, (s=21), (s=21 or s=22), (s=21 or s=22 or s=23) from bug SETTINGS enable_analyzer = 1;; select s, (s=21 or s=22 or s=23) from bug; -select s, (s=21 or s=22 or s=23) from bug SETTINGS allow_experimental_analyzer = 1;; -explain query tree select s, (s=21 or s=22 or s=23) from bug SETTINGS allow_experimental_analyzer = 1;; +select s, (s=21 or s=22 or s=23) from bug SETTINGS enable_analyzer = 1;; +explain query tree select s, (s=21 or s=22 or s=23) from bug SETTINGS enable_analyzer = 1;; set optimize_min_equality_disjunction_chain_length = 3; select * from bug; select * from bug where (k =1 or k=2 or k =3) and (s=21 or s=22 or s=23); -select * from bug where (k =1 or k=2 or k =3) and (s=21 or s=22 or s=23) SETTINGS allow_experimental_analyzer = 1; -explain query tree select * from bug where (k =1 or k=2 or k =3) and (s=21 or s=22 or s=23) SETTINGS allow_experimental_analyzer = 1;; +select * from bug where (k =1 or k=2 or k =3) and (s=21 or s=22 or s=23) SETTINGS enable_analyzer = 1; +explain query tree select * from bug where (k =1 or k=2 or k =3) and (s=21 or s=22 or s=23) SETTINGS enable_analyzer = 1;; select * from (select * from bug where k=1 or k=2 or k=3) where (s=21 or s=22 or s=23); -select * from (select * from bug where k=1 or k=2 or k=3) where (s=21 or s=22 or s=23) SETTINGS allow_experimental_analyzer = 1;; -explain query tree select * from (select * from bug where k=1 or k=2 or k=3) where (s=21 or s=22 or s=23) SETTINGS allow_experimental_analyzer = 1;; +select * from (select * from bug where k=1 or k=2 or k=3) where (s=21 or s=22 or s=23) SETTINGS enable_analyzer = 1;; +explain query tree select * from (select * from bug where k=1 or k=2 or k=3) where (s=21 or s=22 or s=23) SETTINGS enable_analyzer = 1;; select k, (k=1 or k=2 or k=3), s, (s=21), (s=21 or s=22), (s=21 or s=22 or s=23) from bug; -select k, (k=1 or k=2 or k=3), s, (s=21), (s=21 or s=22), (s=21 or s=22 or s=23) from bug SETTINGS allow_experimental_analyzer = 1;; -explain query tree select k, (k=1 or k=2 or k=3), s, (s=21), (s=21 or s=22), (s=21 or s=22 or s=23) from bug SETTINGS allow_experimental_analyzer = 1;; +select k, (k=1 or k=2 or k=3), s, (s=21), (s=21 or s=22), (s=21 or s=22 or s=23) from bug SETTINGS enable_analyzer = 1;; +explain query tree select k, (k=1 or k=2 or k=3), s, (s=21), (s=21 or s=22), (s=21 or s=22 or s=23) from bug SETTINGS enable_analyzer = 1;; select s, (s=21 or s=22 or s=23) from bug; -select s, (s=21 or s=22 or s=23) from bug SETTINGS allow_experimental_analyzer = 1;; -explain query tree select s, (s=21 or s=22 or s=23) from bug SETTINGS allow_experimental_analyzer = 1;; +select s, (s=21 or s=22 or s=23) from bug SETTINGS enable_analyzer = 1;; +explain query tree select s, (s=21 or s=22 or s=23) from bug SETTINGS enable_analyzer = 1;; select s, (s=21 or 22=s or 23=s) from bug; -select s, (s=21 or 22=s or 23=s) from bug SETTINGS allow_experimental_analyzer = 1;; -explain query tree select s, (s=21 or 22=s or 23=s) from bug SETTINGS allow_experimental_analyzer = 1;; +select s, (s=21 or 22=s or 23=s) from bug SETTINGS enable_analyzer = 1;; +explain query tree select s, (s=21 or 22=s or 23=s) from bug SETTINGS enable_analyzer = 1;; DROP TABLE bug; diff --git a/tests/queries/0_stateless/00754_distributed_optimize_skip_select_on_unused_shards.sh b/tests/queries/0_stateless/00754_distributed_optimize_skip_select_on_unused_shards.sh index 09f20284402..989096a26d6 100755 --- a/tests/queries/0_stateless/00754_distributed_optimize_skip_select_on_unused_shards.sh +++ b/tests/queries/0_stateless/00754_distributed_optimize_skip_select_on_unused_shards.sh @@ -25,83 +25,83 @@ ${CLICKHOUSE_CLIENT} --query "SELECT count(*) FROM distributed WHERE a = 0 AND b | grep -F -q "All connection tries failed" && echo 'OK' || echo 'FAIL' # Should pass now -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed WHERE a = 0 AND b = 0; " # Should still fail because of matching unavailable shard -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed WHERE a = 2 AND b = 2; " 2>&1 \ | grep -F -q "All connection tries failed" && echo 'OK' || echo 'FAIL' # Try more complext expressions for constant folding - all should pass. -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed WHERE a = 1 AND a = 0 AND b = 0; " -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed WHERE a IN (0, 1) AND b IN (0, 1); " -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed WHERE a = 0 AND b = 0 OR a = 1 AND b = 1; " # TODO: should pass one day. -#${CLICKHOUSE_CLIENT} -n --query=" +#${CLICKHOUSE_CLIENT} --query=" # SET optimize_skip_unused_shards = 1; # SELECT count(*) FROM distributed WHERE a = 0 AND b >= 0 AND b <= 1; #" -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed WHERE a = 0 AND b = 0 AND c = 0; " -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed WHERE a = 0 AND b = 0 AND c != 10; " -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed WHERE a = 0 AND b = 0 AND (a+b)*b != 12; " -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed WHERE (a = 0 OR a = 1) AND (b = 0 OR b = 1); " # These ones should fail. -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed WHERE a = 0 AND b <= 1; " 2>&1 \ | grep -F -q "All connection tries failed" && echo 'OK' || echo 'FAIL' -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed WHERE a = 0 AND c = 0; " 2>&1 \ | grep -F -q "All connection tries failed" && echo 'OK' || echo 'FAIL' -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed WHERE a = 0 OR a = 1 AND b = 0; " 2>&1 \ | grep -F -q "All connection tries failed" && echo 'OK' || echo 'FAIL' -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed WHERE a = 0 AND b = 0 OR a = 2 AND b = 2; " 2>&1 \ | grep -F -q "All connection tries failed" && echo 'OK' || echo 'FAIL' -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed WHERE a = 0 AND b = 0 OR c = 0; " 2>&1 \ | grep -F -q "All connection tries failed" && echo 'OK' || echo 'FAIL' diff --git a/tests/queries/0_stateless/00754_distributed_optimize_skip_select_on_unused_shards_with_prewhere.sh b/tests/queries/0_stateless/00754_distributed_optimize_skip_select_on_unused_shards_with_prewhere.sh index 035907bddd7..b3dff2ea69a 100755 --- a/tests/queries/0_stateless/00754_distributed_optimize_skip_select_on_unused_shards_with_prewhere.sh +++ b/tests/queries/0_stateless/00754_distributed_optimize_skip_select_on_unused_shards_with_prewhere.sh @@ -30,73 +30,73 @@ ${CLICKHOUSE_CLIENT} --query "SELECT count(*) FROM distributed_00754 PREWHERE a | grep -F -q "All connection tries failed" && echo 'OK' || echo 'FAIL' # Should pass now -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed_00754 PREWHERE a = 0 AND b = 0; " # Should still fail because of matching unavailable shard -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed_00754 PREWHERE a = 2 AND b = 2; " 2>&1 \ | grep -F -q "All connection tries failed" && echo 'OK' || echo 'FAIL' # Try more complex expressions for constant folding - all should pass. -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed_00754 PREWHERE a = 1 AND a = 0 WHERE b = 0; " -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed_00754 PREWHERE a = 1 WHERE b = 1 AND length(c) = 5; " -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed_00754 PREWHERE a IN (0, 1) AND b IN (0, 1) WHERE c LIKE '%l%'; " -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed_00754 PREWHERE a IN (0, 1) WHERE b IN (0, 1) AND c LIKE '%l%'; " -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed_00754 PREWHERE a = 0 AND b = 0 OR a = 1 AND b = 1 WHERE c LIKE '%l%'; " -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed_00754 PREWHERE (a = 0 OR a = 1) WHERE (b = 0 OR b = 1); " # These should fail. -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed_00754 PREWHERE a = 0 AND b <= 1; " 2>&1 \ | grep -F -q "All connection tries failed" && echo 'OK' || echo 'FAIL' -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed_00754 PREWHERE a = 0 WHERE c LIKE '%l%'; " 2>&1 \ | grep -F -q "All connection tries failed" && echo 'OK' || echo 'FAIL' -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed_00754 PREWHERE a = 0 OR a = 1 AND b = 0; " 2>&1 \ | grep -F -q "All connection tries failed" && echo 'OK' || echo 'FAIL' -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed_00754 PREWHERE a = 0 AND b = 0 OR a = 2 AND b = 2; " 2>&1 \ | grep -F -q "All connection tries failed" && echo 'OK' || echo 'FAIL' -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET optimize_skip_unused_shards = 1; SELECT count(*) FROM distributed_00754 PREWHERE a = 0 AND b = 0 OR c LIKE '%l%'; " 2>&1 \ | grep -F -q "All connection tries failed" && echo 'OK' || echo 'FAIL' diff --git a/tests/queries/0_stateless/00757_enum_defaults_const.sql b/tests/queries/0_stateless/00757_enum_defaults_const.sql index 64271a37473..048c9dee88f 100644 --- a/tests/queries/0_stateless/00757_enum_defaults_const.sql +++ b/tests/queries/0_stateless/00757_enum_defaults_const.sql @@ -1,3 +1,3 @@ -SET allow_experimental_analyzer=0; +SET enable_analyzer=0; select os_name, count() from (SELECT CAST('iphone' AS Enum8('iphone' = 1, 'android' = 2)) AS os_name) group by os_name WITH TOTALS; select toNullable(os_name) AS os_name, count() from (SELECT CAST('iphone' AS Enum8('iphone' = 1, 'android' = 2)) AS os_name) group by os_name WITH TOTALS; diff --git a/tests/queries/0_stateless/00757_enum_defaults_const_analyzer.sql b/tests/queries/0_stateless/00757_enum_defaults_const_analyzer.sql index bf079539019..c202ed630db 100644 --- a/tests/queries/0_stateless/00757_enum_defaults_const_analyzer.sql +++ b/tests/queries/0_stateless/00757_enum_defaults_const_analyzer.sql @@ -1,3 +1,3 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; select os_name, count() from (SELECT CAST('iphone' AS Enum8('iphone' = 1, 'android' = 2)) AS os_name) group by os_name WITH TOTALS; select toNullable(os_name) AS os_name, count() from (SELECT CAST('iphone' AS Enum8('iphone' = 1, 'android' = 2)) AS os_name) group by os_name WITH TOTALS; diff --git a/tests/queries/0_stateless/00800_low_cardinality_join.sql b/tests/queries/0_stateless/00800_low_cardinality_join.sql index ecb5194253c..fc5f5d1860c 100644 --- a/tests/queries/0_stateless/00800_low_cardinality_join.sql +++ b/tests/queries/0_stateless/00800_low_cardinality_join.sql @@ -1,4 +1,4 @@ -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; set joined_subquery_requires_alias = 0; select * from (select dummy as val from system.one) any left join (select dummy as val from system.one) using val; diff --git a/tests/queries/0_stateless/00800_low_cardinality_merge_join.sql.j2 b/tests/queries/0_stateless/00800_low_cardinality_merge_join.sql.j2 index 8e2037480c7..a5f5e9b33b8 100644 --- a/tests/queries/0_stateless/00800_low_cardinality_merge_join.sql.j2 +++ b/tests/queries/0_stateless/00800_low_cardinality_merge_join.sql.j2 @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; {% for join_algorithm in ['partial_merge', 'full_sorting_merge', 'grace_hash'] -%} diff --git a/tests/queries/0_stateless/00814_replicated_minimalistic_part_header_zookeeper.sh b/tests/queries/0_stateless/00814_replicated_minimalistic_part_header_zookeeper.sh index 12d889a7137..8f7a1a9ae98 100755 --- a/tests/queries/0_stateless/00814_replicated_minimalistic_part_header_zookeeper.sh +++ b/tests/queries/0_stateless/00814_replicated_minimalistic_part_header_zookeeper.sh @@ -10,7 +10,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) SHARD=$($CLICKHOUSE_CLIENT --query "Select getMacro('shard')") REPLICA=$($CLICKHOUSE_CLIENT --query "Select getMacro('replica')") -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " DROP TABLE IF EXISTS part_header_r1; DROP TABLE IF EXISTS part_header_r2; @@ -62,7 +62,7 @@ do [[ $count1 == 1 && $count2 == 1 ]] && break done -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " SELECT '*** Test part removal ***'; SELECT '*** replica 1 ***'; diff --git a/tests/queries/0_stateless/00818_alias_bug_4110.sql b/tests/queries/0_stateless/00818_alias_bug_4110.sql index d057bacc908..1242a3b605d 100644 --- a/tests/queries/0_stateless/00818_alias_bug_4110.sql +++ b/tests/queries/0_stateless/00818_alias_bug_4110.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; select s.a as a, s.a + 1 as b from (select 10 as a) s; select s.a + 1 as a, s.a as b from (select 10 as a) s; diff --git a/tests/queries/0_stateless/00818_inner_join_bug_3567.sql b/tests/queries/0_stateless/00818_inner_join_bug_3567.sql index 2dec5ce3221..3b4b3cd77d3 100644 --- a/tests/queries/0_stateless/00818_inner_join_bug_3567.sql +++ b/tests/queries/0_stateless/00818_inner_join_bug_3567.sql @@ -1,5 +1,5 @@ SET output_format_pretty_color = 1; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS table1; DROP TABLE IF EXISTS table2; diff --git a/tests/queries/0_stateless/00819_full_join_wrong_columns_in_block.sql b/tests/queries/0_stateless/00819_full_join_wrong_columns_in_block.sql index 3c0246619da..8b3d3ec1bc1 100644 --- a/tests/queries/0_stateless/00819_full_join_wrong_columns_in_block.sql +++ b/tests/queries/0_stateless/00819_full_join_wrong_columns_in_block.sql @@ -1,6 +1,6 @@ SET any_join_distinct_right_table_keys = 1; SET joined_subquery_requires_alias = 0; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT * FROM (SELECT 1 AS a, 'x' AS b) join (SELECT 1 as a, 'y' as b) using a; SELECT * FROM (SELECT 1 AS a, 'x' AS b) left join (SELECT 1 as a, 'y' as b) using a; diff --git a/tests/queries/0_stateless/00820_multiple_joins.sql b/tests/queries/0_stateless/00820_multiple_joins.sql index 5c7a7bebb0b..b4197570cfa 100644 --- a/tests/queries/0_stateless/00820_multiple_joins.sql +++ b/tests/queries/0_stateless/00820_multiple_joins.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS table1; DROP TABLE IF EXISTS table2; diff --git a/tests/queries/0_stateless/00820_multiple_joins_subquery_requires_alias.sql b/tests/queries/0_stateless/00820_multiple_joins_subquery_requires_alias.sql index 3da2cad4eff..538e6967ff3 100644 --- a/tests/queries/0_stateless/00820_multiple_joins_subquery_requires_alias.sql +++ b/tests/queries/0_stateless/00820_multiple_joins_subquery_requires_alias.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS table1; DROP TABLE IF EXISTS table2; diff --git a/tests/queries/0_stateless/00830_join_overwrite.sql b/tests/queries/0_stateless/00830_join_overwrite.sql index bc3662528db..f51a152ea46 100644 --- a/tests/queries/0_stateless/00830_join_overwrite.sql +++ b/tests/queries/0_stateless/00830_join_overwrite.sql @@ -13,7 +13,7 @@ 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; +SET enable_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; diff --git a/tests/queries/0_stateless/00834_kill_mutation_replicated_zookeeper.sh b/tests/queries/0_stateless/00834_kill_mutation_replicated_zookeeper.sh index 16ad08deeb2..3e6b339cb57 100755 --- a/tests/queries/0_stateless/00834_kill_mutation_replicated_zookeeper.sh +++ b/tests/queries/0_stateless/00834_kill_mutation_replicated_zookeeper.sh @@ -1,5 +1,6 @@ #!/usr/bin/env bash -# Tags: replica, no-debug +# Tags: replica, no-debug, no-fasttest +# no-fasttest: Waiting for failed mutations is slow: https://github.com/ClickHouse/ClickHouse/issues/67936 CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/00835_if_generic_case.sql b/tests/queries/0_stateless/00835_if_generic_case.sql index 051fad14603..e06e610dbb5 100644 --- a/tests/queries/0_stateless/00835_if_generic_case.sql +++ b/tests/queries/0_stateless/00835_if_generic_case.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT toDateTime('2000-01-01 00:00:00', 'Asia/Istanbul') AS x, toDate('2000-01-02') AS y, x > y ? x : y AS z; SELECT materialize(toDateTime('2000-01-01 00:00:00', 'Asia/Istanbul')) AS x, toDate('2000-01-02') AS y, x > y ? x : y AS z; diff --git a/tests/queries/0_stateless/00837_minmax_index.sh b/tests/queries/0_stateless/00837_minmax_index.sh index e4de0b9ebfc..ff487f50ee0 100755 --- a/tests/queries/0_stateless/00837_minmax_index.sh +++ b/tests/queries/0_stateless/00837_minmax_index.sh @@ -8,7 +8,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) $CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS minmax_idx;" -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" CREATE TABLE minmax_idx ( u64 UInt64, diff --git a/tests/queries/0_stateless/00838_unique_index.sh b/tests/queries/0_stateless/00838_unique_index.sh index b267b6a8eb3..a3aba4f26b6 100755 --- a/tests/queries/0_stateless/00838_unique_index.sh +++ b/tests/queries/0_stateless/00838_unique_index.sh @@ -7,7 +7,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) $CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS set_idx;" -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" CREATE TABLE set_idx ( u64 UInt64, diff --git a/tests/queries/0_stateless/00848_join_use_nulls_segfault.sql b/tests/queries/0_stateless/00848_join_use_nulls_segfault.sql index 2f6cca0284c..275968236ae 100644 --- a/tests/queries/0_stateless/00848_join_use_nulls_segfault.sql +++ b/tests/queries/0_stateless/00848_join_use_nulls_segfault.sql @@ -1,5 +1,5 @@ SET any_join_distinct_right_table_keys = 1; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS t1_00848; DROP TABLE IF EXISTS t2_00848; diff --git a/tests/queries/0_stateless/00849_multiple_comma_join_2.sql b/tests/queries/0_stateless/00849_multiple_comma_join_2.sql index 6530f691087..250dd4a47ab 100644 --- a/tests/queries/0_stateless/00849_multiple_comma_join_2.sql +++ b/tests/queries/0_stateless/00849_multiple_comma_join_2.sql @@ -12,7 +12,7 @@ CREATE TABLE t2 (a UInt32, b Nullable(Int32)) ENGINE = Memory; CREATE TABLE t3 (a UInt32, b Nullable(Int32)) ENGINE = Memory; CREATE TABLE t4 (a UInt32, b Nullable(Int32)) ENGINE = Memory; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; --- EXPLAIN SYNTAX (old AST based optimization) SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( @@ -62,56 +62,56 @@ SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explai --- EXPLAIN QUERY TREE SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( - EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2 WHERE t1.a = t2.a) SETTINGS allow_experimental_analyzer = 1; + EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2 WHERE t1.a = t2.a) SETTINGS enable_analyzer = 1; SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( - EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2 WHERE t1.b = t2.b) SETTINGS allow_experimental_analyzer = 1; + EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2 WHERE t1.b = t2.b) SETTINGS enable_analyzer = 1; SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( - EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3 WHERE t1.a = t2.a AND t1.a = t3.a) SETTINGS allow_experimental_analyzer = 1; + EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3 WHERE t1.a = t2.a AND t1.a = t3.a) SETTINGS enable_analyzer = 1; SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( - EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3 WHERE t1.b = t2.b AND t1.b = t3.b) SETTINGS allow_experimental_analyzer = 1; + EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3 WHERE t1.b = t2.b AND t1.b = t3.b) SETTINGS enable_analyzer = 1; SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( - EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3, t4 WHERE t1.a = t2.a AND t1.a = t3.a AND t1.a = t4.a) SETTINGS allow_experimental_analyzer = 1; + EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3, t4 WHERE t1.a = t2.a AND t1.a = t3.a AND t1.a = t4.a) SETTINGS enable_analyzer = 1; SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( - EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3, t4 WHERE t1.b = t2.b AND t1.b = t3.b AND t1.b = t4.b) SETTINGS allow_experimental_analyzer = 1; + EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3, t4 WHERE t1.b = t2.b AND t1.b = t3.b AND t1.b = t4.b) SETTINGS enable_analyzer = 1; SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( - EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3, t4 WHERE t2.a = t1.a AND t2.a = t3.a AND t2.a = t4.a) SETTINGS allow_experimental_analyzer = 1; + EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3, t4 WHERE t2.a = t1.a AND t2.a = t3.a AND t2.a = t4.a) SETTINGS enable_analyzer = 1; SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( - EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3, t4 WHERE t3.a = t1.a AND t3.a = t2.a AND t3.a = t4.a) SETTINGS allow_experimental_analyzer = 1; + EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3, t4 WHERE t3.a = t1.a AND t3.a = t2.a AND t3.a = t4.a) SETTINGS enable_analyzer = 1; SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( - EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3, t4 WHERE t4.a = t1.a AND t4.a = t2.a AND t4.a = t3.a) SETTINGS allow_experimental_analyzer = 1; + EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3, t4 WHERE t4.a = t1.a AND t4.a = t2.a AND t4.a = t3.a) SETTINGS enable_analyzer = 1; SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( - EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3, t4 WHERE t1.a = t2.a AND t2.a = t3.a AND t3.a = t4.a) SETTINGS allow_experimental_analyzer = 1; + EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3, t4 WHERE t1.a = t2.a AND t2.a = t3.a AND t3.a = t4.a) SETTINGS enable_analyzer = 1; SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( - EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3, t4) SETTINGS allow_experimental_analyzer = 1; + EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2, t3, t4) SETTINGS enable_analyzer = 1; SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( - EXPLAIN QUERY TREE SELECT t1.a FROM t1 CROSS JOIN t2 CROSS JOIN t3 CROSS JOIN t4) SETTINGS allow_experimental_analyzer = 1; + EXPLAIN QUERY TREE SELECT t1.a FROM t1 CROSS JOIN t2 CROSS JOIN t3 CROSS JOIN t4) SETTINGS enable_analyzer = 1; SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( - EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2 CROSS JOIN t3) SETTINGS allow_experimental_analyzer = 1; + EXPLAIN QUERY TREE SELECT t1.a FROM t1, t2 CROSS JOIN t3) SETTINGS enable_analyzer = 1; SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( - EXPLAIN QUERY TREE SELECT t1.a FROM t1 JOIN t2 USING a CROSS JOIN t3) SETTINGS allow_experimental_analyzer = 1; + EXPLAIN QUERY TREE SELECT t1.a FROM t1 JOIN t2 USING a CROSS JOIN t3) SETTINGS enable_analyzer = 1; SELECT countIf(explain like '%COMMA%' OR explain like '%CROSS%'), countIf(explain like '%INNER%') FROM ( - EXPLAIN QUERY TREE SELECT t1.a FROM t1 JOIN t2 ON t1.a = t2.a CROSS JOIN t3) SETTINGS allow_experimental_analyzer = 1; + EXPLAIN QUERY TREE SELECT t1.a FROM t1 JOIN t2 ON t1.a = t2.a CROSS JOIN t3) SETTINGS enable_analyzer = 1; INSERT INTO t1 values (1,1), (2,2), (3,3), (4,4); INSERT INTO t2 values (1,1), (1, Null); INSERT INTO t3 values (1,1), (1, Null); INSERT INTO t4 values (1,1), (1, Null); -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT 'SELECT * FROM t1, t2'; SELECT * FROM t1, t2 diff --git a/tests/queries/0_stateless/00855_join_with_array_join.sql b/tests/queries/0_stateless/00855_join_with_array_join.sql index c278ff0738a..2d5bc6cb1f6 100644 --- a/tests/queries/0_stateless/00855_join_with_array_join.sql +++ b/tests/queries/0_stateless/00855_join_with_array_join.sql @@ -1,5 +1,5 @@ SET joined_subquery_requires_alias = 0; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT ax, c FROM (SELECT [1,2] ax, 0 c) ARRAY JOIN ax JOIN (SELECT 0 c) USING (c); SELECT ax, c FROM (SELECT [3,4] ax, 0 c) JOIN (SELECT 0 c) USING (c) ARRAY JOIN ax; diff --git a/tests/queries/0_stateless/00858_issue_4756.sql b/tests/queries/0_stateless/00858_issue_4756.sql index 9eacd5ef364..3f6ab037c02 100644 --- a/tests/queries/0_stateless/00858_issue_4756.sql +++ b/tests/queries/0_stateless/00858_issue_4756.sql @@ -1,4 +1,4 @@ -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; set distributed_product_mode = 'local'; drop table if exists shard1; diff --git a/tests/queries/0_stateless/00897_flatten.sql b/tests/queries/0_stateless/00897_flatten.sql index 0d67a1708fd..45d1a225a49 100644 --- a/tests/queries/0_stateless/00897_flatten.sql +++ b/tests/queries/0_stateless/00897_flatten.sql @@ -2,7 +2,7 @@ SELECT flatten(arrayJoin([[[1, 2, 3], [4, 5]], [[6], [7, 8]]])); SELECT arrayFlatten(arrayJoin([[[[]], [[1], [], [2, 3]]], [[[4]]]])); SELECT flatten(arrayMap(x -> arrayMap(y -> arrayMap(z -> range(x), range(x)), range(x)), range(number))) FROM numbers(6); SELECT flatten(arrayMap(x -> arrayMap(y -> arrayMap(z -> range(z), range(y)), range(x)), range(number))) FROM numbers(6); -SELECT flatten(arrayMap(x -> arrayMap(x -> arrayMap(x -> range(x), range(x)), range(x)), range(number))) FROM numbers(6) SETTINGS allow_experimental_analyzer=1; +SELECT flatten(arrayMap(x -> arrayMap(x -> arrayMap(x -> range(x), range(x)), range(x)), range(number))) FROM numbers(6) SETTINGS enable_analyzer=1; SELECT arrayFlatten([[[1, 2, 3], [4, 5]], [[6], [7, 8]]]); SELECT flatten([[[]]]); SELECT arrayFlatten([]); diff --git a/tests/queries/0_stateless/00900_long_parquet_load.sh b/tests/queries/0_stateless/00900_long_parquet_load.sh index 1bafb033f56..3a7022ac0cf 100755 --- a/tests/queries/0_stateless/00900_long_parquet_load.sh +++ b/tests/queries/0_stateless/00900_long_parquet_load.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-fasttest, no-debug +# Tags: long, no-fasttest, no-debug, no-asan, no-msan, no-tsan # # Load all possible .parquet files found in submodules. diff --git a/tests/queries/0_stateless/00907_set_index_max_rows.sh b/tests/queries/0_stateless/00907_set_index_max_rows.sh index 3707aaf2ca6..bdd0f36346f 100755 --- a/tests/queries/0_stateless/00907_set_index_max_rows.sh +++ b/tests/queries/0_stateless/00907_set_index_max_rows.sh @@ -6,7 +6,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) $CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS set_idx;" -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" CREATE TABLE set_idx ( u64 UInt64, diff --git a/tests/queries/0_stateless/00908_bloom_filter_index.reference b/tests/queries/0_stateless/00908_bloom_filter_index.reference index c0cbd6c0335..e2e13a9ed12 100644 --- a/tests/queries/0_stateless/00908_bloom_filter_index.reference +++ b/tests/queries/0_stateless/00908_bloom_filter_index.reference @@ -28,6 +28,8 @@ "rows_read": 3, 8 aŠ±Š²Š³Š“ŠµŃ‘Š¶ "rows_read": 2, +13 +1 1 column-oriented 2 column-oriented "rows_read": 4, diff --git a/tests/queries/0_stateless/00908_bloom_filter_index.sh b/tests/queries/0_stateless/00908_bloom_filter_index.sh index 88fc7944236..3bd169dd6df 100755 --- a/tests/queries/0_stateless/00908_bloom_filter_index.sh +++ b/tests/queries/0_stateless/00908_bloom_filter_index.sh @@ -12,7 +12,7 @@ $CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS bloom_filter_idx3;" # NGRAM BF -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" CREATE TABLE bloom_filter_idx ( k UInt64, @@ -22,7 +22,7 @@ CREATE TABLE bloom_filter_idx ORDER BY k SETTINGS index_granularity = 2, index_granularity_bytes = '10Mi';" -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" CREATE TABLE bloom_filter_idx2 ( k UInt64, @@ -103,9 +103,13 @@ $CLICKHOUSE_CLIENT --optimize_or_like_chain 0 --query="SELECT * FROM bloom_filte $CLICKHOUSE_CLIENT --optimize_or_like_chain 0 --query="SELECT * FROM bloom_filter_idx WHERE (s, lower(s)) IN (('aŠ±Š²Š³Š“ŠµŃ‘Š¶', 'aŠ±Š²Š³Š“ŠµŃ‘Š¶'), ('abc', 'cba')) ORDER BY k" $CLICKHOUSE_CLIENT --optimize_or_like_chain 0 --query="SELECT * FROM bloom_filter_idx WHERE (s, lower(s)) IN (('aŠ±Š²Š³Š“ŠµŃ‘Š¶', 'aŠ±Š²Š³Š“ŠµŃ‘Š¶'), ('abc', 'cba')) ORDER BY k FORMAT JSON" | grep "rows_read" +# Weird conditions not supported by the index. +$CLICKHOUSE_CLIENT --optimize_or_like_chain 0 --query="SELECT count() FROM bloom_filter_idx WHERE (s = 'asd') = (s = 'asd')" +$CLICKHOUSE_CLIENT --optimize_or_like_chain 0 --query="SELECT count() FROM bloom_filter_idx WHERE has(['asd', 'some string'], s)" + # TOKEN BF -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" CREATE TABLE bloom_filter_idx3 ( k UInt64, @@ -143,7 +147,7 @@ $CLICKHOUSE_CLIENT --query="DROP TABLE bloom_filter_idx2" $CLICKHOUSE_CLIENT --query="DROP TABLE bloom_filter_idx3" $CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS bloom_filter_idx_na;" -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" CREATE TABLE bloom_filter_idx_na ( na Array(Array(String)), @@ -152,7 +156,7 @@ CREATE TABLE bloom_filter_idx_na ORDER BY na" 2>&1 | grep -c 'DB::Exception: Unexpected type Array(Array(String)) of bloom filter index' # NGRAM BF with IPv6 -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" CREATE TABLE bloom_filter_ipv6_idx ( foo IPv6, diff --git a/tests/queries/0_stateless/00940_order_by_read_in_order_query_plan.sql b/tests/queries/0_stateless/00940_order_by_read_in_order_query_plan.sql index 532539206f7..0421cadf868 100644 --- a/tests/queries/0_stateless/00940_order_by_read_in_order_query_plan.sql +++ b/tests/queries/0_stateless/00940_order_by_read_in_order_query_plan.sql @@ -1,4 +1,4 @@ -SET optimize_read_in_order = 1, query_plan_read_in_order = 1, allow_experimental_analyzer = 0; +SET optimize_read_in_order = 1, query_plan_read_in_order = 1, enable_analyzer = 0; drop table if exists tab; drop table if exists tab2; diff --git a/tests/queries/0_stateless/00942_mutate_index.sh b/tests/queries/0_stateless/00942_mutate_index.sh index 6ebb30c25b9..e1e23639e85 100755 --- a/tests/queries/0_stateless/00942_mutate_index.sh +++ b/tests/queries/0_stateless/00942_mutate_index.sh @@ -6,7 +6,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) $CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS minmax_idx;" -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" CREATE TABLE minmax_idx ( u64 UInt64, diff --git a/tests/queries/0_stateless/00943_materialize_index.sh b/tests/queries/0_stateless/00943_materialize_index.sh index 6ff7d34a9d7..e4a585fce97 100755 --- a/tests/queries/0_stateless/00943_materialize_index.sh +++ b/tests/queries/0_stateless/00943_materialize_index.sh @@ -7,7 +7,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) $CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS minmax_idx;" -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" CREATE TABLE minmax_idx ( u64 UInt64, @@ -34,7 +34,7 @@ $CLICKHOUSE_CLIENT --query="INSERT INTO minmax_idx VALUES $CLICKHOUSE_CLIENT --query="SELECT count() FROM minmax_idx WHERE i64 = 2 SETTINGS merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability = 0.0;" $CLICKHOUSE_CLIENT --query="SELECT count() FROM minmax_idx WHERE i64 = 2 SETTINGS merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability = 0.0 FORMAT JSON" | grep "rows_read" -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" ALTER TABLE minmax_idx ADD INDEX idx (i64, u64 * i64) TYPE minmax GRANULARITY 1 SETTINGS mutations_sync = 2;" $CLICKHOUSE_CLIENT --query="ALTER TABLE minmax_idx MATERIALIZE INDEX idx IN PARTITION 1 SETTINGS mutations_sync = 2;" diff --git a/tests/queries/0_stateless/00944_clear_index_in_partition.sh b/tests/queries/0_stateless/00944_clear_index_in_partition.sh index 4655077960f..a12536da239 100755 --- a/tests/queries/0_stateless/00944_clear_index_in_partition.sh +++ b/tests/queries/0_stateless/00944_clear_index_in_partition.sh @@ -7,7 +7,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) $CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS minmax_idx;" -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" CREATE TABLE minmax_idx ( u64 UInt64, diff --git a/tests/queries/0_stateless/00945_bloom_filter_index.reference b/tests/queries/0_stateless/00945_bloom_filter_index.reference index c0c2254648e..9d9b49b29c9 100644 --- a/tests/queries/0_stateless/00945_bloom_filter_index.reference +++ b/tests/queries/0_stateless/00945_bloom_filter_index.reference @@ -14,6 +14,11 @@ 0 2 2 +18 +100 +100 +3 +100 1 1 1 @@ -222,3 +227,5 @@ 1 value1 1 value2 2 value3 +1 +1 diff --git a/tests/queries/0_stateless/00945_bloom_filter_index.sql b/tests/queries/0_stateless/00945_bloom_filter_index.sql index 4c26988574a..6e3819e74d3 100644 --- a/tests/queries/0_stateless/00945_bloom_filter_index.sql +++ b/tests/queries/0_stateless/00945_bloom_filter_index.sql @@ -25,6 +25,15 @@ WITH ((1, 2), (2, 3)) AS liter_prepared_set SELECT COUNT() FROM single_column_bl WITH ((1, 1), (2, 2)) AS liter_prepared_set SELECT COUNT() FROM single_column_bloom_filter WHERE (i32, i64) IN liter_prepared_set SETTINGS max_rows_to_read = 6; WITH ((1, (1, 1)), (2, (2, 2))) AS liter_prepared_set SELECT COUNT() FROM single_column_bloom_filter WHERE (i64, (i64, i32)) IN liter_prepared_set SETTINGS max_rows_to_read = 6; +-- Check that indexHint() works (but it doesn't work with COUNT()). +SELECT SUM(ignore(*) + 1) FROM single_column_bloom_filter WHERE indexHint(i32 in (3, 15, 50)); + +-- The index doesn't understand expressions like these, but it shouldn't break the query. +SELECT COUNT() FROM single_column_bloom_filter WHERE (i32 = 200) = (i32 = 200); +SELECT SUM(ignore(*) + 1) FROM single_column_bloom_filter WHERE indexHint((i32 = 200) != (i32 = 200)); +SELECT COUNT() FROM single_column_bloom_filter WHERE indexOf([10, 20, 30], i32) != 0; +SELECT COUNT() FROM single_column_bloom_filter WHERE has([100, 200, 300], 200); + DROP TABLE IF EXISTS single_column_bloom_filter; @@ -365,3 +374,10 @@ SELECT id, ary[indexOf(ary, 'value2')] FROM test_bf_indexOf WHERE ary[indexOf(ar SELECT id, ary[indexOf(ary, 'value3')] FROM test_bf_indexOf WHERE ary[indexOf(ary, 'value3')] = 'value3' ORDER BY id FORMAT TSV; DROP TABLE IF EXISTS test_bf_indexOf; + +-- Test for bug #65597 +DROP TABLE IF EXISTS test_bf_cast; +CREATE TABLE test_bf_cast (c Int32, INDEX x1 (c) type bloom_filter) ENGINE = MergeTree ORDER BY c AS SELECT 1; +SELECT count() FROM test_bf_cast WHERE cast(c = 1 OR c = 9999 AS Bool) SETTINGS use_skip_indexes=0; +SELECT count() FROM test_bf_cast WHERE cast(c = 1 OR c = 9999 AS Bool) SETTINGS use_skip_indexes=1; +DROP TABLE test_bf_cast; diff --git a/tests/queries/0_stateless/00961_check_table.reference b/tests/queries/0_stateless/00961_check_table.reference index a0a054898b9..686285bb6aa 100644 --- a/tests/queries/0_stateless/00961_check_table.reference +++ b/tests/queries/0_stateless/00961_check_table.reference @@ -14,4 +14,4 @@ ======== 201902_4_5_1 1 ======== -201801_1_1_0 1 +201801_1_1_2 1 diff --git a/tests/queries/0_stateless/00961_check_table.sql b/tests/queries/0_stateless/00961_check_table.sql index a6abe8103d5..fc3c5435670 100644 --- a/tests/queries/0_stateless/00961_check_table.sql +++ b/tests/queries/0_stateless/00961_check_table.sql @@ -39,6 +39,6 @@ CHECK TABLE mt_table PARTITION 201902 SETTINGS max_threads = 1; SELECT '========'; -CHECK TABLE mt_table PART '201801_1_1_0'; +CHECK TABLE mt_table PART '201801_1_1_2'; DROP TABLE IF EXISTS mt_table; diff --git a/tests/queries/0_stateless/00964_bloom_index_string_functions.sh b/tests/queries/0_stateless/00964_bloom_index_string_functions.sh index e2ec7fd42e4..9e410f09b13 100755 --- a/tests/queries/0_stateless/00964_bloom_index_string_functions.sh +++ b/tests/queries/0_stateless/00964_bloom_index_string_functions.sh @@ -7,7 +7,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) $CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS bloom_filter_idx;" # NGRAM BF -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" CREATE TABLE bloom_filter_idx ( k UInt64, diff --git a/tests/queries/0_stateless/00965_set_index_string_functions.sh b/tests/queries/0_stateless/00965_set_index_string_functions.sh index 8892fb11752..0f29c3dd2f2 100755 --- a/tests/queries/0_stateless/00965_set_index_string_functions.sh +++ b/tests/queries/0_stateless/00965_set_index_string_functions.sh @@ -6,7 +6,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) $CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS set_idx;" -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" CREATE TABLE set_idx ( k UInt64, diff --git a/tests/queries/0_stateless/00965_shard_unresolvable_addresses.sql b/tests/queries/0_stateless/00965_shard_unresolvable_addresses.sql index 16b62c37d80..f2afb974a06 100644 --- a/tests/queries/0_stateless/00965_shard_unresolvable_addresses.sql +++ b/tests/queries/0_stateless/00965_shard_unresolvable_addresses.sql @@ -1,9 +1,10 @@ --- Tags: shard, no-parallel +-- Tags: shard, no-fasttest +-- no-fasttest: Slow timeouts SET prefer_localhost_replica = 1; +SET connections_with_failover_max_tries=1; +SET connect_timeout_with_failover_ms=2000; +SET connect_timeout_with_failover_secure_ms=2000; SELECT count() FROM remote('127.0.0.1,localhos', system.one); -- { serverError ALL_CONNECTION_TRIES_FAILED } SELECT count() FROM remote('127.0.0.1|localhos', system.one); - --- Clear cache to avoid future errors in the logs -SYSTEM DROP DNS CACHE diff --git a/tests/queries/0_stateless/00974_primary_key_for_lowCardinality.sh b/tests/queries/0_stateless/00974_primary_key_for_lowCardinality.sh index 389d433c7e2..ba260042f47 100755 --- a/tests/queries/0_stateless/00974_primary_key_for_lowCardinality.sh +++ b/tests/queries/0_stateless/00974_primary_key_for_lowCardinality.sh @@ -8,7 +8,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) $CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS lowString;" $CLICKHOUSE_CLIENT --query="DROP TABLE IF EXISTS string;" -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" create table lowString ( a LowCardinality(String), @@ -18,7 +18,7 @@ ENGINE = MergeTree() PARTITION BY toYYYYMM(b) ORDER BY (a)" -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" create table string ( a String, diff --git a/tests/queries/0_stateless/00974_query_profiler.sql b/tests/queries/0_stateless/00974_query_profiler.sql index 24e4241b813..cd2f65eb94a 100644 --- a/tests/queries/0_stateless/00974_query_profiler.sql +++ b/tests/queries/0_stateless/00974_query_profiler.sql @@ -2,7 +2,9 @@ -- Tag no-fasttest: Not sure why fail even in sequential mode. Disabled for now to make some progress. SET allow_introspection_functions = 1; +SET trace_profile_events = 0; -- This can inhibit profiler from working, because it prevents sending samples from different profilers concurrently. +SET query_profiler_cpu_time_period_ns = 0; SET query_profiler_real_time_period_ns = 100000000; SET log_queries = 1; SELECT sleep(0.5), ignore('test real time query profiler'); diff --git a/tests/queries/0_stateless/00990_hasToken.sh b/tests/queries/0_stateless/00990_hasToken.sh index 6a1d4ff5ccf..d79472aa5a5 100755 --- a/tests/queries/0_stateless/00990_hasToken.sh +++ b/tests/queries/0_stateless/00990_hasToken.sh @@ -6,4 +6,4 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # We should have correct env vars from shell_config.sh to run this test -python3 "$CURDIR"/00990_hasToken.python | ${CLICKHOUSE_CLIENT} --max_query_size 1048576 -nm +python3 "$CURDIR"/00990_hasToken.python | ${CLICKHOUSE_CLIENT} --max_query_size 1048576 -m diff --git a/tests/queries/0_stateless/01000_subquery_requires_alias.sql b/tests/queries/0_stateless/01000_subquery_requires_alias.sql index 3cd522a8389..38ba1798dc1 100644 --- a/tests/queries/0_stateless/01000_subquery_requires_alias.sql +++ b/tests/queries/0_stateless/01000_subquery_requires_alias.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET joined_subquery_requires_alias = 1; SELECT * FROM (SELECT 1 as A, 2 as B) X diff --git a/tests/queries/0_stateless/01013_totals_without_aggregation.sql b/tests/queries/0_stateless/01013_totals_without_aggregation.sql index ab656cd92b5..08be45754bb 100644 --- a/tests/queries/0_stateless/01013_totals_without_aggregation.sql +++ b/tests/queries/0_stateless/01013_totals_without_aggregation.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT 11 AS n GROUP BY n WITH TOTALS; SELECT 12 AS n GROUP BY n WITH ROLLUP; diff --git a/tests/queries/0_stateless/01018_ambiguous_column.sql b/tests/queries/0_stateless/01018_ambiguous_column.sql index e9e754ed7a8..b2e0e8fc522 100644 --- a/tests/queries/0_stateless/01018_ambiguous_column.sql +++ b/tests/queries/0_stateless/01018_ambiguous_column.sql @@ -1,5 +1,5 @@ SET output_format_pretty_color=1; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; select * from system.one cross join system.one; select * from system.one cross join system.one r; diff --git a/tests/queries/0_stateless/01030_storage_url_syntax.sql b/tests/queries/0_stateless/01030_storage_url_syntax.sql index 0eb89af8462..084486b61ee 100644 --- a/tests/queries/0_stateless/01030_storage_url_syntax.sql +++ b/tests/queries/0_stateless/01030_storage_url_syntax.sql @@ -1,3 +1,6 @@ +-- Tags: no-fasttest +-- no-fasttest: Timeout for the first query (CANNOT_DETECT_FORMAT) is too slow: https://github.com/ClickHouse/ClickHouse/issues/67939 + drop table if exists test_table_url_syntax ; create table test_table_url_syntax (id UInt32) ENGINE = URL('') diff --git a/tests/queries/0_stateless/01047_window_view_parser_inner_table.sql b/tests/queries/0_stateless/01047_window_view_parser_inner_table.sql index f17f3ac63b0..de436f1b28e 100644 --- a/tests/queries/0_stateless/01047_window_view_parser_inner_table.sql +++ b/tests/queries/0_stateless/01047_window_view_parser_inner_table.sql @@ -1,5 +1,5 @@ SET send_logs_level = 'fatal'; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SET allow_experimental_window_view = 1; DROP DATABASE IF EXISTS {CLICKHOUSE_DATABASE:Identifier}; set allow_deprecated_database_ordinary=1; diff --git a/tests/queries/0_stateless/01048_window_view_parser.sql b/tests/queries/0_stateless/01048_window_view_parser.sql index adcb4a6364d..bcd80e37c00 100644 --- a/tests/queries/0_stateless/01048_window_view_parser.sql +++ b/tests/queries/0_stateless/01048_window_view_parser.sql @@ -1,7 +1,7 @@ -- Tags: no-parallel SET send_logs_level = 'fatal'; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SET allow_experimental_window_view = 1; DROP DATABASE IF EXISTS test_01048; set allow_deprecated_database_ordinary=1; diff --git a/tests/queries/0_stateless/01049_join_low_card_bug_long.reference.j2 b/tests/queries/0_stateless/01049_join_low_card_bug_long.reference.j2 index 2ebe5c373b2..341e77b1d78 100644 --- a/tests/queries/0_stateless/01049_join_low_card_bug_long.reference.j2 +++ b/tests/queries/0_stateless/01049_join_low_card_bug_long.reference.j2 @@ -1,5 +1,5 @@ -- { echoOn } -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; {% for join_algorithm in ['default', 'partial_merge', 'parallel_hash'] -%} SET join_algorithm = '{{ join_algorithm }}'; SET join_use_nulls = 0; @@ -313,7 +313,7 @@ SELECT r.lc, materialize(r.lc), toTypeName(r.lc), toTypeName(materialize(r.lc)) str str LowCardinality(Nullable(String)) LowCardinality(Nullable(String)) str_r str_r LowCardinality(Nullable(String)) LowCardinality(Nullable(String)) \N \N LowCardinality(Nullable(String)) LowCardinality(Nullable(String)) -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; {% for join_algorithm in ['default', 'partial_merge', 'parallel_hash'] -%} SET join_algorithm = '{{ join_algorithm }}'; SET join_use_nulls = 0; diff --git a/tests/queries/0_stateless/01049_join_low_card_bug_long.sql.j2 b/tests/queries/0_stateless/01049_join_low_card_bug_long.sql.j2 index 64ec34ef1bf..7e7b5cb1fed 100644 --- a/tests/queries/0_stateless/01049_join_low_card_bug_long.sql.j2 +++ b/tests/queries/0_stateless/01049_join_low_card_bug_long.sql.j2 @@ -23,9 +23,9 @@ INSERT INTO nl VALUES (0, 'str'), (2, 'str_l'); INSERT INTO l_lc VALUES (0, 'str'), (2, 'str_l'); -- { echoOn } -{% for allow_experimental_analyzer in [0, 1] -%} +{% for enable_analyzer in [0, 1] -%} -SET allow_experimental_analyzer = {{ allow_experimental_analyzer }}; +SET enable_analyzer = {{ enable_analyzer }}; {% for join_algorithm in ['default', 'partial_merge', 'parallel_hash'] -%} SET join_algorithm = '{{ join_algorithm }}'; diff --git a/tests/queries/0_stateless/01049_window_view_window_functions.reference b/tests/queries/0_stateless/01049_window_view_window_functions.reference index 2d49664b280..073301104d2 100644 --- a/tests/queries/0_stateless/01049_window_view_window_functions.reference +++ b/tests/queries/0_stateless/01049_window_view_window_functions.reference @@ -67,3 +67,7 @@ SELECT toDateTime(hopEnd(toDateTime('2020-01-09 12:00:01', 'US/Samoa'), INTERVAL 2020-01-10 00:00:00 SELECT hopEnd(hop(toDateTime('2019-01-09 12:00:01', 'US/Samoa'), INTERVAL '1' DAY, INTERVAL '3' DAY, 'US/Samoa')); 2019-01-10 00:00:00 +SELECT hopStart(tuple()); -- { serverError ILLEGAL_COLUMN } +SELECT hopEnd(tuple()); -- { serverError ILLEGAL_COLUMN } +SELECT tumbleStart(tuple()); -- { serverError ILLEGAL_COLUMN } +SELECT tumbleEnd(tuple()); -- { serverError ILLEGAL_COLUMN } diff --git a/tests/queries/0_stateless/01049_window_view_window_functions.sql b/tests/queries/0_stateless/01049_window_view_window_functions.sql index 617019bd2c6..fb2b4b4949a 100644 --- a/tests/queries/0_stateless/01049_window_view_window_functions.sql +++ b/tests/queries/0_stateless/01049_window_view_window_functions.sql @@ -36,3 +36,8 @@ SELECT hopEnd(toDateTime('2020-01-09 12:00:01', 'US/Samoa'), INTERVAL '1' DAY, I SELECT toDateTime(hopEnd(toDateTime('2020-01-09 12:00:01', 'US/Samoa'), INTERVAL '1' DAY, INTERVAL '3' DAY, 'US/Samoa'), 'US/Samoa'); SELECT toDateTime(hopEnd(toDateTime('2020-01-09 12:00:01', 'US/Samoa'), INTERVAL '1' DAY, INTERVAL '3' DAY, 'US/Samoa'), 'US/Samoa'); SELECT hopEnd(hop(toDateTime('2019-01-09 12:00:01', 'US/Samoa'), INTERVAL '1' DAY, INTERVAL '3' DAY, 'US/Samoa')); + +SELECT hopStart(tuple()); -- { serverError ILLEGAL_COLUMN } +SELECT hopEnd(tuple()); -- { serverError ILLEGAL_COLUMN } +SELECT tumbleStart(tuple()); -- { serverError ILLEGAL_COLUMN } +SELECT tumbleEnd(tuple()); -- { serverError ILLEGAL_COLUMN } diff --git a/tests/queries/0_stateless/01050_window_view_parser_tumble.sql b/tests/queries/0_stateless/01050_window_view_parser_tumble.sql index c52a6fefacb..9c4d312bab6 100644 --- a/tests/queries/0_stateless/01050_window_view_parser_tumble.sql +++ b/tests/queries/0_stateless/01050_window_view_parser_tumble.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SET allow_experimental_window_view = 1; DROP TABLE IF EXISTS mt; diff --git a/tests/queries/0_stateless/01051_window_view_parser_hop.sql b/tests/queries/0_stateless/01051_window_view_parser_hop.sql index b37e4ed3095..569e3b2a6bf 100644 --- a/tests/queries/0_stateless/01051_window_view_parser_hop.sql +++ b/tests/queries/0_stateless/01051_window_view_parser_hop.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SET allow_experimental_window_view = 1; DROP TABLE IF EXISTS mt; 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 4f8482558c8..c473bf766b0 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 @@ -6,7 +6,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh opts=( - "--allow_experimental_analyzer=0" + "--enable_analyzer=0" ) $CLICKHOUSE_CLIENT "${opts[@]}" < /dev/null } -function run_until_deadline_and_at_least_times() +function run_with_time_and_iterations_limits() { set -e - local deadline=$1; shift + local min_time=$1; shift + local max_time=$1; shift local min_iterations=$1; shift + local max_iterations=$1; shift local function_to_run=$1; shift + # if [ "${1:-X}" = "1" ]; then + # set -x + # fi + local started_time - started_time=$(get_now) - local i=0 + started_time=$SECONDS + local iteration=0 while true do - $function_to_run $i "$@" + $function_to_run $iteration "$@" - [[ $(get_now) -lt $deadline ]] || break + [[ $SECONDS -lt $max_time ]] || break + [[ $SECONDS -lt $min_time ]] || [[ $iteration -lt $max_iterations ]] || break - i=$(($i + 1)) + iteration=$(($iteration + 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 + [[ $iteration -gt $min_iterations ]] || echo "$iteration/$min_iterations : not enough iterations of $function_to_run has been made from $started_time until $max_time" >&2 + set +x } function insert_commit_action() @@ -155,42 +157,44 @@ function select_insert_action() ROLLBACK;" } -MAIN_TIME_PART=400 -SECOND_TIME_PART=30 +MIN_SECONDS=5 +MAX_SECONDS=400 WAIT_FINISH=60 -LAST_TIME_GAP=10 -if [[ $((MAIN_TIME_PART + SECOND_TIME_PART + WAIT_FINISH + LAST_TIME_GAP)) -ge 600 ]]; then +if [[ $((MAX_SECONDS + WAIT_FINISH)) -ge 550 ]]; then echo "time sttings are wrong" 2>&1 exit 1 fi -START_TIME=$(get_now) -STOP_TIME=$((START_TIME + MAIN_TIME_PART)) -SECOND_STOP_TIME=$((STOP_TIME + SECOND_TIME_PART)) -MIN_ITERATIONS=20 +START_TIME=$SECONDS +MIN_TIME=$((START_TIME + MIN_SECONDS)) +MAX_TIME=$((START_TIME + MAX_SECONDS)) +MIN_ITERATIONS=15 +MAX_ITERATIONS=200 -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_with_time_and_iterations_limits $MIN_TIME $MAX_TIME $MIN_ITERATIONS $MAX_ITERATIONS insert_commit_action 1 & PID_1=$! +run_with_time_and_iterations_limits $MIN_TIME $MAX_TIME $MIN_ITERATIONS $MAX_ITERATIONS insert_commit_action 2 & PID_2=$! +run_with_time_and_iterations_limits $MIN_TIME $MAX_TIME $MIN_ITERATIONS $MAX_ITERATIONS insert_rollback_action 3 & PID_3=$! -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 $MIN_ITERATIONS select_insert_action & PID_6=$! +run_with_time_and_iterations_limits $MIN_TIME $MAX_TIME $MIN_ITERATIONS $MAX_ITERATIONS optimize_action & PID_4=$! +run_with_time_and_iterations_limits $MIN_TIME $MAX_TIME $MIN_ITERATIONS $MAX_ITERATIONS select_action & PID_5=$! +run_with_time_and_iterations_limits $MIN_TIME $MAX_TIME $MIN_ITERATIONS $MAX_ITERATIONS select_insert_action & PID_6=$! 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 +run_with_time_and_iterations_limits $MIN_TIME $MAX_TIME $MIN_ITERATIONS $MAX_ITERATIONS select_action & PID_7=$! +run_with_time_and_iterations_limits $MIN_TIME $MAX_TIME $MIN_ITERATIONS $MAX_ITERATIONS select_insert_action & PID_8=$! +is_pid_exist $PID_1 || echo "insert_commit_action is not running" 2>&1 +is_pid_exist $PID_2 || echo "second insert_commit_action is not running" 2>&1 +is_pid_exist $PID_3 || echo "insert_rollback_action is not running" 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_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 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 diff --git a/tests/queries/0_stateless/01172_transaction_counters.reference b/tests/queries/0_stateless/01172_transaction_counters.reference index 24083d7d40b..0fd73c7bcec 100644 --- a/tests/queries/0_stateless/01172_transaction_counters.reference +++ b/tests/queries/0_stateless/01172_transaction_counters.reference @@ -16,25 +16,25 @@ 7 all_3_3_0 (0,0,'00000000-0000-0000-0000-000000000000') 0 7 all_4_4_0 (0,0,'00000000-0000-0000-0000-000000000000') 0 8 1 -1 1 AddPart 1 1 1 1 all_1_1_0 -2 1 Begin 1 1 1 1 -2 1 AddPart 1 1 1 1 all_2_2_0 -2 1 Rollback 1 1 1 1 -3 1 Begin 1 1 1 1 -3 1 AddPart 1 1 1 1 all_3_3_0 -3 1 Commit 1 1 1 0 -1 1 LockPart 1 1 1 1 all_2_2_0 -4 1 Begin 1 1 1 1 -4 1 AddPart 1 1 1 1 all_4_4_0 -4 1 Commit 1 1 1 0 -5 1 Begin 1 1 1 1 -5 1 AddPart 1 1 1 1 all_5_5_0 -5 1 LockPart 1 1 1 1 all_1_1_0 -5 1 LockPart 1 1 1 1 all_3_3_0 -5 1 LockPart 1 1 1 1 all_4_4_0 -5 1 LockPart 1 1 1 1 all_5_5_0 -5 1 UnlockPart 1 1 1 1 all_1_1_0 -5 1 UnlockPart 1 1 1 1 all_3_3_0 -5 1 UnlockPart 1 1 1 1 all_4_4_0 -5 1 UnlockPart 1 1 1 1 all_5_5_0 -5 1 Rollback 1 1 1 1 +1 AddPart 1 1 1 1 all_1_1_0 +2 Begin 1 1 1 1 +2 AddPart 1 1 1 1 all_2_2_0 +2 Rollback 1 1 1 1 +3 Begin 1 1 1 1 +3 AddPart 1 1 1 1 all_3_3_0 +3 Commit 1 1 1 0 +1 LockPart 1 1 1 1 all_2_2_0 +4 Begin 1 1 1 1 +4 AddPart 1 1 1 1 all_4_4_0 +4 Commit 1 1 1 0 +5 Begin 1 1 1 1 +5 AddPart 1 1 1 1 all_5_5_0 +5 LockPart 1 1 1 1 all_1_1_0 +5 LockPart 1 1 1 1 all_3_3_0 +5 LockPart 1 1 1 1 all_4_4_0 +5 LockPart 1 1 1 1 all_5_5_0 +5 UnlockPart 1 1 1 1 all_1_1_0 +5 UnlockPart 1 1 1 1 all_3_3_0 +5 UnlockPart 1 1 1 1 all_4_4_0 +5 UnlockPart 1 1 1 1 all_5_5_0 +5 Rollback 1 1 1 1 diff --git a/tests/queries/0_stateless/01172_transaction_counters.sql b/tests/queries/0_stateless/01172_transaction_counters.sql index a809e4196e9..581b45cd15c 100644 --- a/tests/queries/0_stateless/01172_transaction_counters.sql +++ b/tests/queries/0_stateless/01172_transaction_counters.sql @@ -42,7 +42,6 @@ rollback; system flush logs; select indexOf((select arraySort(groupUniqArray(tid)) from system.transactions_info_log where database=currentDatabase() and table='txn_counters'), tid), - (toDecimal64(now64(6), 6) - toDecimal64(event_time, 6)) < 100, type, thread_id!=0, length(query_id)=length(queryID()) or type='Commit' and query_id='', -- ignore fault injection after commit diff --git a/tests/queries/0_stateless/01232_untuple.sql b/tests/queries/0_stateless/01232_untuple.sql index 391d08ab859..e2049e0a453 100644 --- a/tests/queries/0_stateless/01232_untuple.sql +++ b/tests/queries/0_stateless/01232_untuple.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET enable_named_columns_in_function_tuple = 1; select untuple((* except (b),)) 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 index aea91a0bf6b..3c7b9038e1f 100755 --- a/tests/queries/0_stateless/01246_buffer_flush.sh +++ b/tests/queries/0_stateless/01246_buffer_flush.sh @@ -27,7 +27,7 @@ function wait_until() function get_buffer_delay() { local buffer_insert_id=$1 && shift - query "SYSTEM FLUSH LOGS" + $CLICKHOUSE_CLIENT -q "SYSTEM FLUSH LOGS" query " WITH (SELECT event_time_microseconds FROM system.query_log WHERE current_database = '$CLICKHOUSE_DATABASE' AND type = 'QueryStart' AND query_id = '$buffer_insert_id') AS begin_, diff --git a/tests/queries/0_stateless/01300_group_by_other_keys_having.reference b/tests/queries/0_stateless/01300_group_by_other_keys_having.reference index f861da3da2b..23028e3fb6e 100644 --- a/tests/queries/0_stateless/01300_group_by_other_keys_having.reference +++ b/tests/queries/0_stateless/01300_group_by_other_keys_having.reference @@ -3,7 +3,7 @@ SELECT round(avg(log(2) * number), 6) AS k FROM numbers(10000000) GROUP BY (numb 3465735.9028 3465735.9028 3465736.595947 -SELECT round(avg(log(2) * number), 6) AS k FROM numbers(10000000) GROUP BY (number % 2) * (number % 3), number % 3, number % 2 HAVING avg(log(2) * number) > 3465735.3 ORDER BY k SETTINGS allow_experimental_analyzer=1; +SELECT round(avg(log(2) * number), 6) AS k FROM numbers(10000000) GROUP BY (number % 2) * (number % 3), number % 3, number % 2 HAVING avg(log(2) * number) > 3465735.3 ORDER BY k SETTINGS enable_analyzer=1; 3465735.9028 3465735.9028 3465736.595947 @@ -11,7 +11,7 @@ SELECT round(avg(log(2) * number), 6) AS k FROM numbers(10000000) GROUP BY numbe 3465734.169932 3465734.863079 3465735.556226 -SELECT round(avg(log(2) * number), 6) AS k FROM numbers(10000000) GROUP BY number % 5, ((number % 5) * (number % 5)) HAVING ((number % 5) * (number % 5)) < 5 ORDER BY k SETTINGS allow_experimental_analyzer=1; +SELECT round(avg(log(2) * number), 6) AS k FROM numbers(10000000) GROUP BY number % 5, ((number % 5) * (number % 5)) HAVING ((number % 5) * (number % 5)) < 5 ORDER BY k SETTINGS enable_analyzer=1; 3465734.169932 3465734.863079 3465735.556226 @@ -19,7 +19,7 @@ SELECT (number % 5) * (number % 5) AS k FROM numbers(10000000) GROUP BY number % 0 1 4 -SELECT (number % 5) * (number % 5) AS k FROM numbers(10000000) GROUP BY number % 5, ((number % 5) * (number % 5)) HAVING ((number % 5) * (number % 5)) < 5 ORDER BY k SETTINGS allow_experimental_analyzer=1; +SELECT (number % 5) * (number % 5) AS k FROM numbers(10000000) GROUP BY number % 5, ((number % 5) * (number % 5)) HAVING ((number % 5) * (number % 5)) < 5 ORDER BY k SETTINGS enable_analyzer=1; 0 1 4 diff --git a/tests/queries/0_stateless/01300_group_by_other_keys_having.sql b/tests/queries/0_stateless/01300_group_by_other_keys_having.sql index 203e8322ad9..a73b122a9f9 100644 --- a/tests/queries/0_stateless/01300_group_by_other_keys_having.sql +++ b/tests/queries/0_stateless/01300_group_by_other_keys_having.sql @@ -1,16 +1,16 @@ set optimize_group_by_function_keys = 1; set optimize_syntax_fuse_functions = 0; -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; -- { echoOn } SELECT round(avg(log(2) * number), 6) AS k FROM numbers(10000000) GROUP BY (number % 2) * (number % 3), number % 3, number % 2 HAVING avg(log(2) * number) > 3465735.3 ORDER BY k; -SELECT round(avg(log(2) * number), 6) AS k FROM numbers(10000000) GROUP BY (number % 2) * (number % 3), number % 3, number % 2 HAVING avg(log(2) * number) > 3465735.3 ORDER BY k SETTINGS allow_experimental_analyzer=1; +SELECT round(avg(log(2) * number), 6) AS k FROM numbers(10000000) GROUP BY (number % 2) * (number % 3), number % 3, number % 2 HAVING avg(log(2) * number) > 3465735.3 ORDER BY k SETTINGS enable_analyzer=1; SELECT round(avg(log(2) * number), 6) AS k FROM numbers(10000000) GROUP BY number % 5, ((number % 5) * (number % 5)) HAVING ((number % 5) * (number % 5)) < 5 ORDER BY k; -SELECT round(avg(log(2) * number), 6) AS k FROM numbers(10000000) GROUP BY number % 5, ((number % 5) * (number % 5)) HAVING ((number % 5) * (number % 5)) < 5 ORDER BY k SETTINGS allow_experimental_analyzer=1; +SELECT round(avg(log(2) * number), 6) AS k FROM numbers(10000000) GROUP BY number % 5, ((number % 5) * (number % 5)) HAVING ((number % 5) * (number % 5)) < 5 ORDER BY k SETTINGS enable_analyzer=1; SELECT (number % 5) * (number % 5) AS k FROM numbers(10000000) GROUP BY number % 5, ((number % 5) * (number % 5)) HAVING ((number % 5) * (number % 5)) < 5 ORDER BY k; -SELECT (number % 5) * (number % 5) AS k FROM numbers(10000000) GROUP BY number % 5, ((number % 5) * (number % 5)) HAVING ((number % 5) * (number % 5)) < 5 ORDER BY k SETTINGS allow_experimental_analyzer=1; +SELECT (number % 5) * (number % 5) AS k FROM numbers(10000000) GROUP BY number % 5, ((number % 5) * (number % 5)) HAVING ((number % 5) * (number % 5)) < 5 ORDER BY k SETTINGS enable_analyzer=1; -- { echoOff } diff --git a/tests/queries/0_stateless/01323_redundant_functions_in_order_by.reference b/tests/queries/0_stateless/01323_redundant_functions_in_order_by.reference index d47f12ff4d1..c2c37cc4de6 100644 --- a/tests/queries/0_stateless/01323_redundant_functions_in_order_by.reference +++ b/tests/queries/0_stateless/01323_redundant_functions_in_order_by.reference @@ -65,7 +65,7 @@ QUERY id: 0 SORT id: 12, sort_direction: ASCENDING, with_fill: 0 EXPRESSION COLUMN id: 7, column_name: number, result_type: UInt64, source_id: 8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT groupArray(x) FROM ( @@ -99,7 +99,7 @@ QUERY id: 0 SORT id: 12, sort_direction: ASCENDING, with_fill: 0 EXPRESSION COLUMN id: 7, column_name: number, result_type: UInt64, source_id: 8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT groupArray(x) FROM ( @@ -141,7 +141,7 @@ QUERY id: 0 SORT id: 15, sort_direction: ASCENDING, with_fill: 0 EXPRESSION COLUMN id: 7, column_name: number, result_type: UInt64, source_id: 8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT key, a, @@ -203,7 +203,7 @@ QUERY id: 0 SORT id: 25, sort_direction: ASCENDING, with_fill: 0 EXPRESSION COLUMN id: 26, column_name: key, result_type: UInt64, source_id: 5 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT key, a @@ -229,7 +229,7 @@ QUERY id: 0 SORT id: 7, sort_direction: ASCENDING, with_fill: 0 EXPRESSION COLUMN id: 4, column_name: a, result_type: UInt8, source_id: 3 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT key, a @@ -262,7 +262,7 @@ QUERY id: 0 LIST id: 11, nodes: 2 COLUMN id: 2, column_name: key, result_type: UInt64, source_id: 3 COLUMN id: 4, column_name: a, result_type: UInt8, source_id: 3 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 QUERY id: 0 PROJECTION COLUMNS key UInt64 @@ -285,7 +285,7 @@ QUERY id: 0 SORT id: 10, sort_direction: ASCENDING, with_fill: 0 EXPRESSION COLUMN id: 2, column_name: key, result_type: UInt64, source_id: 3 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 QUERY id: 0 PROJECTION COLUMNS t1.id UInt64 @@ -314,7 +314,7 @@ QUERY id: 0 SORT id: 14, sort_direction: ASCENDING, with_fill: 0 EXPRESSION COLUMN id: 15, column_name: id, result_type: UInt64, source_id: 5 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 [0,1,2] [0,1,2] [0,1,2] diff --git a/tests/queries/0_stateless/01323_redundant_functions_in_order_by.sql b/tests/queries/0_stateless/01323_redundant_functions_in_order_by.sql index 738ad581e3d..fb1eed1666e 100644 --- a/tests/queries/0_stateless/01323_redundant_functions_in_order_by.sql +++ b/tests/queries/0_stateless/01323_redundant_functions_in_order_by.sql @@ -8,37 +8,37 @@ INSERT INTO test SELECT number, number, toString(number), number from numbers(4) set optimize_redundant_functions_in_order_by = 1; SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY x, exp(x)); -SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY x, exp(x)) SETTINGS allow_experimental_analyzer=1; +SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY x, exp(x)) SETTINGS enable_analyzer=1; SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY x, exp(exp(x))); -SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY x, exp(exp(x))) SETTINGS allow_experimental_analyzer=1; +SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY x, exp(exp(x))) SETTINGS enable_analyzer=1; SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY exp(x), x); -SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY exp(x), x) SETTINGS allow_experimental_analyzer=1; +SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY exp(x), x) SETTINGS enable_analyzer=1; SELECT * FROM (SELECT number + 2 AS key FROM numbers(4)) s FULL JOIN test t USING(key) ORDER BY s.key, t.key; -SELECT * FROM (SELECT number + 2 AS key FROM numbers(4)) s FULL JOIN test t USING(key) ORDER BY s.key, t.key SETTINGS allow_experimental_analyzer=1; +SELECT * FROM (SELECT number + 2 AS key FROM numbers(4)) s FULL JOIN test t USING(key) ORDER BY s.key, t.key SETTINGS enable_analyzer=1; SELECT key, a FROM test ORDER BY key, a, exp(key + a); -SELECT key, a FROM test ORDER BY key, a, exp(key + a) SETTINGS allow_experimental_analyzer=1; +SELECT key, a FROM test ORDER BY key, a, exp(key + a) SETTINGS enable_analyzer=1; SELECT key, a FROM test ORDER BY key, exp(key + a); -SELECT key, a FROM test ORDER BY key, exp(key + a) SETTINGS allow_experimental_analyzer=1; +SELECT key, a FROM test ORDER BY key, exp(key + a) SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY x, exp(x)); -EXPLAIN QUERY TREE run_passes=1 SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY x, exp(x)) settings allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY x, exp(x)) settings enable_analyzer=1; EXPLAIN SYNTAX SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY x, exp(exp(x))); -EXPLAIN QUERY TREE run_passes=1 SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY x, exp(exp(x))) settings allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY x, exp(exp(x))) settings enable_analyzer=1; EXPLAIN SYNTAX SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY exp(x), x); -EXPLAIN QUERY TREE run_passes=1 SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY exp(x), x) settings allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT groupArray(x) from (SELECT number as x FROM numbers(3) ORDER BY exp(x), x) settings enable_analyzer=1; EXPLAIN SYNTAX SELECT * FROM (SELECT number + 2 AS key FROM numbers(4)) s FULL JOIN test t USING(key) ORDER BY s.key, t.key; -EXPLAIN QUERY TREE run_passes=1 SELECT * FROM (SELECT number + 2 AS key FROM numbers(4)) s FULL JOIN test t USING(key) ORDER BY s.key, t.key settings allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT * FROM (SELECT number + 2 AS key FROM numbers(4)) s FULL JOIN test t USING(key) ORDER BY s.key, t.key settings enable_analyzer=1; EXPLAIN SYNTAX SELECT key, a FROM test ORDER BY key, a, exp(key + a); -EXPLAIN QUERY TREE run_passes=1 SELECT key, a FROM test ORDER BY key, a, exp(key + a) settings allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT key, a FROM test ORDER BY key, a, exp(key + a) settings enable_analyzer=1; EXPLAIN SYNTAX SELECT key, a FROM test ORDER BY key, exp(key + a); -EXPLAIN QUERY TREE run_passes=1 SELECT key, a FROM test ORDER BY key, exp(key + a) settings allow_experimental_analyzer=1; -EXPLAIN QUERY TREE run_passes=1 SELECT key FROM test GROUP BY key ORDER BY avg(a), key settings allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT key, a FROM test ORDER BY key, exp(key + a) settings enable_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT key FROM test GROUP BY key ORDER BY avg(a), key settings enable_analyzer=1; DROP TABLE IF EXISTS t1; DROP TABLE IF EXISTS t2; CREATE TABLE t1 (id UInt64) ENGINE = MergeTree() ORDER BY id; CREATE TABLE t2 (id UInt64) ENGINE = MergeTree() ORDER BY id; -EXPLAIN QUERY TREE run_passes=1 SELECT * FROM t1 INNER JOIN t2 ON t1.id = t2.id ORDER BY t1.id, t2.id settings allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT * FROM t1 INNER JOIN t2 ON t1.id = t2.id ORDER BY t1.id, t2.id settings enable_analyzer=1; set optimize_redundant_functions_in_order_by = 0; diff --git a/tests/queries/0_stateless/01353_low_cardinality_join_types.sql b/tests/queries/0_stateless/01353_low_cardinality_join_types.sql index 93953f1d74a..2aa42f33fd3 100644 --- a/tests/queries/0_stateless/01353_low_cardinality_join_types.sql +++ b/tests/queries/0_stateless/01353_low_cardinality_join_types.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; set join_algorithm = 'hash'; select '-'; @@ -75,7 +75,7 @@ from (select toLowCardinality(number) k, toLowCardinality(toString(number)) s fr full join (select toLowCardinality(number+1) k, toLowCardinality(toString(number+1)) s from numbers(2)) as js2 using k order by js1.k, js2.k; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; set join_algorithm = 'hash'; select '-'; diff --git a/tests/queries/0_stateless/01376_GROUP_BY_injective_elimination_dictGet.reference b/tests/queries/0_stateless/01376_GROUP_BY_injective_elimination_dictGet.reference index 6de0a5be0a5..2d16e71f099 100644 --- a/tests/queries/0_stateless/01376_GROUP_BY_injective_elimination_dictGet.reference +++ b/tests/queries/0_stateless/01376_GROUP_BY_injective_elimination_dictGet.reference @@ -21,4 +21,4 @@ QUERY id: 0 GROUP BY LIST id: 10, nodes: 1 COLUMN id: 6, column_name: number, result_type: UInt64, source_id: 7 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 diff --git a/tests/queries/0_stateless/01376_GROUP_BY_injective_elimination_dictGet.sql b/tests/queries/0_stateless/01376_GROUP_BY_injective_elimination_dictGet.sql index a868b38b4d7..08ca9ed3c2d 100644 --- a/tests/queries/0_stateless/01376_GROUP_BY_injective_elimination_dictGet.sql +++ b/tests/queries/0_stateless/01376_GROUP_BY_injective_elimination_dictGet.sql @@ -38,7 +38,7 @@ EXPLAIN QUERY TREE SELECT dictGet('dictdb_01376.dict_exists', 'value', number) as val FROM numbers(2) GROUP BY val -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; DROP DICTIONARY dictdb_01376.dict_exists; DROP TABLE dictdb_01376.table_for_dict; diff --git a/tests/queries/0_stateless/01414_mutations_and_errors_zookeeper.sh b/tests/queries/0_stateless/01414_mutations_and_errors_zookeeper.sh index 31e2cc395aa..5cdd6057050 100755 --- a/tests/queries/0_stateless/01414_mutations_and_errors_zookeeper.sh +++ b/tests/queries/0_stateless/01414_mutations_and_errors_zookeeper.sh @@ -1,5 +1,6 @@ #!/usr/bin/env bash -# Tags: zookeeper, no-parallel +# Tags: zookeeper, no-parallel, no-fasttest +# no-fasttest: Waiting for failed mutations is slow: https://github.com/ClickHouse/ClickHouse/issues/67936 CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/01428_nullable_asof_join.sql b/tests/queries/0_stateless/01428_nullable_asof_join.sql index f07a26edd97..41f6ba8a03d 100644 --- a/tests/queries/0_stateless/01428_nullable_asof_join.sql +++ b/tests/queries/0_stateless/01428_nullable_asof_join.sql @@ -18,13 +18,13 @@ SELECT a.pk, b.pk, a.dt, b.dt, toTypeName(a.pk), toTypeName(b.pk), toTypeName(ma FROM (SELECT toUInt8(number) > 0 as pk, toUInt8(number) as dt FROM numbers(3)) a ASOF LEFT JOIN (SELECT 1 as pk, toNullable(0) as dt) b USING(pk, dt) -ORDER BY a.dt SETTINGS allow_experimental_analyzer = 0; +ORDER BY a.dt SETTINGS enable_analyzer = 0; SELECT a.pk, b.pk, a.dt, b.dt, toTypeName(a.pk), toTypeName(b.pk), toTypeName(materialize(a.dt)), toTypeName(materialize(b.dt)) FROM (SELECT toUInt8(number) > 0 as pk, toUInt8(number) as dt FROM numbers(3)) a ASOF LEFT JOIN (SELECT 1 as pk, toNullable(0) as dt) b USING(pk, dt) -ORDER BY a.dt SETTINGS allow_experimental_analyzer = 1; +ORDER BY a.dt SETTINGS enable_analyzer = 1; SELECT a.pk, b.pk, a.dt, b.dt, toTypeName(a.pk), toTypeName(b.pk), toTypeName(materialize(a.dt)), toTypeName(materialize(b.dt)) FROM (SELECT toUInt8(number) > 0 as pk, toNullable(toUInt8(number)) as dt FROM numbers(3)) a @@ -70,25 +70,25 @@ SELECT a.pk, b.pk, a.dt, b.dt, toTypeName(a.pk), toTypeName(b.pk), toTypeName(ma FROM (SELECT toUInt8(number) > 0 as pk, toNullable(toUInt8(number)) as dt FROM numbers(3)) a ASOF JOIN (SELECT 1 as pk, 2 as dt) b USING(pk, dt) -ORDER BY a.dt SETTINGS allow_experimental_analyzer = 0; +ORDER BY a.dt SETTINGS enable_analyzer = 0; SELECT a.pk, b.pk, a.dt, b.dt, toTypeName(a.pk), toTypeName(b.pk), toTypeName(materialize(a.dt)), toTypeName(materialize(b.dt)) FROM (SELECT toUInt8(number) > 0 as pk, toNullable(toUInt8(number)) as dt FROM numbers(3)) a ASOF JOIN (SELECT 1 as pk, 2 as dt) b USING(pk, dt) -ORDER BY a.dt SETTINGS allow_experimental_analyzer = 1; +ORDER BY a.dt SETTINGS enable_analyzer = 1; SELECT a.pk, b.pk, a.dt, b.dt, toTypeName(a.pk), toTypeName(b.pk), toTypeName(materialize(a.dt)), toTypeName(materialize(b.dt)) FROM (SELECT toUInt8(number) > 0 as pk, toUInt8(number) as dt FROM numbers(3)) a ASOF JOIN (SELECT 1 as pk, toNullable(0) as dt) b USING(pk, dt) -ORDER BY a.dt SETTINGS allow_experimental_analyzer = 0; +ORDER BY a.dt SETTINGS enable_analyzer = 0; SELECT a.pk, b.pk, a.dt, b.dt, toTypeName(a.pk), toTypeName(b.pk), toTypeName(materialize(a.dt)), toTypeName(materialize(b.dt)) FROM (SELECT toUInt8(number) > 0 as pk, toUInt8(number) as dt FROM numbers(3)) a ASOF JOIN (SELECT 1 as pk, toNullable(0) as dt) b USING(pk, dt) -ORDER BY a.dt SETTINGS allow_experimental_analyzer = 1; +ORDER BY a.dt SETTINGS enable_analyzer = 1; SELECT a.pk, b.pk, a.dt, b.dt, toTypeName(a.pk), toTypeName(b.pk), toTypeName(materialize(a.dt)), toTypeName(materialize(b.dt)) FROM (SELECT toUInt8(number) > 0 as pk, toNullable(toUInt8(number)) as dt FROM numbers(3)) a diff --git a/tests/queries/0_stateless/01455_opentelemetry_distributed.reference b/tests/queries/0_stateless/01455_opentelemetry_distributed.reference index 2920b387aa2..2b08db1f27f 100644 --- a/tests/queries/0_stateless/01455_opentelemetry_distributed.reference +++ b/tests/queries/0_stateless/01455_opentelemetry_distributed.reference @@ -1,10 +1,10 @@ ===http=== -{"query":"select 1 from remote('127.0.0.2', system, one) settings allow_experimental_analyzer = 1 format Null\n","status":"QueryFinish","tracestate":"some custom state","sorted_by_start_time":1} +{"query":"select 1 from remote('127.0.0.2', system, one) settings enable_analyzer = 1 format Null\n","status":"QueryFinish","tracestate":"some custom state","sorted_by_start_time":1} {"query":"DESC TABLE system.one","status":"QueryFinish","tracestate":"some custom state","sorted_by_start_time":1} {"query":"SELECT 1 AS `1` FROM `system`.`one` AS `__table1`","status":"QueryFinish","tracestate":"some custom state","sorted_by_start_time":1} {"query":"DESC TABLE system.one","query_status":"QueryFinish","tracestate":"some custom state","sorted_by_finish_time":1} {"query":"SELECT 1 AS `1` FROM `system`.`one` AS `__table1`","query_status":"QueryFinish","tracestate":"some custom state","sorted_by_finish_time":1} -{"query":"select 1 from remote('127.0.0.2', system, one) settings allow_experimental_analyzer = 1 format Null\n","query_status":"QueryFinish","tracestate":"some custom state","sorted_by_finish_time":1} +{"query":"select 1 from remote('127.0.0.2', system, one) settings enable_analyzer = 1 format Null\n","query_status":"QueryFinish","tracestate":"some custom state","sorted_by_finish_time":1} {"total spans":"3","unique spans":"3","unique non-zero parent spans":"3"} {"initial query spans with proper parent":"2"} {"unique non-empty tracestate values":"1"} diff --git a/tests/queries/0_stateless/01455_opentelemetry_distributed.sh b/tests/queries/0_stateless/01455_opentelemetry_distributed.sh index e19810dd7f1..2b6da6132ed 100755 --- a/tests/queries/0_stateless/01455_opentelemetry_distributed.sh +++ b/tests/queries/0_stateless/01455_opentelemetry_distributed.sh @@ -10,7 +10,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) function check_log { ${CLICKHOUSE_CLIENT} --format=JSONEachRow -nq " -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; system flush logs; -- Show queries sorted by start time. @@ -75,7 +75,7 @@ select uniqExact(value) "'"'"unique non-empty tracestate values"'"'" # Generate some random trace id so that the prevous runs of the test do not interfere. echo "===http===" -trace_id=$(${CLICKHOUSE_CLIENT} -q "select lower(hex(reverse(reinterpretAsString(generateUUIDv4())))) settings allow_experimental_analyzer = 1") +trace_id=$(${CLICKHOUSE_CLIENT} -q "select lower(hex(reverse(reinterpretAsString(generateUUIDv4())))) settings enable_analyzer = 1") # Check that the HTTP traceparent is read, and then passed through `remote` # table function. We expect 4 queries -- one initial, one SELECT and two @@ -85,7 +85,7 @@ ${CLICKHOUSE_CURL} \ --header "traceparent: 00-$trace_id-0000000000000073-01" \ --header "tracestate: some custom state" "$CLICKHOUSE_URL" \ --get \ - --data-urlencode "query=select 1 from remote('127.0.0.2', system, one) settings allow_experimental_analyzer = 1 format Null" + --data-urlencode "query=select 1 from remote('127.0.0.2', system, one) settings enable_analyzer = 1 format Null" check_log diff --git a/tests/queries/0_stateless/01456_modify_column_type_via_add_drop_update.sql b/tests/queries/0_stateless/01456_modify_column_type_via_add_drop_update.sql index b7cbfc92c26..a2e4804f12e 100644 --- a/tests/queries/0_stateless/01456_modify_column_type_via_add_drop_update.sql +++ b/tests/queries/0_stateless/01456_modify_column_type_via_add_drop_update.sql @@ -1,5 +1,3 @@ --- Tags: no-parallel - DROP TABLE IF EXISTS tbl; CREATE TABLE tbl(a String, b UInt32, c Float64, d Int64, e UInt8) ENGINE=MergeTree ORDER BY tuple(); INSERT INTO tbl SELECT number, number * 2, number * 3, number * 4, number * 5 FROM system.numbers LIMIT 10; diff --git a/tests/queries/0_stateless/01476_right_full_join_switch.sql b/tests/queries/0_stateless/01476_right_full_join_switch.sql index dfbdec47e1f..7c8c3157844 100644 --- a/tests/queries/0_stateless/01476_right_full_join_switch.sql +++ b/tests/queries/0_stateless/01476_right_full_join_switch.sql @@ -13,7 +13,7 @@ INSERT INTO nr VALUES (2, NULL); SET join_use_nulls = 0; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- t.x is supertupe for `x` from left and right since `x` is inside `USING`. SELECT t.x, l.s, r.s, toTypeName(l.s), toTypeName(r.s) FROM t AS l LEFT JOIN nr AS r USING (x) ORDER BY t.x; @@ -28,7 +28,7 @@ SELECT t.x, l.s, r.s, toTypeName(l.s), toTypeName(r.s) FROM nr AS l FULL JOIN t SELECT '-'; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; -- t.x is supertupe for `x` from left and right since `x` is inside `USING`. SELECT t.x, l.s, r.s, toTypeName(l.s), toTypeName(r.s) FROM t AS l LEFT JOIN nr AS r USING (x) ORDER BY t.x; diff --git a/tests/queries/0_stateless/01477_lc_in_merge_join_left_key.sql.j2 b/tests/queries/0_stateless/01477_lc_in_merge_join_left_key.sql.j2 index 6eafd41b411..b43db222e43 100644 --- a/tests/queries/0_stateless/01477_lc_in_merge_join_left_key.sql.j2 +++ b/tests/queries/0_stateless/01477_lc_in_merge_join_left_key.sql.j2 @@ -10,11 +10,11 @@ CREATE TABLE nr (`x` Nullable(UInt32), `s` Nullable(String)) ENGINE = Memory; INSERT INTO t VALUES (1, 'l'); INSERT INTO nr VALUES (2, NULL); -{% for allow_experimental_analyzer in [0, 1] -%} +{% for enable_analyzer in [0, 1] -%} -SET allow_experimental_analyzer = {{ allow_experimental_analyzer }}; +SET enable_analyzer = {{ enable_analyzer }}; -{% if allow_experimental_analyzer -%} +{% if enable_analyzer -%} SELECT '- analyzer -'; {% endif -%} diff --git a/tests/queries/0_stateless/01479_cross_join_9855.sql b/tests/queries/0_stateless/01479_cross_join_9855.sql index 9dcf209a1cd..19cd0ab18fd 100644 --- a/tests/queries/0_stateless/01479_cross_join_9855.sql +++ b/tests/queries/0_stateless/01479_cross_join_9855.sql @@ -2,8 +2,8 @@ SET cross_to_inner_join_rewrite = 1; SELECT count() FROM numbers(4) AS n1, numbers(3) AS n2 -WHERE n1.number > (select avg(n.number) from numbers(3) n) SETTINGS allow_experimental_analyzer=0; +WHERE n1.number > (select avg(n.number) from numbers(3) n) SETTINGS enable_analyzer=0; SELECT count() FROM numbers(4) AS n1, numbers(3) AS n2, numbers(6) AS n3 -WHERE n1.number > (select avg(n.number) from numbers(3) n) SETTINGS allow_experimental_analyzer=0; +WHERE n1.number > (select avg(n.number) from numbers(3) n) SETTINGS enable_analyzer=0; diff --git a/tests/queries/0_stateless/01508_explain_header.sql b/tests/queries/0_stateless/01508_explain_header.sql index a9f876068aa..03452e4bdac 100644 --- a/tests/queries/0_stateless/01508_explain_header.sql +++ b/tests/queries/0_stateless/01508_explain_header.sql @@ -1,3 +1,3 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; explain header = 1 select 1 as x; diff --git a/tests/queries/0_stateless/01526_max_untracked_memory.sh b/tests/queries/0_stateless/01526_max_untracked_memory.sh index b2bad637422..173f6fbe2e6 100755 --- a/tests/queries/0_stateless/01526_max_untracked_memory.sh +++ b/tests/queries/0_stateless/01526_max_untracked_memory.sh @@ -14,7 +14,7 @@ min_trace_entries=2 # do not use _, they should be escaped for LIKE query_id_tcp_prefix="01526-tcp-memory-tracking-$RANDOM-$$" -${CLICKHOUSE_CLIENT} --log_queries=1 --max_threads=1 --max_untracked_memory=0 --memory_profiler_sample_probability=1 -q "with '$query_id_tcp_prefix' as __id $query FORMAT Null" +${CLICKHOUSE_CLIENT} --log_queries=1 --max_threads=1 --max_untracked_memory=0 --memory_profiler_sample_probability=1 --trace_profile_events 0 -q "with '$query_id_tcp_prefix' as __id $query FORMAT Null" ${CLICKHOUSE_CLIENT} -q "SYSTEM FLUSH LOGS" query_id_tcp="$(${CLICKHOUSE_CLIENT} -q "SELECT DISTINCT query_id FROM system.query_log WHERE current_database = currentDatabase() AND query LIKE '%$query_id_tcp_prefix%'")" ${CLICKHOUSE_CLIENT} -q "SELECT count()>=$min_trace_entries FROM system.trace_log WHERE query_id = '$query_id_tcp' AND abs(size) < 4e6 AND event_time >= now() - interval 1 hour" @@ -23,7 +23,7 @@ ${CLICKHOUSE_CLIENT} -q "SELECT count()>=$min_trace_entries FROM system.trace_lo # query_id cannot be longer then 28 bytes query_id_http="01526_http_${RANDOM}_$$" -echo "$query" | ${CLICKHOUSE_CURL} -sSg -o /dev/null "${CLICKHOUSE_URL}&query_id=$query_id_http&max_untracked_memory=0&memory_profiler_sample_probability=1&max_threads=1" -d @- +echo "$query" | ${CLICKHOUSE_CURL} -sSg -o /dev/null "${CLICKHOUSE_URL}&query_id=$query_id_http&max_untracked_memory=0&memory_profiler_sample_probability=1&max_threads=1&trace_profile_events=0" -d @- ${CLICKHOUSE_CLIENT} -q "SYSTEM FLUSH LOGS" # at least 2, one allocation, one deallocation # (but actually even more) diff --git a/tests/queries/0_stateless/01556_explain_select_with_union_query.sql b/tests/queries/0_stateless/01556_explain_select_with_union_query.sql index bbd96ef5c69..d8278e1887a 100644 --- a/tests/queries/0_stateless/01556_explain_select_with_union_query.sql +++ b/tests/queries/0_stateless/01556_explain_select_with_union_query.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET union_default_mode = 'DISTINCT'; set enable_global_with_statement = 1; diff --git a/tests/queries/0_stateless/01561_clickhouse_client_stage.reference b/tests/queries/0_stateless/01561_clickhouse_client_stage.reference index 2631199cbab..b6200464293 100644 --- a/tests/queries/0_stateless/01561_clickhouse_client_stage.reference +++ b/tests/queries/0_stateless/01561_clickhouse_client_stage.reference @@ -1,15 +1,15 @@ -execute: --allow_experimental_analyzer=1 +execute: --enable_analyzer=1 "foo" 1 -execute: --allow_experimental_analyzer=1 --stage fetch_columns +execute: --enable_analyzer=1 --stage fetch_columns "__table1.dummy" 0 -execute: --allow_experimental_analyzer=1 --stage with_mergeable_state +execute: --enable_analyzer=1 --stage with_mergeable_state "1_UInt8" 1 -execute: --allow_experimental_analyzer=1 --stage with_mergeable_state_after_aggregation +execute: --enable_analyzer=1 --stage with_mergeable_state_after_aggregation "1_UInt8" 1 -execute: --allow_experimental_analyzer=1 --stage complete +execute: --enable_analyzer=1 --stage complete "foo" 1 diff --git a/tests/queries/0_stateless/01561_clickhouse_client_stage.sh b/tests/queries/0_stateless/01561_clickhouse_client_stage.sh index 99267458421..79c9bb6ae10 100755 --- a/tests/queries/0_stateless/01561_clickhouse_client_stage.sh +++ b/tests/queries/0_stateless/01561_clickhouse_client_stage.sh @@ -6,7 +6,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh opts=( - "--allow_experimental_analyzer=1" + "--enable_analyzer=1" ) function execute_query() diff --git a/tests/queries/0_stateless/01591_window_functions.sql b/tests/queries/0_stateless/01591_window_functions.sql index b821ba13721..db727599d2c 100644 --- a/tests/queries/0_stateless/01591_window_functions.sql +++ b/tests/queries/0_stateless/01591_window_functions.sql @@ -1,6 +1,6 @@ -- Tags: long -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- Too slow SET max_bytes_before_external_sort = 0; diff --git a/tests/queries/0_stateless/01592_window_functions.reference b/tests/queries/0_stateless/01592_window_functions.reference index 06ec67ee82d..ec957dd7a02 100644 --- a/tests/queries/0_stateless/01592_window_functions.reference +++ b/tests/queries/0_stateless/01592_window_functions.reference @@ -79,16 +79,3 @@ iPhone 900 Smartphone 500 500 Kindle Fire 150 Tablet 150 350 Samsung Galaxy Tab 200 Tablet 175 350 iPad 700 Tablet 350 350 ----- Q8 ---- -Lenovo Thinkpad Laptop 700 1 0 -Sony VAIO Laptop 700 1 0 -Dell Vostro Laptop 800 3 0.6666666666666666 -HP Elite Laptop 1200 4 1 -Microsoft Lumia Smartphone 200 1 0 -HTC One Smartphone 400 2 0.3333333333333333 -Nexus Smartphone 500 3 0.6666666666666666 -iPhone Smartphone 900 4 1 -Kindle Fire Tablet 150 1 0 -Samsung Galaxy Tab Tablet 200 2 0.5 -iPad Tablet 700 3 1 -Others Unknow 200 1 0 diff --git a/tests/queries/0_stateless/01592_window_functions.sql b/tests/queries/0_stateless/01592_window_functions.sql index a660fcca7b2..c6bb23bc7cf 100644 --- a/tests/queries/0_stateless/01592_window_functions.sql +++ b/tests/queries/0_stateless/01592_window_functions.sql @@ -101,26 +101,7 @@ SELECT FROM products INNER JOIN product_groups USING (group_id)) t order by group_name, product_name, price; -select '---- Q8 ----'; -INSERT INTO product_groups VALUES (4, 'Unknow'); -INSERT INTO products (product_id,product_name, group_id,price) VALUES (12, 'Others', 4, 200); - -SELECT * -FROM -( - SELECT - product_name, - group_name, - price, - rank() OVER (PARTITION BY group_name ORDER BY price ASC) AS rank, - percent_rank() OVER (PARTITION BY group_name ORDER BY price ASC) AS percent - FROM products - INNER JOIN product_groups USING (group_id) -) AS t -ORDER BY - group_name ASC, - price ASC, - product_name ASC; - drop table product_groups; drop table products; + + diff --git a/tests/queries/0_stateless/01600_detach_permanently.sh b/tests/queries/0_stateless/01600_detach_permanently.sh index 679e9a749ee..be405f8e7d9 100755 --- a/tests/queries/0_stateless/01600_detach_permanently.sh +++ b/tests/queries/0_stateless/01600_detach_permanently.sh @@ -111,8 +111,8 @@ clickhouse_local "INSERT INTO db_ordinary.src SELECT * FROM numbers(10)" clickhouse_local "SELECT if(count() = 10, 'MV is working', 'MV failed') FROM db_ordinary.src_mv_with_inner" clickhouse_local "DETACH VIEW db_ordinary.src_mv_with_inner PERMANENTLY; INSERT INTO db_ordinary.src SELECT * FROM numbers(10)" --stacktrace -clickhouse_local "SELECT if(count() = 10, 'MV can be detached permanently', 'MV detach failed') FROM db_ordinary.src_mv_with_inner SETTINGS allow_experimental_analyzer = 0" 2>&1 | grep -c "db_ordinary.src_mv_with_inner does not exist" -clickhouse_local "SELECT if(count() = 10, 'MV can be detached permanently', 'MV detach failed') FROM db_ordinary.src_mv_with_inner SETTINGS allow_experimental_analyzer = 1" 2>&1 | grep -c "Unknown table expression identifier 'db_ordinary.src_mv_with_inner'" +clickhouse_local "SELECT if(count() = 10, 'MV can be detached permanently', 'MV detach failed') FROM db_ordinary.src_mv_with_inner SETTINGS enable_analyzer = 0" 2>&1 | grep -c "db_ordinary.src_mv_with_inner does not exist" +clickhouse_local "SELECT if(count() = 10, 'MV can be detached permanently', 'MV detach failed') FROM db_ordinary.src_mv_with_inner SETTINGS enable_analyzer = 1" 2>&1 | grep -c "Unknown table expression identifier 'db_ordinary.src_mv_with_inner'" ## Quite silly: ATTACH MATERIALIZED VIEW don't work with short syntax (w/o select), but i can attach it using ATTACH TABLE ... clickhouse_local "ATTACH TABLE db_ordinary.src_mv_with_inner" diff --git a/tests/queries/0_stateless/01600_multiple_left_join_with_aliases.sql b/tests/queries/0_stateless/01600_multiple_left_join_with_aliases.sql index ae296e18560..2945622fbed 100644 --- a/tests/queries/0_stateless/01600_multiple_left_join_with_aliases.sql +++ b/tests/queries/0_stateless/01600_multiple_left_join_with_aliases.sql @@ -1,9 +1,4 @@ --- Tags: no-parallel - -drop database if exists test_01600; -create database test_01600; - -CREATE TABLE test_01600.base +CREATE TABLE base ( `id` UInt64, `id2` UInt64, @@ -14,7 +9,7 @@ ENGINE=MergeTree() PARTITION BY d ORDER BY (id,id2,d); -CREATE TABLE test_01600.derived1 +CREATE TABLE derived1 ( `id1` UInt64, `d1` UInt64, @@ -25,7 +20,7 @@ PARTITION BY d1 ORDER BY (id1, d1) ; -CREATE TABLE test_01600.derived2 +CREATE TABLE derived2 ( `id2` UInt64, `d2` UInt64, @@ -36,19 +31,17 @@ PARTITION BY d2 ORDER BY (id2, d2) ; -select +select base.id as `base.id`, derived2.value2 as `derived2.value2`, derived1.value1 as `derived1.value1` -from test_01600.base as base -left join test_01600.derived2 as derived2 on base.id2 = derived2.id2 -left join test_01600.derived1 as derived1 on base.id = derived1.id1; +from base as base +left join derived2 as derived2 on base.id2 = derived2.id2 +left join derived1 as derived1 on base.id = derived1.id1; SELECT base.id AS `base.id`, derived1.value1 AS `derived1.value1` -FROM test_01600.base AS base -LEFT JOIN test_01600.derived1 AS derived1 ON base.id = derived1.id1; - -drop database test_01600; +FROM base AS base +LEFT JOIN derived1 AS derived1 ON base.id = derived1.id1; diff --git a/tests/queries/0_stateless/01600_parts_states_metrics_long.sh b/tests/queries/0_stateless/01600_parts_states_metrics_long.sh index 47b5a4dea13..0a9f94cc451 100755 --- a/tests/queries/0_stateless/01600_parts_states_metrics_long.sh +++ b/tests/queries/0_stateless/01600_parts_states_metrics_long.sh @@ -1,50 +1,63 @@ #!/usr/bin/env bash -# Tags: long +# Tags: long, no-parallel 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))" +function query() +{ + # NOTE: database_atomic_wait_for_drop_and_detach_synchronously needed only for local env, CI has it ON + ${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&database_atomic_wait_for_drop_and_detach_synchronously=1" -d "$*" +} + # The query is not atomic - it can compare states between system.parts and system.metrics from different points in time. # So, there is inherent race condition. But it should get expected result eventually. # In case of test failure, this code will do infinite loop and timeout. verify() { - for i in {1..5000} - do - result=$( $CLICKHOUSE_CLIENT --query="$verify_sql" ) - [ "$result" = "1" ] && echo "$result" && break - sleep 0.1 + local result - if [[ $i -eq 5000 ]] - then - $CLICKHOUSE_CLIENT " - SELECT sumIf(value, metric = 'PartsActive'), sumIf(value, metric = 'PartsOutdated') FROM system.metrics; - SELECT sum(active), sum(NOT active) FROM system.parts; - SELECT sum(active), sum(NOT active) FROM system.projection_parts; - SELECT count() FROM system.dropped_tables_parts; - " + for _ in {1..100}; do + # NOTE: database = $CLICKHOUSE_DATABASE is unwanted + result=$( query "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 + ))" + ) + + if [ "$result" = "1" ]; then + echo "$result" + return fi + + sleep 0.5 done + + $CLICKHOUSE_CLIENT -q " + SELECT sumIf(value, metric = 'PartsActive'), sumIf(value, metric = 'PartsOutdated') FROM system.metrics; + SELECT sum(active), sum(NOT active) FROM system.parts; + SELECT sum(active), sum(NOT active) FROM system.projection_parts; + SELECT count() FROM system.dropped_tables_parts; + " } -$CLICKHOUSE_CLIENT --database_atomic_wait_for_drop_and_detach_synchronously=1 --query="DROP TABLE IF EXISTS test_table" -$CLICKHOUSE_CLIENT --query="CREATE TABLE test_table (data Date) ENGINE = MergeTree PARTITION BY toYear(data) ORDER BY data;" +query "DROP TABLE IF EXISTS test_table" +query "CREATE TABLE test_table (data Date) ENGINE = MergeTree PARTITION BY toYear(data) ORDER BY data;" -$CLICKHOUSE_CLIENT --query="INSERT INTO test_table VALUES ('1992-01-01')" +query "INSERT INTO test_table VALUES ('1992-01-01')" verify -$CLICKHOUSE_CLIENT --query="INSERT INTO test_table VALUES ('1992-01-02')" +query "INSERT INTO test_table VALUES ('1992-01-02')" verify -$CLICKHOUSE_CLIENT --query="OPTIMIZE TABLE test_table FINAL" +query "OPTIMIZE TABLE test_table FINAL" verify -$CLICKHOUSE_CLIENT --database_atomic_wait_for_drop_and_detach_synchronously=1 --query="DROP TABLE test_table" +query "DROP TABLE test_table" verify diff --git a/tests/queries/0_stateless/01622_constraints_simple_optimization.reference b/tests/queries/0_stateless/01622_constraints_simple_optimization.reference index 84c872856ff..64a9db37a68 100644 --- a/tests/queries/0_stateless/01622_constraints_simple_optimization.reference +++ b/tests/queries/0_stateless/01622_constraints_simple_optimization.reference @@ -52,7 +52,7 @@ QUERY id: 0 LIST id: 5, nodes: 2 COLUMN id: 6, column_name: c, result_type: Int64, source_id: 3 CONSTANT id: 7, constant_value: UInt64_100, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT count() FROM constraint_test_constants WHERE c > 100 @@ -70,7 +70,7 @@ QUERY id: 0 LIST id: 5, nodes: 2 COLUMN id: 6, column_name: c, result_type: Int64, source_id: 3 CONSTANT id: 7, constant_value: UInt64_100, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT count() FROM constraint_test_constants QUERY id: 0 @@ -81,4 +81,4 @@ QUERY id: 0 FUNCTION id: 2, function_name: count, function_type: aggregate, result_type: UInt64 JOIN TREE TABLE id: 3, alias: __table1, table_name: default.constraint_test_constants - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 diff --git a/tests/queries/0_stateless/01622_constraints_simple_optimization.sql b/tests/queries/0_stateless/01622_constraints_simple_optimization.sql index acde02e2c67..e549467de02 100644 --- a/tests/queries/0_stateless/01622_constraints_simple_optimization.sql +++ b/tests/queries/0_stateless/01622_constraints_simple_optimization.sql @@ -101,10 +101,10 @@ SELECT count() FROM constraint_test_constants WHERE 11 <= a; ---> assumption -> EXPLAIN SYNTAX SELECT count() FROM constraint_test_constants WHERE (a > 100 OR b > 100 OR c > 100) AND (a <= 100 OR b > 100 OR c > 100); -- EXPLAIN QUERY TREE SELECT count() FROM constraint_test_constants WHERE (a > 100 OR b > 100 OR c > 100) AND (a <= 100 OR b > 100 OR c > 100); ---> the order of the generated checks is not consistent EXPLAIN SYNTAX SELECT count() FROM constraint_test_constants WHERE (a > 100 OR b > 100 OR c > 100) AND (a <= 100 OR b > 100 OR c > 100) AND (NOT b > 100 OR c > 100); -EXPLAIN QUERY TREE SELECT count() FROM constraint_test_constants WHERE (a > 100 OR b > 100 OR c > 100) AND (a <= 100 OR b > 100 OR c > 100) AND (NOT b > 100 OR c > 100) SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT count() FROM constraint_test_constants WHERE (a > 100 OR b > 100 OR c > 100) AND (a <= 100 OR b > 100 OR c > 100) AND (NOT b > 100 OR c > 100) SETTINGS enable_analyzer = 1; EXPLAIN SYNTAX SELECT count() FROM constraint_test_constants WHERE (a > 100 OR b > 100 OR c > 100) AND (a <= 100 OR b > 100 OR c > 100) AND (NOT b > 100 OR c > 100) AND (c > 100); -EXPLAIN QUERY TREE SELECT count() FROM constraint_test_constants WHERE (a > 100 OR b > 100 OR c > 100) AND (a <= 100 OR b > 100 OR c > 100) AND (NOT b > 100 OR c > 100) AND (c > 100) SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT count() FROM constraint_test_constants WHERE (a > 100 OR b > 100 OR c > 100) AND (a <= 100 OR b > 100 OR c > 100) AND (NOT b > 100 OR c > 100) AND (c > 100) SETTINGS enable_analyzer = 1; EXPLAIN SYNTAX SELECT count() FROM constraint_test_constants WHERE (a > 100 OR b > 100 OR c > 100) AND (a <= 100 OR b > 100 OR c > 100) AND (NOT b > 100 OR c > 100) AND (c <= 100); -EXPLAIN QUERY TREE SELECT count() FROM constraint_test_constants WHERE (a > 100 OR b > 100 OR c > 100) AND (a <= 100 OR b > 100 OR c > 100) AND (NOT b > 100 OR c > 100) AND (c <= 100) SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT count() FROM constraint_test_constants WHERE (a > 100 OR b > 100 OR c > 100) AND (a <= 100 OR b > 100 OR c > 100) AND (NOT b > 100 OR c > 100) AND (c <= 100) SETTINGS enable_analyzer = 1; DROP TABLE constraint_test_constants; diff --git a/tests/queries/0_stateless/01622_constraints_where_optimization.reference b/tests/queries/0_stateless/01622_constraints_where_optimization.reference index 3f6e8211f1a..09a6dd3d0e7 100644 --- a/tests/queries/0_stateless/01622_constraints_where_optimization.reference +++ b/tests/queries/0_stateless/01622_constraints_where_optimization.reference @@ -11,7 +11,7 @@ QUERY id: 0 TABLE id: 3, alias: __table1, table_name: default.t_constraints_where WHERE CONSTANT id: 4, constant_value: UInt64_0, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT count() FROM t_constraints_where WHERE 0 @@ -25,7 +25,7 @@ QUERY id: 0 TABLE id: 3, alias: __table1, table_name: default.t_constraints_where WHERE CONSTANT id: 4, constant_value: UInt64_0, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT count() FROM t_constraints_where WHERE 0 @@ -39,7 +39,7 @@ QUERY id: 0 TABLE id: 3, alias: __table1, table_name: default.t_constraints_where WHERE CONSTANT id: 4, constant_value: UInt64_0, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT count() FROM t_constraints_where WHERE b < 8 @@ -57,7 +57,7 @@ QUERY id: 0 LIST id: 5, nodes: 2 COLUMN id: 6, column_name: b, result_type: UInt32, source_id: 3 CONSTANT id: 7, constant_value: UInt64_8, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT count() FROM t_constraints_where PREWHERE (b > 20) OR (b < 8) @@ -75,7 +75,7 @@ QUERY id: 0 LIST id: 5, nodes: 2 COLUMN id: 6, column_name: b, result_type: UInt32, source_id: 3 CONSTANT id: 7, constant_value: UInt64_8, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT count() FROM t_constraints_where QUERY id: 0 @@ -86,4 +86,4 @@ QUERY id: 0 FUNCTION id: 2, function_name: count, function_type: aggregate, result_type: UInt64 JOIN TREE TABLE id: 3, alias: __table1, table_name: default.t_constraints_where - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 diff --git a/tests/queries/0_stateless/01622_constraints_where_optimization.sql b/tests/queries/0_stateless/01622_constraints_where_optimization.sql index d41b1988bdd..63803ec8ce6 100644 --- a/tests/queries/0_stateless/01622_constraints_where_optimization.sql +++ b/tests/queries/0_stateless/01622_constraints_where_optimization.sql @@ -9,15 +9,15 @@ CREATE TABLE t_constraints_where(a UInt32, b UInt32, CONSTRAINT c1 ASSUME b >= 5 INSERT INTO t_constraints_where VALUES (1, 7); EXPLAIN SYNTAX SELECT count() FROM t_constraints_where WHERE b > 15; -- assumption -> 0 -EXPLAIN QUERY TREE SELECT count() FROM t_constraints_where WHERE b > 15 SETTINGS allow_experimental_analyzer = 1; -- assumption -> 0 +EXPLAIN QUERY TREE SELECT count() FROM t_constraints_where WHERE b > 15 SETTINGS enable_analyzer = 1; -- assumption -> 0 EXPLAIN SYNTAX SELECT count() FROM t_constraints_where WHERE b = 20; -- assumption -> 0 -EXPLAIN QUERY TREE SELECT count() FROM t_constraints_where WHERE b = 20 SETTINGS allow_experimental_analyzer = 1; -- assumption -> 0 +EXPLAIN QUERY TREE SELECT count() FROM t_constraints_where WHERE b = 20 SETTINGS enable_analyzer = 1; -- assumption -> 0 EXPLAIN SYNTAX SELECT count() FROM t_constraints_where WHERE b < 2; -- assumption -> 0 -EXPLAIN QUERY TREE SELECT count() FROM t_constraints_where WHERE b < 2 SETTINGS allow_experimental_analyzer = 1; -- assumption -> 0 +EXPLAIN QUERY TREE SELECT count() FROM t_constraints_where WHERE b < 2 SETTINGS enable_analyzer = 1; -- assumption -> 0 EXPLAIN SYNTAX SELECT count() FROM t_constraints_where WHERE b > 20 OR b < 8; -- assumption -> remove (b < 20) -EXPLAIN QUERY TREE SELECT count() FROM t_constraints_where WHERE b > 20 OR b < 8 SETTINGS allow_experimental_analyzer = 1; -- assumption -> remove (b < 20) +EXPLAIN QUERY TREE SELECT count() FROM t_constraints_where WHERE b > 20 OR b < 8 SETTINGS enable_analyzer = 1; -- assumption -> remove (b < 20) EXPLAIN SYNTAX SELECT count() FROM t_constraints_where PREWHERE b > 20 OR b < 8; -- assumption -> remove (b < 20) -EXPLAIN QUERY TREE SELECT count() FROM t_constraints_where PREWHERE b > 20 OR b < 8 SETTINGS allow_experimental_analyzer = 1; -- assumption -> remove (b < 20) +EXPLAIN QUERY TREE SELECT count() FROM t_constraints_where PREWHERE b > 20 OR b < 8 SETTINGS enable_analyzer = 1; -- assumption -> remove (b < 20) DROP TABLE t_constraints_where; @@ -26,6 +26,6 @@ CREATE TABLE t_constraints_where(a UInt32, b UInt32, CONSTRAINT c1 ASSUME b < 10 INSERT INTO t_constraints_where VALUES (1, 7); EXPLAIN SYNTAX SELECT count() FROM t_constraints_where WHERE b = 1 OR b < 18 OR b > 5; -- assumption -> (b < 20) -> 0; -EXPLAIN QUERY TREE SELECT count() FROM t_constraints_where WHERE b = 1 OR b < 18 OR b > 5 SETTINGS allow_experimental_analyzer = 1; -- assumption -> (b < 20) -> 0; +EXPLAIN QUERY TREE SELECT count() FROM t_constraints_where WHERE b = 1 OR b < 18 OR b > 5 SETTINGS enable_analyzer = 1; -- assumption -> (b < 20) -> 0; DROP TABLE t_constraints_where; diff --git a/tests/queries/0_stateless/01623_constraints_column_swap.reference b/tests/queries/0_stateless/01623_constraints_column_swap.reference index d504a86365b..b49884b4798 100644 --- a/tests/queries/0_stateless/01623_constraints_column_swap.reference +++ b/tests/queries/0_stateless/01623_constraints_column_swap.reference @@ -27,7 +27,7 @@ QUERY id: 0 LIST id: 12, nodes: 2 COLUMN id: 13, column_name: b, result_type: UInt64, source_id: 5 CONSTANT id: 14, constant_value: UInt64_1, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT cityHash64(a) + 10, b + 3 @@ -57,7 +57,7 @@ QUERY id: 0 LIST id: 12, nodes: 2 COLUMN id: 13, column_name: b, result_type: UInt64, source_id: 5 CONSTANT id: 14, constant_value: UInt64_1, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT (b AS `cityHash64(a)`) + 10, (b AS b) + 3 @@ -87,7 +87,7 @@ QUERY id: 0 LIST id: 12, nodes: 2 COLUMN id: 13, column_name: b, result_type: UInt64, source_id: 5 CONSTANT id: 14, constant_value: UInt64_0, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT (b AS `cityHash64(a)`) + 10, (b AS b) + 3 @@ -117,7 +117,7 @@ QUERY id: 0 LIST id: 12, nodes: 2 COLUMN id: 13, column_name: b, result_type: UInt64, source_id: 5 CONSTANT id: 14, constant_value: UInt64_0, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT (b AS `cityHash64(a)`) + 10, (b AS b) + 3 @@ -147,7 +147,7 @@ QUERY id: 0 LIST id: 12, nodes: 2 COLUMN id: 13, column_name: b, result_type: UInt64, source_id: 5 CONSTANT id: 14, constant_value: UInt64_1, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT (b AS `cityHash64(a)`) + 10 FROM column_swap_test_test WHERE b = 0 @@ -169,7 +169,7 @@ QUERY id: 0 LIST id: 8, nodes: 2 COLUMN id: 9, column_name: b, result_type: UInt64, source_id: 5 CONSTANT id: 10, constant_value: UInt64_0, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT (cityHash64(a) AS `cityHash64(a)`) + 10, a @@ -201,7 +201,7 @@ QUERY id: 0 LIST id: 13, nodes: 1 COLUMN id: 14, column_name: a, result_type: String, source_id: 7 CONSTANT id: 15, constant_value: UInt64_0, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT (cityHash64(a) AS b) + 10, a @@ -233,7 +233,7 @@ QUERY id: 0 LIST id: 13, nodes: 1 COLUMN id: 14, column_name: a, result_type: String, source_id: 7 CONSTANT id: 15, constant_value: UInt64_0, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT a AS `substring(reverse(b), 1, 1)`, a AS a @@ -255,7 +255,7 @@ QUERY id: 0 LIST id: 6, nodes: 2 COLUMN id: 7, column_name: a, result_type: String, source_id: 3 CONSTANT id: 8, constant_value: \'c\', constant_value_type: String - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT a AS `substring(reverse(b), 1, 1)`, a AS a @@ -277,7 +277,7 @@ QUERY id: 0 LIST id: 6, nodes: 2 COLUMN id: 7, column_name: a, result_type: String, source_id: 3 CONSTANT id: 8, constant_value: \'c\', constant_value_type: String - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT a AS t1, a AS t2 @@ -299,7 +299,7 @@ QUERY id: 0 LIST id: 6, nodes: 2 COLUMN id: 7, column_name: a, result_type: String, source_id: 3 CONSTANT id: 8, constant_value: \'c\', constant_value_type: String - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT a AS `substring(reverse(b), 1, 1)` FROM column_swap_test_test WHERE a = \'c\' @@ -317,7 +317,7 @@ QUERY id: 0 LIST id: 5, nodes: 2 COLUMN id: 6, column_name: a, result_type: String, source_id: 3 CONSTANT id: 7, constant_value: \'c\', constant_value_type: String - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT a FROM t_bad_constraint QUERY id: 0 @@ -328,4 +328,4 @@ QUERY id: 0 COLUMN id: 2, column_name: a, result_type: UInt32, source_id: 3 JOIN TREE TABLE id: 3, alias: __table1, table_name: default.t_bad_constraint - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 diff --git a/tests/queries/0_stateless/01623_constraints_column_swap.sql b/tests/queries/0_stateless/01623_constraints_column_swap.sql index 242be87938d..ccd387c9f8d 100644 --- a/tests/queries/0_stateless/01623_constraints_column_swap.sql +++ b/tests/queries/0_stateless/01623_constraints_column_swap.sql @@ -14,22 +14,22 @@ INSERT INTO column_swap_test_test VALUES (1, 'cat', 1), (2, 'dog', 2); INSERT INTO column_swap_test_test SELECT number AS i, format('test {} kek {}', toString(number), toString(number + 10)) AS a, 1 AS b FROM system.numbers LIMIT 1000000; EXPLAIN SYNTAX SELECT cityHash64(a) + 10, b + 3 FROM column_swap_test_test WHERE cityHash64(a) = 1; -EXPLAIN QUERY TREE SELECT cityHash64(a) + 10, b + 3 FROM column_swap_test_test WHERE cityHash64(a) = 1 SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT cityHash64(a) + 10, b + 3 FROM column_swap_test_test WHERE cityHash64(a) = 1 SETTINGS enable_analyzer = 1; EXPLAIN SYNTAX SELECT cityHash64(a) + 10, b + 3 FROM column_swap_test_test PREWHERE cityHash64(a) = 1; -EXPLAIN QUERY TREE SELECT cityHash64(a) + 10, b + 3 FROM column_swap_test_test PREWHERE cityHash64(a) = 1 SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT cityHash64(a) + 10, b + 3 FROM column_swap_test_test PREWHERE cityHash64(a) = 1 SETTINGS enable_analyzer = 1; EXPLAIN SYNTAX SELECT cityHash64(a) + 10, b + 3 FROM column_swap_test_test WHERE cityHash64(a) = 0; -EXPLAIN QUERY TREE SELECT cityHash64(a) + 10, b + 3 FROM column_swap_test_test WHERE cityHash64(a) = 0 SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT cityHash64(a) + 10, b + 3 FROM column_swap_test_test WHERE cityHash64(a) = 0 SETTINGS enable_analyzer = 1; EXPLAIN SYNTAX SELECT cityHash64(a) + 10, b + 3 FROM column_swap_test_test WHERE b = 0; -EXPLAIN QUERY TREE SELECT cityHash64(a) + 10, b + 3 FROM column_swap_test_test WHERE b = 0 SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT cityHash64(a) + 10, b + 3 FROM column_swap_test_test WHERE b = 0 SETTINGS enable_analyzer = 1; EXPLAIN SYNTAX SELECT cityHash64(a) + 10, b + 3 FROM column_swap_test_test WHERE b = 1; -EXPLAIN QUERY TREE SELECT cityHash64(a) + 10, b + 3 FROM column_swap_test_test WHERE b = 1 SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT cityHash64(a) + 10, b + 3 FROM column_swap_test_test WHERE b = 1 SETTINGS enable_analyzer = 1; EXPLAIN SYNTAX SELECT cityHash64(a) + 10 FROM column_swap_test_test WHERE cityHash64(a) = 0; -EXPLAIN QUERY TREE SELECT cityHash64(a) + 10 FROM column_swap_test_test WHERE cityHash64(a) = 0 SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT cityHash64(a) + 10 FROM column_swap_test_test WHERE cityHash64(a) = 0 SETTINGS enable_analyzer = 1; EXPLAIN SYNTAX SELECT cityHash64(a) + 10, a FROM column_swap_test_test WHERE cityHash64(a) = 0; -EXPLAIN QUERY TREE SELECT cityHash64(a) + 10, a FROM column_swap_test_test WHERE cityHash64(a) = 0 SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT cityHash64(a) + 10, a FROM column_swap_test_test WHERE cityHash64(a) = 0 SETTINGS enable_analyzer = 1; EXPLAIN SYNTAX SELECT b + 10, a FROM column_swap_test_test WHERE b = 0; -EXPLAIN QUERY TREE SELECT b + 10, a FROM column_swap_test_test WHERE b = 0 SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT b + 10, a FROM column_swap_test_test WHERE b = 0 SETTINGS enable_analyzer = 1; DROP TABLE column_swap_test_test; @@ -37,13 +37,13 @@ CREATE TABLE column_swap_test_test (i Int64, a String, b String, CONSTRAINT c1 A INSERT INTO column_swap_test_test SELECT number AS i, toString(number) AS a, format('test {} kek {}', toString(number), toString(number + 10)) b FROM system.numbers LIMIT 1000000; EXPLAIN SYNTAX SELECT substring(reverse(b), 1, 1), a FROM column_swap_test_test WHERE a = 'c'; -EXPLAIN QUERY TREE SELECT substring(reverse(b), 1, 1), a FROM column_swap_test_test WHERE a = 'c' SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT substring(reverse(b), 1, 1), a FROM column_swap_test_test WHERE a = 'c' SETTINGS enable_analyzer = 1; EXPLAIN SYNTAX SELECT substring(reverse(b), 1, 1), a FROM column_swap_test_test WHERE substring(reverse(b), 1, 1) = 'c'; -EXPLAIN QUERY TREE SELECT substring(reverse(b), 1, 1), a FROM column_swap_test_test WHERE substring(reverse(b), 1, 1) = 'c' SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT substring(reverse(b), 1, 1), a FROM column_swap_test_test WHERE substring(reverse(b), 1, 1) = 'c' SETTINGS enable_analyzer = 1; EXPLAIN SYNTAX SELECT substring(reverse(b), 1, 1) AS t1, a AS t2 FROM column_swap_test_test WHERE substring(reverse(b), 1, 1) = 'c'; -EXPLAIN QUERY TREE SELECT substring(reverse(b), 1, 1) AS t1, a AS t2 FROM column_swap_test_test WHERE substring(reverse(b), 1, 1) = 'c' SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT substring(reverse(b), 1, 1) AS t1, a AS t2 FROM column_swap_test_test WHERE substring(reverse(b), 1, 1) = 'c' SETTINGS enable_analyzer = 1; EXPLAIN SYNTAX SELECT substring(reverse(b), 1, 1) FROM column_swap_test_test WHERE substring(reverse(b), 1, 1) = 'c'; -EXPLAIN QUERY TREE SELECT substring(reverse(b), 1, 1) FROM column_swap_test_test WHERE substring(reverse(b), 1, 1) = 'c' SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT substring(reverse(b), 1, 1) FROM column_swap_test_test WHERE substring(reverse(b), 1, 1) = 'c' SETTINGS enable_analyzer = 1; DROP TABLE column_swap_test_test; @@ -54,6 +54,6 @@ CREATE TABLE t_bad_constraint(a UInt32, s String, CONSTRAINT c1 ASSUME a = toUIn INSERT INTO t_bad_constraint SELECT number, randomPrintableASCII(100) FROM numbers(10000); EXPLAIN SYNTAX SELECT a FROM t_bad_constraint; -EXPLAIN QUERY TREE SELECT a FROM t_bad_constraint SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT a FROM t_bad_constraint SETTINGS enable_analyzer = 1; DROP TABLE t_bad_constraint; diff --git a/tests/queries/0_stateless/01646_rewrite_sum_if.sql b/tests/queries/0_stateless/01646_rewrite_sum_if.sql index da341a3f7db..fd98a157ac6 100644 --- a/tests/queries/0_stateless/01646_rewrite_sum_if.sql +++ b/tests/queries/0_stateless/01646_rewrite_sum_if.sql @@ -34,7 +34,7 @@ SELECT sum(if(number % 2 == 0, 0, 1)) FROM numbers(100); SELECT sum(if(number % 2 == 0 as cond_expr, 0 as zero_expr, 1 as one_expr) as if_expr), sum(cond_expr), sum(if_expr), one_expr, zero_expr FROM numbers(100); SELECT countIf(number % 2 != 0) FROM numbers(100); -set allow_experimental_analyzer = true; +set enable_analyzer = true; EXPLAIN QUERY TREE run_passes=1 SELECT sumIf(123, number % 2 == 0) FROM numbers(100); EXPLAIN QUERY TREE run_passes=1 SELECT sum(if(number % 2 == 0, 123, 0)) FROM numbers(100); diff --git a/tests/queries/0_stateless/01651_bugs_from_15889.sql b/tests/queries/0_stateless/01651_bugs_from_15889.sql index dd31f2941ef..b98feaa1f65 100644 --- a/tests/queries/0_stateless/01651_bugs_from_15889.sql +++ b/tests/queries/0_stateless/01651_bugs_from_15889.sql @@ -111,5 +111,5 @@ WITH ( ) AS t) SELECT if(dateDiff('second', toDateTime(time_with_microseconds), toDateTime(t)) = -9223372036854775808, 'ok', ''); -set joined_subquery_requires_alias=0, allow_experimental_analyzer=0; -- the query is invalid with a new analyzer +set joined_subquery_requires_alias=0, enable_analyzer=0; -- the query is invalid with a new analyzer SELECT number, number / 2 AS n, j1, j2 FROM remote('127.0.0.{2,3}', system.numbers) GLOBAL ANY LEFT JOIN (SELECT number / 3 AS n, number AS j1, 'Hello' AS j2 FROM system.numbers LIMIT 1048577) USING (n) LIMIT 10 format Null; diff --git a/tests/queries/0_stateless/01655_plan_optimizations.sh b/tests/queries/0_stateless/01655_plan_optimizations.sh index 4bd0eb7d908..42cdac8c01f 100755 --- a/tests/queries/0_stateless/01655_plan_optimizations.sh +++ b/tests/queries/0_stateless/01655_plan_optimizations.sh @@ -26,13 +26,13 @@ $CLICKHOUSE_CLIENT -q " settings enable_optimize_predicate_expression=0" echo "> filter should be pushed down after aggregating, column after aggregation is const" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=0 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=0 -q " explain actions = 1 select s, y, y != 0 from (select sum(x) as s, y from ( select number as x, number + 1 as y from numbers(10)) group by y ) where y != 0 settings enable_optimize_predicate_expression=0" | grep -o "Aggregating\|Filter\|COLUMN Const(UInt8) -> notEquals(y, 0)" echo "> (analyzer) filter should be pushed down after aggregating, column after aggregation is const" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=1 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=1 -q " explain actions = 1 select s, y, y != 0 from (select sum(x) as s, y from ( select number as x, number + 1 as y from numbers(10)) group by y ) where y != 0 @@ -44,14 +44,14 @@ $CLICKHOUSE_CLIENT -q " settings enable_optimize_predicate_expression=0" echo "> one condition of filter should be pushed down after aggregating, other condition is aliased" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=0 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=0 -q " explain actions = 1 select s, y from ( select sum(x) as s, y from (select number as x, number + 1 as y from numbers(10)) group by y ) where y != 0 and s != 4 settings enable_optimize_predicate_expression=0" | grep -o "Aggregating\|Filter column\|Filter column: notEquals(y, 0)\|ALIAS notEquals(s, 4) :: 4 -> and(notEquals(y, 0), notEquals(s, 4)) UInt8 : 2" echo "> (analyzer) one condition of filter should be pushed down after aggregating, other condition is aliased" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=1 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=1 -q " explain actions = 1 select s, y from ( select sum(x) as s, y from (select number as x, number + 1 as y from numbers(10)) group by y ) where y != 0 and s != 4 @@ -64,14 +64,14 @@ $CLICKHOUSE_CLIENT -q " settings enable_optimize_predicate_expression=0" echo "> one condition of filter should be pushed down after aggregating, other condition is casted" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=0 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=0 -q " explain actions = 1 select s, y from ( select sum(x) as s, y from (select number as x, number + 1 as y from numbers(10)) group by y ) where y != 0 and s - 4 settings enable_optimize_predicate_expression=0" | grep -o "Aggregating\|Filter column\|Filter column: notEquals(y, 0)\|FUNCTION and(minus(s, 4) :: 5, 1 :: 3) -> and(notEquals(y, 0), minus(s, 4))" echo "> (analyzer) one condition of filter should be pushed down after aggregating, other condition is casted" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=1 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=1 -q " explain actions = 1 select s, y from ( select sum(x) as s, y from (select number as x, number + 1 as y from numbers(10)) group by y ) where y != 0 and s - 4 @@ -84,14 +84,14 @@ $CLICKHOUSE_CLIENT -q " settings enable_optimize_predicate_expression=0" echo "> one condition of filter should be pushed down after aggregating, other two conditions are ANDed" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=0 --convert_query_to_cnf=0 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=0 --convert_query_to_cnf=0 -q " explain actions = 1 select s, y from ( select sum(x) as s, y from (select number as x, number + 1 as y from numbers(10)) group by y ) where y != 0 and s - 8 and s - 4 settings enable_optimize_predicate_expression=0" | grep -o "Aggregating\|Filter column\|Filter column: notEquals(y, 0)\|FUNCTION and(minus(s, 8) :: 5, minus(s, 4) :: 2) -> and(notEquals(y, 0), minus(s, 8), minus(s, 4))" echo "> (analyzer) one condition of filter should be pushed down after aggregating, other two conditions are ANDed" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=1 --convert_query_to_cnf=0 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=1 --convert_query_to_cnf=0 -q " explain actions = 1 select s, y from ( select sum(x) as s, y from (select number as x, number + 1 as y from numbers(10)) group by y ) where y != 0 and s - 8 and s - 4 @@ -104,14 +104,14 @@ $CLICKHOUSE_CLIENT -q " settings enable_optimize_predicate_expression=0" echo "> two conditions of filter should be pushed down after aggregating and ANDed, one condition is aliased" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=0 --convert_query_to_cnf=0 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=0 --convert_query_to_cnf=0 -q " explain actions = 1 select s, y from ( select sum(x) as s, y from (select number as x, number + 1 as y from numbers(10)) group by y ) where y != 0 and s != 8 and y - 4 settings enable_optimize_predicate_expression=0" | grep -o "Aggregating\|Filter column\|Filter column: and(notEquals(y, 0), minus(y, 4))\|ALIAS notEquals(s, 8) :: 4 -> and(notEquals(y, 0), notEquals(s, 8), minus(y, 4))" echo "> (analyzer) two conditions of filter should be pushed down after aggregating and ANDed, one condition is aliased" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=1 --convert_query_to_cnf=0 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=1 --convert_query_to_cnf=0 -q " explain actions = 1 select s, y from ( select sum(x) as s, y from (select number as x, number + 1 as y from numbers(10)) group by y ) where y != 0 and s != 8 and y - 4 @@ -124,13 +124,13 @@ $CLICKHOUSE_CLIENT -q " settings enable_optimize_predicate_expression=0" echo "> filter is split, one part is filtered before ARRAY JOIN" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=0 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=0 -q " explain actions = 1 select x, y from ( select range(number) as x, number + 1 as y from numbers(3) ) array join x where y != 2 and x != 0" | grep -o "Filter column: and(notEquals(y, 2), notEquals(x, 0))\|ARRAY JOIN x\|Filter column: notEquals(y, 2)" echo "> (analyzer) filter is split, one part is filtered before ARRAY JOIN" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=1 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=1 -q " explain actions = 1 select x, y from ( select range(number) as x, number + 1 as y from numbers(3) ) array join x where y != 2 and x != 0" | @@ -154,14 +154,14 @@ $CLICKHOUSE_CLIENT -q " # settings enable_optimize_predicate_expression=0" echo "> filter is pushed down before Distinct" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=0 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=0 -q " explain actions = 1 select x, y from ( select distinct x, y from (select number % 2 as x, number % 3 as y from numbers(10)) ) where y != 2 settings enable_optimize_predicate_expression=0" | grep -o "Distinct\|Filter column: notEquals(y, 2)" echo "> (analyzer) filter is pushed down before Distinct" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=1 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=1 -q " explain actions = 1 select x, y from ( select distinct x, y from (select number % 2 as x, number % 3 as y from numbers(10)) ) where y != 2 @@ -174,14 +174,14 @@ $CLICKHOUSE_CLIENT -q " settings enable_optimize_predicate_expression=0" echo "> filter is pushed down before sorting steps" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=0 --convert_query_to_cnf=0 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=0 --convert_query_to_cnf=0 -q " explain actions = 1 select x, y from ( select number % 2 as x, number % 3 as y from numbers(6) order by y desc ) where x != 0 and y != 0 settings enable_optimize_predicate_expression = 0" | grep -o "Sorting\|Filter column: and(notEquals(x, 0), notEquals(y, 0))" echo "> (analyzer) filter is pushed down before sorting steps" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=1 --convert_query_to_cnf=0 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=1 --convert_query_to_cnf=0 -q " explain actions = 1 select x, y from ( select number % 2 as x, number % 3 as y from numbers(6) order by y desc ) where x != 0 and y != 0 @@ -194,14 +194,14 @@ $CLICKHOUSE_CLIENT -q " settings enable_optimize_predicate_expression = 0" echo "> filter is pushed down before TOTALS HAVING and aggregating" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=0 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=0 -q " explain actions = 1 select * from ( select y, sum(x) from (select number as x, number % 4 as y from numbers(10)) group by y with totals ) where y != 2 settings enable_optimize_predicate_expression=0" | grep -o "TotalsHaving\|Aggregating\|Filter column: notEquals(y, 2)" echo "> (analyzer) filter is pushed down before TOTALS HAVING and aggregating" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=1 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=1 -q " explain actions = 1 select * from ( select y, sum(x) from (select number as x, number % 4 as y from numbers(10)) group by y with totals ) where y != 2 @@ -224,14 +224,14 @@ $CLICKHOUSE_CLIENT -q " ) where number != 2 settings enable_optimize_predicate_expression=0" echo "> one condition of filter is pushed down before LEFT JOIN" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=0 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=0 -q " explain actions = 1 select number as a, r.b from numbers(4) as l any left join ( select number + 2 as b from numbers(3) ) as r on a = r.b where a != 1 and b != 2 settings enable_optimize_predicate_expression = 0" | grep -o "Join\|Filter column: notEquals(number, 1)" echo "> (analyzer) one condition of filter is pushed down before LEFT JOIN" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=1 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=1 -q " explain actions = 1 select number as a, r.b from numbers(4) as l any left join ( select number + 2 as b from numbers(3) @@ -243,14 +243,14 @@ $CLICKHOUSE_CLIENT -q " ) as r on a = r.b where a != 1 and b != 2 settings enable_optimize_predicate_expression = 0" | sort echo "> one condition of filter is pushed down before INNER JOIN" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=0 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=0 -q " explain actions = 1 select number as a, r.b from numbers(4) as l any inner join ( select number + 2 as b from numbers(3) ) as r on a = r.b where a != 1 and b != 2 settings enable_optimize_predicate_expression = 0" | grep -o "Join\|Filter column: and(notEquals(number, 1), notEquals(number, 2))\|Filter column: and(notEquals(b, 2), notEquals(b, 1))" echo "> (analyzer) one condition of filter is pushed down before INNER JOIN" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=1 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=1 -q " explain actions = 1 select number as a, r.b from numbers(4) as l any inner join ( select number + 2 as b from numbers(3) @@ -274,12 +274,12 @@ $CLICKHOUSE_CLIENT -q " echo "> function calculation should be done after sorting and limit (if possible)" echo "> Expression should be divided into two subexpressions and only one of them should be moved after Sorting" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=0 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=0 -q " explain actions = 1 select number as n, sipHash64(n) from numbers(100) order by number + 1 limit 5" | sed 's/^ *//g' | grep -o "^ *\(Expression (.*Before ORDER BY.*)\|Sorting\|FUNCTION \w\+\)" echo "> (analyzer) function calculation should be done after sorting and limit (if possible)" echo "> Expression should be divided into two subexpressions and only one of them should be moved after Sorting" -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=1 -q " +$CLICKHOUSE_CLIENT --enable_analyzer=1 -q " explain actions = 1 select number as n, sipHash64(n) from numbers(100) order by number + 1 limit 5" | sed 's/^ *//g' | grep -o "^ *\(Expression (.*Before ORDER BY.*)\|Sorting\|FUNCTION \w\+\)" echo "> this query should be executed without throwing an exception" 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 2193fc7a8f4..c6620184a43 100644 --- a/tests/queries/0_stateless/01655_plan_optimizations_merge_filters.sql +++ b/tests/queries/0_stateless/01655_plan_optimizations_merge_filters.sql @@ -1,7 +1,7 @@ set query_plan_merge_filters=1; -set allow_experimental_analyzer=1; +set enable_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%'; -set allow_experimental_analyzer=0; +set enable_analyzer=0; 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/01655_plan_optimizations_optimize_read_in_window_order.reference b/tests/queries/0_stateless/01655_plan_optimizations_optimize_read_in_window_order.reference index 7c2753124b3..c8b3c5ca954 100644 --- a/tests/queries/0_stateless/01655_plan_optimizations_optimize_read_in_window_order.reference +++ b/tests/queries/0_stateless/01655_plan_optimizations_optimize_read_in_window_order.reference @@ -1,23 +1,23 @@ Partial sorting plan optimize_read_in_window_order=0 Sort description: n ASC, x ASC - optimize_read_in_window_order=0, allow_experimental_analyzer=1 + optimize_read_in_window_order=0, enable_analyzer=1 Sort description: n ASC, x ASC optimize_read_in_window_order=1 Prefix sort description: n ASC Result sort description: n ASC, x ASC - optimize_read_in_window_order=1, allow_experimental_analyzer=1 + optimize_read_in_window_order=1, enable_analyzer=1 Prefix sort description: __table1.n ASC Result sort description: __table1.n ASC, __table1.x ASC No sorting plan optimize_read_in_window_order=0 Sort description: n ASC, x ASC - optimize_read_in_window_order=0, allow_experimental_analyzer=1 + optimize_read_in_window_order=0, enable_analyzer=1 Sort description: __table1.n ASC, __table1.x ASC optimize_read_in_window_order=1 Prefix sort description: n ASC, x ASC Result sort description: n ASC, x ASC - optimize_read_in_window_order=1, allow_experimental_analyzer=1 + optimize_read_in_window_order=1, enable_analyzer=1 Prefix sort description: __table1.n ASC, __table1.x ASC Result sort description: __table1.n ASC, __table1.x ASC Complex ORDER BY diff --git a/tests/queries/0_stateless/01655_plan_optimizations_optimize_read_in_window_order.sh b/tests/queries/0_stateless/01655_plan_optimizations_optimize_read_in_window_order.sh index 24c8cf5052e..d74ea328d04 100755 --- a/tests/queries/0_stateless/01655_plan_optimizations_optimize_read_in_window_order.sh +++ b/tests/queries/0_stateless/01655_plan_optimizations_optimize_read_in_window_order.sh @@ -19,25 +19,25 @@ $CLICKHOUSE_CLIENT -q "optimize table ${name}_n_x final" echo 'Partial sorting plan' echo ' optimize_read_in_window_order=0' -$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n SETTINGS optimize_read_in_order=0,optimize_read_in_window_order=0,allow_experimental_analyzer=0" | grep -i "sort description" -echo ' optimize_read_in_window_order=0, allow_experimental_analyzer=1' -$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n SETTINGS optimize_read_in_order=0,optimize_read_in_window_order=0,allow_experimental_analyzer=0" | grep -i "sort description" +$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n SETTINGS optimize_read_in_order=0,optimize_read_in_window_order=0,enable_analyzer=0" | grep -i "sort description" +echo ' optimize_read_in_window_order=0, enable_analyzer=1' +$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n SETTINGS optimize_read_in_order=0,optimize_read_in_window_order=0,enable_analyzer=0" | grep -i "sort description" echo ' optimize_read_in_window_order=1' -$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n SETTINGS optimize_read_in_order=1,allow_experimental_analyzer=0" | grep -i "sort description" -echo ' optimize_read_in_window_order=1, allow_experimental_analyzer=1' -$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n SETTINGS optimize_read_in_order=1,allow_experimental_analyzer=1" | grep -i "sort description" +$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n SETTINGS optimize_read_in_order=1,enable_analyzer=0" | grep -i "sort description" +echo ' optimize_read_in_window_order=1, enable_analyzer=1' +$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n SETTINGS optimize_read_in_order=1,enable_analyzer=1" | grep -i "sort description" echo 'No sorting plan' echo ' optimize_read_in_window_order=0' -$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n_x SETTINGS optimize_read_in_order=0,optimize_read_in_window_order=0,allow_experimental_analyzer=0" | grep -i "sort description" -echo ' optimize_read_in_window_order=0, allow_experimental_analyzer=1' -$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n_x SETTINGS optimize_read_in_order=0,optimize_read_in_window_order=0,allow_experimental_analyzer=1" | grep -i "sort description" +$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n_x SETTINGS optimize_read_in_order=0,optimize_read_in_window_order=0,enable_analyzer=0" | grep -i "sort description" +echo ' optimize_read_in_window_order=0, enable_analyzer=1' +$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n_x SETTINGS optimize_read_in_order=0,optimize_read_in_window_order=0,enable_analyzer=1" | grep -i "sort description" echo ' optimize_read_in_window_order=1' -$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n_x SETTINGS optimize_read_in_order=1,allow_experimental_analyzer=0" | grep -i "sort description" -echo ' optimize_read_in_window_order=1, allow_experimental_analyzer=1' -$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n_x SETTINGS optimize_read_in_order=1,allow_experimental_analyzer=1" | grep -i "sort description" +$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n_x SETTINGS optimize_read_in_order=1,enable_analyzer=0" | grep -i "sort description" +echo ' optimize_read_in_window_order=1, enable_analyzer=1' +$CLICKHOUSE_CLIENT -q "explain plan actions=1, description=1 select n, sum(x) OVER (ORDER BY n, x ROWS BETWEEN 100 PRECEDING AND CURRENT ROW) from ${name}_n_x SETTINGS optimize_read_in_order=1,enable_analyzer=1" | grep -i "sort description" echo 'Complex ORDER BY' $CLICKHOUSE_CLIENT -q "CREATE TABLE ${name}_complex (unique1 Int32, unique2 Int32, ten Int32) ENGINE=MergeTree ORDER BY tuple() SETTINGS index_granularity = 8192" diff --git a/tests/queries/0_stateless/01656_test_query_log_factories_info.reference b/tests/queries/0_stateless/01656_test_query_log_factories_info.reference index 47b3133ceca..44531c19ab7 100644 --- a/tests/queries/0_stateless/01656_test_query_log_factories_info.reference +++ b/tests/queries/0_stateless/01656_test_query_log_factories_info.reference @@ -17,7 +17,7 @@ used_functions ['repeat'] arraySort(used_data_type_families) -['Array','Int32','Nullable','String'] +['Int32','Nullable','String'] used_database_engines ['Atomic'] diff --git a/tests/queries/0_stateless/01671_merge_join_and_constants.sql b/tests/queries/0_stateless/01671_merge_join_and_constants.sql index 7a84bd4e97a..a2153bf0093 100644 --- a/tests/queries/0_stateless/01671_merge_join_and_constants.sql +++ b/tests/queries/0_stateless/01671_merge_join_and_constants.sql @@ -1,5 +1,5 @@ SET output_format_pretty_color=1; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS table1; DROP TABLE IF EXISTS table2; diff --git a/tests/queries/0_stateless/01685_ssd_cache_dictionary_complex_key.sh b/tests/queries/0_stateless/01685_ssd_cache_dictionary_complex_key.sh index 5583a9dd5e7..55061b9a643 100755 --- a/tests/queries/0_stateless/01685_ssd_cache_dictionary_complex_key.sh +++ b/tests/queries/0_stateless/01685_ssd_cache_dictionary_complex_key.sh @@ -1,15 +1,11 @@ #!/usr/bin/env bash -# Tags: no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh $CLICKHOUSE_CLIENT -n --query=" - DROP DATABASE IF EXISTS 01685_database_for_cache_dictionary; - CREATE DATABASE 01685_database_for_cache_dictionary; - - CREATE TABLE 01685_database_for_cache_dictionary.complex_key_simple_attributes_source_table + CREATE TABLE complex_key_simple_attributes_source_table ( id UInt64, id_key String, @@ -18,11 +14,11 @@ $CLICKHOUSE_CLIENT -n --query=" ) ENGINE = TinyLog; - INSERT INTO 01685_database_for_cache_dictionary.complex_key_simple_attributes_source_table VALUES(0, 'id_key_0', 'value_0', 'value_second_0'); - INSERT INTO 01685_database_for_cache_dictionary.complex_key_simple_attributes_source_table VALUES(1, 'id_key_1', 'value_1', 'value_second_1'); - INSERT INTO 01685_database_for_cache_dictionary.complex_key_simple_attributes_source_table VALUES(2, 'id_key_2', 'value_2', 'value_second_2'); + INSERT INTO complex_key_simple_attributes_source_table VALUES(0, 'id_key_0', 'value_0', 'value_second_0'); + INSERT INTO complex_key_simple_attributes_source_table VALUES(1, 'id_key_1', 'value_1', 'value_second_1'); + INSERT INTO complex_key_simple_attributes_source_table VALUES(2, 'id_key_2', 'value_2', 'value_second_2'); - CREATE DICTIONARY 01685_database_for_cache_dictionary.cache_dictionary_complex_key_simple_attributes + CREATE DICTIONARY cache_dictionary_complex_key_simple_attributes ( id UInt64, id_key String, @@ -30,32 +26,32 @@ $CLICKHOUSE_CLIENT -n --query=" value_second String DEFAULT 'value_second_default' ) PRIMARY KEY id, id_key - SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'complex_key_simple_attributes_source_table' DB '01685_database_for_cache_dictionary')) + SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'complex_key_simple_attributes_source_table' DB '${CLICKHOUSE_DATABASE}')) LIFETIME(MIN 1 MAX 1000) - LAYOUT(COMPLEX_KEY_SSD_CACHE(BLOCK_SIZE 4096 FILE_SIZE 8192 PATH '$USER_FILES_PATH/0d')); + LAYOUT(COMPLEX_KEY_SSD_CACHE(BLOCK_SIZE 4096 FILE_SIZE 8192 PATH '$USER_FILES_PATH/${CLICKHOUSE_DATABASE}_dic')); SELECT 'Dictionary cache_dictionary_complex_key_simple_attributes'; SELECT 'dictGet existing value'; - SELECT dictGet('01685_database_for_cache_dictionary.cache_dictionary_complex_key_simple_attributes', 'value_first', (number, concat('id_key_', toString(number)))) as value_first, - dictGet('01685_database_for_cache_dictionary.cache_dictionary_complex_key_simple_attributes', 'value_second', (number, concat('id_key_', toString(number)))) as value_second FROM system.numbers LIMIT 3; + SELECT dictGet('cache_dictionary_complex_key_simple_attributes', 'value_first', (number, concat('id_key_', toString(number)))) as value_first, + dictGet('cache_dictionary_complex_key_simple_attributes', 'value_second', (number, concat('id_key_', toString(number)))) as value_second FROM system.numbers LIMIT 3; SELECT 'dictGet with non existing value'; - SELECT dictGet('01685_database_for_cache_dictionary.cache_dictionary_complex_key_simple_attributes', 'value_first', (number, concat('id_key_', toString(number)))) as value_first, - dictGet('01685_database_for_cache_dictionary.cache_dictionary_complex_key_simple_attributes', 'value_second', (number, concat('id_key_', toString(number)))) as value_second FROM system.numbers LIMIT 4; + SELECT dictGet('cache_dictionary_complex_key_simple_attributes', 'value_first', (number, concat('id_key_', toString(number)))) as value_first, + dictGet('cache_dictionary_complex_key_simple_attributes', 'value_second', (number, concat('id_key_', toString(number)))) as value_second FROM system.numbers LIMIT 4; SELECT 'dictGetOrDefault existing value'; - SELECT dictGetOrDefault('01685_database_for_cache_dictionary.cache_dictionary_complex_key_simple_attributes', 'value_first', (number, concat('id_key_', toString(number))), toString('default')) as value_first, - dictGetOrDefault('01685_database_for_cache_dictionary.cache_dictionary_complex_key_simple_attributes', 'value_second', (number, concat('id_key_', toString(number))), toString('default')) as value_second FROM system.numbers LIMIT 3; + SELECT dictGetOrDefault('cache_dictionary_complex_key_simple_attributes', 'value_first', (number, concat('id_key_', toString(number))), toString('default')) as value_first, + dictGetOrDefault('cache_dictionary_complex_key_simple_attributes', 'value_second', (number, concat('id_key_', toString(number))), toString('default')) as value_second FROM system.numbers LIMIT 3; SELECT 'dictGetOrDefault non existing value'; - SELECT dictGetOrDefault('01685_database_for_cache_dictionary.cache_dictionary_complex_key_simple_attributes', 'value_first', (number, concat('id_key_', toString(number))), toString('default')) as value_first, - dictGetOrDefault('01685_database_for_cache_dictionary.cache_dictionary_complex_key_simple_attributes', 'value_second', (number, concat('id_key_', toString(number))), toString('default')) as value_second FROM system.numbers LIMIT 4; + SELECT dictGetOrDefault('cache_dictionary_complex_key_simple_attributes', 'value_first', (number, concat('id_key_', toString(number))), toString('default')) as value_first, + dictGetOrDefault('cache_dictionary_complex_key_simple_attributes', 'value_second', (number, concat('id_key_', toString(number))), toString('default')) as value_second FROM system.numbers LIMIT 4; SELECT 'dictHas'; - SELECT dictHas('01685_database_for_cache_dictionary.cache_dictionary_complex_key_simple_attributes', (number, concat('id_key_', toString(number)))) FROM system.numbers LIMIT 4; + SELECT dictHas('cache_dictionary_complex_key_simple_attributes', (number, concat('id_key_', toString(number)))) FROM system.numbers LIMIT 4; SELECT 'select all values as input stream'; - SELECT * FROM 01685_database_for_cache_dictionary.cache_dictionary_complex_key_simple_attributes ORDER BY id; + SELECT * FROM cache_dictionary_complex_key_simple_attributes ORDER BY id; - DROP DICTIONARY 01685_database_for_cache_dictionary.cache_dictionary_complex_key_simple_attributes; - DROP TABLE 01685_database_for_cache_dictionary.complex_key_simple_attributes_source_table; + DROP DICTIONARY cache_dictionary_complex_key_simple_attributes; + DROP TABLE complex_key_simple_attributes_source_table; - CREATE TABLE 01685_database_for_cache_dictionary.complex_key_complex_attributes_source_table + CREATE TABLE complex_key_complex_attributes_source_table ( id UInt64, id_key String, @@ -64,11 +60,11 @@ $CLICKHOUSE_CLIENT -n --query=" ) ENGINE = TinyLog; - INSERT INTO 01685_database_for_cache_dictionary.complex_key_complex_attributes_source_table VALUES(0, 'id_key_0', 'value_0', 'value_second_0'); - INSERT INTO 01685_database_for_cache_dictionary.complex_key_complex_attributes_source_table VALUES(1, 'id_key_1', 'value_1', NULL); - INSERT INTO 01685_database_for_cache_dictionary.complex_key_complex_attributes_source_table VALUES(2, 'id_key_2', 'value_2', 'value_second_2'); + INSERT INTO complex_key_complex_attributes_source_table VALUES(0, 'id_key_0', 'value_0', 'value_second_0'); + INSERT INTO complex_key_complex_attributes_source_table VALUES(1, 'id_key_1', 'value_1', NULL); + INSERT INTO complex_key_complex_attributes_source_table VALUES(2, 'id_key_2', 'value_2', 'value_second_2'); - CREATE DICTIONARY 01685_database_for_cache_dictionary.cache_dictionary_complex_key_complex_attributes + CREATE DICTIONARY cache_dictionary_complex_key_complex_attributes ( id UInt64, id_key String, @@ -77,29 +73,28 @@ $CLICKHOUSE_CLIENT -n --query=" value_second Nullable(String) DEFAULT 'value_second_default' ) PRIMARY KEY id, id_key - SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'complex_key_complex_attributes_source_table' DB '01685_database_for_cache_dictionary')) + SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'complex_key_complex_attributes_source_table' DB '${CLICKHOUSE_DATABASE}')) LIFETIME(MIN 1 MAX 1000) LAYOUT(COMPLEX_KEY_SSD_CACHE(BLOCK_SIZE 4096 FILE_SIZE 8192 PATH '$USER_FILES_PATH/1d')); SELECT 'Dictionary cache_dictionary_complex_key_complex_attributes'; SELECT 'dictGet existing value'; - SELECT dictGet('01685_database_for_cache_dictionary.cache_dictionary_complex_key_complex_attributes', 'value_first', (number, concat('id_key_', toString(number)))) as value_first, - dictGet('01685_database_for_cache_dictionary.cache_dictionary_complex_key_complex_attributes', 'value_second', (number, concat('id_key_', toString(number)))) as value_second FROM system.numbers LIMIT 3; + SELECT dictGet('cache_dictionary_complex_key_complex_attributes', 'value_first', (number, concat('id_key_', toString(number)))) as value_first, + dictGet('cache_dictionary_complex_key_complex_attributes', 'value_second', (number, concat('id_key_', toString(number)))) as value_second FROM system.numbers LIMIT 3; SELECT 'dictGet with non existing value'; - SELECT dictGet('01685_database_for_cache_dictionary.cache_dictionary_complex_key_complex_attributes', 'value_first', (number, concat('id_key_', toString(number)))) as value_first, - dictGet('01685_database_for_cache_dictionary.cache_dictionary_complex_key_complex_attributes', 'value_second', (number, concat('id_key_', toString(number)))) as value_second FROM system.numbers LIMIT 4; + SELECT dictGet('cache_dictionary_complex_key_complex_attributes', 'value_first', (number, concat('id_key_', toString(number)))) as value_first, + dictGet('cache_dictionary_complex_key_complex_attributes', 'value_second', (number, concat('id_key_', toString(number)))) as value_second FROM system.numbers LIMIT 4; SELECT 'dictGetOrDefault existing value'; - SELECT dictGetOrDefault('01685_database_for_cache_dictionary.cache_dictionary_complex_key_complex_attributes', 'value_first', (number, concat('id_key_', toString(number))), toString('default')) as value_first, - dictGetOrDefault('01685_database_for_cache_dictionary.cache_dictionary_complex_key_complex_attributes', 'value_second', (number, concat('id_key_', toString(number))), toString('default')) as value_second FROM system.numbers LIMIT 3; + SELECT dictGetOrDefault('cache_dictionary_complex_key_complex_attributes', 'value_first', (number, concat('id_key_', toString(number))), toString('default')) as value_first, + dictGetOrDefault('cache_dictionary_complex_key_complex_attributes', 'value_second', (number, concat('id_key_', toString(number))), toString('default')) as value_second FROM system.numbers LIMIT 3; SELECT 'dictGetOrDefault non existing value'; - SELECT dictGetOrDefault('01685_database_for_cache_dictionary.cache_dictionary_complex_key_complex_attributes', 'value_first', (number, concat('id_key_', toString(number))), toString('default')) as value_first, - dictGetOrDefault('01685_database_for_cache_dictionary.cache_dictionary_complex_key_complex_attributes', 'value_second', (number, concat('id_key_', toString(number))), toString('default')) as value_second FROM system.numbers LIMIT 4; + SELECT dictGetOrDefault('cache_dictionary_complex_key_complex_attributes', 'value_first', (number, concat('id_key_', toString(number))), toString('default')) as value_first, + dictGetOrDefault('cache_dictionary_complex_key_complex_attributes', 'value_second', (number, concat('id_key_', toString(number))), toString('default')) as value_second FROM system.numbers LIMIT 4; SELECT 'dictHas'; - SELECT dictHas('01685_database_for_cache_dictionary.cache_dictionary_complex_key_complex_attributes', (number, concat('id_key_', toString(number)))) FROM system.numbers LIMIT 4; + SELECT dictHas('cache_dictionary_complex_key_complex_attributes', (number, concat('id_key_', toString(number)))) FROM system.numbers LIMIT 4; SELECT 'select all values as input stream'; - SELECT * FROM 01685_database_for_cache_dictionary.cache_dictionary_complex_key_complex_attributes ORDER BY id; + SELECT * FROM cache_dictionary_complex_key_complex_attributes ORDER BY id; - DROP DICTIONARY 01685_database_for_cache_dictionary.cache_dictionary_complex_key_complex_attributes; - DROP TABLE 01685_database_for_cache_dictionary.complex_key_complex_attributes_source_table; - - DROP DATABASE 01685_database_for_cache_dictionary;" + DROP DICTIONARY cache_dictionary_complex_key_complex_attributes; + DROP TABLE complex_key_complex_attributes_source_table; +" diff --git a/tests/queries/0_stateless/01710_aggregate_projection_with_normalized_states.reference b/tests/queries/0_stateless/01710_aggregate_projection_with_normalized_states.reference index 25aa9dc5dec..37993873983 100644 --- a/tests/queries/0_stateless/01710_aggregate_projection_with_normalized_states.reference +++ b/tests/queries/0_stateless/01710_aggregate_projection_with_normalized_states.reference @@ -1,2 +1,3 @@ 3 950 990 500 2000 +[950] [999] diff --git a/tests/queries/0_stateless/01710_aggregate_projection_with_normalized_states.sql b/tests/queries/0_stateless/01710_aggregate_projection_with_normalized_states.sql index e023c0991b3..956bf3711a2 100644 --- a/tests/queries/0_stateless/01710_aggregate_projection_with_normalized_states.sql +++ b/tests/queries/0_stateless/01710_aggregate_projection_with_normalized_states.sql @@ -9,7 +9,8 @@ CREATE TABLE r ( s Int64, PROJECTION p (SELECT a, quantilesTimingMerge(0.5, 0.95, 0.99)(q), sum(s) GROUP BY a) -) Engine=SummingMergeTree order by (x, a); +) Engine=SummingMergeTree order by (x, a) +SETTINGS deduplicate_merge_projection_mode = 'drop'; -- should set it to rebuild once projection is supported with SummingMergeTree insert into r select number%100 x, @@ -28,4 +29,6 @@ FROM cluster('test_cluster_two_shards', currentDatabase(), r) WHERE a = 'x' settings prefer_localhost_replica=0; +SELECT quantilesTimingMerge(0.95)(q), quantilesTimingMerge(toInt64(1))(q) FROM remote('127.0.0.{1,2}', currentDatabase(), r); + DROP TABLE r; diff --git a/tests/queries/0_stateless/01710_projection_vertical_merges.sql b/tests/queries/0_stateless/01710_projection_vertical_merges.sql index 0f80d659e92..0d745e44b10 100644 --- a/tests/queries/0_stateless/01710_projection_vertical_merges.sql +++ b/tests/queries/0_stateless/01710_projection_vertical_merges.sql @@ -1,4 +1,5 @@ --- Tags: long +-- Tags: long, no-parallel +-- set no-parallel tag is to prevent timeout of this test drop table if exists t; diff --git a/tests/queries/0_stateless/01721_join_implicit_cast_long.sql.j2 b/tests/queries/0_stateless/01721_join_implicit_cast_long.sql.j2 index db004c13d96..3c2fa9877db 100644 --- a/tests/queries/0_stateless/01721_join_implicit_cast_long.sql.j2 +++ b/tests/queries/0_stateless/01721_join_implicit_cast_long.sql.j2 @@ -37,8 +37,8 @@ SELECT a, t1.a, t2.a FROM t1 FULL JOIN t2 USING (a) ORDER BY (t1.a, t2.a); {{ is SELECT '= left ='; SELECT a, t1.a, t2.a FROM t1 LEFT JOIN t2 USING (a) ORDER BY (t1.a, t2.a); SELECT '= right ='; -SELECT a, t1.a, t2.a FROM t1 RIGHT JOIN t2 USING (a) ORDER BY (t1.a, t2.a) SETTINGS allow_experimental_analyzer = 0; {{ is_implemented(join_algorithm) }} -SELECT a, t1.a, t2.a FROM t1 RIGHT JOIN t2 USING (a) ORDER BY (t1.a, t2.a) SETTINGS allow_experimental_analyzer = 1; {{ is_implemented(join_algorithm) }} +SELECT a, t1.a, t2.a FROM t1 RIGHT JOIN t2 USING (a) ORDER BY (t1.a, t2.a) SETTINGS enable_analyzer = 0; {{ is_implemented(join_algorithm) }} +SELECT a, t1.a, t2.a FROM t1 RIGHT JOIN t2 USING (a) ORDER BY (t1.a, t2.a) SETTINGS enable_analyzer = 1; {{ is_implemented(join_algorithm) }} SELECT '= inner ='; SELECT a, t1.a, t2.a FROM t1 INNER JOIN t2 USING (a) ORDER BY (t1.a, t2.a); @@ -119,8 +119,8 @@ SELECT a, t1.a, t2.a FROM t1 FULL JOIN t2 USING (a) ORDER BY (t1.a, t2.a); {{ is SELECT '= left ='; SELECT a, t1.a, t2.a FROM t1 LEFT JOIN t2 USING (a) ORDER BY (t1.a, t2.a); SELECT '= right ='; -SELECT a, t1.a, t2.a FROM t1 RIGHT JOIN t2 USING (a) ORDER BY (t1.a, t2.a) SETTINGS allow_experimental_analyzer = 0; {{ is_implemented(join_algorithm) }} -SELECT a, t1.a, t2.a FROM t1 RIGHT JOIN t2 USING (a) ORDER BY (t1.a, t2.a) SETTINGS allow_experimental_analyzer = 1; {{ is_implemented(join_algorithm) }} +SELECT a, t1.a, t2.a FROM t1 RIGHT JOIN t2 USING (a) ORDER BY (t1.a, t2.a) SETTINGS enable_analyzer = 0; {{ is_implemented(join_algorithm) }} +SELECT a, t1.a, t2.a FROM t1 RIGHT JOIN t2 USING (a) ORDER BY (t1.a, t2.a) SETTINGS enable_analyzer = 1; {{ is_implemented(join_algorithm) }} SELECT '= inner ='; SELECT a, t1.a, t2.a FROM t1 INNER JOIN t2 USING (a) ORDER BY (t1.a, t2.a); @@ -166,8 +166,8 @@ SELECT '= types ='; SELECT any(toTypeName(a)) == 'Nullable(Int32)' AND any(toTypeName(t2.a)) == 'Nullable(Int32)' FROM t1 FULL JOIN t2 USING (a); {{ is_implemented(join_algorithm) }} SELECT any(toTypeName(a)) == 'Int32' AND any(toTypeName(t2.a)) == 'Nullable(Int32)' FROM t1 LEFT JOIN t2 USING (a); -SELECT any(toTypeName(a)) == 'Nullable(Int32)' AND any(toTypeName(t2.a)) == 'Int32' FROM t1 RIGHT JOIN t2 USING (a) SETTINGS allow_experimental_analyzer = 0; {{ is_implemented(join_algorithm) }} -SELECT any(toTypeName(a)) == 'Int32' AND any(toTypeName(t2.a)) == 'Int32' FROM t1 RIGHT JOIN t2 USING (a) SETTINGS allow_experimental_analyzer = 1; {{ is_implemented(join_algorithm) }} +SELECT any(toTypeName(a)) == 'Nullable(Int32)' AND any(toTypeName(t2.a)) == 'Int32' FROM t1 RIGHT JOIN t2 USING (a) SETTINGS enable_analyzer = 0; {{ is_implemented(join_algorithm) }} +SELECT any(toTypeName(a)) == 'Int32' AND any(toTypeName(t2.a)) == 'Int32' FROM t1 RIGHT JOIN t2 USING (a) SETTINGS enable_analyzer = 1; {{ is_implemented(join_algorithm) }} SELECT any(toTypeName(a)) == 'Int32' AND any(toTypeName(t2.a)) == 'Int32' FROM t1 INNER JOIN t2 USING (a); diff --git a/tests/queries/0_stateless/01739_index_hint.reference b/tests/queries/0_stateless/01739_index_hint.reference index 21f4edc0049..b921dc6c1f9 100644 --- a/tests/queries/0_stateless/01739_index_hint.reference +++ b/tests/queries/0_stateless/01739_index_hint.reference @@ -35,9 +35,9 @@ SELECT count() FROM XXXX WHERE indexHint(t = toDateTime(0)) SETTINGS optimize_us drop table XXXX; CREATE TABLE XXXX (p Nullable(Int64), k Decimal(76, 39)) ENGINE = MergeTree PARTITION BY toDate(p) ORDER BY k SETTINGS index_granularity = 1, allow_nullable_key = 1; INSERT INTO XXXX FORMAT Values ('2020-09-01 00:01:02', 1), ('2020-09-01 20:01:03', 2), ('2020-09-02 00:01:03', 3); -SELECT count() FROM XXXX WHERE indexHint(p = 1.) SETTINGS optimize_use_implicit_projections = 1, allow_experimental_analyzer=0; +SELECT count() FROM XXXX WHERE indexHint(p = 1.) SETTINGS optimize_use_implicit_projections = 1, enable_analyzer=0; 0 -- TODO: optimize_use_implicit_projections ignores indexHint (with analyzer) because source columns might be aliased. -SELECT count() FROM XXXX WHERE indexHint(p = 1.) SETTINGS optimize_use_implicit_projections = 1, allow_experimental_analyzer=1; +SELECT count() FROM XXXX WHERE indexHint(p = 1.) SETTINGS optimize_use_implicit_projections = 1, enable_analyzer=1; 3 drop table XXXX; diff --git a/tests/queries/0_stateless/01739_index_hint.sql b/tests/queries/0_stateless/01739_index_hint.sql index 1eca65f0892..b208063e7c4 100644 --- a/tests/queries/0_stateless/01739_index_hint.sql +++ b/tests/queries/0_stateless/01739_index_hint.sql @@ -38,8 +38,8 @@ CREATE TABLE XXXX (p Nullable(Int64), k Decimal(76, 39)) ENGINE = MergeTree PART INSERT INTO XXXX FORMAT Values ('2020-09-01 00:01:02', 1), ('2020-09-01 20:01:03', 2), ('2020-09-02 00:01:03', 3); -SELECT count() FROM XXXX WHERE indexHint(p = 1.) SETTINGS optimize_use_implicit_projections = 1, allow_experimental_analyzer=0; +SELECT count() FROM XXXX WHERE indexHint(p = 1.) SETTINGS optimize_use_implicit_projections = 1, enable_analyzer=0; -- TODO: optimize_use_implicit_projections ignores indexHint (with analyzer) because source columns might be aliased. -SELECT count() FROM XXXX WHERE indexHint(p = 1.) SETTINGS optimize_use_implicit_projections = 1, allow_experimental_analyzer=1; +SELECT count() FROM XXXX WHERE indexHint(p = 1.) SETTINGS optimize_use_implicit_projections = 1, enable_analyzer=1; drop table XXXX; diff --git a/tests/queries/0_stateless/01756_optimize_skip_unused_shards_rewrite_in.reference b/tests/queries/0_stateless/01756_optimize_skip_unused_shards_rewrite_in.reference index 28dbb9215a8..fc1ddcae595 100644 --- a/tests/queries/0_stateless/01756_optimize_skip_unused_shards_rewrite_in.reference +++ b/tests/queries/0_stateless/01756_optimize_skip_unused_shards_rewrite_in.reference @@ -49,7 +49,7 @@ order by query; tuple(2) select 'optimize_skip_unused_shards_rewrite_in(2,)'; optimize_skip_unused_shards_rewrite_in(2,) -with (select currentDatabase()) as id_2 select *, ignore(id_2) from dist_01756 where dummy in (2,); +with (select currentDatabase()) as id_2 select *, ignore(id_2) from dist_01756 where dummy in (2); system flush logs; select splitByString('IN', query)[-1] from system.query_log where event_date >= yesterday() and @@ -59,10 +59,10 @@ select splitByString('IN', query)[-1] from system.query_log where query like concat('%', currentDatabase(), '%AS%id_2%') and type = 'QueryFinish' order by query; - tuple(2) + (2) select 'optimize_skip_unused_shards_rewrite_in(0,)'; optimize_skip_unused_shards_rewrite_in(0,) -with (select currentDatabase()) as id_00 select *, ignore(id_00) from dist_01756 where dummy in (0,); +with (select currentDatabase()) as id_00 select *, ignore(id_00) from dist_01756 where dummy in (0); 0 0 system flush logs; select splitByString('IN', query)[-1] from system.query_log where @@ -73,7 +73,7 @@ select splitByString('IN', query)[-1] from system.query_log where query like concat('%', currentDatabase(), '%AS%id_00%') and type = 'QueryFinish' order by query; - tuple(0) + (0) -- signed column select 'signed column'; signed column @@ -103,7 +103,7 @@ errors -- optimize_skip_unused_shards does not support non-constants select * from dist_01756 where dummy in (select * from system.one); -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } -- this is a constant for analyzer -select * from dist_01756 where dummy in (toUInt8(0)) settings allow_experimental_analyzer=0; -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } +select * from dist_01756 where dummy in (toUInt8(0)) settings enable_analyzer=0; -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } -- NOT IN does not supported select * from dist_01756 where dummy not in (0, 2); -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } -- @@ -140,7 +140,7 @@ select * from dist_01756_str where key in ('0', '2'); select * from dist_01756_str where key in (0, 2); 0 -- analyzer does support this -select * from dist_01756_str where key in ('0', Null) settings allow_experimental_analyzer=0; -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } +select * from dist_01756_str where key in ('0', Null) settings enable_analyzer=0; -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } -- select * from dist_01756_str where key in (0, 2); -- { serverError TYPE_MISMATCH } -- select * from dist_01756_str where key in (0, Null); -- { serverError TYPE_MISMATCH } diff --git a/tests/queries/0_stateless/01756_optimize_skip_unused_shards_rewrite_in.sql b/tests/queries/0_stateless/01756_optimize_skip_unused_shards_rewrite_in.sql index 9a1a00cc0a1..0759fb93a44 100644 --- a/tests/queries/0_stateless/01756_optimize_skip_unused_shards_rewrite_in.sql +++ b/tests/queries/0_stateless/01756_optimize_skip_unused_shards_rewrite_in.sql @@ -63,7 +63,7 @@ select splitByString('IN', query)[-1] from system.query_log where order by query; select 'optimize_skip_unused_shards_rewrite_in(2,)'; -with (select currentDatabase()) as id_2 select *, ignore(id_2) from dist_01756 where dummy in (2,); +with (select currentDatabase()) as id_2 select *, ignore(id_2) from dist_01756 where dummy in (2); system flush logs; select splitByString('IN', query)[-1] from system.query_log where event_date >= yesterday() and @@ -75,7 +75,7 @@ select splitByString('IN', query)[-1] from system.query_log where order by query; select 'optimize_skip_unused_shards_rewrite_in(0,)'; -with (select currentDatabase()) as id_00 select *, ignore(id_00) from dist_01756 where dummy in (0,); +with (select currentDatabase()) as id_00 select *, ignore(id_00) from dist_01756 where dummy in (0); system flush logs; select splitByString('IN', query)[-1] from system.query_log where event_date >= yesterday() and @@ -113,7 +113,7 @@ select 'errors'; -- optimize_skip_unused_shards does not support non-constants select * from dist_01756 where dummy in (select * from system.one); -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } -- this is a constant for analyzer -select * from dist_01756 where dummy in (toUInt8(0)) settings allow_experimental_analyzer=0; -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } +select * from dist_01756 where dummy in (toUInt8(0)) settings enable_analyzer=0; -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } -- NOT IN does not supported select * from dist_01756 where dummy not in (0, 2); -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } @@ -146,7 +146,7 @@ create table dist_01756_str as data_01756_str engine=Distributed(test_cluster_tw select * from dist_01756_str where key in ('0', '2'); select * from dist_01756_str where key in (0, 2); -- analyzer does support this -select * from dist_01756_str where key in ('0', Null) settings allow_experimental_analyzer=0; -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } +select * from dist_01756_str where key in ('0', Null) settings enable_analyzer=0; -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } -- select * from dist_01756_str where key in (0, 2); -- { serverError TYPE_MISMATCH } -- select * from dist_01756_str where key in (0, Null); -- { serverError TYPE_MISMATCH } diff --git a/tests/queries/0_stateless/01757_optimize_skip_unused_shards_limit.sql b/tests/queries/0_stateless/01757_optimize_skip_unused_shards_limit.sql index 3853ccb4080..6fcf98d47cb 100644 --- a/tests/queries/0_stateless/01757_optimize_skip_unused_shards_limit.sql +++ b/tests/queries/0_stateless/01757_optimize_skip_unused_shards_limit.sql @@ -21,9 +21,9 @@ select * from dist_01757 where dummy = 0 or dummy = 1 format Null settings optim -- and negative -- disabled for analyzer cause new implementation consider `dummy = 0 and dummy = 1` as constant False. -select * from dist_01757 where dummy = 0 and dummy = 1 settings optimize_skip_unused_shards_limit=1, allow_experimental_analyzer=0; -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } -select * from dist_01757 where dummy = 0 and dummy = 2 and dummy = 3 settings optimize_skip_unused_shards_limit=1, allow_experimental_analyzer=0; -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } -select * from dist_01757 where dummy = 0 and dummy = 2 and dummy = 3 settings optimize_skip_unused_shards_limit=2, allow_experimental_analyzer=0; -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } +select * from dist_01757 where dummy = 0 and dummy = 1 settings optimize_skip_unused_shards_limit=1, enable_analyzer=0; -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } +select * from dist_01757 where dummy = 0 and dummy = 2 and dummy = 3 settings optimize_skip_unused_shards_limit=1, enable_analyzer=0; -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } +select * from dist_01757 where dummy = 0 and dummy = 2 and dummy = 3 settings optimize_skip_unused_shards_limit=2, enable_analyzer=0; -- { serverError UNABLE_TO_SKIP_UNUSED_SHARDS } -- and select * from dist_01757 where dummy = 0 and dummy = 1 settings optimize_skip_unused_shards_limit=2; 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 a7f04921f1f..c6bccde8590 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 @@ -1,5 +1,3 @@ --- Tags: no-parallel - DROP TABLE IF EXISTS ddl_dictonary_test_source; CREATE TABLE ddl_dictonary_test_source ( diff --git a/tests/queries/0_stateless/01763_filter_push_down_bugs.sql b/tests/queries/0_stateless/01763_filter_push_down_bugs.sql index 367baef142b..aa610e19c12 100644 --- a/tests/queries/0_stateless/01763_filter_push_down_bugs.sql +++ b/tests/queries/0_stateless/01763_filter_push_down_bugs.sql @@ -57,13 +57,13 @@ EXPLAIN indexes=1 SELECT id, delete_time FROM t1 CROSS JOIN ( SELECT delete_time FROM t2 -) AS d WHERE create_time < delete_time AND id = 101 SETTINGS allow_experimental_analyzer=0; +) AS d WHERE create_time < delete_time AND id = 101 SETTINGS enable_analyzer=0; EXPLAIN indexes=1 SELECT id, delete_time FROM t1 CROSS JOIN ( SELECT delete_time FROM t2 -) AS d WHERE create_time < delete_time AND id = 101 SETTINGS allow_experimental_analyzer=1; +) AS d WHERE create_time < delete_time AND id = 101 SETTINGS enable_analyzer=1; DROP TABLE IF EXISTS t1; DROP TABLE IF EXISTS t2; diff --git a/tests/queries/0_stateless/01764_table_function_dictionary.sql b/tests/queries/0_stateless/01764_table_function_dictionary.sql index 76e7213b367..e37f8d2a290 100644 --- a/tests/queries/0_stateless/01764_table_function_dictionary.sql +++ b/tests/queries/0_stateless/01764_table_function_dictionary.sql @@ -1,5 +1,3 @@ --- Tags: no-parallel - DROP TABLE IF EXISTS table_function_dictionary_source_table; CREATE TABLE table_function_dictionary_source_table ( @@ -18,7 +16,7 @@ CREATE DICTIONARY table_function_dictionary_test_dictionary value UInt64 DEFAULT 0 ) PRIMARY KEY id -SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'table_function_dictionary_source_table')) +SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' DATABASE currentDatabase() TABLE 'table_function_dictionary_source_table')) LAYOUT(DIRECT()); SELECT * FROM dictionary('table_function_dictionary_test_dictionary'); diff --git a/tests/queries/0_stateless/01786_explain_merge_tree.sh b/tests/queries/0_stateless/01786_explain_merge_tree.sh index e3b28acdc41..828012f56bc 100755 --- a/tests/queries/0_stateless/01786_explain_merge_tree.sh +++ b/tests/queries/0_stateless/01786_explain_merge_tree.sh @@ -7,7 +7,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) for i in $(seq 0 1) do - CH_CLIENT="$CLICKHOUSE_CLIENT --optimize_move_to_prewhere=1 --convert_query_to_cnf=0 --optimize_read_in_order=1 --allow_experimental_analyzer=$i" + CH_CLIENT="$CLICKHOUSE_CLIENT --optimize_move_to_prewhere=1 --convert_query_to_cnf=0 --optimize_read_in_order=1 --enable_analyzer=$i" $CH_CLIENT -q "drop table if exists test_index" $CH_CLIENT -q "drop table if exists idx" diff --git a/tests/queries/0_stateless/01811_storage_buffer_flush_parameters.sh b/tests/queries/0_stateless/01811_storage_buffer_flush_parameters.sh new file mode 100755 index 00000000000..6a5949741ab --- /dev/null +++ b/tests/queries/0_stateless/01811_storage_buffer_flush_parameters.sh @@ -0,0 +1,47 @@ +#!/usr/bin/env bash + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +function wait_with_limit() +{ + local limit=$1 && shift + local expr=$1 && shift + + for ((i = 0; i < limit; ++i)); do + if eval "$expr"; then + break + fi + sleep 1 + done +} + +$CLICKHOUSE_CLIENT -nm -q " + drop table if exists data_01811; + drop table if exists buffer_01811; + + + create table data_01811 (key Int) Engine=Memory(); + /* Buffer with flush_rows=1000 */ + create table buffer_01811 (key Int) Engine=Buffer(currentDatabase(), data_01811, + /* num_layers= */ 1, + /* min_time= */ 1, /* max_time= */ 86400, + /* min_rows= */ 1e9, /* max_rows= */ 1e6, + /* min_bytes= */ 0, /* max_bytes= */ 4e6, + /* flush_time= */ 86400, /* flush_rows= */ 10, /* flush_bytes= */0 + ); + + insert into buffer_01811 select * from numbers(10); + insert into buffer_01811 select * from numbers(10); +" + +# wait for background buffer flush +wait_with_limit 30 '[[ $($CLICKHOUSE_CLIENT -q "select count() from data_01811") -gt 0 ]]' + +$CLICKHOUSE_CLIENT -nm -q "select count() from data_01811" + +$CLICKHOUSE_CLIENT -nm -q " + drop table buffer_01811; + drop table data_01811; +" diff --git a/tests/queries/0_stateless/01811_storage_buffer_flush_parameters.sql b/tests/queries/0_stateless/01811_storage_buffer_flush_parameters.sql deleted file mode 100644 index dac68ad4ae8..00000000000 --- a/tests/queries/0_stateless/01811_storage_buffer_flush_parameters.sql +++ /dev/null @@ -1,22 +0,0 @@ -drop table if exists data_01811; -drop table if exists buffer_01811; - -create table data_01811 (key Int) Engine=Memory(); -/* Buffer with flush_rows=1000 */ -create table buffer_01811 (key Int) Engine=Buffer(currentDatabase(), data_01811, - /* num_layers= */ 1, - /* min_time= */ 1, /* max_time= */ 86400, - /* min_rows= */ 1e9, /* max_rows= */ 1e6, - /* min_bytes= */ 0, /* max_bytes= */ 4e6, - /* flush_time= */ 86400, /* flush_rows= */ 10, /* flush_bytes= */0 -); - -insert into buffer_01811 select * from numbers(10); -insert into buffer_01811 select * from numbers(10); - --- wait for background buffer flush -select sleep(3) format Null; -select count() from data_01811; - -drop table buffer_01811; -drop table data_01811; diff --git a/tests/queries/0_stateless/01823_explain_json.sh b/tests/queries/0_stateless/01823_explain_json.sh index 39128773069..356a317ae57 100755 --- a/tests/queries/0_stateless/01823_explain_json.sh +++ b/tests/queries/0_stateless/01823_explain_json.sh @@ -6,7 +6,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh opts=( - "--allow_experimental_analyzer=1" + "--enable_analyzer=1" ) $CLICKHOUSE_CLIENT "${opts[@]}" -q "EXPLAIN json = 1, description = 0 SELECT 1 UNION ALL SELECT 2 FORMAT TSVRaw" echo "--------" diff --git a/tests/queries/0_stateless/01852_multiple_joins_with_union_join.sql b/tests/queries/0_stateless/01852_multiple_joins_with_union_join.sql index 8c6937eb581..4387a697a83 100644 --- a/tests/queries/0_stateless/01852_multiple_joins_with_union_join.sql +++ b/tests/queries/0_stateless/01852_multiple_joins_with_union_join.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS v1; DROP TABLE IF EXISTS v2; diff --git a/tests/queries/0_stateless/01872_functions_to_subcolumns_analyzer.sql b/tests/queries/0_stateless/01872_functions_to_subcolumns_analyzer.sql index b544f6829cf..032d83890ec 100644 --- a/tests/queries/0_stateless/01872_functions_to_subcolumns_analyzer.sql +++ b/tests/queries/0_stateless/01872_functions_to_subcolumns_analyzer.sql @@ -1,6 +1,6 @@ DROP TABLE IF EXISTS t_func_to_subcolumns; -SET allow_experimental_analyzer = 1; +SET enable_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)) diff --git a/tests/queries/0_stateless/01890_cross_join_explain_crash.sql b/tests/queries/0_stateless/01890_cross_join_explain_crash.sql index bb2bc606870..79aea3884b7 100644 --- a/tests/queries/0_stateless/01890_cross_join_explain_crash.sql +++ b/tests/queries/0_stateless/01890_cross_join_explain_crash.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET joined_subquery_requires_alias = 0; select * FROM (SELECT 1), (SELECT 1), (SELECT 1); diff --git a/tests/queries/0_stateless/01903_correct_block_size_prediction_with_default.reference b/tests/queries/0_stateless/01903_correct_block_size_prediction_with_default.reference index b70a1cb7c75..2c66db91737 100644 --- a/tests/queries/0_stateless/01903_correct_block_size_prediction_with_default.reference +++ b/tests/queries/0_stateless/01903_correct_block_size_prediction_with_default.reference @@ -1,3 +1,9 @@ 8 +8 +1 4 4 +1 +4 +4 +1 diff --git a/tests/queries/0_stateless/01903_correct_block_size_prediction_with_default.sh b/tests/queries/0_stateless/01903_correct_block_size_prediction_with_default.sh new file mode 100755 index 00000000000..1482730af2c --- /dev/null +++ b/tests/queries/0_stateless/01903_correct_block_size_prediction_with_default.sh @@ -0,0 +1,37 @@ +#!/usr/bin/env bash +# Tags: no-random-merge-tree-settings, no-random-settings + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +sql="toUInt16OrNull(arrayFirst((v, k) -> (k = '4Id'), arr[2], arr[1]))" + +# Create the table and fill it +$CLICKHOUSE_CLIENT -n --query=" + CREATE TABLE test_extract(str String, arr Array(Array(String)) ALIAS extractAllGroupsHorizontal(str, '\\W(\\w+)=(\"[^\"]*?\"|[^\",}]*)')) ENGINE=MergeTree() PARTITION BY tuple() ORDER BY tuple(); + INSERT INTO test_extract (str) WITH range(8) as range_arr, arrayMap(x-> concat(toString(x),'Id'), range_arr) as key, arrayMap(x -> rand() % 8, range_arr) as val, arrayStringConcat(arrayMap((x,y) -> concat(x,'=',toString(y)), key, val),',') as str SELECT str FROM numbers(500000); + ALTER TABLE test_extract ADD COLUMN 15Id Nullable(UInt16) DEFAULT $sql;" + +function test() +{ + # Execute two queries and compare if they have similar memory usage: + # The first query uses the default column value, while the second explicitly uses the same SQL as the default value. + # Follow https://github.com/ClickHouse/ClickHouse/issues/17317 for more info about the issue + where=$1 + + uuid_1=$(cat /proc/sys/kernel/random/uuid) + $CLICKHOUSE_CLIENT --query="SELECT uniq(15Id) FROM test_extract $where SETTINGS max_threads=1" --query_id=$uuid_1 + uuid_2=$(cat /proc/sys/kernel/random/uuid) + $CLICKHOUSE_CLIENT --query="SELECT uniq($sql) FROM test_extract $where SETTINGS max_threads=1" --query_id=$uuid_2 + $CLICKHOUSE_CLIENT -n --query=" + SYSTEM FLUSH LOGS; + WITH memory_1 AS (SELECT memory_usage FROM system.query_log WHERE current_database = currentDatabase() AND query_id='$uuid_1' AND type = 'QueryFinish' as memory_1), + memory_2 AS (SELECT memory_usage FROM system.query_log WHERE current_database = currentDatabase() AND query_id='$uuid_2' AND type = 'QueryFinish' as memory_2) + SELECT memory_1.memory_usage <= 1.2 * memory_2.memory_usage OR + memory_2.memory_usage <= 1.2 * memory_1.memory_usage FROM memory_1, memory_2;" +} + +test "" +test "PREWHERE 15Id < 4" +test "WHERE 15Id < 4" diff --git a/tests/queries/0_stateless/01903_correct_block_size_prediction_with_default.sql b/tests/queries/0_stateless/01903_correct_block_size_prediction_with_default.sql deleted file mode 100644 index 2eec08635eb..00000000000 --- a/tests/queries/0_stateless/01903_correct_block_size_prediction_with_default.sql +++ /dev/null @@ -1,13 +0,0 @@ --- Tags: no-random-merge-tree-settings - -CREATE TABLE test_extract(str String, arr Array(Array(String)) ALIAS extractAllGroupsHorizontal(str, '\\W(\\w+)=("[^"]*?"|[^",}]*)')) ENGINE=MergeTree() PARTITION BY tuple() ORDER BY tuple(); - -INSERT INTO test_extract (str) WITH range(8) as range_arr, arrayMap(x-> concat(toString(x),'Id'), range_arr) as key, arrayMap(x -> rand() % 8, range_arr) as val, arrayStringConcat(arrayMap((x,y) -> concat(x,'=',toString(y)), key, val),',') as str SELECT str FROM numbers(500000); - -ALTER TABLE test_extract ADD COLUMN `15Id` Nullable(UInt16) DEFAULT toUInt16OrNull(arrayFirst((v, k) -> (k = '4Id'), arr[2], arr[1])); - -SELECT uniq(15Id) FROM test_extract SETTINGS max_threads=1, max_memory_usage=100000000; - -SELECT uniq(15Id) FROM test_extract PREWHERE 15Id < 4 SETTINGS max_threads=1, max_memory_usage=100000000; - -SELECT uniq(15Id) FROM test_extract WHERE 15Id < 4 SETTINGS max_threads=1, max_memory_usage=100000000; diff --git a/tests/queries/0_stateless/01913_names_of_tuple_literal.sql b/tests/queries/0_stateless/01913_names_of_tuple_literal.sql index 879f4c91587..d6dda4fda9b 100644 --- a/tests/queries/0_stateless/01913_names_of_tuple_literal.sql +++ b/tests/queries/0_stateless/01913_names_of_tuple_literal.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT ((1, 2), (2, 3), (3, 4)) FORMAT TSVWithNames; SELECT ((1, 2), (2, 3), (3, 4)) FORMAT TSVWithNames SETTINGS legacy_column_name_of_tuple_literal = 1; diff --git a/tests/queries/0_stateless/01925_join_materialized_columns.sql b/tests/queries/0_stateless/01925_join_materialized_columns.sql index abb6fda3bfb..ce71cfed8a3 100644 --- a/tests/queries/0_stateless/01925_join_materialized_columns.sql +++ b/tests/queries/0_stateless/01925_join_materialized_columns.sql @@ -32,8 +32,8 @@ SELECT t1.dt, t2.dt FROM t1 JOIN t2 ON t1.foo = t2.bar ORDER BY t1.dt; SELECT '-'; SELECT * FROM t1 ALL JOIN t2 ON t1.dt = t2.dt ORDER BY t1.time, t2.time; SELECT '-'; -SELECT * FROM t1 ALL JOIN t2 USING (dt) ORDER BY t1.time, t2.time settings allow_experimental_analyzer=0; -SELECT * FROM t1 ALL JOIN t2 USING (dt) ORDER BY t1.time, t2.time settings allow_experimental_analyzer=1; +SELECT * FROM t1 ALL JOIN t2 USING (dt) ORDER BY t1.time, t2.time settings enable_analyzer=0; +SELECT * FROM t1 ALL JOIN t2 USING (dt) ORDER BY t1.time, t2.time settings enable_analyzer=1; SELECT '-'; SELECT * FROM t1 JOIN t2 ON t1.dt1 = t2.dt2 ORDER BY t1.time, t1.dimension_1, t2.time, t2.dimension_2; SELECT '-'; @@ -53,5 +53,5 @@ SELECT t1.time as talias FROM t1 JOIN t2 ON talias = t2.time_alias; SELECT t2.time as talias FROM t1 JOIN t2 ON t1.time = talias; SELECT t2.time as talias FROM t1 JOIN t2 ON t1.time_alias = talias; SELECT time as talias FROM t1 JOIN t2 ON t1.time = talias; -- { serverError AMBIGUOUS_COLUMN_NAME, INVALID_JOIN_ON_EXPRESSION } -SELECT time as talias FROM t1 JOIN t2 ON talias = t2.time settings allow_experimental_analyzer=0; -- { serverError AMBIGUOUS_COLUMN_NAME } -SELECT time as talias FROM t1 JOIN t2 ON talias = t2.time settings allow_experimental_analyzer=1; +SELECT time as talias FROM t1 JOIN t2 ON talias = t2.time settings enable_analyzer=0; -- { serverError AMBIGUOUS_COLUMN_NAME } +SELECT time as talias FROM t1 JOIN t2 ON talias = t2.time settings enable_analyzer=1; diff --git a/tests/queries/0_stateless/01925_test_storage_merge_aliases_analyzer.sql b/tests/queries/0_stateless/01925_test_storage_merge_aliases_analyzer.sql index 31035aa80cd..28b9c8650af 100644 --- a/tests/queries/0_stateless/01925_test_storage_merge_aliases_analyzer.sql +++ b/tests/queries/0_stateless/01925_test_storage_merge_aliases_analyzer.sql @@ -1,7 +1,7 @@ -- Tags: no-parallel drop table if exists merge; -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; create table merge ( dt Date, diff --git a/tests/queries/0_stateless/01936_three_parts_identifiers_in_wrong_places.sql b/tests/queries/0_stateless/01936_three_parts_identifiers_in_wrong_places.sql index f344b7007d0..4efbb461501 100644 --- a/tests/queries/0_stateless/01936_three_parts_identifiers_in_wrong_places.sql +++ b/tests/queries/0_stateless/01936_three_parts_identifiers_in_wrong_places.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT dictGet(t.nest.a, concat(currentDatabase(), '.dict.dict'), 's', number) FROM numbers(5); -- { serverError INVALID_IDENTIFIER } diff --git a/tests/queries/0_stateless/01942_dateTimeToSnowflakeID.sql b/tests/queries/0_stateless/01942_dateTimeToSnowflakeID.sql index 945b399157f..0154265ef72 100644 --- a/tests/queries/0_stateless/01942_dateTimeToSnowflakeID.sql +++ b/tests/queries/0_stateless/01942_dateTimeToSnowflakeID.sql @@ -1,5 +1,5 @@ SET session_timezone = 'UTC'; -- disable timezone randomization -SET allow_experimental_analyzer = 1; -- The old path formats the result with different whitespaces +SET enable_analyzer = 1; -- The old path formats the result with different whitespaces SELECT '-- Negative tests'; SELECT dateTimeToSnowflakeID(); -- {serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH} diff --git a/tests/queries/0_stateless/01942_snowflakeIDToDateTime.sql b/tests/queries/0_stateless/01942_snowflakeIDToDateTime.sql index 48316691c71..41e5beb9c16 100644 --- a/tests/queries/0_stateless/01942_snowflakeIDToDateTime.sql +++ b/tests/queries/0_stateless/01942_snowflakeIDToDateTime.sql @@ -1,5 +1,5 @@ SET session_timezone = 'UTC'; -- disable timezone randomization -SET allow_experimental_analyzer = 1; -- The old path formats the result with different whitespaces +SET enable_analyzer = 1; -- The old path formats the result with different whitespaces SELECT '-- Negative tests'; SELECT snowflakeIDToDateTime(); -- {serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH} diff --git a/tests/queries/0_stateless/01943_non_deterministic_order_key.sql b/tests/queries/0_stateless/01943_non_deterministic_order_key.sql index 781ea1b1cc7..94aa423c552 100644 --- a/tests/queries/0_stateless/01943_non_deterministic_order_key.sql +++ b/tests/queries/0_stateless/01943_non_deterministic_order_key.sql @@ -1,3 +1,4 @@ CREATE TABLE a (number UInt64) ENGINE = MergeTree ORDER BY if(now() > toDateTime('2020-06-01 13:31:40'), toInt64(number), -number); -- { serverError BAD_ARGUMENTS } CREATE TABLE b (number UInt64) ENGINE = MergeTree ORDER BY now() > toDateTime(number); -- { serverError BAD_ARGUMENTS } CREATE TABLE c (number UInt64) ENGINE = MergeTree ORDER BY now(); -- { serverError BAD_ARGUMENTS } +CREATE TABLE d (number UInt64) ENGINE = MergeTree ORDER BY now() + 1 + 1 + number; -- { serverError BAD_ARGUMENTS } diff --git a/tests/queries/0_stateless/01945_system_warnings.sh b/tests/queries/0_stateless/01945_system_warnings.sh index 249c3218bcc..63403ce2893 100755 --- a/tests/queries/0_stateless/01945_system_warnings.sh +++ b/tests/queries/0_stateless/01945_system_warnings.sh @@ -1,5 +1,4 @@ #!/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/01952_optimize_distributed_group_by_sharding_key.reference b/tests/queries/0_stateless/01952_optimize_distributed_group_by_sharding_key.reference index 5acefdb365e..e786532f25a 100644 --- a/tests/queries/0_stateless/01952_optimize_distributed_group_by_sharding_key.reference +++ b/tests/queries/0_stateless/01952_optimize_distributed_group_by_sharding_key.reference @@ -71,7 +71,7 @@ Expression (Projection) Expression ((Before ORDER BY + (Convert VIEW subquery result to VIEW table structure + (Materialize constants after VIEW subquery + (Projection + Before ORDER BY))))) ReadFromSystemNumbers ReadFromRemote (Read from remote replica) -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; explain select distinct k1 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)); -- not optimized Expression (Project names) Distinct (DISTINCT) diff --git a/tests/queries/0_stateless/01952_optimize_distributed_group_by_sharding_key.sql b/tests/queries/0_stateless/01952_optimize_distributed_group_by_sharding_key.sql index adf55a9dd7f..960fd227a39 100644 --- a/tests/queries/0_stateless/01952_optimize_distributed_group_by_sharding_key.sql +++ b/tests/queries/0_stateless/01952_optimize_distributed_group_by_sharding_key.sql @@ -4,7 +4,7 @@ set optimize_skip_unused_shards=1; set optimize_distributed_group_by_sharding_key=1; set prefer_localhost_replica=1; -set allow_experimental_analyzer = 0; +set enable_analyzer = 0; -- { echo } explain select distinct k1 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)); -- not optimized @@ -17,7 +17,7 @@ explain select distinct k1, k2 from remote('127.{1,2}', view(select 1 k1, 2 k2, explain select distinct on (k1) k2 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)) order by v; -- not optimized explain select distinct on (k1, k2) v from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)) order by v; -- optimized -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; explain select distinct k1 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)); -- not optimized explain select distinct k1, k2 from remote('127.{1,2}', view(select 1 k1, 2 k2, 3 v from numbers(2)), cityHash64(k1, k2)); -- optimized diff --git a/tests/queries/0_stateless/02000_join_on_const.reference b/tests/queries/0_stateless/02000_join_on_const.reference index 848ecedf9e3..3bd1633ce32 100644 --- a/tests/queries/0_stateless/02000_join_on_const.reference +++ b/tests/queries/0_stateless/02000_join_on_const.reference @@ -33,23 +33,23 @@ 2 2 2 2 -- { echoOn } -SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS allow_experimental_analyzer = 1; +SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; 1 0 2 2 -SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS allow_experimental_analyzer = 1; +SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; 2 2 0 3 -SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS allow_experimental_analyzer = 1; +SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; 1 0 2 2 0 3 -SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS allow_experimental_analyzer = 1; +SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; 1 0 2 0 -SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS allow_experimental_analyzer = 1; +SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; 0 2 0 3 -SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS allow_experimental_analyzer = 1; +SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; 1 0 2 0 0 2 diff --git a/tests/queries/0_stateless/02000_join_on_const.sql b/tests/queries/0_stateless/02000_join_on_const.sql index 2c1152e0ae6..da70973ed87 100644 --- a/tests/queries/0_stateless/02000_join_on_const.sql +++ b/tests/queries/0_stateless/02000_join_on_const.sql @@ -56,30 +56,30 @@ SELECT * FROM t1 RIGHT JOIN t2 ON NULL SETTINGS join_algorithm = 'auto'; -- { se SELECT * FROM t1 FULL JOIN t2 ON NULL SETTINGS join_algorithm = 'partial_merge'; -- { serverError INVALID_JOIN_ON_EXPRESSION,NOT_IMPLEMENTED } -- mixing of constant and non-constant expressions in ON is not allowed -SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 == 1 SETTINGS allow_experimental_analyzer = 0; -- { serverError AMBIGUOUS_COLUMN_NAME } -SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 == 1 SETTINGS allow_experimental_analyzer = 1; -SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 == 2 SETTINGS allow_experimental_analyzer = 0; -- { serverError AMBIGUOUS_COLUMN_NAME } -SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 == 2 SETTINGS allow_experimental_analyzer = 1; +SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 == 1 SETTINGS enable_analyzer = 0; -- { serverError AMBIGUOUS_COLUMN_NAME } +SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 == 1 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 == 2 SETTINGS enable_analyzer = 0; -- { serverError AMBIGUOUS_COLUMN_NAME } +SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 == 2 SETTINGS enable_analyzer = 1; -SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 != 1 SETTINGS allow_experimental_analyzer = 0; -- { serverError INVALID_JOIN_ON_EXPRESSION } -SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 != 1 SETTINGS allow_experimental_analyzer = 1; +SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 != 1 SETTINGS enable_analyzer = 0; -- { serverError INVALID_JOIN_ON_EXPRESSION } +SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 != 1 SETTINGS enable_analyzer = 1; SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND NULL; -- { serverError INVALID_JOIN_ON_EXPRESSION } SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 'aaa'; -- { serverError INVALID_JOIN_ON_EXPRESSION,ILLEGAL_TYPE_OF_ARGUMENT } SELECT * FROM t1 JOIN t2 ON 'aaa'; -- { serverError INVALID_JOIN_ON_EXPRESSION } -SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 0 SETTINGS allow_experimental_analyzer = 0; -- { serverError INVALID_JOIN_ON_EXPRESSION } -SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 0 SETTINGS allow_experimental_analyzer = 1; -SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 SETTINGS allow_experimental_analyzer = 0; -- { serverError INVALID_JOIN_ON_EXPRESSION } -SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 SETTINGS allow_experimental_analyzer = 1; +SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 0 SETTINGS enable_analyzer = 0; -- { serverError INVALID_JOIN_ON_EXPRESSION } +SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 0 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 SETTINGS enable_analyzer = 0; -- { serverError INVALID_JOIN_ON_EXPRESSION } +SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 SETTINGS enable_analyzer = 1; -- { echoOn } -SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS allow_experimental_analyzer = 1; -SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS allow_experimental_analyzer = 1; -SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS allow_experimental_analyzer = 1; +SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; -SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS allow_experimental_analyzer = 1; -SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS allow_experimental_analyzer = 1; -SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS allow_experimental_analyzer = 1; +SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; SELECT * FROM (SELECT 1 as a) as t1 INNER JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; SELECT * FROM (SELECT 1 as a) as t1 LEFT JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; @@ -94,27 +94,26 @@ SELECT a + 1 FROM (SELECT 1 as x) as t1 LEFT JOIN ( SELECT 1 AS a ) AS t2 ON TRUE -SETTINGS allow_experimental_analyzer=1, join_use_nulls=1; +SETTINGS enable_analyzer=1, join_use_nulls=1; SELECT a + 1, x + 1, toTypeName(a), toTypeName(x) FROM (SELECT 1 as x) as t1 LEFT JOIN ( SELECT sum(number) as a from numbers(3) GROUP BY NULL) AS t2 ON TRUE -SETTINGS allow_experimental_analyzer=1, join_use_nulls=1; +SETTINGS enable_analyzer=1, join_use_nulls=1; SELECT a + 1, x + 1, toTypeName(a), toTypeName(x) FROM (SELECT 1 as x) as t1 RIGHT JOIN ( SELECT sum(number) as a from numbers(3) GROUP BY NULL) AS t2 ON TRUE -SETTINGS allow_experimental_analyzer=1, join_use_nulls=1; +SETTINGS enable_analyzer=1, join_use_nulls=1; SELECT a + 1, x + 1, toTypeName(a), toTypeName(x) FROM (SELECT 1 as x) as t1 FULL JOIN ( SELECT sum(number) as a from numbers(3) GROUP BY NULL) AS t2 ON TRUE -SETTINGS allow_experimental_analyzer=1, join_use_nulls=1; +SETTINGS enable_analyzer=1, join_use_nulls=1; DROP TABLE IF EXISTS t1; DROP TABLE IF EXISTS t2; - diff --git a/tests/queries/0_stateless/02002_row_level_filter_bug.sh b/tests/queries/0_stateless/02002_row_level_filter_bug.sh index d15a26f48f5..557aa738217 100755 --- a/tests/queries/0_stateless/02002_row_level_filter_bug.sh +++ b/tests/queries/0_stateless/02002_row_level_filter_bug.sh @@ -1,11 +1,12 @@ #!/usr/bin/env bash -# Tags: no-parallel -# Tag no-parallel: create user CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh +TEST_ROLE="${CLICKHOUSE_DATABASE}_role" +TEST_USER="${CLICKHOUSE_DATABASE}_user" +TEST_POLICY="${CLICKHOUSE_DATABASE}_policy" $CLICKHOUSE_CLIENT --query "drop table if exists test_table" $CLICKHOUSE_CLIENT --query "CREATE TABLE test_table @@ -37,26 +38,26 @@ arrayJoin(['AWD','ZZZ']) as team, arrayJoin([3183,3106,0,3130,3108,3126,3109,3107,3182,3180,3129,3128,3125,3266]) as a FROM numbers(600);" -$CLICKHOUSE_CLIENT --query "DROP ROLE IF exists AWD;" -$CLICKHOUSE_CLIENT --query "create role AWD;" -$CLICKHOUSE_CLIENT --query "REVOKE ALL ON *.* FROM AWD;" +$CLICKHOUSE_CLIENT --query "DROP ROLE IF EXISTS ${TEST_ROLE};" +$CLICKHOUSE_CLIENT --query "create role ${TEST_ROLE};" +$CLICKHOUSE_CLIENT --query "REVOKE ALL ON *.* FROM ${TEST_ROLE};" -$CLICKHOUSE_CLIENT --query "DROP USER IF EXISTS AWD_user;" -$CLICKHOUSE_CLIENT --query "CREATE USER AWD_user IDENTIFIED WITH plaintext_password BY 'AWD_pwd' DEFAULT ROLE AWD;" +$CLICKHOUSE_CLIENT --query "DROP USER IF EXISTS ${TEST_USER};" +$CLICKHOUSE_CLIENT --query "CREATE USER ${TEST_USER} IDENTIFIED WITH plaintext_password BY 'AWD_pwd' DEFAULT ROLE ${TEST_ROLE};" -$CLICKHOUSE_CLIENT --query "GRANT SELECT ON test_table TO AWD;" +$CLICKHOUSE_CLIENT --query "GRANT SELECT ON test_table TO ${TEST_ROLE};" -$CLICKHOUSE_CLIENT --query "DROP ROW POLICY IF EXISTS ttt_bu_test_table_AWD ON test_table;" -$CLICKHOUSE_CLIENT --query "CREATE ROW POLICY ttt_bu_test_table_AWD ON test_table FOR SELECT USING team = 'AWD' TO AWD;" +$CLICKHOUSE_CLIENT --query "DROP ROW POLICY IF EXISTS ${TEST_POLICY} ON test_table;" +$CLICKHOUSE_CLIENT --query "CREATE ROW POLICY ${TEST_POLICY} ON test_table FOR SELECT USING team = 'AWD' TO ${TEST_ROLE};" -$CLICKHOUSE_CLIENT --user=AWD_user --password=AWD_pwd --query " +$CLICKHOUSE_CLIENT --user=${TEST_USER} --password=AWD_pwd --query " SELECT count() AS count FROM test_table WHERE t_date = '2021-07-15' AND c = 'aur' AND a=3130; " -$CLICKHOUSE_CLIENT --user=AWD_user --password=AWD_pwd --query " +$CLICKHOUSE_CLIENT --user=${TEST_USER} --password=AWD_pwd --query " SELECT team, a, @@ -70,8 +71,12 @@ GROUP BY t_date; " -$CLICKHOUSE_CLIENT --user=AWD_user --password=AWD_pwd --query " +$CLICKHOUSE_CLIENT --user=${TEST_USER} --password=AWD_pwd --query " SELECT count() AS count FROM test_table WHERE (t_date = '2021-07-15') AND (c = 'aur') AND (a = 313) " + +$CLICKHOUSE_CLIENT --query "DROP ROLE IF EXISTS ${TEST_ROLE};" +$CLICKHOUSE_CLIENT --query "DROP USER IF EXISTS ${TEST_USER};" +$CLICKHOUSE_CLIENT --query "DROP ROW POLICY IF EXISTS ${TEST_POLICY} ON test_table;" diff --git a/tests/queries/0_stateless/02003_memory_limit_in_client.sh b/tests/queries/0_stateless/02003_memory_limit_in_client.sh index 96028f4847a..15cacbff8c5 100755 --- a/tests/queries/0_stateless/02003_memory_limit_in_client.sh +++ b/tests/queries/0_stateless/02003_memory_limit_in_client.sh @@ -4,21 +4,21 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -$CLICKHOUSE_CLIENT --max_memory_usage_in_client=1 -n -q "SELECT arrayMap(x -> range(x), range(number)) FROM numbers(1000) -- { clientError MEMORY_LIMIT_EXCEEDED }" -$CLICKHOUSE_CLIENT --max_memory_usage_in_client=0 -n -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client=1 -q "SELECT arrayMap(x -> range(x), range(number)) FROM numbers(1000) -- { clientError MEMORY_LIMIT_EXCEEDED }" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client=0 -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" -$CLICKHOUSE_CLIENT --max_memory_usage_in_client='5K' -n -q "SELECT arrayMap(x -> range(x), range(number)) FROM numbers(1000) -- { clientError MEMORY_LIMIT_EXCEEDED }" -$CLICKHOUSE_CLIENT --max_memory_usage_in_client='5k' -n -q "SELECT arrayMap(x -> range(x), range(number)) FROM numbers(1000) -- { clientError MEMORY_LIMIT_EXCEEDED }" -$CLICKHOUSE_CLIENT --max_memory_usage_in_client='1M' -n -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" -$CLICKHOUSE_CLIENT --max_memory_usage_in_client='23G' -n -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" -$CLICKHOUSE_CLIENT --max_memory_usage_in_client='11T' -n -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client='5K' -q "SELECT arrayMap(x -> range(x), range(number)) FROM numbers(1000) -- { clientError MEMORY_LIMIT_EXCEEDED }" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client='5k' -q "SELECT arrayMap(x -> range(x), range(number)) FROM numbers(1000) -- { clientError MEMORY_LIMIT_EXCEEDED }" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client='1M' -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client='23G' -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client='11T' -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" -$CLICKHOUSE_CLIENT --max_memory_usage_in_client='2P' -n -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_INPUT_ASSERTION_FAILED" -$CLICKHOUSE_CLIENT --max_memory_usage_in_client='2.1p' -n -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_INPUT_ASSERTION_FAILED" -$CLICKHOUSE_CLIENT --max_memory_usage_in_client='10E' -n -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_INPUT_ASSERTION_FAILED" -$CLICKHOUSE_CLIENT --max_memory_usage_in_client='10.2e' -n -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_INPUT_ASSERTION_FAILED" -$CLICKHOUSE_CLIENT --max_memory_usage_in_client='-1.1T' -n -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_NUMBER" -$CLICKHOUSE_CLIENT --max_memory_usage_in_client='-1' -n -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_NUMBER" -$CLICKHOUSE_CLIENT --max_memory_usage_in_client='1m' -n -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_INPUT_ASSERTION_FAILED" -$CLICKHOUSE_CLIENT --max_memory_usage_in_client='14g' -n -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_INPUT_ASSERTION_FAILED" -$CLICKHOUSE_CLIENT --max_memory_usage_in_client='11t' -n -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_INPUT_ASSERTION_FAILED" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client='2P' -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_INPUT_ASSERTION_FAILED" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client='2.1p' -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_INPUT_ASSERTION_FAILED" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client='10E' -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_INPUT_ASSERTION_FAILED" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client='10.2e' -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_INPUT_ASSERTION_FAILED" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client='-1.1T' -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_NUMBER" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client='-1' -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_NUMBER" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client='1m' -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_INPUT_ASSERTION_FAILED" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client='14g' -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_INPUT_ASSERTION_FAILED" +$CLICKHOUSE_CLIENT --max_memory_usage_in_client='11t' -q "SELECT * FROM (SELECT * FROM system.numbers LIMIT 600000) as num WHERE num.number=60000" 2>&1 | grep -c -F "CANNOT_PARSE_INPUT_ASSERTION_FAILED" diff --git a/tests/queries/0_stateless/02015_column_default_dict_get_identifier.sql b/tests/queries/0_stateless/02015_column_default_dict_get_identifier.sql index 24a3b631388..046d0c42180 100644 --- a/tests/queries/0_stateless/02015_column_default_dict_get_identifier.sql +++ b/tests/queries/0_stateless/02015_column_default_dict_get_identifier.sql @@ -1,9 +1,6 @@ -- Tags: no-parallel -DROP DATABASE IF EXISTS 02015_db; -CREATE DATABASE 02015_db; - -CREATE TABLE 02015_db.test_table +CREATE TABLE test_table ( key_column UInt64, data_column_1 UInt64, @@ -12,9 +9,9 @@ CREATE TABLE 02015_db.test_table ENGINE = MergeTree ORDER BY key_column; -INSERT INTO 02015_db.test_table VALUES (0, 0, 0); +INSERT INTO test_table VALUES (0, 0, 0); -CREATE DICTIONARY 02015_db.test_dictionary +CREATE DICTIONARY test_dictionary ( key_column UInt64 DEFAULT 0, data_column_1 UInt64 DEFAULT 1, @@ -22,20 +19,18 @@ CREATE DICTIONARY 02015_db.test_dictionary ) PRIMARY KEY key_column LAYOUT(DIRECT()) -SOURCE(CLICKHOUSE(DB '02015_db' TABLE 'test_table')); +SOURCE(CLICKHOUSE(DB currentDatabase() TABLE 'test_table')); -CREATE TABLE 02015_db.test_table_default +CREATE TABLE test_table_default ( - data_1 DEFAULT dictGetUInt64('02015_db.test_dictionary', 'data_column_1', toUInt64(0)), - data_2 DEFAULT dictGet(02015_db.test_dictionary, 'data_column_2', toUInt64(0)) + data_1 DEFAULT dictGetUInt64('test_dictionary', 'data_column_1', toUInt64(0)), + data_2 DEFAULT dictGet(test_dictionary, 'data_column_2', toUInt64(0)) ) ENGINE=TinyLog; -INSERT INTO 02015_db.test_table_default(data_1) VALUES (5); -SELECT * FROM 02015_db.test_table_default; +INSERT INTO test_table_default(data_1) VALUES (5); +SELECT * FROM test_table_default; -DROP TABLE 02015_db.test_table_default; -DROP DICTIONARY 02015_db.test_dictionary; -DROP TABLE 02015_db.test_table; - -DROP DATABASE 02015_db; +DROP TABLE test_table_default; +DROP DICTIONARY test_dictionary; +DROP TABLE test_table; diff --git a/tests/queries/0_stateless/02021_create_database_with_comment.sh b/tests/queries/0_stateless/02021_create_database_with_comment.sh index f77397dc482..d87b0794c91 100755 --- a/tests/queries/0_stateless/02021_create_database_with_comment.sh +++ b/tests/queries/0_stateless/02021_create_database_with_comment.sh @@ -20,7 +20,7 @@ function test_db_comments() local ENGINE_NAME="$1" echo "engine : ${ENGINE_NAME}" - $CLICKHOUSE_CLIENT --allow_deprecated_database_ordinary=1 -nm <& /dev/null +$CLICKHOUSE_CLIENT --print-profile-events --profile-events-delay-ms=-1 -q 'select 1; select 1' >& /dev/null echo $? echo 'regression test for overlap profile events snapshots between queries' -$CLICKHOUSE_CLIENT --print-profile-events --profile-events-delay-ms=-1 -n -q 'select 1; select 1' |& grep -F -o '[ 0 ] SelectedRows: 1 (increment)' +$CLICKHOUSE_CLIENT --print-profile-events --profile-events-delay-ms=-1 -q 'select 1; select 1' |& grep -F -o '[ 0 ] SelectedRows: 1 (increment)' echo 'regression test for overlap profile events snapshots between queries (clickhouse-local)' -$CLICKHOUSE_LOCAL --print-profile-events --profile-events-delay-ms=-1 -n -q 'select 1; select 1' |& grep -F -o '[ 0 ] SelectedRows: 1 (increment)' +$CLICKHOUSE_LOCAL --print-profile-events --profile-events-delay-ms=-1 -q 'select 1; select 1' |& grep -F -o '[ 0 ] SelectedRows: 1 (increment)' echo 'print everything' profile_events="$( @@ -35,5 +35,5 @@ profile_events="$( test "$profile_events" -gt 1 && echo OK || echo "FAIL ($profile_events)" echo 'check that ProfileEvents is new for each query' -sleep_function_calls=$($CLICKHOUSE_CLIENT --print-profile-events --profile-events-delay-ms=-1 -n -q 'select sleep(1); select 1' |& grep -c 'SleepFunctionCalls') +sleep_function_calls=$($CLICKHOUSE_CLIENT --print-profile-events --profile-events-delay-ms=-1 -q 'select sleep(1); select 1' |& grep -c 'SleepFunctionCalls') test "$sleep_function_calls" -eq 1 && echo OK || echo "FAIL ($sleep_function_calls)" diff --git a/tests/queries/0_stateless/02099_tsv_raw_format.reference b/tests/queries/0_stateless/02099_tsv_raw_format_1.reference similarity index 94% rename from tests/queries/0_stateless/02099_tsv_raw_format.reference rename to tests/queries/0_stateless/02099_tsv_raw_format_1.reference index de46cf8dff7..3ac175e51f6 100644 --- a/tests/queries/0_stateless/02099_tsv_raw_format.reference +++ b/tests/queries/0_stateless/02099_tsv_raw_format_1.reference @@ -109,5 +109,3 @@ UInt64 String Date 2 \N nSome text -b1cad4eb4be08a40387c9de70d02fcc2 - -b1cad4eb4be08a40387c9de70d02fcc2 - diff --git a/tests/queries/0_stateless/02099_tsv_raw_format.sh b/tests/queries/0_stateless/02099_tsv_raw_format_1.sh similarity index 69% rename from tests/queries/0_stateless/02099_tsv_raw_format.sh rename to tests/queries/0_stateless/02099_tsv_raw_format_1.sh index a69c96ab613..bd1f8731717 100755 --- a/tests/queries/0_stateless/02099_tsv_raw_format.sh +++ b/tests/queries/0_stateless/02099_tsv_raw_format_1.sh @@ -1,6 +1,9 @@ #!/usr/bin/env bash # Tags: long +# This test was split in two due to long runtimes in sanitizers. +# The other part is 02099_tsv_raw_format_2.sh. + CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh @@ -46,16 +49,3 @@ echo 'nSome text' | $CLICKHOUSE_CLIENT -q "INSERT INTO test_nullable_string_0209 $CLICKHOUSE_CLIENT -q "SELECT * FROM test_nullable_string_02099" $CLICKHOUSE_CLIENT -q "DROP TABLE test_nullable_string_02099" - - -$CLICKHOUSE_CLIENT -q "DROP TABLE IF EXISTS test_parallel_parsing_02099" -$CLICKHOUSE_CLIENT -q "CREATE TABLE test_parallel_parsing_02099 (x UInt64, a Array(UInt64), s String) ENGINE=Memory()"; -$CLICKHOUSE_CLIENT -q "SELECT number AS x, range(number % 50) AS a, toString(a) AS s FROM numbers(1000000) FORMAT TSVRaw" | $CLICKHOUSE_CLIENT --input_format_parallel_parsing=0 -q "INSERT INTO test_parallel_parsing_02099 FORMAT TSVRaw" -$CLICKHOUSE_CLIENT -q "SELECT * FROM test_parallel_parsing_02099 ORDER BY x" | md5sum - -$CLICKHOUSE_CLIENT -q "TRUNCATE TABLE test_parallel_parsing_02099" - -$CLICKHOUSE_CLIENT -q "SELECT number AS x, range(number % 50) AS a, toString(a) AS s FROM numbers(1000000) FORMAT TSVRaw" | $CLICKHOUSE_CLIENT --input_format_parallel_parsing=1 -q "INSERT INTO test_parallel_parsing_02099 FORMAT TSVRaw" -$CLICKHOUSE_CLIENT -q "SELECT * FROM test_parallel_parsing_02099 ORDER BY x" | md5sum - -$CLICKHOUSE_CLIENT -q "DROP TABLE test_parallel_parsing_02099" diff --git a/tests/queries/0_stateless/02099_tsv_raw_format_2.reference b/tests/queries/0_stateless/02099_tsv_raw_format_2.reference new file mode 100644 index 00000000000..4682749c21f --- /dev/null +++ b/tests/queries/0_stateless/02099_tsv_raw_format_2.reference @@ -0,0 +1,2 @@ +c8ff17885084035ea1aebd95fee2efb6 - +c8ff17885084035ea1aebd95fee2efb6 - diff --git a/tests/queries/0_stateless/02099_tsv_raw_format_2.sh b/tests/queries/0_stateless/02099_tsv_raw_format_2.sh new file mode 100755 index 00000000000..9f57eea42f2 --- /dev/null +++ b/tests/queries/0_stateless/02099_tsv_raw_format_2.sh @@ -0,0 +1,22 @@ +#!/usr/bin/env bash +# Tags: long + +# This test was split in two due to long runtimes in sanitizers. +# The other part is 02099_tsv_raw_format_1.sh. +# +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +$CLICKHOUSE_CLIENT -q "DROP TABLE IF EXISTS test_parallel_parsing_02099" +$CLICKHOUSE_CLIENT -q "CREATE TABLE test_parallel_parsing_02099 (x UInt64, a Array(UInt64), s String) ENGINE=Memory()"; +$CLICKHOUSE_CLIENT -q "SELECT number AS x, range(number % 50) AS a, toString(a) AS s FROM numbers(100000) FORMAT TSVRaw" | $CLICKHOUSE_CLIENT --input_format_parallel_parsing=0 -q "INSERT INTO test_parallel_parsing_02099 FORMAT TSVRaw" +$CLICKHOUSE_CLIENT -q "SELECT * FROM test_parallel_parsing_02099 ORDER BY x" | md5sum + +$CLICKHOUSE_CLIENT -q "TRUNCATE TABLE test_parallel_parsing_02099" + +$CLICKHOUSE_CLIENT -q "SELECT number AS x, range(number % 50) AS a, toString(a) AS s FROM numbers(100000) FORMAT TSVRaw" | $CLICKHOUSE_CLIENT --input_format_parallel_parsing=1 -q "INSERT INTO test_parallel_parsing_02099 FORMAT TSVRaw" +$CLICKHOUSE_CLIENT -q "SELECT * FROM test_parallel_parsing_02099 ORDER BY x" | md5sum + +$CLICKHOUSE_CLIENT -q "DROP TABLE test_parallel_parsing_02099" + diff --git a/tests/queries/0_stateless/02115_map_contains_analyzer.sql b/tests/queries/0_stateless/02115_map_contains_analyzer.sql index 46e02eca4f0..00285404987 100644 --- a/tests/queries/0_stateless/02115_map_contains_analyzer.sql +++ b/tests/queries/0_stateless/02115_map_contains_analyzer.sql @@ -5,7 +5,7 @@ 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; +SET enable_analyzer = 1; 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; diff --git a/tests/queries/0_stateless/02116_tuple_element_analyzer.sql b/tests/queries/0_stateless/02116_tuple_element_analyzer.sql index 5aeb72c9ee4..ef3729bdc95 100644 --- a/tests/queries/0_stateless/02116_tuple_element_analyzer.sql +++ b/tests/queries/0_stateless/02116_tuple_element_analyzer.sql @@ -4,7 +4,7 @@ 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; +SET enable_analyzer = 1; SELECT t1.1 FROM t_tuple_element; EXPLAIN QUERY TREE dump_tree = 0, dump_ast = 1 SELECT t1.1 FROM t_tuple_element; 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 cfae4fee6c2..32e8b2f4312 100644 --- a/tests/queries/0_stateless/02117_show_create_table_system.reference +++ b/tests/queries/0_stateless/02117_show_create_table_system.reference @@ -52,6 +52,8 @@ CREATE TABLE system.clusters `database_shard_name` String, `database_replica_name` String, `is_active` Nullable(UInt8), + `replication_lag` Nullable(UInt32), + `recovery_time` Nullable(UInt64), `name` String ALIAS cluster ) ENGINE = SystemClusters diff --git a/tests/queries/0_stateless/02122_join_group_by_timeout.reference b/tests/queries/0_stateless/02122_join_group_by_timeout.reference index f314e22e519..6500560e8fc 100644 --- a/tests/queries/0_stateless/02122_join_group_by_timeout.reference +++ b/tests/queries/0_stateless/02122_join_group_by_timeout.reference @@ -1,4 +1,6 @@ -Code: 159 -0 +Code: 159 +query_duration 1 +0 +query_duration 1 Code: 159 0 diff --git a/tests/queries/0_stateless/02122_join_group_by_timeout.sh b/tests/queries/0_stateless/02122_join_group_by_timeout.sh index 8380c5dbd0c..b4644878544 100755 --- a/tests/queries/0_stateless/02122_join_group_by_timeout.sh +++ b/tests/queries/0_stateless/02122_join_group_by_timeout.sh @@ -1,27 +1,23 @@ #!/usr/bin/env bash -# Tags: no-debug - -# no-debug: Query is canceled by timeout after max_execution_time, -# but sending an exception to the client may hang -# for more than MAX_PROCESS_WAIT seconds in a slow debug build, -# and test will fail. CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -MAX_PROCESS_WAIT=5 - -IS_SANITIZER=$($CLICKHOUSE_CLIENT -q "SELECT count() FROM system.warnings WHERE message like '%built with sanitizer%'") -if [ "$IS_SANITIZER" -gt 0 ]; then - # Query may hang for more than 5 seconds, especially in tsan build - MAX_PROCESS_WAIT=15 +TIMEOUT=5 +IS_SANITIZER_OR_DEBUG=$($CLICKHOUSE_CLIENT -q "SELECT count() FROM system.warnings WHERE message like '%built with sanitizer%' or message like '%built in debug mode%'") +if [ "$IS_SANITIZER_OR_DEBUG" -gt 0 ]; then + # Increase the timeout due to in debug/sanitizers build: + # - client is slow + # - stacktrace resolving is slow + TIMEOUT=15 fi # TCP CLIENT: As of today (02/12/21) uses PullingAsyncPipelineExecutor ### Should be cancelled after 1 second and return a 159 exception (timeout) -timeout -s KILL $MAX_PROCESS_WAIT $CLICKHOUSE_CLIENT --max_execution_time 1 -q \ - "SELECT * FROM +query_id=$(random_str 12) +$CLICKHOUSE_CLIENT --query_id "$query_id" --max_execution_time 1 -q " + SELECT * FROM ( SELECT a.name as n FROM @@ -34,28 +30,35 @@ timeout -s KILL $MAX_PROCESS_WAIT $CLICKHOUSE_CLIENT --max_execution_time 1 -q \ GROUP BY n ) LIMIT 20 - FORMAT Null" 2>&1 | grep -o "Code: 159" | sort | uniq + FORMAT Null +" 2>&1 | grep -m1 -o "Code: 159" +$CLICKHOUSE_CLIENT -q "system flush logs" +${CLICKHOUSE_CURL} -q -sS "$CLICKHOUSE_URL" -d "select 'query_duration', round(query_duration_ms/1000) from system.query_log where current_database = '$CLICKHOUSE_DATABASE' and query_id = '$query_id' and type != 'QueryStart'" + ### Should stop pulling data and return what has been generated already (return code 0) -timeout -s KILL $MAX_PROCESS_WAIT $CLICKHOUSE_CLIENT -q \ - "SELECT a.name as n - FROM - ( - SELECT 'Name' as name, number FROM system.numbers LIMIT 2000000 - ) AS a, - ( - SELECT 'Name' as name2, number FROM system.numbers LIMIT 2000000 - ) as b - FORMAT Null - SETTINGS max_execution_time = 1, timeout_overflow_mode = 'break' - " +query_id=$(random_str 12) +$CLICKHOUSE_CLIENT --query_id "$query_id" -q " + SELECT a.name as n + FROM + ( + SELECT 'Name' as name, number FROM system.numbers LIMIT 2000000 + ) AS a, + ( + SELECT 'Name' as name2, number FROM system.numbers LIMIT 2000000 + ) as b + FORMAT Null + SETTINGS max_execution_time = 1, timeout_overflow_mode = 'break' +" echo $? +$CLICKHOUSE_CLIENT -q "system flush logs" +${CLICKHOUSE_CURL} -q -sS "$CLICKHOUSE_URL" -d "select 'query_duration', round(query_duration_ms/1000) from system.query_log where current_database = '$CLICKHOUSE_DATABASE' and query_id = '$query_id' and type != 'QueryStart'" # HTTP CLIENT: As of today (02/12/21) uses PullingPipelineExecutor ### Should be cancelled after 1 second and return a 159 exception (timeout) -${CLICKHOUSE_CURL} -q --max-time $MAX_PROCESS_WAIT -sS "$CLICKHOUSE_URL&max_execution_time=1" -d \ - "SELECT * FROM +${CLICKHOUSE_CURL} -q --max-time $TIMEOUT -sS "$CLICKHOUSE_URL&max_execution_time=1" -d " + SELECT * FROM ( SELECT a.name as n FROM @@ -68,12 +71,13 @@ ${CLICKHOUSE_CURL} -q --max-time $MAX_PROCESS_WAIT -sS "$CLICKHOUSE_URL&max_exec GROUP BY n ) LIMIT 20 - FORMAT Null" 2>&1 | grep -o "Code: 159" | sort | uniq + FORMAT Null +" 2>&1 | grep -o "Code: 159" | sort | uniq ### Should stop pulling data and return what has been generated already (return code 0) -${CLICKHOUSE_CURL} -q --max-time $MAX_PROCESS_WAIT -sS "$CLICKHOUSE_URL" -d \ - "SELECT a.name as n +${CLICKHOUSE_CURL} -q --max-time $TIMEOUT -sS "$CLICKHOUSE_URL" -d " + SELECT a.name as n FROM ( SELECT 'Name' as name, number FROM system.numbers LIMIT 2000000 @@ -83,5 +87,5 @@ ${CLICKHOUSE_CURL} -q --max-time $MAX_PROCESS_WAIT -sS "$CLICKHOUSE_URL" -d \ ) as b FORMAT Null SETTINGS max_execution_time = 1, timeout_overflow_mode = 'break' - " +" echo $? diff --git a/tests/queries/0_stateless/02125_query_views_log_window_function.sql b/tests/queries/0_stateless/02125_query_views_log_window_function.sql index fff1e943c58..3f15cf5b998 100644 --- a/tests/queries/0_stateless/02125_query_views_log_window_function.sql +++ b/tests/queries/0_stateless/02125_query_views_log_window_function.sql @@ -1,4 +1,4 @@ -set allow_experimental_analyzer = 0; +set enable_analyzer = 0; set allow_experimental_window_view = 1; CREATE TABLE data ( `id` UInt64, `timestamp` DateTime) ENGINE = Memory; diff --git a/tests/queries/0_stateless/02129_window_functions_disable_optimizations.sql b/tests/queries/0_stateless/02129_window_functions_disable_optimizations.sql index 0f12bc2eb9c..20a4f2bcf5f 100644 --- a/tests/queries/0_stateless/02129_window_functions_disable_optimizations.sql +++ b/tests/queries/0_stateless/02129_window_functions_disable_optimizations.sql @@ -29,4 +29,4 @@ HAVING sum(log(2) * number) > 346.57353 ORDER BY k; SELECT round(sum(log(2) * number), 6) AS k FROM numbers(10000) GROUP BY (number % 2) * (number % 3), number % 3, number % 2 HAVING sum(log(2) * number) > 346.57353 ORDER BY k -SETTINGS allow_experimental_analyzer=1; +SETTINGS enable_analyzer=1; diff --git a/tests/queries/0_stateless/02136_scalar_read_rows_json.sh b/tests/queries/0_stateless/02136_scalar_read_rows_json.sh index 1fe345d266d..00f2c92161f 100755 --- a/tests/queries/0_stateless/02136_scalar_read_rows_json.sh +++ b/tests/queries/0_stateless/02136_scalar_read_rows_json.sh @@ -7,4 +7,4 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) echo "#1" ${CLICKHOUSE_CLIENT} --query='SELECT count() FROM numbers(100) FORMAT JSON;' | grep -a -v "elapsed" echo "#2" -${CLICKHOUSE_CLIENT} --query='SELECT (SELECT max(number), count(number) FROM numbers(100000) as n) SETTINGS max_block_size = 65505, allow_experimental_analyzer = 1 FORMAT JSON;' | grep -a -v "elapsed" | grep -v "_subquery" +${CLICKHOUSE_CLIENT} --query='SELECT (SELECT max(number), count(number) FROM numbers(100000) as n) SETTINGS max_block_size = 65505, enable_analyzer = 1 FORMAT JSON;' | grep -a -v "elapsed" | grep -v "_subquery" 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 ae8c39b49bc..7bbdecf5501 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 @@ -22,20 +22,20 @@ SELECT toStartOfMonth(date) as d, i FROM t_read_in_order ORDER BY d, -i LIMIT 5; EXPLAIN PIPELINE SELECT toStartOfMonth(date) as d, i FROM t_read_in_order ORDER BY d, -i LIMIT 5; SELECT date, i FROM t_read_in_order WHERE date = '2020-10-11' ORDER BY i LIMIT 5; -EXPLAIN PIPELINE SELECT date, i FROM t_read_in_order WHERE date = '2020-10-11' ORDER BY i LIMIT 5 settings allow_experimental_analyzer=0; -EXPLAIN PIPELINE SELECT date, i FROM t_read_in_order WHERE date = '2020-10-11' ORDER BY i LIMIT 5 settings allow_experimental_analyzer=1; +EXPLAIN PIPELINE SELECT date, i FROM t_read_in_order WHERE date = '2020-10-11' ORDER BY i LIMIT 5 settings enable_analyzer=0; +EXPLAIN PIPELINE SELECT date, i FROM t_read_in_order WHERE date = '2020-10-11' ORDER BY i LIMIT 5 settings enable_analyzer=1; SELECT * FROM t_read_in_order WHERE date = '2020-10-11' ORDER BY i, v LIMIT 5; -EXPLAIN PIPELINE SELECT * FROM t_read_in_order WHERE date = '2020-10-11' ORDER BY i, v LIMIT 5 settings allow_experimental_analyzer=0; -EXPLAIN PIPELINE SELECT * FROM t_read_in_order WHERE date = '2020-10-11' ORDER BY i, v LIMIT 5 settings allow_experimental_analyzer=1; +EXPLAIN PIPELINE SELECT * FROM t_read_in_order WHERE date = '2020-10-11' ORDER BY i, v LIMIT 5 settings enable_analyzer=0; +EXPLAIN PIPELINE SELECT * FROM t_read_in_order WHERE date = '2020-10-11' ORDER BY i, v LIMIT 5 settings enable_analyzer=1; INSERT INTO t_read_in_order SELECT '2020-10-12', number, number FROM numbers(100000); SELECT date, i FROM t_read_in_order WHERE date = '2020-10-12' ORDER BY i LIMIT 5; EXPLAIN SYNTAX SELECT date, i FROM t_read_in_order WHERE date = '2020-10-12' ORDER BY i DESC LIMIT 5; -EXPLAIN PIPELINE SELECT date, i FROM t_read_in_order WHERE date = '2020-10-12' ORDER BY i DESC LIMIT 5 settings allow_experimental_analyzer=0; -EXPLAIN PIPELINE SELECT date, i FROM t_read_in_order WHERE date = '2020-10-12' ORDER BY i DESC LIMIT 5 settings allow_experimental_analyzer=1; +EXPLAIN PIPELINE SELECT date, i FROM t_read_in_order WHERE date = '2020-10-12' ORDER BY i DESC LIMIT 5 settings enable_analyzer=0; +EXPLAIN PIPELINE SELECT date, i FROM t_read_in_order WHERE date = '2020-10-12' ORDER BY i DESC LIMIT 5 settings enable_analyzer=1; SELECT date, i FROM t_read_in_order WHERE date = '2020-10-12' ORDER BY i DESC LIMIT 5; DROP TABLE IF EXISTS t_read_in_order; 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 index 6d715604d93..007dae6e427 100755 --- a/tests/queries/0_stateless/02151_hash_table_sizes_stats_joins.sh +++ b/tests/queries/0_stateless/02151_hash_table_sizes_stats_joins.sh @@ -58,7 +58,7 @@ $CLICKHOUSE_CLIENT -q "SYSTEM FLUSH LOGS" for i in "${!queries_without_preallocation[@]}"; do $CLICKHOUSE_CLIENT --param_query_id="${queries_without_preallocation[$i]}" -q " -- the old analyzer is not supported - SELECT sum(if(getSetting('allow_experimental_analyzer'), ProfileEvents['HashJoinPreallocatedElementsInHashTables'] = 0, 1)) + SELECT sum(if(getSetting('enable_analyzer'), ProfileEvents['HashJoinPreallocatedElementsInHashTables'] = 0, 1)) FROM system.query_log WHERE event_date >= yesterday() AND query_id = {query_id:String} AND current_database = currentDatabase() AND type = 'QueryFinish' " @@ -67,7 +67,7 @@ done for i in "${!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)) + SELECT sum(if(getSetting('enable_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' " diff --git a/tests/queries/0_stateless/02154_dictionary_get_http_json.sh b/tests/queries/0_stateless/02154_dictionary_get_http_json.sh index fbaf67fff2f..bcd9f4f5c1a 100755 --- a/tests/queries/0_stateless/02154_dictionary_get_http_json.sh +++ b/tests/queries/0_stateless/02154_dictionary_get_http_json.sh @@ -32,7 +32,7 @@ $CLICKHOUSE_CLIENT -q """ echo """ SELECT dictGet(02154_test_dictionary, 'value', toUInt64(0)), dictGet(02154_test_dictionary, 'value', toUInt64(1)) - SETTINGS allow_experimental_analyzer = 1 + SETTINGS enable_analyzer = 1 FORMAT JSON """ | ${CLICKHOUSE_CURL} -sSg "${CLICKHOUSE_URL}&wait_end_of_query=1&output_format_write_statistics=0" -d @- diff --git a/tests/queries/0_stateless/02156_storage_merge_prewhere.sql b/tests/queries/0_stateless/02156_storage_merge_prewhere.sql index 4f010ebadfd..2e5066fc488 100644 --- a/tests/queries/0_stateless/02156_storage_merge_prewhere.sql +++ b/tests/queries/0_stateless/02156_storage_merge_prewhere.sql @@ -24,8 +24,8 @@ INSERT INTO t_02156_mt1 SELECT number, toString(number) FROM numbers(10000); INSERT INTO t_02156_mt2 SELECT number, toString(number) FROM numbers(10000); INSERT INTO t_02156_log SELECT number, toString(number) FROM numbers(10000); -SELECT replaceRegexpAll(explain, '__table1\.|_UInt8', '') FROM (EXPLAIN actions=1 SELECT count() FROM t_02156_merge1 WHERE k = 3 AND notEmpty(v)) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%' settings allow_experimental_analyzer=1; -SELECT replaceRegexpAll(explain, '__table1\.|_UInt8', '') FROM (EXPLAIN actions=1 SELECT count() FROM t_02156_merge1 WHERE k = 3 AND notEmpty(v)) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%' settings allow_experimental_analyzer=0; +SELECT replaceRegexpAll(explain, '__table1\.|_UInt8', '') FROM (EXPLAIN actions=1 SELECT count() FROM t_02156_merge1 WHERE k = 3 AND notEmpty(v)) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%' settings enable_analyzer=1; +SELECT replaceRegexpAll(explain, '__table1\.|_UInt8', '') FROM (EXPLAIN actions=1 SELECT count() FROM t_02156_merge1 WHERE k = 3 AND notEmpty(v)) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%' settings enable_analyzer=0; SELECT count() FROM t_02156_merge1 WHERE k = 3 AND notEmpty(v); SELECT replaceRegexpAll(explain, '__table1\.|_UInt8', '') FROM (EXPLAIN actions=1 SELECT count() FROM t_02156_merge2 WHERE k = 3 AND notEmpty(v)) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; diff --git a/tests/queries/0_stateless/02164_clickhouse_local_interactive_exception.expect b/tests/queries/0_stateless/02164_clickhouse_local_interactive_exception.expect deleted file mode 100755 index add977c4fce..00000000000 --- a/tests/queries/0_stateless/02164_clickhouse_local_interactive_exception.expect +++ /dev/null @@ -1,29 +0,0 @@ -#!/usr/bin/expect -f - -set basedir [file dirname $argv0] -set basename [file tail $argv0] -if {[info exists env(CLICKHOUSE_TMP)]} { - set CLICKHOUSE_TMP $env(CLICKHOUSE_TMP) -} else { - set CLICKHOUSE_TMP "." -} -exp_internal -f $CLICKHOUSE_TMP/$basename.debuglog 0 - -log_user 0 -set timeout 20 -match_max 100000 - -expect_after { - -i $any_spawn_id eof { exp_continue } - -i $any_spawn_id timeout { exit 1 } -} - -spawn bash -c "source $basedir/../shell_config.sh ; \$CLICKHOUSE_LOCAL --disable_suggestion" - -expect ":) " -send -- "insert into table function null() format TSV some trash here 123 \n 456\r" -expect "CANNOT_PARSE_INPUT_ASSERTION_FAILED" -expect ":) " - -send -- "" -expect eof diff --git a/tests/queries/0_stateless/02164_clickhouse_local_interactive_exception.python b/tests/queries/0_stateless/02164_clickhouse_local_interactive_exception.python new file mode 100644 index 00000000000..4c2df9556a1 --- /dev/null +++ b/tests/queries/0_stateless/02164_clickhouse_local_interactive_exception.python @@ -0,0 +1,95 @@ +import pty +import os +import shlex +import time +import multiprocessing + +COMPLETION_TIMEOUT_SECONDS = 30 +DEBUG_LOG = os.path.join( + os.environ["CLICKHOUSE_TMP"], + os.path.basename(os.path.abspath(__file__)).strip(".python") + ".debuglog", +) + +STATE_MAP = { + -1: "process did not start", + 0: "all good", + 1: "process started and said ':)'", + 2: "prompt search was started", + 3: "prompt is missing", +} + + +def run_with_timeout(func, args, 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 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 expect(text, master, debug_log_fd, output=""): + while not text in output: + output_b = os.read(master, 4096) + output += output_b.decode() + debug_log_fd.write(repr(output_b) + "\n") + debug_log_fd.flush() + + return output + + +def test_completion(program, argv, prompt, state=None): + shell_pid, master = pty.fork() + if shell_pid == 0: + os.execv(program, argv) + else: + try: + debug_log_fd = open(DEBUG_LOG, "a") + + expect(":)", master, debug_log_fd) + + state.value = 1 + os.write(master, bytes(prompt.encode())) + expect(prompt[:-10], master, debug_log_fd) + + time.sleep(0.01) + os.write(master, b"\r") + state.value = 2 + + output = expect("CANNOT_PARSE_INPUT_ASSERTION_FAILED", master, debug_log_fd) + expect(":)", master, debug_log_fd, output) + + print("OK") + state.value = 0 + finally: + os.close(master) + debug_log_fd.close() + + +if __name__ == "__main__": + clickhouse_local = os.environ["CLICKHOUSE_LOCAL"] + args = shlex.split(clickhouse_local) + args.append("--disable_suggestion") + args.append("--highlight=0") + run_with_timeout( + test_completion, + [ + args[0], + args, + "insert into table function null() format TSV some trash here 123 \n 456", + ], + COMPLETION_TIMEOUT_SECONDS, + ) diff --git a/tests/queries/0_stateless/02164_clickhouse_local_interactive_exception.reference b/tests/queries/0_stateless/02164_clickhouse_local_interactive_exception.reference index e69de29bb2d..d86bac9de59 100644 --- a/tests/queries/0_stateless/02164_clickhouse_local_interactive_exception.reference +++ b/tests/queries/0_stateless/02164_clickhouse_local_interactive_exception.reference @@ -0,0 +1 @@ +OK diff --git a/tests/queries/0_stateless/02174_cte_scalar_cache.sql b/tests/queries/0_stateless/02174_cte_scalar_cache.sql index 86cfff21446..d14475c843a 100644 --- a/tests/queries/0_stateless/02174_cte_scalar_cache.sql +++ b/tests/queries/0_stateless/02174_cte_scalar_cache.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; WITH ( SELECT sleep(0.0001) FROM system.one ) as a1, @@ -28,7 +28,7 @@ WITH ( SELECT sleep(0.0001) FROM system.one ) as a5 SELECT '02177_CTE_NEW_ANALYZER', a1, a2, a3, a4, a5 FROM system.numbers LIMIT 100 FORMAT Null -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; SYSTEM FLUSH LOGS; SELECT diff --git a/tests/queries/0_stateless/02174_cte_scalar_cache_mv.sql b/tests/queries/0_stateless/02174_cte_scalar_cache_mv.sql index ca54b9e1400..a2be0a11faf 100644 --- a/tests/queries/0_stateless/02174_cte_scalar_cache_mv.sql +++ b/tests/queries/0_stateless/02174_cte_scalar_cache_mv.sql @@ -14,7 +14,7 @@ CREATE MATERIALIZED VIEW mv1 TO t2 AS FROM t1 LIMIT 5; -set allow_experimental_analyzer = 0; +set enable_analyzer = 0; -- FIRST INSERT INSERT INTO t1 @@ -61,7 +61,7 @@ WHERE AND event_date >= yesterday() AND event_time > now() - interval 10 minute; truncate table t2; -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; -- FIRST INSERT ANALYZER INSERT INTO t1 @@ -100,7 +100,7 @@ WHERE DROP TABLE mv1; -set allow_experimental_analyzer = 0; +set enable_analyzer = 0; CREATE TABLE t3 (z Int64) ENGINE = Memory; CREATE MATERIALIZED VIEW mv2 TO t3 AS @@ -134,7 +134,7 @@ WHERE AND event_date >= yesterday() AND event_time > now() - interval 10 minute; truncate table t3; -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; -- SECOND INSERT ANALYZER INSERT INTO t1 @@ -162,7 +162,7 @@ WHERE DROP TABLE mv2; -set allow_experimental_analyzer = 0; +set enable_analyzer = 0; CREATE TABLE t4 (z Int64) ENGINE = Memory; CREATE MATERIALIZED VIEW mv3 TO t4 AS @@ -197,7 +197,7 @@ WHERE AND event_date >= yesterday() AND event_time > now() - interval 10 minute; truncate table t4; -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; -- THIRD INSERT ANALYZER INSERT INTO t1 diff --git a/tests/queries/0_stateless/02184_hash_functions_and_ip_types.sql b/tests/queries/0_stateless/02184_hash_functions_and_ip_types.sql index c90c8b90c9e..e7d1909cae6 100644 --- a/tests/queries/0_stateless/02184_hash_functions_and_ip_types.sql +++ b/tests/queries/0_stateless/02184_hash_functions_and_ip_types.sql @@ -1,6 +1,6 @@ -- Tags: no-fasttest -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT toIPv4('1.2.3.4') AS ipv4, diff --git a/tests/queries/0_stateless/02210_toColumnTypeName_toLowCardinality_const.reference b/tests/queries/0_stateless/02210_toColumnTypeName_toLowCardinality_const.reference index 1e3d3a50562..e3978020431 100644 --- a/tests/queries/0_stateless/02210_toColumnTypeName_toLowCardinality_const.reference +++ b/tests/queries/0_stateless/02210_toColumnTypeName_toLowCardinality_const.reference @@ -1 +1 @@ -Const(ColumnLowCardinality) +Const(LowCardinality(UInt8)) diff --git a/tests/queries/0_stateless/02221_parallel_replicas_bug.sh b/tests/queries/0_stateless/02221_parallel_replicas_bug.sh index 3c44a2a7ba7..a382b3859f3 100755 --- a/tests/queries/0_stateless/02221_parallel_replicas_bug.sh +++ b/tests/queries/0_stateless/02221_parallel_replicas_bug.sh @@ -4,4 +4,4 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} --allow_experimental_parallel_reading_from_replicas=1 --parallel_replicas_for_non_replicated_merge_tree=1 -nm < "$CURDIR"/01099_parallel_distributed_insert_select.sql > /dev/null +${CLICKHOUSE_CLIENT} --allow_experimental_parallel_reading_from_replicas=1 --parallel_replicas_for_non_replicated_merge_tree=1 -m < "$CURDIR"/01099_parallel_distributed_insert_select.sql > /dev/null diff --git a/tests/queries/0_stateless/02221_system_zookeeper_unrestricted.sh b/tests/queries/0_stateless/02221_system_zookeeper_unrestricted.sh index db94c59d2de..e23a272a4e8 100755 --- a/tests/queries/0_stateless/02221_system_zookeeper_unrestricted.sh +++ b/tests/queries/0_stateless/02221_system_zookeeper_unrestricted.sh @@ -8,7 +8,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) ${CLICKHOUSE_CLIENT} --query="DROP TABLE IF EXISTS sample_table" ${CLICKHOUSE_CLIENT} --query="DROP TABLE IF EXISTS sample_table_2" -${CLICKHOUSE_CLIENT} -n -q" +${CLICKHOUSE_CLIENT} -q" CREATE TABLE sample_table ( key UInt64 ) @@ -16,7 +16,7 @@ ENGINE ReplicatedMergeTree('/clickhouse/$CLICKHOUSE_TEST_ZOOKEEPER_PREFIX/02221_ ORDER BY tuple(); " -${CLICKHOUSE_CLIENT} -n -q" +${CLICKHOUSE_CLIENT} -q" CREATE TABLE sample_table_2 ( key UInt64 ) diff --git a/tests/queries/0_stateless/02221_system_zookeeper_unrestricted_like.sh b/tests/queries/0_stateless/02221_system_zookeeper_unrestricted_like.sh index c62ec14b340..6381d811d5d 100755 --- a/tests/queries/0_stateless/02221_system_zookeeper_unrestricted_like.sh +++ b/tests/queries/0_stateless/02221_system_zookeeper_unrestricted_like.sh @@ -8,7 +8,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) ${CLICKHOUSE_CLIENT} --query="DROP TABLE IF EXISTS sample_table;" ${CLICKHOUSE_CLIENT} --query="DROP TABLE IF EXISTS sample_table_2;" -${CLICKHOUSE_CLIENT} -n --query="CREATE TABLE sample_table ( +${CLICKHOUSE_CLIENT} --query="CREATE TABLE sample_table ( key UInt64 ) ENGINE ReplicatedMergeTree('/clickhouse/$CLICKHOUSE_TEST_ZOOKEEPER_PREFIX/02221_system_zookeeper_unrestricted_like', '1') @@ -16,7 +16,7 @@ ORDER BY tuple(); DROP TABLE IF EXISTS sample_table SYNC;" -${CLICKHOUSE_CLIENT} -n --query "CREATE TABLE sample_table_2 ( +${CLICKHOUSE_CLIENT} --query "CREATE TABLE sample_table_2 ( key UInt64 ) ENGINE ReplicatedMergeTree('/clickhouse/$CLICKHOUSE_TEST_ZOOKEEPER_PREFIX/02221_system_zookeeper_unrestricted_like_2', '1') diff --git a/tests/queries/0_stateless/02225_parallel_distributed_insert_select_view.sh b/tests/queries/0_stateless/02225_parallel_distributed_insert_select_view.sh index 376a49fd820..63111cc32e4 100755 --- a/tests/queries/0_stateless/02225_parallel_distributed_insert_select_view.sh +++ b/tests/queries/0_stateless/02225_parallel_distributed_insert_select_view.sh @@ -6,7 +6,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " drop table if exists dst_02225; drop table if exists src_02225; create table dst_02225 (key Int) engine=Memory(); @@ -14,7 +14,7 @@ create table src_02225 (key Int) engine=Memory(); insert into src_02225 values (1); " -$CLICKHOUSE_CLIENT --param_database=$CLICKHOUSE_DATABASE -nm -q " +$CLICKHOUSE_CLIENT --param_database=$CLICKHOUSE_DATABASE -m -q " truncate table dst_02225; insert into function remote('127.{1,2}', currentDatabase(), dst_02225, key) select * from remote('127.{1,2}', view(select * from {database:Identifier}.src_02225), key) @@ -29,7 +29,7 @@ settings parallel_distributed_insert_select=2, max_distributed_depth=1; select * from dst_02225; " -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " drop table src_02225; drop table dst_02225; " diff --git a/tests/queries/0_stateless/02226_analyzer_or_like_combine.reference b/tests/queries/0_stateless/02226_analyzer_or_like_combine.reference index 0ff24b39709..61dfd1c6779 100644 --- a/tests/queries/0_stateless/02226_analyzer_or_like_combine.reference +++ b/tests/queries/0_stateless/02226_analyzer_or_like_combine.reference @@ -40,7 +40,7 @@ QUERY id: 0 LIST id: 3, nodes: 1 CONSTANT id: 4, constant_value: \'ŠŸŃ€ŠøŠ²ŠµŃ‚, World\', constant_value_type: String CONSTANT id: 16, constant_value: \'world%\', constant_value_type: String - SETTINGS optimize_or_like_chain=0 allow_experimental_analyzer=1 + SETTINGS optimize_or_like_chain=0 enable_analyzer=1 SELECT materialize(\'ŠŸŃ€ŠøŠ²ŠµŃ‚, World\') AS s WHERE multiMatchAny(s, [\'^hell\', \'(?i)ŠæрŠøŠ²ŠµŃ‚\', \'(?i)^world\']) OR false SETTINGS optimize_or_like_chain = 1 @@ -68,7 +68,7 @@ QUERY id: 0 CONSTANT id: 4, constant_value: \'ŠŸŃ€ŠøŠ²ŠµŃ‚, World\', constant_value_type: String CONSTANT id: 10, constant_value: Array_[\'^hell\', \'(?i)ŠæрŠøŠ²ŠµŃ‚\', \'(?i)^world\'], constant_value_type: Array(String) CONSTANT id: 11, constant_value: UInt64_0, constant_value_type: UInt8 - SETTINGS optimize_or_like_chain=1 allow_experimental_analyzer=1 + SETTINGS optimize_or_like_chain=1 enable_analyzer=1 SELECT materialize(\'ŠŸŃ€ŠøŠ²ŠµŃ‚, World\') AS s1, materialize(\'ŠŸŃ€ŠøŠ²ŠµŃ‚, World\') AS s2 diff --git a/tests/queries/0_stateless/02226_analyzer_or_like_combine.sql b/tests/queries/0_stateless/02226_analyzer_or_like_combine.sql index fbebfc6d281..b23e5640b8f 100644 --- a/tests/queries/0_stateless/02226_analyzer_or_like_combine.sql +++ b/tests/queries/0_stateless/02226_analyzer_or_like_combine.sql @@ -1,7 +1,7 @@ EXPLAIN SYNTAX SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s WHERE (s LIKE 'hell%') OR (s ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s ILIKE 'world%') SETTINGS optimize_or_like_chain = 0; -EXPLAIN QUERY TREE run_passes=1 SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s WHERE (s LIKE 'hell%') OR (s ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s ILIKE 'world%') SETTINGS optimize_or_like_chain = 0, allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE run_passes=1 SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s WHERE (s LIKE 'hell%') OR (s ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s ILIKE 'world%') SETTINGS optimize_or_like_chain = 0, enable_analyzer = 1; EXPLAIN SYNTAX SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s WHERE (s LIKE 'hell%') OR (s ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s ILIKE 'world%') SETTINGS optimize_or_like_chain = 1; -EXPLAIN QUERY TREE run_passes=1 SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s WHERE (s LIKE 'hell%') OR (s ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s ILIKE 'world%') SETTINGS optimize_or_like_chain = 1, allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE run_passes=1 SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s WHERE (s LIKE 'hell%') OR (s ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s ILIKE 'world%') SETTINGS optimize_or_like_chain = 1, enable_analyzer = 1; EXPLAIN SYNTAX SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s1, materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s2 WHERE (s1 LIKE 'hell%') OR (s2 ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s1 ILIKE 'world%') SETTINGS optimize_or_like_chain = 1; EXPLAIN SYNTAX SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s1, materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s2 WHERE (s1 LIKE 'hell%') OR (s2 ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s1 ILIKE 'world%') SETTINGS optimize_or_like_chain = 1, allow_hyperscan = 0; @@ -11,18 +11,18 @@ EXPLAIN SYNTAX SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s1, materialize('ŠŸ SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, optimized World') AS s WHERE (s LIKE 'hell%') OR (s LIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s ILIKE '%world') SETTINGS optimize_or_like_chain = 1; -SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, optimized World') AS s WHERE (s LIKE 'hell%') OR (s LIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s ILIKE '%world') SETTINGS optimize_or_like_chain = 1, allow_experimental_analyzer = 1; +SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, optimized World') AS s WHERE (s LIKE 'hell%') OR (s LIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s ILIKE '%world') SETTINGS optimize_or_like_chain = 1, enable_analyzer = 1; SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s WHERE (s LIKE 'hell%') OR (s LIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s ILIKE '%world') SETTINGS optimize_or_like_chain = 0; -SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s WHERE (s LIKE 'hell%') OR (s LIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s ILIKE '%world') SETTINGS optimize_or_like_chain = 0, allow_experimental_analyzer = 1; +SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s WHERE (s LIKE 'hell%') OR (s LIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s ILIKE '%world') SETTINGS optimize_or_like_chain = 0, enable_analyzer = 1; SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, optimized World') AS s WHERE (s LIKE 'hell%') OR (s ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s LIKE 'world%') SETTINGS optimize_or_like_chain = 1; -SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, optimized World') AS s WHERE (s LIKE 'hell%') OR (s ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s LIKE 'world%') SETTINGS optimize_or_like_chain = 1, allow_experimental_analyzer = 1; +SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, optimized World') AS s WHERE (s LIKE 'hell%') OR (s ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s LIKE 'world%') SETTINGS optimize_or_like_chain = 1, enable_analyzer = 1; SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s WHERE (s LIKE 'hell%') OR (s ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s LIKE 'world%') SETTINGS optimize_or_like_chain = 0; -SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s WHERE (s LIKE 'hell%') OR (s ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s LIKE 'world%') SETTINGS optimize_or_like_chain = 0, allow_experimental_analyzer = 1; +SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s WHERE (s LIKE 'hell%') OR (s ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s LIKE 'world%') SETTINGS optimize_or_like_chain = 0, enable_analyzer = 1; -SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s WHERE (s LIKE 'hell%') OR (s ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s ILIKE 'world%') SETTINGS optimize_or_like_chain = 1, allow_experimental_analyzer = 1; +SELECT materialize('ŠŸŃ€ŠøŠ²ŠµŃ‚, World') AS s WHERE (s LIKE 'hell%') OR (s ILIKE '%ŠæрŠøŠ²ŠµŃ‚%') OR (s ILIKE 'world%') SETTINGS optimize_or_like_chain = 1, enable_analyzer = 1; -- Aliases diff --git a/tests/queries/0_stateless/02226_parallel_reading_from_replicas_benchmark.sh b/tests/queries/0_stateless/02226_parallel_reading_from_replicas_benchmark.sh index bc90f4b2c11..177b373641f 100755 --- a/tests/queries/0_stateless/02226_parallel_reading_from_replicas_benchmark.sh +++ b/tests/queries/0_stateless/02226_parallel_reading_from_replicas_benchmark.sh @@ -4,7 +4,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " drop table if exists data_02226; create table data_02226 (key Int) engine=MergeTree() order by key as select * from numbers(1); @@ -24,7 +24,7 @@ opts=( $CLICKHOUSE_BENCHMARK --query "select * from remote('127.1', $CLICKHOUSE_DATABASE, data_02226)" "${opts[@]}" >& /dev/null ret=$? -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " drop table data_02226; " diff --git a/tests/queries/0_stateless/02227_union_match_by_name.sql b/tests/queries/0_stateless/02227_union_match_by_name.sql index 6a19add1d37..489c3d976ea 100644 --- a/tests/queries/0_stateless/02227_union_match_by_name.sql +++ b/tests/queries/0_stateless/02227_union_match_by_name.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- { echoOn } diff --git a/tests/queries/0_stateless/02228_unquoted_dates_in_csv_schema_inference.reference b/tests/queries/0_stateless/02228_unquoted_dates_in_csv_schema_inference.reference index be82d744a3b..56293ca0e5d 100644 --- a/tests/queries/0_stateless/02228_unquoted_dates_in_csv_schema_inference.reference +++ b/tests/queries/0_stateless/02228_unquoted_dates_in_csv_schema_inference.reference @@ -1 +1 @@ -c1 Nullable(DateTime64(9)) +c1 Nullable(DateTime) diff --git a/tests/queries/0_stateless/02232_allow_only_replicated_engine.sh b/tests/queries/0_stateless/02232_allow_only_replicated_engine.sh index d1a3825d286..e47a3033681 100755 --- a/tests/queries/0_stateless/02232_allow_only_replicated_engine.sh +++ b/tests/queries/0_stateless/02232_allow_only_replicated_engine.sh @@ -12,9 +12,9 @@ ${CLICKHOUSE_CLIENT} -q "GRANT CREATE TABLE ON ${CLICKHOUSE_DATABASE}_db.* TO us ${CLICKHOUSE_CLIENT} -q "GRANT TABLE ENGINE ON Memory, TABLE ENGINE ON MergeTree, TABLE ENGINE ON ReplicatedMergeTree TO user_${CLICKHOUSE_DATABASE}" ${CLICKHOUSE_CLIENT} -q "CREATE DATABASE ${CLICKHOUSE_DATABASE}_db engine = Replicated('/clickhouse/databases/${CLICKHOUSE_TEST_ZOOKEEPER_PREFIX}/${CLICKHOUSE_DATABASE}_db', '{shard}', '{replica}')" ${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --user "user_${CLICKHOUSE_DATABASE}" --query "CREATE TABLE ${CLICKHOUSE_DATABASE}_db.tab_memory (x UInt32) engine = Memory;" -${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --user "user_${CLICKHOUSE_DATABASE}" -n --query "CREATE TABLE ${CLICKHOUSE_DATABASE}_db.tab_mt (x UInt32) engine = MergeTree order by x;" 2>&1 | grep -o "Only tables with a Replicated engine" -${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none -n --query "CREATE TABLE ${CLICKHOUSE_DATABASE}_db.tab_mt (x UInt32) engine = MergeTree order by x;" -${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --user "user_${CLICKHOUSE_DATABASE}" -n --query "CREATE TABLE ${CLICKHOUSE_DATABASE}_db.tab_rmt (x UInt32) engine = ReplicatedMergeTree order by x;" +${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --user "user_${CLICKHOUSE_DATABASE}" --query "CREATE TABLE ${CLICKHOUSE_DATABASE}_db.tab_mt (x UInt32) engine = MergeTree order by x;" 2>&1 | grep -o "Only tables with a Replicated engine" +${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --query "CREATE TABLE ${CLICKHOUSE_DATABASE}_db.tab_mt (x UInt32) engine = MergeTree order by x;" +${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --user "user_${CLICKHOUSE_DATABASE}" --query "CREATE TABLE ${CLICKHOUSE_DATABASE}_db.tab_rmt (x UInt32) engine = ReplicatedMergeTree order by x;" ${CLICKHOUSE_CLIENT} --query "DROP DATABASE ${CLICKHOUSE_DATABASE}_db" ${CLICKHOUSE_CLIENT} -q "DROP USER user_${CLICKHOUSE_DATABASE}" diff --git a/tests/queries/0_stateless/02233_with_total_empty_chunk.sql b/tests/queries/0_stateless/02233_with_total_empty_chunk.sql index d59319ac75e..c70b35df459 100644 --- a/tests/queries/0_stateless/02233_with_total_empty_chunk.sql +++ b/tests/queries/0_stateless/02233_with_total_empty_chunk.sql @@ -1,3 +1,3 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT (NULL, NULL, NULL, NULL, NULL, NULL, NULL) FROM numbers(0) GROUP BY number WITH TOTALS HAVING sum(number) <= arrayJoin([]) -- { serverError ILLEGAL_TYPE_OF_COLUMN_FOR_FILTER }; diff --git a/tests/queries/0_stateless/02234_clickhouse_local_test_mode.sh b/tests/queries/0_stateless/02234_clickhouse_local_test_mode.sh index a6c47d80fa9..d387fda746c 100755 --- a/tests/queries/0_stateless/02234_clickhouse_local_test_mode.sh +++ b/tests/queries/0_stateless/02234_clickhouse_local_test_mode.sh @@ -4,6 +4,5 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh - -$CLICKHOUSE_LOCAL --query="SELECT n SETTINGS allow_experimental_analyzer = 1" 2>&1 | grep -q "Code: 47. DB::Exception:" && echo 'OK' || echo 'FAIL' -$CLICKHOUSE_LOCAL --query="SELECT n SETTINGS allow_experimental_analyzer = 0" 2>&1 | grep -q "Code: 47. DB::Exception:" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_LOCAL --query="SELECT n SETTINGS enable_analyzer = 1" 2>&1 | grep -q "Code: 47. DB::Exception:" && echo 'OK' || echo 'FAIL' +$CLICKHOUSE_LOCAL --query="SELECT n SETTINGS enable_analyzer = 0" 2>&1 | grep -q "Code: 47. DB::Exception:" && echo 'OK' || echo 'FAIL' diff --git a/tests/queries/0_stateless/02245_s3_virtual_columns.reference b/tests/queries/0_stateless/02245_s3_virtual_columns.reference index 09383c51888..3822f6ffa0f 100644 --- a/tests/queries/0_stateless/02245_s3_virtual_columns.reference +++ b/tests/queries/0_stateless/02245_s3_virtual_columns.reference @@ -11,5 +11,5 @@ create table test_02245_2 (a UInt64, _path Int32) engine = S3(s3_conn, filename= insert into test_02245_2 select 1, 2 settings s3_truncate_on_insert=1; select * from test_02245_2; 1 2 -select _path from test_02245_2; -2 +select _path, isNotNull(_etag) from test_02245_2; +2 1 diff --git a/tests/queries/0_stateless/02245_s3_virtual_columns.sql b/tests/queries/0_stateless/02245_s3_virtual_columns.sql index e86344d2094..a66b212e5c7 100644 --- a/tests/queries/0_stateless/02245_s3_virtual_columns.sql +++ b/tests/queries/0_stateless/02245_s3_virtual_columns.sql @@ -12,4 +12,4 @@ drop table if exists test_02245_2; create table test_02245_2 (a UInt64, _path Int32) engine = S3(s3_conn, filename='test_02245_2', format=Parquet); insert into test_02245_2 select 1, 2 settings s3_truncate_on_insert=1; select * from test_02245_2; -select _path from test_02245_2; +select _path, isNotNull(_etag) from test_02245_2; diff --git a/tests/queries/0_stateless/02250_ON_CLUSTER_grant.sh b/tests/queries/0_stateless/02250_ON_CLUSTER_grant.sh index 66417e9694a..09f9c0c8a98 100755 --- a/tests/queries/0_stateless/02250_ON_CLUSTER_grant.sh +++ b/tests/queries/0_stateless/02250_ON_CLUSTER_grant.sh @@ -6,7 +6,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) function cleanup() { - $CLICKHOUSE_CLIENT -nmq " + $CLICKHOUSE_CLIENT -mq " DROP USER IF EXISTS with_on_cluster_$CLICKHOUSE_TEST_UNIQUE_NAME; DROP USER IF EXISTS without_on_cluster_$CLICKHOUSE_TEST_UNIQUE_NAME; DROP DATABASE IF EXISTS db_with_on_cluster_$CLICKHOUSE_TEST_UNIQUE_NAME; @@ -15,7 +15,7 @@ function cleanup() cleanup trap cleanup EXIT -$CLICKHOUSE_CLIENT -nmq " +$CLICKHOUSE_CLIENT -mq " CREATE USER with_on_cluster_$CLICKHOUSE_TEST_UNIQUE_NAME; CREATE USER without_on_cluster_$CLICKHOUSE_TEST_UNIQUE_NAME; diff --git a/tests/queries/0_stateless/02262_column_ttl.sh b/tests/queries/0_stateless/02262_column_ttl.sh index b5e29c9b2a1..c620d3b6d9c 100755 --- a/tests/queries/0_stateless/02262_column_ttl.sh +++ b/tests/queries/0_stateless/02262_column_ttl.sh @@ -14,7 +14,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # note, that this should be written in .sh since we need $CLICKHOUSE_DATABASE # not 'default' to catch text_log -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " drop table if exists ttl_02262; drop table if exists this_text_log; @@ -31,7 +31,7 @@ $CLICKHOUSE_CLIENT -nm -q " ttl_02262_uuid=$($CLICKHOUSE_CLIENT -q "select uuid from system.tables where database = '$CLICKHOUSE_DATABASE' and name = 'ttl_02262'") -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " -- OPTIMIZE TABLE x FINAL will be done in background -- attach to it's log, via table UUID in query_id (see merger/mutator code). create materialized view this_text_log engine=Memory() as diff --git a/tests/queries/0_stateless/02267_join_dup_columns_issue36199.sql b/tests/queries/0_stateless/02267_join_dup_columns_issue36199.sql index fbcc374ba10..ecb2198237f 100644 --- a/tests/queries/0_stateless/02267_join_dup_columns_issue36199.sql +++ b/tests/queries/0_stateless/02267_join_dup_columns_issue36199.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SET join_algorithm = 'hash'; @@ -22,7 +22,7 @@ SELECT * FROM ( SELECT count('a'), count('b'), count('c'), 2 AS x ) as t1 RIGHT SELECT 'y', * FROM (SELECT count('y'), count('y'), 2 AS x) AS t1 RIGHT JOIN (SELECT count('x'), count('y'), 3 AS x) AS t2 ON t1.x = t2.x; SELECT * FROM (SELECT arrayJoin([NULL]), 9223372036854775806, arrayJoin([NULL]), NULL AS x) AS t1 RIGHT JOIN (SELECT arrayJoin([arrayJoin([10000000000.])]), NULL AS x) AS t2 ON t1.x = t2.x; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET join_algorithm = 'hash'; SELECT * FROM ( SELECT 2 AS x ) AS t1 RIGHT JOIN ( SELECT count('x'), count('y'), 2 AS x ) AS t2 ON t1.x = t2.x; diff --git a/tests/queries/0_stateless/02271_fix_column_matcher_and_column_transformer.sql b/tests/queries/0_stateless/02271_fix_column_matcher_and_column_transformer.sql index f8faa3e653b..ab89d98c608 100644 --- a/tests/queries/0_stateless/02271_fix_column_matcher_and_column_transformer.sql +++ b/tests/queries/0_stateless/02271_fix_column_matcher_and_column_transformer.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS github_events; diff --git a/tests/queries/0_stateless/02286_parallel_final.sh b/tests/queries/0_stateless/02286_parallel_final.sh index 0ac510208f3..47dfad42e11 100755 --- a/tests/queries/0_stateless/02286_parallel_final.sh +++ b/tests/queries/0_stateless/02286_parallel_final.sh @@ -9,7 +9,7 @@ echo "Test intersecting ranges" test_random_values() { layers=$1 - $CLICKHOUSE_CLIENT -n -q " + $CLICKHOUSE_CLIENT -q " drop table if exists tbl_8parts_${layers}granules_rnd; create table tbl_8parts_${layers}granules_rnd (key1 UInt32, sign Int8) engine = CollapsingMergeTree(sign) order by (key1) partition by (key1 % 8); insert into tbl_8parts_${layers}granules_rnd select number, 1 from numbers_mt($((layers * 8 * 8192))); @@ -29,7 +29,7 @@ echo "Test non intersecting ranges" test_sequential_values() { layers=$1 - $CLICKHOUSE_CLIENT -n -q " + $CLICKHOUSE_CLIENT -q " drop table if exists tbl_8parts_${layers}granules_seq; create table tbl_8parts_${layers}granules_seq (key1 UInt32, sign Int8) engine = CollapsingMergeTree(sign) order by (key1) partition by (key1 / $((layers * 8192)))::UInt64; insert into tbl_8parts_${layers}granules_seq select number, 1 from numbers_mt($((layers * 8 * 8192))); diff --git a/tests/queries/0_stateless/02303_query_kind.reference b/tests/queries/0_stateless/02303_query_kind.reference index 9f1c026f889..c26242098e3 100644 --- a/tests/queries/0_stateless/02303_query_kind.reference +++ b/tests/queries/0_stateless/02303_query_kind.reference @@ -1,4 +1,4 @@ -clickhouse-client --allow_experimental_analyzer=1 --query_kind secondary_query -q explain plan header=1 select toString(dummy) as dummy from system.one group by dummy +clickhouse-client --enable_analyzer=1 --query_kind secondary_query -q explain plan header=1 select toString(dummy) as dummy from system.one group by dummy Expression ((Project names + Projection)) Header: dummy String Aggregating @@ -7,7 +7,7 @@ Header: dummy String Header: toString(__table1.dummy) String ReadFromStorage (SystemOne) Header: dummy UInt8 -clickhouse-local --allow_experimental_analyzer=1 --query_kind secondary_query -q explain plan header=1 select toString(dummy) as dummy from system.one group by dummy +clickhouse-local --enable_analyzer=1 --query_kind secondary_query -q explain plan header=1 select toString(dummy) as dummy from system.one group by dummy Expression ((Project names + Projection)) Header: dummy String Aggregating @@ -16,7 +16,7 @@ Header: dummy String Header: toString(__table1.dummy) String ReadFromStorage (SystemOne) Header: dummy UInt8 -clickhouse-client --allow_experimental_analyzer=1 --query_kind initial_query -q explain plan header=1 select toString(dummy) as dummy from system.one group by dummy +clickhouse-client --enable_analyzer=1 --query_kind initial_query -q explain plan header=1 select toString(dummy) as dummy from system.one group by dummy Expression ((Project names + Projection)) Header: dummy String Aggregating @@ -25,7 +25,7 @@ Header: dummy String Header: __table1.dummy UInt8 ReadFromStorage (SystemOne) Header: dummy UInt8 -clickhouse-local --allow_experimental_analyzer=1 --query_kind initial_query -q explain plan header=1 select toString(dummy) as dummy from system.one group by dummy +clickhouse-local --enable_analyzer=1 --query_kind initial_query -q explain plan header=1 select toString(dummy) as dummy from system.one group by dummy Expression ((Project names + Projection)) Header: dummy String Aggregating diff --git a/tests/queries/0_stateless/02303_query_kind.sh b/tests/queries/0_stateless/02303_query_kind.sh index 1d883a2dcc7..7fe491b9deb 100755 --- a/tests/queries/0_stateless/02303_query_kind.sh +++ b/tests/queries/0_stateless/02303_query_kind.sh @@ -5,7 +5,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CUR_DIR"/../shell_config.sh opts=( - "--allow_experimental_analyzer=1" + "--enable_analyzer=1" ) function run_query() diff --git a/tests/queries/0_stateless/02310_clickhouse_local_INSERT_progress_profile_events.expect b/tests/queries/0_stateless/02310_clickhouse_local_INSERT_progress_profile_events.expect index d5b2a278180..5c731ea6d89 100755 --- a/tests/queries/0_stateless/02310_clickhouse_local_INSERT_progress_profile_events.expect +++ b/tests/queries/0_stateless/02310_clickhouse_local_INSERT_progress_profile_events.expect @@ -1,5 +1,5 @@ #!/usr/bin/expect -f -# Tags: no-debug, no-tsan, no-msan, no-asan, no-ubsan, no-s3-storage +# Tags: no-debug, no-tsan, no-msan, no-asan, no-ubsan, no-s3-storage, no-cpu-aarch64 # ^ it can be slower than 60 seconds # This is the regression for the concurrent access in ProgressIndication, diff --git a/tests/queries/0_stateless/02313_dump_column_structure_low_cardinality.reference b/tests/queries/0_stateless/02313_dump_column_structure_low_cardinality.reference index fa7f1799c31..6b1e4743867 100644 --- a/tests/queries/0_stateless/02313_dump_column_structure_low_cardinality.reference +++ b/tests/queries/0_stateless/02313_dump_column_structure_low_cardinality.reference @@ -1 +1 @@ -Array(LowCardinality(String)), Const(size = 1, Array(size = 1, UInt64(size = 1), ColumnLowCardinality(size = 2, UInt8(size = 2), ColumnUnique(size = 3, String(size = 3))))) +Array(LowCardinality(String)), Const(size = 1, Array(size = 1, UInt64(size = 1), LowCardinality(size = 2, UInt8(size = 2), Unique(size = 3, String(size = 3))))) diff --git a/tests/queries/0_stateless/02313_filesystem_cache_seeks.sh b/tests/queries/0_stateless/02313_filesystem_cache_seeks.sh index b54e3d7f805..b7adde6fcbb 100755 --- a/tests/queries/0_stateless/02313_filesystem_cache_seeks.sh +++ b/tests/queries/0_stateless/02313_filesystem_cache_seeks.sh @@ -8,16 +8,20 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CUR_DIR"/../shell_config.sh +client_opts=( + --distributed_ddl_output_mode 'null_status_on_timeout' +) + 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" - $CLICKHOUSE_CLIENT --query "DROP TABLE IF EXISTS test_02313" + $CLICKHOUSE_CLIENT "${client_opts[@]}" --query "DROP TABLE IF EXISTS test_02313" > /dev/null - $CLICKHOUSE_CLIENT --query "CREATE TABLE test_02313 (id Int32, val String) + $CLICKHOUSE_CLIENT "${client_opts[@]}" --query "CREATE TABLE test_02313 (id Int32, val String) ENGINE = MergeTree() ORDER BY tuple() - SETTINGS storage_policy = '$STORAGE_POLICY'" + SETTINGS storage_policy = '$STORAGE_POLICY'" > /dev/null $CLICKHOUSE_CLIENT --enable_filesystem_cache_on_write_operations=0 -n --query "INSERT INTO test_02313 SELECT * FROM @@ -32,6 +36,6 @@ for STORAGE_POLICY in 's3_cache' 'local_cache' 's3_cache_multi' 'azure_cache'; d $CLICKHOUSE_CLIENT --query "SELECT * FROM test_02313 WHERE val LIKE concat('%', randomPrintableASCII(3), '%') FORMAT Null" $CLICKHOUSE_CLIENT --query "SELECT * FROM test_02313 WHERE val LIKE concat('%', randomPrintableASCII(3), '%') FORMAT Null" - $CLICKHOUSE_CLIENT --query "DROP TABLE test_02313" + $CLICKHOUSE_CLIENT "${client_opts[@]}" --query "DROP TABLE test_02313" > /dev/null done diff --git a/tests/queries/0_stateless/02315_grouping_constant_folding.reference b/tests/queries/0_stateless/02315_grouping_constant_folding.reference index 31816318a42..7b8c75b2304 100644 --- a/tests/queries/0_stateless/02315_grouping_constant_folding.reference +++ b/tests/queries/0_stateless/02315_grouping_constant_folding.reference @@ -27,7 +27,7 @@ SELECT count() AS amount, a, b, GROUPING(a, b) FROM test02315 GROUP BY ROLLUP(a, 5 0 0 2 5 1 0 2 10 0 0 0 -SELECT count() AS amount, a, b, GROUPING(a, b) FROM test02315 GROUP BY GROUPING SETS ((a, b), (a, a), ()) ORDER BY (amount, a, b) SETTINGS force_grouping_standard_compatibility=0, allow_experimental_analyzer=1; +SELECT count() AS amount, a, b, GROUPING(a, b) FROM test02315 GROUP BY GROUPING SETS ((a, b), (a, a), ()) ORDER BY (amount, a, b) SETTINGS force_grouping_standard_compatibility=0, enable_analyzer=1; 1 0 0 3 1 0 2 3 1 0 4 3 diff --git a/tests/queries/0_stateless/02315_grouping_constant_folding.sql b/tests/queries/0_stateless/02315_grouping_constant_folding.sql index f992aa0da32..5e305d2e6c5 100644 --- a/tests/queries/0_stateless/02315_grouping_constant_folding.sql +++ b/tests/queries/0_stateless/02315_grouping_constant_folding.sql @@ -9,7 +9,7 @@ SELECT count() AS amount, a, b, GROUPING(a, b) FROM test02315 GROUP BY GROUPING SELECT count() AS amount, a, b, GROUPING(a, b) FROM test02315 GROUP BY ROLLUP(a, b) ORDER BY (amount, a, b) SETTINGS force_grouping_standard_compatibility=0; -SELECT count() AS amount, a, b, GROUPING(a, b) FROM test02315 GROUP BY GROUPING SETS ((a, b), (a, a), ()) ORDER BY (amount, a, b) SETTINGS force_grouping_standard_compatibility=0, allow_experimental_analyzer=1; +SELECT count() AS amount, a, b, GROUPING(a, b) FROM test02315 GROUP BY GROUPING SETS ((a, b), (a, a), ()) ORDER BY (amount, a, b) SETTINGS force_grouping_standard_compatibility=0, enable_analyzer=1; -- { echoOff } DROP TABLE test02315; diff --git a/tests/queries/0_stateless/02317_distinct_in_order_optimization_explain.sh b/tests/queries/0_stateless/02317_distinct_in_order_optimization_explain.sh index 07c7bc4af56..953485c3a1f 100755 --- a/tests/queries/0_stateless/02317_distinct_in_order_optimization_explain.sh +++ b/tests/queries/0_stateless/02317_distinct_in_order_optimization_explain.sh @@ -23,99 +23,99 @@ $CLICKHOUSE_CLIENT -q "insert into distinct_in_order_explain select number % num $CLICKHOUSE_CLIENT -q "select '-- disable optimize_distinct_in_order'" $CLICKHOUSE_CLIENT -q "select '-- distinct all primary key columns -> ordinary distinct'" -$CLICKHOUSE_CLIENT -nq "$DISABLE_OPTIMIZATION;explain pipeline select distinct * from distinct_in_order_explain" | eval $FIND_DISTINCT +$CLICKHOUSE_CLIENT -q "$DISABLE_OPTIMIZATION;explain pipeline select distinct * from distinct_in_order_explain" | eval $FIND_DISTINCT $CLICKHOUSE_CLIENT -q "select '-- enable optimize_distinct_in_order'" $CLICKHOUSE_CLIENT -q "select '-- distinct with all primary key columns -> pre-distinct optimization only'" -$CLICKHOUSE_CLIENT -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct * from distinct_in_order_explain" | eval $FIND_DISTINCT +$CLICKHOUSE_CLIENT -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct * from distinct_in_order_explain" | eval $FIND_DISTINCT $CLICKHOUSE_CLIENT -q "select '-- distinct with primary key prefix -> pre-distinct optimization only'" -$CLICKHOUSE_CLIENT -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, c from distinct_in_order_explain" | eval $FIND_DISTINCT +$CLICKHOUSE_CLIENT -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, c from distinct_in_order_explain" | eval $FIND_DISTINCT $CLICKHOUSE_CLIENT -q "select '-- distinct with primary key prefix and order by column in distinct -> pre-distinct and final distinct optimization'" -$CLICKHOUSE_CLIENT -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, c from distinct_in_order_explain order by c" | eval $FIND_DISTINCT +$CLICKHOUSE_CLIENT -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, c from distinct_in_order_explain order by c" | eval $FIND_DISTINCT $CLICKHOUSE_CLIENT -q "select '-- distinct with primary key prefix and order by the same columns -> pre-distinct and final distinct optimization'" -$CLICKHOUSE_CLIENT -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, b from distinct_in_order_explain order by a, b" | eval $FIND_DISTINCT +$CLICKHOUSE_CLIENT -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, b from distinct_in_order_explain order by a, b" | eval $FIND_DISTINCT $CLICKHOUSE_CLIENT -q "select '-- distinct with primary key prefix and order by columns are prefix of distinct columns -> pre-distinct and final distinct optimization'" -$CLICKHOUSE_CLIENT -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, b from distinct_in_order_explain order by a" | eval $FIND_DISTINCT +$CLICKHOUSE_CLIENT -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, b from distinct_in_order_explain order by a" | eval $FIND_DISTINCT $CLICKHOUSE_CLIENT -q "select '-- distinct with primary key prefix and order by column in distinct but non-primary key prefix -> pre-distinct and final distinct optimization'" -$CLICKHOUSE_CLIENT -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, b, c from distinct_in_order_explain order by c" | eval $FIND_DISTINCT +$CLICKHOUSE_CLIENT -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, b, c from distinct_in_order_explain order by c" | eval $FIND_DISTINCT $CLICKHOUSE_CLIENT -q "select '-- distinct with primary key prefix and order by column _not_ in distinct -> pre-distinct optimization only'" -$CLICKHOUSE_CLIENT -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, c from distinct_in_order_explain order by b" | eval $FIND_DISTINCT +$CLICKHOUSE_CLIENT -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, c from distinct_in_order_explain order by b" | eval $FIND_DISTINCT $CLICKHOUSE_CLIENT -q "select '-- distinct with non-primary key prefix -> ordinary distinct'" -$CLICKHOUSE_CLIENT -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct b, c from distinct_in_order_explain" | eval $FIND_DISTINCT +$CLICKHOUSE_CLIENT -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct b, c from distinct_in_order_explain" | eval $FIND_DISTINCT $CLICKHOUSE_CLIENT -q "select '-- distinct with non-primary key prefix and order by column in distinct -> final distinct optimization only'" -$CLICKHOUSE_CLIENT -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct b, c from distinct_in_order_explain order by b" | eval $FIND_DISTINCT +$CLICKHOUSE_CLIENT -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct b, c from distinct_in_order_explain order by b" | eval $FIND_DISTINCT $CLICKHOUSE_CLIENT -q "select '-- distinct with non-primary key prefix and order by column _not_ in distinct -> ordinary distinct'" -$CLICKHOUSE_CLIENT -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct b, c from distinct_in_order_explain order by a" | eval $FIND_DISTINCT +$CLICKHOUSE_CLIENT -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct b, c from distinct_in_order_explain order by a" | eval $FIND_DISTINCT $CLICKHOUSE_CLIENT -q "select '-- distinct with non-primary key prefix and order by _const_ column in distinct -> ordinary distinct'" -$CLICKHOUSE_CLIENT -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct b, 1 as x from distinct_in_order_explain order by x" | eval $FIND_DISTINCT +$CLICKHOUSE_CLIENT -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct b, 1 as x from distinct_in_order_explain order by x" | eval $FIND_DISTINCT echo "-- Check reading in order for distinct" echo "-- disabled, distinct columns match sorting key" -$CLICKHOUSE_CLIENT --max_threads=0 -nq "$DISABLE_OPTIMIZATION;explain pipeline select distinct a, b from distinct_in_order_explain" | eval $FIND_READING_DEFAULT +$CLICKHOUSE_CLIENT --max_threads=0 -q "$DISABLE_OPTIMIZATION;explain pipeline select distinct a, b from distinct_in_order_explain" | eval $FIND_READING_DEFAULT echo "-- enabled, distinct columns match sorting key" # read_in_order_two_level_merge_threshold is set here to avoid repeating MergeTreeInOrder in output -$CLICKHOUSE_CLIENT --read_in_order_two_level_merge_threshold=2 -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, b from distinct_in_order_explain" | eval $FIND_READING_IN_ORDER +$CLICKHOUSE_CLIENT --read_in_order_two_level_merge_threshold=2 -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, b from distinct_in_order_explain" | eval $FIND_READING_IN_ORDER echo "-- enabled, distinct columns form prefix of sorting key" -$CLICKHOUSE_CLIENT --read_in_order_two_level_merge_threshold=2 -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, b from distinct_in_order_explain" | eval $FIND_READING_IN_ORDER +$CLICKHOUSE_CLIENT --read_in_order_two_level_merge_threshold=2 -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, b from distinct_in_order_explain" | eval $FIND_READING_IN_ORDER echo "-- enabled, distinct columns DON't form prefix of sorting key" -$CLICKHOUSE_CLIENT --max_threads=0 -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct b from distinct_in_order_explain" | eval $FIND_READING_DEFAULT +$CLICKHOUSE_CLIENT --max_threads=0 -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct b from distinct_in_order_explain" | eval $FIND_READING_DEFAULT echo "-- enabled, distinct columns contains constant columns, non-const columns form prefix of sorting key" -$CLICKHOUSE_CLIENT --read_in_order_two_level_merge_threshold=2 -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct 1, a from distinct_in_order_explain" | eval $FIND_READING_IN_ORDER +$CLICKHOUSE_CLIENT --read_in_order_two_level_merge_threshold=2 -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct 1, a from distinct_in_order_explain" | eval $FIND_READING_IN_ORDER echo "-- enabled, distinct columns contains constant columns, non-const columns match prefix of sorting key" -$CLICKHOUSE_CLIENT --read_in_order_two_level_merge_threshold=2 -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct 1, b, a from distinct_in_order_explain" | eval $FIND_READING_IN_ORDER +$CLICKHOUSE_CLIENT --read_in_order_two_level_merge_threshold=2 -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct 1, b, a from distinct_in_order_explain" | eval $FIND_READING_IN_ORDER echo "-- enabled, only part of distinct columns form prefix of sorting key" -$CLICKHOUSE_CLIENT --max_threads=0 -nq "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, c from distinct_in_order_explain" | eval $FIND_READING_DEFAULT +$CLICKHOUSE_CLIENT --max_threads=0 -q "$ENABLE_OPTIMIZATION;explain pipeline select distinct a, c from distinct_in_order_explain" | eval $FIND_READING_DEFAULT echo "=== disable new analyzer ===" -DISABLE_ANALYZER="set allow_experimental_analyzer=0" +DISABLE_ANALYZER="set enable_analyzer=0" echo "-- enabled, check that sorting properties are propagated from ReadFromMergeTree till preliminary distinct" -$CLICKHOUSE_CLIENT -nq "$DISABLE_ANALYZER;$ENABLE_OPTIMIZATION;explain plan sorting=1 select distinct b, a from distinct_in_order_explain where a > 0" | eval $FIND_SORTING_PROPERTIES +$CLICKHOUSE_CLIENT -q "$DISABLE_ANALYZER;$ENABLE_OPTIMIZATION;explain plan sorting=1 select distinct b, a from distinct_in_order_explain where a > 0" | eval $FIND_SORTING_PROPERTIES echo "-- check that reading in order optimization for ORDER BY and DISTINCT applied correctly in the same query" ENABLE_READ_IN_ORDER="set optimize_read_in_order=1" echo "-- disabled, check that sorting description for ReadFromMergeTree match ORDER BY columns" -$CLICKHOUSE_CLIENT -nq "$DISABLE_ANALYZER;$DISABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a" | eval $FIND_SORTING_PROPERTIES +$CLICKHOUSE_CLIENT -q "$DISABLE_ANALYZER;$DISABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a" | eval $FIND_SORTING_PROPERTIES echo "-- enabled, check that ReadFromMergeTree sorting description is overwritten by DISTINCT optimization i.e. it contains columns from DISTINCT clause" -$CLICKHOUSE_CLIENT -nq "$DISABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a" | eval $FIND_SORTING_PROPERTIES +$CLICKHOUSE_CLIENT -q "$DISABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a" | eval $FIND_SORTING_PROPERTIES echo "-- enabled, check that ReadFromMergeTree sorting description is overwritten by DISTINCT optimization, but direction used from ORDER BY clause" -$CLICKHOUSE_CLIENT -nq "$DISABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a DESC" | eval $FIND_SORTING_PROPERTIES +$CLICKHOUSE_CLIENT -q "$DISABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a DESC" | eval $FIND_SORTING_PROPERTIES echo "-- enabled, check that ReadFromMergeTree sorting description is NOT overwritten by DISTINCT optimization (1), - it contains columns from ORDER BY clause" -$CLICKHOUSE_CLIENT -nq "$DISABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct a from distinct_in_order_explain order by a, b" | eval $FIND_SORTING_PROPERTIES +$CLICKHOUSE_CLIENT -q "$DISABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct a from distinct_in_order_explain order by a, b" | eval $FIND_SORTING_PROPERTIES echo "-- enabled, check that ReadFromMergeTree sorting description is NOT overwritten by DISTINCT optimization (2), - direction used from ORDER BY clause" -$CLICKHOUSE_CLIENT -nq "$DISABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a DESC, b DESC" | eval $FIND_SORTING_PROPERTIES +$CLICKHOUSE_CLIENT -q "$DISABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a DESC, b DESC" | eval $FIND_SORTING_PROPERTIES echo "-- enabled, check that disabling other 'read in order' optimizations do not disable distinct in order optimization" -$CLICKHOUSE_CLIENT -nq "$DISABLE_ANALYZER;$ENABLE_OPTIMIZATION;set optimize_read_in_order=0;set optimize_aggregation_in_order=0;set optimize_read_in_window_order=0;explain plan sorting=1 select distinct a,b from distinct_in_order_explain" | eval $FIND_SORTING_PROPERTIES +$CLICKHOUSE_CLIENT -q "$DISABLE_ANALYZER;$ENABLE_OPTIMIZATION;set optimize_read_in_order=0;set optimize_aggregation_in_order=0;set optimize_read_in_window_order=0;explain plan sorting=1 select distinct a,b from distinct_in_order_explain" | eval $FIND_SORTING_PROPERTIES echo "=== enable new analyzer ===" -ENABLE_ANALYZER="set allow_experimental_analyzer=1" +ENABLE_ANALYZER="set enable_analyzer=1" echo "-- enabled, check that sorting properties are propagated from ReadFromMergeTree till preliminary distinct" -$CLICKHOUSE_CLIENT -nq "$ENABLE_ANALYZER;$ENABLE_OPTIMIZATION;explain plan sorting=1 select distinct b, a from distinct_in_order_explain where a > 0 settings optimize_move_to_prewhere=1" | eval $FIND_SORTING_PROPERTIES +$CLICKHOUSE_CLIENT -q "$ENABLE_ANALYZER;$ENABLE_OPTIMIZATION;explain plan sorting=1 select distinct b, a from distinct_in_order_explain where a > 0 settings optimize_move_to_prewhere=1" | eval $FIND_SORTING_PROPERTIES echo "-- disabled, check that sorting description for ReadFromMergeTree match ORDER BY columns" -$CLICKHOUSE_CLIENT -nq "$ENABLE_ANALYZER;$DISABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a" | eval $FIND_SORTING_PROPERTIES +$CLICKHOUSE_CLIENT -q "$ENABLE_ANALYZER;$DISABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a" | eval $FIND_SORTING_PROPERTIES echo "-- enabled, check that ReadFromMergeTree sorting description is overwritten by DISTINCT optimization i.e. it contains columns from DISTINCT clause" -$CLICKHOUSE_CLIENT -nq "$ENABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a" | eval $FIND_SORTING_PROPERTIES +$CLICKHOUSE_CLIENT -q "$ENABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a" | eval $FIND_SORTING_PROPERTIES echo "-- enabled, check that ReadFromMergeTree sorting description is overwritten by DISTINCT optimization, but direction used from ORDER BY clause" -$CLICKHOUSE_CLIENT -nq "$ENABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a DESC" | eval $FIND_SORTING_PROPERTIES +$CLICKHOUSE_CLIENT -q "$ENABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a DESC" | eval $FIND_SORTING_PROPERTIES echo "-- enabled, check that ReadFromMergeTree sorting description is NOT overwritten by DISTINCT optimization (1), - it contains columns from ORDER BY clause" -$CLICKHOUSE_CLIENT -nq "$ENABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct a from distinct_in_order_explain order by a, b" | eval $FIND_SORTING_PROPERTIES +$CLICKHOUSE_CLIENT -q "$ENABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct a from distinct_in_order_explain order by a, b" | eval $FIND_SORTING_PROPERTIES echo "-- enabled, check that ReadFromMergeTree sorting description is NOT overwritten by DISTINCT optimization (2), - direction used from ORDER BY clause" -$CLICKHOUSE_CLIENT -nq "$ENABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a DESC, b DESC" | eval $FIND_SORTING_PROPERTIES +$CLICKHOUSE_CLIENT -q "$ENABLE_ANALYZER;$ENABLE_OPTIMIZATION;$ENABLE_READ_IN_ORDER;explain plan sorting=1 select distinct b, a from distinct_in_order_explain order by a DESC, b DESC" | eval $FIND_SORTING_PROPERTIES echo "-- enabled, check that disabling other 'read in order' optimizations do not disable distinct in order optimization" -$CLICKHOUSE_CLIENT -nq "$ENABLE_ANALYZER;$ENABLE_OPTIMIZATION;set optimize_read_in_order=0;set optimize_aggregation_in_order=0;set optimize_read_in_window_order=0;explain plan sorting=1 select distinct a,b from distinct_in_order_explain" | eval $FIND_SORTING_PROPERTIES +$CLICKHOUSE_CLIENT -q "$ENABLE_ANALYZER;$ENABLE_OPTIMIZATION;set optimize_read_in_order=0;set optimize_aggregation_in_order=0;set optimize_read_in_window_order=0;explain plan sorting=1 select distinct a,b from distinct_in_order_explain" | eval $FIND_SORTING_PROPERTIES $CLICKHOUSE_CLIENT -q "drop table if exists distinct_in_order_explain sync" diff --git a/tests/queries/0_stateless/02319_lightweight_delete_on_merge_tree.sql b/tests/queries/0_stateless/02319_lightweight_delete_on_merge_tree.sql index f82f79dbe44..6491253cd5f 100644 --- a/tests/queries/0_stateless/02319_lightweight_delete_on_merge_tree.sql +++ b/tests/queries/0_stateless/02319_lightweight_delete_on_merge_tree.sql @@ -102,7 +102,7 @@ ALTER TABLE t_proj ADD PROJECTION p_1 (SELECT avg(a), avg(b), count()) SETTINGS INSERT INTO t_proj SELECT number + 1, number + 1 FROM numbers(1000); -DELETE FROM t_proj WHERE a < 100; -- { serverError NOT_IMPLEMENTED } +DELETE FROM t_proj WHERE a < 100; -- { serverError SUPPORT_IS_DISABLED } SELECT avg(a), avg(b), count() FROM t_proj; diff --git a/tests/queries/0_stateless/02325_dates_schema_inference.reference b/tests/queries/0_stateless/02325_dates_schema_inference.reference index c8eebd3262e..124f105220d 100644 --- a/tests/queries/0_stateless/02325_dates_schema_inference.reference +++ b/tests/queries/0_stateless/02325_dates_schema_inference.reference @@ -1,29 +1,29 @@ JSONEachRow x Nullable(Date) x Nullable(DateTime64(9)) -x Nullable(DateTime64(9)) +x Nullable(DateTime) x Array(Nullable(Date)) -x Array(Nullable(DateTime64(9))) -x Array(Nullable(DateTime64(9))) -x Tuple(\n date1 Nullable(DateTime64(9)),\n date2 Nullable(Date)) -x Array(Nullable(DateTime64(9))) -x Array(Nullable(DateTime64(9))) -x Nullable(DateTime64(9)) +x Array(Nullable(DateTime)) +x Array(Nullable(DateTime)) +x Tuple(\n date1 Nullable(DateTime),\n date2 Nullable(Date)) +x Array(Nullable(DateTime)) +x Array(Nullable(DateTime)) +x Nullable(DateTime) x Array(Nullable(String)) x Nullable(String) x Array(Nullable(String)) -x Tuple(\n key1 Array(Array(Nullable(DateTime64(9)))),\n key2 Array(Array(Nullable(String)))) +x Tuple(\n key1 Array(Array(Nullable(DateTime))),\n key2 Array(Array(Nullable(String)))) CSV c1 Nullable(Date) c1 Nullable(DateTime64(9)) -c1 Nullable(DateTime64(9)) +c1 Nullable(DateTime) c1 Array(Nullable(Date)) -c1 Array(Nullable(DateTime64(9))) -c1 Array(Nullable(DateTime64(9))) -c1 Map(String, Nullable(DateTime64(9))) -c1 Array(Nullable(DateTime64(9))) -c1 Array(Nullable(DateTime64(9))) -c1 Nullable(DateTime64(9)) +c1 Array(Nullable(DateTime)) +c1 Array(Nullable(DateTime)) +c1 Map(String, Nullable(DateTime)) +c1 Array(Nullable(DateTime)) +c1 Array(Nullable(DateTime)) +c1 Nullable(DateTime) c1 Array(Nullable(String)) c1 Nullable(String) c1 Array(Nullable(String)) @@ -31,14 +31,14 @@ c1 Map(String, Array(Array(Nullable(String)))) TSV c1 Nullable(Date) c1 Nullable(DateTime64(9)) -c1 Nullable(DateTime64(9)) +c1 Nullable(DateTime) c1 Array(Nullable(Date)) -c1 Array(Nullable(DateTime64(9))) -c1 Array(Nullable(DateTime64(9))) -c1 Map(String, Nullable(DateTime64(9))) -c1 Array(Nullable(DateTime64(9))) -c1 Array(Nullable(DateTime64(9))) -c1 Nullable(DateTime64(9)) +c1 Array(Nullable(DateTime)) +c1 Array(Nullable(DateTime)) +c1 Map(String, Nullable(DateTime)) +c1 Array(Nullable(DateTime)) +c1 Array(Nullable(DateTime)) +c1 Nullable(DateTime) c1 Array(Nullable(String)) c1 Nullable(String) c1 Array(Nullable(String)) @@ -46,14 +46,14 @@ c1 Map(String, Array(Array(Nullable(String)))) Values c1 Nullable(Date) c1 Nullable(DateTime64(9)) -c1 Nullable(DateTime64(9)) +c1 Nullable(DateTime) c1 Array(Nullable(Date)) -c1 Array(Nullable(DateTime64(9))) -c1 Array(Nullable(DateTime64(9))) -c1 Map(String, Nullable(DateTime64(9))) -c1 Array(Nullable(DateTime64(9))) -c1 Array(Nullable(DateTime64(9))) -c1 Nullable(DateTime64(9)) +c1 Array(Nullable(DateTime)) +c1 Array(Nullable(DateTime)) +c1 Map(String, Nullable(DateTime)) +c1 Array(Nullable(DateTime)) +c1 Array(Nullable(DateTime)) +c1 Nullable(DateTime) c1 Array(Nullable(String)) c1 Nullable(String) c1 Array(Nullable(String)) diff --git a/tests/queries/0_stateless/02335_column_ttl_expired_column_optimization.sh b/tests/queries/0_stateless/02335_column_ttl_expired_column_optimization.sh index 96f80d65878..490f8361682 100755 --- a/tests/queries/0_stateless/02335_column_ttl_expired_column_optimization.sh +++ b/tests/queries/0_stateless/02335_column_ttl_expired_column_optimization.sh @@ -6,7 +6,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) data_path="$CLICKHOUSE_TMP/local" -$CLICKHOUSE_LOCAL --path "$data_path" -nm -q " +$CLICKHOUSE_LOCAL --path "$data_path" -m -q " create table ttl_02335 ( date Date, key Int, diff --git a/tests/queries/0_stateless/02337_analyzer_columns_basic.sql b/tests/queries/0_stateless/02337_analyzer_columns_basic.sql index 167eecc6fb8..c132a69ac21 100644 --- a/tests/queries/0_stateless/02337_analyzer_columns_basic.sql +++ b/tests/queries/0_stateless/02337_analyzer_columns_basic.sql @@ -1,6 +1,6 @@ -- Tags: no-parallel -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- Empty from section diff --git a/tests/queries/0_stateless/02337_multiple_joins_original_names.sql b/tests/queries/0_stateless/02337_multiple_joins_original_names.sql index 63bbfe6873d..37c7077b56e 100644 --- a/tests/queries/0_stateless/02337_multiple_joins_original_names.sql +++ b/tests/queries/0_stateless/02337_multiple_joins_original_names.sql @@ -1,6 +1,6 @@ -- https://github.com/ClickHouse/ClickHouse/issues/34697 -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT table1_id FROM ( SELECT first.table1_id diff --git a/tests/queries/0_stateless/02338_analyzer_constants_basic.sql b/tests/queries/0_stateless/02338_analyzer_constants_basic.sql index 6d6249538a4..536202dc2cf 100644 --- a/tests/queries/0_stateless/02338_analyzer_constants_basic.sql +++ b/tests/queries/0_stateless/02338_analyzer_constants_basic.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DESCRIBE (SELECT 1); SELECT 1; diff --git a/tests/queries/0_stateless/02339_analyzer_matcher_basic.sql b/tests/queries/0_stateless/02339_analyzer_matcher_basic.sql index a09325fa43b..57c6a9479ec 100644 --- a/tests/queries/0_stateless/02339_analyzer_matcher_basic.sql +++ b/tests/queries/0_stateless/02339_analyzer_matcher_basic.sql @@ -1,6 +1,6 @@ -- Tags: no-parallel -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT 'Matchers without FROM section'; diff --git a/tests/queries/0_stateless/02340_analyzer_functions.sql b/tests/queries/0_stateless/02340_analyzer_functions.sql index 101a5bfcc86..bd018302913 100644 --- a/tests/queries/0_stateless/02340_analyzer_functions.sql +++ b/tests/queries/0_stateless/02340_analyzer_functions.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DESCRIBE (SELECT 1 + 1); SELECT 1 + 1; diff --git a/tests/queries/0_stateless/02341_analyzer_aliases_basics.sql b/tests/queries/0_stateless/02341_analyzer_aliases_basics.sql index 9f21db8e659..8bed0c8bc1e 100644 --- a/tests/queries/0_stateless/02341_analyzer_aliases_basics.sql +++ b/tests/queries/0_stateless/02341_analyzer_aliases_basics.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT 'Aliases to constants'; diff --git a/tests/queries/0_stateless/02341_global_join_cte.reference b/tests/queries/0_stateless/02341_global_join_cte.reference index f2cfe994ffa..4f854ca3bef 100644 --- a/tests/queries/0_stateless/02341_global_join_cte.reference +++ b/tests/queries/0_stateless/02341_global_join_cte.reference @@ -1,6 +1,6 @@ -- { echo } -with rhs as (select * from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one))) select lhs.d2 from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one)) lhs global join rhs using (d1) order by rhs.d2 settings allow_experimental_analyzer=0; -- { serverError ALIAS_REQUIRED } -with rhs as (select * from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one))) select lhs.d2 from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one)) lhs global join rhs using (d1) order by rhs.d2 settings allow_experimental_analyzer=1; -- It works with analyzer; rhs is an alias itself. +with rhs as (select * from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one))) select lhs.d2 from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one)) lhs global join rhs using (d1) order by rhs.d2 settings enable_analyzer=0; -- { serverError ALIAS_REQUIRED } +with rhs as (select * from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one))) select lhs.d2 from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one)) lhs global join rhs using (d1) order by rhs.d2 settings enable_analyzer=1; -- It works with analyzer; rhs is an alias itself. 0 0 0 diff --git a/tests/queries/0_stateless/02341_global_join_cte.sql b/tests/queries/0_stateless/02341_global_join_cte.sql index b9b906afd70..f6acd822f14 100644 --- a/tests/queries/0_stateless/02341_global_join_cte.sql +++ b/tests/queries/0_stateless/02341_global_join_cte.sql @@ -1,5 +1,5 @@ -- { echo } -with rhs as (select * from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one))) select lhs.d2 from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one)) lhs global join rhs using (d1) order by rhs.d2 settings allow_experimental_analyzer=0; -- { serverError ALIAS_REQUIRED } -with rhs as (select * from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one))) select lhs.d2 from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one)) lhs global join rhs using (d1) order by rhs.d2 settings allow_experimental_analyzer=1; -- It works with analyzer; rhs is an alias itself. +with rhs as (select * from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one))) select lhs.d2 from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one)) lhs global join rhs using (d1) order by rhs.d2 settings enable_analyzer=0; -- { serverError ALIAS_REQUIRED } +with rhs as (select * from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one))) select lhs.d2 from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one)) lhs global join rhs using (d1) order by rhs.d2 settings enable_analyzer=1; -- It works with analyzer; rhs is an alias itself. with rhs as (select * from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one))) select lhs.d2 from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one)) lhs global join rhs using (d1) order by rhs.d2 settings joined_subquery_requires_alias=0; with rhs_ as (select * from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one))) select lhs.d2 from remote('127.{1,2}', view(select dummy d1, dummy d2 from system.one)) lhs global join rhs_ rhs using (d1) order by rhs.d2 settings joined_subquery_requires_alias=0; diff --git a/tests/queries/0_stateless/02342_analyzer_compound_types.sql b/tests/queries/0_stateless/02342_analyzer_compound_types.sql index 0fd96928496..36617aab2f8 100644 --- a/tests/queries/0_stateless/02342_analyzer_compound_types.sql +++ b/tests/queries/0_stateless/02342_analyzer_compound_types.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT 'Constant tuple'; diff --git a/tests/queries/0_stateless/02342_window_view_different_struct.sql b/tests/queries/0_stateless/02342_window_view_different_struct.sql index a5b2b8daa5a..9c35459ecef 100644 --- a/tests/queries/0_stateless/02342_window_view_different_struct.sql +++ b/tests/queries/0_stateless/02342_window_view_different_struct.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SET allow_experimental_window_view = 1; DROP TABLE IF EXISTS data_02342; diff --git a/tests/queries/0_stateless/02343_analyzer_column_transformers_strict.sql b/tests/queries/0_stateless/02343_analyzer_column_transformers_strict.sql index 7e323c570b8..b55cb85c0da 100644 --- a/tests/queries/0_stateless/02343_analyzer_column_transformers_strict.sql +++ b/tests/queries/0_stateless/02343_analyzer_column_transformers_strict.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02343_analyzer_lambdas.sql b/tests/queries/0_stateless/02343_analyzer_lambdas.sql index 80fa47fc325..07f382700ae 100644 --- a/tests/queries/0_stateless/02343_analyzer_lambdas.sql +++ b/tests/queries/0_stateless/02343_analyzer_lambdas.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02343_analyzer_lambdas_issue_28083.sql b/tests/queries/0_stateless/02343_analyzer_lambdas_issue_28083.sql index 3b780e1dec3..dff0d7f82ca 100644 --- a/tests/queries/0_stateless/02343_analyzer_lambdas_issue_28083.sql +++ b/tests/queries/0_stateless/02343_analyzer_lambdas_issue_28083.sql @@ -1,11 +1,11 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; select so, r from - (select [('y',0),('n',1)] as cg, + (select [('y',0),('n',1)] as cg, if( arrayMap( x -> x.1, cg ) != ['y', 'n'], 'y', 'n') as so, - arrayFilter( x -> x.1 = so , cg) as r + arrayFilter( x -> x.1 = so , cg) as r ); select diff --git a/tests/queries/0_stateless/02343_analyzer_lambdas_issue_36677.sql b/tests/queries/0_stateless/02343_analyzer_lambdas_issue_36677.sql index b07f3f33ac3..da0b4e8ef57 100644 --- a/tests/queries/0_stateless/02343_analyzer_lambdas_issue_36677.sql +++ b/tests/queries/0_stateless/02343_analyzer_lambdas_issue_36677.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT arraySum(x -> ((x.1) / ((x.2) * (x.2))), arrayZip(mag, magerr)) / arraySum(x -> (1. / (x * x)), magerr) AS weightedmeanmag, diff --git a/tests/queries/0_stateless/02344_analyzer_multiple_aliases_for_expression.sql b/tests/queries/0_stateless/02344_analyzer_multiple_aliases_for_expression.sql index ee02b79cc32..3c7ea467734 100644 --- a/tests/queries/0_stateless/02344_analyzer_multiple_aliases_for_expression.sql +++ b/tests/queries/0_stateless/02344_analyzer_multiple_aliases_for_expression.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02345_analyzer_subqueries.sql b/tests/queries/0_stateless/02345_analyzer_subqueries.sql index c0cc242b57b..d1ec9b58e27 100644 --- a/tests/queries/0_stateless/02345_analyzer_subqueries.sql +++ b/tests/queries/0_stateless/02345_analyzer_subqueries.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02346_fulltext_index_bug52019.sql b/tests/queries/0_stateless/02346_fulltext_index_bug52019.sql index a643df65095..e29c3c51e5e 100644 --- a/tests/queries/0_stateless/02346_fulltext_index_bug52019.sql +++ b/tests/queries/0_stateless/02346_fulltext_index_bug52019.sql @@ -13,8 +13,8 @@ ORDER BY k SETTINGS index_granularity = 2, index_granularity_bytes = '10Mi'; INSERT INTO tab (k) VALUES (0); -SELECT * FROM tab PREWHERE (s[NULL]) = 'Click a03' SETTINGS allow_experimental_analyzer=1; -SELECT * FROM tab PREWHERE (s[1]) = 'Click a03' SETTINGS allow_experimental_analyzer=1; -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } -SELECT * FROM tab PREWHERE (s['foo']) = 'Click a03' SETTINGS allow_experimental_analyzer=1; +SELECT * FROM tab PREWHERE (s[NULL]) = 'Click a03' SETTINGS enable_analyzer=1; +SELECT * FROM tab PREWHERE (s[1]) = 'Click a03' SETTINGS enable_analyzer=1; -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +SELECT * FROM tab PREWHERE (s['foo']) = 'Click a03' SETTINGS enable_analyzer=1; DROP TABLE tab; diff --git a/tests/queries/0_stateless/02346_fulltext_index_match_predicate.sql b/tests/queries/0_stateless/02346_fulltext_index_match_predicate.sql index 7f36c423a41..a20c8dc9afd 100644 --- a/tests/queries/0_stateless/02346_fulltext_index_match_predicate.sql +++ b/tests/queries/0_stateless/02346_fulltext_index_match_predicate.sql @@ -31,7 +31,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 0; + enable_analyzer = 0; SELECT * FROM @@ -42,7 +42,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 1; + enable_analyzer = 1; SELECT '---'; @@ -61,7 +61,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 0; + enable_analyzer = 0; SELECT * FROM @@ -72,7 +72,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 1; + enable_analyzer = 1; SELECT '---'; @@ -91,7 +91,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 0; + enable_analyzer = 0; SELECT * FROM @@ -102,6 +102,6 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 1; + enable_analyzer = 1; DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_bugs.reference b/tests/queries/0_stateless/02354_vector_search_bugs.reference index d2c2d7e2fb7..a27b086e118 100644 --- a/tests/queries/0_stateless/02354_vector_search_bugs.reference +++ b/tests/queries/0_stateless/02354_vector_search_bugs.reference @@ -1,17 +1,5 @@ Issue #52258: Empty Arrays or Arrays with default values are rejected -- Annoy -- Usearch It is possible to create parts with different Array vector sizes but there will be an error at query time -- Annoy -- Usearch Correctness of index with > 1 mark -- Annoy 1 [1,0] 0 9000 [9000,0] 0 -1 (1,0) 0 -9000 (9000,0) 0 -- Usearch -1 [1,0] 0 -9000 [9000,0] 0 -1 (1,0) 0 -9000 (9000,0) 0 diff --git a/tests/queries/0_stateless/02354_vector_search_bugs.sql b/tests/queries/0_stateless/02354_vector_search_bugs.sql index db0c06c4e16..7c66b4b8e45 100644 --- a/tests/queries/0_stateless/02354_vector_search_bugs.sql +++ b/tests/queries/0_stateless/02354_vector_search_bugs.sql @@ -1,61 +1,22 @@ --- Tags: no-fasttest, no-ubsan, no-cpu-aarch64, no-ordinary-database, no-asan +-- Tags: no-fasttest, no-ordinary-database --- Tests vector search in ClickHouse, i.e. Annoy and Usearch indexes. Both index types share similarities in implementation and usage, --- therefore they are tested in a single file. +-- Tests various bugs and special cases for vector indexes. --- This file contains tests for various bugs and special cases - -SET allow_experimental_annoy_index = 1; -SET allow_experimental_usearch_index = 1; - -SET allow_experimental_analyzer = 1; -- 0 vs. 1 produce slightly different error codes, make it future-proof +SET allow_experimental_vector_similarity_index = 1; +SET enable_analyzer = 1; -- 0 vs. 1 produce slightly different error codes, make it future-proof DROP TABLE IF EXISTS tab; SELECT 'Issue #52258: Empty Arrays or Arrays with default values are rejected'; -SELECT '- Annoy'; - -CREATE TABLE tab (id UInt64, vec Array(Float32), INDEX idx vec TYPE annoy()) ENGINE = MergeTree() ORDER BY (id); +CREATE TABLE tab (id UInt64, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree() ORDER BY id; INSERT INTO tab VALUES (1, []); -- { serverError INCORRECT_DATA } INSERT INTO tab (id) VALUES (1); -- { serverError INCORRECT_DATA } DROP TABLE tab; -CREATE TABLE tab (id UInt64, vec Tuple(Float32, Float32), INDEX idx vec TYPE annoy()) ENGINE = MergeTree() ORDER BY (id); -INSERT INTO tab (id) VALUES (1); -- works fine, takes on default tuple (0.0, 0.0) -DROP TABLE tab; - -SELECT '- Usearch'; - -CREATE TABLE tab (id UInt64, vec Array(Float32), INDEX idx vec TYPE usearch()) ENGINE = MergeTree() ORDER BY (id); -INSERT INTO tab VALUES (1, []); -- { serverError INCORRECT_DATA } -INSERT INTO tab (id) VALUES (1); -- { serverError INCORRECT_DATA } -DROP TABLE tab; - -CREATE TABLE tab (id UInt64, vec Tuple(Float32, Float32), INDEX idx vec TYPE usearch()) ENGINE = MergeTree() ORDER BY (id); -INSERT INTO tab (id) VALUES (1); -- works fine, takes on default tuple (0.0, 0.0) -DROP TABLE tab; - SELECT 'It is possible to create parts with different Array vector sizes but there will be an error at query time'; -SELECT '- Annoy'; - -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE annoy()) ENGINE = MergeTree ORDER BY id; -SYSTEM STOP MERGES tab; -INSERT INTO tab values (0, [2.2, 2.3]) (1, [3.1, 3.2]); -INSERT INTO tab values (2, [2.2, 2.3, 2.4]) (3, [3.1, 3.2, 3.3]); - -WITH [0.0, 2.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab -ORDER BY L2Distance(vec, reference_vec) -LIMIT 3; -- { serverError SIZES_OF_ARRAYS_DONT_MATCH } - -DROP TABLE tab; - -SELECT '- Usearch'; - -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE usearch()) ENGINE = MergeTree ORDER BY id; +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id; SYSTEM STOP MERGES tab; INSERT INTO tab values (0, [2.2, 2.3]) (1, [3.1, 3.2]); INSERT INTO tab values (2, [2.2, 2.3, 2.4]) (3, [3.1, 3.2, 3.3]); @@ -70,9 +31,7 @@ DROP TABLE tab; SELECT 'Correctness of index with > 1 mark'; -SELECT '- Annoy'; - -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE annoy()) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity_bytes=0, min_rows_for_wide_part = 0, min_bytes_for_wide_part = 0, index_granularity=8192; -- disable adaptive granularity due to bug +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity_bytes = 0, min_rows_for_wide_part = 0, min_bytes_for_wide_part = 0, index_granularity = 8192; -- disable adaptive granularity due to bug INSERT INTO tab SELECT number, [toFloat32(number), 0.0] from numbers(10000); WITH [1.0, 0.0] AS reference_vec @@ -88,58 +47,3 @@ ORDER BY L2Distance(vec, reference_vec) LIMIT 1; DROP TABLE tab; - --- same, but with Tuples -CREATE TABLE tab(id Int32, vec Tuple(Float32, Float32), INDEX idx vec TYPE annoy()) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity_bytes=0, min_rows_for_wide_part = 0, min_bytes_for_wide_part = 0, index_granularity=8192; -- disable adaptive granularity due to bug -INSERT INTO tab SELECT number, (toFloat32(number), 0.0) from numbers(10000); - -WITH (1.0, 0.0) AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab -ORDER BY L2Distance(vec, reference_vec) -LIMIT 1; - -WITH (9000.0, 0.0) AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab -ORDER BY L2Distance(vec, reference_vec) -LIMIT 1; - -DROP TABLE tab; - -SELECT '- Usearch'; - -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE usearch()) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity_bytes=0, min_rows_for_wide_part = 0, min_bytes_for_wide_part = 0, index_granularity=8192; -- disable adaptive granularity due to bug -INSERT INTO tab SELECT number, [toFloat32(number), 0.0] from numbers(10000); - -WITH [1.0, 0.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab -ORDER BY L2Distance(vec, reference_vec) -LIMIT 1; - -WITH [9000.0, 0.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab -ORDER BY L2Distance(vec, reference_vec) -LIMIT 1; - -DROP TABLE tab; - --- same, but with Tuples -CREATE TABLE tab(id Int32, vec Tuple(Float32, Float32), INDEX idx vec TYPE usearch()) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity_bytes=0, min_rows_for_wide_part = 0, min_bytes_for_wide_part = 0, index_granularity=8192; -- disable adaptive granularity due to bug -INSERT INTO tab SELECT number, (toFloat32(number), 0.0) from numbers(10000); - -WITH (1.0, 0.0) AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab -ORDER BY L2Distance(vec, reference_vec) -LIMIT 1; - -WITH (9000.0, 0.0) AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab -ORDER BY L2Distance(vec, reference_vec) -LIMIT 1; - -DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_default_granularity.reference b/tests/queries/0_stateless/02354_vector_search_default_granularity.reference index 2f97ed72c52..ab3cc71ff6b 100644 --- a/tests/queries/0_stateless/02354_vector_search_default_granularity.reference +++ b/tests/queries/0_stateless/02354_vector_search_default_granularity.reference @@ -1,7 +1,2 @@ -Test the default index granularity for vector search indexes (CREATE TABLE AND ALTER TABLE), should be 100 million for Annoy and USearch -- Annoy -100000000 -100000000 -- Usearch 100000000 100000000 diff --git a/tests/queries/0_stateless/02354_vector_search_default_granularity.sql b/tests/queries/0_stateless/02354_vector_search_default_granularity.sql index f15554505f8..acb69cb6ff8 100644 --- a/tests/queries/0_stateless/02354_vector_search_default_granularity.sql +++ b/tests/queries/0_stateless/02354_vector_search_default_granularity.sql @@ -1,35 +1,18 @@ --- Tags: no-fasttest, no-ubsan, no-cpu-aarch64, no-ordinary-database, no-asan +-- Tags: no-fasttest, no-ordinary-database --- Tests vector search in ClickHouse, i.e. Annoy and Usearch indexes. Both index types share similarities in implementation and usage, --- therefore they are tested in a single file. +-- Tests that vector search indexes use a (non-standard) index granularity of 100 mio by default. --- This file contains tests for the non-standard default granularity of vector search indexes. - -SET allow_experimental_annoy_index = 1; -SET allow_experimental_usearch_index = 1; - -SELECT 'Test the default index granularity for vector search indexes (CREATE TABLE AND ALTER TABLE), should be 100 million for Annoy and USearch'; - -SELECT '- Annoy'; +SET allow_experimental_vector_similarity_index = 1; +-- After CREATE TABLE DROP TABLE IF EXISTS tab; -CREATE TABLE tab (id Int32, vec Array(Float32), INDEX idx(vec) TYPE annoy) ENGINE=MergeTree ORDER BY id; -SELECT granularity FROM system.data_skipping_indices WHERE database = currentDatabase() AND table = 'tab' AND name = 'idx'; - -DROP TABLE tab; -CREATE TABLE tab (id Int32, vec Array(Float32)) ENGINE=MergeTree ORDER BY id; -ALTER TABLE tab ADD INDEX idx(vec) TYPE annoy; -SELECT granularity FROM system.data_skipping_indices WHERE database = currentDatabase() AND table = 'tab' AND name = 'idx'; - -SELECT '- Usearch'; - -DROP TABLE tab; -CREATE TABLE tab (id Int32, vec Array(Float32), INDEX idx(vec) TYPE usearch) ENGINE=MergeTree ORDER BY id; -SELECT granularity FROM system.data_skipping_indices WHERE database = currentDatabase() AND table = 'tab' AND name = 'idx'; - -DROP TABLE tab; -CREATE TABLE tab (id Int32, vec Array(Float32)) ENGINE=MergeTree ORDER BY id; -ALTER TABLE tab ADD INDEX idx(vec) TYPE usearch; +CREATE TABLE tab (id Int32, vec Array(Float32), INDEX idx(vec) TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id; +SELECT granularity FROM system.data_skipping_indices WHERE database = currentDatabase() AND table = 'tab' AND name = 'idx'; + +-- After ALTER TABLE +DROP TABLE tab; +CREATE TABLE tab (id Int32, vec Array(Float32)) ENGINE = MergeTree ORDER BY id; +ALTER TABLE tab ADD INDEX idx(vec) TYPE vector_similarity('hnsw', 'L2Distance'); SELECT granularity FROM system.data_skipping_indices WHERE database = currentDatabase() AND table = 'tab' AND name = 'idx'; DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_detach_attach.reference b/tests/queries/0_stateless/02354_vector_search_detach_attach.reference new file mode 100644 index 00000000000..80eb091922e --- /dev/null +++ b/tests/queries/0_stateless/02354_vector_search_detach_attach.reference @@ -0,0 +1,3 @@ +5 [0,2] 0 +6 [0,2.1] 0.09999990463256836 +7 [0,2.2] 0.20000004768371582 diff --git a/tests/queries/0_stateless/02354_vector_search_detach_attach.sql b/tests/queries/0_stateless/02354_vector_search_detach_attach.sql new file mode 100644 index 00000000000..f92eaddbbed --- /dev/null +++ b/tests/queries/0_stateless/02354_vector_search_detach_attach.sql @@ -0,0 +1,20 @@ +-- Tags: no-fasttest, no-ordinary-database + +-- Tests that vector similarity indexes can be detached/attached. + +SET allow_experimental_vector_similarity_index = 1; + +DROP TABLE IF EXISTS tab; +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity = 8192; +INSERT INTO tab VALUES (0, [1.0, 0.0]), (1, [1.1, 0.0]), (2, [1.2, 0.0]), (3, [1.3, 0.0]), (4, [1.4, 0.0]), (5, [0.0, 2.0]), (6, [0.0, 2.1]), (7, [0.0, 2.2]), (8, [0.0, 2.3]), (9, [0.0, 2.4]); + +DETACH TABLE tab SYNC; +ATTACH TABLE tab; + +WITH [0.0, 2.0] AS reference_vec +SELECT id, vec, L2Distance(vec, reference_vec) +FROM tab +ORDER BY L2Distance(vec, reference_vec) +LIMIT 3; + +DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_index_creation_negative.reference b/tests/queries/0_stateless/02354_vector_search_index_creation_negative.reference index 43bc49e8adc..f18daa6e02e 100644 --- a/tests/queries/0_stateless/02354_vector_search_index_creation_negative.reference +++ b/tests/queries/0_stateless/02354_vector_search_index_creation_negative.reference @@ -1,8 +1,8 @@ -At most two index arguments -1st argument (distance function) must be String -Rejects unsupported distance functions -2nd argument (Annoy: number of trees, USearch: scalar kind) must be UInt64 (Annoy) / String (Usearch) -Rejects unsupported scalar kinds (only Usearch) +Two or six index arguments +1st argument (method) must be String and hnsw +2nd argument (distance function) must be String and L2Distance or cosineDistance +3nd argument (quantization), if given, must be String and f32, f16, ... +4nd argument (M), if given, must be UInt64 and > 1 Must be created on single column -Must be created on Array(Float32) or Tuple(Float32, Float, ...) columns +Must be created on Array(Float32) columns Rejects INSERTs of Arrays with different sizes diff --git a/tests/queries/0_stateless/02354_vector_search_index_creation_negative.sql b/tests/queries/0_stateless/02354_vector_search_index_creation_negative.sql index 6a4d6448629..de9d37e1000 100644 --- a/tests/queries/0_stateless/02354_vector_search_index_creation_negative.sql +++ b/tests/queries/0_stateless/02354_vector_search_index_creation_negative.sql @@ -1,63 +1,45 @@ --- Tags: no-fasttest, no-ubsan, no-cpu-aarch64, no-ordinary-database, no-asan +-- Tags: no-fasttest, no-ordinary-database --- Tests vector search in ClickHouse, i.e. Annoy and Usearch indexes. Both index types share similarities in implementation and usage, --- therefore they are tested in a single file. +-- Tests that various conditions are checked during creation of vector search indexes. --- This file tests that various conditions are checked during creation of vector search indexes. - -SET allow_experimental_annoy_index = 1; -SET allow_experimental_usearch_index = 1; +SET allow_experimental_vector_similarity_index = 1; DROP TABLE IF EXISTS tab; -SELECT 'At most two index arguments'; -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE annoy('too', 'many', 'arguments')) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_QUERY } -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE usearch('too', 'many', 'args')) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_QUERY } +SELECT 'Two or six index arguments'; +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_QUERY } +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity()) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_QUERY } +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('cant_have_one_arg')) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_QUERY } +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('cant', 'have', 'three_args')) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_QUERY } +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('cant', 'have', 'more', 'than', 'six', 'args', '!')) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_QUERY } -SELECT '1st argument (distance function) must be String'; -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE annoy(3)) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_QUERY } -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE usearch(3)) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_QUERY } +SELECT '1st argument (method) must be String and hnsw'; +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity(3, 'L2Distance')) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_QUERY } +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('not_hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_DATA } -SELECT 'Rejects unsupported distance functions'; -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE annoy('invalidDistance')) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_DATA } -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE usearch('invalidDistance')) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_DATA } +SELECT '2nd argument (distance function) must be String and L2Distance or cosineDistance'; +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 3)) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_QUERY } +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'invalid_distance')) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_DATA } -SELECT '2nd argument (Annoy: number of trees, USearch: scalar kind) must be UInt64 (Annoy) / String (Usearch)'; -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE annoy('L2Distance', 'not an UInt64')) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_QUERY } -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE usearch(3)) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_QUERY } - -SELECT 'Rejects unsupported scalar kinds (only Usearch)'; -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE usearch('L2Distance', 'invalidKind')) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_DATA } +SELECT '3nd argument (quantization), if given, must be String and f32, f16, ...'; +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance', 1, 1, 1, 1)) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_QUERY } +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance', 'invalid', 2, 1, 1)) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_DATA } +SELECT '4nd argument (M), if given, must be UInt64 and > 1'; +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance', 'f32', 'invalid', 1, 1)) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_QUERY } +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance', 'f32', 1, 1, 1)) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_DATA } SELECT 'Must be created on single column'; -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx (vec, id) TYPE annoy()) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_NUMBER_OF_COLUMNS } -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx (vec, id) TYPE usearch()) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_NUMBER_OF_COLUMNS } - -SELECT 'Must be created on Array(Float32) or Tuple(Float32, Float, ...) columns'; +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx (vec, id) TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id; -- { serverError INCORRECT_NUMBER_OF_COLUMNS } +SELECT 'Must be created on Array(Float32) columns'; SET allow_suspicious_low_cardinality_types = 1; - -CREATE TABLE tab(id Int32, vec Float32, INDEX idx vec TYPE annoy()) ENGINE = MergeTree ORDER BY id; -- { serverError ILLEGAL_COLUMN } -CREATE TABLE tab(id Int32, vec Float32, INDEX idx vec TYPE usearch()) ENGINE = MergeTree ORDER BY id; -- { serverError ILLEGAL_COLUMN } - -CREATE TABLE tab(id Int32, vec Array(Float64), INDEX idx vec TYPE annoy()) ENGINE = MergeTree ORDER BY id; -- { serverError ILLEGAL_COLUMN } -CREATE TABLE tab(id Int32, vec Array(Float64), INDEX idx vec TYPE usearch()) ENGINE = MergeTree ORDER BY id; -- { serverError ILLEGAL_COLUMN } - -CREATE TABLE tab(id Int32, vec Tuple(Float64), INDEX idx vec TYPE annoy()) ENGINE = MergeTree ORDER BY id; -- { serverError ILLEGAL_COLUMN } -CREATE TABLE tab(id Int32, vec Tuple(Float64), INDEX idx vec TYPE usearch()) ENGINE = MergeTree ORDER BY id; -- { serverError ILLEGAL_COLUMN } - -CREATE TABLE tab(id Int32, vec LowCardinality(Float32), INDEX idx vec TYPE annoy()) ENGINE = MergeTree ORDER BY id; -- { serverError ILLEGAL_COLUMN } -CREATE TABLE tab(id Int32, vec LowCardinality(Float32), INDEX idx vec TYPE usearch()) ENGINE = MergeTree ORDER BY id; -- { serverError ILLEGAL_COLUMN } - -CREATE TABLE tab(id Int32, vec Nullable(Float32), INDEX idx vec TYPE annoy()) ENGINE = MergeTree ORDER BY id; -- { serverError ILLEGAL_COLUMN } -CREATE TABLE tab(id Int32, vec Nullable(Float32), INDEX idx vec TYPE usearch()) ENGINE = MergeTree ORDER BY id; -- { serverError ILLEGAL_COLUMN } +CREATE TABLE tab(id Int32, vec UInt64, INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id; -- { serverError ILLEGAL_COLUMN } +CREATE TABLE tab(id Int32, vec Float32, INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id; -- { serverError ILLEGAL_COLUMN } +CREATE TABLE tab(id Int32, vec Array(Float64), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id; -- { serverError ILLEGAL_COLUMN } +CREATE TABLE tab(id Int32, vec LowCardinality(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id; -- { serverError ILLEGAL_COLUMN } +CREATE TABLE tab(id Int32, vec Nullable(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id; -- { serverError ILLEGAL_COLUMN } SELECT 'Rejects INSERTs of Arrays with different sizes'; - -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE annoy()) ENGINE = MergeTree ORDER BY id; -INSERT INTO tab values (0, [2.2, 2.3]) (1, [3.1, 3.2, 3.3]); -- { serverError INCORRECT_DATA } -DROP TABLE tab; - -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE usearch()) ENGINE = MergeTree ORDER BY id; +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id; INSERT INTO tab values (0, [2.2, 2.3]) (1, [3.1, 3.2, 3.3]); -- { serverError INCORRECT_DATA } DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_legacy_index_compatibility.reference b/tests/queries/0_stateless/02354_vector_search_legacy_index_compatibility.reference new file mode 100644 index 00000000000..030bfa9b1bd --- /dev/null +++ b/tests/queries/0_stateless/02354_vector_search_legacy_index_compatibility.reference @@ -0,0 +1,2 @@ +Annoy +Usearch diff --git a/tests/queries/0_stateless/02354_vector_search_legacy_index_compatibility.sql b/tests/queries/0_stateless/02354_vector_search_legacy_index_compatibility.sql new file mode 100644 index 00000000000..0889aa74f7a --- /dev/null +++ b/tests/queries/0_stateless/02354_vector_search_legacy_index_compatibility.sql @@ -0,0 +1,43 @@ +-- Indexes of type 'annoy' or 'usearch' are no longer supported. +-- Test what happens when ClickHouse encounters tables with the old index type. + +DROP TABLE IF EXISTS tab; + +SELECT 'Annoy'; + +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX vec_idx vec TYPE annoy()) ENGINE = MergeTree ORDER BY id; + +INSERT INTO tab VALUES (0, [1.0, 0.0]), (1, [1.1, 0.0]), (2, [1.2, 0.0]), (3, [1.3, 0.0]), (4, [1.4, 0.0]), (5, [0.0, 2.0]), (6, [0.0, 2.1]), (7, [0.0, 2.2]), (8, [0.0, 2.3]), (9, [0.0, 2.4]); -- { serverError ILLEGAL_INDEX } + +WITH [0.0, 2.0] AS reference_vec +SELECT id, vec, L2Distance(vec, reference_vec) +FROM tab +ORDER BY L2Distance(vec, reference_vec) +LIMIT 3; +-- (*) The search succeeds because the index contains no data (i.e. some shortcut) +-- If it had data (can't really test in SQL tests ...), this statement would also return an error, trust me. + +-- Detach and attach should work. +DETACH TABLE tab; +ATTACH TABLE tab; + +DROP TABLE tab; + +SELECT 'Usearch'; + +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX vec_idx vec TYPE usearch()) ENGINE = MergeTree ORDER BY id; + +INSERT INTO tab VALUES (0, [1.0, 0.0]), (1, [1.1, 0.0]), (2, [1.2, 0.0]), (3, [1.3, 0.0]), (4, [1.4, 0.0]), (5, [0.0, 2.0]), (6, [0.0, 2.1]), (7, [0.0, 2.2]), (8, [0.0, 2.3]), (9, [0.0, 2.4]); -- { serverError ILLEGAL_INDEX } + +WITH [0.0, 2.0] AS reference_vec +SELECT id, vec, L2Distance(vec, reference_vec) +FROM tab +ORDER BY L2Distance(vec, reference_vec) +LIMIT 3; +-- see above: (*) + +-- Detach and attach should work. +DETACH TABLE tab; +ATTACH TABLE tab; + +DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_multiple_indexes.reference b/tests/queries/0_stateless/02354_vector_search_multiple_indexes.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/02354_vector_search_multiple_indexes.sql b/tests/queries/0_stateless/02354_vector_search_multiple_indexes.sql new file mode 100644 index 00000000000..f1cfc041233 --- /dev/null +++ b/tests/queries/0_stateless/02354_vector_search_multiple_indexes.sql @@ -0,0 +1,14 @@ +-- Tags: no-fasttest, no-ordinary-database + +-- Tests that multiple vector similarity indexes can be created on the same column (even if that makes no sense) + +SET allow_experimental_vector_similarity_index = 1; + +DROP TABLE IF EXISTS tab; +CREATE TABLE tab (id Int32, vec Array(Float32), PRIMARY KEY id, INDEX vec_idx(vec) TYPE vector_similarity('hnsw', 'L2Distance')); + +ALTER TABLE tab ADD INDEX idx(vec) TYPE minmax; +ALTER TABLE tab ADD INDEX vec_idx1(vec) TYPE vector_similarity('hnsw', 'cosineDistance'); +ALTER TABLE tab ADD INDEX vec_idx2(vec) TYPE vector_similarity('hnsw', 'L2Distance'); -- silly but creating the same index also works for non-vector indexes ... + +DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_queries.reference b/tests/queries/0_stateless/02354_vector_search_queries.reference index 41c1915ecc3..7c8e4c0ca59 100644 --- a/tests/queries/0_stateless/02354_vector_search_queries.reference +++ b/tests/queries/0_stateless/02354_vector_search_queries.reference @@ -1,41 +1,14 @@ -ARRAY, 10 rows, index_granularity = 8192, GRANULARITY = 1 million --> 1 granule, 1 indexed block -- Annoy: WHERE-type +10 rows, index_granularity = 8192, GRANULARITY = 1 million --> 1 granule, 1 indexed block +- ORDER-BY-type 5 [0,2] 0 6 [0,2.1] 0.09999990463256836 7 [0,2.2] 0.20000004768371582 -- Annoy: ORDER-BY-type -5 [0,2] 0 -6 [0,2.1] 0.09999990463256836 -7 [0,2.2] 0.20000004768371582 -- Usearch: WHERE-type -5 [0,2] 0 -6 [0,2.1] 0.09999990463256836 -7 [0,2.2] 0.20000004768371582 -- Usearch: ORDER-BY-type -5 [0,2] 0 -6 [0,2.1] 0.09999990463256836 -7 [0,2.2] 0.20000004768371582 -- Annoy: WHERE-type, EXPLAIN -Expression ((Projection + Before ORDER BY)) - Limit (preliminary LIMIT (without OFFSET)) - Expression - ReadFromMergeTree (default.tab_annoy) - Indexes: - PrimaryKey - Condition: true - Parts: 1/1 - Granules: 1/1 - Skip - Name: idx - Description: annoy GRANULARITY 100000000 - Parts: 1/1 - Granules: 1/1 -- Annoy: ORDER-BY-type, EXPLAIN +- ORDER-BY-type, EXPLAIN Expression (Projection) Limit (preliminary LIMIT (without OFFSET)) Sorting (Sorting for ORDER BY) Expression (Before ORDER BY) - ReadFromMergeTree (default.tab_annoy) + ReadFromMergeTree (default.tab) Indexes: PrimaryKey Condition: true @@ -43,78 +16,20 @@ Expression (Projection) Granules: 1/1 Skip Name: idx - Description: annoy GRANULARITY 100000000 + Description: vector_similarity GRANULARITY 100000000 Parts: 1/1 Granules: 1/1 -- Usearch: WHERE-type, EXPLAIN -Expression ((Projection + Before ORDER BY)) - Limit (preliminary LIMIT (without OFFSET)) - Expression - ReadFromMergeTree (default.tab_usearch) - Indexes: - PrimaryKey - Condition: true - Parts: 1/1 - Granules: 1/1 - Skip - Name: idx - Description: usearch GRANULARITY 100000000 - Parts: 1/1 - Granules: 1/1 -- Usearch: ORDER-BY-type, EXPLAIN +12 rows, index_granularity = 3, GRANULARITY = 2 --> 4 granules, 2 indexed block +- ORDER-BY-type +6 [0,2] 0 +7 [0,2.1] 0.09999990463256836 +8 [0,2.2] 0.20000004768371582 +- ORDER-BY-type, EXPLAIN Expression (Projection) Limit (preliminary LIMIT (without OFFSET)) Sorting (Sorting for ORDER BY) Expression (Before ORDER BY) - ReadFromMergeTree (default.tab_usearch) - Indexes: - PrimaryKey - Condition: true - Parts: 1/1 - Granules: 1/1 - Skip - Name: idx - Description: usearch GRANULARITY 100000000 - Parts: 1/1 - Granules: 1/1 -ARRAY vectors, 12 rows, index_granularity = 3, GRANULARITY = 2 --> 4 granules, 2 indexed block -- Annoy: WHERE-type -6 [0,2] 0 -7 [0,2.1] 0.09999990463256836 -8 [0,2.2] 0.20000004768371582 -- Annoy: ORDER-BY-type -6 [0,2] 0 -7 [0,2.1] 0.09999990463256836 -8 [0,2.2] 0.20000004768371582 -- Usearch: WHERE-type -6 [0,2] 0 -7 [0,2.1] 0.09999990463256836 -8 [0,2.2] 0.20000004768371582 -- Usearch: ORDER-BY-type -6 [0,2] 0 -7 [0,2.1] 0.09999990463256836 -8 [0,2.2] 0.20000004768371582 -- Annoy: WHERE-type, EXPLAIN -Expression ((Projection + Before ORDER BY)) - Limit (preliminary LIMIT (without OFFSET)) - Expression - ReadFromMergeTree (default.tab_annoy) - Indexes: - PrimaryKey - Condition: true - Parts: 1/1 - Granules: 4/4 - Skip - Name: idx - Description: annoy GRANULARITY 2 - Parts: 1/1 - Granules: 1/4 -- Annoy: ORDER-BY-type, EXPLAIN -Expression (Projection) - Limit (preliminary LIMIT (without OFFSET)) - Sorting (Sorting for ORDER BY) - Expression (Before ORDER BY) - ReadFromMergeTree (default.tab_annoy) + ReadFromMergeTree (default.tab) Indexes: PrimaryKey Condition: true @@ -122,77 +37,20 @@ Expression (Projection) Granules: 4/4 Skip Name: idx - Description: annoy GRANULARITY 2 + Description: vector_similarity GRANULARITY 2 Parts: 1/1 Granules: 2/4 -- Usearch: WHERE-type, EXPLAIN -Expression ((Projection + Before ORDER BY)) - Limit (preliminary LIMIT (without OFFSET)) - Expression - ReadFromMergeTree (default.tab_usearch) - Indexes: - PrimaryKey - Condition: true - Parts: 1/1 - Granules: 4/4 - Skip - Name: idx - Description: usearch GRANULARITY 2 - Parts: 1/1 - Granules: 1/4 -- Usearch: ORDER-BY-type, EXPLAIN -Expression (Projection) - Limit (preliminary LIMIT (without OFFSET)) - Sorting (Sorting for ORDER BY) - Expression (Before ORDER BY) - ReadFromMergeTree (default.tab_usearch) - Indexes: - PrimaryKey - Condition: true - Parts: 1/1 - Granules: 4/4 - Skip - Name: idx - Description: usearch GRANULARITY 2 - Parts: 1/1 - Granules: 2/4 -TUPLE vectors and special cases -- Annoy: WHERE-type -0 (4.6,2.3) 0.5527864045000421 -1 (2,3.2) 0.15200169244542905 -2 (4.2,3.4) 0.37080174340866845 -- Annoy: ORDER-BY-type -6 (1,9.3) 0.005731362878640178 -1 (2,3.2) 0.15200169244542905 -7 (5.5,4.7) 0.3503476876550442 -- Usearch: WHERE-type -0 (4.6,2.3) 0.5527864045000421 -1 (2,3.2) 0.15200169244542905 -2 (4.2,3.4) 0.37080174340866845 -- Usearch: ORDER-BY-type -6 (1,9.3) 0.005731362878640178 -1 (2,3.2) 0.15200169244542905 -7 (5.5,4.7) 0.3503476876550442 -- Special case: MaximumDistance is negative -- Special case: MaximumDistance is negative -- Special case: setting "annoy_index_search_k_nodes" +Special cases +- ORDER-BY-type +6 [1,9.3] 0.005731362878640178 +1 [2,3.2] 0.15200169244542905 +7 [5.5,4.7] 0.3503476876550442 - Special case: setting "max_limit_for_ann_queries" Expression (Projection) Limit (preliminary LIMIT (without OFFSET)) Sorting (Sorting for ORDER BY) Expression (Before ORDER BY) - ReadFromMergeTree (default.tab_annoy) - Indexes: - PrimaryKey - Condition: true - Parts: 1/1 - Granules: 4/4 -- Special case: setting "max_limit_for_ann_queries" -Expression (Projection) - Limit (preliminary LIMIT (without OFFSET)) - Sorting (Sorting for ORDER BY) - Expression (Before ORDER BY) - ReadFromMergeTree (default.tab_usearch) + ReadFromMergeTree (default.tab) Indexes: PrimaryKey Condition: true diff --git a/tests/queries/0_stateless/02354_vector_search_queries.sql b/tests/queries/0_stateless/02354_vector_search_queries.sql index 64051aa8544..dbf0fca32ab 100644 --- a/tests/queries/0_stateless/02354_vector_search_queries.sql +++ b/tests/queries/0_stateless/02354_vector_search_queries.sql @@ -1,238 +1,80 @@ --- Tags: no-fasttest, no-ubsan, no-cpu-aarch64, no-ordinary-database, no-asan +-- Tags: no-fasttest, no-ordinary-database --- Tests vector search in ClickHouse, i.e. Annoy and Usearch indexes. Both index types share similarities in implementation and usage, --- therefore they are tested in a single file. +-- Tests various simple approximate nearest neighborhood (ANN) queries that utilize vector search indexes. --- This file tests various simple approximate nearest neighborhood (ANN) queries that utilize vector search indexes. +SET allow_experimental_vector_similarity_index = 1; -SET allow_experimental_annoy_index = 1; -SET allow_experimental_usearch_index = 1; +SET enable_analyzer = 0; -SELECT 'ARRAY, 10 rows, index_granularity = 8192, GRANULARITY = 1 million --> 1 granule, 1 indexed block'; +SELECT '10 rows, index_granularity = 8192, GRANULARITY = 1 million --> 1 granule, 1 indexed block'; -DROP TABLE IF EXISTS tab_annoy; -DROP TABLE IF EXISTS tab_usearch; +DROP TABLE IF EXISTS tab; -CREATE TABLE tab_annoy(id Int32, vec Array(Float32), INDEX idx vec TYPE annoy()) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity = 8192; -INSERT INTO tab_annoy VALUES (0, [1.0, 0.0]), (1, [1.1, 0.0]), (2, [1.2, 0.0]), (3, [1.3, 0.0]), (4, [1.4, 0.0]), (5, [0.0, 2.0]), (6, [0.0, 2.1]), (7, [0.0, 2.2]), (8, [0.0, 2.3]), (9, [0.0, 2.4]); - -CREATE TABLE tab_usearch(id Int32, vec Array(Float32), INDEX idx vec TYPE usearch()) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity = 8192; -INSERT INTO tab_usearch VALUES (0, [1.0, 0.0]), (1, [1.1, 0.0]), (2, [1.2, 0.0]), (3, [1.3, 0.0]), (4, [1.4, 0.0]), (5, [0.0, 2.0]), (6, [0.0, 2.1]), (7, [0.0, 2.2]), (8, [0.0, 2.3]), (9, [0.0, 2.4]); +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity = 8192; +INSERT INTO tab VALUES (0, [1.0, 0.0]), (1, [1.1, 0.0]), (2, [1.2, 0.0]), (3, [1.3, 0.0]), (4, [1.4, 0.0]), (5, [0.0, 2.0]), (6, [0.0, 2.1]), (7, [0.0, 2.2]), (8, [0.0, 2.3]), (9, [0.0, 2.4]); -SELECT '- Annoy: WHERE-type'; +SELECT '- ORDER-BY-type'; WITH [0.0, 2.0] AS reference_vec SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_annoy -WHERE L2Distance(vec, reference_vec) < 1.0 -LIMIT 3; - -SELECT '- Annoy: ORDER-BY-type'; -WITH [0.0, 2.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_annoy +FROM tab ORDER BY L2Distance(vec, reference_vec) LIMIT 3; -SELECT '- Usearch: WHERE-type'; +SELECT '- ORDER-BY-type, EXPLAIN'; +EXPLAIN indexes = 1 WITH [0.0, 2.0] AS reference_vec SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_usearch -WHERE L2Distance(vec, reference_vec) < 1.0 -LIMIT 3; - -SELECT '- Usearch: ORDER-BY-type'; -WITH [0.0, 2.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_usearch +FROM tab ORDER BY L2Distance(vec, reference_vec) LIMIT 3; -SELECT '- Annoy: WHERE-type, EXPLAIN'; -EXPLAIN indexes=1 -WITH [0.0, 2.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_annoy -WHERE L2Distance(vec, reference_vec) < 1.0 -LIMIT 3; +DROP TABLE tab; -SELECT '- Annoy: ORDER-BY-type, EXPLAIN'; -EXPLAIN indexes=1 + +SELECT '12 rows, index_granularity = 3, GRANULARITY = 2 --> 4 granules, 2 indexed block'; + +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance') GRANULARITY 2) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity = 3; +INSERT INTO tab VALUES (0, [1.0, 0.0]), (1, [1.1, 0.0]), (2, [1.2, 0.0]), (3, [1.3, 0.0]), (4, [1.4, 0.0]), (5, [1.5, 0.0]), (6, [0.0, 2.0]), (7, [0.0, 2.1]), (8, [0.0, 2.2]), (9, [0.0, 2.3]), (10, [0.0, 2.4]), (11, [0.0, 2.5]); + +SELECT '- ORDER-BY-type'; WITH [0.0, 2.0] AS reference_vec SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_annoy +FROM tab ORDER BY L2Distance(vec, reference_vec) LIMIT 3; -SELECT '- Usearch: WHERE-type, EXPLAIN'; -EXPLAIN indexes=1 +SELECT '- ORDER-BY-type, EXPLAIN'; +EXPLAIN indexes = 1 WITH [0.0, 2.0] AS reference_vec SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_usearch -WHERE L2Distance(vec, reference_vec) < 1.0 -LIMIT 3; - -SELECT '- Usearch: ORDER-BY-type, EXPLAIN'; -EXPLAIN indexes=1 -WITH [0.0, 2.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_usearch +FROM tab ORDER BY L2Distance(vec, reference_vec) LIMIT 3; -DROP TABLE tab_annoy; -DROP TABLE tab_usearch; +DROP TABLE tab; -SELECT 'ARRAY vectors, 12 rows, index_granularity = 3, GRANULARITY = 2 --> 4 granules, 2 indexed block'; +SELECT 'Special cases'; -- Not a systematic test, just to check that no bad things happen. +-- Test with non-default metric, M, ef_construction, ef_search -CREATE TABLE tab_annoy(id Int32, vec Array(Float32), INDEX idx vec TYPE annoy() GRANULARITY 2) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity = 3; -INSERT INTO tab_annoy VALUES (0, [1.0, 0.0]), (1, [1.1, 0.0]), (2, [1.2, 0.0]), (3, [1.3, 0.0]), (4, [1.4, 0.0]), (5, [1.5, 0.0]), (6, [0.0, 2.0]), (7, [0.0, 2.1]), (8, [0.0, 2.2]), (9, [0.0, 2.3]), (10, [0.0, 2.4]), (11, [0.0, 2.5]); +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'cosineDistance', 'f32', 42, 99, 66) GRANULARITY 2) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity = 3; +INSERT INTO tab VALUES (0, [4.6, 2.3]), (1, [2.0, 3.2]), (2, [4.2, 3.4]), (3, [5.3, 2.9]), (4, [2.4, 5.2]), (5, [5.3, 2.3]), (6, [1.0, 9.3]), (7, [5.5, 4.7]), (8, [6.4, 3.5]), (9, [5.3, 2.5]), (10, [6.4, 3.4]), (11, [6.4, 3.2]); -CREATE TABLE tab_usearch(id Int32, vec Array(Float32), INDEX idx vec TYPE usearch() GRANULARITY 2) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity = 3; -INSERT INTO tab_usearch VALUES (0, [1.0, 0.0]), (1, [1.1, 0.0]), (2, [1.2, 0.0]), (3, [1.3, 0.0]), (4, [1.4, 0.0]), (5, [1.5, 0.0]), (6, [0.0, 2.0]), (7, [0.0, 2.1]), (8, [0.0, 2.2]), (9, [0.0, 2.3]), (10, [0.0, 2.4]), (11, [0.0, 2.5]); - -SELECT '- Annoy: WHERE-type'; +SELECT '- ORDER-BY-type'; WITH [0.0, 2.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_annoy -WHERE L2Distance(vec, reference_vec) < 1.0 -LIMIT 3; - -SELECT '- Annoy: ORDER-BY-type'; -WITH [0.0, 2.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_annoy -ORDER BY L2Distance(vec, reference_vec) -LIMIT 3; - -SELECT '- Usearch: WHERE-type'; -WITH [0.0, 2.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_usearch -WHERE L2Distance(vec, reference_vec) < 1.0 -LIMIT 3; - -SELECT '- Usearch: ORDER-BY-type'; -WITH [0.0, 2.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_usearch -ORDER BY L2Distance(vec, reference_vec) -LIMIT 3; - -SELECT '- Annoy: WHERE-type, EXPLAIN'; -EXPLAIN indexes=1 -WITH [0.0, 2.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_annoy -WHERE L2Distance(vec, reference_vec) < 1.0 -LIMIT 3; - -SELECT '- Annoy: ORDER-BY-type, EXPLAIN'; -EXPLAIN indexes=1 -WITH [0.0, 2.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_annoy -ORDER BY L2Distance(vec, reference_vec) -LIMIT 3; - -SELECT '- Usearch: WHERE-type, EXPLAIN'; -EXPLAIN indexes=1 -WITH [0.0, 2.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_usearch -WHERE L2Distance(vec, reference_vec) < 1.0 -LIMIT 3; - -SELECT '- Usearch: ORDER-BY-type, EXPLAIN'; -EXPLAIN indexes=1 -WITH [0.0, 2.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab_usearch -ORDER BY L2Distance(vec, reference_vec) -LIMIT 3; - -DROP TABLE tab_annoy; -DROP TABLE tab_usearch; - - -SELECT 'TUPLE vectors and special cases'; --- Not a systematic test, just to check that no bad things happen. --- Just for jun, use metric = 'cosineDistance' (Annoy/Usearch), tree_count = 200 (Annoy), scalarKind = 'f64' (Usearch) - -CREATE TABLE tab_annoy(id Int32, vec Tuple(Float32, Float32), INDEX idx vec TYPE annoy('cosineDistance', 200) GRANULARITY 2) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity = 3; -INSERT INTO tab_annoy VALUES (0, (4.6, 2.3)), (1, (2.0, 3.2)), (2, (4.2, 3.4)), (3, (5.3, 2.9)), (4, (2.4, 5.2)), (5, (5.3, 2.3)), (6, (1.0, 9.3)), (7, (5.5, 4.7)), (8, (6.4, 3.5)), (9, (5.3, 2.5)), (10, (6.4, 3.4)), (11, (6.4, 3.2)); - -CREATE TABLE tab_usearch(id Int32, vec Tuple(Float32, Float32), INDEX idx vec TYPE usearch('cosineDistance', 'f64') GRANULARITY 2) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity = 3; -INSERT INTO tab_usearch VALUES (0, (4.6, 2.3)), (1, (2.0, 3.2)), (2, (4.2, 3.4)), (3, (5.3, 2.9)), (4, (2.4, 5.2)), (5, (5.3, 2.3)), (6, (1.0, 9.3)), (7, (5.5, 4.7)), (8, (6.4, 3.5)), (9, (5.3, 2.5)), (10, (6.4, 3.4)), (11, (6.4, 3.2)); - -SELECT '- Annoy: WHERE-type'; -WITH (0.0, 2.0) AS reference_vec SELECT id, vec, cosineDistance(vec, reference_vec) -FROM tab_annoy -WHERE cosineDistance(vec, reference_vec) < 1.0 -LIMIT 3; - -SELECT '- Annoy: ORDER-BY-type'; -WITH (0.0, 2.0) AS reference_vec -SELECT id, vec, cosineDistance(vec, reference_vec) -FROM tab_annoy +FROM tab ORDER BY cosineDistance(vec, reference_vec) LIMIT 3; -SELECT '- Usearch: WHERE-type'; -WITH (0.0, 2.0) AS reference_vec -SELECT id, vec, cosineDistance(vec, reference_vec) -FROM tab_usearch -WHERE cosineDistance(vec, reference_vec) < 1.0 -LIMIT 3; - -SELECT '- Usearch: ORDER-BY-type'; -WITH (0.0, 2.0) AS reference_vec -SELECT id, vec, cosineDistance(vec, reference_vec) -FROM tab_usearch -ORDER BY cosineDistance(vec, reference_vec) -LIMIT 3; - -SELECT '- Special case: MaximumDistance is negative'; -WITH (0.0, 2.0) as reference_vec -SELECT id, vec, cosineDistance(vec, reference_vec) -FROM tab_annoy -WHERE cosineDistance(vec, reference_vec) < -1.0 -LIMIT 3; -- { serverError INCORRECT_QUERY } - -SELECT '- Special case: MaximumDistance is negative'; -WITH (0.0, 2.0) as reference_vec -SELECT id, vec, cosineDistance(vec, reference_vec) -FROM tab_usearch -WHERE cosineDistance(vec, reference_vec) < -1.0 -LIMIT 3; -- { serverError INCORRECT_QUERY } - -SELECT '- Special case: setting "annoy_index_search_k_nodes"'; -WITH (0.0, 2.0) as reference_vec -SELECT id, vec, cosineDistance(vec, reference_vec) -FROM tab_annoy -ORDER BY cosineDistance(vec, reference_vec) -LIMIT 3 -SETTINGS annoy_index_search_k_nodes=0; -- searches zero nodes --> no results - SELECT '- Special case: setting "max_limit_for_ann_queries"'; EXPLAIN indexes=1 -WITH (0.0, 2.0) as reference_vec +WITH [0.0, 2.0] as reference_vec SELECT id, vec, cosineDistance(vec, reference_vec) -FROM tab_annoy +FROM tab ORDER BY cosineDistance(vec, reference_vec) LIMIT 3 -SETTINGS max_limit_for_ann_queries=2; -- LIMIT 3 > 2 --> don't use the ann index +SETTINGS max_limit_for_ann_queries = 2; -- LIMIT 3 > 2 --> don't use the ann index -SELECT '- Special case: setting "max_limit_for_ann_queries"'; -EXPLAIN indexes=1 -WITH (0.0, 2.0) as reference_vec -SELECT id, vec, cosineDistance(vec, reference_vec) -FROM tab_usearch -ORDER BY cosineDistance(vec, reference_vec) -LIMIT 3 -SETTINGS max_limit_for_ann_queries=2; -- LIMIT 3 > 2 --> don't use the ann index - -DROP TABLE tab_annoy; -DROP TABLE tab_usearch; +DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_unquoted_index_parameters.reference b/tests/queries/0_stateless/02354_vector_search_unquoted_index_parameters.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/02354_vector_search_unquoted_index_parameters.sql b/tests/queries/0_stateless/02354_vector_search_unquoted_index_parameters.sql new file mode 100644 index 00000000000..da6494bf831 --- /dev/null +++ b/tests/queries/0_stateless/02354_vector_search_unquoted_index_parameters.sql @@ -0,0 +1,23 @@ +-- Tags: no-fasttest, no-ordinary-database + +SET allow_experimental_vector_similarity_index = 1; + +-- Tests that quoted and unquoted parameters can be passed to vector search indexes. + +DROP TABLE IF EXISTS tab1; +DROP TABLE IF EXISTS tab2; + +CREATE TABLE tab1 (id Int32, vec Array(Float32), PRIMARY KEY id, INDEX vec_idx(vec) TYPE vector_similarity('hnsw', 'L2Distance')); +CREATE TABLE tab2 (id Int32, vec Array(Float32), PRIMARY KEY id, INDEX vec_idx(vec) TYPE vector_similarity(hnsw, L2Distance)); + +DROP TABLE tab1; +DROP TABLE tab2; + +CREATE TABLE tab1 (id Int32, vec Array(Float32), PRIMARY KEY id); +CREATE TABLE tab2 (id Int32, vec Array(Float32), PRIMARY KEY id); + +ALTER TABLE tab1 ADD INDEX idx1(vec) TYPE vector_similarity('hnsw', 'L2Distance'); +ALTER TABLE tab2 ADD INDEX idx2(vec) TYPE vector_similarity(hnsw, L2Distance); + +DROP TABLE tab1; +DROP TABLE tab2; diff --git a/tests/queries/0_stateless/02355_column_type_name_lc.reference b/tests/queries/0_stateless/02355_column_type_name_lc.reference index 234a072299f..50c25a86b2f 100644 --- a/tests/queries/0_stateless/02355_column_type_name_lc.reference +++ b/tests/queries/0_stateless/02355_column_type_name_lc.reference @@ -1 +1 @@ -ColumnLowCardinality +LowCardinality(String) diff --git a/tests/queries/0_stateless/02361_fsync_profile_events.sh b/tests/queries/0_stateless/02361_fsync_profile_events.sh index 98c9cf9b7b4..73bf3fa120a 100755 --- a/tests/queries/0_stateless/02361_fsync_profile_events.sh +++ b/tests/queries/0_stateless/02361_fsync_profile_events.sh @@ -6,7 +6,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " drop table if exists data_fsync_pe; create table data_fsync_pe (key Int) engine=MergeTree() @@ -27,7 +27,7 @@ for i in {1..100}; do $CLICKHOUSE_CLIENT --query_id "$query_id" -q "insert into data_fsync_pe values (1)" read -r FileSync FileOpen DirectorySync FileSyncElapsedMicroseconds DirectorySyncElapsedMicroseconds <<<"$( - $CLICKHOUSE_CLIENT -nm --param_query_id "$query_id" -q " + $CLICKHOUSE_CLIENT -m --param_query_id "$query_id" -q " system flush logs; select diff --git a/tests/queries/0_stateless/02364_window_view_segfault.sh b/tests/queries/0_stateless/02364_window_view_segfault.sh index 4173b6c6b8d..833d8967558 100755 --- a/tests/queries/0_stateless/02364_window_view_segfault.sh +++ b/tests/queries/0_stateless/02364_window_view_segfault.sh @@ -6,7 +6,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh opts=( - "--allow_experimental_analyzer=0" + "--enable_analyzer=0" ) ${CLICKHOUSE_CLIENT} "${opts[@]}" --multiline --query """ diff --git a/tests/queries/0_stateless/02366_explain_query_tree.sql b/tests/queries/0_stateless/02366_explain_query_tree.sql index c38b2d819d1..82621ec90f7 100644 --- a/tests/queries/0_stateless/02366_explain_query_tree.sql +++ b/tests/queries/0_stateless/02366_explain_query_tree.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; EXPLAIN QUERY TREE run_passes = 0 SELECT 1; diff --git a/tests/queries/0_stateless/02367_analyzer_table_alias_columns.sql b/tests/queries/0_stateless/02367_analyzer_table_alias_columns.sql index f41680cd9f4..a6408abca9b 100644 --- a/tests/queries/0_stateless/02367_analyzer_table_alias_columns.sql +++ b/tests/queries/0_stateless/02367_analyzer_table_alias_columns.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02368_analyzer_table_functions.sql b/tests/queries/0_stateless/02368_analyzer_table_functions.sql index 456e095c6c1..cc65848a5d8 100644 --- a/tests/queries/0_stateless/02368_analyzer_table_functions.sql +++ b/tests/queries/0_stateless/02368_analyzer_table_functions.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT c1, c2, c3, c4 FROM format('CSV', '1,2,"[1,2,3]","[[\'abc\'], [], [\'d\', \'e\']]"'); SELECT f.c1, f.c2, f.c3, f.c4 FROM format('CSV', '1,2,"[1,2,3]","[[\'abc\'], [], [\'d\', \'e\']]"') AS f; diff --git a/tests/queries/0_stateless/02369_analyzer_array_join_function.sql b/tests/queries/0_stateless/02369_analyzer_array_join_function.sql index e60ec7e71a9..f84c96ee6a6 100644 --- a/tests/queries/0_stateless/02369_analyzer_array_join_function.sql +++ b/tests/queries/0_stateless/02369_analyzer_array_join_function.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT arrayJoin([1, 2, 3]); diff --git a/tests/queries/0_stateless/02370_analyzer_in_function.sql b/tests/queries/0_stateless/02370_analyzer_in_function.sql index a6e4400e101..9fb05ada829 100644 --- a/tests/queries/0_stateless/02370_analyzer_in_function.sql +++ b/tests/queries/0_stateless/02370_analyzer_in_function.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT 1 IN 1; SELECT 1 IN (1); diff --git a/tests/queries/0_stateless/02371_analyzer_join_cross.sql b/tests/queries/0_stateless/02371_analyzer_join_cross.sql index 3624a1d2282..660dc19770a 100644 --- a/tests/queries/0_stateless/02371_analyzer_join_cross.sql +++ b/tests/queries/0_stateless/02371_analyzer_join_cross.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET single_join_prefer_left_table = 0; DROP TABLE IF EXISTS test_table_join_1; diff --git a/tests/queries/0_stateless/02372_analyzer_join.sql.j2 b/tests/queries/0_stateless/02372_analyzer_join.sql.j2 index 45ae63b9a49..fb75ddeac09 100644 --- a/tests/queries/0_stateless/02372_analyzer_join.sql.j2 +++ b/tests/queries/0_stateless/02372_analyzer_join.sql.j2 @@ -1,6 +1,6 @@ -- Tags: long -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET single_join_prefer_left_table = 0; DROP TABLE IF EXISTS test_table_join_1; 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 bcec6d178a8..16a37ee7f50 100644 --- a/tests/queries/0_stateless/02373_analyzer_join_use_nulls.sql +++ b/tests/queries/0_stateless/02373_analyzer_join_use_nulls.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET join_use_nulls = 1; DROP TABLE IF EXISTS test_table_join_1; diff --git a/tests/queries/0_stateless/02374_analyzer_array_join.sql b/tests/queries/0_stateless/02374_analyzer_array_join.sql index 8c26df1806e..fe1508f5f78 100644 --- a/tests/queries/0_stateless/02374_analyzer_array_join.sql +++ b/tests/queries/0_stateless/02374_analyzer_array_join.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02374_analyzer_join_using.sql.j2 b/tests/queries/0_stateless/02374_analyzer_join_using.sql.j2 index 26fb52716ff..2c64efc7fc8 100644 --- a/tests/queries/0_stateless/02374_analyzer_join_using.sql.j2 +++ b/tests/queries/0_stateless/02374_analyzer_join_using.sql.j2 @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table_join_1; CREATE TABLE test_table_join_1 diff --git a/tests/queries/0_stateless/02374_combine_multi_if_and_count_if_opt.sql b/tests/queries/0_stateless/02374_combine_multi_if_and_count_if_opt.sql index 4371c2e5641..05472e5e35a 100644 --- a/tests/queries/0_stateless/02374_combine_multi_if_and_count_if_opt.sql +++ b/tests/queries/0_stateless/02374_combine_multi_if_and_count_if_opt.sql @@ -4,7 +4,7 @@ create table m (a int) engine Log; insert into m values (1); -set allow_experimental_analyzer = true, optimize_rewrite_sum_if_to_count_if=1; +set enable_analyzer = true, optimize_rewrite_sum_if_to_count_if=1; EXPLAIN QUERY TREE select sum(multiIf(a = 1, 1, 0)) from m; diff --git a/tests/queries/0_stateless/02375_analyzer_union.sql b/tests/queries/0_stateless/02375_analyzer_union.sql index 5e41f07d217..cf9a99ffbba 100644 --- a/tests/queries/0_stateless/02375_analyzer_union.sql +++ b/tests/queries/0_stateless/02375_analyzer_union.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02376_analyzer_in_function_subquery.sql b/tests/queries/0_stateless/02376_analyzer_in_function_subquery.sql index 72a4edb8567..295d8a8f97e 100644 --- a/tests/queries/0_stateless/02376_analyzer_in_function_subquery.sql +++ b/tests/queries/0_stateless/02376_analyzer_in_function_subquery.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02377_analyzer_in_function_set.sql b/tests/queries/0_stateless/02377_analyzer_in_function_set.sql index e3cbcf75a9c..00aa40ccf17 100644 --- a/tests/queries/0_stateless/02377_analyzer_in_function_set.sql +++ b/tests/queries/0_stateless/02377_analyzer_in_function_set.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02377_extend_protocol_with_query_parameters.sh b/tests/queries/0_stateless/02377_extend_protocol_with_query_parameters.sh index 71e3b6961f8..46396d38747 100755 --- a/tests/queries/0_stateless/02377_extend_protocol_with_query_parameters.sh +++ b/tests/queries/0_stateless/02377_extend_protocol_with_query_parameters.sh @@ -24,7 +24,7 @@ $CLICKHOUSE_CLIENT \ table_name="t_02377_extend_protocol_with_query_parameters_$RANDOM$RANDOM" -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " create table $table_name( id Int64, arr Array(UInt8), @@ -57,17 +57,17 @@ $CLICKHOUSE_CLIENT \ # it is possible to set parameter for the current session -$CLICKHOUSE_CLIENT -n -q "set param_n = 42; select {n: UInt8}" +$CLICKHOUSE_CLIENT -q "set param_n = 42; select {n: UInt8}" # and it will not be visible to other sessions -$CLICKHOUSE_CLIENT -n -q "select {n: UInt8} -- { serverError 456 }" +$CLICKHOUSE_CLIENT -q "select {n: UInt8} -- { serverError 456 }" # the same parameter could be set multiple times within one session (new value overrides the previous one) -$CLICKHOUSE_CLIENT -n -q "set param_n = 12; set param_n = 13; select {n: UInt8}" +$CLICKHOUSE_CLIENT -q "set param_n = 12; set param_n = 13; select {n: UInt8}" # multiple different parameters could be defined within each session -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " set param_a = 13, param_b = 'str'; set param_c = '2022-08-04 18:30:53'; set param_d = '{\'10\': [11, 12], \'13\': [14, 15]}'; diff --git a/tests/queries/0_stateless/02377_optimize_sorting_by_input_stream_properties_explain.sh b/tests/queries/0_stateless/02377_optimize_sorting_by_input_stream_properties_explain.sh index c223fcc86bc..974f10e2f24 100755 --- a/tests/queries/0_stateless/02377_optimize_sorting_by_input_stream_properties_explain.sh +++ b/tests/queries/0_stateless/02377_optimize_sorting_by_input_stream_properties_explain.sh @@ -15,14 +15,14 @@ FIND_SORTMODE="$GREP_SORTMODE | $TRIM_LEADING_SPACES" function explain_sorting { echo "-- QUERY: "$1 - $CLICKHOUSE_CLIENT --merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability=0.0 -nq "$1" | eval $FIND_SORTING + $CLICKHOUSE_CLIENT --merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability=0.0 -q "$1" | eval $FIND_SORTING } function explain_sortmode { echo "-- QUERY: "$1 - $CLICKHOUSE_CLIENT --allow_experimental_analyzer=0 --merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability=0.0 -nq "$1" | eval $FIND_SORTMODE + $CLICKHOUSE_CLIENT --enable_analyzer=0 --merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability=0.0 -nq "$1" | eval $FIND_SORTMODE echo "-- QUERY (analyzer): "$1 - $CLICKHOUSE_CLIENT --allow_experimental_analyzer=1 --merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability=0.0 -nq "$1" | eval $FIND_SORTMODE + $CLICKHOUSE_CLIENT --enable_analyzer=1 --merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability=0.0 -nq "$1" | eval $FIND_SORTMODE } $CLICKHOUSE_CLIENT -q "drop table if exists optimize_sorting sync" diff --git a/tests/queries/0_stateless/02378_analyzer_projection_names.sql b/tests/queries/0_stateless/02378_analyzer_projection_names.sql index f41afe6a950..7b3099c4f36 100644 --- a/tests/queries/0_stateless/02378_analyzer_projection_names.sql +++ b/tests/queries/0_stateless/02378_analyzer_projection_names.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET single_join_prefer_left_table = 0; DROP TABLE IF EXISTS test_table; diff --git a/tests/queries/0_stateless/02378_part_log_profile_events.sql b/tests/queries/0_stateless/02378_part_log_profile_events.sql index 38d3f8b4c05..eec76d6f50e 100644 --- a/tests/queries/0_stateless/02378_part_log_profile_events.sql +++ b/tests/queries/0_stateless/02378_part_log_profile_events.sql @@ -39,7 +39,7 @@ SYSTEM FLUSH LOGS; SELECT if(count() == 2, 'Ok', 'Error: ' || toString(count())), - if(SUM(ProfileEvents['MergedRows']) == 512, 'Ok', 'Error: ' || toString(SUM(ProfileEvents['MergedRows']))), + if(SUM(ProfileEvents['MutatedRows']) == 512, 'Ok', 'Error: ' || toString(SUM(ProfileEvents['MutatedRows']))), if(SUM(ProfileEvents['FileOpen']) > 1, 'Ok', 'Error: ' || toString(SUM(ProfileEvents['FileOpen']))) FROM system.part_log WHERE event_time > now() - INTERVAL 10 MINUTE diff --git a/tests/queries/0_stateless/02379_analyzer_subquery_depth.sql b/tests/queries/0_stateless/02379_analyzer_subquery_depth.sql index 5699a15aead..40303e0f92a 100644 --- a/tests/queries/0_stateless/02379_analyzer_subquery_depth.sql +++ b/tests/queries/0_stateless/02379_analyzer_subquery_depth.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT (SELECT a FROM (SELECT 1 AS a)) SETTINGS max_subquery_depth = 1; -- { serverError TOO_DEEP_SUBQUERIES } SELECT (SELECT a FROM (SELECT 1 AS a)) SETTINGS max_subquery_depth = 2; diff --git a/tests/queries/0_stateless/02380_analyzer_join_sample.sql b/tests/queries/0_stateless/02380_analyzer_join_sample.sql index e417f47d173..bc77f3623ae 100644 --- a/tests/queries/0_stateless/02380_analyzer_join_sample.sql +++ b/tests/queries/0_stateless/02380_analyzer_join_sample.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table_join_1; CREATE TABLE test_table_join_1 diff --git a/tests/queries/0_stateless/02381_analyzer_join_final.sql b/tests/queries/0_stateless/02381_analyzer_join_final.sql index 57fc3aedd8f..0db81ac7728 100644 --- a/tests/queries/0_stateless/02381_analyzer_join_final.sql +++ b/tests/queries/0_stateless/02381_analyzer_join_final.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table_join_1; CREATE TABLE test_table_join_1 diff --git a/tests/queries/0_stateless/02381_join_dup_columns_in_plan.sql.j2 b/tests/queries/0_stateless/02381_join_dup_columns_in_plan.sql.j2 index ca4af4df6b6..4e88b67e3dd 100644 --- a/tests/queries/0_stateless/02381_join_dup_columns_in_plan.sql.j2 +++ b/tests/queries/0_stateless/02381_join_dup_columns_in_plan.sql.j2 @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- { echoOn } diff --git a/tests/queries/0_stateless/02382_analyzer_matcher_join_using.sql b/tests/queries/0_stateless/02382_analyzer_matcher_join_using.sql index 7983b05a69e..6a0b58e7b28 100644 --- a/tests/queries/0_stateless/02382_analyzer_matcher_join_using.sql +++ b/tests/queries/0_stateless/02382_analyzer_matcher_join_using.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table_join_1; CREATE TABLE test_table_join_1 diff --git a/tests/queries/0_stateless/02383_analyzer_merge_tree_self_join.sql b/tests/queries/0_stateless/02383_analyzer_merge_tree_self_join.sql index c22a0f4244b..fbd6fe4db4f 100644 --- a/tests/queries/0_stateless/02383_analyzer_merge_tree_self_join.sql +++ b/tests/queries/0_stateless/02383_analyzer_merge_tree_self_join.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table_join_1; CREATE TABLE test_table_join_1 diff --git a/tests/queries/0_stateless/02384_analyzer_dict_get_join_get.sql b/tests/queries/0_stateless/02384_analyzer_dict_get_join_get.sql index f4619f20765..a8ad5c4d957 100644 --- a/tests/queries/0_stateless/02384_analyzer_dict_get_join_get.sql +++ b/tests/queries/0_stateless/02384_analyzer_dict_get_join_get.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02385_analyzer_aliases_compound_expression.sql b/tests/queries/0_stateless/02385_analyzer_aliases_compound_expression.sql index 861ada9623a..025e064d23c 100644 --- a/tests/queries/0_stateless/02385_analyzer_aliases_compound_expression.sql +++ b/tests/queries/0_stateless/02385_analyzer_aliases_compound_expression.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT cast(tuple(1, 'Value'), 'Tuple(first UInt64, second String)') AS value, value.first, value.second; diff --git a/tests/queries/0_stateless/02386_analyzer_in_function_nested_subqueries.sql b/tests/queries/0_stateless/02386_analyzer_in_function_nested_subqueries.sql index c8ca3ff21d4..64d24aae1b1 100644 --- a/tests/queries/0_stateless/02386_analyzer_in_function_nested_subqueries.sql +++ b/tests/queries/0_stateless/02386_analyzer_in_function_nested_subqueries.sql @@ -1,3 +1,3 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT (NULL IN (SELECT 9223372036854775806 IN (SELECT 65536), inf, NULL IN (NULL))) IN (SELECT NULL IN (NULL)); diff --git a/tests/queries/0_stateless/02387_analyzer_cte.sql b/tests/queries/0_stateless/02387_analyzer_cte.sql index 1f10ac10438..149eab7f741 100644 --- a/tests/queries/0_stateless/02387_analyzer_cte.sql +++ b/tests/queries/0_stateless/02387_analyzer_cte.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02388_analyzer_recursive_lambda.sql b/tests/queries/0_stateless/02388_analyzer_recursive_lambda.sql index 9fd2f73703d..31d6f91a39d 100644 --- a/tests/queries/0_stateless/02388_analyzer_recursive_lambda.sql +++ b/tests/queries/0_stateless/02388_analyzer_recursive_lambda.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; WITH x -> plus(lambda(1), x) AS lambda SELECT lambda(1048576); -- { serverError UNSUPPORTED_METHOD }; diff --git a/tests/queries/0_stateless/02389_analyzer_nested_lambda.sql b/tests/queries/0_stateless/02389_analyzer_nested_lambda.sql index 8e3777ebc15..51a10adae6a 100644 --- a/tests/queries/0_stateless/02389_analyzer_nested_lambda.sql +++ b/tests/queries/0_stateless/02389_analyzer_nested_lambda.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- { echoOn } diff --git a/tests/queries/0_stateless/02404_data.CSV b/tests/queries/0_stateless/02404_data.CSV new file mode 100644 index 00000000000..2d8b5c8daa8 --- /dev/null +++ b/tests/queries/0_stateless/02404_data.CSV @@ -0,0 +1,10 @@ +0,"1970-01-01" +1,"1970-01-02" +2,"1970-01-03" +3,"1970-01-04" +4,"1970-01-05" +5,"1970-01-06" +6,"1970-01-07" +7,"1970-01-08" +8,"1970-01-09" +9,"1970-01-10" diff --git a/tests/queries/0_stateless/02404_data.CSVWithNames b/tests/queries/0_stateless/02404_data.CSVWithNames new file mode 100644 index 00000000000..34647008916 --- /dev/null +++ b/tests/queries/0_stateless/02404_data.CSVWithNames @@ -0,0 +1,11 @@ +"number","toDate(number)" +0,"1970-01-01" +1,"1970-01-02" +2,"1970-01-03" +3,"1970-01-04" +4,"1970-01-05" +5,"1970-01-06" +6,"1970-01-07" +7,"1970-01-08" +8,"1970-01-09" +9,"1970-01-10" diff --git a/tests/queries/0_stateless/02404_data.CustomSeparated b/tests/queries/0_stateless/02404_data.CustomSeparated new file mode 100644 index 00000000000..f3ae1663536 --- /dev/null +++ b/tests/queries/0_stateless/02404_data.CustomSeparated @@ -0,0 +1,10 @@ +0 1970-01-01 +1 1970-01-02 +2 1970-01-03 +3 1970-01-04 +4 1970-01-05 +5 1970-01-06 +6 1970-01-07 +7 1970-01-08 +8 1970-01-09 +9 1970-01-10 diff --git a/tests/queries/0_stateless/02404_data.JSONCompactEachRow b/tests/queries/0_stateless/02404_data.JSONCompactEachRow new file mode 100644 index 00000000000..de2e0986aab --- /dev/null +++ b/tests/queries/0_stateless/02404_data.JSONCompactEachRow @@ -0,0 +1,10 @@ +["0", "1970-01-01"] +["1", "1970-01-02"] +["2", "1970-01-03"] +["3", "1970-01-04"] +["4", "1970-01-05"] +["5", "1970-01-06"] +["6", "1970-01-07"] +["7", "1970-01-08"] +["8", "1970-01-09"] +["9", "1970-01-10"] diff --git a/tests/queries/0_stateless/02404_data.JSONEachRow b/tests/queries/0_stateless/02404_data.JSONEachRow new file mode 100644 index 00000000000..e77256ac7fc --- /dev/null +++ b/tests/queries/0_stateless/02404_data.JSONEachRow @@ -0,0 +1,10 @@ +{"number":"0","toDate(number)":"1970-01-01"} +{"number":"1","toDate(number)":"1970-01-02"} +{"number":"2","toDate(number)":"1970-01-03"} +{"number":"3","toDate(number)":"1970-01-04"} +{"number":"4","toDate(number)":"1970-01-05"} +{"number":"5","toDate(number)":"1970-01-06"} +{"number":"6","toDate(number)":"1970-01-07"} +{"number":"7","toDate(number)":"1970-01-08"} +{"number":"8","toDate(number)":"1970-01-09"} +{"number":"9","toDate(number)":"1970-01-10"} diff --git a/tests/queries/0_stateless/02404_data.TSKV b/tests/queries/0_stateless/02404_data.TSKV new file mode 100644 index 00000000000..70f7ad33c8b --- /dev/null +++ b/tests/queries/0_stateless/02404_data.TSKV @@ -0,0 +1,10 @@ +number=0 toDate(number)=1970-01-01 +number=1 toDate(number)=1970-01-02 +number=2 toDate(number)=1970-01-03 +number=3 toDate(number)=1970-01-04 +number=4 toDate(number)=1970-01-05 +number=5 toDate(number)=1970-01-06 +number=6 toDate(number)=1970-01-07 +number=7 toDate(number)=1970-01-08 +number=8 toDate(number)=1970-01-09 +number=9 toDate(number)=1970-01-10 diff --git a/tests/queries/0_stateless/02404_data.TSV b/tests/queries/0_stateless/02404_data.TSV new file mode 100644 index 00000000000..f3ae1663536 --- /dev/null +++ b/tests/queries/0_stateless/02404_data.TSV @@ -0,0 +1,10 @@ +0 1970-01-01 +1 1970-01-02 +2 1970-01-03 +3 1970-01-04 +4 1970-01-05 +5 1970-01-06 +6 1970-01-07 +7 1970-01-08 +8 1970-01-09 +9 1970-01-10 diff --git a/tests/queries/0_stateless/02404_data.TSVWithNames b/tests/queries/0_stateless/02404_data.TSVWithNames new file mode 100644 index 00000000000..23310234a8c --- /dev/null +++ b/tests/queries/0_stateless/02404_data.TSVWithNames @@ -0,0 +1,11 @@ +number toDate(number) +0 1970-01-01 +1 1970-01-02 +2 1970-01-03 +3 1970-01-04 +4 1970-01-05 +5 1970-01-06 +6 1970-01-07 +7 1970-01-08 +8 1970-01-09 +9 1970-01-10 diff --git a/tests/queries/0_stateless/02404_data.Values b/tests/queries/0_stateless/02404_data.Values new file mode 100644 index 00000000000..d9a621d7ec9 --- /dev/null +++ b/tests/queries/0_stateless/02404_data.Values @@ -0,0 +1 @@ +(0,'1970-01-01'),(1,'1970-01-02'),(2,'1970-01-03'),(3,'1970-01-04'),(4,'1970-01-05'),(5,'1970-01-06'),(6,'1970-01-07'),(7,'1970-01-08'),(8,'1970-01-09'),(9,'1970-01-10') \ No newline at end of file diff --git a/tests/queries/0_stateless/02404_schema_inference_cache_respect_format_settings.reference b/tests/queries/0_stateless/02404_schema_inference_cache_respect_format_settings.reference index 049603328d9..3d6b1021916 100644 --- a/tests/queries/0_stateless/02404_schema_inference_cache_respect_format_settings.reference +++ b/tests/queries/0_stateless/02404_schema_inference_cache_respect_format_settings.reference @@ -4,7 +4,7 @@ c2 Nullable(Date) c1 Nullable(Float64) c2 Nullable(Date) c1 Nullable(Int64) -c2 Nullable(DateTime64(9)) +c2 Nullable(DateTime) c1 UInt8 c2 Nullable(Date) 4 @@ -14,7 +14,7 @@ toDate(number) Nullable(Date) number Nullable(Float64) toDate(number) Nullable(Date) number Nullable(Int64) -toDate(number) Nullable(DateTime64(9)) +toDate(number) Nullable(DateTime) number Nullable(Int64) toDate(number) Nullable(Date) 4 @@ -24,7 +24,7 @@ c2 Nullable(Date) c1 Nullable(Float64) c2 Nullable(Date) c1 Nullable(Int64) -c2 Nullable(DateTime64(9)) +c2 Nullable(DateTime) c1 UInt8 c2 Nullable(Date) 4 @@ -34,7 +34,7 @@ toDate(number) Nullable(Date) number Nullable(Float64) toDate(number) Nullable(Date) number Nullable(Int64) -toDate(number) Nullable(DateTime64(9)) +toDate(number) Nullable(DateTime) number Nullable(Int64) toDate(number) Nullable(Date) 4 @@ -44,7 +44,7 @@ toDate(number) Nullable(Date) number Nullable(Float64) toDate(number) Nullable(Date) number Nullable(Int64) -toDate(number) Nullable(DateTime64(9)) +toDate(number) Nullable(DateTime) number Nullable(Int64) toDate(number) Nullable(Date) 4 @@ -54,7 +54,7 @@ c2 Nullable(Date) c1 Nullable(Float64) c2 Nullable(Date) c1 Nullable(Int64) -c2 Nullable(DateTime64(9)) +c2 Nullable(DateTime) c1 UInt8 c2 Nullable(Date) 4 @@ -64,7 +64,7 @@ toDate(number) Nullable(Date) number Nullable(Float64) toDate(number) Nullable(Date) number Nullable(Int64) -toDate(number) Nullable(DateTime64(9)) +toDate(number) Nullable(DateTime) number Nullable(Int64) toDate(number) Nullable(Date) 4 @@ -74,7 +74,7 @@ c2 Nullable(Date) c1 Nullable(Float64) c2 Nullable(Date) c1 Nullable(Int64) -c2 Nullable(DateTime64(9)) +c2 Nullable(DateTime) c1 UInt8 c2 Nullable(Date) 4 @@ -84,7 +84,7 @@ c2 Nullable(Date) c1 Nullable(Float64) c2 Nullable(Date) c1 Nullable(Int64) -c2 Nullable(DateTime64(9)) +c2 Nullable(DateTime) c1 UInt8 c2 Nullable(Date) 4 diff --git a/tests/queries/0_stateless/02417_load_marks_async.sh b/tests/queries/0_stateless/02417_load_marks_async.sh index 950656e7ab6..bcede9e4f5e 100755 --- a/tests/queries/0_stateless/02417_load_marks_async.sh +++ b/tests/queries/0_stateless/02417_load_marks_async.sh @@ -7,7 +7,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) ${CLICKHOUSE_CLIENT} -q "DROP TABLE IF EXISTS test;" -${CLICKHOUSE_CLIENT} -n -q " +${CLICKHOUSE_CLIENT} -q " CREATE TABLE test ( n0 UInt64, diff --git a/tests/queries/0_stateless/02420_final_setting_analyzer.reference b/tests/queries/0_stateless/02420_final_setting_analyzer.reference index 780a6e5de68..763ecad84ab 100644 --- a/tests/queries/0_stateless/02420_final_setting_analyzer.reference +++ b/tests/queries/0_stateless/02420_final_setting_analyzer.reference @@ -1,5 +1,5 @@ -- { echoOn } -set allow_experimental_analyzer=1; +set enable_analyzer=1; SYSTEM STOP MERGES tbl; -- simple test case create table if not exists replacing_mt (x String) engine=ReplacingMergeTree() ORDER BY x; diff --git a/tests/queries/0_stateless/02420_final_setting_analyzer.sql b/tests/queries/0_stateless/02420_final_setting_analyzer.sql index cbdec017602..89fff094825 100644 --- a/tests/queries/0_stateless/02420_final_setting_analyzer.sql +++ b/tests/queries/0_stateless/02420_final_setting_analyzer.sql @@ -1,5 +1,5 @@ -- { echoOn } -set allow_experimental_analyzer=1; +set enable_analyzer=1; SYSTEM STOP MERGES tbl; -- simple test case diff --git a/tests/queries/0_stateless/02421_decimal_in_precision_issue_41125.sql b/tests/queries/0_stateless/02421_decimal_in_precision_issue_41125.sql index fde893626c1..5a4e23c10bd 100644 --- a/tests/queries/0_stateless/02421_decimal_in_precision_issue_41125.sql +++ b/tests/queries/0_stateless/02421_decimal_in_precision_issue_41125.sql @@ -21,7 +21,7 @@ SELECT count() == 1 FROM dtest WHERE b IN toDecimal64('44.4000', 4); SELECT count() == 1 FROM dtest WHERE b IN toDecimal128('44.4000', 4); SELECT count() == 1 FROM dtest WHERE b IN toDecimal256('44.4000', 4); -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT count() == 0 FROM (SELECT '33.3' :: Decimal(9, 1) AS a WHERE a IN ('33.33' :: Decimal(9, 2))); diff --git a/tests/queries/0_stateless/02421_explain_subquery.sql b/tests/queries/0_stateless/02421_explain_subquery.sql index 2970003cb1c..02f45e8cc3b 100644 --- a/tests/queries/0_stateless/02421_explain_subquery.sql +++ b/tests/queries/0_stateless/02421_explain_subquery.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT count() > 3 FROM (EXPLAIN PIPELINE header = 1 SELECT * FROM system.numbers ORDER BY number DESC) WHERE explain LIKE '%Header: number UInt64%'; SELECT count() > 0 FROM (EXPLAIN PLAN SELECT * FROM system.numbers ORDER BY number DESC) WHERE explain ILIKE '%Sort%'; @@ -32,7 +32,7 @@ SELECT count() == 1 FROM (EXPLAIN ESTIMATE SELECT sum(a) FROM t1); DROP TABLE t1; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT count() > 3 FROM (EXPLAIN PIPELINE header = 1 SELECT * FROM system.numbers ORDER BY number DESC) WHERE explain LIKE '%Header: \_\_table1.number UInt64%'; SELECT count() > 0 FROM (EXPLAIN PLAN SELECT * FROM system.numbers ORDER BY number DESC) WHERE explain ILIKE '%Sort%'; diff --git a/tests/queries/0_stateless/02421_simple_queries_for_opentelemetry.sh b/tests/queries/0_stateless/02421_simple_queries_for_opentelemetry.sh index 98b571c5968..91e85eabcb8 100755 --- a/tests/queries/0_stateless/02421_simple_queries_for_opentelemetry.sh +++ b/tests/queries/0_stateless/02421_simple_queries_for_opentelemetry.sh @@ -9,7 +9,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # $2 - query function execute_query() { - ${CLICKHOUSE_CLIENT} --opentelemetry_start_trace_probability=1 --query_id $1 -nq " + ${CLICKHOUSE_CLIENT} --opentelemetry_start_trace_probability=1 --query_id $1 -q " ${2} " } @@ -18,7 +18,7 @@ function execute_query() # so we only to check the db.statement only function check_query_span_query_only() { -${CLICKHOUSE_CLIENT} -nq " +${CLICKHOUSE_CLIENT} -q " SYSTEM FLUSH LOGS; SELECT attribute['db.statement'] as query FROM system.opentelemetry_span_log @@ -31,7 +31,7 @@ ${CLICKHOUSE_CLIENT} -nq " function check_query_span() { -${CLICKHOUSE_CLIENT} -nq " +${CLICKHOUSE_CLIENT} -q " SYSTEM FLUSH LOGS; SELECT attribute['db.statement'] as query, attribute['clickhouse.read_rows'] as read_rows, @@ -47,7 +47,7 @@ ${CLICKHOUSE_CLIENT} -nq " # # Set up # -${CLICKHOUSE_CLIENT} -nq " +${CLICKHOUSE_CLIENT} -q " DROP TABLE IF EXISTS ${CLICKHOUSE_DATABASE}.opentelemetry_test; CREATE TABLE ${CLICKHOUSE_DATABASE}.opentelemetry_test (id UInt64) Engine=MergeTree Order By id; " @@ -79,4 +79,4 @@ check_query_span $query_id # ${CLICKHOUSE_CLIENT} -q " DROP TABLE IF EXISTS ${CLICKHOUSE_DATABASE}.opentelemetry_test; -" \ No newline at end of file +" diff --git a/tests/queries/0_stateless/02428_decimal_in_floating_point_literal.sql b/tests/queries/0_stateless/02428_decimal_in_floating_point_literal.sql index a84cb5572ba..a0d92115188 100644 --- a/tests/queries/0_stateless/02428_decimal_in_floating_point_literal.sql +++ b/tests/queries/0_stateless/02428_decimal_in_floating_point_literal.sql @@ -30,7 +30,7 @@ SELECT count() == 1 FROM decimal_in_float_test WHERE a NOT IN (33.333); SELECT count() == 1 FROM decimal_in_float_test WHERE b IN (44.44); SELECT count() == 1 FROM decimal_in_float_test WHERE b NOT IN (44.4,44.444); -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT toDecimal32(1.555,3) IN (1.5551); diff --git a/tests/queries/0_stateless/02428_parameterized_view.sh b/tests/queries/0_stateless/02428_parameterized_view.sh index c6f0927db36..95f82db5454 100755 --- a/tests/queries/0_stateless/02428_parameterized_view.sh +++ b/tests/queries/0_stateless/02428_parameterized_view.sh @@ -72,8 +72,8 @@ $CLICKHOUSE_CLIENT -q "INSERT INTO ${CLICKHOUSE_TEST_UNIQUE_NAME}.Catalog VALUES $CLICKHOUSE_CLIENT -q "INSERT INTO ${CLICKHOUSE_TEST_UNIQUE_NAME}.Catalog VALUES ('Paper', 20, 1)" $CLICKHOUSE_CLIENT -q "CREATE VIEW ${CLICKHOUSE_TEST_UNIQUE_NAME}.pv1 AS SELECT * FROM ${CLICKHOUSE_TEST_UNIQUE_NAME}.Catalog WHERE Price={price:UInt64}" $CLICKHOUSE_CLIENT -q "SELECT Price FROM ${CLICKHOUSE_TEST_UNIQUE_NAME}.pv1(price=20)" -$CLICKHOUSE_CLIENT -q "SELECT Price FROM \`${CLICKHOUSE_TEST_UNIQUE_NAME}.pv1\`(price=20) SETTINGS allow_experimental_analyzer = 0" 2>&1 | grep -Fq "UNKNOWN_FUNCTION" && echo 'ERROR' || echo 'OK' -$CLICKHOUSE_CLIENT -q "SELECT Price FROM \`${CLICKHOUSE_TEST_UNIQUE_NAME}.pv1\`(price=20) SETTINGS allow_experimental_analyzer = 1" +$CLICKHOUSE_CLIENT -q "SELECT Price FROM \`${CLICKHOUSE_TEST_UNIQUE_NAME}.pv1\`(price=20) SETTINGS enable_analyzer = 0" 2>&1 | grep -Fq "UNKNOWN_FUNCTION" && echo 'ERROR' || echo 'OK' +$CLICKHOUSE_CLIENT -q "SELECT Price FROM \`${CLICKHOUSE_TEST_UNIQUE_NAME}.pv1\`(price=20) SETTINGS enable_analyzer = 1" $CLICKHOUSE_CLIENT -q "INSERT INTO test_02428_Catalog VALUES ('Book2', 30, 8)" diff --git a/tests/queries/0_stateless/02434_cancel_insert_when_client_dies.sh b/tests/queries/0_stateless/02434_cancel_insert_when_client_dies.sh index dca8dae22c3..1548bef857f 100755 --- a/tests/queries/0_stateless/02434_cancel_insert_when_client_dies.sh +++ b/tests/queries/0_stateless/02434_cancel_insert_when_client_dies.sh @@ -1,5 +1,6 @@ #!/usr/bin/env bash -# Tags: no-random-settings, no-asan, no-msan, no-tsan, no-debug +# Tags: no-random-settings, no-asan, no-msan, no-tsan, no-debug, no-fasttest +# no-fasttest: The test runs for 40 seconds # shellcheck disable=SC2009 CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) diff --git a/tests/queries/0_stateless/02435_rollback_cancelled_queries.sh b/tests/queries/0_stateless/02435_rollback_cancelled_queries.sh index ba652013a57..1bc7d4cd1d3 100755 --- a/tests/queries/0_stateless/02435_rollback_cancelled_queries.sh +++ b/tests/queries/0_stateless/02435_rollback_cancelled_queries.sh @@ -1,5 +1,6 @@ #!/usr/bin/env bash -# Tags: no-random-settings, no-ordinary-database +# Tags: no-random-settings, no-ordinary-database, no-fasttest +# no-fasttest: The test is slow (too many small blocks) # shellcheck disable=SC2009 CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) diff --git a/tests/queries/0_stateless/02445_replicated_db_alter_partition.sh b/tests/queries/0_stateless/02445_replicated_db_alter_partition.sh index 4d9048354a1..f716a6aa779 100755 --- a/tests/queries/0_stateless/02445_replicated_db_alter_partition.sh +++ b/tests/queries/0_stateless/02445_replicated_db_alter_partition.sh @@ -1,5 +1,8 @@ #!/usr/bin/env bash +# Tags: no-fasttest +# no-fasttest: Slow timeouts + CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh diff --git a/tests/queries/0_stateless/02447_drop_database_replica.sh b/tests/queries/0_stateless/02447_drop_database_replica.sh index c6bf298f944..abe99398a56 100755 --- a/tests/queries/0_stateless/02447_drop_database_replica.sh +++ b/tests/queries/0_stateless/02447_drop_database_replica.sh @@ -1,8 +1,9 @@ #!/usr/bin/env bash -# Tags: no-parallel +# Tags: no-parallel, no-fasttest # no-parallel: This test is not parallel because when we execute system-wide SYSTEM DROP REPLICA, # other tests might shut down the storage in parallel and the test will fail. +# no-fasttest: It has several tests with timeouts for inactive replicas CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02451_order_by_monotonic.sh b/tests/queries/0_stateless/02451_order_by_monotonic.sh index 7d1356b4445..fa0a37678f0 100755 --- a/tests/queries/0_stateless/02451_order_by_monotonic.sh +++ b/tests/queries/0_stateless/02451_order_by_monotonic.sh @@ -5,7 +5,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh opts=( - "--allow_experimental_analyzer=1" + "--enable_analyzer=1" ) function explain_sort_description() diff --git a/tests/queries/0_stateless/02458_insert_select_progress_tcp.sh b/tests/queries/0_stateless/02458_insert_select_progress_tcp.sh index ae3ea017fbb..178da822d41 100755 --- a/tests/queries/0_stateless/02458_insert_select_progress_tcp.sh +++ b/tests/queries/0_stateless/02458_insert_select_progress_tcp.sh @@ -4,7 +4,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " drop table if exists insert_select_progress_tcp; create table insert_select_progress_tcp(s UInt16) engine = MergeTree order by s; " diff --git a/tests/queries/0_stateless/02459_group_by_all.sql b/tests/queries/0_stateless/02459_group_by_all.sql index 4f08ee331a4..8281c201edf 100644 --- a/tests/queries/0_stateless/02459_group_by_all.sql +++ b/tests/queries/0_stateless/02459_group_by_all.sql @@ -21,7 +21,7 @@ select substring(a, 1, 3), substring(substring(substring(a, c, count(b)), 1, cou select substring(a, 1, 3), substring(a, 1, count(b)) from group_by_all group by all; select count(b) AS len, substring(a, 1, 3), substring(a, 1, len) from group_by_all group by all; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; select a, count(b) from group_by_all group by all order by a; select substring(a, 1, 3), count(b) from group_by_all group by all; diff --git a/tests/queries/0_stateless/02474_analyzer_subqueries_table_expression_modifiers.sql b/tests/queries/0_stateless/02474_analyzer_subqueries_table_expression_modifiers.sql index 5ac8c79d4ed..e86c867043d 100644 --- a/tests/queries/0_stateless/02474_analyzer_subqueries_table_expression_modifiers.sql +++ b/tests/queries/0_stateless/02474_analyzer_subqueries_table_expression_modifiers.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT * FROM (SELECT 1) FINAL; -- { serverError UNSUPPORTED_METHOD } SELECT * FROM (SELECT 1) SAMPLE 1/2; -- { serverError UNSUPPORTED_METHOD } diff --git a/tests/queries/0_stateless/02475_analyzer_join_tree_subquery.sql b/tests/queries/0_stateless/02475_analyzer_join_tree_subquery.sql index eda90529166..c9e7ac19157 100644 --- a/tests/queries/0_stateless/02475_analyzer_join_tree_subquery.sql +++ b/tests/queries/0_stateless/02475_analyzer_join_tree_subquery.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; WITH subquery AS (SELECT sum(number) FROM numbers(10)) SELECT * FROM subquery; diff --git a/tests/queries/0_stateless/02475_analyzer_subquery_compound_expression.sql b/tests/queries/0_stateless/02475_analyzer_subquery_compound_expression.sql index f96c834c057..fc9e9d44b42 100644 --- a/tests/queries/0_stateless/02475_analyzer_subquery_compound_expression.sql +++ b/tests/queries/0_stateless/02475_analyzer_subquery_compound_expression.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT cast(tuple(1, 2), 'Tuple(value_1 UInt64, value_2 UInt64)') AS value, value.value_1, value.value_2; diff --git a/tests/queries/0_stateless/02475_or_function_alias_and_const_where.sql b/tests/queries/0_stateless/02475_or_function_alias_and_const_where.sql index ddb0f70c6de..53b97a1dd2a 100644 --- a/tests/queries/0_stateless/02475_or_function_alias_and_const_where.sql +++ b/tests/queries/0_stateless/02475_or_function_alias_and_const_where.sql @@ -1,2 +1,2 @@ SELECT (number = 1) AND (number = 2) AS value, sum(value) OVER () FROM numbers(1) WHERE 1; -SELECT (number = 1) AND (number = 2) AS value, sum(value) OVER () FROM numbers(1) WHERE 1 SETTINGS allow_experimental_analyzer=1; \ No newline at end of file +SELECT (number = 1) AND (number = 2) AS value, sum(value) OVER () FROM numbers(1) WHERE 1 SETTINGS enable_analyzer=1; diff --git a/tests/queries/0_stateless/02476_analyzer_identifier_hints.sh b/tests/queries/0_stateless/02476_analyzer_identifier_hints.sh index 0702c146426..92f519a9f8a 100755 --- a/tests/queries/0_stateless/02476_analyzer_identifier_hints.sh +++ b/tests/queries/0_stateless/02476_analyzer_identifier_hints.sh @@ -6,7 +6,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " DROP TABLE IF EXISTS test_table; CREATE TABLE test_table ( @@ -17,64 +17,64 @@ $CLICKHOUSE_CLIENT -n -q " INSERT INTO test_table VALUES (0, 'Value'); "; -$CLICKHOUSE_CLIENT -q "SELECT value_ FROM test_table SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT value_ FROM test_table SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT test_table.value_ FROM test_table SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT test_table.value_ FROM test_table SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['test_table.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT test_tabl.value_ FROM test_table SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT test_tabl.value_ FROM test_table SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['test_table.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT test_table.value_ FROM test_table AS test_table_alias SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT test_table.value_ FROM test_table AS test_table_alias SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['test_table.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT test_tabl.value_ FROM test_table AS test_table_alias SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT test_tabl.value_ FROM test_table AS test_table_alias SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['test_table.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT test_table_alias.value_ FROM test_table AS test_table_alias SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT test_table_alias.value_ FROM test_table AS test_table_alias SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['test_table_alias.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT test_table_alia.value_ FROM test_table AS test_table_alias SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT test_table_alia.value_ FROM test_table AS test_table_alias SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['test_table_alias.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT value_ FROM (SELECT 1 AS value) SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT value_ FROM (SELECT 1 AS value) SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT subquery.value_ FROM (SELECT 1 AS value) AS subquery SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT subquery.value_ FROM (SELECT 1 AS value) AS subquery SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['subquery.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT subquer.value_ FROM (SELECT 1 AS value) AS subquery SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT subquer.value_ FROM (SELECT 1 AS value) AS subquery SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['subquery.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT 1 AS value) SELECT value_ FROM cte_subquery SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT 1 AS value) SELECT value_ FROM cte_subquery SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT 1 AS value) SELECT cte_subquery.value_ FROM cte_subquery SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT 1 AS value) SELECT cte_subquery.value_ FROM cte_subquery SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['cte_subquery.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT 1 AS value) SELECT cte_subquer.value_ FROM cte_subquery SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT 1 AS value) SELECT cte_subquer.value_ FROM cte_subquery SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['cte_subquery.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT 1 AS value) SELECT cte_subquery_alias.value_ FROM cte_subquery AS cte_subquery_alias SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT 1 AS value) SELECT cte_subquery_alias.value_ FROM cte_subquery AS cte_subquery_alias SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['cte_subquery_alias.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT 1 AS value) SELECT cte_subquery_alia.value_ FROM cte_subquery AS cte_subquery_alias SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT 1 AS value) SELECT cte_subquery_alia.value_ FROM cte_subquery AS cte_subquery_alias SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['cte_subquery_alias.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT 1 AS constant_value, constant_valu SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT 1 AS constant_value, constant_valu SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['constant_value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT 1 AS constant_value, constant_valu SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT 1 AS constant_value, constant_valu SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['constant_value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT 1 AS constant_value, arrayMap(lambda_argument -> lambda_argument + constant_valu, [1, 2, 3]) SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT 1 AS constant_value, arrayMap(lambda_argument -> lambda_argument + constant_valu, [1, 2, 3]) SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['constant_value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "WITH 1 AS constant_value SELECT (SELECT constant_valu) SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "WITH 1 AS constant_value SELECT (SELECT constant_valu) SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['constant_value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " DROP TABLE IF EXISTS test_table_compound; CREATE TABLE test_table_compound ( @@ -85,64 +85,64 @@ $CLICKHOUSE_CLIENT -n -q " INSERT INTO test_table_compound VALUES (0, tuple('Value_1')); "; -$CLICKHOUSE_CLIENT -q "SELECT value.value_ FROM test_table_compound SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT value.value_ FROM test_table_compound SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT test_table_compound.value.value_ FROM test_table_compound SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT test_table_compound.value.value_ FROM test_table_compound SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['test_table_compound.value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT test_tabl_compound.value.value_ FROM test_table_compound SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT test_tabl_compound.value.value_ FROM test_table_compound SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['test_table_compound.value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT test_table_compound.value.value_ FROM test_table_compound AS test_table_compound_alias SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT test_table_compound.value.value_ FROM test_table_compound AS test_table_compound_alias SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['test_table_compound.value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT test_tabl_compound.value.value_ FROM test_table_compound AS test_table_compound_alias SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT test_tabl_compound.value.value_ FROM test_table_compound AS test_table_compound_alias SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['test_table_compound.value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT test_table_compound_alias.value.value_ FROM test_table_compound AS test_table_compound_alias SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT test_table_compound_alias.value.value_ FROM test_table_compound AS test_table_compound_alias SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['test_table_compound_alias.value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT test_table_compound_alia.value.value_ FROM test_table_compound AS test_table_compound_alias SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT test_table_compound_alia.value.value_ FROM test_table_compound AS test_table_compound_alias SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['test_table_compound_alias.value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT value.value_ FROM (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT value.value_ FROM (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT subquery.value.value_ FROM (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) AS subquery SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT subquery.value.value_ FROM (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) AS subquery SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['subquery.value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT subquer.value.value_ FROM (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) AS subquery SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT subquer.value.value_ FROM (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) AS subquery SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['subquery.value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) SELECT value.value_ FROM cte_subquery SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) SELECT value.value_ FROM cte_subquery SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) SELECT cte_subquery.value.value_ FROM cte_subquery SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) SELECT cte_subquery.value.value_ FROM cte_subquery SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['cte_subquery.value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) SELECT cte_subquer.value.value_ FROM cte_subquery SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) SELECT cte_subquer.value.value_ FROM cte_subquery SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['cte_subquery.value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) SELECT cte_subquery_alias.value.value_ FROM cte_subquery AS cte_subquery_alias SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) SELECT cte_subquery_alias.value.value_ FROM cte_subquery AS cte_subquery_alias SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['cte_subquery_alias.value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) SELECT cte_subquery_alia.value.value_ FROM cte_subquery AS cte_subquery_alias SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "WITH cte_subquery AS (SELECT cast(tuple(1), 'Tuple(value_1 String)') AS value) SELECT cte_subquery_alia.value.value_ FROM cte_subquery AS cte_subquery_alias SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['cte_subquery_alias.value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT cast(tuple(1), 'Tuple(value_1 String)') AS constant_value, constant_value.value_ SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT cast(tuple(1), 'Tuple(value_1 String)') AS constant_value, constant_value.value_ SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['constant_value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT cast(tuple(1), 'Tuple(value_1 String)') AS constant_value, constant_valu.value_ SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT cast(tuple(1), 'Tuple(value_1 String)') AS constant_value, constant_valu.value_ SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['constant_value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT cast(tuple(1), 'Tuple(value_1 String)') AS constant_value, arrayMap(lambda_argument -> lambda_argument + constant_value.value_, [1, 2, 3]) SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT cast(tuple(1), 'Tuple(value_1 String)') AS constant_value, arrayMap(lambda_argument -> lambda_argument + constant_value.value_, [1, 2, 3]) SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['constant_value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "WITH cast(tuple(1), 'Tuple(value_1 String)') AS constant_value SELECT (SELECT constant_value.value_) SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "WITH cast(tuple(1), 'Tuple(value_1 String)') AS constant_value SELECT (SELECT constant_value.value_) SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['constant_value.value_1'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " DROP TABLE IF EXISTS test_table_1; CREATE TABLE test_table_1 ( @@ -162,30 +162,30 @@ $CLICKHOUSE_CLIENT -n -q " INSERT INTO test_table_2 VALUES (0, 'Value'); "; -$CLICKHOUSE_CLIENT -q "SELECT test_table_1.value_ FROM test_table_1 INNER JOIN test_table_2 ON test_table_1.id = test_table_2.id SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT test_table_1.value_ FROM test_table_1 INNER JOIN test_table_2 ON test_table_1.id = test_table_2.id SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['test_table_1.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT test_table_2.value_ FROM test_table_1 INNER JOIN test_table_2 ON test_table_1.id = test_table_2.id SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT test_table_2.value_ FROM test_table_1 INNER JOIN test_table_2 ON test_table_1.id = test_table_2.id SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['test_table_2.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT t1.value_ FROM test_table_1 AS t1 INNER JOIN test_table_2 AS t2 ON t1.id = t2.id SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT t1.value_ FROM test_table_1 AS t1 INNER JOIN test_table_2 AS t2 ON t1.id = t2.id SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['t1.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT t2.value_ FROM test_table_1 AS t1 INNER JOIN test_table_2 AS t2 ON t1.id = t2.id SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT t2.value_ FROM test_table_1 AS t1 INNER JOIN test_table_2 AS t2 ON t1.id = t2.id SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['t2.value'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT [1] AS a, a.size1 SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT [1] AS a, a.size1 SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['a.size0'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT ((1))::Tuple(a Tuple(b UInt32)) AS t, t.c SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT ((1))::Tuple(a Tuple(b UInt32)) AS t, t.c SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['t.a'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT ((1))::Tuple(a Tuple(b UInt32)) AS t, t.a.c SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT ((1))::Tuple(a Tuple(b UInt32)) AS t, t.a.c SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['t.a.b'\]" &>/dev/null; $CLICKHOUSE_CLIENT -q "SELECT 1"; -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " DROP TABLE test_table; DROP TABLE test_table_compound; DROP TABLE test_table_1; diff --git a/tests/queries/0_stateless/02476_analyzer_join_with_unused_columns.sql b/tests/queries/0_stateless/02476_analyzer_join_with_unused_columns.sql index ca937e01238..feb6786ffb6 100644 --- a/tests/queries/0_stateless/02476_analyzer_join_with_unused_columns.sql +++ b/tests/queries/0_stateless/02476_analyzer_join_with_unused_columns.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT subquery_1.id, subquery_2.id FROM (SELECT 1 AS id, 2 AS value) AS subquery_1, (SELECT 3 AS id, 4 AS value) AS subquery_2; diff --git a/tests/queries/0_stateless/02476_fuse_sum_count.sql b/tests/queries/0_stateless/02476_fuse_sum_count.sql index 315bbd10a65..2319e81440b 100644 --- a/tests/queries/0_stateless/02476_fuse_sum_count.sql +++ b/tests/queries/0_stateless/02476_fuse_sum_count.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET optimize_syntax_fuse_functions = 1; DROP TABLE IF EXISTS fuse_tbl; diff --git a/tests/queries/0_stateless/02477_analyzer_array_join_with_join.sql b/tests/queries/0_stateless/02477_analyzer_array_join_with_join.sql index 7e4ecb98ae8..3d2cc1b1620 100644 --- a/tests/queries/0_stateless/02477_analyzer_array_join_with_join.sql +++ b/tests/queries/0_stateless/02477_analyzer_array_join_with_join.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02477_analyzer_ast_key_condition_crash.sql b/tests/queries/0_stateless/02477_analyzer_ast_key_condition_crash.sql index 53f3a9b23ec..2fc1cc45ce0 100644 --- a/tests/queries/0_stateless/02477_analyzer_ast_key_condition_crash.sql +++ b/tests/queries/0_stateless/02477_analyzer_ast_key_condition_crash.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02477_analyzer_function_hints.sh b/tests/queries/0_stateless/02477_analyzer_function_hints.sh index f83935e47fb..23cc5651257 100755 --- a/tests/queries/0_stateless/02477_analyzer_function_hints.sh +++ b/tests/queries/0_stateless/02477_analyzer_function_hints.sh @@ -6,22 +6,22 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -q "SELECT plu(1, 1) SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT plu(1, 1) SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['plus'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT uniqExac(1, 1) SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT uniqExac(1, 1) SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['uniqExact'\]" &>/dev/null; $CLICKHOUSE_CLIENT -q "DROP FUNCTION IF EXISTS test_user_defined_function_$CLICKHOUSE_DATABASE;" $CLICKHOUSE_CLIENT -q "CREATE FUNCTION test_user_defined_function_$CLICKHOUSE_DATABASE AS x -> x + 1;" -$CLICKHOUSE_CLIENT -q "SELECT test_user_defined_function_${CLICKHOUSE_DATABASE}A(1) SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT test_user_defined_function_${CLICKHOUSE_DATABASE}A(1) SETTINGS enable_analyzer = 1;" 2>&1 \ | grep -E "Maybe you meant: \[.*'test_user_defined_function_$CLICKHOUSE_DATABASE'.*\]" &>/dev/null; $CLICKHOUSE_CLIENT -q "DROP FUNCTION test_user_defined_function_$CLICKHOUSE_DATABASE"; -$CLICKHOUSE_CLIENT -q "WITH (x -> x + 1) AS lambda_function SELECT lambda_functio(1) SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "WITH (x -> x + 1) AS lambda_function SELECT lambda_functio(1) SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['lambda_function'\]" &>/dev/null; -$CLICKHOUSE_CLIENT -q "SELECT row_numbe() OVER (PARTITION BY 1) SETTINGS allow_experimental_analyzer = 1;" 2>&1 \ +$CLICKHOUSE_CLIENT -q "SELECT row_numbe() OVER (PARTITION BY 1) SETTINGS enable_analyzer = 1;" 2>&1 \ | grep "Maybe you meant: \['row_number'\]" &>/dev/null; $CLICKHOUSE_CLIENT -q "SELECT 1"; diff --git a/tests/queries/0_stateless/02477_exists_fuzz_43478.sql b/tests/queries/0_stateless/02477_exists_fuzz_43478.sql index 8ec876eb252..c225befed0e 100644 --- a/tests/queries/0_stateless/02477_exists_fuzz_43478.sql +++ b/tests/queries/0_stateless/02477_exists_fuzz_43478.sql @@ -1,3 +1,3 @@ create table test_rows_compact_part__fuzz_11 (x UInt32) engine = MergeTree order by x; insert into test_rows_compact_part__fuzz_11 select 1; -select 1 from test_rows_compact_part__fuzz_11 where exists(select 1) settings allow_experimental_analyzer=1; +select 1 from test_rows_compact_part__fuzz_11 where exists(select 1) settings enable_analyzer=1; diff --git a/tests/queries/0_stateless/02477_fuse_quantiles.sql b/tests/queries/0_stateless/02477_fuse_quantiles.sql index c0719d771d7..8ddc029f75f 100644 --- a/tests/queries/0_stateless/02477_fuse_quantiles.sql +++ b/tests/queries/0_stateless/02477_fuse_quantiles.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET optimize_syntax_fuse_functions = 1; DROP TABLE IF EXISTS fuse_tbl; diff --git a/tests/queries/0_stateless/02477_logical_expressions_optimizer_low_cardinality.reference b/tests/queries/0_stateless/02477_logical_expressions_optimizer_low_cardinality.reference index 11c178ac0d0..0ce90cde108 100644 --- a/tests/queries/0_stateless/02477_logical_expressions_optimizer_low_cardinality.reference +++ b/tests/queries/0_stateless/02477_logical_expressions_optimizer_low_cardinality.reference @@ -19,7 +19,7 @@ QUERY id: 0 COLUMN id: 2, column_name: a, result_type: LowCardinality(String), source_id: 3 CONSTANT id: 8, constant_value: Tuple_(\'x\', \'y\'), constant_value_type: Tuple(String, String) CONSTANT id: 9, constant_value: UInt64_0, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE (a = \'x\') OR (\'y\' = a) @@ -41,7 +41,7 @@ QUERY id: 0 COLUMN id: 2, column_name: a, result_type: LowCardinality(String), source_id: 3 CONSTANT id: 8, constant_value: Tuple_(\'x\', \'y\'), constant_value_type: Tuple(String, String) CONSTANT id: 9, constant_value: UInt64_0, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE (a != \'x\') AND (a != \'y\') @@ -63,7 +63,7 @@ QUERY id: 0 COLUMN id: 2, column_name: a, result_type: LowCardinality(String), source_id: 3 CONSTANT id: 8, constant_value: Tuple_(\'x\', \'y\'), constant_value_type: Tuple(String, String) CONSTANT id: 9, constant_value: \'UInt8\', constant_value_type: String - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE (a != \'x\') AND (\'y\' != a) @@ -85,7 +85,7 @@ QUERY id: 0 COLUMN id: 2, column_name: a, result_type: LowCardinality(String), source_id: 3 CONSTANT id: 8, constant_value: Tuple_(\'x\', \'y\'), constant_value_type: Tuple(String, String) CONSTANT id: 9, constant_value: \'UInt8\', constant_value_type: String - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE (b = 0) OR (b = 1) @@ -111,7 +111,7 @@ QUERY id: 0 LIST id: 11, nodes: 2 COLUMN id: 8, column_name: b, result_type: UInt32, source_id: 3 CONSTANT id: 12, constant_value: UInt64_1, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE (b != 0) AND (b != 1) @@ -137,4 +137,4 @@ QUERY id: 0 LIST id: 11, nodes: 2 COLUMN id: 8, column_name: b, result_type: UInt32, source_id: 3 CONSTANT id: 12, constant_value: UInt64_1, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 diff --git a/tests/queries/0_stateless/02477_logical_expressions_optimizer_low_cardinality.sql b/tests/queries/0_stateless/02477_logical_expressions_optimizer_low_cardinality.sql index 976b21a7e29..b328e9658d0 100644 --- a/tests/queries/0_stateless/02477_logical_expressions_optimizer_low_cardinality.sql +++ b/tests/queries/0_stateless/02477_logical_expressions_optimizer_low_cardinality.sql @@ -5,21 +5,21 @@ CREATE TABLE t_logical_expressions_optimizer_low_cardinality (a LowCardinality(S -- LowCardinality case, ignore optimize_min_equality_disjunction_chain_length limit, optimizer applied -- Chain of OR equals EXPLAIN SYNTAX SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE a = 'x' OR a = 'y'; -EXPLAIN QUERY TREE SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE a = 'x' OR a = 'y' SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE a = 'x' OR a = 'y' SETTINGS enable_analyzer = 1; EXPLAIN SYNTAX SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE a = 'x' OR 'y' = a; -EXPLAIN QUERY TREE SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE a = 'x' OR 'y' = a SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE a = 'x' OR 'y' = a SETTINGS enable_analyzer = 1; -- Chain of AND notEquals EXPLAIN SYNTAX SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE a <> 'x' AND a <> 'y'; -EXPLAIN QUERY TREE SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE a <> 'x' AND a <> 'y' SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE a <> 'x' AND a <> 'y' SETTINGS enable_analyzer = 1; EXPLAIN SYNTAX SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE a <> 'x' AND 'y' <> a; -EXPLAIN QUERY TREE SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE a <> 'x' AND 'y' <> a SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE a <> 'x' AND 'y' <> a SETTINGS enable_analyzer = 1; -- Non-LowCardinality case, optimizer not applied for short chains -- Chain of OR equals EXPLAIN SYNTAX SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE b = 0 OR b = 1; -EXPLAIN QUERY TREE SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE b = 0 OR b = 1 SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE b = 0 OR b = 1 SETTINGS enable_analyzer = 1; -- Chain of AND notEquals EXPLAIN SYNTAX SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE b <> 0 AND b <> 1; -EXPLAIN QUERY TREE SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE b <> 0 AND b <> 1 SETTINGS allow_experimental_analyzer = 1; +EXPLAIN QUERY TREE SELECT a FROM t_logical_expressions_optimizer_low_cardinality WHERE b <> 0 AND b <> 1 SETTINGS enable_analyzer = 1; DROP TABLE t_logical_expressions_optimizer_low_cardinality; diff --git a/tests/queries/0_stateless/02478_analyzer_table_expression_aliases.sql b/tests/queries/0_stateless/02478_analyzer_table_expression_aliases.sql index 66f50f7b26d..a1eb88c634d 100644 --- a/tests/queries/0_stateless/02478_analyzer_table_expression_aliases.sql +++ b/tests/queries/0_stateless/02478_analyzer_table_expression_aliases.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02478_window_frame_type_groups.sql b/tests/queries/0_stateless/02478_window_frame_type_groups.sql index f762bcb61ee..a01e1813cc1 100644 --- a/tests/queries/0_stateless/02478_window_frame_type_groups.sql +++ b/tests/queries/0_stateless/02478_window_frame_type_groups.sql @@ -1,7 +1,7 @@ -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT toUInt64(dense_rank(1) OVER (ORDER BY 100 ASC GROUPS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING)) FROM numbers(10); -- { serverError NOT_IMPLEMENTED } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT toUInt64(dense_rank(1) OVER (ORDER BY 100 ASC GROUPS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING)) FROM numbers(10); -- { serverError NOT_IMPLEMENTED } diff --git a/tests/queries/0_stateless/02479_analyzer_aggregation_crash.sql b/tests/queries/0_stateless/02479_analyzer_aggregation_crash.sql index c931a3ab634..1e890740665 100644 --- a/tests/queries/0_stateless/02479_analyzer_aggregation_crash.sql +++ b/tests/queries/0_stateless/02479_analyzer_aggregation_crash.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET compile_aggregate_expressions = 1; SET min_count_to_compile_aggregate_expression = 0; diff --git a/tests/queries/0_stateless/02479_analyzer_aggregation_totals_rollup_crash_fix.sql b/tests/queries/0_stateless/02479_analyzer_aggregation_totals_rollup_crash_fix.sql index 6cd3e6a9385..004e61ee1a2 100644 --- a/tests/queries/0_stateless/02479_analyzer_aggregation_totals_rollup_crash_fix.sql +++ b/tests/queries/0_stateless/02479_analyzer_aggregation_totals_rollup_crash_fix.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT anyLast(number) FROM numbers(1) GROUP BY number WITH ROLLUP WITH TOTALS; diff --git a/tests/queries/0_stateless/02479_analyzer_join_with_constants.sql b/tests/queries/0_stateless/02479_analyzer_join_with_constants.sql index 9f77cf39f47..503bbe63347 100644 --- a/tests/queries/0_stateless/02479_analyzer_join_with_constants.sql +++ b/tests/queries/0_stateless/02479_analyzer_join_with_constants.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT * FROM (SELECT 1 AS id) AS t1 INNER JOIN (SELECT 1 AS id) AS t2 ON t1.id = t2.id AND 1; diff --git a/tests/queries/0_stateless/02479_mysql_connect_to_self.sql b/tests/queries/0_stateless/02479_mysql_connect_to_self.sql index cf2220073d3..b8ed7e0c03e 100644 --- a/tests/queries/0_stateless/02479_mysql_connect_to_self.sql +++ b/tests/queries/0_stateless/02479_mysql_connect_to_self.sql @@ -7,7 +7,7 @@ DROP TABLE IF EXISTS foo; CREATE TABLE foo (key UInt32, a String, b Int64, c String) ENGINE = TinyLog; INSERT INTO foo VALUES (1, 'one', -1, 'een'), (2, 'two', -2, 'twee'), (3, 'three', -3, 'drie'), (4, 'four', -4, 'vier'), (5, 'five', -5, 'vijf'); -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT '---'; SELECT * FROM mysql('127.0.0.1:9004', currentDatabase(), foo, 'default', '', SETTINGS connect_timeout = 100, connection_wait_timeout = 100) ORDER BY key; diff --git a/tests/queries/0_stateless/02480_analyzer_alias_nullptr.sql b/tests/queries/0_stateless/02480_analyzer_alias_nullptr.sql index f6b381e5c70..07503de1b10 100644 --- a/tests/queries/0_stateless/02480_analyzer_alias_nullptr.sql +++ b/tests/queries/0_stateless/02480_analyzer_alias_nullptr.sql @@ -1,3 +1,3 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT min(b), x AS b FROM (SELECT max(number) FROM numbers(1)); -- { serverError UNKNOWN_IDENTIFIER } diff --git a/tests/queries/0_stateless/02480_tets_show_full.sh b/tests/queries/0_stateless/02480_tets_show_full.sh index 5f5040ba128..50184857a1f 100755 --- a/tests/queries/0_stateless/02480_tets_show_full.sh +++ b/tests/queries/0_stateless/02480_tets_show_full.sh @@ -7,7 +7,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) database=$($CLICKHOUSE_CLIENT -q 'SELECT currentDatabase()') -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " DROP TABLE IF EXISTS test_02480_table; DROP VIEW IF EXISTS test_02480_view; CREATE TABLE test_02480_table (id Int64) ENGINE=MergeTree ORDER BY id; diff --git a/tests/queries/0_stateless/02480_tlp_nan.reference b/tests/queries/0_stateless/02480_tlp_nan.reference index befd1f66564..29e5bffa7f0 100644 --- a/tests/queries/0_stateless/02480_tlp_nan.reference +++ b/tests/queries/0_stateless/02480_tlp_nan.reference @@ -1,21 +1,21 @@ -- {echo} -SELECT sqrt(-1) as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=1; +SELECT sqrt(-1) as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=1; nan 0 1 0 -SELECT sqrt(-1) as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=0; +SELECT sqrt(-1) as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=0; nan 0 1 0 -SELECT -inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=1; +SELECT -inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=1; -inf 0 1 0 -SELECT -inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=0; +SELECT -inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=0; -inf 0 1 0 -SELECT NULL as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=1; +SELECT NULL as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=1; \N \N \N 1 -SELECT NULL as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=0; +SELECT NULL as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=0; \N \N \N 1 -SELECT inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=1; +SELECT inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=1; inf 0 1 0 -SELECT inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=0; +SELECT inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=0; inf 0 1 0 -SELECT nan as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=1; +SELECT nan as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=1; nan 0 1 0 -SELECT nan as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=0; +SELECT nan as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=0; nan 0 1 0 diff --git a/tests/queries/0_stateless/02480_tlp_nan.sql b/tests/queries/0_stateless/02480_tlp_nan.sql index e24bc9a9830..55318e0cb30 100644 --- a/tests/queries/0_stateless/02480_tlp_nan.sql +++ b/tests/queries/0_stateless/02480_tlp_nan.sql @@ -1,15 +1,15 @@ -- {echo} -SELECT sqrt(-1) as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=1; -SELECT sqrt(-1) as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=0; +SELECT sqrt(-1) as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=1; +SELECT sqrt(-1) as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=0; -SELECT -inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=1; -SELECT -inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=0; +SELECT -inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=1; +SELECT -inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=0; -SELECT NULL as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=1; -SELECT NULL as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=0; +SELECT NULL as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=1; +SELECT NULL as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=0; -SELECT inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=1; -SELECT inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=0; +SELECT inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=1; +SELECT inf as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=0; -SELECT nan as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=1; -SELECT nan as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS allow_experimental_analyzer=0; +SELECT nan as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=1; +SELECT nan as x, not(x), not(not(x)), (not(x)) IS NULL SETTINGS enable_analyzer=0; diff --git a/tests/queries/0_stateless/02481_aggregation_in_order_plan.sql b/tests/queries/0_stateless/02481_aggregation_in_order_plan.sql index 490060dee6f..139e0ed4b77 100644 --- a/tests/queries/0_stateless/02481_aggregation_in_order_plan.sql +++ b/tests/queries/0_stateless/02481_aggregation_in_order_plan.sql @@ -5,5 +5,5 @@ insert into tab select 0, number % 3, 2 - intDiv(number, 3), (number % 3 + 1) * insert into tab select 0, number % 3, 2 - intDiv(number, 3), (number % 3 + 1) * 100 from numbers(6); select a, any(b), c, d from tab where b = 1 group by a, c, d order by c, d settings optimize_aggregation_in_order=1, query_plan_aggregation_in_order=1; -select * from (explain actions = 1, sorting=1 select a, any(b), c, d from tab where b = 1 group by a, c, d settings optimize_aggregation_in_order=1, query_plan_aggregation_in_order=1) where explain like '%ReadFromMergeTree%' or explain like '%Aggregating%' or explain like '%Order:%' settings allow_experimental_analyzer=0; -select * from (explain actions = 1, sorting=1 select a, any(b), c, d from tab where b = 1 group by a, c, d settings optimize_aggregation_in_order=1, query_plan_aggregation_in_order=1) where explain like '%ReadFromMergeTree%' or explain like '%Aggregating%' or explain like '%Order:%' settings allow_experimental_analyzer=1; +select * from (explain actions = 1, sorting=1 select a, any(b), c, d from tab where b = 1 group by a, c, d settings optimize_aggregation_in_order=1, query_plan_aggregation_in_order=1) where explain like '%ReadFromMergeTree%' or explain like '%Aggregating%' or explain like '%Order:%' settings enable_analyzer=0; +select * from (explain actions = 1, sorting=1 select a, any(b), c, d from tab where b = 1 group by a, c, d settings optimize_aggregation_in_order=1, query_plan_aggregation_in_order=1) where explain like '%ReadFromMergeTree%' or explain like '%Aggregating%' or explain like '%Order:%' settings enable_analyzer=1; diff --git a/tests/queries/0_stateless/02481_analyzer_join_alias_unknown_identifier_crash.sql b/tests/queries/0_stateless/02481_analyzer_join_alias_unknown_identifier_crash.sql index 0c5f0eba750..8b5b272f547 100644 --- a/tests/queries/0_stateless/02481_analyzer_join_alias_unknown_identifier_crash.sql +++ b/tests/queries/0_stateless/02481_analyzer_join_alias_unknown_identifier_crash.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table_join_1; CREATE TABLE test_table_join_1 diff --git a/tests/queries/0_stateless/02481_analyzer_optimize_aggregation_arithmetics.sql b/tests/queries/0_stateless/02481_analyzer_optimize_aggregation_arithmetics.sql index ca91d137bf4..e68de0af522 100644 --- a/tests/queries/0_stateless/02481_analyzer_optimize_aggregation_arithmetics.sql +++ b/tests/queries/0_stateless/02481_analyzer_optimize_aggregation_arithmetics.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET optimize_arithmetic_operations_in_aggregate_functions = 1; -- { echoOn } diff --git a/tests/queries/0_stateless/02481_analyzer_optimize_grouping_sets_keys.sql b/tests/queries/0_stateless/02481_analyzer_optimize_grouping_sets_keys.sql index fef71fdf94f..8e6b132f5b7 100644 --- a/tests/queries/0_stateless/02481_analyzer_optimize_grouping_sets_keys.sql +++ b/tests/queries/0_stateless/02481_analyzer_optimize_grouping_sets_keys.sql @@ -1,4 +1,4 @@ -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; set optimize_syntax_fuse_functions = 0; EXPLAIN QUERY TREE run_passes=1 diff --git a/tests/queries/0_stateless/02482_load_parts_refcounts.sh b/tests/queries/0_stateless/02482_load_parts_refcounts.sh index 5303824d97c..4dc7a7fd99b 100755 --- a/tests/queries/0_stateless/02482_load_parts_refcounts.sh +++ b/tests/queries/0_stateless/02482_load_parts_refcounts.sh @@ -5,7 +5,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -n --query " +$CLICKHOUSE_CLIENT --query " DROP TABLE IF EXISTS load_parts_refcounts SYNC; CREATE TABLE load_parts_refcounts (id UInt32) diff --git a/tests/queries/0_stateless/02483_cuturlparameter_with_arrays.sql b/tests/queries/0_stateless/02483_cuturlparameter_with_arrays.sql index 6d64d2685b7..0cf95d10a88 100644 --- a/tests/queries/0_stateless/02483_cuturlparameter_with_arrays.sql +++ b/tests/queries/0_stateless/02483_cuturlparameter_with_arrays.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- { echoOn } diff --git a/tests/queries/0_stateless/02483_elapsed_time.sh b/tests/queries/0_stateless/02483_elapsed_time.sh index fdb23d6da01..274b30cefb9 100755 --- a/tests/queries/0_stateless/02483_elapsed_time.sh +++ b/tests/queries/0_stateless/02483_elapsed_time.sh @@ -17,7 +17,7 @@ EXCEPTION_BEFORE_START_QUERY="WITH FROM system.numbers WHERE number IN (sub) ) - SETTINGS enable_global_with_statement = 0, allow_experimental_analyzer = 1" + SETTINGS enable_global_with_statement = 0, enable_analyzer = 1" # For this query the system.query_log needs to show ExceptionBeforeStart and elapsed seconds <= 1.0 diff --git a/tests/queries/0_stateless/02489_analyzer_indexes.sql b/tests/queries/0_stateless/02489_analyzer_indexes.sql index b5438ddd31a..dcf18016da6 100644 --- a/tests/queries/0_stateless/02489_analyzer_indexes.sql +++ b/tests/queries/0_stateless/02489_analyzer_indexes.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table @@ -59,4 +59,3 @@ WHERE t1.id = 1 AND t1.value_1 = '1' AND t1.value_2 = '1' AND t1.value_3 = '1' SETTINGS force_primary_key = 1, force_data_skipping_indices = 'value_1_idx, value_2_idx, value_3_idx'; DROP TABLE test_table; - diff --git a/tests/queries/0_stateless/02490_benchmark_max_consecutive_errors.sh b/tests/queries/0_stateless/02490_benchmark_max_consecutive_errors.sh index f747b3156a5..df7e9386662 100755 --- a/tests/queries/0_stateless/02490_benchmark_max_consecutive_errors.sh +++ b/tests/queries/0_stateless/02490_benchmark_max_consecutive_errors.sh @@ -11,5 +11,6 @@ if [ "$RES" -eq 10 ] then echo "$RES" else + echo "$RES" cat "${CLICKHOUSE_TMP}/${CLICKHOUSE_DATABASE}.log" fi diff --git a/tests/queries/0_stateless/02493_analyzer_sum_if_to_count_if.sql b/tests/queries/0_stateless/02493_analyzer_sum_if_to_count_if.sql index f1dbfa1f32a..171e080961a 100644 --- a/tests/queries/0_stateless/02493_analyzer_sum_if_to_count_if.sql +++ b/tests/queries/0_stateless/02493_analyzer_sum_if_to_count_if.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET optimize_rewrite_sum_if_to_count_if = 1; EXPLAIN QUERY TREE (SELECT sumIf(1, (number % 2) == 0) FROM numbers(10)); diff --git a/tests/queries/0_stateless/02493_analyzer_table_functions_untuple.sql b/tests/queries/0_stateless/02493_analyzer_table_functions_untuple.sql index bdbe65c643b..c9687783dc9 100644 --- a/tests/queries/0_stateless/02493_analyzer_table_functions_untuple.sql +++ b/tests/queries/0_stateless/02493_analyzer_table_functions_untuple.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT number FROM numbers(untuple(tuple(1))); diff --git a/tests/queries/0_stateless/02493_analyzer_uniq_injective_functions_elimination.sql b/tests/queries/0_stateless/02493_analyzer_uniq_injective_functions_elimination.sql index 5a3b2379fde..ca37c6f3833 100644 --- a/tests/queries/0_stateless/02493_analyzer_uniq_injective_functions_elimination.sql +++ b/tests/queries/0_stateless/02493_analyzer_uniq_injective_functions_elimination.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1, optimize_injective_functions_inside_uniq = 1; +SET enable_analyzer = 1, optimize_injective_functions_inside_uniq = 1; -- Simple test EXPLAIN QUERY TREE SELECT uniqCombined(tuple('')) FROM numbers(1); diff --git a/tests/queries/0_stateless/02494_analyzer_compound_expression_crash_fix.sql b/tests/queries/0_stateless/02494_analyzer_compound_expression_crash_fix.sql index 3e6f9f42724..20b0bdd46b0 100644 --- a/tests/queries/0_stateless/02494_analyzer_compound_expression_crash_fix.sql +++ b/tests/queries/0_stateless/02494_analyzer_compound_expression_crash_fix.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table ( diff --git a/tests/queries/0_stateless/02494_query_cache_explain.sql b/tests/queries/0_stateless/02494_query_cache_explain.sql index bf376b47fdb..decdd92c477 100644 --- a/tests/queries/0_stateless/02494_query_cache_explain.sql +++ b/tests/queries/0_stateless/02494_query_cache_explain.sql @@ -1,7 +1,7 @@ -- Tags: no-parallel -- Tag no-parallel: Messes with internal cache -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET query_cache_system_table_handling = 'save'; SYSTEM DROP QUERY CACHE; diff --git a/tests/queries/0_stateless/02494_query_cache_nested_query_bug.sh b/tests/queries/0_stateless/02494_query_cache_nested_query_bug.sh index 24803ed7420..9a07ddf0386 100755 --- a/tests/queries/0_stateless/02494_query_cache_nested_query_bug.sh +++ b/tests/queries/0_stateless/02494_query_cache_nested_query_bug.sh @@ -15,8 +15,8 @@ ${CLICKHOUSE_CLIENT} --query "CREATE TABLE tab (a UInt64) ENGINE=MergeTree() ORD ${CLICKHOUSE_CLIENT} --query "INSERT INTO tab VALUES (1) (2) (3)" ${CLICKHOUSE_CLIENT} --query "INSERT INTO tab VALUES (3) (4) (5)" -SETTINGS_NO_ANALYZER="SETTINGS use_query_cache=1, max_threads=1, allow_experimental_analyzer=0, merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability=0.0" -SETTINGS_ANALYZER="SETTINGS use_query_cache=1, max_threads=1, allow_experimental_analyzer=1, merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability=0.0" +SETTINGS_NO_ANALYZER="SETTINGS use_query_cache=1, max_threads=1, enable_analyzer=0, merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability=0.0" +SETTINGS_ANALYZER="SETTINGS use_query_cache=1, max_threads=1, enable_analyzer=1, merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability=0.0" # Verify that the first query does two aggregations and the second query zero aggregations. Since query cache is currently not integrated # with EXPLAIN PLAN, we need to check the logs. diff --git a/tests/queries/0_stateless/02494_query_cache_tag.reference b/tests/queries/0_stateless/02494_query_cache_tag.reference new file mode 100644 index 00000000000..f7be5c06ecf --- /dev/null +++ b/tests/queries/0_stateless/02494_query_cache_tag.reference @@ -0,0 +1,12 @@ +1 +SELECT 1 SETTINGS use_query_cache = true, query_cache_tag = \'abc\' abc +--- +1 +1 +SELECT 1 SETTINGS use_query_cache = true +SELECT 1 SETTINGS use_query_cache = true, query_cache_tag = \'abc\' abc +--- +1 +1 +SELECT 1 SETTINGS use_query_cache = true abc +SELECT 1 SETTINGS use_query_cache = true def diff --git a/tests/queries/0_stateless/02494_query_cache_tag.sql b/tests/queries/0_stateless/02494_query_cache_tag.sql new file mode 100644 index 00000000000..62d36f6ebe6 --- /dev/null +++ b/tests/queries/0_stateless/02494_query_cache_tag.sql @@ -0,0 +1,34 @@ +-- Tags: no-parallel +-- Tag no-parallel: Messes with internal cache + +SYSTEM DROP QUERY CACHE; + +-- Store the result a single query with a tag in the query cache and check that the system table knows about the tag +SELECT 1 SETTINGS use_query_cache = true, query_cache_tag = 'abc'; + +SELECT query, tag FROM system.query_cache; + +SELECT '---'; + +SYSTEM DROP QUERY CACHE; + +-- Store the result of the same query with two different tags. The cache should store two entries. +SELECT 1 SETTINGS use_query_cache = true; -- default query_cache_tag = '' +SELECT 1 SETTINGS use_query_cache = true, query_cache_tag = 'abc'; +SELECT query, tag FROM system.query_cache ORDER BY ALL; + +SELECT '---'; + +SYSTEM DROP QUERY CACHE; + +-- Like before but the tag is set standalone. + +SET query_cache_tag = 'abc'; +SELECT 1 SETTINGS use_query_cache = true; + +SET query_cache_tag = 'def'; +SELECT 1 SETTINGS use_query_cache = true; + +SELECT query, tag FROM system.query_cache ORDER BY ALL; + +SYSTEM DROP QUERY CACHE; diff --git a/tests/queries/0_stateless/02495_analyzer_storage_join.sql b/tests/queries/0_stateless/02495_analyzer_storage_join.sql index 7e6c03971f9..1148cd68338 100644 --- a/tests/queries/0_stateless/02495_analyzer_storage_join.sql +++ b/tests/queries/0_stateless/02495_analyzer_storage_join.sql @@ -2,7 +2,7 @@ DROP TABLE IF EXISTS t; DROP TABLE IF EXISTS t1; DROP TABLE IF EXISTS tj; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET single_join_prefer_left_table = 0; CREATE TABLE tj (key2 UInt64, key1 Int64, a UInt64, b UInt64, x UInt64, y UInt64) ENGINE = Join(ALL, RIGHT, key1, key2); diff --git a/tests/queries/0_stateless/02495_sum_if_to_count_if_bug.sql b/tests/queries/0_stateless/02495_sum_if_to_count_if_bug.sql index 0791b374668..c00c0ba4e07 100644 --- a/tests/queries/0_stateless/02495_sum_if_to_count_if_bug.sql +++ b/tests/queries/0_stateless/02495_sum_if_to_count_if_bug.sql @@ -1,4 +1,3 @@ select sum(if((number % NULL) = 2, 0, 1)) FROM numbers(1024) settings optimize_rewrite_sum_if_to_count_if=0; -select sum(if((number % NULL) = 2, 0, 1)) FROM numbers(1024) settings optimize_rewrite_sum_if_to_count_if=1, allow_experimental_analyzer=0; -select sum(if((number % NULL) = 2, 0, 1)) FROM numbers(1024) settings optimize_rewrite_sum_if_to_count_if=1, allow_experimental_analyzer=1; - +select sum(if((number % NULL) = 2, 0, 1)) FROM numbers(1024) settings optimize_rewrite_sum_if_to_count_if=1, enable_analyzer=0; +select sum(if((number % NULL) = 2, 0, 1)) FROM numbers(1024) settings optimize_rewrite_sum_if_to_count_if=1, enable_analyzer=1; diff --git a/tests/queries/0_stateless/02496_remove_redundant_sorting.reference b/tests/queries/0_stateless/02496_remove_redundant_sorting.reference index 77ef213b36d..4d004f2f78f 100644 --- a/tests/queries/0_stateless/02496_remove_redundant_sorting.reference +++ b/tests/queries/0_stateless/02496_remove_redundant_sorting.reference @@ -465,6 +465,37 @@ Expression ((Projection + Before ORDER BY)) ReadFromStorage (SystemOne) -- execute Float64 9007199254740994 +-- presence of an inner OFFSET retains the ORDER BY +-- query +WITH + t1 AS ( + SELECT a, b + FROM + VALUES ( + 'b UInt32, a Int32', + (1, 1), + (2, 0) + ) + ) +SELECT + SUM(a) +FROM ( + SELECT a, b + FROM t1 + ORDER BY 1 DESC, 2 + OFFSET 1 +) t2 +-- explain +Expression ((Projection + Before ORDER BY)) + Aggregating + Expression (Before GROUP BY) + Offset + Expression (Projection) + Sorting (Sorting for ORDER BY) + Expression ((Before ORDER BY + (Projection + Before ORDER BY))) + ReadFromStorage (Values) +-- execute +0 -- disable common optimization to avoid functions to be lifted up (liftUpFunctions optimization), needed for testing with stateful function -- neighbor() as stateful function prevents removing inner ORDER BY since its result depends on order -- query diff --git a/tests/queries/0_stateless/02496_remove_redundant_sorting.sh b/tests/queries/0_stateless/02496_remove_redundant_sorting.sh index 661b32fce72..6e132c55628 100755 --- a/tests/queries/0_stateless/02496_remove_redundant_sorting.sh +++ b/tests/queries/0_stateless/02496_remove_redundant_sorting.sh @@ -8,8 +8,8 @@ if [ -z ${ENABLE_ANALYZER+x} ]; then ENABLE_ANALYZER=0 fi -DISABLE_OPTIMIZATION="SET allow_experimental_analyzer=$ENABLE_ANALYZER;SET query_plan_remove_redundant_sorting=0;SET optimize_duplicate_order_by_and_distinct=0" -ENABLE_OPTIMIZATION="SET allow_experimental_analyzer=$ENABLE_ANALYZER;SET query_plan_remove_redundant_sorting=1;SET optimize_duplicate_order_by_and_distinct=0" +DISABLE_OPTIMIZATION="SET enable_analyzer=$ENABLE_ANALYZER;SET query_plan_remove_redundant_sorting=0;SET optimize_duplicate_order_by_and_distinct=0" +ENABLE_OPTIMIZATION="SET enable_analyzer=$ENABLE_ANALYZER;SET query_plan_remove_redundant_sorting=1;SET optimize_duplicate_order_by_and_distinct=0" echo "-- Disabled query_plan_remove_redundant_sorting" echo "-- ORDER BY clauses in subqueries are untouched" @@ -26,15 +26,15 @@ FROM ORDER BY number DESC ) ORDER BY number ASC" -$CLICKHOUSE_CLIENT -nq "$DISABLE_OPTIMIZATION;EXPLAIN $query" +$CLICKHOUSE_CLIENT -q "$DISABLE_OPTIMIZATION;EXPLAIN $query" function run_query { echo "-- query" echo "$1" echo "-- explain" - $CLICKHOUSE_CLIENT -nq "$ENABLE_OPTIMIZATION;EXPLAIN $1" + $CLICKHOUSE_CLIENT -q "$ENABLE_OPTIMIZATION;EXPLAIN $1" echo "-- execute" - $CLICKHOUSE_CLIENT -nq "$ENABLE_OPTIMIZATION;$1" + $CLICKHOUSE_CLIENT -q "$ENABLE_OPTIMIZATION;$1" } echo "-- Enabled query_plan_remove_redundant_sorting" @@ -302,6 +302,27 @@ FROM )" run_query "$query" +echo "-- presence of an inner OFFSET retains the ORDER BY" +query="WITH + t1 AS ( + SELECT a, b + FROM + VALUES ( + 'b UInt32, a Int32', + (1, 1), + (2, 0) + ) + ) +SELECT + SUM(a) +FROM ( + SELECT a, b + FROM t1 + ORDER BY 1 DESC, 2 + OFFSET 1 +) t2" +run_query "$query" + echo "-- disable common optimization to avoid functions to be lifted up (liftUpFunctions optimization), needed for testing with stateful function" ENABLE_OPTIMIZATION="SET query_plan_enable_optimizations=0;$ENABLE_OPTIMIZATION" echo "-- neighbor() as stateful function prevents removing inner ORDER BY since its result depends on order" diff --git a/tests/queries/0_stateless/02496_remove_redundant_sorting_analyzer.reference b/tests/queries/0_stateless/02496_remove_redundant_sorting_analyzer.reference index b6a2e3182df..dd5ac7bf706 100644 --- a/tests/queries/0_stateless/02496_remove_redundant_sorting_analyzer.reference +++ b/tests/queries/0_stateless/02496_remove_redundant_sorting_analyzer.reference @@ -464,6 +464,36 @@ Expression ((Project names + Projection)) ReadFromStorage (SystemOne) -- execute Float64 9007199254740994 +-- presence of an inner OFFSET retains the ORDER BY +-- query +WITH + t1 AS ( + SELECT a, b + FROM + VALUES ( + 'b UInt32, a Int32', + (1, 1), + (2, 0) + ) + ) +SELECT + SUM(a) +FROM ( + SELECT a, b + FROM t1 + ORDER BY 1 DESC, 2 + OFFSET 1 +) t2 +-- explain +Expression ((Project names + Projection)) + Aggregating + Expression ((Before GROUP BY + (Change column names to column identifiers + Project names))) + Offset + Sorting (Sorting for ORDER BY) + Expression ((Before ORDER BY + (Projection + (Change column names to column identifiers + (Project names + (Projection + Change column names to column identifiers)))))) + ReadFromStorage (Values) +-- execute +0 -- disable common optimization to avoid functions to be lifted up (liftUpFunctions optimization), needed for testing with stateful function -- neighbor() as stateful function prevents removing inner ORDER BY since its result depends on order -- query diff --git a/tests/queries/0_stateless/02497_analyzer_sum_if_count_if_pass_crash_fix.sql b/tests/queries/0_stateless/02497_analyzer_sum_if_count_if_pass_crash_fix.sql index 51522565014..7533a333225 100644 --- a/tests/queries/0_stateless/02497_analyzer_sum_if_count_if_pass_crash_fix.sql +++ b/tests/queries/0_stateless/02497_analyzer_sum_if_count_if_pass_crash_fix.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET optimize_rewrite_sum_if_to_count_if = 1; SELECT sum(if((number % 2) = 0 AS cond_expr, 1 AS one_expr, 0 AS zero_expr) AS if_expr), sum(cond_expr), sum(if_expr), one_expr, zero_expr FROM numbers(100); diff --git a/tests/queries/0_stateless/02497_having_without_actual_aggregation_bug.sql b/tests/queries/0_stateless/02497_having_without_actual_aggregation_bug.sql index b28cbd4861e..e5fd26e484e 100644 --- a/tests/queries/0_stateless/02497_having_without_actual_aggregation_bug.sql +++ b/tests/queries/0_stateless/02497_having_without_actual_aggregation_bug.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; select number from numbers_mt(10) having number >= 9; diff --git a/tests/queries/0_stateless/02497_if_transform_strings_to_enum.sql b/tests/queries/0_stateless/02497_if_transform_strings_to_enum.sql index 131eac390f1..cc2d839fbc1 100644 --- a/tests/queries/0_stateless/02497_if_transform_strings_to_enum.sql +++ b/tests/queries/0_stateless/02497_if_transform_strings_to_enum.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET optimize_if_transform_strings_to_enum = 1; SELECT transform(number, [2, 4, 6], ['google', 'censor.net', 'yahoo'], 'other') FROM system.numbers LIMIT 10; diff --git a/tests/queries/0_stateless/02497_storage_file_reader_selection.sh b/tests/queries/0_stateless/02497_storage_file_reader_selection.sh index aa43e81f131..27243dd47fa 100755 --- a/tests/queries/0_stateless/02497_storage_file_reader_selection.sh +++ b/tests/queries/0_stateless/02497_storage_file_reader_selection.sh @@ -13,6 +13,6 @@ $CLICKHOUSE_LOCAL --storage_file_read_method=mmap --print-profile-events -q "SEL $CLICKHOUSE_LOCAL --storage_file_read_method=pread --print-profile-events -q "SELECT * FROM file($DATA_FILE) FORMAT Null" 2>&1 | grep -F -q "CreatedReadBufferMMap" && echo 'Fail' || echo 0 $CLICKHOUSE_LOCAL --storage_file_read_method=pread --print-profile-events -q "SELECT * FROM file($DATA_FILE) FORMAT Null" 2>&1 | grep -F -q "CreatedReadBufferOrdinary" && echo 1 || echo 'Fail' -$CLICKHOUSE_CLIENT --storage_file_read_method=mmap -nq "SELECT * FROM file('/dev/null', 'LineAsString') FORMAT Null -- { serverError BAD_ARGUMENTS }" +$CLICKHOUSE_CLIENT --storage_file_read_method=mmap -q "SELECT * FROM file('/dev/null', 'LineAsString') FORMAT Null -- { serverError BAD_ARGUMENTS }" rm $DATA_FILE diff --git a/tests/queries/0_stateless/02497_storage_join_right_assert.sql b/tests/queries/0_stateless/02497_storage_join_right_assert.sql index 8f9134e9504..eabaa236379 100644 --- a/tests/queries/0_stateless/02497_storage_join_right_assert.sql +++ b/tests/queries/0_stateless/02497_storage_join_right_assert.sql @@ -7,8 +7,8 @@ CREATE TABLE t2 (key UInt64, a UInt64) ENGINE = Join(ALL, RIGHT, key); INSERT INTO t1 VALUES (1, 1), (2, 2); INSERT INTO t2 VALUES (2, 2), (3, 3); -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT * FROM t1 ALL RIGHT JOIN t2 USING (key) ORDER BY key; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT * FROM t1 ALL RIGHT JOIN t2 USING (key) ORDER BY key; diff --git a/tests/queries/0_stateless/02498_analyzer_aggregate_functions_arithmetic_operations_pass_fix.sql b/tests/queries/0_stateless/02498_analyzer_aggregate_functions_arithmetic_operations_pass_fix.sql index 8491018eb72..76c44f9e0cb 100644 --- a/tests/queries/0_stateless/02498_analyzer_aggregate_functions_arithmetic_operations_pass_fix.sql +++ b/tests/queries/0_stateless/02498_analyzer_aggregate_functions_arithmetic_operations_pass_fix.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET optimize_arithmetic_operations_in_aggregate_functions = 1; DROP TABLE IF EXISTS test_table; diff --git a/tests/queries/0_stateless/02498_analyzer_settings_push_down.sql b/tests/queries/0_stateless/02498_analyzer_settings_push_down.sql index 67623869f0a..472ab358d97 100644 --- a/tests/queries/0_stateless/02498_analyzer_settings_push_down.sql +++ b/tests/queries/0_stateless/02498_analyzer_settings_push_down.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET optimize_functions_to_subcolumns = 0; DROP TABLE IF EXISTS test_table; diff --git a/tests/queries/0_stateless/02498_storage_join_key_positions.sql.j2 b/tests/queries/0_stateless/02498_storage_join_key_positions.sql.j2 index e2dad61a93e..e814b8eaf3f 100644 --- a/tests/queries/0_stateless/02498_storage_join_key_positions.sql.j2 +++ b/tests/queries/0_stateless/02498_storage_join_key_positions.sql.j2 @@ -13,7 +13,7 @@ INSERT INTO tjj VALUES (11, 11, 11, 1000), (21, 21, 21, 2000), (31, 31, 31, 3000 {% for use_analyzer in [0, 1] -%} -SET allow_experimental_analyzer = '{{ use_analyzer }}'; +SET enable_analyzer = '{{ use_analyzer }}'; SELECT '--- using ---'; SELECT * FROM t1 ALL INNER JOIN tj USING (key1, key2, key3) ORDER BY key1; diff --git a/tests/queries/0_stateless/02499_analyzer_aggregate_function_lambda_crash_fix.sql b/tests/queries/0_stateless/02499_analyzer_aggregate_function_lambda_crash_fix.sql index f2698512112..7ac817aecde 100644 --- a/tests/queries/0_stateless/02499_analyzer_aggregate_function_lambda_crash_fix.sql +++ b/tests/queries/0_stateless/02499_analyzer_aggregate_function_lambda_crash_fix.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT count((t, x_0, x_1) -> ((key_2, x_0, x_1) IN (NULL, NULL, '0.3'))) FROM numbers(10); -- { serverError UNSUPPORTED_METHOD } SELECT count((t, x_0, x_1) -> ((key_2, x_0, x_1) IN (NULL, NULL, '0.3'))) OVER (PARTITION BY id) FROM numbers(10); -- { serverError UNSUPPORTED_METHOD } diff --git a/tests/queries/0_stateless/02499_analyzer_set_index.sql b/tests/queries/0_stateless/02499_analyzer_set_index.sql index f90ae61541f..52d96cfcabf 100644 --- a/tests/queries/0_stateless/02499_analyzer_set_index.sql +++ b/tests/queries/0_stateless/02499_analyzer_set_index.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02500_analyzer_storage_view_crash_fix.sql b/tests/queries/0_stateless/02500_analyzer_storage_view_crash_fix.sql index 8f4d14b95cc..f0484a68566 100644 --- a/tests/queries/0_stateless/02500_analyzer_storage_view_crash_fix.sql +++ b/tests/queries/0_stateless/02500_analyzer_storage_view_crash_fix.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02500_remove_redundant_distinct.sh b/tests/queries/0_stateless/02500_remove_redundant_distinct.sh index c4f0994cd13..6fd42fa940a 100755 --- a/tests/queries/0_stateless/02500_remove_redundant_distinct.sh +++ b/tests/queries/0_stateless/02500_remove_redundant_distinct.sh @@ -9,8 +9,8 @@ if [ -z ${ENABLE_ANALYZER+x} ]; then fi OPTIMIZATION_SETTING="query_plan_remove_redundant_distinct" -DISABLE_OPTIMIZATION="set allow_experimental_analyzer=$ENABLE_ANALYZER;SET $OPTIMIZATION_SETTING=0;SET optimize_duplicate_order_by_and_distinct=0" -ENABLE_OPTIMIZATION="set allow_experimental_analyzer=$ENABLE_ANALYZER;SET $OPTIMIZATION_SETTING=1;SET optimize_duplicate_order_by_and_distinct=0" +DISABLE_OPTIMIZATION="set enable_analyzer=$ENABLE_ANALYZER;SET $OPTIMIZATION_SETTING=0;SET optimize_duplicate_order_by_and_distinct=0" +ENABLE_OPTIMIZATION="set enable_analyzer=$ENABLE_ANALYZER;SET $OPTIMIZATION_SETTING=1;SET optimize_duplicate_order_by_and_distinct=0" echo "-- Disabled $OPTIMIZATION_SETTING" query="SELECT DISTINCT * @@ -24,15 +24,15 @@ FROM ) )" -$CLICKHOUSE_CLIENT -nq "$DISABLE_OPTIMIZATION;EXPLAIN $query" +$CLICKHOUSE_CLIENT -q "$DISABLE_OPTIMIZATION;EXPLAIN $query" function run_query { echo "-- query" echo "$1" echo "-- explain" - $CLICKHOUSE_CLIENT -nq "$ENABLE_OPTIMIZATION;EXPLAIN $1" + $CLICKHOUSE_CLIENT -q "$ENABLE_OPTIMIZATION;EXPLAIN $1" echo "-- execute" - $CLICKHOUSE_CLIENT -nq "$ENABLE_OPTIMIZATION;$1" + $CLICKHOUSE_CLIENT -q "$ENABLE_OPTIMIZATION;$1" } echo "-- Enabled $OPTIMIZATION_SETTING" diff --git a/tests/queries/0_stateless/02501_analyzer_expired_context_crash_fix.sql b/tests/queries/0_stateless/02501_analyzer_expired_context_crash_fix.sql index b9ec14501bd..e2c940c829a 100644 --- a/tests/queries/0_stateless/02501_analyzer_expired_context_crash_fix.sql +++ b/tests/queries/0_stateless/02501_analyzer_expired_context_crash_fix.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02502_analyzer_insert_select_crash_fix.sql b/tests/queries/0_stateless/02502_analyzer_insert_select_crash_fix.sql index 4643f65988a..a438276bda5 100644 --- a/tests/queries/0_stateless/02502_analyzer_insert_select_crash_fix.sql +++ b/tests/queries/0_stateless/02502_analyzer_insert_select_crash_fix.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02503_join_switch_alias_fuzz.sql b/tests/queries/0_stateless/02503_join_switch_alias_fuzz.sql index 28d64bf3881..113a8493de6 100644 --- a/tests/queries/0_stateless/02503_join_switch_alias_fuzz.sql +++ b/tests/queries/0_stateless/02503_join_switch_alias_fuzz.sql @@ -1,4 +1,4 @@ SELECT * FROM (SELECT 1 AS id, '' AS test) AS a LEFT JOIN (SELECT test, 1 AS id, NULL AS test) AS b ON b.id = a.id -SETTINGS join_algorithm = 'auto', max_rows_in_join = 1, allow_experimental_analyzer = 1 +SETTINGS join_algorithm = 'auto', max_rows_in_join = 1, enable_analyzer = 1 ; diff --git a/tests/queries/0_stateless/02513_analyzer_duplicate_alias_crash_fix.sql b/tests/queries/0_stateless/02513_analyzer_duplicate_alias_crash_fix.sql index fb50ea2c4ca..e54252b5c5f 100644 --- a/tests/queries/0_stateless/02513_analyzer_duplicate_alias_crash_fix.sql +++ b/tests/queries/0_stateless/02513_analyzer_duplicate_alias_crash_fix.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT toUInt64(NULL) AS x FROM (SELECT 1) HAVING x IN (SELECT NULL FROM (SELECT x IN (SELECT x IN (SELECT 1), x IN (SELECT 1) FROM (SELECT 1 WHERE x IN (SELECT NULL FROM (SELECT NULL)))))); diff --git a/tests/queries/0_stateless/02513_analyzer_sort_msan.sql b/tests/queries/0_stateless/02513_analyzer_sort_msan.sql index e5beccaff2a..b86a15e9ef0 100644 --- a/tests/queries/0_stateless/02513_analyzer_sort_msan.sql +++ b/tests/queries/0_stateless/02513_analyzer_sort_msan.sql @@ -1,6 +1,6 @@ DROP TABLE IF EXISTS products; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; CREATE TABLE products (`price` UInt32) ENGINE = Memory; INSERT INTO products VALUES (1); diff --git a/tests/queries/0_stateless/02514_analyzer_drop_join_on.sql b/tests/queries/0_stateless/02514_analyzer_drop_join_on.sql index 2406be13aa8..df84e2f50b2 100644 --- a/tests/queries/0_stateless/02514_analyzer_drop_join_on.sql +++ b/tests/queries/0_stateless/02514_analyzer_drop_join_on.sql @@ -15,7 +15,7 @@ INSERT INTO c VALUES (1, 1, 'a'), (2, 2, 'b'), (3, 3, 'c'); CREATE TABLE d (k UInt64, d1 UInt64, d2 String) ENGINE = Memory; INSERT INTO d VALUES (1, 1, 'a'), (2, 2, 'b'), (3, 3, 'c'); -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- { echoOn } diff --git a/tests/queries/0_stateless/02514_database_replicated_no_arguments_for_rmt.sh b/tests/queries/0_stateless/02514_database_replicated_no_arguments_for_rmt.sh index a0f228e6af4..c1aa24943c1 100755 --- a/tests/queries/0_stateless/02514_database_replicated_no_arguments_for_rmt.sh +++ b/tests/queries/0_stateless/02514_database_replicated_no_arguments_for_rmt.sh @@ -14,8 +14,8 @@ ${CLICKHOUSE_CLIENT} -q "CREATE USER user_${CLICKHOUSE_DATABASE} settings databa ${CLICKHOUSE_CLIENT} -q "GRANT CREATE TABLE ON ${CLICKHOUSE_DATABASE}_db.* TO user_${CLICKHOUSE_DATABASE}" ${CLICKHOUSE_CLIENT} -q "GRANT TABLE ENGINE ON ReplicatedMergeTree TO user_${CLICKHOUSE_DATABASE}" ${CLICKHOUSE_CLIENT} -q "CREATE DATABASE ${CLICKHOUSE_DATABASE}_db engine = Replicated('/clickhouse/databases/${CLICKHOUSE_TEST_ZOOKEEPER_PREFIX}/${CLICKHOUSE_DATABASE}_db', '{shard}', '{replica}')" -${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --user "user_${CLICKHOUSE_DATABASE}" -n --query "CREATE TABLE ${CLICKHOUSE_DATABASE}_db.tab_rmt_ok (x UInt32) engine = ReplicatedMergeTree order by x;" -${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --user "user_${CLICKHOUSE_DATABASE}" -n --query "CREATE TABLE ${CLICKHOUSE_DATABASE}_db.tab_rmt_fail (x UInt32) engine = ReplicatedMergeTree('/clickhouse/tables/$CLICKHOUSE_TEST_ZOOKEEPER_PREFIX/root/{shard}', '{replica}') order by x; -- { serverError 80 }" +${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --user "user_${CLICKHOUSE_DATABASE}" --query "CREATE TABLE ${CLICKHOUSE_DATABASE}_db.tab_rmt_ok (x UInt32) engine = ReplicatedMergeTree order by x;" +${CLICKHOUSE_CLIENT} --distributed_ddl_output_mode=none --user "user_${CLICKHOUSE_DATABASE}" --query "CREATE TABLE ${CLICKHOUSE_DATABASE}_db.tab_rmt_fail (x UInt32) engine = ReplicatedMergeTree('/clickhouse/tables/$CLICKHOUSE_TEST_ZOOKEEPER_PREFIX/root/{shard}', '{replica}') order by x; -- { serverError 80 }" ${CLICKHOUSE_CLIENT} --query "DROP DATABASE ${CLICKHOUSE_DATABASE}_db" ${CLICKHOUSE_CLIENT} -q "DROP USER user_${CLICKHOUSE_DATABASE}" diff --git a/tests/queries/0_stateless/02515_analyzer_null_for_empty.sql b/tests/queries/0_stateless/02515_analyzer_null_for_empty.sql index de21e9b475e..e12f215743b 100644 --- a/tests/queries/0_stateless/02515_analyzer_null_for_empty.sql +++ b/tests/queries/0_stateless/02515_analyzer_null_for_empty.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET aggregate_functions_null_for_empty = 1; SELECT max(aggr) FROM (SELECT max('92233720368547758.06') AS aggr FROM system.one); diff --git a/tests/queries/0_stateless/02516_join_with_totals_and_subquery_bug.sql.j2 b/tests/queries/0_stateless/02516_join_with_totals_and_subquery_bug.sql.j2 index 09447dfce65..a199165a38f 100644 --- a/tests/queries/0_stateless/02516_join_with_totals_and_subquery_bug.sql.j2 +++ b/tests/queries/0_stateless/02516_join_with_totals_and_subquery_bug.sql.j2 @@ -13,7 +13,7 @@ INNER JOIN GROUP BY 1 WITH TOTALS ) AS t2 USING (a) -SETTINGS allow_experimental_analyzer=0; +SETTINGS enable_analyzer=0; SELECT * FROM @@ -30,7 +30,7 @@ INNER JOIN GROUP BY 1 WITH TOTALS ) AS t2 USING (a) -SETTINGS allow_experimental_analyzer=1; +SETTINGS enable_analyzer=1; SELECT a FROM diff --git a/tests/queries/0_stateless/02516_projections_and_context.sql b/tests/queries/0_stateless/02516_projections_and_context.sql index 334544eb4fa..ec14fc0aa9e 100644 --- a/tests/queries/0_stateless/02516_projections_and_context.sql +++ b/tests/queries/0_stateless/02516_projections_and_context.sql @@ -1,10 +1,10 @@ DROP TABLE IF EXISTS test1__fuzz_37; CREATE TABLE test1__fuzz_37 (`i` Date) ENGINE = MergeTree ORDER BY i; insert into test1__fuzz_37 values ('2020-10-10'); -set allow_experimental_analyzer = 0; +set enable_analyzer = 0; SELECT count() FROM test1__fuzz_37 GROUP BY dictHas(NULL, (dictHas(NULL, (('', materialize(NULL)), materialize(NULL))), 'KeyKey')), dictHas('test_dictionary', tuple(materialize('Ke\0'))), tuple(dictHas(NULL, (tuple('Ke\0Ke\0Ke\0Ke\0Ke\0Ke\0\0\0\0Ke\0'), materialize(NULL)))), 'test_dicti\0nary', (('', materialize(NULL)), dictHas(NULL, (dictHas(NULL, tuple(materialize(NULL))), 'KeyKeyKeyKeyKeyKeyKeyKey')), materialize(NULL)); -- { serverError BAD_ARGUMENTS } SELECT count() FROM test1__fuzz_37 GROUP BY dictHas('non_existing_dictionary', materialize('a')); -- { serverError BAD_ARGUMENTS } -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; SELECT count() FROM test1__fuzz_37 GROUP BY dictHas(NULL, (dictHas(NULL, (('', materialize(NULL)), materialize(NULL))), 'KeyKey')), dictHas('test_dictionary', tuple(materialize('Ke\0'))), tuple(dictHas(NULL, (tuple('Ke\0Ke\0Ke\0Ke\0Ke\0Ke\0\0\0\0Ke\0'), materialize(NULL)))), 'test_dicti\0nary', (('', materialize(NULL)), dictHas(NULL, (dictHas(NULL, tuple(materialize(NULL))), 'KeyKeyKeyKeyKeyKeyKeyKey')), materialize(NULL)); -- { serverError BAD_ARGUMENTS } SELECT count() FROM test1__fuzz_37 GROUP BY dictHas('non_existing_dictionary', materialize('a')); -- { serverError BAD_ARGUMENTS } DROP TABLE test1__fuzz_37; diff --git a/tests/queries/0_stateless/02518_rewrite_aggregate_function_with_if.sql b/tests/queries/0_stateless/02518_rewrite_aggregate_function_with_if.sql index fe882da67cb..4ed13307c29 100644 --- a/tests/queries/0_stateless/02518_rewrite_aggregate_function_with_if.sql +++ b/tests/queries/0_stateless/02518_rewrite_aggregate_function_with_if.sql @@ -1,4 +1,4 @@ -set allow_experimental_analyzer = true; +set enable_analyzer = true; -- { echoOn } set optimize_rewrite_aggregate_function_with_if = false; diff --git a/tests/queries/0_stateless/02521_analyzer_aggregation_without_column.sql b/tests/queries/0_stateless/02521_analyzer_aggregation_without_column.sql index 105bce6711c..50bf3cd45e4 100644 --- a/tests/queries/0_stateless/02521_analyzer_aggregation_without_column.sql +++ b/tests/queries/0_stateless/02521_analyzer_aggregation_without_column.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02521_analyzer_array_join_crash.sql b/tests/queries/0_stateless/02521_analyzer_array_join_crash.sql index 7842d47d757..f5d601303a2 100644 --- a/tests/queries/0_stateless/02521_analyzer_array_join_crash.sql +++ b/tests/queries/0_stateless/02521_analyzer_array_join_crash.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02524_fuzz_and_fuss_2.sql b/tests/queries/0_stateless/02524_fuzz_and_fuss_2.sql index 7b49378d4da..a38fb0bd471 100644 --- a/tests/queries/0_stateless/02524_fuzz_and_fuss_2.sql +++ b/tests/queries/0_stateless/02524_fuzz_and_fuss_2.sql @@ -9,6 +9,6 @@ ENGINE = Memory; INSERT INTO data_a_02187 SELECT * FROM system.one -SETTINGS max_block_size = '1', min_insert_block_size_rows = '65536', min_insert_block_size_bytes = '0', max_insert_threads = '0', max_threads = '3', receive_timeout = '10', receive_data_timeout_ms = '10000', connections_with_failover_max_tries = '0', extremes = '1', use_uncompressed_cache = '0', optimize_move_to_prewhere = '1', optimize_move_to_prewhere_if_final = '0', replication_alter_partitions_sync = '2', totals_mode = 'before_having', allow_suspicious_low_cardinality_types = '1', compile_expressions = '1', min_count_to_compile_expression = '0', group_by_two_level_threshold = '100', distributed_aggregation_memory_efficient = '0', distributed_group_by_no_merge = '1', optimize_distributed_group_by_sharding_key = '1', optimize_skip_unused_shards = '1', optimize_skip_unused_shards_rewrite_in = '1', force_optimize_skip_unused_shards = '2', optimize_skip_unused_shards_nesting = '1', force_optimize_skip_unused_shards_nesting = '2', merge_tree_min_rows_for_concurrent_read = '10000', force_primary_key = '1', network_compression_method = 'ZSTD', network_zstd_compression_level = '7', log_queries = '0', log_queries_min_type = 'QUERY_FINISH', distributed_product_mode = 'local', insert_quorum = '2', insert_quorum_timeout = '0', insert_quorum_parallel = '0', select_sequential_consistency = '1', join_use_nulls = '1', any_join_distinct_right_table_keys = '1', preferred_max_column_in_block_size_bytes = '32', distributed_foreground_insert = '1', insert_allow_materialized_columns = '1', use_index_for_in_with_subqueries = '1', joined_subquery_requires_alias = '0', empty_result_for_aggregation_by_empty_set = '1', allow_suspicious_codecs = '1', query_profiler_real_time_period_ns = '0', query_profiler_cpu_time_period_ns = '0', opentelemetry_start_trace_probability = '1', max_rows_to_read = '1000000', read_overflow_mode = 'break', max_rows_to_group_by = '10', group_by_overflow_mode = 'any', max_rows_to_sort = '100', sort_overflow_mode = 'break', max_result_rows = '10', max_execution_time = '3', max_execution_speed = '1', max_bytes_in_join = '100', join_algorithm = 'partial_merge', max_memory_usage = '1099511627776', log_query_threads = '1', send_logs_level = 'fatal', enable_optimize_predicate_expression = '1', prefer_localhost_replica = '1', optimize_read_in_order = '1', optimize_aggregation_in_order = '1', read_in_order_two_level_merge_threshold = '1', allow_introspection_functions = '1', check_query_single_value_result = '1', allow_experimental_live_view = '1', default_table_engine = 'Memory', mutations_sync = '2', convert_query_to_cnf = '0', optimize_arithmetic_operations_in_aggregate_functions = '1', optimize_duplicate_order_by_and_distinct = '0', optimize_multiif_to_if = '0', optimize_functions_to_subcolumns = '1', optimize_using_constraints = '1', optimize_substitute_columns = '1', optimize_append_index = '1', transform_null_in = '1', data_type_default_nullable = '1', cast_keep_nullable = '1', cast_ipv4_ipv6_default_on_conversion_error = '0', system_events_show_zero_values = '1', enable_global_with_statement = '1', optimize_on_insert = '0', optimize_rewrite_sum_if_to_count_if = '1', distributed_ddl_output_mode = 'throw', union_default_mode = 'ALL', optimize_aggregators_of_group_by_keys = '1', optimize_group_by_function_keys = '1', short_circuit_function_evaluation = 'enable', async_insert = '1', enable_filesystem_cache = '0', allow_deprecated_database_ordinary = '1', allow_deprecated_syntax_for_merge_tree = '1', allow_experimental_nlp_functions = '1', allow_experimental_object_type = '1', optimize_use_projections = '1', input_format_null_as_default = '1', input_format_ipv4_default_on_conversion_error = '0', input_format_ipv6_default_on_conversion_error = '0', output_format_json_named_tuples_as_objects = '1', output_format_write_statistics = '0', output_format_pretty_row_numbers = '1'; +SETTINGS max_block_size = '1', min_insert_block_size_rows = '65536', min_insert_block_size_bytes = '0', max_insert_threads = '0', max_threads = '3', receive_timeout = '10', receive_data_timeout_ms = '10000', connections_with_failover_max_tries = '0', extremes = '1', use_uncompressed_cache = '0', optimize_move_to_prewhere = '1', optimize_move_to_prewhere_if_final = '0', replication_alter_partitions_sync = '2', totals_mode = 'before_having', allow_suspicious_low_cardinality_types = '1', compile_expressions = '1', min_count_to_compile_expression = '0', group_by_two_level_threshold = '100', distributed_aggregation_memory_efficient = '0', distributed_group_by_no_merge = '1', optimize_distributed_group_by_sharding_key = '1', optimize_skip_unused_shards = '1', optimize_skip_unused_shards_rewrite_in = '1', force_optimize_skip_unused_shards = '2', optimize_skip_unused_shards_nesting = '1', force_optimize_skip_unused_shards_nesting = '2', merge_tree_min_rows_for_concurrent_read = '10000', force_primary_key = '1', network_compression_method = 'ZSTD', network_zstd_compression_level = '7', log_queries = '0', log_queries_min_type = 'QUERY_FINISH', distributed_product_mode = 'local', insert_quorum = '2', insert_quorum_timeout = '0', insert_quorum_parallel = '0', select_sequential_consistency = '1', join_use_nulls = '1', any_join_distinct_right_table_keys = '1', preferred_max_column_in_block_size_bytes = '32', distributed_foreground_insert = '1', insert_allow_materialized_columns = '1', use_index_for_in_with_subqueries = '1', joined_subquery_requires_alias = '0', empty_result_for_aggregation_by_empty_set = '1', allow_suspicious_codecs = '1', query_profiler_real_time_period_ns = '0', query_profiler_cpu_time_period_ns = '0', opentelemetry_start_trace_probability = '1', max_rows_to_read = '1000000', read_overflow_mode = 'break', max_rows_to_group_by = '10', group_by_overflow_mode = 'any', max_rows_to_sort = '100', sort_overflow_mode = 'break', max_result_rows = '10', max_execution_time = '9', max_execution_speed = '1', max_bytes_in_join = '100', join_algorithm = 'partial_merge', max_memory_usage = '1099511627776', log_query_threads = '1', send_logs_level = 'fatal', enable_optimize_predicate_expression = '1', prefer_localhost_replica = '1', optimize_read_in_order = '1', optimize_aggregation_in_order = '1', read_in_order_two_level_merge_threshold = '1', allow_introspection_functions = '1', check_query_single_value_result = '1', allow_experimental_live_view = '1', default_table_engine = 'Memory', mutations_sync = '2', convert_query_to_cnf = '0', optimize_arithmetic_operations_in_aggregate_functions = '1', optimize_duplicate_order_by_and_distinct = '0', optimize_multiif_to_if = '0', optimize_functions_to_subcolumns = '1', optimize_using_constraints = '1', optimize_substitute_columns = '1', optimize_append_index = '1', transform_null_in = '1', data_type_default_nullable = '1', cast_keep_nullable = '1', cast_ipv4_ipv6_default_on_conversion_error = '0', system_events_show_zero_values = '1', enable_global_with_statement = '1', optimize_on_insert = '0', optimize_rewrite_sum_if_to_count_if = '1', distributed_ddl_output_mode = 'throw', union_default_mode = 'ALL', optimize_aggregators_of_group_by_keys = '1', optimize_group_by_function_keys = '1', short_circuit_function_evaluation = 'enable', async_insert = '1', enable_filesystem_cache = '0', allow_deprecated_database_ordinary = '1', allow_deprecated_syntax_for_merge_tree = '1', allow_experimental_nlp_functions = '1', allow_experimental_object_type = '1', optimize_use_projections = '1', input_format_null_as_default = '1', input_format_ipv4_default_on_conversion_error = '0', input_format_ipv6_default_on_conversion_error = '0', output_format_json_named_tuples_as_objects = '1', output_format_write_statistics = '0', output_format_pretty_row_numbers = '1'; DROP TABLE data_a_02187; diff --git a/tests/queries/0_stateless/02525_analyzer_function_in_crash_fix.sql b/tests/queries/0_stateless/02525_analyzer_function_in_crash_fix.sql index 95b896d38ab..dd1688ad400 100644 --- a/tests/queries/0_stateless/02525_analyzer_function_in_crash_fix.sql +++ b/tests/queries/0_stateless/02525_analyzer_function_in_crash_fix.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02532_analyzer_aggregation_with_rollup.sql b/tests/queries/0_stateless/02532_analyzer_aggregation_with_rollup.sql index 09097eb029f..587ef71df8a 100644 --- a/tests/queries/0_stateless/02532_analyzer_aggregation_with_rollup.sql +++ b/tests/queries/0_stateless/02532_analyzer_aggregation_with_rollup.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT sum(a.number) AS total, 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 72f4ea06184..e69de29bb2d 100644 --- a/tests/queries/0_stateless/02532_send_logs_level_test.reference +++ b/tests/queries/0_stateless/02532_send_logs_level_test.reference @@ -1,4 +0,0 @@ - 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 506ac2331f2..a91e49ddd22 100755 --- a/tests/queries/0_stateless/02532_send_logs_level_test.sh +++ b/tests/queries/0_stateless/02532_send_logs_level_test.sh @@ -8,7 +8,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " drop table if exists data; create table data (key Int) engine=MergeTree order by tuple() settings min_bytes_for_wide_part = '1G', compress_marks = 1; insert into data values (1); @@ -18,6 +18,10 @@ $CLICKHOUSE_CLIENT -nm -q " # instead of "last" value, hence you cannot simply append another # --send_logs_level here. CLICKHOUSE_CLIENT_CLEAN=$(echo ${CLICKHOUSE_CLIENT} | sed 's/'"--send_logs_level=${CLICKHOUSE_CLIENT_SERVER_LOGS_LEVEL}"'/--send_logs_level=test/g') -$CLICKHOUSE_CLIENT_CLEAN -q "select * from data SETTINGS merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability = 0.0;" |& grep -o -e '.*' -e '.*' -$CLICKHOUSE_CLIENT -q "drop table data" +set -e + +trap '$CLICKHOUSE_CLIENT -q "drop table data"' EXIT + +$CLICKHOUSE_CLIENT_CLEAN -q "select * from data SETTINGS merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability = 0.0;" |& (! grep -q -o -e '.*') +$CLICKHOUSE_CLIENT_CLEAN -q "select * from data SETTINGS merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability = 0.0;" |& grep -q -o -e '.*' diff --git a/tests/queries/0_stateless/02534_analyzer_grouping_function.sql b/tests/queries/0_stateless/02534_analyzer_grouping_function.sql index 3163e03d579..ee1cc1d88d1 100644 --- a/tests/queries/0_stateless/02534_analyzer_grouping_function.sql +++ b/tests/queries/0_stateless/02534_analyzer_grouping_function.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table 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 4ae5df9629a..59bbfc96289 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 @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; -- { echoOn } SELECT number, number % 2, sum(number) AS val diff --git a/tests/queries/0_stateless/02535_analyzer_limit_offset.sql b/tests/queries/0_stateless/02535_analyzer_limit_offset.sql index 8f98d823e5c..96aef9557c5 100644 --- a/tests/queries/0_stateless/02535_analyzer_limit_offset.sql +++ b/tests/queries/0_stateless/02535_analyzer_limit_offset.sql @@ -1,3 +1,3 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT number FROM numbers(100) LIMIT 10 OFFSET 10; diff --git a/tests/queries/0_stateless/02538_analyzer_create_table_as_select.sql b/tests/queries/0_stateless/02538_analyzer_create_table_as_select.sql index 168066ce2f9..16634e996fe 100644 --- a/tests/queries/0_stateless/02538_analyzer_create_table_as_select.sql +++ b/tests/queries/0_stateless/02538_analyzer_create_table_as_select.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table_data; CREATE TABLE test_table_data diff --git a/tests/queries/0_stateless/02540_analyzer_matcher_alias_materialized_columns.sql b/tests/queries/0_stateless/02540_analyzer_matcher_alias_materialized_columns.sql index cc622dde8fe..58840796c7e 100644 --- a/tests/queries/0_stateless/02540_analyzer_matcher_alias_materialized_columns.sql +++ b/tests/queries/0_stateless/02540_analyzer_matcher_alias_materialized_columns.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02540_input_format_json_ignore_unknown_keys_in_named_tuple.sh b/tests/queries/0_stateless/02540_input_format_json_ignore_unknown_keys_in_named_tuple.sh index eccac543215..48b3b27680c 100755 --- a/tests/queries/0_stateless/02540_input_format_json_ignore_unknown_keys_in_named_tuple.sh +++ b/tests/queries/0_stateless/02540_input_format_json_ignore_unknown_keys_in_named_tuple.sh @@ -115,7 +115,7 @@ EOL # NOTE: due to [1] we cannot use dot.dot notation, only tupleElement() # # [1]: https://github.com/ClickHouse/ClickHouse/issues/24607 -$CLICKHOUSE_LOCAL --allow_experimental_analyzer=1 "${gharchive_settings[@]}" --structure="${gharchive_structure[*]}" -q " +$CLICKHOUSE_LOCAL --enable_analyzer=1 "${gharchive_settings[@]}" --structure="${gharchive_structure[*]}" -q " SELECT payload.issue.labels.name AS labels, payload.pull_request.merged_by.login AS merged_by diff --git a/tests/queries/0_stateless/02541_analyzer_grouping_sets_crash_fix.sql b/tests/queries/0_stateless/02541_analyzer_grouping_sets_crash_fix.sql index d7af475bbe7..b9aa251bc96 100644 --- a/tests/queries/0_stateless/02541_analyzer_grouping_sets_crash_fix.sql +++ b/tests/queries/0_stateless/02541_analyzer_grouping_sets_crash_fix.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; WITH pow(NULL, 256) AS four SELECT NULL AS two GROUP BY GROUPING SETS ((pow(two, 65536))); diff --git a/tests/queries/0_stateless/02552_analyzer_optimize_group_by_function_keys_crash.sql b/tests/queries/0_stateless/02552_analyzer_optimize_group_by_function_keys_crash.sql index ee9032472a7..85740cd85a2 100644 --- a/tests/queries/0_stateless/02552_analyzer_optimize_group_by_function_keys_crash.sql +++ b/tests/queries/0_stateless/02552_analyzer_optimize_group_by_function_keys_crash.sql @@ -1,3 +1,3 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT NULL GROUP BY tuple('0.0000000007'), count(NULL) OVER (ROWS BETWEEN CURRENT ROW AND UNBOUNDED FOLLOWING) -- { serverError ILLEGAL_AGGREGATION }; diff --git a/tests/queries/0_stateless/02553_type_object_analyzer.sql b/tests/queries/0_stateless/02553_type_object_analyzer.sql index 55482a02ed1..eb4e49757cf 100644 --- a/tests/queries/0_stateless/02553_type_object_analyzer.sql +++ b/tests/queries/0_stateless/02553_type_object_analyzer.sql @@ -1,6 +1,6 @@ SET output_format_json_named_tuples_as_objects = 1; SET allow_experimental_object_type = 1; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS t_json_analyzer; CREATE TABLE t_json_analyzer (a JSON) ENGINE = Memory; 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 f926b9037d2..0eed4a8c592 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 @@ -60,7 +60,7 @@ FROM (day_)) ) AS t ) -WHERE type_1 = 'all' settings allow_experimental_analyzer=0; +WHERE type_1 = 'all' settings enable_analyzer=0; -- Query plan with analyzer has less Filter steps (which is more optimal) EXPLAIN PIPELINE @@ -83,7 +83,7 @@ FROM (day_)) ) AS t ) -WHERE type_1 = 'all' settings allow_experimental_analyzer=1; +WHERE type_1 = 'all' settings enable_analyzer=1; SELECT ''; SELECT '---Result---'; @@ -129,7 +129,7 @@ FROM (day_)) ) AS t ) -WHERE day_ = '2023-01-05' settings allow_experimental_analyzer=0; +WHERE day_ = '2023-01-05' settings enable_analyzer=0; -- Query plan with analyzer has less Filter steps (which is more optimal) EXPLAIN PIPELINE @@ -151,6 +151,6 @@ FROM (day_)) ) AS t ) -WHERE day_ = '2023-01-05' settings allow_experimental_analyzer=1; +WHERE day_ = '2023-01-05' settings enable_analyzer=1; DROP TABLE test_grouping_sets_predicate; diff --git a/tests/queries/0_stateless/02555_davengers_rename_chain.sh b/tests/queries/0_stateless/02555_davengers_rename_chain.sh index 660a95846c4..196507dc72e 100755 --- a/tests/queries/0_stateless/02555_davengers_rename_chain.sh +++ b/tests/queries/0_stateless/02555_davengers_rename_chain.sh @@ -46,7 +46,7 @@ tables["wrong_metadata_compact"]="min_bytes_for_wide_part = 10000000" for table in "${!tables[@]}"; do settings="${tables[$table]}" - $CLICKHOUSE_CLIENT -n --query=" + $CLICKHOUSE_CLIENT --query=" DROP TABLE IF EXISTS $table; CREATE TABLE $table( @@ -69,7 +69,7 @@ for table in "${!tables[@]}"; do wait_column "$table" "\`a1\` UInt64" || exit 2 - $CLICKHOUSE_CLIENT -n --query=" + $CLICKHOUSE_CLIENT --query=" -- { echoOn } SELECT 'ECHO_ALIGNMENT_FIX' FORMAT Null; @@ -82,7 +82,7 @@ for table in "${!tables[@]}"; do wait_mutation_loaded "$table" "b1 TO a" || exit 2 - $CLICKHOUSE_CLIENT -n --query=" + $CLICKHOUSE_CLIENT --query=" -- { echoOn } SELECT 'ECHO_ALIGNMENT_FIX' FORMAT Null; @@ -94,7 +94,7 @@ for table in "${!tables[@]}"; do wait_for_all_mutations "$table" - $CLICKHOUSE_CLIENT -n --query=" + $CLICKHOUSE_CLIENT --query=" -- { echoOn } SELECT 'ECHO_ALIGNMENT_FIX' FORMAT Null; diff --git a/tests/queries/0_stateless/02558_system_processes_elapsed.sh b/tests/queries/0_stateless/02558_system_processes_elapsed.sh index 891ac3cf7bc..8d2615541fd 100755 --- a/tests/queries/0_stateless/02558_system_processes_elapsed.sh +++ b/tests/queries/0_stateless/02558_system_processes_elapsed.sh @@ -9,7 +9,12 @@ while :; do pid=$! sleep 1.5 duration="$($CLICKHOUSE_CLIENT -q "select floor(elapsed) from system.processes where current_database = currentDatabase() and query not like '%system.processes%'")" - kill -INT $pid + # The process might not exist at this point in some exception situations + # maybe it was killed by OOM? + # It safe to skip this iteration. + if ! kill -INT $pid > /dev/null 2>&1; then + continue + fi wait $CLICKHOUSE_CLIENT -q "kill query where current_database = currentDatabase() sync format Null" if [[ $duration -eq 1 ]]; then diff --git a/tests/queries/0_stateless/02560_analyzer_materialized_view.sql b/tests/queries/0_stateless/02560_analyzer_materialized_view.sql index 1f268fe1e16..3fdef366dc9 100644 --- a/tests/queries/0_stateless/02560_analyzer_materialized_view.sql +++ b/tests/queries/0_stateless/02560_analyzer_materialized_view.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/02563_analyzer_merge.sql b/tests/queries/0_stateless/02563_analyzer_merge.sql index 217fb7019c4..6c252c22773 100644 --- a/tests/queries/0_stateless/02563_analyzer_merge.sql +++ b/tests/queries/0_stateless/02563_analyzer_merge.sql @@ -1,6 +1,6 @@ -- Tags: no-parallel -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP DATABASE IF EXISTS 02563_db; CREATE DATABASE 02563_db; diff --git a/tests/queries/0_stateless/02564_analyzer_cross_to_inner.sql b/tests/queries/0_stateless/02564_analyzer_cross_to_inner.sql index a83cd238982..7032559e066 100644 --- a/tests/queries/0_stateless/02564_analyzer_cross_to_inner.sql +++ b/tests/queries/0_stateless/02564_analyzer_cross_to_inner.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS t1; DROP TABLE IF EXISTS t2; diff --git a/tests/queries/0_stateless/02565_analyzer_limit_settings.sql b/tests/queries/0_stateless/02565_analyzer_limit_settings.sql index 7c02c2d0d20..1dd6735e64d 100644 --- a/tests/queries/0_stateless/02565_analyzer_limit_settings.sql +++ b/tests/queries/0_stateless/02565_analyzer_limit_settings.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- { echoOn } SET limit = 0; diff --git a/tests/queries/0_stateless/02566_analyzer_limit_settings_distributed.sql b/tests/queries/0_stateless/02566_analyzer_limit_settings_distributed.sql index 1624344b5a9..a2620f436f4 100644 --- a/tests/queries/0_stateless/02566_analyzer_limit_settings_distributed.sql +++ b/tests/queries/0_stateless/02566_analyzer_limit_settings_distributed.sql @@ -1,6 +1,6 @@ -- Tags: distributed -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT 'limit', * FROM remote('127.1', view(SELECT * FROM numbers(10))) SETTINGS limit=5; SELECT 'offset', * FROM remote('127.1', view(SELECT * FROM numbers(10))) SETTINGS offset=5; diff --git a/tests/queries/0_stateless/02567_and_consistency.reference b/tests/queries/0_stateless/02567_and_consistency.reference index e0014f187a8..7e6d1f24f43 100644 --- a/tests/queries/0_stateless/02567_and_consistency.reference +++ b/tests/queries/0_stateless/02567_and_consistency.reference @@ -8,7 +8,7 @@ true ===== 1 ===== -allow_experimental_analyzer +enable_analyzer true #45440 2086579505 0 1 0 0 diff --git a/tests/queries/0_stateless/02567_and_consistency.sql b/tests/queries/0_stateless/02567_and_consistency.sql index 0eeab99e539..0442a6dad7f 100644 --- a/tests/queries/0_stateless/02567_and_consistency.sql +++ b/tests/queries/0_stateless/02567_and_consistency.sql @@ -50,9 +50,9 @@ SELECT 1 and sin(1); SELECT '====='; -SELECT 'allow_experimental_analyzer'; +SELECT 'enable_analyzer'; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT toBool(sin(SUM(number))) AS x FROM diff --git a/tests/queries/0_stateless/02572_query_views_log_background_thread.sh b/tests/queries/0_stateless/02572_query_views_log_background_thread.sh index a3e428e75c8..22b94e09b58 100755 --- a/tests/queries/0_stateless/02572_query_views_log_background_thread.sh +++ b/tests/queries/0_stateless/02572_query_views_log_background_thread.sh @@ -8,18 +8,21 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} --ignore-error --multiquery --query "drop table if exists buffer_02572; +${CLICKHOUSE_CLIENT} --ignore-error --query "drop table if exists buffer_02572; drop table if exists data_02572; drop table if exists copy_02572; drop table if exists mv_02572;" ${CLICKHOUSE_CLIENT} --query="create table copy_02572 (key Int) engine=Memory();" ${CLICKHOUSE_CLIENT} --query="create table data_02572 (key Int) engine=Memory();" -${CLICKHOUSE_CLIENT} --query="create table buffer_02572 (key Int) engine=Buffer(currentDatabase(), data_02572, 1, 3, 3, 1, 1e9, 1, 1e9);" +${CLICKHOUSE_CLIENT} --query="create table buffer_02572 (key Int) engine=Buffer(currentDatabase(), data_02572, 1, 8, 8, 1, 1e9, 1, 1e9);" ${CLICKHOUSE_CLIENT} --query="create materialized view mv_02572 to copy_02572 as select * from data_02572;" +start=$(date +%s) ${CLICKHOUSE_CLIENT} --query="insert into buffer_02572 values (1);" -# ensure that the flush was not direct -${CLICKHOUSE_CLIENT} --ignore-error --multiquery --query "select * from data_02572; select * from copy_02572;" +if [ $(( $(date +%s) - start )) -gt 6 ]; then # clickhouse test cluster is overloaded, will skip + # ensure that the flush was not direct + ${CLICKHOUSE_CLIENT} --ignore-error --query "select * from data_02572; select * from copy_02572;" +fi # we cannot use OPTIMIZE, this will attach query context, so let's wait for _ in {1..100}; do @@ -28,11 +31,11 @@ for _ in {1..100}; do done -${CLICKHOUSE_CLIENT} --ignore-error --multiquery --query "select * from data_02572; select * from copy_02572;" +${CLICKHOUSE_CLIENT} --ignore-error --query "select * from data_02572; select * from copy_02572;" ${CLICKHOUSE_CLIENT} --query="system flush logs;" ${CLICKHOUSE_CLIENT} --query="select count() > 0, lower(status::String), errorCodeToName(exception_code) from system.query_views_log where view_name = concatWithSeparator('.', currentDatabase(), 'mv_02572') and view_target = concatWithSeparator('.', currentDatabase(), 'copy_02572') - group by 2, 3;" \ No newline at end of file + group by 2, 3;" diff --git a/tests/queries/0_stateless/02576_predicate_push_down_sorting_fix.sql b/tests/queries/0_stateless/02576_predicate_push_down_sorting_fix.sql index 2dade7837b7..486a26613f6 100644 --- a/tests/queries/0_stateless/02576_predicate_push_down_sorting_fix.sql +++ b/tests/queries/0_stateless/02576_predicate_push_down_sorting_fix.sql @@ -1,3 +1,3 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; EXPLAIN header = 1, actions = 1 SELECT number FROM (SELECT number FROM numbers(2) ORDER BY ignore(2)) WHERE ignore(2); diff --git a/tests/queries/0_stateless/02576_rewrite_array_exists_to_has.sql b/tests/queries/0_stateless/02576_rewrite_array_exists_to_has.sql index 5233f2f7e3c..b5a123e3767 100644 --- a/tests/queries/0_stateless/02576_rewrite_array_exists_to_has.sql +++ b/tests/queries/0_stateless/02576_rewrite_array_exists_to_has.sql @@ -1,4 +1,4 @@ -set allow_experimental_analyzer = true; +set enable_analyzer = true; set optimize_rewrite_array_exists_to_has = false; EXPLAIN QUERY TREE run_passes = 1 select arrayExists(x -> x = 5 , materialize(range(10))) from numbers(10); @@ -8,7 +8,7 @@ set optimize_rewrite_array_exists_to_has = true; EXPLAIN QUERY TREE run_passes = 1 select arrayExists(x -> x = 5 , materialize(range(10))) from numbers(10); EXPLAIN QUERY TREE run_passes = 1 select arrayExists(x -> 5 = x , materialize(range(10))) from numbers(10); -set allow_experimental_analyzer = false; +set enable_analyzer = false; set optimize_rewrite_array_exists_to_has = false; EXPLAIN SYNTAX select arrayExists(x -> x = 5 , materialize(range(10))) from numbers(10); diff --git a/tests/queries/0_stateless/02577_analyzer_array_join_calc_twice.sql b/tests/queries/0_stateless/02577_analyzer_array_join_calc_twice.sql index b6bb258db28..0b281dd4f81 100644 --- a/tests/queries/0_stateless/02577_analyzer_array_join_calc_twice.sql +++ b/tests/queries/0_stateless/02577_analyzer_array_join_calc_twice.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT 1 + arrayJoin(a) AS m FROM (SELECT [1, 2, 3] AS a) GROUP BY m; diff --git a/tests/queries/0_stateless/02579_fill_empty_chunk.sql b/tests/queries/0_stateless/02579_fill_empty_chunk.sql index 30942b154c9..aeae98df7a3 100644 --- a/tests/queries/0_stateless/02579_fill_empty_chunk.sql +++ b/tests/queries/0_stateless/02579_fill_empty_chunk.sql @@ -1,7 +1,7 @@ -- this SELECT produces empty chunk in FillingTransform SET enable_positional_arguments = 0; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT 2 AS x, diff --git a/tests/queries/0_stateless/02579_fill_empty_chunk_analyzer.sql b/tests/queries/0_stateless/02579_fill_empty_chunk_analyzer.sql index 8350173f443..144640149ea 100644 --- a/tests/queries/0_stateless/02579_fill_empty_chunk_analyzer.sql +++ b/tests/queries/0_stateless/02579_fill_empty_chunk_analyzer.sql @@ -1,7 +1,7 @@ -- this SELECT produces empty chunk in FillingTransform SET enable_positional_arguments = 0; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- With analyzer this special query has correct output SELECT diff --git a/tests/queries/0_stateless/02581_share_big_sets_between_mutation_tasks.sql b/tests/queries/0_stateless/02581_share_big_sets_between_mutation_tasks.sql index ea1452fc372..ad8bef7fbb7 100644 --- a/tests/queries/0_stateless/02581_share_big_sets_between_mutation_tasks.sql +++ b/tests/queries/0_stateless/02581_share_big_sets_between_mutation_tasks.sql @@ -1,4 +1,5 @@ --- Tags: no-tsan, no-asan, no-ubsan, no-msan +-- Tags: no-tsan, no-asan, no-ubsan, no-msan, no-fasttest +-- no-fasttest: Slow test -- no sanitizers: too slow sometimes DROP TABLE IF EXISTS 02581_trips; diff --git a/tests/queries/0_stateless/02582_analyzer_join_subquery_empty_column_list.sql b/tests/queries/0_stateless/02582_analyzer_join_subquery_empty_column_list.sql index 10e5871cc44..33c9296a0d8 100644 --- a/tests/queries/0_stateless/02582_analyzer_join_subquery_empty_column_list.sql +++ b/tests/queries/0_stateless/02582_analyzer_join_subquery_empty_column_list.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- { echoOn } SELECT a FROM ( select 1 AS a ) AS t1, ( select 2 AS b, 3 AS c) AS t2; @@ -10,5 +10,3 @@ SELECT b FROM ( select 1 AS a UNION ALL select 1 as a ) AS t1, ( select 2 AS b, SELECT c FROM ( select 1 AS a UNION ALL select 1 as a ) AS t1, ( select 2 AS b, 3 AS c UNION ALL select 2 as b, 3 as c) AS t2; SELECT 42 FROM ( select 1 AS a UNION ALL select 1 as a ) AS t1, ( select 2 AS b, 3 AS c UNION ALL select 2 as b, 3 as c) AS t2; SELECT count() FROM ( select 1 AS a UNION ALL select 1 as a ) AS t1, ( select 2 AS b, 3 AS c UNION ALL select 2 as b, 3 as c) AS t2; - - diff --git a/tests/queries/0_stateless/02597_column_update_tricky_expression_and_replication.python b/tests/queries/0_stateless/02597_column_update_tricky_expression_and_replication.python new file mode 100644 index 00000000000..eb0cab9d56f --- /dev/null +++ b/tests/queries/0_stateless/02597_column_update_tricky_expression_and_replication.python @@ -0,0 +1,63 @@ +#!/usr/bin/env python3 + +import os +import sys +from threading import Thread +from queue import Queue + +CURDIR = os.path.dirname(os.path.realpath(__file__)) +sys.path.insert(0, os.path.join(CURDIR, "helpers")) + +from pure_http_client import ClickHouseClient + + +client = ClickHouseClient() + + +client.query("DROP TABLE IF EXISTS test SYNC") +client.query( + """ +CREATE TABLE test +( + c_id String, + p_id String, + d UInt32, +) +Engine = ReplicatedMergeTree('/clickhouse/tables/{database}/test/test_table', '1') +ORDER BY (c_id, p_id) +""" +) + + +def attempt_mutation(q): + try: + client.query( + "ALTER TABLE test DROP COLUMN x SETTINGS mutations_sync=2", + with_retries=False, + ) + except ValueError as e: + assert "BAD_ARGUMENTS" in str(e) + q.put("OK") + + +client.query("INSERT INTO test SELECT '1', '11', '111' FROM numbers(5)") +client.query("SYSTEM ENABLE FAILPOINT infinite_sleep") +client.query( + "ALTER TABLE test UPDATE d = d + sleepEachRow(0.3) where 1 SETTINGS mutations_sync=0" +) +client.query("ALTER TABLE test ADD COLUMN x UInt32 default 0 SETTINGS mutations_sync=0") +client.query("ALTER TABLE test UPDATE d = x + 1 where 1 SETTINGS mutations_sync=0") + +q = Queue() +t = Thread(target=attempt_mutation, args=(q,)) +t.start() +t.join() +assert not q.empty() +assert q.get() == "OK" + +client.query("SYSTEM DISABLE FAILPOINT infinite_sleep") + +client.query("ALTER TABLE test UPDATE x = x + 1 where 1 SETTINGS mutations_sync=2") +client.query("ALTER TABLE test DROP COLUMN x SETTINGS mutations_sync=2") +client.query("SELECT * from test format Null") +client.query("DROP TABLE test") diff --git a/tests/queries/0_stateless/02597_column_update_tricky_expression_and_replication.reference b/tests/queries/0_stateless/02597_column_update_tricky_expression_and_replication.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/02597_column_update_tricky_expression_and_replication.sh b/tests/queries/0_stateless/02597_column_update_tricky_expression_and_replication.sh new file mode 100755 index 00000000000..5be04d99204 --- /dev/null +++ b/tests/queries/0_stateless/02597_column_update_tricky_expression_and_replication.sh @@ -0,0 +1,9 @@ +#!/usr/bin/env bash +# Tags: zookeeper, no-parallel + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +# We should have correct env vars from shell_config.sh to run this test +python3 "$CURDIR"/02597_column_update_tricky_expression_and_replication.python diff --git a/tests/queries/0_stateless/02597_column_update_tricy_expression_and_replication.sql b/tests/queries/0_stateless/02597_column_update_tricy_expression_and_replication.sql deleted file mode 100644 index 34f88b19b7e..00000000000 --- a/tests/queries/0_stateless/02597_column_update_tricy_expression_and_replication.sql +++ /dev/null @@ -1,28 +0,0 @@ -CREATE TABLE test ( - `c_id` String, - `p_id` String, - `d` UInt32 -) -ENGINE = ReplicatedMergeTree('/clickhouse/tables/{database}/test/test_table', '1') -ORDER BY (c_id, p_id); - -INSERT INTO test SELECT '1', '11', '111' FROM numbers(3); - -INSERT INTO test SELECT '2', '22', '22' FROM numbers(3); - -set mutations_sync=0; - -ALTER TABLE test UPDATE d = d + sleepEachRow(0.3) where 1; - -ALTER TABLE test ADD COLUMN x UInt32 default 0; -ALTER TABLE test UPDATE d = x + 1 where 1; -ALTER TABLE test DROP COLUMN x SETTINGS mutations_sync = 2; --{serverError BAD_ARGUMENTS} - -ALTER TABLE test UPDATE x = x + 1 where 1 SETTINGS mutations_sync = 2; - -ALTER TABLE test DROP COLUMN x SETTINGS mutations_sync = 2; - -select * from test format Null; - -DROP TABLE test; - diff --git a/tests/queries/0_stateless/02661_quantile_approx.reference b/tests/queries/0_stateless/02661_quantile_approx.reference index 0ee846a268b..239516f9ac4 100644 --- a/tests/queries/0_stateless/02661_quantile_approx.reference +++ b/tests/queries/0_stateless/02661_quantile_approx.reference @@ -33,10 +33,10 @@ FROM FROM numbers(49999) ); [24902,44518,49999] -select medianGK()(number) from numbers(10) SETTINGS allow_experimental_analyzer = 0; -- { serverError BAD_ARGUMENTS } -select medianGK()(number) from numbers(10) SETTINGS allow_experimental_analyzer = 1; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } -select quantileGK()(number) from numbers(10) SETTINGS allow_experimental_analyzer = 0; -- { serverError BAD_ARGUMENTS } -select quantileGK()(number) from numbers(10) SETTINGS allow_experimental_analyzer = 1; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +select medianGK()(number) from numbers(10) SETTINGS enable_analyzer = 0; -- { serverError BAD_ARGUMENTS } +select medianGK()(number) from numbers(10) SETTINGS enable_analyzer = 1; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +select quantileGK()(number) from numbers(10) SETTINGS enable_analyzer = 0; -- { serverError BAD_ARGUMENTS } +select quantileGK()(number) from numbers(10) SETTINGS enable_analyzer = 1; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } select medianGK(100)(number) from numbers(10); 4 select quantileGK(100)(number) from numbers(10); @@ -47,8 +47,8 @@ select quantileGK(100, 0.5, 0.75)(number) from numbers(10); -- { serverError NUM select quantileGK('abc', 0.5)(number) from numbers(10); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } select quantileGK(1.23, 0.5)(number) from numbers(10); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } select quantileGK(-100, 0.5)(number) from numbers(10); -- { serverError BAD_ARGUMENTS } -select quantilesGK()(number) from numbers(10) SETTINGS allow_experimental_analyzer = 0; -- { serverError BAD_ARGUMENTS } -select quantilesGK()(number) from numbers(10) SETTINGS allow_experimental_analyzer = 1; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +select quantilesGK()(number) from numbers(10) SETTINGS enable_analyzer = 0; -- { serverError BAD_ARGUMENTS } +select quantilesGK()(number) from numbers(10) SETTINGS enable_analyzer = 1; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } select quantilesGK(100)(number) from numbers(10); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } select quantilesGK(100, 0.5)(number) from numbers(10); [4] diff --git a/tests/queries/0_stateless/02661_quantile_approx.sql b/tests/queries/0_stateless/02661_quantile_approx.sql index c0004260fa1..732ce645c98 100644 --- a/tests/queries/0_stateless/02661_quantile_approx.sql +++ b/tests/queries/0_stateless/02661_quantile_approx.sql @@ -1,4 +1,4 @@ -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; -- { echoOn } with arrayJoin([0, 1, 2, 10]) as x select quantilesGK(100, 0.5, 0.4, 0.1)(x); @@ -29,11 +29,11 @@ FROM FROM numbers(49999) ); -select medianGK()(number) from numbers(10) SETTINGS allow_experimental_analyzer = 0; -- { serverError BAD_ARGUMENTS } -select medianGK()(number) from numbers(10) SETTINGS allow_experimental_analyzer = 1; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +select medianGK()(number) from numbers(10) SETTINGS enable_analyzer = 0; -- { serverError BAD_ARGUMENTS } +select medianGK()(number) from numbers(10) SETTINGS enable_analyzer = 1; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } -select quantileGK()(number) from numbers(10) SETTINGS allow_experimental_analyzer = 0; -- { serverError BAD_ARGUMENTS } -select quantileGK()(number) from numbers(10) SETTINGS allow_experimental_analyzer = 1; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +select quantileGK()(number) from numbers(10) SETTINGS enable_analyzer = 0; -- { serverError BAD_ARGUMENTS } +select quantileGK()(number) from numbers(10) SETTINGS enable_analyzer = 1; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } select medianGK(100)(number) from numbers(10); select quantileGK(100)(number) from numbers(10); @@ -43,8 +43,8 @@ select quantileGK('abc', 0.5)(number) from numbers(10); -- { serverError ILLEGAL select quantileGK(1.23, 0.5)(number) from numbers(10); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } select quantileGK(-100, 0.5)(number) from numbers(10); -- { serverError BAD_ARGUMENTS } -select quantilesGK()(number) from numbers(10) SETTINGS allow_experimental_analyzer = 0; -- { serverError BAD_ARGUMENTS } -select quantilesGK()(number) from numbers(10) SETTINGS allow_experimental_analyzer = 1; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } +select quantilesGK()(number) from numbers(10) SETTINGS enable_analyzer = 0; -- { serverError BAD_ARGUMENTS } +select quantilesGK()(number) from numbers(10) SETTINGS enable_analyzer = 1; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } select quantilesGK(100)(number) from numbers(10); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } select quantilesGK(100, 0.5)(number) from numbers(10); diff --git a/tests/queries/0_stateless/02662_first_last_value.reference b/tests/queries/0_stateless/02662_first_last_value.reference index b0783399623..308cbf850b0 100644 --- a/tests/queries/0_stateless/02662_first_last_value.reference +++ b/tests/queries/0_stateless/02662_first_last_value.reference @@ -18,7 +18,7 @@ select last_value(b) ignore nulls from test; 5 select last_value(b) respect nulls from test; \N -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- first value select first_value(b) from test; 3 diff --git a/tests/queries/0_stateless/02662_first_last_value.sql b/tests/queries/0_stateless/02662_first_last_value.sql index 8e429e2e27d..16768bd6f1e 100644 --- a/tests/queries/0_stateless/02662_first_last_value.sql +++ b/tests/queries/0_stateless/02662_first_last_value.sql @@ -15,7 +15,7 @@ select last_value(b) from test; select last_value(b) ignore nulls from test; select last_value(b) respect nulls from test; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- first value select first_value(b) from test; diff --git a/tests/queries/0_stateless/02668_logical_optimizer_removing_redundant_checks.sql b/tests/queries/0_stateless/02668_logical_optimizer_removing_redundant_checks.sql index 7d624195df9..dabdcfd5507 100644 --- a/tests/queries/0_stateless/02668_logical_optimizer_removing_redundant_checks.sql +++ b/tests/queries/0_stateless/02668_logical_optimizer_removing_redundant_checks.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS 02668_logical_optimizer; diff --git a/tests/queries/0_stateless/02674_trivial_count_analyzer.reference b/tests/queries/0_stateless/02674_trivial_count_analyzer.reference index 05feadb58a0..9d7f393e71c 100644 --- a/tests/queries/0_stateless/02674_trivial_count_analyzer.reference +++ b/tests/queries/0_stateless/02674_trivial_count_analyzer.reference @@ -1,5 +1,5 @@ -- { echoOn } -set allow_experimental_analyzer=1; +set enable_analyzer=1; set optimize_trivial_count_query=1; create table m3(a Int64, b UInt64) Engine=MergeTree order by tuple(); select count() from m3; diff --git a/tests/queries/0_stateless/02674_trivial_count_analyzer.sql b/tests/queries/0_stateless/02674_trivial_count_analyzer.sql index 988d1b9ba92..c13a9dc68f2 100644 --- a/tests/queries/0_stateless/02674_trivial_count_analyzer.sql +++ b/tests/queries/0_stateless/02674_trivial_count_analyzer.sql @@ -2,7 +2,7 @@ drop table if exists m3; drop table if exists replacing_m3; -- { echoOn } -set allow_experimental_analyzer=1; +set enable_analyzer=1; set optimize_trivial_count_query=1; create table m3(a Int64, b UInt64) Engine=MergeTree order by tuple(); diff --git a/tests/queries/0_stateless/02675_predicate_push_down_filled_join_fix.sql b/tests/queries/0_stateless/02675_predicate_push_down_filled_join_fix.sql index 930127497ae..6c8932b5d58 100644 --- a/tests/queries/0_stateless/02675_predicate_push_down_filled_join_fix.sql +++ b/tests/queries/0_stateless/02675_predicate_push_down_filled_join_fix.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET single_join_prefer_left_table = 0; SET optimize_move_to_prewhere = 0; diff --git a/tests/queries/0_stateless/02675_profile_events_from_query_log_and_client.reference b/tests/queries/0_stateless/02675_profile_events_from_query_log_and_client.reference index babcecf7004..448eca3e5b1 100644 --- a/tests/queries/0_stateless/02675_profile_events_from_query_log_and_client.reference +++ b/tests/queries/0_stateless/02675_profile_events_from_query_log_and_client.reference @@ -1,11 +1,11 @@ INSERT TO S3 - [ 0 ] S3Clients: 1 - [ 0 ] S3CompleteMultipartUpload: 1 - [ 0 ] S3CreateMultipartUpload: 1 - [ 0 ] S3HeadObject: 2 - [ 0 ] S3ReadRequestsCount: 2 - [ 0 ] S3UploadPart: 1 - [ 0 ] S3WriteRequestsCount: 3 +S3Clients 1 +S3CompleteMultipartUpload 1 +S3CreateMultipartUpload 1 +S3HeadObject 2 +S3ReadRequestsCount 2 +S3UploadPart 1 +Successful write requests 3 CHECK WITH query_log QueryFinish S3CreateMultipartUpload 1 S3UploadPart 1 S3CompleteMultipartUpload 1 S3PutObject 0 CREATE diff --git a/tests/queries/0_stateless/02675_profile_events_from_query_log_and_client.sh b/tests/queries/0_stateless/02675_profile_events_from_query_log_and_client.sh index e346d9893a7..6d770b308b5 100755 --- a/tests/queries/0_stateless/02675_profile_events_from_query_log_and_client.sh +++ b/tests/queries/0_stateless/02675_profile_events_from_query_log_and_client.sh @@ -9,7 +9,23 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) echo "INSERT TO S3" $CLICKHOUSE_CLIENT --print-profile-events --profile-events-delay-ms=-1 -nq " INSERT INTO TABLE FUNCTION s3('http://localhost:11111/test/profile_events.csv', 'test', 'testtest', 'CSV', 'number UInt64') SELECT number FROM numbers(1000000) SETTINGS s3_max_single_part_upload_size = 10, s3_truncate_on_insert = 1; -" 2>&1 | grep -o -e '\ \[\ .*\ \]\ S3.*:\ .*\ ' | grep -v 'Microseconds' | grep -v 'S3DiskConnections' | grep -v 'S3DiskAddresses' | sort +" 2>&1 | $CLICKHOUSE_LOCAL -q " +WITH '(\\w+): (\\d+)' AS pattern, + (SELECT (groupArray(regexpExtract(line, pattern, 1)), + groupArray(regexpExtract(line, pattern, 2)::UInt64))::Map(String, UInt64) + FROM file(stdin, 'LineAsString', 'line String') + WHERE line LIKE '% S3%' + AND line NOT LIKE '%Microseconds%' + AND line NOT LIKE '%S3DiskConnections%' + AND line NOT LIKE '%S3DiskAddresses') AS pe_map +SELECT * FROM ( + SELECT untuple(arrayJoin(pe_map) AS pe) + WHERE tupleElement(pe, 1) not like '%WriteRequests%' + UNION ALL + SELECT 'Successful write requests', + (pe_map['S3WriteRequestsCount'] - pe_map['S3WriteRequestsErrors'])::UInt64 +) ORDER BY 1 +" echo "CHECK WITH query_log" $CLICKHOUSE_CLIENT -nq " @@ -40,19 +56,19 @@ CREATE TABLE times (t DateTime) ENGINE MergeTree ORDER BY t echo "INSERT" $CLICKHOUSE_CLIENT --print-profile-events --profile-events-delay-ms=-1 -nq " INSERT INTO times SELECT now() + INTERVAL 1 day SETTINGS optimize_on_insert = 0; -" 2>&1 | grep -o -e '\ \[\ .*\ \]\ FileOpen:\ .*\ ' +" 2>&1 | grep -o -e ' \[ .* \] FileOpen: .* ' echo "READ" $CLICKHOUSE_CLIENT --print-profile-events --profile-events-delay-ms=-1 -nq " SELECT '1', min(t) FROM times SETTINGS optimize_use_implicit_projections = 1; -" 2>&1 | grep -o -e '\ \[\ .*\ \]\ FileOpen:\ .*\ ' +" 2>&1 | grep -o -e ' \[ .* \] FileOpen: .* ' echo "INSERT and READ INSERT" $CLICKHOUSE_CLIENT --print-profile-events --profile-events-delay-ms=-1 -nq " INSERT INTO times SELECT now() + INTERVAL 2 day SETTINGS optimize_on_insert = 0; SELECT '2', min(t) FROM times SETTINGS optimize_use_implicit_projections = 1; INSERT INTO times SELECT now() + INTERVAL 3 day SETTINGS optimize_on_insert = 0; -" 2>&1 | grep -o -e '\ \[\ .*\ \]\ FileOpen:\ .*\ ' +" 2>&1 | grep -o -e ' \[ .* \] FileOpen: .* ' echo "DROP" $CLICKHOUSE_CLIENT -nq " diff --git a/tests/queries/0_stateless/02676_analyzer_limit_offset.sql b/tests/queries/0_stateless/02676_analyzer_limit_offset.sql index 39c6b85f088..5dbe55e916c 100644 --- a/tests/queries/0_stateless/02676_analyzer_limit_offset.sql +++ b/tests/queries/0_stateless/02676_analyzer_limit_offset.sql @@ -1,4 +1,4 @@ -set allow_experimental_analyzer=1; +set enable_analyzer=1; DROP TABLE IF EXISTS test; CREATE TABLE test (i UInt64) Engine = MergeTree() order by i; diff --git a/tests/queries/0_stateless/02676_distinct_reading_in_order_analyzer.sql b/tests/queries/0_stateless/02676_distinct_reading_in_order_analyzer.sql index f9ff1eed111..6a219cd3781 100644 --- a/tests/queries/0_stateless/02676_distinct_reading_in_order_analyzer.sql +++ b/tests/queries/0_stateless/02676_distinct_reading_in_order_analyzer.sql @@ -1,6 +1,6 @@ drop table if exists t; -set allow_experimental_analyzer=1; +set enable_analyzer=1; create table t (a UInt64, b UInt64) engine=MergeTree() order by (a); insert into t select number % 2, number from numbers(10); diff --git a/tests/queries/0_stateless/02677_analyzer_bitmap_has_any.sql b/tests/queries/0_stateless/02677_analyzer_bitmap_has_any.sql index c06ea009c1d..dc906a92f1e 100644 --- a/tests/queries/0_stateless/02677_analyzer_bitmap_has_any.sql +++ b/tests/queries/0_stateless/02677_analyzer_bitmap_has_any.sql @@ -18,7 +18,7 @@ FROM bitmapHasAny(bitmapBuild([toUInt64(1)]), ( SELECT groupBitmapState(toUInt64(2)) )) has2 -) SETTINGS allow_experimental_analyzer = 0; -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } +) SETTINGS enable_analyzer = 0; -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } SELECT '--------------'; @@ -32,5 +32,4 @@ FROM bitmapHasAny(bitmapBuild([toUInt64(1)]), ( SELECT groupBitmapState(toUInt64(2)) )) has2 -) SETTINGS allow_experimental_analyzer = 1; - +) SETTINGS enable_analyzer = 1; diff --git a/tests/queries/0_stateless/02677_analyzer_compound_expressions.sql b/tests/queries/0_stateless/02677_analyzer_compound_expressions.sql index 6b7fdab8993..90781f70158 100644 --- a/tests/queries/0_stateless/02677_analyzer_compound_expressions.sql +++ b/tests/queries/0_stateless/02677_analyzer_compound_expressions.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; WITH ('a', 'b')::Tuple(c1 String, c2 String) AS t SELECT t.c1, t.c2; diff --git a/tests/queries/0_stateless/02677_get_subcolumn_array_of_tuples.sql b/tests/queries/0_stateless/02677_get_subcolumn_array_of_tuples.sql index 5779821afaa..95665979857 100644 --- a/tests/queries/0_stateless/02677_get_subcolumn_array_of_tuples.sql +++ b/tests/queries/0_stateless/02677_get_subcolumn_array_of_tuples.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS t_get_subcolumn; diff --git a/tests/queries/0_stateless/02679_explain_merge_tree_prewhere_row_policy.sql b/tests/queries/0_stateless/02679_explain_merge_tree_prewhere_row_policy.sql index 4bc7be13490..4911979394a 100644 --- a/tests/queries/0_stateless/02679_explain_merge_tree_prewhere_row_policy.sql +++ b/tests/queries/0_stateless/02679_explain_merge_tree_prewhere_row_policy.sql @@ -10,8 +10,8 @@ INSERT INTO test_table VALUES (0, 'Value'); DROP ROW POLICY IF EXISTS test_row_policy ON test_table; CREATE ROW POLICY test_row_policy ON test_table USING id >= 5 TO ALL; -EXPLAIN header = 1, actions = 1 SELECT id, value FROM test_table PREWHERE id = 5 settings allow_experimental_analyzer=0; -EXPLAIN header = 1, actions = 1 SELECT id, value FROM test_table PREWHERE id = 5 settings allow_experimental_analyzer=1; +EXPLAIN header = 1, actions = 1 SELECT id, value FROM test_table PREWHERE id = 5 settings enable_analyzer=0; +EXPLAIN header = 1, actions = 1 SELECT id, value FROM test_table PREWHERE id = 5 settings enable_analyzer=1; DROP ROW POLICY test_row_policy ON test_table; DROP TABLE test_table; diff --git a/tests/queries/0_stateless/02699_polygons_sym_difference_rollup.sql b/tests/queries/0_stateless/02699_polygons_sym_difference_rollup.sql index 502cca20ab2..680b98fb1bd 100644 --- a/tests/queries/0_stateless/02699_polygons_sym_difference_rollup.sql +++ b/tests/queries/0_stateless/02699_polygons_sym_difference_rollup.sql @@ -1,4 +1,4 @@ SELECT polygonsSymDifferenceCartesian([[[(1., 1.)]] AS x], [x]) GROUP BY x WITH ROLLUP; -SELECT [[(2147483647, 0.), (10.0001, 65535), (1, 255), (1023, 2147483646)]], polygonsSymDifferenceCartesian([[[(2147483647, 0.), (10.0001, 65535), (1023, 2147483646)]]], [[[(1000.0001, 10.0001)]]]) GROUP BY [[(2147483647, 0.), (10.0001, 65535), (1023, 2147483646)]] WITH ROLLUP SETTINGS allow_experimental_analyzer=0; -SELECT [[(2147483647, 0.), (10.0001, 65535), (1, 255), (1023, 2147483646)]], polygonsSymDifferenceCartesian([[[(2147483647, 0.), (10.0001, 65535), (1023, 2147483646)]]], [[[(1000.0001, 10.0001)]]]) GROUP BY [[(2147483647, 0.), (10.0001, 65535), (1023, 2147483646)]] WITH ROLLUP SETTINGS allow_experimental_analyzer=1; +SELECT [[(2147483647, 0.), (10.0001, 65535), (1, 255), (1023, 2147483646)]], polygonsSymDifferenceCartesian([[[(2147483647, 0.), (10.0001, 65535), (1023, 2147483646)]]], [[[(1000.0001, 10.0001)]]]) GROUP BY [[(2147483647, 0.), (10.0001, 65535), (1023, 2147483646)]] WITH ROLLUP SETTINGS enable_analyzer=0; +SELECT [[(2147483647, 0.), (10.0001, 65535), (1, 255), (1023, 2147483646)]], polygonsSymDifferenceCartesian([[[(2147483647, 0.), (10.0001, 65535), (1023, 2147483646)]]], [[[(1000.0001, 10.0001)]]]) GROUP BY [[(2147483647, 0.), (10.0001, 65535), (1023, 2147483646)]] WITH ROLLUP SETTINGS enable_analyzer=1; SELECT polygonsSymDifferenceCartesian([[[(100.0001, 1000.0001), (-20., 20.), (10., 10.), (20., 20.), (20., -20.), (1000.0001, 1.1920928955078125e-7)]],[[(0.0001, 100000000000000000000.)]] AS x],[x]) GROUP BY x WITH ROLLUP; diff --git a/tests/queries/0_stateless/02699_polygons_sym_difference_total.sql b/tests/queries/0_stateless/02699_polygons_sym_difference_total.sql index 0fac4b11320..53d0a3bb543 100644 --- a/tests/queries/0_stateless/02699_polygons_sym_difference_total.sql +++ b/tests/queries/0_stateless/02699_polygons_sym_difference_total.sql @@ -1,2 +1,2 @@ -SET allow_experimental_analyzer=0; +SET enable_analyzer=0; SELECT [(9223372036854775807, 1.1754943508222875e-38)], x, NULL, polygonsSymDifferenceCartesian([[[(1.1754943508222875e-38, 1.1920928955078125e-7), (0.5, 0.5)]], [[(1.1754943508222875e-38, 1.1920928955078125e-7), (1.1754943508222875e-38, 1.1920928955078125e-7)], [(0., 1.0001)]], [[(1., 1.0001)]] AS x], [[[(3.4028234663852886e38, 0.9999)]]]) GROUP BY GROUPING SETS ((x)) WITH TOTALS diff --git a/tests/queries/0_stateless/02699_polygons_sym_difference_total_analyzer.sql b/tests/queries/0_stateless/02699_polygons_sym_difference_total_analyzer.sql index 879e0e5297f..40f610ae5a6 100644 --- a/tests/queries/0_stateless/02699_polygons_sym_difference_total_analyzer.sql +++ b/tests/queries/0_stateless/02699_polygons_sym_difference_total_analyzer.sql @@ -1,2 +1,2 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT [(9223372036854775807, 1.1754943508222875e-38)], x, NULL, polygonsSymDifferenceCartesian([[[(1.1754943508222875e-38, 1.1920928955078125e-7), (0.5, 0.5)]], [[(1.1754943508222875e-38, 1.1920928955078125e-7), (1.1754943508222875e-38, 1.1920928955078125e-7)], [(0., 1.0001)]], [[(1., 1.0001)]] AS x], [[[(3.4028234663852886e38, 0.9999)]]]) GROUP BY GROUPING SETS ((x)) WITH TOTALS diff --git a/tests/queries/0_stateless/02701_invalid_having_NOT_AN_AGGREGATE.sql b/tests/queries/0_stateless/02701_invalid_having_NOT_AN_AGGREGATE.sql index 092bda23164..9cfc4d83058 100644 --- a/tests/queries/0_stateless/02701_invalid_having_NOT_AN_AGGREGATE.sql +++ b/tests/queries/0_stateless/02701_invalid_having_NOT_AN_AGGREGATE.sql @@ -1 +1 @@ -SELECT a, sum(b) FROM (SELECT 1 AS a, 1 AS b, 0 AS c) GROUP BY a HAVING c SETTINGS allow_experimental_analyzer=1 -- { serverError NOT_AN_AGGREGATE } +SELECT a, sum(b) FROM (SELECT 1 AS a, 1 AS b, 0 AS c) GROUP BY a HAVING c SETTINGS enable_analyzer=1 -- { serverError NOT_AN_AGGREGATE } diff --git a/tests/queries/0_stateless/02702_logical_optimizer_with_nulls.sql b/tests/queries/0_stateless/02702_logical_optimizer_with_nulls.sql index 72ab507f541..9e2927334e1 100644 --- a/tests/queries/0_stateless/02702_logical_optimizer_with_nulls.sql +++ b/tests/queries/0_stateless/02702_logical_optimizer_with_nulls.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS 02702_logical_optimizer; diff --git a/tests/queries/0_stateless/02703_explain_query_tree_is_forbidden_with_old_analyzer.sql b/tests/queries/0_stateless/02703_explain_query_tree_is_forbidden_with_old_analyzer.sql index d351bfe402c..c028e74f1b5 100644 --- a/tests/queries/0_stateless/02703_explain_query_tree_is_forbidden_with_old_analyzer.sql +++ b/tests/queries/0_stateless/02703_explain_query_tree_is_forbidden_with_old_analyzer.sql @@ -1,2 +1,2 @@ -set allow_experimental_analyzer=0; +set enable_analyzer=0; EXPLAIN QUERY TREE run_passes = true, dump_passes = true SELECT 1; -- { serverError NOT_IMPLEMENTED } 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 6f43c1ae869..79253648475 100755 --- a/tests/queries/0_stateless/02703_max_local_read_bandwidth.sh +++ b/tests/queries/0_stateless/02703_max_local_read_bandwidth.sh @@ -1,11 +1,12 @@ #!/usr/bin/env bash -# Tags: no-object-storage, no-random-settings, no-random-merge-tree-settings +# Tags: no-object-storage, no-random-settings, no-random-merge-tree-settings, no-fasttest +# no-fasttest: The test is slow CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " drop table if exists data; create table data (key UInt64 CODEC(NONE)) engine=MergeTree() order by tuple() settings min_bytes_for_wide_part=1e9; " @@ -25,7 +26,7 @@ read_methods=( for read_method in "${read_methods[@]}"; do query_id=$(random_str 10) $CLICKHOUSE_CLIENT --query_id "$query_id" -q "select * from data format Null settings max_local_read_bandwidth='1M', local_filesystem_read_method='$read_method'" - $CLICKHOUSE_CLIENT -nm -q " + $CLICKHOUSE_CLIENT -m -q " SYSTEM FLUSH LOGS; SELECT '$read_method', 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 4f6a300c5b3..c5776134673 100755 --- a/tests/queries/0_stateless/02703_max_local_write_bandwidth.sh +++ b/tests/queries/0_stateless/02703_max_local_write_bandwidth.sh @@ -5,7 +5,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " drop table if exists data; create table data (key UInt64 CODEC(NONE)) engine=MergeTree() order by tuple() settings min_bytes_for_wide_part=1e9; " @@ -13,7 +13,7 @@ $CLICKHOUSE_CLIENT -nm -q " query_id=$(random_str 10) # writes 1e6*8 bytes with 1M bandwith it should take (8-1)/1=7 seconds $CLICKHOUSE_CLIENT --query_id "$query_id" -q "insert into data select * from numbers(1e6) settings max_local_write_bandwidth='1M'" -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " SYSTEM FLUSH LOGS; SELECT query_duration_ms >= 7e3, diff --git a/tests/queries/0_stateless/02704_max_backup_bandwidth.sh b/tests/queries/0_stateless/02704_max_backup_bandwidth.sh index 8cb03a93a7a..7e914c4c539 100755 --- a/tests/queries/0_stateless/02704_max_backup_bandwidth.sh +++ b/tests/queries/0_stateless/02704_max_backup_bandwidth.sh @@ -5,7 +5,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " drop table if exists data; create table data (key UInt64 CODEC(NONE)) engine=MergeTree() order by tuple() settings min_bytes_for_wide_part=1e9; " @@ -15,7 +15,7 @@ $CLICKHOUSE_CLIENT -q "insert into data select * from numbers(1e6)" query_id=$(random_str 10) $CLICKHOUSE_CLIENT --query_id "$query_id" -q "backup table data to Disk('backups', '$CLICKHOUSE_DATABASE/data/backup1')" --max_backup_bandwidth=1M > /dev/null -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " SYSTEM FLUSH LOGS; SELECT query_duration_ms >= 7e3, diff --git a/tests/queries/0_stateless/02704_storage_merge_explain_graph_crash.sql b/tests/queries/0_stateless/02704_storage_merge_explain_graph_crash.sql index 44a8fe4f049..db5eddf2a90 100644 --- a/tests/queries/0_stateless/02704_storage_merge_explain_graph_crash.sql +++ b/tests/queries/0_stateless/02704_storage_merge_explain_graph_crash.sql @@ -13,4 +13,4 @@ CREATE TABLE foo2_dist (`Id` UInt32, `Val` String) ENGINE = Distributed(test_sha CREATE TABLE merge1 AS foo ENGINE = Merge(currentDatabase(), '^(foo|foo2_dist)$'); EXPLAIN PIPELINE graph = 1, compact = 1 SELECT * FROM merge1 FORMAT Null; -EXPLAIN PIPELINE graph = 1, compact = 1 SELECT * FROM merge1 FORMAT Null SETTINGS allow_experimental_analyzer=1; +EXPLAIN PIPELINE graph = 1, compact = 1 SELECT * FROM merge1 FORMAT Null SETTINGS enable_analyzer=1; diff --git a/tests/queries/0_stateless/02707_analyzer_nested_lambdas_types.sql b/tests/queries/0_stateless/02707_analyzer_nested_lambdas_types.sql index f9258d61900..320e1111e65 100644 --- a/tests/queries/0_stateless/02707_analyzer_nested_lambdas_types.sql +++ b/tests/queries/0_stateless/02707_analyzer_nested_lambdas_types.sql @@ -1,24 +1,24 @@ SELECT range(1), arrayMap(x -> arrayMap(x -> x, range(x)), [1]) -SETTINGS allow_experimental_analyzer = 0; +SETTINGS enable_analyzer = 0; SELECT range(1), arrayMap(x -> arrayMap(x -> x, range(x)), [1]) -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; SELECT range(1), arrayMap(x -> arrayMap(x -> 1, range(x)), [1]) -SETTINGS allow_experimental_analyzer = 0; +SETTINGS enable_analyzer = 0; SELECT range(1), arrayMap(x -> arrayMap(x -> 1, range(x)), [1]) -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; SELECT range(1), arrayMap(x -> arrayMap(y -> 1, range(x)), [1]) -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; diff --git a/tests/queries/0_stateless/02722_matcher_join_use_nulls.sql.j2 b/tests/queries/0_stateless/02722_matcher_join_use_nulls.sql.j2 index 25451a34867..6a8472fecdf 100644 --- a/tests/queries/0_stateless/02722_matcher_join_use_nulls.sql.j2 +++ b/tests/queries/0_stateless/02722_matcher_join_use_nulls.sql.j2 @@ -8,7 +8,7 @@ INSERT INTO t1 VALUES (1); INSERT INTO t2 VALUES (2, 2); SET join_use_nulls = 1; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- { echoOn } diff --git a/tests/queries/0_stateless/02724_limit_num_mutations.sh b/tests/queries/0_stateless/02724_limit_num_mutations.sh index 60888db0e2e..604cc9ff08e 100755 --- a/tests/queries/0_stateless/02724_limit_num_mutations.sh +++ b/tests/queries/0_stateless/02724_limit_num_mutations.sh @@ -23,7 +23,7 @@ function wait_for_alter() done } -${CLICKHOUSE_CLIENT} -n --query " +${CLICKHOUSE_CLIENT} --query " DROP TABLE IF EXISTS t_limit_mutations SYNC; CREATE TABLE t_limit_mutations (id UInt64, v UInt64) @@ -48,14 +48,14 @@ SELECT count() FROM system.mutations WHERE database = currentDatabase() AND tabl SHOW CREATE TABLE t_limit_mutations; " -${CLICKHOUSE_CLIENT} -n --query " +${CLICKHOUSE_CLIENT} --query " ALTER TABLE t_limit_mutations UPDATE v = 6 WHERE 1 SETTINGS number_of_mutations_to_throw = 100; ALTER TABLE t_limit_mutations MODIFY COLUMN v String SETTINGS number_of_mutations_to_throw = 100, alter_sync = 0; " wait_for_alter "String" -${CLICKHOUSE_CLIENT} -n --query " +${CLICKHOUSE_CLIENT} --query " SELECT * FROM t_limit_mutations ORDER BY id; SELECT count() FROM system.mutations WHERE database = currentDatabase() AND table = 't_limit_mutations' AND NOT is_done; SHOW CREATE TABLE t_limit_mutations; @@ -65,7 +65,7 @@ ${CLICKHOUSE_CLIENT} --query "SYSTEM START MERGES t_limit_mutations" wait_for_mutation "t_limit_mutations" "0000000003" -${CLICKHOUSE_CLIENT} -n --query " +${CLICKHOUSE_CLIENT} --query " SELECT * FROM t_limit_mutations ORDER BY id; SELECT count() FROM system.mutations WHERE database = currentDatabase() AND table = 't_limit_mutations' AND NOT is_done; SHOW CREATE TABLE t_limit_mutations; diff --git a/tests/queries/0_stateless/02725_async_insert_table_setting.sh b/tests/queries/0_stateless/02725_async_insert_table_setting.sh index 13911e8d677..14c2d335275 100755 --- a/tests/queries/0_stateless/02725_async_insert_table_setting.sh +++ b/tests/queries/0_stateless/02725_async_insert_table_setting.sh @@ -4,7 +4,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} -n --query " +${CLICKHOUSE_CLIENT} --query " DROP TABLE IF EXISTS t_mt_async_insert; DROP TABLE IF EXISTS t_mt_sync_insert; @@ -19,7 +19,7 @@ url="${CLICKHOUSE_URL}&async_insert=0&wait_for_async_insert=1" ${CLICKHOUSE_CURL} -sS "$url" -d "INSERT INTO t_mt_async_insert VALUES (1, 'aa'), (2, 'bb')" ${CLICKHOUSE_CURL} -sS "$url" -d "INSERT INTO t_mt_sync_insert VALUES (1, 'aa'), (2, 'bb')" -${CLICKHOUSE_CLIENT} -n --query " +${CLICKHOUSE_CLIENT} --query " SELECT count() FROM t_mt_async_insert; SELECT count() FROM t_mt_sync_insert; diff --git a/tests/queries/0_stateless/02725_cnf_large_check.sql b/tests/queries/0_stateless/02725_cnf_large_check.sql index 0780e6bcdd3..2567636c02c 100644 --- a/tests/queries/0_stateless/02725_cnf_large_check.sql +++ b/tests/queries/0_stateless/02725_cnf_large_check.sql @@ -7,21 +7,21 @@ INSERT INTO 02725_cnf VALUES (0, 0, 0, 0, 0, 0, 0, 0, 0, 0), (0, 0, 0, 0, 0, 0, SELECT count() FROM 02725_cnf WHERE (c5 AND (NOT c0)) OR ((NOT c3) AND (NOT c6) AND (NOT c1) AND (NOT c6)) OR (c7 AND (NOT c3) AND (NOT c5) AND (NOT c7)) OR ((NOT c8) AND c5) OR ((NOT c0)) OR ((NOT c8) AND (NOT c5) AND c1 AND c6 AND c3) OR (c7 AND (NOT c0) AND c6 AND c1 AND (NOT c2)) OR (c3 AND (NOT c9) AND c1) -SETTINGS convert_query_to_cnf = 1, allow_experimental_analyzer = 1; +SETTINGS convert_query_to_cnf = 1, enable_analyzer = 1; SELECT count() FROM 02725_cnf WHERE (c5 AND (NOT c0)) OR ((NOT c3) AND (NOT c6) AND (NOT c1) AND (NOT c6)) OR (c7 AND (NOT c3) AND (NOT c5) AND (NOT c7)) OR ((NOT c8) AND c5) OR ((NOT c0)) OR ((NOT c8) AND (NOT c5) AND c1 AND c6 AND c3) OR (c7 AND (NOT c0) AND c6 AND c1 AND (NOT c2)) OR (c3 AND (NOT c9) AND c1) -SETTINGS convert_query_to_cnf = 1, allow_experimental_analyzer = 0; +SETTINGS convert_query_to_cnf = 1, enable_analyzer = 0; SELECT count() FROM 02725_cnf WHERE ((NOT c2) AND c2 AND (NOT c1)) OR ((NOT c2) AND c3 AND (NOT c5)) OR ((NOT c7) AND (NOT c8)) OR (c9 AND c6 AND c8 AND (NOT c8) AND (NOT c7)) -SETTINGS convert_query_to_cnf = 1, allow_experimental_analyzer = 1; +SETTINGS convert_query_to_cnf = 1, enable_analyzer = 1; SELECT count() FROM 02725_cnf WHERE ((NOT c2) AND c2 AND (NOT c1)) OR ((NOT c2) AND c3 AND (NOT c5)) OR ((NOT c7) AND (NOT c8)) OR (c9 AND c6 AND c8 AND (NOT c8) AND (NOT c7)) -SETTINGS convert_query_to_cnf = 1, allow_experimental_analyzer = 0; +SETTINGS convert_query_to_cnf = 1, enable_analyzer = 0; DROP TABLE 02725_cnf; 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 97d644fa4d6..5d941adcb81 100644 --- a/tests/queries/0_stateless/02726_async_insert_flush_queue.sql +++ b/tests/queries/0_stateless/02726_async_insert_flush_queue.sql @@ -30,7 +30,9 @@ SELECT count() FROM t_async_inserts_flush; SYSTEM FLUSH ASYNC INSERT QUEUE; -SELECT count() FROM system.asynchronous_inserts; +SELECT count() FROM system.asynchronous_inserts +WHERE database = currentDatabase() AND table = 't_async_inserts_flush'; + SELECT count() FROM t_async_inserts_flush; DROP TABLE t_async_inserts_flush; diff --git a/tests/queries/0_stateless/02726_async_insert_flush_stress.sh b/tests/queries/0_stateless/02726_async_insert_flush_stress.sh index 876766d0780..61bbbd620f0 100755 --- a/tests/queries/0_stateless/02726_async_insert_flush_stress.sh +++ b/tests/queries/0_stateless/02726_async_insert_flush_stress.sh @@ -91,5 +91,5 @@ flush1 $TIMEOUT & wait ${CLICKHOUSE_CLIENT} -q "SYSTEM FLUSH ASYNC INSERT QUEUE" -${CLICKHOUSE_CLIENT} -q "SELECT count() FROM system.asynchronous_inserts" +${CLICKHOUSE_CLIENT} -q "SELECT count() FROM system.asynchronous_inserts WHERE database = currentDatabase() AND table = 'async_inserts'" ${CLICKHOUSE_CLIENT} -q "DROP TABLE IF EXISTS async_inserts"; diff --git a/tests/queries/0_stateless/02731_analyzer_join_resolve_nested.sql.j2 b/tests/queries/0_stateless/02731_analyzer_join_resolve_nested.sql.j2 index c2f3c51b17a..4ddf41c4d6d 100644 --- a/tests/queries/0_stateless/02731_analyzer_join_resolve_nested.sql.j2 +++ b/tests/queries/0_stateless/02731_analyzer_join_resolve_nested.sql.j2 @@ -31,7 +31,7 @@ INSERT INTO nnna VALUES (1, [[([([(1,'d')],'d')], 's')]], ['s']); CREATE TABLE nnnb ( x UInt64, t Nested(t Nested(t Nested(t Nested(t UInt32, s String), s String), s String), s String) ) ENGINE = MergeTree ORDER BY x; INSERT INTO nnnb VALUES (1, [[([([(1,'d')],'d')], 's')]], ['s']); -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; {% for join_use_nulls in [0, 1] -%} diff --git a/tests/queries/0_stateless/02731_parallel_replicas_join_subquery.reference b/tests/queries/0_stateless/02731_parallel_replicas_join_subquery.reference index 451f0d6d485..125fa524c4a 100644 --- a/tests/queries/0_stateless/02731_parallel_replicas_join_subquery.reference +++ b/tests/queries/0_stateless/02731_parallel_replicas_join_subquery.reference @@ -33,8 +33,8 @@ =============== QUERIES EXECUTED BY PARALLEL INNER QUERY ALONE =============== 0 3 SELECT `__table1`.`key` AS `key`, `__table1`.`value1` AS `value1`, `__table1`.`value2` AS `value2`, toUInt64(min(`__table1`.`time`)) AS `start_ts` FROM `default`.`join_inner_table` AS `__table1` PREWHERE (`__table1`.`id` = \'833c9e22-c245-4eb5-8745-117a9a1f26b1\') AND (`__table1`.`number` > _CAST(1610517366120, \'UInt64\')) GROUP BY `__table1`.`key`, `__table1`.`value1`, `__table1`.`value2` ORDER BY `__table1`.`key` ASC, `__table1`.`value1` ASC, `__table1`.`value2` ASC LIMIT _CAST(10, \'UInt64\') 0 3 SELECT `key`, `value1`, `value2`, toUInt64(min(`time`)) AS `start_ts` FROM `default`.`join_inner_table` PREWHERE (`id` = \'833c9e22-c245-4eb5-8745-117a9a1f26b1\') AND (`number` > toUInt64(\'1610517366120\')) GROUP BY `key`, `value1`, `value2` ORDER BY `key` ASC, `value1` ASC, `value2` ASC LIMIT 10 -1 1 -- Parallel inner query alone\nSELECT\n key,\n value1,\n value2,\n toUInt64(min(time)) AS start_ts\nFROM join_inner_table\nPREWHERE (id = \'833c9e22-c245-4eb5-8745-117a9a1f26b1\') AND (number > toUInt64(\'1610517366120\'))\nGROUP BY key, value1, value2\nORDER BY key, value1, value2\nLIMIT 10\nSETTINGS allow_experimental_parallel_reading_from_replicas = 1, allow_experimental_analyzer=0; -1 1 -- Parallel inner query alone\nSELECT\n key,\n value1,\n value2,\n toUInt64(min(time)) AS start_ts\nFROM join_inner_table\nPREWHERE (id = \'833c9e22-c245-4eb5-8745-117a9a1f26b1\') AND (number > toUInt64(\'1610517366120\'))\nGROUP BY key, value1, value2\nORDER BY key, value1, value2\nLIMIT 10\nSETTINGS allow_experimental_parallel_reading_from_replicas = 1, allow_experimental_analyzer=1; +1 1 -- Parallel inner query alone\nSELECT\n key,\n value1,\n value2,\n toUInt64(min(time)) AS start_ts\nFROM join_inner_table\nPREWHERE (id = \'833c9e22-c245-4eb5-8745-117a9a1f26b1\') AND (number > toUInt64(\'1610517366120\'))\nGROUP BY key, value1, value2\nORDER BY key, value1, value2\nLIMIT 10\nSETTINGS allow_experimental_parallel_reading_from_replicas = 1, enable_analyzer=0; +1 1 -- Parallel inner query alone\nSELECT\n key,\n value1,\n value2,\n toUInt64(min(time)) AS start_ts\nFROM join_inner_table\nPREWHERE (id = \'833c9e22-c245-4eb5-8745-117a9a1f26b1\') AND (number > toUInt64(\'1610517366120\'))\nGROUP BY key, value1, value2\nORDER BY key, value1, value2\nLIMIT 10\nSETTINGS allow_experimental_parallel_reading_from_replicas = 1, enable_analyzer=1; =============== OUTER QUERY (NO PARALLEL) =============== >T%O ,z< 10 NQTpY# W\\Xx4 10 @@ -61,5 +61,5 @@ t _CAST(1610517366120, \'UInt64\')) GROUP BY `__table3`.`key`, `__table3`.`value1`, `__table3`.`value2`) AS `__table2` USING (`key`) GROUP BY `__table1`.`key`, `__table2`.`value1`, `__table2`.`value2` 0 3 SELECT `key`, `value1`, `value2` FROM `default`.`join_inner_table` PREWHERE (`id` = \'833c9e22-c245-4eb5-8745-117a9a1f26b1\') AND (`number` > toUInt64(\'1610517366120\')) GROUP BY `key`, `value1`, `value2` 0 3 SELECT `value1`, `value2`, count() AS `count` FROM `default`.`join_outer_table` ALL INNER JOIN `_data_` USING (`key`) GROUP BY `key`, `value1`, `value2` -1 1 -- Parallel full query\nSELECT\n value1,\n value2,\n avg(count) AS avg\nFROM\n (\n SELECT\n key,\n value1,\n value2,\n count() AS count\n FROM join_outer_table\n INNER JOIN\n (\n SELECT\n key,\n value1,\n value2,\n toUInt64(min(time)) AS start_ts\n FROM join_inner_table\n PREWHERE (id = \'833c9e22-c245-4eb5-8745-117a9a1f26b1\') AND (number > toUInt64(\'1610517366120\'))\n GROUP BY key, value1, value2\n ) USING (key)\n GROUP BY key, value1, value2\n )\nGROUP BY value1, value2\nORDER BY value1, value2\nSETTINGS allow_experimental_parallel_reading_from_replicas = 1, allow_experimental_analyzer=0; -1 1 -- Parallel full query\nSELECT\n value1,\n value2,\n avg(count) AS avg\nFROM\n (\n SELECT\n key,\n value1,\n value2,\n count() AS count\n FROM join_outer_table\n INNER JOIN\n (\n SELECT\n key,\n value1,\n value2,\n toUInt64(min(time)) AS start_ts\n FROM join_inner_table\n PREWHERE (id = \'833c9e22-c245-4eb5-8745-117a9a1f26b1\') AND (number > toUInt64(\'1610517366120\'))\n GROUP BY key, value1, value2\n ) USING (key)\n GROUP BY key, value1, value2\n )\nGROUP BY value1, value2\nORDER BY value1, value2\nSETTINGS allow_experimental_parallel_reading_from_replicas = 1, allow_experimental_analyzer=1; +1 1 -- Parallel full query\nSELECT\n value1,\n value2,\n avg(count) AS avg\nFROM\n (\n SELECT\n key,\n value1,\n value2,\n count() AS count\n FROM join_outer_table\n INNER JOIN\n (\n SELECT\n key,\n value1,\n value2,\n toUInt64(min(time)) AS start_ts\n FROM join_inner_table\n PREWHERE (id = \'833c9e22-c245-4eb5-8745-117a9a1f26b1\') AND (number > toUInt64(\'1610517366120\'))\n GROUP BY key, value1, value2\n ) USING (key)\n GROUP BY key, value1, value2\n )\nGROUP BY value1, value2\nORDER BY value1, value2\nSETTINGS allow_experimental_parallel_reading_from_replicas = 1, enable_analyzer=0; +1 1 -- Parallel full query\nSELECT\n value1,\n value2,\n avg(count) AS avg\nFROM\n (\n SELECT\n key,\n value1,\n value2,\n count() AS count\n FROM join_outer_table\n INNER JOIN\n (\n SELECT\n key,\n value1,\n value2,\n toUInt64(min(time)) AS start_ts\n FROM join_inner_table\n PREWHERE (id = \'833c9e22-c245-4eb5-8745-117a9a1f26b1\') AND (number > toUInt64(\'1610517366120\'))\n GROUP BY key, value1, value2\n ) USING (key)\n GROUP BY key, value1, value2\n )\nGROUP BY value1, value2\nORDER BY value1, value2\nSETTINGS allow_experimental_parallel_reading_from_replicas = 1, enable_analyzer=1; diff --git a/tests/queries/0_stateless/02731_parallel_replicas_join_subquery.sql b/tests/queries/0_stateless/02731_parallel_replicas_join_subquery.sql index 7693d0da295..8121d60a05b 100644 --- a/tests/queries/0_stateless/02731_parallel_replicas_join_subquery.sql +++ b/tests/queries/0_stateless/02731_parallel_replicas_join_subquery.sql @@ -38,7 +38,7 @@ FROM join_inner_table GROUP BY key, value1, value2 ORDER BY key, value1, value2 LIMIT 10; --- settings allow_experimental_analyzer=0; +-- settings enable_analyzer=0; -- SELECT -- key, @@ -49,7 +49,7 @@ LIMIT 10; -- PREWHERE (id = '833c9e22-c245-4eb5-8745-117a9a1f26b1') AND (number > toUInt64('1610517366120')) -- GROUP BY key, value1, value2 -- ORDER BY key, value1, value2 --- LIMIT 10 settings allow_experimental_analyzer=1; +-- LIMIT 10 settings enable_analyzer=1; SELECT '=============== INNER QUERY (PARALLEL) ==============='; @@ -64,7 +64,7 @@ PREWHERE (id = '833c9e22-c245-4eb5-8745-117a9a1f26b1') AND (number > toUInt64('1 GROUP BY key, value1, value2 ORDER BY key, value1, value2 LIMIT 10 -SETTINGS allow_experimental_parallel_reading_from_replicas = 1, allow_experimental_analyzer=0; +SETTINGS allow_experimental_parallel_reading_from_replicas = 1, enable_analyzer=0; -- Parallel inner query alone SELECT @@ -77,7 +77,7 @@ PREWHERE (id = '833c9e22-c245-4eb5-8745-117a9a1f26b1') AND (number > toUInt64('1 GROUP BY key, value1, value2 ORDER BY key, value1, value2 LIMIT 10 -SETTINGS allow_experimental_parallel_reading_from_replicas = 1, allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 1, enable_analyzer=1; SELECT '=============== QUERIES EXECUTED BY PARALLEL INNER QUERY ALONE ==============='; @@ -184,7 +184,7 @@ FROM ) GROUP BY value1, value2 ORDER BY value1, value2 -SETTINGS allow_experimental_parallel_reading_from_replicas = 1, allow_experimental_analyzer=0; +SETTINGS allow_experimental_parallel_reading_from_replicas = 1, enable_analyzer=0; -- Parallel full query SELECT @@ -214,7 +214,7 @@ FROM ) GROUP BY value1, value2 ORDER BY value1, value2 -SETTINGS allow_experimental_parallel_reading_from_replicas = 1, allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 1, enable_analyzer=1; SYSTEM FLUSH LOGS; diff --git a/tests/queries/0_stateless/02734_optimize_group_by.sql b/tests/queries/0_stateless/02734_optimize_group_by.sql index 28e86c04b0f..626805d028d 100644 --- a/tests/queries/0_stateless/02734_optimize_group_by.sql +++ b/tests/queries/0_stateless/02734_optimize_group_by.sql @@ -1,5 +1,5 @@ -SELECT 'a' AS key, 'b' as value GROUP BY key WITH CUBE SETTINGS allow_experimental_analyzer = 0; -SELECT 'a' AS key, 'b' as value GROUP BY key WITH CUBE SETTINGS allow_experimental_analyzer = 1; +SELECT 'a' AS key, 'b' as value GROUP BY key WITH CUBE SETTINGS enable_analyzer = 0; +SELECT 'a' AS key, 'b' as value GROUP BY key WITH CUBE SETTINGS enable_analyzer = 1; SELECT 'a' AS key, 'b' as value GROUP BY ignore(1) WITH CUBE; diff --git a/tests/queries/0_stateless/02765_queries_with_subqueries_profile_events.reference b/tests/queries/0_stateless/02765_queries_with_subqueries_profile_events.reference index d21f914f0dc..a6c8c0bbc3e 100644 --- a/tests/queries/0_stateless/02765_queries_with_subqueries_profile_events.reference +++ b/tests/queries/0_stateless/02765_queries_with_subqueries_profile_events.reference @@ -1,24 +1,24 @@ -view allow_experimental_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries +view enable_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries 1 0 1 0 1 2 3 -subquery allow_experimental_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries +subquery enable_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries 1 0 0 1 0 2 2 -CSE allow_experimental_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries +CSE enable_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries 1 0 0 1 0 2 2 -CSE_Multi allow_experimental_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries +CSE_Multi enable_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries 1 0 0 1 0 2 2 -CTE allow_experimental_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries +CTE enable_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries 1 0 0 1 0 2 2 -CTE_Multi allow_experimental_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries +CTE_Multi enable_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries 1 0 0 1 0 4 4 -view allow_experimental_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries +view enable_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries 1 1 1 0 1 3 4 -subquery allow_experimental_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries +subquery enable_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries 1 1 0 1 0 2 2 -CSE allow_experimental_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries +CSE enable_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries 1 1 0 1 0 2 2 -CSE_Multi allow_experimental_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries +CSE_Multi enable_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries 1 1 0 1 0 2 2 -CTE allow_experimental_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries +CTE enable_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries 1 1 0 1 0 2 2 -CTE_Multi allow_experimental_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries +CTE_Multi enable_analyzer InsertQuery SelectQuery InsertQueriesWithSubqueries SelectQueriesWithSubqueries QueriesWithSubqueries 1 1 0 1 0 4 4 diff --git a/tests/queries/0_stateless/02765_queries_with_subqueries_profile_events.sh b/tests/queries/0_stateless/02765_queries_with_subqueries_profile_events.sh index 84031ad9081..b7d93b5396c 100755 --- a/tests/queries/0_stateless/02765_queries_with_subqueries_profile_events.sh +++ b/tests/queries/0_stateless/02765_queries_with_subqueries_profile_events.sh @@ -14,14 +14,14 @@ $CLICKHOUSE_CLIENT -n -q " CREATE MATERIALIZED VIEW mv TO output SQL SECURITY NONE AS SELECT * FROM input; " -for allow_experimental_analyzer in 0 1; do +for enable_analyzer in 0 1; do query_id="$(random_str 10)" - $CLICKHOUSE_CLIENT --allow_experimental_analyzer "$allow_experimental_analyzer" --query_id "$query_id" -q "INSERT INTO input SELECT * FROM numbers(1)" + $CLICKHOUSE_CLIENT --enable_analyzer "$enable_analyzer" --query_id "$query_id" -q "INSERT INTO input SELECT * FROM numbers(1)" $CLICKHOUSE_CLIENT -mn -q " SYSTEM FLUSH LOGS; SELECT 1 view, - $allow_experimental_analyzer allow_experimental_analyzer, + $enable_analyzer enable_analyzer, ProfileEvents['InsertQuery'] InsertQuery, ProfileEvents['SelectQuery'] SelectQuery, ProfileEvents['InsertQueriesWithSubqueries'] InsertQueriesWithSubqueries, @@ -34,12 +34,12 @@ for allow_experimental_analyzer in 0 1; do " query_id="$(random_str 10)" - $CLICKHOUSE_CLIENT --allow_experimental_analyzer "$allow_experimental_analyzer" --query_id "$query_id" -q "SELECT * FROM system.one WHERE dummy IN (SELECT * FROM system.one) FORMAT Null" + $CLICKHOUSE_CLIENT --enable_analyzer "$enable_analyzer" --query_id "$query_id" -q "SELECT * FROM system.one WHERE dummy IN (SELECT * FROM system.one) FORMAT Null" $CLICKHOUSE_CLIENT -mn -q " SYSTEM FLUSH LOGS; SELECT 1 subquery, - $allow_experimental_analyzer allow_experimental_analyzer, + $enable_analyzer enable_analyzer, ProfileEvents['InsertQuery'] InsertQuery, ProfileEvents['SelectQuery'] SelectQuery, ProfileEvents['InsertQueriesWithSubqueries'] InsertQueriesWithSubqueries, @@ -51,12 +51,12 @@ for allow_experimental_analyzer in 0 1; do " query_id="$(random_str 10)" - $CLICKHOUSE_CLIENT --allow_experimental_analyzer "$allow_experimental_analyzer" --query_id "$query_id" -q "WITH (SELECT * FROM system.one) AS x SELECT x FORMAT Null" + $CLICKHOUSE_CLIENT --enable_analyzer "$enable_analyzer" --query_id "$query_id" -q "WITH (SELECT * FROM system.one) AS x SELECT x FORMAT Null" $CLICKHOUSE_CLIENT -mn -q " SYSTEM FLUSH LOGS; SELECT 1 CSE, - $allow_experimental_analyzer allow_experimental_analyzer, + $enable_analyzer enable_analyzer, ProfileEvents['InsertQuery'] InsertQuery, ProfileEvents['SelectQuery'] SelectQuery, ProfileEvents['InsertQueriesWithSubqueries'] InsertQueriesWithSubqueries, @@ -68,12 +68,12 @@ for allow_experimental_analyzer in 0 1; do " query_id="$(random_str 10)" - $CLICKHOUSE_CLIENT --allow_experimental_analyzer "$allow_experimental_analyzer" --query_id "$query_id" -q "WITH (SELECT * FROM system.one) AS x SELECT x, x FORMAT Null" + $CLICKHOUSE_CLIENT --enable_analyzer "$enable_analyzer" --query_id "$query_id" -q "WITH (SELECT * FROM system.one) AS x SELECT x, x FORMAT Null" $CLICKHOUSE_CLIENT -mn -q " SYSTEM FLUSH LOGS; SELECT 1 CSE_Multi, - $allow_experimental_analyzer allow_experimental_analyzer, + $enable_analyzer enable_analyzer, ProfileEvents['InsertQuery'] InsertQuery, ProfileEvents['SelectQuery'] SelectQuery, ProfileEvents['InsertQueriesWithSubqueries'] InsertQueriesWithSubqueries, @@ -85,12 +85,12 @@ for allow_experimental_analyzer in 0 1; do " query_id="$(random_str 10)" - $CLICKHOUSE_CLIENT --allow_experimental_analyzer "$allow_experimental_analyzer" --query_id "$query_id" -q "WITH x AS (SELECT * FROM system.one) SELECT * FROM x FORMAT Null" + $CLICKHOUSE_CLIENT --enable_analyzer "$enable_analyzer" --query_id "$query_id" -q "WITH x AS (SELECT * FROM system.one) SELECT * FROM x FORMAT Null" $CLICKHOUSE_CLIENT -mn -q " SYSTEM FLUSH LOGS; SELECT 1 CTE, - $allow_experimental_analyzer allow_experimental_analyzer, + $enable_analyzer enable_analyzer, ProfileEvents['InsertQuery'] InsertQuery, ProfileEvents['SelectQuery'] SelectQuery, ProfileEvents['InsertQueriesWithSubqueries'] InsertQueriesWithSubqueries, @@ -102,12 +102,12 @@ for allow_experimental_analyzer in 0 1; do " query_id="$(random_str 10)" - $CLICKHOUSE_CLIENT --allow_experimental_analyzer "$allow_experimental_analyzer" --query_id "$query_id" -q "WITH x AS (SELECT * FROM system.one) SELECT * FROM x UNION ALL SELECT * FROM x FORMAT Null" + $CLICKHOUSE_CLIENT --enable_analyzer "$enable_analyzer" --query_id "$query_id" -q "WITH x AS (SELECT * FROM system.one) SELECT * FROM x UNION ALL SELECT * FROM x FORMAT Null" $CLICKHOUSE_CLIENT -mn -q " SYSTEM FLUSH LOGS; SELECT 1 CTE_Multi, - $allow_experimental_analyzer allow_experimental_analyzer, + $enable_analyzer enable_analyzer, ProfileEvents['InsertQuery'] InsertQuery, ProfileEvents['SelectQuery'] SelectQuery, ProfileEvents['InsertQueriesWithSubqueries'] InsertQueriesWithSubqueries, diff --git a/tests/queries/0_stateless/02767_into_outfile_extensions_msan.sh b/tests/queries/0_stateless/02767_into_outfile_extensions_msan.sh index 0c5767314d5..2d2ee328a29 100755 --- a/tests/queries/0_stateless/02767_into_outfile_extensions_msan.sh +++ b/tests/queries/0_stateless/02767_into_outfile_extensions_msan.sh @@ -6,6 +6,6 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) out="explain1.$CLICKHOUSE_TEST_UNIQUE_NAME.out" # only EXPLAIN triggers the problem under MSan -$CLICKHOUSE_CLIENT --allow_experimental_analyzer=0 -q "explain select * from numbers(1) into outfile '$out'" +$CLICKHOUSE_CLIENT --enable_analyzer=0 -q "explain select * from numbers(1) into outfile '$out'" cat "$out" rm -f "$out" diff --git a/tests/queries/0_stateless/02771_ignore_data_skipping_indices.sql b/tests/queries/0_stateless/02771_ignore_data_skipping_indices.sql index d86b65c3291..91ca5ef0340 100644 --- a/tests/queries/0_stateless/02771_ignore_data_skipping_indices.sql +++ b/tests/queries/0_stateless/02771_ignore_data_skipping_indices.sql @@ -23,12 +23,12 @@ SELECT * FROM data_02771 SETTINGS ignore_data_skipping_indices='na_idx'; SELECT * FROM data_02771 WHERE x = 1 AND y = 1 SETTINGS ignore_data_skipping_indices='xy_idx',force_data_skipping_indices='xy_idx' ; -- { serverError INDEX_NOT_USED } SELECT * FROM data_02771 WHERE x = 1 AND y = 2 SETTINGS ignore_data_skipping_indices='xy_idx'; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT * from ( EXPLAIN indexes = 1 SELECT * FROM data_02771 WHERE x = 1 AND y = 2 ) WHERE explain NOT LIKE '%Expression%' AND explain NOT LIKE '%Filter%'; SELECT * from ( EXPLAIN indexes = 1 SELECT * FROM data_02771 WHERE x = 1 AND y = 2 SETTINGS ignore_data_skipping_indices='xy_idx' ) WHERE explain NOT LIKE '%Expression%' AND explain NOT LIKE '%Filter%'; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT * from ( EXPLAIN indexes = 1 SELECT * FROM data_02771 WHERE x = 1 AND y = 2 ) WHERE explain NOT LIKE '%Expression%' AND explain NOT LIKE '%Filter%'; SELECT * from ( EXPLAIN indexes = 1 SELECT * FROM data_02771 WHERE x = 1 AND y = 2 SETTINGS ignore_data_skipping_indices='xy_idx' ) WHERE explain NOT LIKE '%Expression%' AND explain NOT LIKE '%Filter%'; diff --git a/tests/queries/0_stateless/02771_parallel_replicas_analyzer.reference b/tests/queries/0_stateless/02771_parallel_replicas_analyzer.reference index 5bf3520ccdb..f60f1e0a376 100644 --- a/tests/queries/0_stateless/02771_parallel_replicas_analyzer.reference +++ b/tests/queries/0_stateless/02771_parallel_replicas_analyzer.reference @@ -8,5 +8,5 @@ 5935810273536892891 7885388429666205427 8124171311239967992 -1 1 -- Simple query with analyzer and pure parallel replicas\nSELECT number\nFROM join_inner_table__fuzz_146_replicated\n SETTINGS\n allow_experimental_analyzer = 1,\n max_parallel_replicas = 2,\n cluster_for_parallel_replicas = \'test_cluster_one_shard_three_replicas_localhost\',\n allow_experimental_parallel_reading_from_replicas = 1; +1 1 -- Simple query with analyzer and pure parallel replicas\nSELECT number\nFROM join_inner_table__fuzz_146_replicated\n SETTINGS\n enable_analyzer = 1,\n max_parallel_replicas = 2,\n cluster_for_parallel_replicas = \'test_cluster_one_shard_three_replicas_localhost\',\n allow_experimental_parallel_reading_from_replicas = 1; 0 2 SELECT `__table1`.`number` AS `number` FROM `default`.`join_inner_table__fuzz_146_replicated` AS `__table1` diff --git a/tests/queries/0_stateless/02771_parallel_replicas_analyzer.sql b/tests/queries/0_stateless/02771_parallel_replicas_analyzer.sql index 88a0d2163d6..e60049f2756 100644 --- a/tests/queries/0_stateless/02771_parallel_replicas_analyzer.sql +++ b/tests/queries/0_stateless/02771_parallel_replicas_analyzer.sql @@ -21,7 +21,7 @@ INSERT INTO join_inner_table__fuzz_146_replicated SELECT number FROM join_inner_table__fuzz_146_replicated SETTINGS - allow_experimental_analyzer = 1, + enable_analyzer = 1, max_parallel_replicas = 2, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_parallel_reading_from_replicas = 1; diff --git a/tests/queries/0_stateless/02771_semi_join_use_nulls.sql.j2 b/tests/queries/0_stateless/02771_semi_join_use_nulls.sql.j2 index 248461a98bb..74e252c785b 100644 --- a/tests/queries/0_stateless/02771_semi_join_use_nulls.sql.j2 +++ b/tests/queries/0_stateless/02771_semi_join_use_nulls.sql.j2 @@ -1,11 +1,11 @@ -{% for allow_experimental_analyzer in [0, 1] -%} +{% for enable_analyzer in [0, 1] -%} {% for join_use_nulls in [0, 1] -%} {% for kind in ['LEFT', 'RIGHT'] -%} {% for strictness in ['SEMI', 'ANTI'] -%} {% for maybe_materialize in ['', 'materialize'] -%} -SET allow_experimental_analyzer = {{ allow_experimental_analyzer }}; +SET enable_analyzer = {{ enable_analyzer }}; SET join_use_nulls = {{ join_use_nulls }}; diff --git a/tests/queries/0_stateless/02783_date_predicate_optimizations.sql b/tests/queries/0_stateless/02783_date_predicate_optimizations.sql index 4da8cebff1c..b127af677ee 100644 --- a/tests/queries/0_stateless/02783_date_predicate_optimizations.sql +++ b/tests/queries/0_stateless/02783_date_predicate_optimizations.sql @@ -11,8 +11,8 @@ INSERT INTO source values ('2021-12-31 23:00:00', 0); SELECT * FROM source WHERE toYYYYMM(ts) = 202112; SELECT * FROM source WHERE toYear(ts) = 2021; -SELECT * FROM source WHERE toYYYYMM(ts) = 202112 SETTINGS allow_experimental_analyzer=1; -SELECT * FROM source WHERE toYear(ts) = 2021 SETTINGS allow_experimental_analyzer=1; +SELECT * FROM source WHERE toYYYYMM(ts) = 202112 SETTINGS enable_analyzer=1; +SELECT * FROM source WHERE toYear(ts) = 2021 SETTINGS enable_analyzer=1; DROP TABLE IF EXISTS source; CREATE TABLE source @@ -46,18 +46,18 @@ SELECT count(*) FROM source WHERE toYear(dt) < 2023; SELECT count(*) FROM source WHERE toYear(dt) <= 2023; SELECT count(*) FROM source WHERE toYear(dt) > 2023; SELECT count(*) FROM source WHERE toYear(dt) >= 2023; -SELECT count(*) FROM source WHERE toYYYYMM(dt) = 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(dt) <> 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(dt) < 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(dt) <= 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(dt) > 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(dt) >= 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(dt) = 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(dt) <> 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(dt) < 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(dt) <= 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(dt) > 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(dt) >= 2023 SETTINGS allow_experimental_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(dt) = 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(dt) <> 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(dt) < 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(dt) <= 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(dt) > 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(dt) >= 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(dt) = 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(dt) <> 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(dt) < 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(dt) <= 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(dt) > 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(dt) >= 2023 SETTINGS enable_analyzer=1; SELECT 'DateTime'; SELECT count(*) FROM source WHERE toYYYYMM(ts) = 202312; @@ -72,18 +72,18 @@ SELECT count(*) FROM source WHERE toYear(ts) < 2023; SELECT count(*) FROM source WHERE toYear(ts) <= 2023; SELECT count(*) FROM source WHERE toYear(ts) > 2023; SELECT count(*) FROM source WHERE toYear(ts) >= 2023; -SELECT count(*) FROM source WHERE toYYYYMM(ts) = 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(ts) <> 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(ts) < 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(ts) <= 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(ts) > 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(ts) >= 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(ts) = 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(ts) <> 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(ts) < 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(ts) <= 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(ts) > 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(ts) >= 2023 SETTINGS allow_experimental_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(ts) = 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(ts) <> 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(ts) < 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(ts) <= 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(ts) > 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(ts) >= 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(ts) = 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(ts) <> 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(ts) < 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(ts) <= 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(ts) > 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(ts) >= 2023 SETTINGS enable_analyzer=1; SELECT 'Date32'; SELECT count(*) FROM source WHERE toYYYYMM(dt_32) = 202312; @@ -98,18 +98,18 @@ SELECT count(*) FROM source WHERE toYear(dt_32) < 2023; SELECT count(*) FROM source WHERE toYear(dt_32) <= 2023; SELECT count(*) FROM source WHERE toYear(dt_32) > 2023; SELECT count(*) FROM source WHERE toYear(dt_32) >= 2023; -SELECT count(*) FROM source WHERE toYYYYMM(dt_32) = 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(dt_32) <> 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(dt_32) < 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(dt_32) <= 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(dt_32) > 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(dt_32) >= 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(dt_32) = 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(dt_32) <> 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(dt_32) < 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(dt_32) <= 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(dt_32) > 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(dt_32) >= 2023 SETTINGS allow_experimental_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(dt_32) = 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(dt_32) <> 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(dt_32) < 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(dt_32) <= 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(dt_32) > 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(dt_32) >= 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(dt_32) = 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(dt_32) <> 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(dt_32) < 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(dt_32) <= 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(dt_32) > 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(dt_32) >= 2023 SETTINGS enable_analyzer=1; SELECT 'DateTime64'; SELECT count(*) FROM source WHERE toYYYYMM(ts_64) = 202312; @@ -124,16 +124,16 @@ SELECT count(*) FROM source WHERE toYear(ts_64) < 2023; SELECT count(*) FROM source WHERE toYear(ts_64) <= 2023; SELECT count(*) FROM source WHERE toYear(ts_64) > 2023; SELECT count(*) FROM source WHERE toYear(ts_64) >= 2023; -SELECT count(*) FROM source WHERE toYYYYMM(ts_64) = 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(ts_64) <> 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(ts_64) < 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(ts_64) <= 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(ts_64) > 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYYYYMM(ts_64) >= 202312 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(ts_64) = 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(ts_64) <> 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(ts_64) < 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(ts_64) <= 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(ts_64) > 2023 SETTINGS allow_experimental_analyzer=1; -SELECT count(*) FROM source WHERE toYear(ts_64) >= 2023 SETTINGS allow_experimental_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(ts_64) = 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(ts_64) <> 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(ts_64) < 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(ts_64) <= 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(ts_64) > 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYYYYMM(ts_64) >= 202312 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(ts_64) = 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(ts_64) <> 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(ts_64) < 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(ts_64) <= 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(ts_64) > 2023 SETTINGS enable_analyzer=1; +SELECT count(*) FROM source WHERE toYear(ts_64) >= 2023 SETTINGS enable_analyzer=1; DROP TABLE source; diff --git a/tests/queries/0_stateless/02783_parallel_replicas_trivial_count_optimization.sh b/tests/queries/0_stateless/02783_parallel_replicas_trivial_count_optimization.sh index 20b3efedd49..bf7170fd7fb 100755 --- a/tests/queries/0_stateless/02783_parallel_replicas_trivial_count_optimization.sh +++ b/tests/queries/0_stateless/02783_parallel_replicas_trivial_count_optimization.sh @@ -34,7 +34,7 @@ function run_query_with_pure_parallel_replicas () { --prefer_localhost_replica 1 \ --cluster_for_parallel_replicas 'test_cluster_one_shard_three_replicas_localhost' \ --allow_experimental_parallel_reading_from_replicas 1 \ - --allow_experimental_analyzer 0 + --enable_analyzer 0 $CLICKHOUSE_CLIENT \ --query "$2" \ @@ -43,7 +43,7 @@ function run_query_with_pure_parallel_replicas () { --prefer_localhost_replica 1 \ --cluster_for_parallel_replicas 'test_cluster_one_shard_three_replicas_localhost' \ --allow_experimental_parallel_reading_from_replicas 1 \ - --allow_experimental_analyzer 1 + --enable_analyzer 1 } function run_query_with_custom_key_parallel_replicas () { @@ -58,7 +58,7 @@ function run_query_with_custom_key_parallel_replicas () { --max_parallel_replicas 3 \ --parallel_replicas_custom_key_filter_type 'default' \ --parallel_replicas_custom_key "$2" \ - --allow_experimental_analyzer 0 + --enable_analyzer 0 $CLICKHOUSE_CLIENT \ --query "$2" \ @@ -66,7 +66,7 @@ function run_query_with_custom_key_parallel_replicas () { --max_parallel_replicas 3 \ --parallel_replicas_custom_key_filter_type 'default' \ --parallel_replicas_custom_key "$2" \ - --allow_experimental_analyzer 1 + --enable_analyzer 1 } $CLICKHOUSE_CLIENT --query " diff --git a/tests/queries/0_stateless/02784_move_all_conditions_to_prewhere_analyzer_asan.sql b/tests/queries/0_stateless/02784_move_all_conditions_to_prewhere_analyzer_asan.sql index 44b9ce4fdc1..3766e5b0c8f 100644 --- a/tests/queries/0_stateless/02784_move_all_conditions_to_prewhere_analyzer_asan.sql +++ b/tests/queries/0_stateless/02784_move_all_conditions_to_prewhere_analyzer_asan.sql @@ -4,7 +4,7 @@ CREATE TABLE t_02784 (c1 UInt64, c2 UInt64) ENGINE=MergeTree() ORDER BY c1 SETTI INSERT INTO t_02784 SELECT number, number FROM numbers(1); -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SET move_all_conditions_to_prewhere=1; SELECT c1, c2 FROM t_02784 WHERE c1 = 0 AND c2 = 0; diff --git a/tests/queries/0_stateless/02785_date_predicate_optimizations_ast_query_tree_rewrite.reference b/tests/queries/0_stateless/02785_date_predicate_optimizations_ast_query_tree_rewrite.reference index fca48238778..c2c77a4aa31 100644 --- a/tests/queries/0_stateless/02785_date_predicate_optimizations_ast_query_tree_rewrite.reference +++ b/tests/queries/0_stateless/02785_date_predicate_optimizations_ast_query_tree_rewrite.reference @@ -39,7 +39,7 @@ QUERY id: 0 LIST id: 22, nodes: 2 COLUMN id: 19, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 23, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE ((date1 < \'1993-01-01\') OR (date1 >= \'1994-01-01\')) AND ((id >= 1) AND (id <= 3)) @@ -81,7 +81,7 @@ QUERY id: 0 LIST id: 22, nodes: 2 COLUMN id: 19, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 23, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE (date1 < \'1993-01-01\') AND ((id >= 1) AND (id <= 3)) @@ -115,7 +115,7 @@ QUERY id: 0 LIST id: 17, nodes: 2 COLUMN id: 14, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 18, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE (date1 >= \'1994-01-01\') AND ((id >= 1) AND (id <= 3)) @@ -149,7 +149,7 @@ QUERY id: 0 LIST id: 17, nodes: 2 COLUMN id: 14, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 18, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE (date1 < \'1994-01-01\') AND ((id >= 1) AND (id <= 3)) @@ -183,7 +183,7 @@ QUERY id: 0 LIST id: 17, nodes: 2 COLUMN id: 14, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 18, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE (date1 >= \'1993-01-01\') AND ((id >= 1) AND (id <= 3)) @@ -217,7 +217,7 @@ QUERY id: 0 LIST id: 17, nodes: 2 COLUMN id: 14, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 18, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE ((date1 >= \'1993-01-01\') AND (date1 < \'1998-01-01\')) AND ((id >= 1) AND (id <= 3)) @@ -259,7 +259,7 @@ QUERY id: 0 LIST id: 22, nodes: 2 COLUMN id: 19, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 23, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE (((date1 >= \'1993-01-01\') AND (date1 < \'1994-01-01\')) OR ((date1 >= \'1994-01-01\') AND (date1 < \'1995-01-01\'))) AND ((id >= 1) AND (id <= 3)) @@ -317,7 +317,7 @@ QUERY id: 0 LIST id: 32, nodes: 2 COLUMN id: 29, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 33, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1, toYear(date1) AS year1 @@ -366,7 +366,7 @@ QUERY id: 0 LIST id: 24, nodes: 2 COLUMN id: 21, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 25, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE (date1 < \'1993-01-01\') AND ((id >= 1) AND (id <= 3)) @@ -400,7 +400,7 @@ QUERY id: 0 LIST id: 17, nodes: 2 COLUMN id: 14, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 18, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t PREWHERE (date1 >= \'1993-01-01\') AND (date1 < \'1994-01-01\') @@ -441,7 +441,7 @@ QUERY id: 0 LIST id: 20, nodes: 2 COLUMN id: 17, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 21, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE ((id >= 1) AND (id <= 3)) AND ((date1 >= \'1993-01-01\') AND (date1 < \'1994-01-01\')) @@ -481,7 +481,7 @@ QUERY id: 0 LIST id: 20, nodes: 2 COLUMN id: 17, column_name: date1, result_type: Date, source_id: 3 CONSTANT id: 21, constant_value: \'1994-01-01\', constant_value_type: String - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE (toYYYYMM(date1) = 199300) AND ((id >= 1) AND (id <= 3)) @@ -518,7 +518,7 @@ QUERY id: 0 LIST id: 19, nodes: 2 COLUMN id: 16, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 20, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE (toYYYYMM(date1) = 199313) AND ((id >= 1) AND (id <= 3)) @@ -555,7 +555,7 @@ QUERY id: 0 LIST id: 19, nodes: 2 COLUMN id: 16, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 20, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE ((date1 >= \'1993-12-01\') AND (date1 < \'1994-01-01\')) AND ((id >= 1) AND (id <= 3)) @@ -597,7 +597,7 @@ QUERY id: 0 LIST id: 22, nodes: 2 COLUMN id: 19, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 23, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE ((date1 >= \'1992-03-01\') AND (date1 < \'1992-04-01\')) AND ((id >= 1) AND (id <= 3)) @@ -639,7 +639,7 @@ QUERY id: 0 LIST id: 22, nodes: 2 COLUMN id: 19, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 23, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE ((date1 < \'1992-03-01\') OR (date1 >= \'1992-04-01\')) AND ((id >= 1) AND (id <= 3)) @@ -681,7 +681,7 @@ QUERY id: 0 LIST id: 22, nodes: 2 COLUMN id: 19, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 23, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE (date1 < \'1992-03-01\') AND ((id >= 1) AND (id <= 3)) @@ -715,7 +715,7 @@ QUERY id: 0 LIST id: 17, nodes: 2 COLUMN id: 14, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 18, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE (date1 >= \'1992-04-01\') AND ((id >= 1) AND (id <= 3)) @@ -749,7 +749,7 @@ QUERY id: 0 LIST id: 17, nodes: 2 COLUMN id: 14, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 18, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE (date1 < \'1992-04-01\') AND ((id >= 1) AND (id <= 3)) @@ -783,7 +783,7 @@ QUERY id: 0 LIST id: 17, nodes: 2 COLUMN id: 14, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 18, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE (date1 >= \'1992-03-01\') AND ((id >= 1) AND (id <= 3)) @@ -817,7 +817,7 @@ QUERY id: 0 LIST id: 17, nodes: 2 COLUMN id: 14, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 18, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date_t WHERE ((date1 >= \'1992-03-01\') OR ((date1 >= \'1993-01-01\') AND (date1 < \'1994-01-01\'))) AND ((id >= 1) AND (id <= 3)) @@ -867,7 +867,7 @@ QUERY id: 0 LIST id: 27, nodes: 2 COLUMN id: 24, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 28, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM datetime_t WHERE ((date1 >= \'1993-01-01 00:00:00\') AND (date1 < \'1994-01-01 00:00:00\')) AND ((id >= 1) AND (id <= 3)) @@ -909,7 +909,7 @@ QUERY id: 0 LIST id: 22, nodes: 2 COLUMN id: 19, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 23, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM datetime_t WHERE ((date1 >= \'1993-12-01 00:00:00\') AND (date1 < \'1994-01-01 00:00:00\')) AND ((id >= 1) AND (id <= 3)) @@ -951,7 +951,7 @@ QUERY id: 0 LIST id: 22, nodes: 2 COLUMN id: 19, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 23, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date32_t WHERE ((date1 >= \'1993-01-01\') AND (date1 < \'1994-01-01\')) AND ((id >= 1) AND (id <= 3)) @@ -993,7 +993,7 @@ QUERY id: 0 LIST id: 22, nodes: 2 COLUMN id: 19, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 23, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM date32_t WHERE ((date1 >= \'1993-12-01\') AND (date1 < \'1994-01-01\')) AND ((id >= 1) AND (id <= 3)) @@ -1035,7 +1035,7 @@ QUERY id: 0 LIST id: 22, nodes: 2 COLUMN id: 19, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 23, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM datetime64_t WHERE ((date1 >= \'1993-01-01 00:00:00\') AND (date1 < \'1994-01-01 00:00:00\')) AND ((id >= 1) AND (id <= 3)) @@ -1077,7 +1077,7 @@ QUERY id: 0 LIST id: 22, nodes: 2 COLUMN id: 19, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 23, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 SELECT value1 FROM datetime64_t WHERE ((date1 >= \'1993-12-01 00:00:00\') AND (date1 < \'1994-01-01 00:00:00\')) AND ((id >= 1) AND (id <= 3)) @@ -1119,4 +1119,4 @@ QUERY id: 0 LIST id: 22, nodes: 2 COLUMN id: 19, column_name: id, result_type: UInt32, source_id: 3 CONSTANT id: 23, constant_value: UInt64_3, constant_value_type: UInt8 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 diff --git a/tests/queries/0_stateless/02785_date_predicate_optimizations_ast_query_tree_rewrite.sql b/tests/queries/0_stateless/02785_date_predicate_optimizations_ast_query_tree_rewrite.sql index 9cc8dd74e5d..5ff62cb4bb3 100644 --- a/tests/queries/0_stateless/02785_date_predicate_optimizations_ast_query_tree_rewrite.sql +++ b/tests/queries/0_stateless/02785_date_predicate_optimizations_ast_query_tree_rewrite.sql @@ -2,74 +2,74 @@ DROP TABLE IF EXISTS date_t; CREATE TABLE date_t (id UInt32, value1 String, date1 Date) ENGINE ReplacingMergeTree() ORDER BY id; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYear(date1) = 1993 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYear(date1) = 1993 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYear(date1) = 1993 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYear(date1) <> 1993 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYear(date1) <> 1993 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYear(date1) <> 1993 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYear(date1) < 1993 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYear(date1) < 1993 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYear(date1) < 1993 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYear(date1) > 1993 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYear(date1) > 1993 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYear(date1) > 1993 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYear(date1) <= 1993 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYear(date1) <= 1993 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYear(date1) <= 1993 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYear(date1) >= 1993 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYear(date1) >= 1993 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYear(date1) >= 1993 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYear(date1) BETWEEN 1993 AND 1997 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYear(date1) BETWEEN 1993 AND 1997 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYear(date1) BETWEEN 1993 AND 1997 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE (toYear(date1) = 1993 OR toYear(date1) = 1994) AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE (toYear(date1) = 1993 OR toYear(date1) = 1994) AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE (toYear(date1) = 1993 OR toYear(date1) = 1994) AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1, toYear(date1) as year1 FROM date_t WHERE year1 = 1993 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1, toYear(date1) as year1 FROM date_t WHERE year1 = 1993 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1, toYear(date1) as year1 FROM date_t WHERE year1 = 1993 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE 1993 > toYear(date1) AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE 1993 > toYear(date1) AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE 1993 > toYear(date1) AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t PREWHERE toYear(date1) = 1993 WHERE id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t PREWHERE toYear(date1) = 1993 WHERE id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t PREWHERE toYear(date1) = 1993 WHERE id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE id BETWEEN 1 AND 3 HAVING toYear(date1) = 1993; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE id BETWEEN 1 AND 3 HAVING toYear(date1) = 1993 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE id BETWEEN 1 AND 3 HAVING toYear(date1) = 1993 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYYYYMM(date1) = 199300 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) = 199300 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) = 199300 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYYYYMM(date1) = 199313 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) = 199313 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) = 199313 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYYYYMM(date1) = 199312 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) = 199312 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) = 199312 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYYYYMM(date1) = 199203 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) = 199203 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) = 199203 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYYYYMM(date1) <> 199203 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) <> 199203 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) <> 199203 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYYYYMM(date1) < 199203 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) < 199203 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) < 199203 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYYYYMM(date1) > 199203 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) > 199203 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) > 199203 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYYYYMM(date1) <= 199203 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) <= 199203 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) <= 199203 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE toYYYYMM(date1) >= 199203 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) >= 199203 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE toYYYYMM(date1) >= 199203 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date_t WHERE (toYYYYMM(date1) >= 199203 OR toYear(date1) = 1993) AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE (toYYYYMM(date1) >= 199203 OR toYear(date1) = 1993) AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date_t WHERE (toYYYYMM(date1) >= 199203 OR toYear(date1) = 1993) AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; DROP TABLE date_t; DROP TABLE IF EXISTS datetime_t; CREATE TABLE datetime_t (id UInt32, value1 String, date1 Datetime) ENGINE ReplacingMergeTree() ORDER BY id; EXPLAIN SYNTAX SELECT value1 FROM datetime_t WHERE toYear(date1) = 1993 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM datetime_t WHERE toYear(date1) = 1993 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM datetime_t WHERE toYear(date1) = 1993 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM datetime_t WHERE toYYYYMM(date1) = 199312 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM datetime_t WHERE toYYYYMM(date1) = 199312 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM datetime_t WHERE toYYYYMM(date1) = 199312 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; DROP TABLE datetime_t; DROP TABLE IF EXISTS date32_t; CREATE TABLE date32_t (id UInt32, value1 String, date1 Date32) ENGINE ReplacingMergeTree() ORDER BY id; EXPLAIN SYNTAX SELECT value1 FROM date32_t WHERE toYear(date1) = 1993 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date32_t WHERE toYear(date1) = 1993 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date32_t WHERE toYear(date1) = 1993 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM date32_t WHERE toYYYYMM(date1) = 199312 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date32_t WHERE toYYYYMM(date1) = 199312 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM date32_t WHERE toYYYYMM(date1) = 199312 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; DROP TABLE date32_t; DROP TABLE IF EXISTS datetime64_t; CREATE TABLE datetime64_t (id UInt32, value1 String, date1 Datetime64) ENGINE ReplacingMergeTree() ORDER BY id; EXPLAIN SYNTAX SELECT value1 FROM datetime64_t WHERE toYear(date1) = 1993 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM datetime64_t WHERE toYear(date1) = 1993 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM datetime64_t WHERE toYear(date1) = 1993 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; EXPLAIN SYNTAX SELECT value1 FROM datetime64_t WHERE toYYYYMM(date1) = 199312 AND id BETWEEN 1 AND 3; -EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM datetime64_t WHERE toYYYYMM(date1) = 199312 AND id BETWEEN 1 AND 3 SETTINGS allow_experimental_analyzer=1; +EXPLAIN QUERY TREE run_passes=1 SELECT value1 FROM datetime64_t WHERE toYYYYMM(date1) = 199312 AND id BETWEEN 1 AND 3 SETTINGS enable_analyzer=1; DROP TABLE datetime64_t; 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 751b2798243..5a37d51233d 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 @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: no-fasttest, no-random-settings, no-replicated-database +# Tags: no-fasttest, no-random-settings, no-replicated-database, no-distributed-cache CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -14,43 +14,59 @@ SETTINGS disk = 's3_disk', min_bytes_for_wide_part = 0; INSERT INTO test_s3 SELECT number, number FROM numbers_mt(1e7); " -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" -${CLICKHOUSE_CLIENT} -nm --query " -SELECT ProfileEvents['DiskConnectionsPreserved'] > 0 -FROM system.query_log -WHERE type = 'QueryFinish' - AND current_database = currentDatabase() - AND query_id='$query_id'; -" + +# This (reusing connections from the pool) is not guaranteed to always happen, +# (due to random time difference between the queries and random activity in parallel) +# but should happen most of the time. + +while true +do + query="SELECT a, b FROM test_s3" + query_id=$(${CLICKHOUSE_CLIENT} --query "select queryID() from ($query) limit 1" 2>&1) + ${CLICKHOUSE_CLIENT} --query "SYSTEM FLUSH LOGS" + + RES=$(${CLICKHOUSE_CLIENT} -nm --query " + SELECT ProfileEvents['DiskConnectionsPreserved'] > 0 + FROM system.query_log + WHERE type = 'QueryFinish' + AND current_database = currentDatabase() + AND query_id='$query_id'; + ") + + [[ $RES -eq 1 ]] && echo "$RES" && break; +done # Test connection pool in ReadWriteBufferFromHTTP -query_id=$(${CLICKHOUSE_CLIENT} -nq " -create table mut (n int, m int, k int) engine=ReplicatedMergeTree('/test/02441/{database}/mut', '1') order by n; -set insert_keeper_fault_injection_probability=0; -insert into mut values (1, 2, 3), (10, 20, 30); +while true +do + query_id=$(${CLICKHOUSE_CLIENT} -nq " + create table mut (n int, m int, k int) engine=ReplicatedMergeTree('/test/02441/{database}/mut', '1') order by n; + set insert_keeper_fault_injection_probability=0; + insert into mut values (1, 2, 3), (10, 20, 30); -system stop merges mut; -alter table mut delete where n = 10; + system stop merges mut; + alter table mut delete where n = 10; -select queryID() from( - -- a funny way to wait for a MUTATE_PART to be assigned - select sleepEachRow(2) from url('http://localhost:8123/?param_tries={1..10}&query=' || encodeURLComponent( - 'select 1 where ''MUTATE_PART'' not in (select type from system.replication_queue where database=''' || currentDatabase() || ''' and table=''mut'')' - ), 'LineAsString', 's String') - -- queryID() will be returned for each row, since the query above doesn't return anything we need to return a fake row - union all - select 1 -) limit 1 settings max_threads=1; -" 2>&1) -${CLICKHOUSE_CLIENT} --query "SYSTEM FLUSH LOGS" -${CLICKHOUSE_CLIENT} -nm --query " -SELECT ProfileEvents['StorageConnectionsPreserved'] > 0 -FROM system.query_log -WHERE type = 'QueryFinish' - AND current_database = currentDatabase() - AND query_id='$query_id'; -" + select queryID() from( + -- a funny way to wait for a MUTATE_PART to be assigned + select sleepEachRow(2) from url('http://localhost:8123/?param_tries={1..10}&query=' || encodeURLComponent( + 'select 1 where ''MUTATE_PART'' not in (select type from system.replication_queue where database=''' || currentDatabase() || ''' and table=''mut'')' + ), 'LineAsString', 's String') + -- queryID() will be returned for each row, since the query above doesn't return anything we need to return a fake row + union all + select 1 + ) limit 1 settings max_threads=1; + " 2>&1) + ${CLICKHOUSE_CLIENT} --query "SYSTEM FLUSH LOGS" + RES=$(${CLICKHOUSE_CLIENT} -nm --query " + SELECT ProfileEvents['StorageConnectionsPreserved'] > 0 + FROM system.query_log + WHERE type = 'QueryFinish' + AND current_database = currentDatabase() + AND query_id='$query_id'; + ") + + [[ $RES -eq 1 ]] && echo "$RES" && break; +done diff --git a/tests/queries/0_stateless/02792_drop_projection_lwd.sql b/tests/queries/0_stateless/02792_drop_projection_lwd.sql index dcde7dcc600..dad7f7cd028 100644 --- a/tests/queries/0_stateless/02792_drop_projection_lwd.sql +++ b/tests/queries/0_stateless/02792_drop_projection_lwd.sql @@ -7,7 +7,7 @@ CREATE TABLE t_projections_lwd (a UInt32, b UInt32, PROJECTION p (SELECT * ORDER INSERT INTO t_projections_lwd SELECT number, number FROM numbers(100); -- LWD does not work, as expected -DELETE FROM t_projections_lwd WHERE a = 1; -- { serverError NOT_IMPLEMENTED } +DELETE FROM t_projections_lwd WHERE a = 1; -- { serverError SUPPORT_IS_DISABLED } KILL MUTATION WHERE database = currentDatabase() AND table = 't_projections_lwd' SYNC FORMAT Null; -- drop projection diff --git a/tests/queries/0_stateless/02803_remote_cannot_clone_block.sql b/tests/queries/0_stateless/02803_remote_cannot_clone_block.sql index 6d79aa76d18..dd72b990445 100644 --- a/tests/queries/0_stateless/02803_remote_cannot_clone_block.sql +++ b/tests/queries/0_stateless/02803_remote_cannot_clone_block.sql @@ -6,7 +6,7 @@ SELECT * FROM system.numbers LIMIT 10000; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT * FROM diff --git a/tests/queries/0_stateless/02805_distributed_queries_timeouts.sql b/tests/queries/0_stateless/02805_distributed_queries_timeouts.sql index f6bccc99977..98aeac36243 100644 --- a/tests/queries/0_stateless/02805_distributed_queries_timeouts.sql +++ b/tests/queries/0_stateless/02805_distributed_queries_timeouts.sql @@ -1,3 +1,5 @@ +-- Tags: no-fasttest +-- no-fasttest: Timeouts are slow 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; diff --git a/tests/queries/0_stateless/02808_custom_disk_with_user_defined_name.sh b/tests/queries/0_stateless/02808_custom_disk_with_user_defined_name.sh index 333bc1bc25d..63fa60bd548 100755 --- a/tests/queries/0_stateless/02808_custom_disk_with_user_defined_name.sh +++ b/tests/queries/0_stateless/02808_custom_disk_with_user_defined_name.sh @@ -13,7 +13,7 @@ DROP TABLE IF EXISTS test; CREATE TABLE test (a Int32, b String) ENGINE = MergeTree() ORDER BY tuple() SETTINGS disk = disk(name = 's3_disk', type = cache, max_size = '100Ki', path = ${CLICKHOUSE_TEST_UNIQUE_NAME}, disk = s3_disk); -""" 2>&1 | grep -q "Disk with name \`s3_disk\` already exist" && echo 'OK' || echo 'FAIL' +""" 2>&1 | grep -q "Disk \`s3_disk\` already exists and is described by the config" && echo 'OK' || echo 'FAIL' disk_name="${CLICKHOUSE_TEST_UNIQUE_NAME}" diff --git a/tests/queries/0_stateless/02812_bug_with_unused_join_columns.sql b/tests/queries/0_stateless/02812_bug_with_unused_join_columns.sql index 6c801b5b73e..d791b8f3367 100644 --- a/tests/queries/0_stateless/02812_bug_with_unused_join_columns.sql +++ b/tests/queries/0_stateless/02812_bug_with_unused_join_columns.sql @@ -1 +1 @@ -SELECT concat(func.name, comb.name) AS x FROM system.functions AS func JOIN system.aggregate_function_combinators AS comb using name WHERE is_aggregate settings allow_experimental_analyzer=1; +SELECT concat(func.name, comb.name) AS x FROM system.functions AS func JOIN system.aggregate_function_combinators AS comb using name WHERE is_aggregate settings enable_analyzer=1; diff --git a/tests/queries/0_stateless/02815_analyzer_aggregate_functions_of_group_by_keys.sql b/tests/queries/0_stateless/02815_analyzer_aggregate_functions_of_group_by_keys.sql index ca03cbb6f9f..dfb885f5a69 100644 --- a/tests/queries/0_stateless/02815_analyzer_aggregate_functions_of_group_by_keys.sql +++ b/tests/queries/0_stateless/02815_analyzer_aggregate_functions_of_group_by_keys.sql @@ -1,4 +1,4 @@ -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; set optimize_move_functions_out_of_any = 0; SELECT 'set optimize_aggregators_of_group_by_keys = 1'; diff --git a/tests/queries/0_stateless/02815_join_algorithm_setting.sql b/tests/queries/0_stateless/02815_join_algorithm_setting.sql index a8733eebc91..4a5ae784b31 100644 --- a/tests/queries/0_stateless/02815_join_algorithm_setting.sql +++ b/tests/queries/0_stateless/02815_join_algorithm_setting.sql @@ -104,9 +104,9 @@ JOIN ( SELECT k AS key, k + 100 AS key2 FROM t2 ) AS t2 ON t1.key = t2.key OR t1 -- But for CROSS choose `hash` algorithm even though it's not enabled SELECT * FROM ( SELECT number AS key, number * 10 AS key2 FROM numbers_mt(10) ) AS t1 CROSS JOIN ( SELECT k AS key, k + 100 AS key2 FROM t2 ) AS t2 FORMAT Null -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; -- ... (not for old analyzer) SELECT * FROM ( SELECT number AS key, number * 10 AS key2 FROM numbers_mt(10) ) AS t1 CROSS JOIN ( SELECT k AS key, k + 100 AS key2 FROM t2 ) AS t2 FORMAT Null -SETTINGS allow_experimental_analyzer = 0; -- { serverError NOT_IMPLEMENTED } +SETTINGS enable_analyzer = 0; -- { serverError NOT_IMPLEMENTED } diff --git a/tests/queries/0_stateless/02834_analyzer_with_statement_references.sql b/tests/queries/0_stateless/02834_analyzer_with_statement_references.sql index 29ed6e3f0da..ce1eaa7ae77 100644 --- a/tests/queries/0_stateless/02834_analyzer_with_statement_references.sql +++ b/tests/queries/0_stateless/02834_analyzer_with_statement_references.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; WITH test_aliases AS (SELECT number FROM numbers(20)), alias2 AS (SELECT number FROM test_aliases) SELECT number FROM alias2 SETTINGS enable_global_with_statement = 1; diff --git a/tests/queries/0_stateless/02834_remote_session_log.sh b/tests/queries/0_stateless/02834_remote_session_log.sh index 3bedfb6c9ee..0581cb36136 100755 --- a/tests/queries/0_stateless/02834_remote_session_log.sh +++ b/tests/queries/0_stateless/02834_remote_session_log.sh @@ -44,6 +44,16 @@ for interface in 'TCP' 'HTTP' 'MySQL' do LOGIN_COUNT=`${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'LoginSuccess' AND interface = '${interface}'"` CORRESPONDING_LOGOUT_RECORDS_COUNT=`${CLICKHOUSE_CLIENT} -q "SELECT COUNT(*) FROM (SELECT ${SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'LoginSuccess' AND interface = '${interface}' INTERSECT SELECT ${SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'Logout' AND interface = '${interface}')"` + # The client can exit sooner than the server records its disconnection and closes the session. + # When the client disconnects, two processes happen at the same time and are in the race condition: + # - the client application exits and returns control to the shell; + # - the server closes the session and records the logout event to the session log. + # We cannot expect that after the control is returned to the shell, the server records the logout event. + while [ "$LOGIN_COUNT" != "$CORRESPONDING_LOGOUT_RECORDS_COUNT" ] + do + sleep 0.1 + CORRESPONDING_LOGOUT_RECORDS_COUNT=`${CLICKHOUSE_CLIENT} -q "SELECT COUNT(*) FROM (SELECT ${SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'LoginSuccess' AND interface = '${interface}' INTERSECT SELECT ${SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'Logout' AND interface = '${interface}')"` + done if [ "$LOGIN_COUNT" == "$CORRESPONDING_LOGOUT_RECORDS_COUNT" ]; then echo "${interface} Login and logout count is equal" diff --git a/tests/queries/0_stateless/02835_join_step_explain.sql b/tests/queries/0_stateless/02835_join_step_explain.sql index d0475fa14b6..1cdd3684a0b 100644 --- a/tests/queries/0_stateless/02835_join_step_explain.sql +++ b/tests/queries/0_stateless/02835_join_step_explain.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table_1; CREATE TABLE test_table_1 diff --git a/tests/queries/0_stateless/02840_merge__table_or_filter.reference b/tests/queries/0_stateless/02840_merge__table_or_filter.reference index ff5e0865a22..21b54a0121e 100644 --- a/tests/queries/0_stateless/02840_merge__table_or_filter.reference +++ b/tests/queries/0_stateless/02840_merge__table_or_filter.reference @@ -1,38 +1,38 @@ -- { echoOn } -select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v1') settings allow_experimental_analyzer=0, convert_query_to_cnf=0; +select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v1') settings enable_analyzer=0, convert_query_to_cnf=0; v1 1 v1 2 -select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v2') settings allow_experimental_analyzer=0, convert_query_to_cnf=0; +select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v2') settings enable_analyzer=0, convert_query_to_cnf=0; v1 1 v2 2 -select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v3') settings allow_experimental_analyzer=0, convert_query_to_cnf=0; +select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v3') settings enable_analyzer=0, convert_query_to_cnf=0; v1 1 -select _table, key from m where (value = 10 and _table = 'v3') or (value = 20 and _table = 'v3') settings allow_experimental_analyzer=0, convert_query_to_cnf=0; -select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v1') settings allow_experimental_analyzer=0, convert_query_to_cnf=1; +select _table, key from m where (value = 10 and _table = 'v3') or (value = 20 and _table = 'v3') settings enable_analyzer=0, convert_query_to_cnf=0; +select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v1') settings enable_analyzer=0, convert_query_to_cnf=1; v1 1 v1 2 -select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v2') settings allow_experimental_analyzer=0, convert_query_to_cnf=1; +select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v2') settings enable_analyzer=0, convert_query_to_cnf=1; v1 1 v2 2 -select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v3') settings allow_experimental_analyzer=0, convert_query_to_cnf=1; +select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v3') settings enable_analyzer=0, convert_query_to_cnf=1; v1 1 -select _table, key from m where (value = 10 and _table = 'v3') or (value = 20 and _table = 'v3') settings allow_experimental_analyzer=0, convert_query_to_cnf=1; -select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v1') settings allow_experimental_analyzer=1, convert_query_to_cnf=0; +select _table, key from m where (value = 10 and _table = 'v3') or (value = 20 and _table = 'v3') settings enable_analyzer=0, convert_query_to_cnf=1; +select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v1') settings enable_analyzer=1, convert_query_to_cnf=0; v1 1 v1 2 -select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v2') settings allow_experimental_analyzer=1, convert_query_to_cnf=0; +select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v2') settings enable_analyzer=1, convert_query_to_cnf=0; v1 1 v2 2 -select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v3') settings allow_experimental_analyzer=1, convert_query_to_cnf=0; +select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v3') settings enable_analyzer=1, convert_query_to_cnf=0; v1 1 -select _table, key from m where (value = 10 and _table = 'v3') or (value = 20 and _table = 'v3') settings allow_experimental_analyzer=1, convert_query_to_cnf=0; -select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v1') settings allow_experimental_analyzer=1, convert_query_to_cnf=1; +select _table, key from m where (value = 10 and _table = 'v3') or (value = 20 and _table = 'v3') settings enable_analyzer=1, convert_query_to_cnf=0; +select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v1') settings enable_analyzer=1, convert_query_to_cnf=1; v1 1 v1 2 -select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v2') settings allow_experimental_analyzer=1, convert_query_to_cnf=1; +select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v2') settings enable_analyzer=1, convert_query_to_cnf=1; v1 1 v2 2 -select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v3') settings allow_experimental_analyzer=1, convert_query_to_cnf=1; +select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v3') settings enable_analyzer=1, convert_query_to_cnf=1; v1 1 -select _table, key from m where (value = 10 and _table = 'v3') or (value = 20 and _table = 'v3') settings allow_experimental_analyzer=1, convert_query_to_cnf=1; +select _table, key from m where (value = 10 and _table = 'v3') or (value = 20 and _table = 'v3') settings enable_analyzer=1, convert_query_to_cnf=1; diff --git a/tests/queries/0_stateless/02840_merge__table_or_filter.sql.j2 b/tests/queries/0_stateless/02840_merge__table_or_filter.sql.j2 index 286e4545ef7..f7413dc3ee6 100644 --- a/tests/queries/0_stateless/02840_merge__table_or_filter.sql.j2 +++ b/tests/queries/0_stateless/02840_merge__table_or_filter.sql.j2 @@ -27,10 +27,10 @@ select _table, key from m where (value = 10 and _table = 'v3') or (value = 20 an set max_threads=1; -- { echoOn } {% for settings in [ - 'allow_experimental_analyzer=0, convert_query_to_cnf=0', - 'allow_experimental_analyzer=0, convert_query_to_cnf=1', - 'allow_experimental_analyzer=1, convert_query_to_cnf=0', - 'allow_experimental_analyzer=1, convert_query_to_cnf=1' + 'enable_analyzer=0, convert_query_to_cnf=0', + 'enable_analyzer=0, convert_query_to_cnf=1', + 'enable_analyzer=1, convert_query_to_cnf=0', + 'enable_analyzer=1, convert_query_to_cnf=1' ] %} select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v1') settings {{ settings }}; select _table, key from m where (value = 10 and _table = 'v1') or (value = 20 and _table = 'v2') settings {{ settings }}; diff --git a/tests/queries/0_stateless/02841_local_assert.sh b/tests/queries/0_stateless/02841_local_assert.sh index a167c09da1f..dc49007b0f6 100755 --- a/tests/queries/0_stateless/02841_local_assert.sh +++ b/tests/queries/0_stateless/02841_local_assert.sh @@ -7,12 +7,12 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CUR_DIR"/../shell_config.sh echo "create table test (x UInt64) engine=Memory; -insert into test from infile 'data'; -- {clientError BAD_ARGUMENTS}" | $CLICKHOUSE_LOCAL -nm +insert into test from infile 'data'; -- {clientError BAD_ARGUMENTS}" | $CLICKHOUSE_LOCAL -m echo "create table test (x UInt64) engine=Memory; -insert into test from infile 'data';" | $CLICKHOUSE_LOCAL -nm --ignore-error +insert into test from infile 'data';" | $CLICKHOUSE_LOCAL -m --ignore-error echo "create table test (x UInt64) engine=Memory; insert into test from infile 'data'; -- {clientError BAD_ARGUMENTS} -select 1" | $CLICKHOUSE_LOCAL -nm +select 1" | $CLICKHOUSE_LOCAL -m diff --git a/tests/queries/0_stateless/02841_not_ready_set_bug.sh b/tests/queries/0_stateless/02841_not_ready_set_bug.sh index 9b2f3b0698e..556e2f52de2 100755 --- a/tests/queries/0_stateless/02841_not_ready_set_bug.sh +++ b/tests/queries/0_stateless/02841_not_ready_set_bug.sh @@ -11,3 +11,20 @@ $CLICKHOUSE_CLIENT --max_threads=2 --max_result_rows=1 --result_overflow_mode=br $CLICKHOUSE_CLIENT -q "SELECT * FROM system.tables WHERE 1 in (SELECT number from numbers(2)) AND database = currentDatabase() format Null" $CLICKHOUSE_CLIENT -q "SELECT xor(1, 0) FROM system.parts WHERE 1 IN (SELECT 1) FORMAT Null" + +# (Not all of these tests are effective because some of these tables are empty.) +$CLICKHOUSE_CLIENT -nq " + select * from system.columns where table in (select '123'); + select * from system.replicas where database in (select '123'); + select * from system.data_skipping_indices where database in (select '123'); + select * from system.databases where name in (select '123'); + select * from system.mutations where table in (select '123'); + select * from system.part_moves_between_shards where database in (select '123'); + select * from system.replication_queue where database in (select '123'); + select * from system.distribution_queue where database in (select '123'); +" +$CLICKHOUSE_CLIENT -nq " + create table a (x Int8) engine MergeTree order by x; + insert into a values (1); + select * from mergeTreeIndex(currentDatabase(), 'a') where part_name in (select '123'); +" diff --git a/tests/queries/0_stateless/02841_valid_json_and_xml_on_http_exception.sh b/tests/queries/0_stateless/02841_valid_json_and_xml_on_http_exception.sh index c47fe5c7e94..3dda63e1e49 100755 --- a/tests/queries/0_stateless/02841_valid_json_and_xml_on_http_exception.sh +++ b/tests/queries/0_stateless/02841_valid_json_and_xml_on_http_exception.sh @@ -4,7 +4,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -CH_URL_BASE="$CLICKHOUSE_URL&http_write_exception_in_output_format=1&allow_experimental_analyzer=0" +CH_URL_BASE="$CLICKHOUSE_URL&http_write_exception_in_output_format=1&enable_analyzer=0" for wait_end_of_query in 0 1 do diff --git a/tests/queries/0_stateless/02841_with_clause_resolve.sql b/tests/queries/0_stateless/02841_with_clause_resolve.sql index b416446461b..fe94a26110d 100644 --- a/tests/queries/0_stateless/02841_with_clause_resolve.sql +++ b/tests/queries/0_stateless/02841_with_clause_resolve.sql @@ -1,16 +1,16 @@ -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; WITH -- Input 44100 AS sample_frequency , number AS tick , tick / sample_frequency AS time - + -- Delay , (time, wave, delay_, decay, count) -> arraySum(n1 -> wave(time - delay_ * n1), range(count)) AS delay , delay(time, (time -> 0.5), 0.2, 0.5, 5) AS kick - + SELECT kick @@ -23,29 +23,29 @@ WITH 44100 AS sample_frequency , number AS tick , tick / sample_frequency AS time - + -- Output control , 1 AS master_volume , level -> least(1.0, greatest(-1.0, level)) AS clamp , level -> (clamp(level) * 0x7FFF * master_volume)::Int16 AS output , x -> (x, x) AS mono - + -- Basic waves , time -> sin(time * 2 * pi()) AS sine_wave , time -> time::UInt64 % 2 * 2 - 1 AS square_wave , time -> (time - floor(time)) * 2 - 1 AS sawtooth_wave , time -> abs(sawtooth_wave(time)) * 2 - 1 AS triangle_wave - + -- Helpers , (from, to, wave, time) -> from + ((wave(time) + 1) / 2) * (to - from) AS lfo , (from, to, steps, time) -> from + floor((time - floor(time)) * steps) / steps * (to - from) AS step_lfo , (from, to, steps, time) -> exp(step_lfo(log(from), log(to), steps, time)) AS exp_step_lfo - + -- Noise , time -> cityHash64(time) / 0xFFFFFFFFFFFFFFFF AS uniform_noise , time -> erf(uniform_noise(time)) AS white_noise , time -> cityHash64(time) % 2 ? 1 : -1 AS bernoulli_noise - + -- Distortion , (x, amount) -> clamp(x * amount) AS clipping , (x, amount) -> clamp(x > 0 ? pow(x, amount) : -pow(-x, amount)) AS power_distortion @@ -53,10 +53,10 @@ WITH , (time, sample_frequency) -> round(time * sample_frequency) / sample_frequency AS desample , (time, wave, amount) -> (time - floor(time) < (1 - amount)) ? wave(time * (1 - amount)) : 0 AS thin , (time, wave, amount) -> wave(floor(time) + pow(time - floor(time), amount)) AS skew - + -- Combining , (a, b, weight) -> a * (1 - weight) + b * weight AS combine - + -- Envelopes , (time, offset, attack, hold, release) -> time < offset ? 0 @@ -64,7 +64,7 @@ WITH : (time < offset + attack + hold ? 1 : (time < offset + attack + hold + release ? (offset + attack + hold + release - time) / release : 0))) AS envelope - + , (bpm, time, offset, attack, hold, release) -> envelope( time * (bpm / 60) - floor(time * (bpm / 60)), @@ -72,20 +72,20 @@ WITH attack, hold, release) AS running_envelope - + -- Sequencers , (sequence, time) -> sequence[1 + time::UInt64 % length(sequence)] AS sequencer - + -- Delay , (time, wave, delay, decay, count) -> arraySum(n -> wave(time - delay * n) * pow(decay, n), range(count)) AS delay - - + + , delay(time, (time -> power_distortion(sine_wave(time * 80 + sine_wave(time * 2)), lfo(0.5, 1, sine_wave, time / 16)) * running_envelope(60, time, 0, 0.0, 0.01, 0.1)), 0.2, 0.5, 5) AS kick - + SELECT - + (output( kick + delay(time, (time -> @@ -95,7 +95,7 @@ SELECT lfo(1, 0.75, triangle_wave, time / 8))), 0.2, 0.5, 10) * lfo(0.5, 1, triangle_wave, time / 7) - + + delay(time, (time -> power_distortion( sine_wave(time * sequencer([50, 100, 200, 400], time / 2) + 1 * sine_wave(time * sequencer([50, 100, 200], time / 4) + 1/4)) @@ -103,16 +103,16 @@ SELECT lfo(1, 0.75, triangle_wave, time / 8))), 0.2, 0.5, 10) * lfo(0.5, 1, triangle_wave, 16 + time / 11) - + + delay(time, (time -> white_noise(time) * running_envelope(60, time, 0.75, 0.01, 0.01, 0.1)), 0.2, 0.5, 10) * lfo(0.5, 1, triangle_wave, 24 + time / 13) - + + sine_wave(time * 100 + 1 * sine_wave(time * 10 + 1/4)) * running_envelope(120, time, 0, 0.01, 0.01, 0.1) ), - + output( kick + delay(time + 0.01, (time -> @@ -122,7 +122,7 @@ SELECT lfo(1, 0.75, triangle_wave, time / 8))), 0.2, 0.5, 10) * lfo(0.5, 1, triangle_wave, time / 7) - + + delay(time - 0.01, (time -> power_distortion( sine_wave(time * sequencer([50, 100, 200, 400], time / 2) + 1 * sine_wave(time * sequencer([50, 100, 200], time / 4) + 1/4)) @@ -130,12 +130,12 @@ SELECT lfo(1, 0.75, triangle_wave, time / 8))), 0.2, 0.5, 10) * lfo(0.5, 1, triangle_wave, 16 + time / 11) - + + delay(time + 0.005, (time -> white_noise(time) * running_envelope(60, time, 0.75, 0.01, 0.01, 0.1)), 0.2, 0.5, 10) * lfo(0.5, 1, triangle_wave, 24 + time / 13) )) - + FROM system.numbers LIMIT 10; diff --git a/tests/queries/0_stateless/02845_threads_count_in_distributed_queries.sql.j2 b/tests/queries/0_stateless/02845_threads_count_in_distributed_queries.sql.j2 index ffdd4e3400e..26b858d8c00 100644 --- a/tests/queries/0_stateless/02845_threads_count_in_distributed_queries.sql.j2 +++ b/tests/queries/0_stateless/02845_threads_count_in_distributed_queries.sql.j2 @@ -1,5 +1,9 @@ +-- Tags: no-parallel, no-fasttest +-- ^ because query_thread_log is not guaranteed to be written under high load +-- (when the queue is full, events are silently dropped) + -- enforce some defaults to be sure that the env settings will not affect the test -SET max_threads=5, async_socket_for_remote=1, prefer_localhost_replica=1, optimize_read_in_order=1, load_marks_asynchronously=0, local_filesystem_read_method='pread', remote_filesystem_read_method='read'; +SET max_threads=5, async_socket_for_remote=1, prefer_localhost_replica=1, optimize_read_in_order=1, load_marks_asynchronously=0, local_filesystem_read_method='pread', remote_filesystem_read_method='read', trace_profile_events=0; -- we use query_thread_log to check peak thread usage -- after https://github.com/ClickHouse/ClickHouse/issues/53417 there is a simpler way to check it diff --git a/tests/queries/0_stateless/02861_filter_pushdown_const_bug.sql b/tests/queries/0_stateless/02861_filter_pushdown_const_bug.sql index ea52df5d4b4..c9cdab0cea2 100644 --- a/tests/queries/0_stateless/02861_filter_pushdown_const_bug.sql +++ b/tests/queries/0_stateless/02861_filter_pushdown_const_bug.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS t1; diff --git a/tests/queries/0_stateless/02864_statistics_bugs.reference b/tests/queries/0_stateless/02864_statistics_bugs.reference new file mode 100644 index 00000000000..a7eeae9def6 --- /dev/null +++ b/tests/queries/0_stateless/02864_statistics_bugs.reference @@ -0,0 +1,3 @@ +10 +11 +0 diff --git a/tests/queries/0_stateless/02864_statistics_bugs.sql b/tests/queries/0_stateless/02864_statistics_bugs.sql new file mode 100644 index 00000000000..01bbe221b0f --- /dev/null +++ b/tests/queries/0_stateless/02864_statistics_bugs.sql @@ -0,0 +1,27 @@ +SET allow_experimental_statistics = 1; +SET allow_statistics_optimize = 1; +SET mutations_sync = 1; + +DROP TABLE IF EXISTS bug_67742; +CREATE TABLE bug_67742 (a Float64 STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); +INSERT INTO bug_67742 SELECT number FROM system.numbers LIMIT 10000; +SELECT count(*) FROM bug_67742 WHERE a < '10'; +DROP TABLE bug_67742; + +DROP TABLE IF EXISTS bug_67742; +CREATE TABLE bug_67742 (a Int32 STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); +INSERT INTO bug_67742 SELECT number FROM system.numbers LIMIT 10000; +SELECT count(*) FROM bug_67742 WHERE a < '10.5'; -- { serverError TYPE_MISMATCH } +DROP TABLE bug_67742; + +DROP TABLE IF EXISTS bug_67742; +CREATE TABLE bug_67742 (a Int32 STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); +INSERT INTO bug_67742 SELECT number FROM system.numbers LIMIT 10000; +SELECT count(*) FROM bug_67742 WHERE a < 10.5; +DROP TABLE bug_67742; + +DROP TABLE IF EXISTS bug_67742; +CREATE TABLE bug_67742 (a Int16 STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); +INSERT INTO bug_67742 SELECT number FROM system.numbers LIMIT 10000; +SELECT count(*) FROM bug_67742 WHERE a < '9999999999999999999999999'; +DROP TABLE bug_67742; diff --git a/tests/queries/0_stateless/02864_statistics_count_min_sketch.reference b/tests/queries/0_stateless/02864_statistics_count_min_sketch.reference deleted file mode 100644 index 02c41656a36..00000000000 --- a/tests/queries/0_stateless/02864_statistics_count_min_sketch.reference +++ /dev/null @@ -1,14 +0,0 @@ -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 deleted file mode 100644 index c730aa7b4a7..00000000000 --- a/tests/queries/0_stateless/02864_statistics_count_min_sketch.sql +++ /dev/null @@ -1,70 +0,0 @@ --- 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 index a7ff5caa0b0..0e453b0ee8a 100644 --- a/tests/queries/0_stateless/02864_statistics_ddl.reference +++ b/tests/queries/0_stateless/02864_statistics_ddl.reference @@ -1,31 +1,6 @@ -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 +CREATE TABLE default.tab\n(\n `f64` Float64,\n `f64_tdigest` Float64 STATISTICS(tdigest),\n `f32` Float32,\n `s` String,\n `a` Array(Float64)\n)\nENGINE = MergeTree\nORDER BY tuple()\nSETTINGS index_granularity = 8192 +CREATE TABLE default.tab\n(\n `f64` Float64 STATISTICS(tdigest, uniq),\n `f64_tdigest` Float64 STATISTICS(tdigest),\n `f32` Float32 STATISTICS(tdigest, uniq),\n `s` String,\n `a` Array(Float64)\n)\nENGINE = MergeTree\nORDER BY tuple()\nSETTINGS index_granularity = 8192 +CREATE TABLE default.tab\n(\n `f64` Float64 STATISTICS(tdigest, uniq),\n `f64_tdigest` Float64 STATISTICS(tdigest),\n `f32` Float32 STATISTICS(tdigest, uniq),\n `s` String,\n `a` Array(Float64)\n)\nENGINE = MergeTree\nORDER BY tuple()\nSETTINGS index_granularity = 8192 +CREATE TABLE default.tab\n(\n `f64` Float64 STATISTICS(tdigest, uniq),\n `f64_tdigest` Float64 STATISTICS(tdigest),\n `f32` Float32 STATISTICS(tdigest, uniq),\n `s` String,\n `a` Array(Float64)\n)\nENGINE = MergeTree\nORDER BY tuple()\nSETTINGS index_granularity = 8192 +CREATE TABLE default.tab\n(\n `f64` Float64 STATISTICS(tdigest, uniq),\n `f64_tdigest` Float64 STATISTICS(tdigest),\n `f32` Float32 STATISTICS(tdigest, uniq),\n `s` String,\n `a` Array(Float64)\n)\nENGINE = MergeTree\nORDER BY tuple()\nSETTINGS index_granularity = 8192 +CREATE TABLE default.tab\n(\n `f64` Float64,\n `f64_tdigest` Float64 STATISTICS(tdigest),\n `f32` Float32,\n `s` String,\n `a` Array(Float64)\n)\nENGINE = MergeTree\nORDER BY tuple()\nSETTINGS index_granularity = 8192 diff --git a/tests/queries/0_stateless/02864_statistics_ddl.sql b/tests/queries/0_stateless/02864_statistics_ddl.sql index fe612efe2ac..32b56a842b7 100644 --- a/tests/queries/0_stateless/02864_statistics_ddl.sql +++ b/tests/queries/0_stateless/02864_statistics_ddl.sql @@ -1,59 +1,195 @@ --- Tests that various DDL statements create/drop/materialize statistics +-- Tags: no-fasttest +-- no-fasttest: 'count_min' sketches need a 3rd party library + +-- Tests that DDL statements which create / drop / materialize statistics + +SET mutations_sync = 1; DROP TABLE IF EXISTS tab; +-- Error case: Can't create statistics when allow_experimental_statistics = 0 +CREATE TABLE tab (col Float64 STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); -- { serverError INCORRECT_QUERY } + SET allow_experimental_statistics = 1; -SET allow_statistics_optimize = 1; + +-- Error case: Unknown statistics types are rejected +CREATE TABLE tab (col Float64 STATISTICS(no_statistics_type)) Engine = MergeTree() ORDER BY tuple(); -- { serverError INCORRECT_QUERY } + +-- Error case: The same statistics type can't exist more than once on a column +CREATE TABLE tab (col Float64 STATISTICS(tdigest, tdigest)) Engine = MergeTree() ORDER BY tuple(); -- { serverError INCORRECT_QUERY } + +SET allow_suspicious_low_cardinality_types = 1; + +-- Statistics can only be created on columns of specific data types (depending on the statistics kind), (*) + +-- tdigest requires data_type.isValueRepresentedByInteger +-- These types work: +CREATE TABLE tab (col UInt8 STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col UInt256 STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Float32 STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Decimal32(3) STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Date STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Date32 STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col DateTime STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col DateTime64 STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Enum('hello', 'world') STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col IPv4 STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Nullable(UInt8) STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col LowCardinality(UInt8) STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col LowCardinality(Nullable(UInt8)) STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +-- These types don't work: +CREATE TABLE tab (col String STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col FixedString(1) STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col Array(Float64) STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col Tuple(Float64, Float64) STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col Map(UInt64, UInt64) STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col UUID STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col IPv6 STATISTICS(tdigest)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } + +-- uniq requires data_type.isValueRepresentedByInteger +-- These types work: +CREATE TABLE tab (col UInt8 STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col UInt256 STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Float32 STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Decimal32(3) STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Date STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Date32 STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col DateTime STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col DateTime64 STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Enum('hello', 'world') STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col IPv4 STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Nullable(UInt8) STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col LowCardinality(UInt8) STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col LowCardinality(Nullable(UInt8)) STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +-- These types don't work: +CREATE TABLE tab (col String STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col FixedString(1) STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col Array(Float64) STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col Tuple(Float64, Float64) STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col Map(UInt64, UInt64) STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col UUID STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col IPv6 STATISTICS(uniq)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } + +-- count_min requires data_type.isValueRepresentedByInteger or data_type = (Fixed)String +-- These types work: +CREATE TABLE tab (col UInt8 STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col UInt256 STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Float32 STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Decimal32(3) STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Date STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Date32 STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col DateTime STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col DateTime64 STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Enum('hello', 'world') STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col IPv4 STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col Nullable(UInt8) STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col LowCardinality(UInt8) STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col LowCardinality(Nullable(UInt8)) STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col String STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +CREATE TABLE tab (col FixedString(1) STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); DROP TABLE tab; +-- These types don't work: +CREATE TABLE tab (col Array(Float64) STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col Tuple(Float64, Float64) STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col Map(UInt64, UInt64) STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col UUID STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } +CREATE TABLE tab (col IPv6 STATISTICS(count_min)) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } + +-- CREATE TABLE was easy, ALTER is more fun 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; + f64 Float64, + f64_tdigest Float64 STATISTICS(tdigest), + f32 Float32, + s String, + a Array(Float64) +) +Engine = MergeTree() +ORDER BY tuple(); +-- Error case: Unknown statistics types are rejected +-- (relevant for ADD and MODIFY) +ALTER TABLE tab ADD STATISTICS f64 TYPE no_statistics_type; -- { serverError INCORRECT_QUERY } +ALTER TABLE tab ADD STATISTICS IF NOT EXISTS f64 TYPE no_statistics_type; -- { serverError INCORRECT_QUERY } +ALTER TABLE tab MODIFY STATISTICS f64 TYPE no_statistics_type; -- { serverError INCORRECT_QUERY } +-- for some reason, ALTER TABLE tab MODIFY STATISTICS IF EXISTS is not supported + +-- Error case: The same statistics type can't exist more than once on a column +-- (relevant for ADD and MODIFY) +-- Create the same statistics object twice +ALTER TABLE tab ADD STATISTICS f64 TYPE tdigest, tdigest; -- { serverError INCORRECT_QUERY } +ALTER TABLE tab ADD STATISTICS IF NOT EXISTS f64 TYPE tdigest, tdigest; -- { serverError INCORRECT_QUERY } +ALTER TABLE tab MODIFY STATISTICS f64 TYPE tdigest, tdigest; -- { serverError INCORRECT_QUERY } +-- Create an statistics which exists already +ALTER TABLE tab ADD STATISTICS f64_tdigest TYPE tdigest; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab ADD STATISTICS IF NOT EXISTS f64_tdigest TYPE tdigest; -- no-op +ALTER TABLE tab MODIFY STATISTICS f64_tdigest TYPE tdigest; -- no-op + +-- Error case: Column does not exist +-- (relevant for ADD, MODIFY, DROP, CLEAR, and MATERIALIZE) +-- Note that the results are unfortunately quite inconsistent ... +ALTER TABLE tab ADD STATISTICS no_such_column TYPE tdigest; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab ADD STATISTICS IF NOT EXISTS no_such_column TYPE tdigest; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab MODIFY STATISTICS no_such_column TYPE tdigest; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab DROP STATISTICS no_such_column; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab DROP STATISTICS IF EXISTS no_such_column; -- no-op +ALTER TABLE tab CLEAR STATISTICS no_such_column; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab CLEAR STATISTICS IF EXISTS no_such_column; -- no-op +ALTER TABLE tab MATERIALIZE STATISTICS no_such_column; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab MATERIALIZE STATISTICS IF EXISTS no_such_column; -- { serverError ILLEGAL_STATISTICS } + +-- Error case: Column exists but has no statistics +-- (relevant for MODIFY, DROP, CLEAR, and MATERIALIZE) +-- Note that the results are unfortunately quite inconsistent ... +ALTER TABLE tab MODIFY STATISTICS s TYPE tdigest; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab DROP STATISTICS s; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab DROP STATISTICS IF EXISTS s; -- no-op +ALTER TABLE tab CLEAR STATISTICS s; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab CLEAR STATISTICS IF EXISTS s; -- no-op +ALTER TABLE tab MATERIALIZE STATISTICS s; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab MATERIALIZE STATISTICS IF EXISTS s; -- { serverError ILLEGAL_STATISTICS } + +-- We don't check systematically that that statistics can only be created via ALTER ADD STATISTICS on columns of specific data types (the +-- internal type validation code is tested already above, (*)). Only do a rudimentary check for each statistics type with a data type that +-- works and one that doesn't work. +-- tdigest +-- Works: +ALTER TABLE tab ADD STATISTICS f64 TYPE tdigest; ALTER TABLE tab DROP STATISTICS f64; +ALTER TABLE tab MODIFY STATISTICS f64 TYPE tdigest; ALTER TABLE tab DROP STATISTICS f64; +-- Doesn't work: +ALTER TABLE tab ADD STATISTICS a TYPE tdigest; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab MODIFY STATISTICS a TYPE tdigest; -- { serverError ILLEGAL_STATISTICS } +-- uniq +-- Works: +ALTER TABLE tab ADD STATISTICS f64 TYPE uniq; ALTER TABLE tab DROP STATISTICS f64; +ALTER TABLE tab MODIFY STATISTICS f64 TYPE count_min; ALTER TABLE tab DROP STATISTICS f64; +-- Doesn't work: +ALTER TABLE tab ADD STATISTICS a TYPE uniq; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab MODIFY STATISTICS a TYPE uniq; -- { serverError ILLEGAL_STATISTICS } +-- count_min +-- Works: +ALTER TABLE tab ADD STATISTICS f64 TYPE count_min; ALTER TABLE tab DROP STATISTICS f64; +ALTER TABLE tab MODIFY STATISTICS f64 TYPE count_min; ALTER TABLE tab DROP STATISTICS f64; +-- Doesn't work: +ALTER TABLE tab ADD STATISTICS a TYPE count_min; -- { serverError ILLEGAL_STATISTICS } +ALTER TABLE tab MODIFY STATISTICS a TYPE count_min; -- { serverError ILLEGAL_STATISTICS } + +-- Any data type changes on columns with statistics are disallowed, for simplicity even if the new data type is compatible with all existing +-- statistics objects (e.g. tdigest can be created on Float64 and UInt64) +ALTER TABLE tab MODIFY COLUMN f64_tdigest UInt64; -- { serverError ALTER_OF_COLUMN_IS_FORBIDDEN } + +-- Finally, do a full-circle test of a good case. Print table definition after each step. +-- Intentionally specifying _two_ columns and _two_ statistics types to have that also tested. +SHOW CREATE TABLE tab; +ALTER TABLE tab ADD STATISTICS f64, f32 TYPE tdigest, uniq; +SHOW CREATE TABLE tab; +ALTER TABLE tab MODIFY STATISTICS f64, f32 TYPE tdigest, uniq; +SHOW CREATE TABLE tab; +ALTER TABLE tab CLEAR STATISTICS f64, f32; +SHOW CREATE TABLE tab; +ALTER TABLE tab MATERIALIZE STATISTICS f64, f32; +SHOW CREATE TABLE tab; +ALTER TABLE tab DROP STATISTICS f64, f32; 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; +DROP TABLE tab; diff --git a/tests/queries/0_stateless/02864_statistics_delayed_materialization_in_merge.reference b/tests/queries/0_stateless/02864_statistics_delayed_materialization_in_merge.reference new file mode 100644 index 00000000000..eb5e685597c --- /dev/null +++ b/tests/queries/0_stateless/02864_statistics_delayed_materialization_in_merge.reference @@ -0,0 +1,12 @@ +After insert + Prewhere info + Prewhere filter + Prewhere filter column: and(less(b, 10_UInt8), less(a, 10_UInt8)) (removed) +After merge + Prewhere info + Prewhere filter + Prewhere filter column: and(less(a, 10_UInt8), less(b, 10_UInt8)) (removed) +After truncate, insert, and materialize + Prewhere info + Prewhere filter + Prewhere filter column: and(less(a, 10_UInt8), less(b, 10_UInt8)) (removed) diff --git a/tests/queries/0_stateless/02864_statistics_delayed_materialization_in_merge.sql b/tests/queries/0_stateless/02864_statistics_delayed_materialization_in_merge.sql new file mode 100644 index 00000000000..d469a4c2036 --- /dev/null +++ b/tests/queries/0_stateless/02864_statistics_delayed_materialization_in_merge.sql @@ -0,0 +1,36 @@ +-- Tests delayed materialization of statistics in merge instead of during insert (setting 'materialize_statistics_on_insert = 0'). +-- (The concrete statistics type, column data type and predicate type don't matter) + +-- Checks by the predicate evaluation order in EXPLAIN. This is quite fragile, a better approach would be helpful (maybe 'send_logs_level'?) + +DROP TABLE IF EXISTS tab; + +SET allow_experimental_statistics = 1; +SET allow_statistics_optimize = 1; +SET enable_analyzer = 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 'After insert'; +SELECT replaceRegexpAll(explain, '__table1\.', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE b < 10 and a < 10) WHERE explain LIKE '%Prewhere%'; -- checks b first, then a (statistics not used) + +OPTIMIZE TABLE tab FINAL; +SELECT 'After merge'; +SELECT replaceRegexpAll(explain, '__table1\.', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE b < 10 and a < 10) WHERE explain LIKE '%Prewhere%'; -- checks a first, then b (statistics used) + +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 'After truncate, insert, and materialize'; +SELECT replaceRegexpAll(explain, '__table1\.', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE b < 10 and a < 10) WHERE explain LIKE '%Prewhere%'; -- checks a first, then b (statistics used) + +DROP TABLE tab; diff --git a/tests/queries/0_stateless/02864_statistics_exception.sql b/tests/queries/0_stateless/02864_statistics_exception.sql deleted file mode 100644 index 289ffee6600..00000000000 --- a/tests/queries/0_stateless/02864_statistics_exception.sql +++ /dev/null @@ -1,55 +0,0 @@ --- Tests creating/dropping/materializing statistics produces the right exceptions. - -DROP TABLE IF EXISTS tab; - --- Can't create statistics when allow_experimental_statistics = 0 -CREATE TABLE tab -( - a Float64 STATISTICS(tdigest) -) Engine = MergeTree() ORDER BY tuple(); -- { serverError INCORRECT_QUERY } - -SET allow_experimental_statistics = 1; - --- The same type of statistics can't exist more than once on a column -CREATE TABLE tab -( - a Float64 STATISTICS(tdigest, tdigest) -) Engine = MergeTree() ORDER BY tuple(); -- { serverError INCORRECT_QUERY } - --- Unknown statistics types are rejected -CREATE TABLE tab -( - a Float64 STATISTICS(no_statistics_type) -) Engine = MergeTree() ORDER BY tuple(); -- { serverError INCORRECT_QUERY } - --- tDigest statistics can only be created on numeric columns -CREATE TABLE tab -( - a String STATISTICS(tdigest), -) Engine = MergeTree() ORDER BY tuple(); -- { serverError ILLEGAL_STATISTICS } - -CREATE TABLE tab -( - a Float64, - b String -) Engine = MergeTree() ORDER BY tuple(); - -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 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 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 tab; diff --git a/tests/queries/0_stateless/02864_statistics_materialize_in_merge.reference b/tests/queries/0_stateless/02864_statistics_materialize_in_merge.reference deleted file mode 100644 index 5e969cf41cb..00000000000 --- a/tests/queries/0_stateless/02864_statistics_materialize_in_merge.reference +++ /dev/null @@ -1,10 +0,0 @@ -10 -10 -10 -statistics not used Condition less(b, 10_UInt8) moved to PREWHERE -statistics not used Condition less(a, 10_UInt8) moved to PREWHERE -statistics used after merge Condition less(a, 10_UInt8) moved to PREWHERE -statistics used after merge Condition less(b, 10_UInt8) moved to PREWHERE -statistics used after materialize Condition less(a, 10_UInt8) moved to PREWHERE -statistics used after materialize Condition less(b, 10_UInt8) moved to PREWHERE -2 0 diff --git a/tests/queries/0_stateless/02864_statistics_materialize_in_merge.sql b/tests/queries/0_stateless/02864_statistics_materialize_in_merge.sql deleted file mode 100644 index 3e15ec1148e..00000000000 --- a/tests/queries/0_stateless/02864_statistics_materialize_in_merge.sql +++ /dev/null @@ -1,52 +0,0 @@ --- 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_predicates.reference b/tests/queries/0_stateless/02864_statistics_predicates.reference new file mode 100644 index 00000000000..ffbd7269e05 --- /dev/null +++ b/tests/queries/0_stateless/02864_statistics_predicates.reference @@ -0,0 +1,98 @@ +u64 and = +10 +10 +10 +10 +0 +0 +0 +0 +10 +10 +10 +10 +u64 and < +70 +70 +70 +70 +80 +80 +80 +80 +70 +70 +70 +70 +f64 and = +10 +10 +10 +10 +0 +0 +0 +0 +10 +10 +10 +10 +0 +0 +0 +0 +f64 and < +70 +70 +70 +70 +80 +80 +80 +80 +70 +70 +70 +70 +80 +80 +80 +80 +dt and = +0 +0 +0 +0 +10 +10 +10 +10 +dt and < +10000 +10000 +10000 +10000 +70 +70 +70 +70 +b and = +5000 +5000 +5000 +5000 +5000 +5000 +5000 +5000 +5000 +5000 +5000 +5000 +0 +0 +0 +0 +s and = +10 +10 diff --git a/tests/queries/0_stateless/02864_statistics_predicates.sql b/tests/queries/0_stateless/02864_statistics_predicates.sql new file mode 100644 index 00000000000..779116cf19a --- /dev/null +++ b/tests/queries/0_stateless/02864_statistics_predicates.sql @@ -0,0 +1,214 @@ +-- Tags: no-fasttest +-- no-fasttest: 'count_min' sketches need a 3rd party library + +-- Tests the cross product of all predicates with all right-hand sides on all data types and all statistics types. + +SET allow_experimental_statistics = 1; +SET allow_statistics_optimize = 1; + +DROP TABLE IF EXISTS tab; + +CREATE TABLE tab +( + u64 UInt64, + u64_tdigest UInt64 STATISTICS(tdigest), + u64_count_min UInt64 STATISTICS(count_min), + u64_uniq UInt64 STATISTICS(uniq), + f64 Float64, + f64_tdigest Float64 STATISTICS(tdigest), + f64_count_min Float64 STATISTICS(count_min), + f64_uniq Float64 STATISTICS(uniq), + dt DateTime, + dt_tdigest DateTime STATISTICS(tdigest), + dt_count_min DateTime STATISTICS(count_min), + dt_uniq DateTime STATISTICS(uniq), + b Bool, + b_tdigest Bool STATISTICS(tdigest), + b_count_min Bool STATISTICS(count_min), + b_uniq Bool STATISTICS(uniq), + s String, + -- s_tdigest String STATISTICS(tdigest), -- not supported by tdigest + s_count_min String STATISTICS(count_min) + -- s_uniq String STATISTICS(uniq), -- not supported by uniq +) Engine = MergeTree() ORDER BY tuple() +SETTINGS min_bytes_for_wide_part = 0; + +INSERT INTO tab +-- SELECT number % 10000, number % 1000, -(number % 100) FROM system.numbers LIMIT 10000; +SELECT number % 1000, + number % 1000, + number % 1000, + number % 1000, + number % 1000, + number % 1000, + number % 1000, + number % 1000, + number % 1000, + number % 1000, + number % 1000, + number % 1000, + number % 2, + number % 2, + number % 2, + number % 2, + toString(number % 1000), + toString(number % 1000) +FROM system.numbers LIMIT 10000; + +-- u64 ---------------------------------------------------- + +SELECT 'u64 and ='; + +SELECT count(*) FROM tab WHERE u64 = 7; +SELECT count(*) FROM tab WHERE u64_tdigest = 7; +SELECT count(*) FROM tab WHERE u64_count_min = 7; +SELECT count(*) FROM tab WHERE u64_uniq = 7; + +SELECT count(*) FROM tab WHERE u64 = 7.7; +SELECT count(*) FROM tab WHERE u64_tdigest = 7.7; +SELECT count(*) FROM tab WHERE u64_count_min = 7.7; +SELECT count(*) FROM tab WHERE u64_uniq = 7.7; + +SELECT count(*) FROM tab WHERE u64 = '7'; +SELECT count(*) FROM tab WHERE u64_tdigest = '7'; +SELECT count(*) FROM tab WHERE u64_count_min = '7'; +SELECT count(*) FROM tab WHERE u64_uniq = '7'; + +SELECT count(*) FROM tab WHERE u64 = '7.7'; -- { serverError TYPE_MISMATCH } +SELECT count(*) FROM tab WHERE u64_tdigest = '7.7'; -- { serverError TYPE_MISMATCH } +SELECT count(*) FROM tab WHERE u64_count_min = '7.7'; -- { serverError TYPE_MISMATCH } +SELECT count(*) FROM tab WHERE u64_uniq = '7.7'; -- { serverError TYPE_MISMATCH } + +SELECT 'u64 and <'; + +SELECT count(*) FROM tab WHERE u64 < 7; +SELECT count(*) FROM tab WHERE u64_tdigest < 7; +SELECT count(*) FROM tab WHERE u64_count_min < 7; +SELECT count(*) FROM tab WHERE u64_uniq < 7; + +SELECT count(*) FROM tab WHERE u64 < 7.7; +SELECT count(*) FROM tab WHERE u64_tdigest < 7.7; +SELECT count(*) FROM tab WHERE u64_count_min < 7.7; +SELECT count(*) FROM tab WHERE u64_uniq < 7.7; + +SELECT count(*) FROM tab WHERE u64 < '7'; +SELECT count(*) FROM tab WHERE u64_tdigest < '7'; +SELECT count(*) FROM tab WHERE u64_count_min < '7'; +SELECT count(*) FROM tab WHERE u64_uniq < '7'; + +SELECT count(*) FROM tab WHERE u64 < '7.7'; -- { serverError TYPE_MISMATCH } +SELECT count(*) FROM tab WHERE u64_tdigest < '7.7'; -- { serverError TYPE_MISMATCH } +SELECT count(*) FROM tab WHERE u64_count_min < '7.7'; -- { serverError TYPE_MISMATCH } +SELECT count(*) FROM tab WHERE u64_uniq < '7.7'; -- { serverError TYPE_MISMATCH } + +-- f64 ---------------------------------------------------- + +SELECT 'f64 and ='; + +SELECT count(*) FROM tab WHERE f64 = 7; +SELECT count(*) FROM tab WHERE f64_tdigest = 7; +SELECT count(*) FROM tab WHERE f64_count_min = 7; +SELECT count(*) FROM tab WHERE f64_uniq = 7; + +SELECT count(*) FROM tab WHERE f64 = 7.7; +SELECT count(*) FROM tab WHERE f64_tdigest = 7.7; +SELECT count(*) FROM tab WHERE f64_count_min = 7.7; +SELECT count(*) FROM tab WHERE f64_uniq = 7.7; + +SELECT count(*) FROM tab WHERE f64 = '7'; +SELECT count(*) FROM tab WHERE f64_tdigest = '7'; +SELECT count(*) FROM tab WHERE f64_count_min = '7'; +SELECT count(*) FROM tab WHERE f64_uniq = '7'; + +SELECT count(*) FROM tab WHERE f64 = '7.7'; +SELECT count(*) FROM tab WHERE f64_tdigest = '7.7'; +SELECT count(*) FROM tab WHERE f64_count_min = '7.7'; +SELECT count(*) FROM tab WHERE f64_uniq = '7.7'; + +SELECT 'f64 and <'; + +SELECT count(*) FROM tab WHERE f64 < 7; +SELECT count(*) FROM tab WHERE f64_tdigest < 7; +SELECT count(*) FROM tab WHERE f64_count_min < 7; +SELECT count(*) FROM tab WHERE f64_uniq < 7; + +SELECT count(*) FROM tab WHERE f64 < 7.7; +SELECT count(*) FROM tab WHERE f64_tdigest < 7.7; +SELECT count(*) FROM tab WHERE f64_count_min < 7.7; +SELECT count(*) FROM tab WHERE f64_uniq < 7.7; + +SELECT count(*) FROM tab WHERE f64 < '7'; +SELECT count(*) FROM tab WHERE f64_tdigest < '7'; +SELECT count(*) FROM tab WHERE f64_count_min < '7'; +SELECT count(*) FROM tab WHERE f64_uniq < '7'; + +SELECT count(*) FROM tab WHERE f64 < '7.7'; +SELECT count(*) FROM tab WHERE f64_tdigest < '7.7'; +SELECT count(*) FROM tab WHERE f64_count_min < '7.7'; +SELECT count(*) FROM tab WHERE f64_uniq < '7.7'; + +-- dt ---------------------------------------------------- + +SELECT 'dt and ='; + +SELECT count(*) FROM tab WHERE dt = '2024-08-08 11:12:13'; +SELECT count(*) FROM tab WHERE dt_tdigest = '2024-08-08 11:12:13'; +SELECT count(*) FROM tab WHERE dt_count_min = '2024-08-08 11:12:13'; +SELECT count(*) FROM tab WHERE dt_uniq = '2024-08-08 11:12:13'; + +SELECT count(*) FROM tab WHERE dt = 7; +SELECT count(*) FROM tab WHERE dt_tdigest = 7; +SELECT count(*) FROM tab WHERE dt_count_min = 7; +SELECT count(*) FROM tab WHERE dt_uniq = 7; + +SELECT 'dt and <'; + +SELECT count(*) FROM tab WHERE dt < '2024-08-08 11:12:13'; +SELECT count(*) FROM tab WHERE dt_tdigest < '2024-08-08 11:12:13'; +SELECT count(*) FROM tab WHERE dt_count_min < '2024-08-08 11:12:13'; +SELECT count(*) FROM tab WHERE dt_uniq < '2024-08-08 11:12:13'; + +SELECT count(*) FROM tab WHERE dt < 7; +SELECT count(*) FROM tab WHERE dt_tdigest < 7; +SELECT count(*) FROM tab WHERE dt_count_min < 7; +SELECT count(*) FROM tab WHERE dt_uniq < 7; + +-- b ---------------------------------------------------- + +SELECT 'b and ='; + +SELECT count(*) FROM tab WHERE b = true; +SELECT count(*) FROM tab WHERE b_tdigest = true; +SELECT count(*) FROM tab WHERE b_count_min = true; +SELECT count(*) FROM tab WHERE b_uniq = true; + +SELECT count(*) FROM tab WHERE b = 'true'; +SELECT count(*) FROM tab WHERE b_tdigest = 'true'; +SELECT count(*) FROM tab WHERE b_count_min = 'true'; +SELECT count(*) FROM tab WHERE b_uniq = 'true'; + +SELECT count(*) FROM tab WHERE b = 1; +SELECT count(*) FROM tab WHERE b_tdigest = 1; +SELECT count(*) FROM tab WHERE b_count_min = 1; +SELECT count(*) FROM tab WHERE b_uniq = 1; + +SELECT count(*) FROM tab WHERE b = 1.1; +SELECT count(*) FROM tab WHERE b_tdigest = 1.1; +SELECT count(*) FROM tab WHERE b_count_min = 1.1; +SELECT count(*) FROM tab WHERE b_uniq = 1.1; + +-- s ---------------------------------------------------- + +SELECT 's and ='; + +SELECT count(*) FROM tab WHERE s = 7; -- { serverError NO_COMMON_TYPE } +-- SELECT count(*) FROM tab WHERE s_tdigest = 7; -- not supported +SELECT count(*) FROM tab WHERE s_count_min = 7; -- { serverError NO_COMMON_TYPE } +-- SELECT count(*) FROM tab WHERE s_uniq = 7; -- not supported + +SELECT count(*) FROM tab WHERE s = '7'; +-- SELECT count(*) FROM tab WHERE s_tdigest = '7'; -- not supported +SELECT count(*) FROM tab WHERE s_count_min = '7'; +-- SELECT count(*) FROM tab WHERE s_uniq = '7'; -- not supported + +DROP TABLE tab; diff --git a/tests/queries/0_stateless/02864_statistics_uniq.reference b/tests/queries/0_stateless/02864_statistics_uniq.reference deleted file mode 100644 index 77786dbdd8c..00000000000 --- a/tests/queries/0_stateless/02864_statistics_uniq.reference +++ /dev/null @@ -1,35 +0,0 @@ -CREATE TABLE default.t1\n(\n `a` Float64 STATISTICS(tdigest),\n `b` Int64 STATISTICS(tdigest),\n `c` Int64 STATISTICS(tdigest, uniq),\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), equals(c, 0), less(b, 10)) (removed) - Prewhere info - Prewhere filter - Prewhere filter column: and(equals(c, 11), less(a, 10), less(b, 10)) (removed) -After merge - Prewhere info - Prewhere filter - Prewhere filter column: and(less(a, 10), equals(c, 0), less(b, 10)) (removed) - Prewhere info - Prewhere filter - Prewhere filter column: and(equals(c, 11), less(a, 10), less(b, 10)) (removed) -After modify TDigest - Prewhere info - Prewhere filter - Prewhere filter column: and(less(a, 10), equals(c, 11), less(b, 10)) (removed) - Prewhere info - Prewhere filter - Prewhere filter column: and(less(a, 10), equals(c, 0), less(b, 10)) (removed) - Prewhere info - Prewhere filter - Prewhere filter column: and(less(c, -1), less(a, 10), less(b, 10)) (removed) -After drop - Prewhere info - Prewhere filter - Prewhere filter column: and(less(a, 10), equals(c, 11), less(b, 10)) (removed) - Prewhere info - Prewhere filter - Prewhere filter column: and(less(a, 10), equals(c, 0), less(b, 10)) (removed) - Prewhere info - Prewhere filter - Prewhere filter column: and(less(a, 10), less(c, -1), less(b, 10)) (removed) diff --git a/tests/queries/0_stateless/02864_statistics_uniq.sql b/tests/queries/0_stateless/02864_statistics_uniq.sql deleted file mode 100644 index 0f5f353c045..00000000000 --- a/tests/queries/0_stateless/02864_statistics_uniq.sql +++ /dev/null @@ -1,73 +0,0 @@ -DROP TABLE IF EXISTS t1; - -SET allow_experimental_statistics = 1; -SET allow_statistics_optimize = 1; -SET mutations_sync = 1; - -CREATE TABLE t1 -( - a Float64 STATISTICS(tdigest), - b Int64 STATISTICS(tdigest), - c Int64 STATISTICS(tdigest, uniq), - pk String, -) Engine = MergeTree() ORDER BY pk -SETTINGS min_bytes_for_wide_part = 0; - -SHOW CREATE TABLE t1; - -INSERT INTO t1 select number, -number, number/1000, generateUUIDv4() FROM system.numbers LIMIT 10000; -INSERT INTO t1 select 0, 0, 11, generateUUIDv4(); - -SELECT 'After insert'; -SELECT replaceRegexpAll(explain, '__table1.|_UInt8|_Int8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM t1 WHERE b < 10 and c = 0 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; -SELECT replaceRegexpAll(explain, '__table1.|_UInt8|_Int8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM t1 WHERE b < 10 and c = 11 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; -OPTIMIZE TABLE t1 FINAL; - -SELECT 'After merge'; -SELECT replaceRegexpAll(explain, '__table1.|_UInt8|_Int8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM t1 WHERE b < 10 and c = 0 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; -SELECT replaceRegexpAll(explain, '__table1.|_UInt8|_Int8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM t1 WHERE b < 10 and c = 11 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; - -SELECT 'After modify TDigest'; -ALTER TABLE t1 MODIFY STATISTICS c TYPE TDigest; -ALTER TABLE t1 MATERIALIZE STATISTICS c; - -SELECT replaceRegexpAll(explain, '__table1.|_UInt8|_Int8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM t1 WHERE b < 10 and c = 11 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; -SELECT replaceRegexpAll(explain, '__table1.|_UInt8|_Int8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM t1 WHERE b < 10 and c = 0 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; -SELECT replaceRegexpAll(explain, '__table1.|_UInt8|_Int8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM t1 WHERE b < 10 and c < -1 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; - - -ALTER TABLE t1 DROP STATISTICS c; - -SELECT 'After drop'; -SELECT replaceRegexpAll(explain, '__table1.|_UInt8|_Int8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM t1 WHERE b < 10 and c = 11 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; -SELECT replaceRegexpAll(explain, '__table1.|_UInt8|_Int8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM t1 WHERE b < 10 and c = 0 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; -SELECT replaceRegexpAll(explain, '__table1.|_UInt8|_Int8', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM t1 WHERE b < 10 and c < -1 and a < 10) WHERE explain LIKE '%Prewhere%' OR explain LIKE '%Filter column%'; - -DROP TABLE IF EXISTS t1; -DROP TABLE IF EXISTS t2; -SET allow_suspicious_low_cardinality_types=1; -CREATE TABLE t2 -( - a Float64 STATISTICS(tdigest), - b Int64 STATISTICS(tdigest), - c LowCardinality(Int64) STATISTICS(tdigest, uniq), - pk String, -) Engine = MergeTree() ORDER BY pk -SETTINGS min_bytes_for_wide_part = 0; -INSERT INTO t2 select number, -number, number/1000, generateUUIDv4() FROM system.numbers LIMIT 10000; - -DROP TABLE IF EXISTS t2; -DROP TABLE IF EXISTS t3; - -CREATE TABLE t3 -( - a Float64 STATISTICS(tdigest), - b Int64 STATISTICS(tdigest), - c Nullable(Int64) STATISTICS(tdigest, uniq), - pk String, -) Engine = MergeTree() ORDER BY pk -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/02864_statistics_usage.reference b/tests/queries/0_stateless/02864_statistics_usage.reference new file mode 100644 index 00000000000..a9f669b88c1 --- /dev/null +++ b/tests/queries/0_stateless/02864_statistics_usage.reference @@ -0,0 +1,20 @@ +After insert + Prewhere info + Prewhere filter + Prewhere filter column: and(less(a, 10_UInt8), less(b, 10_UInt8)) (removed) +After drop statistic + Prewhere info + Prewhere filter + Prewhere filter column: and(less(b, 10_UInt8), less(a, 10_UInt8)) (removed) +After add and materialize statistic + Prewhere info + Prewhere filter + Prewhere filter column: and(less(a, 10_UInt8), less(b, 10_UInt8)) (removed) +After merge + Prewhere info + Prewhere filter + Prewhere filter column: and(less(a, 10_UInt8), less(b, 10_UInt8)) (removed) +After rename + Prewhere info + Prewhere filter + Prewhere filter column: and(less(a, 10_UInt8), less(c, 10_UInt8)) (removed) diff --git a/tests/queries/0_stateless/02864_statistics_usage.sql b/tests/queries/0_stateless/02864_statistics_usage.sql new file mode 100644 index 00000000000..4956bd27e87 --- /dev/null +++ b/tests/queries/0_stateless/02864_statistics_usage.sql @@ -0,0 +1,42 @@ +-- Test that the optimizer picks up column statistics +-- (The concrete statistics type, column data type and predicate type don't matter) + +-- Checks by the predicate evaluation order in EXPLAIN. This is quite fragile, a better approach would be helpful (maybe 'send_logs_level'?) + +SET allow_experimental_statistics = 1; +SET allow_statistics_optimize = 1; +SET mutations_sync = 1; +SET enable_analyzer = 1; + +DROP TABLE IF EXISTS tab; + +CREATE TABLE tab +( + a Float64 STATISTICS(tdigest), + b Int64 STATISTICS(tdigest) +) Engine = MergeTree() ORDER BY tuple() +SETTINGS min_bytes_for_wide_part = 0; + +INSERT INTO tab select number, -number FROM system.numbers LIMIT 10000; +SELECT 'After insert'; +SELECT replaceRegexpAll(explain, '__table1\.', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE b < 10 and a < 10) WHERE explain LIKE '%Prewhere%'; -- checks a first, then b (statistics used) + +ALTER TABLE tab DROP STATISTICS a, b; +SELECT 'After drop statistic'; +SELECT replaceRegexpAll(explain, '__table1\.', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE b < 10 and a < 10) WHERE explain LIKE '%Prewhere%'; -- checks b first, then a (statistics not used) + +ALTER TABLE tab ADD STATISTICS a, b TYPE tdigest; +ALTER TABLE tab MATERIALIZE STATISTICS a, b; +INSERT INTO tab select number, -number FROM system.numbers LIMIT 10000; +SELECT 'After add and materialize statistic'; +SELECT replaceRegexpAll(explain, '__table1\.', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE b < 10 and a < 10) WHERE explain LIKE '%Prewhere%'; -- checks a first, then b (statistics used) + +OPTIMIZE TABLE tab FINAL; +SELECT 'After merge'; +SELECT replaceRegexpAll(explain, '__table1\.', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE b < 10 and a < 10) WHERE explain LIKE '%Prewhere%'; -- checks a first, then b (statistics used) + +ALTER TABLE tab RENAME COLUMN b TO c; +SELECT 'After rename'; +SELECT replaceRegexpAll(explain, '__table1\.', '') FROM (EXPLAIN actions=1 SELECT count(*) FROM tab WHERE c < 10 and a < 10) WHERE explain LIKE '%Prewhere%'; -- checks a first, then c (statistics used) + +DROP TABLE IF EXISTS tab; diff --git a/tests/queries/0_stateless/02866_size_of_marks_skip_idx_explain.sql b/tests/queries/0_stateless/02866_size_of_marks_skip_idx_explain.sql index b916c5ca13d..b3adf38710a 100644 --- a/tests/queries/0_stateless/02866_size_of_marks_skip_idx_explain.sql +++ b/tests/queries/0_stateless/02866_size_of_marks_skip_idx_explain.sql @@ -4,7 +4,7 @@ SET optimize_move_to_prewhere = 1; SET convert_query_to_cnf = 0; SET optimize_read_in_order = 1; -SET allow_experimental_analyzer = 1; -- slightly different operator names than w/o +SET enable_analyzer = 1; -- slightly different operator names than w/o DROP TABLE IF EXISTS test_skip_idx; diff --git a/tests/queries/0_stateless/02868_distinct_to_count_optimization.reference b/tests/queries/0_stateless/02868_distinct_to_count_optimization.reference index c2075f72f33..3dedcedbcee 100644 --- a/tests/queries/0_stateless/02868_distinct_to_count_optimization.reference +++ b/tests/queries/0_stateless/02868_distinct_to_count_optimization.reference @@ -6,7 +6,7 @@ FROM SELECT DISTINCT a FROM test_rewrite_uniq_to_count ) -SETTINGS allow_experimental_analyzer = 0 +SETTINGS enable_analyzer = 0 3 QUERY id: 0 PROJECTION COLUMNS @@ -23,7 +23,7 @@ QUERY id: 0 COLUMN id: 5, column_name: a, result_type: UInt8, source_id: 6 JOIN TREE TABLE id: 6, alias: __table2, table_name: default.test_rewrite_uniq_to_count - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 2. test distinct with subquery alias 3 SELECT count() @@ -32,7 +32,7 @@ FROM SELECT DISTINCT a FROM test_rewrite_uniq_to_count ) AS t -SETTINGS allow_experimental_analyzer = 0 +SETTINGS enable_analyzer = 0 3 QUERY id: 0 PROJECTION COLUMNS @@ -49,7 +49,7 @@ QUERY id: 0 COLUMN id: 5, column_name: a, result_type: UInt8, source_id: 6 JOIN TREE TABLE id: 6, alias: __table2, table_name: default.test_rewrite_uniq_to_count - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 3. test distinct with compound column name 3 SELECT count() @@ -58,7 +58,7 @@ FROM SELECT DISTINCT a FROM test_rewrite_uniq_to_count ) AS t -SETTINGS allow_experimental_analyzer = 0 +SETTINGS enable_analyzer = 0 3 QUERY id: 0 PROJECTION COLUMNS @@ -75,7 +75,7 @@ QUERY id: 0 COLUMN id: 5, column_name: a, result_type: UInt8, source_id: 6 JOIN TREE TABLE id: 6, alias: __table2, table_name: default.test_rewrite_uniq_to_count - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 4. test distinct with select expression alias 3 SELECT count() @@ -84,7 +84,7 @@ FROM SELECT DISTINCT a AS alias_of_a FROM test_rewrite_uniq_to_count ) AS t -SETTINGS allow_experimental_analyzer = 0 +SETTINGS enable_analyzer = 0 3 QUERY id: 0 PROJECTION COLUMNS @@ -101,7 +101,7 @@ QUERY id: 0 COLUMN id: 5, column_name: a, result_type: UInt8, source_id: 6 JOIN TREE TABLE id: 6, alias: __table2, table_name: default.test_rewrite_uniq_to_count - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 5. test simple group by 3 SELECT count() @@ -113,7 +113,7 @@ FROM FROM test_rewrite_uniq_to_count GROUP BY a ) -SETTINGS allow_experimental_analyzer = 0 +SETTINGS enable_analyzer = 0 3 QUERY id: 0 PROJECTION COLUMNS @@ -133,7 +133,7 @@ QUERY id: 0 GROUP BY LIST id: 7, nodes: 1 COLUMN id: 5, column_name: a, result_type: UInt8, source_id: 6 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 6. test group by with subquery alias 3 SELECT count() @@ -145,7 +145,7 @@ FROM FROM test_rewrite_uniq_to_count GROUP BY a ) AS t -SETTINGS allow_experimental_analyzer = 0 +SETTINGS enable_analyzer = 0 3 QUERY id: 0 PROJECTION COLUMNS @@ -165,7 +165,7 @@ QUERY id: 0 GROUP BY LIST id: 7, nodes: 1 COLUMN id: 5, column_name: a, result_type: UInt8, source_id: 6 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 7. test group by with compound column name 3 SELECT count() @@ -177,7 +177,7 @@ FROM FROM test_rewrite_uniq_to_count GROUP BY a ) AS t -SETTINGS allow_experimental_analyzer = 0 +SETTINGS enable_analyzer = 0 3 QUERY id: 0 PROJECTION COLUMNS @@ -197,7 +197,7 @@ QUERY id: 0 GROUP BY LIST id: 7, nodes: 1 COLUMN id: 5, column_name: a, result_type: UInt8, source_id: 6 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 8. test group by with select expression alias 3 SELECT count() @@ -209,7 +209,7 @@ FROM FROM test_rewrite_uniq_to_count GROUP BY alias_of_a ) AS t -SETTINGS allow_experimental_analyzer = 0 +SETTINGS enable_analyzer = 0 3 QUERY id: 0 PROJECTION COLUMNS @@ -229,4 +229,4 @@ QUERY id: 0 GROUP BY LIST id: 7, nodes: 1 COLUMN id: 5, column_name: a, result_type: UInt8, source_id: 6 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 diff --git a/tests/queries/0_stateless/02868_distinct_to_count_optimization.sql b/tests/queries/0_stateless/02868_distinct_to_count_optimization.sql index 66431b7c36b..d30bade4dd5 100644 --- a/tests/queries/0_stateless/02868_distinct_to_count_optimization.sql +++ b/tests/queries/0_stateless/02868_distinct_to_count_optimization.sql @@ -16,53 +16,53 @@ set optimize_uniq_to_count=true; SELECT '1. test simple distinct'; -SELECT uniq(a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) settings allow_experimental_analyzer=0; -EXPLAIN SYNTAX SELECT uniq(a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) settings allow_experimental_analyzer=0; -SELECT uniq(a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) settings allow_experimental_analyzer=1; -EXPLAIN QUERY TREE SELECT uniq(a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) settings allow_experimental_analyzer=1; +SELECT uniq(a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) settings enable_analyzer=0; +EXPLAIN SYNTAX SELECT uniq(a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) settings enable_analyzer=0; +SELECT uniq(a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) settings enable_analyzer=1; +EXPLAIN QUERY TREE SELECT uniq(a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) settings enable_analyzer=1; SELECT '2. test distinct with subquery alias'; -SELECT uniq(t.a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) t settings allow_experimental_analyzer=0; -EXPLAIN SYNTAX SELECT uniq(a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) t settings allow_experimental_analyzer=0; -SELECT uniq(t.a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) t settings allow_experimental_analyzer=1; -EXPLAIN QUERY TREE SELECT uniq(t.a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) t settings allow_experimental_analyzer=1; +SELECT uniq(t.a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) t settings enable_analyzer=0; +EXPLAIN SYNTAX SELECT uniq(a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) t settings enable_analyzer=0; +SELECT uniq(t.a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) t settings enable_analyzer=1; +EXPLAIN QUERY TREE SELECT uniq(t.a) FROM (SELECT DISTINCT a FROM test_rewrite_uniq_to_count) t settings enable_analyzer=1; SELECT '3. test distinct with compound column name'; -SELECT uniq(a) FROM (SELECT DISTINCT test_rewrite_uniq_to_count.a FROM test_rewrite_uniq_to_count) t settings allow_experimental_analyzer=0; -EXPLAIN SYNTAX SELECT uniq(a) FROM (SELECT DISTINCT test_rewrite_uniq_to_count.a FROM test_rewrite_uniq_to_count) t settings allow_experimental_analyzer=0; -SELECT uniq(a) FROM (SELECT DISTINCT test_rewrite_uniq_to_count.a FROM test_rewrite_uniq_to_count) t settings allow_experimental_analyzer=1; -EXPLAIN QUERY TREE SELECT uniq(a) FROM (SELECT DISTINCT test_rewrite_uniq_to_count.a FROM test_rewrite_uniq_to_count) t settings allow_experimental_analyzer=1; +SELECT uniq(a) FROM (SELECT DISTINCT test_rewrite_uniq_to_count.a FROM test_rewrite_uniq_to_count) t settings enable_analyzer=0; +EXPLAIN SYNTAX SELECT uniq(a) FROM (SELECT DISTINCT test_rewrite_uniq_to_count.a FROM test_rewrite_uniq_to_count) t settings enable_analyzer=0; +SELECT uniq(a) FROM (SELECT DISTINCT test_rewrite_uniq_to_count.a FROM test_rewrite_uniq_to_count) t settings enable_analyzer=1; +EXPLAIN QUERY TREE SELECT uniq(a) FROM (SELECT DISTINCT test_rewrite_uniq_to_count.a FROM test_rewrite_uniq_to_count) t settings enable_analyzer=1; SELECT '4. test distinct with select expression alias'; -SELECT uniq(alias_of_a) FROM (SELECT DISTINCT a as alias_of_a FROM test_rewrite_uniq_to_count) t settings allow_experimental_analyzer=0; -EXPLAIN SYNTAX SELECT uniq(alias_of_a) FROM (SELECT DISTINCT a as alias_of_a FROM test_rewrite_uniq_to_count) t settings allow_experimental_analyzer=0; -SELECT uniq(alias_of_a) FROM (SELECT DISTINCT a as alias_of_a FROM test_rewrite_uniq_to_count) t settings allow_experimental_analyzer=1; -EXPLAIN QUERY TREE SELECT uniq(alias_of_a) FROM (SELECT DISTINCT a as alias_of_a FROM test_rewrite_uniq_to_count) t settings allow_experimental_analyzer=1; +SELECT uniq(alias_of_a) FROM (SELECT DISTINCT a as alias_of_a FROM test_rewrite_uniq_to_count) t settings enable_analyzer=0; +EXPLAIN SYNTAX SELECT uniq(alias_of_a) FROM (SELECT DISTINCT a as alias_of_a FROM test_rewrite_uniq_to_count) t settings enable_analyzer=0; +SELECT uniq(alias_of_a) FROM (SELECT DISTINCT a as alias_of_a FROM test_rewrite_uniq_to_count) t settings enable_analyzer=1; +EXPLAIN QUERY TREE SELECT uniq(alias_of_a) FROM (SELECT DISTINCT a as alias_of_a FROM test_rewrite_uniq_to_count) t settings enable_analyzer=1; SELECT '5. test simple group by'; -SELECT uniq(a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) settings allow_experimental_analyzer=0; -EXPLAIN SYNTAX SELECT uniq(a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) settings allow_experimental_analyzer=0; -SELECT uniq(a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) settings allow_experimental_analyzer=1; -EXPLAIN QUERY TREE SELECT uniq(a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) settings allow_experimental_analyzer=1; +SELECT uniq(a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) settings enable_analyzer=0; +EXPLAIN SYNTAX SELECT uniq(a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) settings enable_analyzer=0; +SELECT uniq(a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) settings enable_analyzer=1; +EXPLAIN QUERY TREE SELECT uniq(a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) settings enable_analyzer=1; SELECT '6. test group by with subquery alias'; -SELECT uniq(t.a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings allow_experimental_analyzer=0; -EXPLAIN SYNTAX SELECT uniq(t.a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings allow_experimental_analyzer=0; -SELECT uniq(t.a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings allow_experimental_analyzer=1; -EXPLAIN QUERY TREE SELECT uniq(t.a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings allow_experimental_analyzer=1; +SELECT uniq(t.a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings enable_analyzer=0; +EXPLAIN SYNTAX SELECT uniq(t.a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings enable_analyzer=0; +SELECT uniq(t.a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings enable_analyzer=1; +EXPLAIN QUERY TREE SELECT uniq(t.a) FROM (SELECT a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings enable_analyzer=1; SELECT '7. test group by with compound column name'; -SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings allow_experimental_analyzer=0; -EXPLAIN SYNTAX SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings allow_experimental_analyzer=0; -SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings allow_experimental_analyzer=1; -EXPLAIN QUERY TREE SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings allow_experimental_analyzer=1; +SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings enable_analyzer=0; +EXPLAIN SYNTAX SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings enable_analyzer=0; +SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings enable_analyzer=1; +EXPLAIN QUERY TREE SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY a) t settings enable_analyzer=1; SELECT '8. test group by with select expression alias'; -SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY alias_of_a) t settings allow_experimental_analyzer=0; -EXPLAIN SYNTAX SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY alias_of_a) t settings allow_experimental_analyzer=0; -SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY alias_of_a) t settings allow_experimental_analyzer=1; -EXPLAIN QUERY TREE SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY alias_of_a) t settings allow_experimental_analyzer=1; +SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY alias_of_a) t settings enable_analyzer=0; +EXPLAIN SYNTAX SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY alias_of_a) t settings enable_analyzer=0; +SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY alias_of_a) t settings enable_analyzer=1; +EXPLAIN QUERY TREE SELECT uniq(t.alias_of_a) FROM (SELECT a as alias_of_a, sum(b) FROM test_rewrite_uniq_to_count GROUP BY alias_of_a) t settings enable_analyzer=1; drop table if exists test_rewrite_uniq_to_count; diff --git a/tests/queries/0_stateless/02868_no_merge_across_partitions_final_with_lonely.sh b/tests/queries/0_stateless/02868_no_merge_across_partitions_final_with_lonely.sh index 4bc29ce4233..be0ef4e2648 100755 --- a/tests/queries/0_stateless/02868_no_merge_across_partitions_final_with_lonely.sh +++ b/tests/queries/0_stateless/02868_no_merge_across_partitions_final_with_lonely.sh @@ -7,7 +7,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} -nm -q """ +${CLICKHOUSE_CLIENT} -m -q """ DROP TABLE IF EXISTS with_lonely; CREATE TABLE with_lonely @@ -23,7 +23,7 @@ ORDER BY (id); """ create_optimize_partition() { - ${CLICKHOUSE_CLIENT} -nm -q """ + ${CLICKHOUSE_CLIENT} -m -q """ INSERT INTO with_lonely SELECT number, '$1', number*10, 0 FROM numbers(10); INSERT INTO with_lonely SELECT number+500000, '$1', number*10, 1 FROM numbers(10); """ @@ -39,7 +39,7 @@ create_optimize_partition "2022-10-29" create_optimize_partition "2022-10-30" create_optimize_partition "2022-10-31" -${CLICKHOUSE_CLIENT} -nm -q """ +${CLICKHOUSE_CLIENT} -m -q """ SYSTEM STOP MERGES with_lonely; INSERT INTO with_lonely SELECT number, '2022-11-01', number*10, 0 FROM numbers(10); diff --git a/tests/queries/0_stateless/02871_clickhouse_client_restart_pager.sh b/tests/queries/0_stateless/02871_clickhouse_client_restart_pager.sh index cc4ce9b122e..418e439e44b 100755 --- a/tests/queries/0_stateless/02871_clickhouse_client_restart_pager.sh +++ b/tests/queries/0_stateless/02871_clickhouse_client_restart_pager.sh @@ -7,7 +7,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # head by default print 10 rows, but it is not enough to query 11 rows, since # we need to overflow the default pipe size, hence just 1 million of rows (it # should be around 6 MiB in text representation, should be definitelly enough). -$CLICKHOUSE_CLIENT --ignore-error -nm --pager head -q " +$CLICKHOUSE_CLIENT --ignore-error -m --pager head -q " select * from numbers(1e6); -- { clientError CANNOT_WRITE_TO_FILE_DESCRIPTOR } select * from numbers(1e6); -- { clientError CANNOT_WRITE_TO_FILE_DESCRIPTOR } " diff --git a/tests/queries/0_stateless/02875_clickhouse_local_multiquery.sh b/tests/queries/0_stateless/02875_clickhouse_local_multiquery.sh index 3f2b732e71b..3a7d861262e 100755 --- a/tests/queries/0_stateless/02875_clickhouse_local_multiquery.sh +++ b/tests/queries/0_stateless/02875_clickhouse_local_multiquery.sh @@ -9,7 +9,7 @@ $CLICKHOUSE_CLIENT -q "select 1; select 2;" $CLICKHOUSE_LOCAL -q "select 1; select 2;" # -n is a no-op -$CLICKHOUSE_CLIENT -n -q "select 1; select 2;" -$CLICKHOUSE_LOCAL -n -q "select 1; select 2;" +$CLICKHOUSE_CLIENT -q "select 1; select 2;" +$CLICKHOUSE_LOCAL -q "select 1; select 2;" exit 0 diff --git a/tests/queries/0_stateless/02875_merge_engine_set_index.sh b/tests/queries/0_stateless/02875_merge_engine_set_index.sh index 355d83167a6..f40696c31a9 100755 --- a/tests/queries/0_stateless/02875_merge_engine_set_index.sh +++ b/tests/queries/0_stateless/02875_merge_engine_set_index.sh @@ -7,7 +7,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -nq " +$CLICKHOUSE_CLIENT -q " CREATE TABLE t1 ( a UInt32, @@ -57,7 +57,7 @@ ORDER BY b DESC FORMAT Null;" -$CLICKHOUSE_CLIENT -nq " +$CLICKHOUSE_CLIENT -q " SYSTEM FLUSH LOGS; SELECT ProfileEvents['SelectedMarks'] diff --git a/tests/queries/0_stateless/02878_use_structure_from_insertion_table_with_explicit_insert_columns.sh b/tests/queries/0_stateless/02878_use_structure_from_insertion_table_with_explicit_insert_columns.sh index 8bdaa47c111..dd08724456b 100755 --- a/tests/queries/0_stateless/02878_use_structure_from_insertion_table_with_explicit_insert_columns.sh +++ b/tests/queries/0_stateless/02878_use_structure_from_insertion_table_with_explicit_insert_columns.sh @@ -5,7 +5,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh $CLICKHOUSE_LOCAL -q "select 42 as x format Native" > $CLICKHOUSE_TEST_UNIQUE_NAME.native -$CLICKHOUSE_LOCAL -n -q " +$CLICKHOUSE_LOCAL -q " create table test (x UInt64, y UInt64) engine=Memory; insert into test (x) select * from file('$CLICKHOUSE_TEST_UNIQUE_NAME.native'); insert into test (y) select * from file('$CLICKHOUSE_TEST_UNIQUE_NAME.native'); diff --git a/tests/queries/0_stateless/02879_use_structure_from_insertion_table_with_defaults.sh b/tests/queries/0_stateless/02879_use_structure_from_insertion_table_with_defaults.sh index 315bbcd544f..c7270b65e19 100755 --- a/tests/queries/0_stateless/02879_use_structure_from_insertion_table_with_defaults.sh +++ b/tests/queries/0_stateless/02879_use_structure_from_insertion_table_with_defaults.sh @@ -5,7 +5,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh $CLICKHOUSE_LOCAL -q "select 1 as x format Native" > $CLICKHOUSE_TEST_UNIQUE_NAME.native -$CLICKHOUSE_LOCAL -n -q " +$CLICKHOUSE_LOCAL -q " create table test (x UInt64, y UInt64 default 42) engine=Memory; insert into test select * from file('$CLICKHOUSE_TEST_UNIQUE_NAME.native'); select * from test; diff --git a/tests/queries/0_stateless/02882_primary_key_index_in_function_different_types.sql b/tests/queries/0_stateless/02882_primary_key_index_in_function_different_types.sql index 1b1a7607344..83b38955734 100644 --- a/tests/queries/0_stateless/02882_primary_key_index_in_function_different_types.sql +++ b/tests/queries/0_stateless/02882_primary_key_index_in_function_different_types.sql @@ -7,14 +7,14 @@ CREATE TABLE test_table INSERT INTO test_table SELECT number, number FROM numbers(10); -set allow_experimental_analyzer = 0; +set enable_analyzer = 0; EXPLAIN indexes = 1, description=0 SELECT id FROM test_table WHERE id <= 10 AND value IN (SELECT 5); EXPLAIN indexes = 1, description=0 SELECT id FROM test_table WHERE id <= 10 AND value IN (SELECT '5'); EXPLAIN indexes = 1, description=0 SELECT id FROM test_table WHERE id <= 10 AND value IN (SELECT toUInt8(number) FROM numbers(5)); EXPLAIN indexes = 1, description=0 SELECT id FROM test_table WHERE id <= 10 AND value IN (SELECT toString(number) FROM numbers(5)); -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; EXPLAIN indexes = 1, description=0 SELECT id FROM test_table WHERE id <= 10 AND value IN (SELECT 5); EXPLAIN indexes = 1, description=0 SELECT id FROM test_table WHERE id <= 10 AND value IN (SELECT '5'); diff --git a/tests/queries/0_stateless/02883_named_collections_override.sh b/tests/queries/0_stateless/02883_named_collections_override.sh index a08c795127d..915ce280226 100755 --- a/tests/queries/0_stateless/02883_named_collections_override.sh +++ b/tests/queries/0_stateless/02883_named_collections_override.sh @@ -8,7 +8,7 @@ u1="${CLICKHOUSE_TEST_UNIQUE_NAME}_collection1" u2="${CLICKHOUSE_TEST_UNIQUE_NAME}_collection2" u3="${CLICKHOUSE_TEST_UNIQUE_NAME}_collection3" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " DROP NAMED COLLECTION IF EXISTS $u1; DROP NAMED COLLECTION IF EXISTS $u2; diff --git a/tests/queries/0_stateless/02884_async_insert_native_protocol_1.sh b/tests/queries/0_stateless/02884_async_insert_native_protocol_1.sh index 7f583087336..791515c82d6 100755 --- a/tests/queries/0_stateless/02884_async_insert_native_protocol_1.sh +++ b/tests/queries/0_stateless/02884_async_insert_native_protocol_1.sh @@ -7,7 +7,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) set -e -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " DROP TABLE IF EXISTS t_async_insert_native_1; CREATE TABLE t_async_insert_native_1 (id UInt64, s String) ENGINE = MergeTree ORDER BY id; " @@ -22,7 +22,7 @@ echo '{"id": 1, "s": "aaa"}' \ | $CLICKHOUSE_CLIENT $async_insert_options -q 'INSERT INTO t_async_insert_native_1 FORMAT JSONEachRow {"id": 2, "s": "bbb"}' 2>&1 \ | grep -o "NOT_IMPLEMENTED" -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " SELECT sum(length(entries.bytes)) FROM system.asynchronous_inserts WHERE database = '$CLICKHOUSE_DATABASE' AND table = 't_async_insert_native_1'; diff --git a/tests/queries/0_stateless/02884_async_insert_native_protocol_2.sh b/tests/queries/0_stateless/02884_async_insert_native_protocol_2.sh index b9b1854eaef..a8a9209ee68 100755 --- a/tests/queries/0_stateless/02884_async_insert_native_protocol_2.sh +++ b/tests/queries/0_stateless/02884_async_insert_native_protocol_2.sh @@ -6,7 +6,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) set -e -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " DROP TABLE IF EXISTS t_async_insert_native_2; CREATE TABLE t_async_insert_native_2 (id UInt64, s String) ENGINE = MergeTree ORDER BY id; " @@ -18,7 +18,7 @@ echo "(3, 'ccc') (4, 'ddd') (5, 'eee')" | $CLICKHOUSE_CLIENT $async_insert_optio wait -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " SELECT * FROM t_async_insert_native_2 ORDER BY id; SYSTEM FLUSH LOGS; diff --git a/tests/queries/0_stateless/02884_async_insert_native_protocol_3.sh b/tests/queries/0_stateless/02884_async_insert_native_protocol_3.sh index c9d399607d0..229f13eb821 100755 --- a/tests/queries/0_stateless/02884_async_insert_native_protocol_3.sh +++ b/tests/queries/0_stateless/02884_async_insert_native_protocol_3.sh @@ -7,7 +7,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) set -e -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " DROP TABLE IF EXISTS t_async_insert_native_3; CREATE TABLE t_async_insert_native_3 (id UInt64, s String) ENGINE = MergeTree ORDER BY id; " @@ -21,7 +21,7 @@ $CLICKHOUSE_CLIENT $async_insert_options -q "INSERT INTO t_async_insert_native_3 wait -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " SELECT format, length(entries.bytes) FROM system.asynchronous_inserts WHERE database = '$CLICKHOUSE_DATABASE' AND table = 't_async_insert_native_3' ORDER BY format; diff --git a/tests/queries/0_stateless/02884_async_insert_native_protocol_4.sh b/tests/queries/0_stateless/02884_async_insert_native_protocol_4.sh index 9118c11315c..e84c1ca8899 100755 --- a/tests/queries/0_stateless/02884_async_insert_native_protocol_4.sh +++ b/tests/queries/0_stateless/02884_async_insert_native_protocol_4.sh @@ -4,7 +4,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " DROP TABLE IF EXISTS t_async_insert_native_4; CREATE TABLE t_async_insert_native_4 (id UInt64) ENGINE = MergeTree ORDER BY id; " @@ -20,7 +20,7 @@ echo "(2) (3) (4) (5)" | $CLICKHOUSE_CLIENT_WITH_LOG $async_insert_options --asy -q 'INSERT INTO t_async_insert_native_4 FORMAT Values' 2>&1 \ | grep -c "too much data" -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " SELECT * FROM t_async_insert_native_4 ORDER BY id; SYSTEM FLUSH LOGS; diff --git a/tests/queries/0_stateless/02884_create_view_with_sql_security_option.reference b/tests/queries/0_stateless/02884_create_view_with_sql_security_option.reference index a03343c8cb3..39e7aad87e0 100644 --- a/tests/queries/0_stateless/02884_create_view_with_sql_security_option.reference +++ b/tests/queries/0_stateless/02884_create_view_with_sql_security_option.reference @@ -27,6 +27,7 @@ OK OK 100 100 +OK ===== TestGrants ===== OK OK diff --git a/tests/queries/0_stateless/02884_create_view_with_sql_security_option.sh b/tests/queries/0_stateless/02884_create_view_with_sql_security_option.sh index cc4e76a9ed9..fadbbff7f34 100755 --- a/tests/queries/0_stateless/02884_create_view_with_sql_security_option.sh +++ b/tests/queries/0_stateless/02884_create_view_with_sql_security_option.sh @@ -199,6 +199,8 @@ ${CLICKHOUSE_CLIENT} --user $user2 --query "INSERT INTO source SELECT * FROM gen ${CLICKHOUSE_CLIENT} --query "SELECT count() FROM destination1" ${CLICKHOUSE_CLIENT} --query "SELECT count() FROM destination2" +(( $(${CLICKHOUSE_CLIENT} --query "ALTER TABLE test_table MODIFY SQL SECURITY INVOKER" 2>&1 | grep -c "is not supported") >= 1 )) && echo "OK" || echo "UNEXPECTED" + echo "===== TestGrants =====" ${CLICKHOUSE_CLIENT} --query "GRANT CREATE ON *.* TO $user1" ${CLICKHOUSE_CLIENT} --query "GRANT SELECT ON $db.test_table TO $user1, $user2" diff --git a/tests/queries/0_stateless/02885_ephemeral_columns_from_file.sh b/tests/queries/0_stateless/02885_ephemeral_columns_from_file.sh index 2917ec86957..065658d4d56 100755 --- a/tests/queries/0_stateless/02885_ephemeral_columns_from_file.sh +++ b/tests/queries/0_stateless/02885_ephemeral_columns_from_file.sh @@ -9,7 +9,7 @@ $CLICKHOUSE_LOCAL -q "select * from file('$CLICKHOUSE_TEST_UNIQUE_NAME.jsonl', a $CLICKHOUSE_LOCAL -q "select * from file('$CLICKHOUSE_TEST_UNIQUE_NAME.jsonl', auto, 'x UInt64 Alias y, y UInt64')" 2>&1 | grep -c "BAD_ARGUMENTS" $CLICKHOUSE_LOCAL -q "select * from file('$CLICKHOUSE_TEST_UNIQUE_NAME.jsonl', auto, 'x UInt64 Materialized 42, y UInt64')" 2>&1 | grep -c "BAD_ARGUMENTS" -$CLICKHOUSE_LOCAL -n -q " +$CLICKHOUSE_LOCAL -q " create table test (x UInt64 Ephemeral, y UInt64 default x + 1) engine=Memory; insert into test (x, y) select * from file('$CLICKHOUSE_TEST_UNIQUE_NAME.jsonl'); select * from test; diff --git a/tests/queries/0_stateless/02890_named_tuple_functions.sql b/tests/queries/0_stateless/02890_named_tuple_functions.sql index 0033102bd53..6724462562a 100644 --- a/tests/queries/0_stateless/02890_named_tuple_functions.sql +++ b/tests/queries/0_stateless/02890_named_tuple_functions.sql @@ -1,5 +1,5 @@ set enable_named_columns_in_function_tuple = 1; -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; drop table if exists x; create table x (i int, j int) engine MergeTree order by i; diff --git a/tests/queries/0_stateless/02890_untuple_column_names.reference b/tests/queries/0_stateless/02890_untuple_column_names.reference index 13a85c70138..f91a63fa15f 100644 --- a/tests/queries/0_stateless/02890_untuple_column_names.reference +++ b/tests/queries/0_stateless/02890_untuple_column_names.reference @@ -57,7 +57,7 @@ t.1: 1 Row 1: ā”€ā”€ā”€ā”€ā”€ā”€ t.1: 1 --- tuple() with enable_named_columns_in_function_tuple = 1 and allow_experimental_analyzer = 1 keeps the column names +-- tuple() with enable_named_columns_in_function_tuple = 1 and enable_analyzer = 1 keeps the column names Row 1: ā”€ā”€ā”€ā”€ā”€ā”€ t.a: 1 diff --git a/tests/queries/0_stateless/02890_untuple_column_names.sql b/tests/queries/0_stateless/02890_untuple_column_names.sql index cd490ca3522..9773e2e53a0 100644 --- a/tests/queries/0_stateless/02890_untuple_column_names.sql +++ b/tests/queries/0_stateless/02890_untuple_column_names.sql @@ -6,43 +6,43 @@ SELECT '-- tuple element alias'; -SELECT untuple(tuple(1)::Tuple(a Int)), untuple(tuple('s')::Tuple(a String)) FORMAT Vertical SETTINGS allow_experimental_analyzer = 0; -SELECT untuple(tuple(1)::Tuple(a Int)), untuple(tuple('s')::Tuple(a String)) FORMAT Vertical SETTINGS allow_experimental_analyzer = 1; +SELECT untuple(tuple(1)::Tuple(a Int)), untuple(tuple('s')::Tuple(a String)) FORMAT Vertical SETTINGS enable_analyzer = 0; +SELECT untuple(tuple(1)::Tuple(a Int)), untuple(tuple('s')::Tuple(a String)) FORMAT Vertical SETTINGS enable_analyzer = 1; -SELECT untuple(tuple(1)::Tuple(a Int)), untuple(tuple(1)::Tuple(a Int)) FORMAT Vertical SETTINGS allow_experimental_analyzer = 0; -- { serverError DUPLICATE_COLUMN } -SELECT untuple(tuple(1)::Tuple(a Int)), untuple(tuple(1)::Tuple(a Int)) FORMAT Vertical SETTINGS allow_experimental_analyzer = 1; -- Bug: doesn't throw an exception +SELECT untuple(tuple(1)::Tuple(a Int)), untuple(tuple(1)::Tuple(a Int)) FORMAT Vertical SETTINGS enable_analyzer = 0; -- { serverError DUPLICATE_COLUMN } +SELECT untuple(tuple(1)::Tuple(a Int)), untuple(tuple(1)::Tuple(a Int)) FORMAT Vertical SETTINGS enable_analyzer = 1; -- Bug: doesn't throw an exception SELECT '-- tuple element alias + untuple() alias'; -SELECT untuple(tuple(1)::Tuple(a Int)) x, untuple(tuple('s')::Tuple(a String)) y FORMAT Vertical SETTINGS allow_experimental_analyzer = 0; -SELECT untuple(tuple(1)::Tuple(a Int)) x, untuple(tuple('s')::Tuple(a String)) y FORMAT Vertical SETTINGS allow_experimental_analyzer = 1; +SELECT untuple(tuple(1)::Tuple(a Int)) x, untuple(tuple('s')::Tuple(a String)) y FORMAT Vertical SETTINGS enable_analyzer = 0; +SELECT untuple(tuple(1)::Tuple(a Int)) x, untuple(tuple('s')::Tuple(a String)) y FORMAT Vertical SETTINGS enable_analyzer = 1; -SELECT untuple(tuple(1)::Tuple(a Int)) x, untuple(tuple(1)::Tuple(a Int)) x FORMAT Vertical SETTINGS allow_experimental_analyzer = 0; -- { serverError DUPLICATE_COLUMN } -SELECT untuple(tuple(1)::Tuple(a Int)) x, untuple(tuple(1)::Tuple(a Int)) x FORMAT Vertical SETTINGS allow_experimental_analyzer = 1; -- Bug: doesn't throw an exception +SELECT untuple(tuple(1)::Tuple(a Int)) x, untuple(tuple(1)::Tuple(a Int)) x FORMAT Vertical SETTINGS enable_analyzer = 0; -- { serverError DUPLICATE_COLUMN } +SELECT untuple(tuple(1)::Tuple(a Int)) x, untuple(tuple(1)::Tuple(a Int)) x FORMAT Vertical SETTINGS enable_analyzer = 1; -- Bug: doesn't throw an exception SELECT '-- untuple() alias'; -SELECT untuple(tuple(1)::Tuple(Int)) x, untuple(tuple('s')::Tuple(String)) y FORMAT Vertical SETTINGS allow_experimental_analyzer = 0; -SELECT untuple(tuple(1)::Tuple(Int)) x, untuple(tuple('s')::Tuple(String)) y FORMAT Vertical SETTINGS allow_experimental_analyzer = 1; +SELECT untuple(tuple(1)::Tuple(Int)) x, untuple(tuple('s')::Tuple(String)) y FORMAT Vertical SETTINGS enable_analyzer = 0; +SELECT untuple(tuple(1)::Tuple(Int)) x, untuple(tuple('s')::Tuple(String)) y FORMAT Vertical SETTINGS enable_analyzer = 1; -SELECT untuple(tuple(1)::Tuple(Int)) x, untuple(tuple(1)::Tuple(Int)) x FORMAT Vertical SETTINGS allow_experimental_analyzer = 0; -- { serverError DUPLICATE_COLUMN } -SELECT untuple(tuple(1)::Tuple(Int)) x, untuple(tuple(1)::Tuple(Int)) x FORMAT Vertical SETTINGS allow_experimental_analyzer = 1; -- Bug: doesn't throw an exception +SELECT untuple(tuple(1)::Tuple(Int)) x, untuple(tuple(1)::Tuple(Int)) x FORMAT Vertical SETTINGS enable_analyzer = 0; -- { serverError DUPLICATE_COLUMN } +SELECT untuple(tuple(1)::Tuple(Int)) x, untuple(tuple(1)::Tuple(Int)) x FORMAT Vertical SETTINGS enable_analyzer = 1; -- Bug: doesn't throw an exception SELECT '-- no aliases'; -SELECT untuple(tuple(1)::Tuple(Int)), untuple(tuple('s')::Tuple(String)) FORMAT Vertical SETTINGS allow_experimental_analyzer = 0; -SELECT untuple(tuple(1)::Tuple(Int)), untuple(tuple('s')::Tuple(String)) FORMAT Vertical SETTINGS allow_experimental_analyzer = 1; +SELECT untuple(tuple(1)::Tuple(Int)), untuple(tuple('s')::Tuple(String)) FORMAT Vertical SETTINGS enable_analyzer = 0; +SELECT untuple(tuple(1)::Tuple(Int)), untuple(tuple('s')::Tuple(String)) FORMAT Vertical SETTINGS enable_analyzer = 1; -SELECT untuple(tuple(1)::Tuple(Int)), untuple(tuple(1)::Tuple(Int)) FORMAT Vertical SETTINGS allow_experimental_analyzer = 0; -- { serverError DUPLICATE_COLUMN } -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 untuple(tuple(1)::Tuple(Int)), untuple(tuple(1)::Tuple(Int)) FORMAT Vertical SETTINGS enable_analyzer = 0; -- { serverError DUPLICATE_COLUMN } +SELECT untuple(tuple(1)::Tuple(Int)), untuple(tuple(1)::Tuple(Int)) FORMAT Vertical SETTINGS enable_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, 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 untuple(tuple(1 as a)) as t FORMAT Vertical SETTINGS enable_analyzer = 0, enable_named_columns_in_function_tuple = 0; +SELECT untuple(tuple(1 as a)) as t FORMAT Vertical SETTINGS enable_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 '-- tuple() with enable_named_columns_in_function_tuple = 1 and enable_analyzer = 1 keeps the column names'; +SELECT untuple(tuple(1 as a)) as t FORMAT Vertical SETTINGS enable_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; -SELECT untuple(JSONExtract('{"key": "value"}', 'Tuple(key String)')) x FORMAT Vertical SETTINGS allow_experimental_analyzer = 1; +SELECT untuple(JSONExtract('{"key": "value"}', 'Tuple(key String)')) x FORMAT Vertical SETTINGS enable_analyzer = 0; +SELECT untuple(JSONExtract('{"key": "value"}', 'Tuple(key String)')) x FORMAT Vertical SETTINGS enable_analyzer = 1; diff --git a/tests/queries/0_stateless/02895_npy_output_format.sh b/tests/queries/0_stateless/02895_npy_output_format.sh index a364e447062..74000bc298f 100755 --- a/tests/queries/0_stateless/02895_npy_output_format.sh +++ b/tests/queries/0_stateless/02895_npy_output_format.sh @@ -9,7 +9,7 @@ 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}/ -${CLICKHOUSE_CLIENT} -n -q --ignore-error " +${CLICKHOUSE_CLIENT} -q --ignore-error " DROP DATABASE IF EXISTS npy_output_02895; CREATE DATABASE IF NOT EXISTS npy_output_02895; diff --git a/tests/queries/0_stateless/02895_peak_memory_usage_http_headers_regression.sh b/tests/queries/0_stateless/02895_peak_memory_usage_http_headers_regression.sh index d6775927f35..b4656c9e321 100755 --- a/tests/queries/0_stateless/02895_peak_memory_usage_http_headers_regression.sh +++ b/tests/queries/0_stateless/02895_peak_memory_usage_http_headers_regression.sh @@ -4,7 +4,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " DROP TABLE IF EXISTS data; DROP TABLE IF EXISTS data2; DROP VIEW IF EXISTS mv1; diff --git a/tests/queries/0_stateless/02903_empty_order_by_throws_error.sh b/tests/queries/0_stateless/02903_empty_order_by_throws_error.sh index 64f5dd1a987..ef631d9ed1b 100755 --- a/tests/queries/0_stateless/02903_empty_order_by_throws_error.sh +++ b/tests/queries/0_stateless/02903_empty_order_by_throws_error.sh @@ -5,13 +5,13 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CUR_DIR"/../shell_config.sh # setting disabled and no order by or primary key; expect error -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" DROP TABLE IF EXISTS test_empty_order_by; CREATE TABLE test_empty_order_by(a UInt8) ENGINE = MergeTree() SETTINGS index_granularity = 8192; " 2>&1 \ | grep -F -q "You must provide an ORDER BY or PRIMARY KEY expression in the table definition." && echo 'OK' || echo 'FAIL' # setting disabled and primary key in table definition -$CLICKHOUSE_CLIENT -n --query=" +$CLICKHOUSE_CLIENT --query=" DROP TABLE IF EXISTS test_empty_order_by; CREATE TABLE test_empty_order_by(a UInt8) ENGINE = MergeTree() PRIMARY KEY a SETTINGS index_granularity = 8192; SHOW CREATE TABLE test_empty_order_by; diff --git a/tests/queries/0_stateless/02903_rmt_retriable_merge_exception.sh b/tests/queries/0_stateless/02903_rmt_retriable_merge_exception.sh index 5065da371a8..b77e5b0b402 100755 --- a/tests/queries/0_stateless/02903_rmt_retriable_merge_exception.sh +++ b/tests/queries/0_stateless/02903_rmt_retriable_merge_exception.sh @@ -15,7 +15,7 @@ if [[ $($CLICKHOUSE_CLIENT -q "select count()>0 from system.clusters where clust cluster=test_cluster_database_replicated fi -$CLICKHOUSE_CLIENT -nm --distributed_ddl_output_mode=none -q " +$CLICKHOUSE_CLIENT -m --distributed_ddl_output_mode=none -q " drop table if exists rmt1; drop table if exists rmt2; @@ -46,7 +46,7 @@ part_name='%' # wait while there be at least one 'No active replica has part all_0_1_1 or covering part' in logs for _ in {0..50}; do - no_active_repilica_messages=$($CLICKHOUSE_CLIENT -nm -q " + no_active_repilica_messages=$($CLICKHOUSE_CLIENT -m -q " system flush logs; select count() @@ -65,7 +65,7 @@ for _ in {0..50}; do sleep 1 done -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " system start pulling replication log rmt2; system flush logs; diff --git a/tests/queries/0_stateless/02904_empty_order_by_with_setting_enabled.sh b/tests/queries/0_stateless/02904_empty_order_by_with_setting_enabled.sh index 7ac9b488be5..5f9dc6ea077 100755 --- a/tests/queries/0_stateless/02904_empty_order_by_with_setting_enabled.sh +++ b/tests/queries/0_stateless/02904_empty_order_by_with_setting_enabled.sh @@ -5,7 +5,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CUR_DIR"/../shell_config.sh # setting enabled and no order by or primary key -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET create_table_empty_primary_key_by_default = true; DROP TABLE IF EXISTS test_empty_order_by; CREATE TABLE test_empty_order_by(a UInt8) ENGINE = MergeTree() SETTINGS index_granularity = 8192; @@ -13,7 +13,7 @@ ${CLICKHOUSE_CLIENT} -n --query=" " 2>&1 \ | grep -F -q "ORDER BY tuple()" && echo 'OK' || echo 'FAIL' # setting enabled and per-column primary key -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET create_table_empty_primary_key_by_default = true; DROP TABLE IF EXISTS test_empty_order_by; CREATE TABLE test_empty_order_by(a UInt8 PRIMARY KEY, b String PRIMARY KEY) ENGINE = MergeTree() SETTINGS index_granularity = 8192; @@ -21,7 +21,7 @@ ${CLICKHOUSE_CLIENT} -n --query=" " 2>&1 \ | grep -F -q "ORDER BY (a, b)" && echo 'OK' || echo 'FAIL' # setting enabled and primary key in table definition (not per-column or order by) -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET create_table_empty_primary_key_by_default = true; DROP TABLE IF EXISTS test_empty_order_by; CREATE TABLE test_empty_order_by(a UInt8, b String) ENGINE = MergeTree() PRIMARY KEY (a) SETTINGS index_granularity = 8192; @@ -29,7 +29,7 @@ ${CLICKHOUSE_CLIENT} -n --query=" " 2>&1 \ | grep -F -q "ORDER BY a" && echo 'OK' || echo 'FAIL' # setting enabled and order by in table definition (no primary key) -${CLICKHOUSE_CLIENT} -n --query=" +${CLICKHOUSE_CLIENT} --query=" SET create_table_empty_primary_key_by_default = true; DROP TABLE IF EXISTS test_empty_order_by; CREATE TABLE test_empty_order_by(a UInt8, b String) ENGINE = MergeTree() ORDER BY (a, b) SETTINGS index_granularity = 8192; diff --git a/tests/queries/0_stateless/02907_backup_mv_with_no_inner_table.sh b/tests/queries/0_stateless/02907_backup_mv_with_no_inner_table.sh index 30ec50fa20f..e37f1e51c74 100755 --- a/tests/queries/0_stateless/02907_backup_mv_with_no_inner_table.sh +++ b/tests/queries/0_stateless/02907_backup_mv_with_no_inner_table.sh @@ -5,7 +5,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " drop table if exists src; create table src (a Int32) engine = MergeTree() order by tuple(); @@ -15,14 +15,14 @@ create materialized view mv (a Int32) engine = MergeTree() order by tuple() as s uuid=$(${CLICKHOUSE_CLIENT} --query "select uuid from system.tables where table='mv' and database == currentDatabase()") inner_table=".inner_id.${uuid}" -${CLICKHOUSE_CLIENT} -nm --query "drop table \`$inner_table\` sync" +${CLICKHOUSE_CLIENT} -m --query "drop table \`$inner_table\` sync" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " set send_logs_level = 'error'; backup table ${CLICKHOUSE_DATABASE}.\`mv\` to Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}'); " | grep -o "BACKUP_CREATED" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " drop table mv; restore table ${CLICKHOUSE_DATABASE}.\`mv\` from Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}'); " | grep -o "RESTORED" diff --git a/tests/queries/0_stateless/02907_backup_mv_with_no_source_table.sh b/tests/queries/0_stateless/02907_backup_mv_with_no_source_table.sh index d59ebe400ee..f950954941f 100755 --- a/tests/queries/0_stateless/02907_backup_mv_with_no_source_table.sh +++ b/tests/queries/0_stateless/02907_backup_mv_with_no_source_table.sh @@ -4,7 +4,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " drop table if exists src; create table src (a Int32) engine = MergeTree() order by tuple(); @@ -15,18 +15,18 @@ drop table if exists mv; create materialized view mv to dst (a Int32) as select * from src; " -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " drop table src; backup database ${CLICKHOUSE_DATABASE} on cluster test_shard_localhost to Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}'); " | grep -o "BACKUP_CREATED" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " drop table mv; set allow_deprecated_database_ordinary=1; restore table ${CLICKHOUSE_DATABASE}.mv on cluster test_shard_localhost from Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}'); " | grep -o "RESTORED" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " drop table if exists src; create table src (a Int32) engine = MergeTree() order by tuple(); @@ -37,13 +37,13 @@ drop table if exists mv; create materialized view mv to dst (a Int32) as select * from src; " -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " drop table src; drop table dst; backup database ${CLICKHOUSE_DATABASE} on cluster test_shard_localhost to Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}2'); " | grep -o "BACKUP_CREATED" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " drop table mv; set allow_deprecated_database_ordinary=1; restore table ${CLICKHOUSE_DATABASE}.mv on cluster test_shard_localhost from Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}2'); diff --git a/tests/queries/0_stateless/02907_backup_restore_default_nullable.sh b/tests/queries/0_stateless/02907_backup_restore_default_nullable.sh index 8ed36a7edd7..dc5793d1638 100755 --- a/tests/queries/0_stateless/02907_backup_restore_default_nullable.sh +++ b/tests/queries/0_stateless/02907_backup_restore_default_nullable.sh @@ -4,7 +4,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " drop table if exists test; set data_type_default_nullable = 0; create table test (test String) ENGINE = MergeTree() ORDER BY tuple(); @@ -13,7 +13,7 @@ backup table ${CLICKHOUSE_DATABASE}.test on cluster test_shard_localhost to Disk ${CLICKHOUSE_CLIENT} --query "show create table test" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " drop table test sync; set data_type_default_nullable = 1; restore table ${CLICKHOUSE_DATABASE}.test on cluster test_shard_localhost from Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}'); diff --git a/tests/queries/0_stateless/02907_backup_restore_flatten_nested.sh b/tests/queries/0_stateless/02907_backup_restore_flatten_nested.sh index 742d24a97eb..eae307add10 100755 --- a/tests/queries/0_stateless/02907_backup_restore_flatten_nested.sh +++ b/tests/queries/0_stateless/02907_backup_restore_flatten_nested.sh @@ -4,7 +4,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " drop table if exists test; set flatten_nested = 0; create table test (test Array(Tuple(foo String, bar Float64))) ENGINE = MergeTree() ORDER BY tuple(); @@ -13,7 +13,7 @@ backup table ${CLICKHOUSE_DATABASE}.test on cluster test_shard_localhost to Disk ${CLICKHOUSE_CLIENT} --query "show create table test" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " drop table if exists test2; set flatten_nested = 0; create table test2 (test Nested(foo String, bar Float64)) ENGINE = MergeTree() ORDER BY tuple(); @@ -22,7 +22,7 @@ backup table ${CLICKHOUSE_DATABASE}.test2 on cluster test_shard_localhost to Dis ${CLICKHOUSE_CLIENT} --query "show create table test2" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " drop table test sync; set flatten_nested = 1; restore table ${CLICKHOUSE_DATABASE}.test on cluster test_shard_localhost from Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}'); @@ -30,7 +30,7 @@ restore table ${CLICKHOUSE_DATABASE}.test on cluster test_shard_localhost from D ${CLICKHOUSE_CLIENT} --query "show create table test" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " drop table test2 sync; set flatten_nested = 1; restore table ${CLICKHOUSE_DATABASE}.test2 on cluster test_shard_localhost from Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}2'); diff --git a/tests/queries/0_stateless/02907_clickhouse_dictionary_bug.sh b/tests/queries/0_stateless/02907_clickhouse_dictionary_bug.sh index 57182050534..2cad15c6fcb 100755 --- a/tests/queries/0_stateless/02907_clickhouse_dictionary_bug.sh +++ b/tests/queries/0_stateless/02907_clickhouse_dictionary_bug.sh @@ -5,7 +5,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} -n -q " +${CLICKHOUSE_CLIENT} -q " DROP DICTIONARY IF EXISTS 02907_dictionary; DROP TABLE IF EXISTS 02907_table; diff --git a/tests/queries/0_stateless/02907_system_backups_profile_events.sh b/tests/queries/0_stateless/02907_system_backups_profile_events.sh index 801056a2844..9a1d5a3db11 100755 --- a/tests/queries/0_stateless/02907_system_backups_profile_events.sh +++ b/tests/queries/0_stateless/02907_system_backups_profile_events.sh @@ -4,7 +4,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " drop table if exists test; create table test (a Int32) engine = MergeTree() order by tuple(); " @@ -12,10 +12,10 @@ create table test (a Int32) engine = MergeTree() order by tuple(); backup_id=${CLICKHOUSE_TEST_UNIQUE_NAME} backup_name="Disk('backups', '$backup_id')"; -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " backup table ${CLICKHOUSE_DATABASE}.test to $backup_name; " | grep -o "BACKUP_CREATED" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " select ProfileEvents['BackupEntriesCollectorMicroseconds'] > 10 from system.backups where name='Disk(\'backups\', \'$backup_id\')' " diff --git a/tests/queries/0_stateless/02908_Npy_files_caching.sh b/tests/queries/0_stateless/02908_Npy_files_caching.sh index 4845f740972..218e13efb95 100755 --- a/tests/queries/0_stateless/02908_Npy_files_caching.sh +++ b/tests/queries/0_stateless/02908_Npy_files_caching.sh @@ -7,13 +7,13 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) $CLICKHOUSE_LOCAL -q "select count() from file('$CURDIR/data_npy/one_dim.npy') settings optimize_count_from_files=0" $CLICKHOUSE_LOCAL -q "select count() from file('$CURDIR/data_npy/one_dim.npy') settings optimize_count_from_files=1" $CLICKHOUSE_LOCAL -q "select count() from file('$CURDIR/data_npy/one_dim.npy', auto, 'array Int64') settings optimize_count_from_files=1" -$CLICKHOUSE_LOCAL -nm -q " +$CLICKHOUSE_LOCAL -m -q " desc file('$CURDIR/data_npy/one_dim.npy'); select number_of_rows from system.schema_inference_cache where format='Npy'; " $CLICKHOUSE_LOCAL -q "select count() from file('$CURDIR/data_npy/npy_big.npy') settings optimize_count_from_files=0" $CLICKHOUSE_LOCAL -q "select count() from file('$CURDIR/data_npy/npy_big.npy') settings optimize_count_from_files=1" -$CLICKHOUSE_LOCAL -nm -q " +$CLICKHOUSE_LOCAL -m -q " desc file('$CURDIR/data_npy/npy_big.npy'); select number_of_rows from system.schema_inference_cache where format='Npy'; " diff --git a/tests/queries/0_stateless/02908_table_ttl_dependency.sh b/tests/queries/0_stateless/02908_table_ttl_dependency.sh index 70136b4a42b..0bc02426f61 100755 --- a/tests/queries/0_stateless/02908_table_ttl_dependency.sh +++ b/tests/queries/0_stateless/02908_table_ttl_dependency.sh @@ -6,7 +6,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " DROP TABLE IF EXISTS 02908_dependent; DROP TABLE IF EXISTS 02908_main; @@ -14,11 +14,11 @@ $CLICKHOUSE_CLIENT -nm -q " CREATE TABLE 02908_dependent (a UInt32, ts DateTime) ENGINE = MergeTree ORDER BY a TTL ts + 1 WHERE a IN (SELECT a FROM ${CLICKHOUSE_DATABASE}.02908_main); " -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " DROP TABLE 02908_main; " 2>&1 | grep -F -q "HAVE_DEPENDENT_OBJECTS" -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " DROP TABLE 02908_dependent; DROP TABLE 02908_main; " diff --git a/tests/queries/0_stateless/02909_settings_in_json_schema_cache.sh b/tests/queries/0_stateless/02909_settings_in_json_schema_cache.sh index 8da144f90ca..75d491642ea 100755 --- a/tests/queries/0_stateless/02909_settings_in_json_schema_cache.sh +++ b/tests/queries/0_stateless/02909_settings_in_json_schema_cache.sh @@ -5,7 +5,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CUR_DIR"/../shell_config.sh echo '{"x" : 42}' > $CLICKHOUSE_TEST_UNIQUE_NAME.json -$CLICKHOUSE_LOCAL -nm -q " +$CLICKHOUSE_LOCAL -m -q " DESC file('$CLICKHOUSE_TEST_UNIQUE_NAME.json') SETTINGS schema_inference_make_columns_nullable=1; DESC file('$CLICKHOUSE_TEST_UNIQUE_NAME.json') SETTINGS schema_inference_make_columns_nullable=0; SELECT count() from system.schema_inference_cache where format = 'JSON' and additional_format_info like '%schema_inference_make_columns_nullable%';" diff --git a/tests/queries/0_stateless/02910_object-json-crash-add-column.sql b/tests/queries/0_stateless/02910_object-json-crash-add-column.sql index b2d64be1676..bda5e958453 100644 --- a/tests/queries/0_stateless/02910_object-json-crash-add-column.sql +++ b/tests/queries/0_stateless/02910_object-json-crash-add-column.sql @@ -1,3 +1,5 @@ +SET allow_experimental_object_type=1; + DROP TABLE IF EXISTS test02910; CREATE TABLE test02910 diff --git a/tests/queries/0_stateless/02911_analyzer_explain_estimate.sql b/tests/queries/0_stateless/02911_analyzer_explain_estimate.sql index b082f2f33b2..77f30ba82ba 100644 --- a/tests/queries/0_stateless/02911_analyzer_explain_estimate.sql +++ b/tests/queries/0_stateless/02911_analyzer_explain_estimate.sql @@ -1,5 +1,5 @@ -- Tags: distributed -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; EXPLAIN ESTIMATE SELECT 0 = 1048577, NULL, groupBitmapOr(bitmapBuild([toInt32(65537)])) FROM cluster(test_cluster_two_shards) WHERE NULL = 1048575; diff --git a/tests/queries/0_stateless/02911_analyzer_order_by_read_in_order_query_plan.sql b/tests/queries/0_stateless/02911_analyzer_order_by_read_in_order_query_plan.sql index 77a72c24f5a..907263168d0 100644 --- a/tests/queries/0_stateless/02911_analyzer_order_by_read_in_order_query_plan.sql +++ b/tests/queries/0_stateless/02911_analyzer_order_by_read_in_order_query_plan.sql @@ -1,4 +1,4 @@ -SET optimize_read_in_order = 1, query_plan_read_in_order = 1, allow_experimental_analyzer = 1; +SET optimize_read_in_order = 1, query_plan_read_in_order = 1, enable_analyzer = 1; drop table if exists tab; drop table if exists tab2; diff --git a/tests/queries/0_stateless/02911_analyzer_remove_unused_projection_columns.sql b/tests/queries/0_stateless/02911_analyzer_remove_unused_projection_columns.sql index 70de63c592d..d567ac6c8f3 100644 --- a/tests/queries/0_stateless/02911_analyzer_remove_unused_projection_columns.sql +++ b/tests/queries/0_stateless/02911_analyzer_remove_unused_projection_columns.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table 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 ee5e1582015..e1e095a6eae 100644 --- a/tests/queries/0_stateless/02911_join_on_nullsafe_optimization.sql +++ b/tests/queries/0_stateless/02911_join_on_nullsafe_optimization.sql @@ -15,7 +15,7 @@ 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; +SET enable_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; diff --git a/tests/queries/0_stateless/02911_support_alias_column_in_indices.sql b/tests/queries/0_stateless/02911_support_alias_column_in_indices.sql index 4d68786d7db..5ab50044e64 100644 --- a/tests/queries/0_stateless/02911_support_alias_column_in_indices.sql +++ b/tests/queries/0_stateless/02911_support_alias_column_in_indices.sql @@ -17,8 +17,8 @@ settings index_granularity = 8192, min_index_granularity_bytes = 1024, index_gra insert into test1 select * from numbers(10); insert into test1 select * from numbers(11, 20); -explain indexes = 1 select * from test1 where a > 10 settings allow_experimental_analyzer = 0; -explain indexes = 1 select * from test1 where a > 10 settings allow_experimental_analyzer = 1; +explain indexes = 1 select * from test1 where a > 10 settings enable_analyzer = 0; +explain indexes = 1 select * from test1 where a > 10 settings enable_analyzer = 1; create table test2 ( @@ -34,7 +34,7 @@ settings index_granularity = 8192, min_index_granularity_bytes = 1024, index_gra insert into test2 select * from numbers(10); insert into test2 select * from numbers(11, 20); -explain indexes = 1 select * from test2 where a2 > 15 settings allow_experimental_analyzer = 0; -explain indexes = 1 select * from test2 where a2 > 15 settings allow_experimental_analyzer = 1; +explain indexes = 1 select * from test2 where a2 > 15 settings enable_analyzer = 0; +explain indexes = 1 select * from test2 where a2 > 15 settings enable_analyzer = 1; drop database 02911_support_alias_column_in_indices; diff --git a/tests/queries/0_stateless/02915_analyzer_fuzz_1.sql b/tests/queries/0_stateless/02915_analyzer_fuzz_1.sql index 94849453063..f1d606ab1cd 100644 --- a/tests/queries/0_stateless/02915_analyzer_fuzz_1.sql +++ b/tests/queries/0_stateless/02915_analyzer_fuzz_1.sql @@ -1,2 +1,2 @@ -set allow_experimental_analyzer=1; +set enable_analyzer=1; SELECT concat('With ', materialize(_CAST('ba\0', 'LowCardinality(FixedString(3))'))) AS `concat('With ', materialize(CAST('ba\\0', 'LowCardinality(FixedString(3))')))` FROM system.one GROUP BY 'With '; diff --git a/tests/queries/0_stateless/02915_analyzer_fuzz_2.sql b/tests/queries/0_stateless/02915_analyzer_fuzz_2.sql index ca9fff68446..8921d36c546 100644 --- a/tests/queries/0_stateless/02915_analyzer_fuzz_2.sql +++ b/tests/queries/0_stateless/02915_analyzer_fuzz_2.sql @@ -1,5 +1,4 @@ SET aggregate_functions_null_for_empty = 1; ---set allow_experimental_analyzer=1; +--set enable_analyzer=1; create table t_delete_projection (x UInt32, y UInt64, projection p (select sum(y))) engine = MergeTree order by tuple() SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; insert into t_delete_projection select number, toString(number) from numbers(8192 * 10); - diff --git a/tests/queries/0_stateless/02915_analyzer_fuzz_5.sql b/tests/queries/0_stateless/02915_analyzer_fuzz_5.sql index 29d06d2c315..d75d4f4eb72 100644 --- a/tests/queries/0_stateless/02915_analyzer_fuzz_5.sql +++ b/tests/queries/0_stateless/02915_analyzer_fuzz_5.sql @@ -1,4 +1,4 @@ -set allow_experimental_analyzer=1; +set enable_analyzer=1; SET max_block_size = 1000; SET max_threads = 4; SET max_rows_to_group_by = 3000, group_by_overflow_mode = 'any'; diff --git a/tests/queries/0_stateless/02915_analyzer_fuzz_6.sql b/tests/queries/0_stateless/02915_analyzer_fuzz_6.sql index b4eb1b4aff4..cc276ec4074 100644 --- a/tests/queries/0_stateless/02915_analyzer_fuzz_6.sql +++ b/tests/queries/0_stateless/02915_analyzer_fuzz_6.sql @@ -1,5 +1,5 @@ set allow_suspicious_low_cardinality_types=1; -set allow_experimental_analyzer=1; +set enable_analyzer=1; create table tab (x LowCardinality(Nullable(Float64))) engine = MergeTree order by x settings allow_nullable_key=1; insert into tab select number from numbers(2); diff --git a/tests/queries/0_stateless/02915_input_table_function_in_subquery.sh b/tests/queries/0_stateless/02915_input_table_function_in_subquery.sh index 80e38338751..7ad38e11e96 100755 --- a/tests/queries/0_stateless/02915_input_table_function_in_subquery.sh +++ b/tests/queries/0_stateless/02915_input_table_function_in_subquery.sh @@ -5,7 +5,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " CREATE TABLE IF NOT EXISTS ts_data_double_raw ( device_id UInt32 NOT NULL CODEC(ZSTD), diff --git a/tests/queries/0_stateless/02915_lazy_loading_of_base_backups.sh b/tests/queries/0_stateless/02915_lazy_loading_of_base_backups.sh index 5f0f41a956b..b6d6ca57768 100755 --- a/tests/queries/0_stateless/02915_lazy_loading_of_base_backups.sh +++ b/tests/queries/0_stateless/02915_lazy_loading_of_base_backups.sh @@ -13,40 +13,40 @@ b_backup="Disk('backups', '$b_backup_id')" c_backup_id=${CLICKHOUSE_TEST_UNIQUE_NAME}_c c_backup="Disk('backups', '$c_backup_id')" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " DROP TABLE IF EXISTS tbl1; DROP TABLE IF EXISTS tbl2; DROP TABLE IF EXISTS tbl3; " -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " CREATE TABLE tbl1 (a Int32) ENGINE = MergeTree() ORDER BY tuple(); " # The following BACKUP command must write backup 'a'. -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " BACKUP DATABASE ${CLICKHOUSE_DATABASE} TO $a_backup SETTINGS id='$a_backup_id'; " | grep -o "BACKUP_CREATED" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " CREATE TABLE tbl2 (a Int32) ENGINE = MergeTree() ORDER BY tuple(); " # The following BACKUP command must read backup 'a' and write backup 'b'. -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " BACKUP DATABASE ${CLICKHOUSE_DATABASE} TO $b_backup SETTINGS id='$b_backup_id', base_backup=$a_backup; " | grep -o "BACKUP_CREATED" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " CREATE TABLE tbl3 (a Int32) ENGINE = MergeTree() ORDER BY tuple(); " # The following BACKUP command must read only backup 'b' (and not 'a') and write backup 'c'. -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " BACKUP DATABASE ${CLICKHOUSE_DATABASE} TO $c_backup SETTINGS id='$c_backup_id', base_backup=$b_backup; " | grep -o "BACKUP_CREATED" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " DROP TABLE tbl1; DROP TABLE tbl2; DROP TABLE tbl3; @@ -57,28 +57,28 @@ r2_restore_id=${CLICKHOUSE_TEST_UNIQUE_NAME}_r2 r3_restore_id=${CLICKHOUSE_TEST_UNIQUE_NAME}_r3 # The following RESTORE command must read all 3 backups 'a', 'b', c' because the table 'tbl1' was in the first backup. -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " RESTORE TABLE ${CLICKHOUSE_DATABASE}.tbl1 FROM $c_backup SETTINGS id='$r1_restore_id'; " | grep -o "RESTORED" # The following RESTORE command must read only 2 backups 'b', c' (and not 'a') because the table 'tbl2' was in the second backup. -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " RESTORE TABLE ${CLICKHOUSE_DATABASE}.tbl2 FROM $c_backup SETTINGS id='$r2_restore_id'; " | grep -o "RESTORED" # The following RESTORE command must read only 1 backup 'c' (and not 'a' or 'b') because the table 'tbl3' was in the third backup. -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " RESTORE TABLE ${CLICKHOUSE_DATABASE}.tbl3 FROM $c_backup SETTINGS id='$r3_restore_id'; " | grep -o "RESTORED" all_ids="['$a_backup_id', '$b_backup_id', '$c_backup_id', '$r1_restore_id', '$r2_restore_id', '$r3_restore_id']" id_prefix_len=`expr "${CLICKHOUSE_TEST_UNIQUE_NAME}_" : '.*'` -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " SELECT substr(id, 1 + $id_prefix_len) as short_id, ProfileEvents['BackupsOpenedForRead'], ProfileEvents['BackupsOpenedForWrite'] FROM system.backups WHERE id IN ${all_ids} ORDER BY short_id " -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " DROP TABLE tbl1; DROP TABLE tbl2; DROP TABLE tbl3; diff --git a/tests/queries/0_stateless/02916_dictionary_access.sh b/tests/queries/0_stateless/02916_dictionary_access.sh index 08ee517ab3b..be62cc027ef 100755 --- a/tests/queries/0_stateless/02916_dictionary_access.sh +++ b/tests/queries/0_stateless/02916_dictionary_access.sh @@ -7,7 +7,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) username="user_${CLICKHOUSE_TEST_UNIQUE_NAME}" dictname="dict_${CLICKHOUSE_TEST_UNIQUE_NAME}" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " CREATE DICTIONARY IF NOT EXISTS ${dictname} ( id UInt64, @@ -23,15 +23,15 @@ ${CLICKHOUSE_CLIENT} -nm --query " SELECT dictGet(${dictname}, 'value', 1); " -$CLICKHOUSE_CLIENT -nm --user="${username}" --query " +$CLICKHOUSE_CLIENT -m --user="${username}" --query " SELECT * FROM dictionary(${dictname}); " 2>&1 | grep -o ACCESS_DENIED | uniq -$CLICKHOUSE_CLIENT -nm --user="${username}" --query " +$CLICKHOUSE_CLIENT -m --user="${username}" --query " SELECT dictGet(${dictname}, 'value', 1); " 2>&1 | grep -o ACCESS_DENIED | uniq -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " DROP DICTIONARY IF EXISTS ${dictname}; DROP USER IF EXISTS ${username}; " diff --git a/tests/queries/0_stateless/02916_joinget_dependency.sh b/tests/queries/0_stateless/02916_joinget_dependency.sh index 6477ae8c967..ff9332cb57f 100755 --- a/tests/queries/0_stateless/02916_joinget_dependency.sh +++ b/tests/queries/0_stateless/02916_joinget_dependency.sh @@ -6,7 +6,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # We test the dependency on the DROP -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " DROP TABLE IF EXISTS Sub_distributed; DROP TABLE IF EXISTS Sub; DROP TABLE IF EXISTS Mapping; @@ -20,8 +20,8 @@ $CLICKHOUSE_CLIENT -q " DROP TABLE Mapping; " 2>&1 | grep -cm1 "HAVE_DEPENDENT_OBJECTS" -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " DROP TABLE Sub_distributed; DROP TABLE Sub; DROP TABLE Mapping; -" \ No newline at end of file +" diff --git a/tests/queries/0_stateless/02918_join_pm_lc_crash.sql b/tests/queries/0_stateless/02918_join_pm_lc_crash.sql index 123208ee981..0326acff4a6 100644 --- a/tests/queries/0_stateless/02918_join_pm_lc_crash.sql +++ b/tests/queries/0_stateless/02918_join_pm_lc_crash.sql @@ -1,31 +1,30 @@ SET joined_subquery_requires_alias = 0, join_algorithm = 'partial_merge'; -SET allow_experimental_analyzer = 0, join_use_nulls = 0; +SET enable_analyzer = 0, join_use_nulls = 0; SELECT * FROM (SELECT dummy AS val FROM system.one) JOIN (SELECT toLowCardinality(toNullable(dummy)) AS val FROM system.one GROUP BY val WITH TOTALS) USING (val); -SET allow_experimental_analyzer = 0, join_use_nulls = 1; +SET enable_analyzer = 0, join_use_nulls = 1; SELECT * FROM (SELECT dummy AS val FROM system.one) JOIN (SELECT toLowCardinality(toNullable(dummy)) AS val FROM system.one GROUP BY val WITH TOTALS) USING (val); -SET allow_experimental_analyzer = 1, join_use_nulls = 0; +SET enable_analyzer = 1, join_use_nulls = 0; SELECT * FROM (SELECT dummy AS val FROM system.one) JOIN (SELECT toLowCardinality(toNullable(dummy)) AS val FROM system.one GROUP BY val WITH TOTALS) USING (val); -SET allow_experimental_analyzer = 1, join_use_nulls = 1; +SET enable_analyzer = 1, join_use_nulls = 1; SELECT * FROM (SELECT dummy AS val FROM system.one) JOIN (SELECT toLowCardinality(toNullable(dummy)) AS val FROM system.one GROUP BY val WITH TOTALS) USING (val); - diff --git a/tests/queries/0_stateless/02918_optimize_count_for_merge_tables.sql b/tests/queries/0_stateless/02918_optimize_count_for_merge_tables.sql index 9feb2aa2ad6..f5647f01ab9 100644 --- a/tests/queries/0_stateless/02918_optimize_count_for_merge_tables.sql +++ b/tests/queries/0_stateless/02918_optimize_count_for_merge_tables.sql @@ -15,7 +15,7 @@ INSERT INTO mt2 VALUES (1); SELECT count() FROM merge; -- can use the trivial count optimization -EXPLAIN SELECT count() FROM merge settings allow_experimental_analyzer=0; +EXPLAIN SELECT count() FROM merge settings enable_analyzer=0; CREATE TABLE mt3 (id UInt64) ENGINE = TinyLog; @@ -24,7 +24,7 @@ INSERT INTO mt2 VALUES (2); SELECT count() FROM merge; -- can't use the trivial count optimization as TinyLog doesn't support it -EXPLAIN SELECT count() FROM merge settings allow_experimental_analyzer=0; +EXPLAIN SELECT count() FROM merge settings enable_analyzer=0; DROP TABLE IF EXISTS mt1; DROP TABLE IF EXISTS mt2; diff --git a/tests/queries/0_stateless/02922_respect_nulls_parser.sql b/tests/queries/0_stateless/02922_respect_nulls_parser.sql index c9a17fdfbfe..ccd67dbe676 100644 --- a/tests/queries/0_stateless/02922_respect_nulls_parser.sql +++ b/tests/queries/0_stateless/02922_respect_nulls_parser.sql @@ -13,7 +13,7 @@ SELECT toDateTimeNonExistingFunction(now()) RESPECT NULLS b; -- { serverError UN SELECT toDateTime(now()) RESPECT NULLS b; -- { serverError SYNTAX_ERROR } SELECT count() from numbers(10) where in(number, (0)) RESPECT NULLS; -- { serverError SYNTAX_ERROR } SELECT if(number > 0, number, 0) respect nulls from numbers(0); -- { serverError SYNTAX_ERROR } -WITH (x -> x + 1) AS lambda SELECT lambda(number) RESPECT NULLS FROM numbers(10) SETTINGS allow_experimental_analyzer = 1; -- { serverError SYNTAX_ERROR } +WITH (x -> x + 1) AS lambda SELECT lambda(number) RESPECT NULLS FROM numbers(10) SETTINGS enable_analyzer = 1; -- { serverError SYNTAX_ERROR } SELECT * from system.one WHERE indexHint(dummy = 1) RESPECT NULLS; -- { serverError SYNTAX_ERROR } SELECT arrayJoin([[3,4,5], [6,7], [2], [1,1]]) IGNORE NULLS; -- { serverError SYNTAX_ERROR } SELECT number, grouping(number % 2, number) RESPECT NULLS AS gr FROM numbers(10) GROUP BY GROUPING SETS ((number), (number % 2)) SETTINGS force_grouping_standard_compatibility = 0; -- { serverError SYNTAX_ERROR } diff --git a/tests/queries/0_stateless/02930_client_file_log_comment.sh b/tests/queries/0_stateless/02930_client_file_log_comment.sh index 50cd587e4b5..393bffcaf59 100755 --- a/tests/queries/0_stateless/02930_client_file_log_comment.sh +++ b/tests/queries/0_stateless/02930_client_file_log_comment.sh @@ -14,7 +14,7 @@ echo -n 'select 4242' >> "$file2" $CLICKHOUSE_CLIENT --queries-file "$file1" "$file2" <<<'select 42' $CLICKHOUSE_CLIENT --log_comment foo --queries-file /dev/stdin <<<'select 424242' -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " system flush logs; select query, log_comment from system.query_log where current_database = '$CLICKHOUSE_DATABASE' and event_date >= yesterday() and query = 'select 42' and type != 'QueryStart'; select query, log_comment from system.query_log where current_database = '$CLICKHOUSE_DATABASE' and event_date >= yesterday() and query = 'select 4242' and type != 'QueryStart'; diff --git a/tests/queries/0_stateless/02931_size_virtual_column_use_structure_from_insertion_table.sh b/tests/queries/0_stateless/02931_size_virtual_column_use_structure_from_insertion_table.sh index d9e4a2c8f8b..76ab56a4570 100755 --- a/tests/queries/0_stateless/02931_size_virtual_column_use_structure_from_insertion_table.sh +++ b/tests/queries/0_stateless/02931_size_virtual_column_use_structure_from_insertion_table.sh @@ -5,7 +5,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh echo "1,2" > $CLICKHOUSE_TEST_UNIQUE_NAME.csv -$CLICKHOUSE_LOCAL -nm -q " +$CLICKHOUSE_LOCAL -m -q " create table test (x UInt64, y UInt32, size UInt64) engine=Memory; insert into test select c1, c2, _size from file('$CLICKHOUSE_TEST_UNIQUE_NAME.csv') settings use_structure_from_insertion_table_in_table_functions=1; select * from test; diff --git a/tests/queries/0_stateless/02932_analyzer_rewrite_sum_column_and_constant.sql b/tests/queries/0_stateless/02932_analyzer_rewrite_sum_column_and_constant.sql index 43dad8eb8e0..5492d061c12 100644 --- a/tests/queries/0_stateless/02932_analyzer_rewrite_sum_column_and_constant.sql +++ b/tests/queries/0_stateless/02932_analyzer_rewrite_sum_column_and_constant.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; -- { echoOn } Select sum(number + 1) from numbers(10); diff --git a/tests/queries/0_stateless/02932_parallel_replicas_fuzzer.sql b/tests/queries/0_stateless/02932_parallel_replicas_fuzzer.sql index 3daaf36188a..e5cbe100a58 100644 --- a/tests/queries/0_stateless/02932_parallel_replicas_fuzzer.sql +++ b/tests/queries/0_stateless/02932_parallel_replicas_fuzzer.sql @@ -35,4 +35,4 @@ ORDER BY nan DESC, _CAST([0, NULL, NULL, NULL, NULL], 'Array(Nullable(UInt8))') DESC FORMAT Null -SETTINGS receive_timeout = 10., receive_data_timeout_ms = 10000, use_hedged_requests = 0, allow_suspicious_low_cardinality_types = 1, max_parallel_replicas = 3, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_parallel_reading_from_replicas = 1, parallel_replicas_for_non_replicated_merge_tree = 1, log_queries = 1, table_function_remote_max_addresses = 200, allow_experimental_analyzer = 1; +SETTINGS receive_timeout = 10., receive_data_timeout_ms = 10000, use_hedged_requests = 0, allow_suspicious_low_cardinality_types = 1, max_parallel_replicas = 3, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_parallel_reading_from_replicas = 1, parallel_replicas_for_non_replicated_merge_tree = 1, log_queries = 1, table_function_remote_max_addresses = 200, enable_analyzer = 1; diff --git a/tests/queries/0_stateless/02933_paste_join.sql b/tests/queries/0_stateless/02933_paste_join.sql index 604078d1c3a..6c5a923d028 100644 --- a/tests/queries/0_stateless/02933_paste_join.sql +++ b/tests/queries/0_stateless/02933_paste_join.sql @@ -41,10 +41,10 @@ INSERT INTO test SELECT number from numbers(6); SELECT * FROM (SELECT number FROM test) PASTE JOIN (SELECT number FROM numbers(6) ORDER BY number) SETTINGS joined_subquery_requires_alias = 0; SELECT * FROM (SELECT number FROM test PASTE JOIN (Select number FROM numbers(7))) PASTE JOIN (SELECT number FROM numbers(6) PASTE JOIN (SELECT number FROM test)) SETTINGS joined_subquery_requires_alias = 0; SELECT * FROM (SELECT number FROM test PASTE JOIN (SELECT number FROM test PASTE JOIN (Select number FROM numbers(7)))) PASTE JOIN (SELECT number FROM numbers(6) PASTE JOIN (SELECT number FROM test)) SETTINGS joined_subquery_requires_alias = 0; -SELECT * FROM (SELECT 1 AS a) PASTE JOIN (SELECT 2 AS b) PASTE JOIN (SELECT 3 AS c) SETTINGS allow_experimental_analyzer = 1; -SELECT * FROM (SELECT 1 AS a) PASTE JOIN (SELECT 2 AS b) PASTE JOIN (SELECT 3 AS a) SETTINGS allow_experimental_analyzer = 1; -- { serverError AMBIGUOUS_COLUMN_NAME } +SELECT * FROM (SELECT 1 AS a) PASTE JOIN (SELECT 2 AS b) PASTE JOIN (SELECT 3 AS c) SETTINGS enable_analyzer = 1; +SELECT * FROM (SELECT 1 AS a) PASTE JOIN (SELECT 2 AS b) PASTE JOIN (SELECT 3 AS a) SETTINGS enable_analyzer = 1; -- { serverError AMBIGUOUS_COLUMN_NAME } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; CREATE TABLE test1 (a Int32) engine=MergeTree order by a; INSERT INTO test1 SELECT * FROM numbers(2); CREATE TABLE test2 (a Int32) engine=MergeTree order by a; diff --git a/tests/queries/0_stateless/02940_system_stacktrace_optimizations.sh b/tests/queries/0_stateless/02940_system_stacktrace_optimizations.sh index 0e23bb6c42b..9d1faf301d3 100755 --- a/tests/queries/0_stateless/02940_system_stacktrace_optimizations.sh +++ b/tests/queries/0_stateless/02940_system_stacktrace_optimizations.sh @@ -9,12 +9,12 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # no message at all echo "thread = 0" -$CLICKHOUSE_CLIENT --allow_repeated_settings --send_logs_level=test -nm -q "select * from system.stack_trace where thread_id = 0" |& grep -F -o 'Send signal to' +$CLICKHOUSE_CLIENT --allow_repeated_settings --send_logs_level=test -m -q "select * from system.stack_trace where thread_id = 0" |& grep -F -o 'Send signal to' # send messages to some threads echo "thread != 0" -$CLICKHOUSE_CLIENT --allow_repeated_settings --send_logs_level=test -nm -q "select * from system.stack_trace where thread_id != 0 format Null" |& grep -F -o 'Send signal to' | grep -v 'Send signal to 0 threads (total)' +$CLICKHOUSE_CLIENT --allow_repeated_settings --send_logs_level=test -m -q "select * from system.stack_trace where thread_id != 0 format Null" |& grep -F -o 'Send signal to' | grep -v 'Send signal to 0 threads (total)' # there is no thread with comm="foo", so no signals will be sent echo "thread_name = 'foo'" -$CLICKHOUSE_CLIENT --allow_repeated_settings --send_logs_level=test -nm -q "select * from system.stack_trace where thread_name = 'foo' format Null" |& grep -F -o 'Send signal to 0 threads (total)' +$CLICKHOUSE_CLIENT --allow_repeated_settings --send_logs_level=test -m -q "select * from system.stack_trace where thread_name = 'foo' format Null" |& grep -F -o 'Send signal to 0 threads (total)' diff --git a/tests/queries/0_stateless/02943_order_by_all.sql b/tests/queries/0_stateless/02943_order_by_all.sql index 0d5e0ea52e4..4ce59e84e6c 100644 --- a/tests/queries/0_stateless/02943_order_by_all.sql +++ b/tests/queries/0_stateless/02943_order_by_all.sql @@ -13,40 +13,40 @@ INSERT INTO order_by_all VALUES ('B', 3), ('C', NULL), ('D', 1), ('A', 2); SELECT '-- no modifiers'; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT a, b FROM order_by_all ORDER BY ALL; SELECT b, a FROM order_by_all ORDER BY ALL; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT a, b FROM order_by_all ORDER BY ALL; SELECT b, a FROM order_by_all ORDER BY ALL; SELECT '-- with ASC/DESC modifiers'; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT a, b FROM order_by_all ORDER BY ALL ASC; SELECT a, b FROM order_by_all ORDER BY ALL DESC; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT a, b FROM order_by_all ORDER BY ALL ASC; SELECT a, b FROM order_by_all ORDER BY ALL DESC; SELECT '-- with NULLS FIRST/LAST modifiers'; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT b, a FROM order_by_all ORDER BY ALL NULLS FIRST; SELECT b, a FROM order_by_all ORDER BY ALL NULLS LAST; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT b, a FROM order_by_all ORDER BY ALL NULLS FIRST; SELECT b, a FROM order_by_all ORDER BY ALL NULLS LAST; SELECT '-- SELECT *'; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT * FROM order_by_all ORDER BY all; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT * FROM order_by_all ORDER BY all; DROP TABLE order_by_all; @@ -65,7 +65,7 @@ INSERT INTO order_by_all VALUES ('B', 3, 10), ('C', NULL, 40), ('D', 1, 20), ('A SELECT ' -- columns'; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT a, b, all FROM order_by_all ORDER BY all; -- { serverError UNEXPECTED_EXPRESSION } SELECT a, b, all FROM order_by_all ORDER BY all SETTINGS enable_order_by_all = false; SELECT a FROM order_by_all ORDER BY all; -- { serverError UNEXPECTED_EXPRESSION } @@ -73,7 +73,7 @@ SELECT a FROM order_by_all ORDER BY all SETTINGS enable_order_by_all = false; SELECT * FROM order_by_all ORDER BY all; -- { serverError UNEXPECTED_EXPRESSION } SELECT * FROM order_by_all ORDER BY all SETTINGS enable_order_by_all = false; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT a, b, all FROM order_by_all ORDER BY all; -- { serverError UNEXPECTED_EXPRESSION } SELECT a, b, all FROM order_by_all ORDER BY all SETTINGS enable_order_by_all = false; SELECT a FROM order_by_all ORDER BY all SETTINGS enable_order_by_all = false; @@ -86,30 +86,30 @@ SELECT * FROM order_by_all ORDER BY all SETTINGS enable_order_by_all = false; SELECT ' -- column aliases'; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT a, b AS all FROM order_by_all ORDER BY all; -- { serverError UNEXPECTED_EXPRESSION } SELECT a, b AS all FROM order_by_all ORDER BY all SETTINGS enable_order_by_all = false; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT a, b AS all FROM order_by_all ORDER BY all; -- { serverError UNEXPECTED_EXPRESSION } SELECT a, b AS all FROM order_by_all ORDER BY all SETTINGS enable_order_by_all = false; SELECT ' -- expressions'; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT format('{} {}', a, b) AS all FROM order_by_all ORDER BY all; -- { serverError UNEXPECTED_EXPRESSION } SELECT format('{} {}', a, b) AS all FROM order_by_all ORDER BY all SETTINGS enable_order_by_all = false; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT format('{} {}', a, b) AS all FROM order_by_all ORDER BY all; -- { serverError UNEXPECTED_EXPRESSION } SELECT format('{} {}', a, b) AS all FROM order_by_all ORDER BY all SETTINGS enable_order_by_all = false; SELECT ' -- ORDER BY ALL loses its special meaning when used in conjunction with other columns'; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT a, b, all FROM order_by_all ORDER BY all, a; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT a, b, all FROM order_by_all ORDER BY all, a; DROP TABLE order_by_all; diff --git a/tests/queries/0_stateless/02943_rmt_alter_metadata_merge_checksum_mismatch.sh b/tests/queries/0_stateless/02943_rmt_alter_metadata_merge_checksum_mismatch.sh index 27950866e81..44af2dbf26f 100755 --- a/tests/queries/0_stateless/02943_rmt_alter_metadata_merge_checksum_mismatch.sh +++ b/tests/queries/0_stateless/02943_rmt_alter_metadata_merge_checksum_mismatch.sh @@ -26,7 +26,7 @@ function wait_part() function restore_failpoints() { # restore entry error with failpoints (to avoid endless errors in logs) - $CLICKHOUSE_CLIENT -nm -q " + $CLICKHOUSE_CLIENT -m -q " system enable failpoint replicated_queue_unfail_entries; system sync replica $failed_replica; system disable failpoint replicated_queue_unfail_entries; @@ -34,7 +34,7 @@ function restore_failpoints() } trap restore_failpoints EXIT -$CLICKHOUSE_CLIENT -nm --insert_keeper_fault_injection_probability=0 -q " +$CLICKHOUSE_CLIENT -m --insert_keeper_fault_injection_probability=0 -q " drop table if exists data_r1; drop table if exists data_r2; @@ -45,7 +45,7 @@ $CLICKHOUSE_CLIENT -nm --insert_keeper_fault_injection_probability=0 -q " " # will fail ALTER_METADATA on one of replicas -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " system enable failpoint replicated_queue_fail_next_entry; alter table data_r1 drop index value_idx settings alter_sync=0; -- part all_0_0_0_1 @@ -80,7 +80,7 @@ fi # This will create MERGE_PARTS, on failed replica it will be fetched from source replica (since it does not have all parts to execute merge) $CLICKHOUSE_CLIENT -q "optimize table $success_replica final settings optimize_throw_if_noop=1, alter_sync=1" # part all_0_0_1_1 -$CLICKHOUSE_CLIENT -nm --insert_keeper_fault_injection_probability=0 -q " +$CLICKHOUSE_CLIENT -m --insert_keeper_fault_injection_probability=0 -q " insert into $success_replica (key) values (2); -- part all_2_2_0 -- Avoid 'Cannot select parts for optimization: Entry for part all_2_2_0 hasn't been read from the replication log yet' system sync replica $success_replica pull; diff --git a/tests/queries/0_stateless/02943_tokenbf_and_ngrambf_indexes_support_match_function.sql b/tests/queries/0_stateless/02943_tokenbf_and_ngrambf_indexes_support_match_function.sql index 42175cbb2c6..5ad54c872fa 100644 --- a/tests/queries/0_stateless/02943_tokenbf_and_ngrambf_indexes_support_match_function.sql +++ b/tests/queries/0_stateless/02943_tokenbf_and_ngrambf_indexes_support_match_function.sql @@ -41,7 +41,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 0; + enable_analyzer = 0; SELECT * FROM @@ -52,7 +52,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 1; + enable_analyzer = 1; SELECT * FROM @@ -63,7 +63,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 0; + enable_analyzer = 0; SELECT * FROM @@ -74,7 +74,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 1; + enable_analyzer = 1; SELECT '---'; @@ -96,7 +96,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 0; + enable_analyzer = 0; SELECT * FROM @@ -107,7 +107,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 1; + enable_analyzer = 1; SELECT * FROM @@ -118,7 +118,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 0; + enable_analyzer = 0; SELECT * FROM @@ -129,7 +129,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 1; + enable_analyzer = 1; SELECT '---'; @@ -150,7 +150,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 0; + enable_analyzer = 0; SELECT * FROM ( @@ -160,7 +160,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 1; + enable_analyzer = 1; SELECT * FROM @@ -171,7 +171,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 0; + enable_analyzer = 0; SELECT * FROM @@ -182,7 +182,7 @@ FROM WHERE explain LIKE '%Granules: %' SETTINGS - allow_experimental_analyzer = 1; + enable_analyzer = 1; DROP TABLE tokenbf_tab; DROP TABLE ngrambf_tab; diff --git a/tests/queries/0_stateless/02944_variant_as_common_type_analyzer.sql b/tests/queries/0_stateless/02944_variant_as_common_type_analyzer.sql index 77bed6cf796..7947c8a0ce1 100644 --- a/tests/queries/0_stateless/02944_variant_as_common_type_analyzer.sql +++ b/tests/queries/0_stateless/02944_variant_as_common_type_analyzer.sql @@ -1,7 +1,7 @@ -- this test is just like 02944_variant_as_common_type, but with different expected output, because -- analyzer changes some return types. Specifically, if(c, x, y) always casts to the common type of -- x and y, even if c is constant. -set allow_experimental_analyzer=1; +set enable_analyzer=1; set allow_experimental_variant_type=1; set use_variant_as_common_type=1; @@ -76,4 +76,3 @@ 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/02947_merge_tree_index_table_3.sh b/tests/queries/0_stateless/02947_merge_tree_index_table_3.sh index 6cb184cb1fe..ec699d974d4 100755 --- a/tests/queries/0_stateless/02947_merge_tree_index_table_3.sh +++ b/tests/queries/0_stateless/02947_merge_tree_index_table_3.sh @@ -6,7 +6,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) user_name="${CLICKHOUSE_DATABASE}_test_user_02947" -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " DROP TABLE IF EXISTS t_merge_tree_index; DROP USER IF EXISTS $user_name; @@ -44,7 +44,7 @@ $CLICKHOUSE_CLIENT --user "$user_name" --password "password" -q "SELECT arr.size $CLICKHOUSE_CLIENT --user "$user_name" --password "password" -q "SELECT b FROM mergeTreeIndex(currentDatabase(), t_merge_tree_index, with_marks = true)" 2>&1 | grep -m1 -o "ACCESS_DENIED" || echo "OK" $CLICKHOUSE_CLIENT --user "$user_name" --password "password" -q "SELECT b.mark FROM mergeTreeIndex(currentDatabase(), t_merge_tree_index, with_marks = true)" 2>&1 | grep -m1 -o "ACCESS_DENIED" || echo "OK" -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " DROP TABLE IF EXISTS t_merge_tree_index; DROP USER IF EXISTS $user_name; " diff --git a/tests/queries/0_stateless/02950_dictionary_ssd_cache_short_circuit.sh b/tests/queries/0_stateless/02950_dictionary_ssd_cache_short_circuit.sh index 3d2fe5d664d..d06aba8a4b6 100755 --- a/tests/queries/0_stateless/02950_dictionary_ssd_cache_short_circuit.sh +++ b/tests/queries/0_stateless/02950_dictionary_ssd_cache_short_circuit.sh @@ -1,15 +1,11 @@ #!/usr/bin/env bash -# Tags: no-parallel CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -n --query=" - DROP DATABASE IF EXISTS 02950_database_for_ssd_cache_dictionary; - CREATE DATABASE 02950_database_for_ssd_cache_dictionary; - - CREATE TABLE 02950_database_for_ssd_cache_dictionary.source_table +$CLICKHOUSE_CLIENT --query=" + CREATE TABLE source_table ( id UInt64, v1 String, @@ -18,9 +14,9 @@ $CLICKHOUSE_CLIENT -n --query=" ) ENGINE = TinyLog; - INSERT INTO 02950_database_for_ssd_cache_dictionary.source_table VALUES (0, 'zero', 'zero', 0), (1, 'one', NULL, 1); + INSERT INTO source_table VALUES (0, 'zero', 'zero', 0), (1, 'one', NULL, 1); - CREATE DICTIONARY 02950_database_for_ssd_cache_dictionary.ssd_cache_dictionary + CREATE DICTIONARY ssd_cache_dictionary ( id UInt64, v1 String, @@ -30,12 +26,11 @@ $CLICKHOUSE_CLIENT -n --query=" PRIMARY KEY id SOURCE(CLICKHOUSE(HOST 'localhost' PORT tcpPort() USER 'default' TABLE 'source_table')) LIFETIME(MIN 1 MAX 1000) - LAYOUT(SSD_CACHE(BLOCK_SIZE 4096 FILE_SIZE 8192 PATH '$CLICKHOUSE_USER_FILES/0d')); + LAYOUT(SSD_CACHE(BLOCK_SIZE 4096 FILE_SIZE 8192 PATH '$CLICKHOUSE_USER_FILES/${CLICKHOUSE_DATABASE}_ssd_dic')); - 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; - SELECT dictGetOrDefault('02950_database_for_ssd_cache_dictionary.ssd_cache_dictionary', 'v3', id+1, intDiv(NULL, id)) FROM 02950_database_for_ssd_cache_dictionary.source_table; + SELECT dictGetOrDefault('ssd_cache_dictionary', ('v1', 'v2'), 0, (intDiv(1, id), intDiv(1, id))) FROM source_table; + SELECT dictGetOrDefault('ssd_cache_dictionary', 'v2', id+1, intDiv(NULL, id)) FROM source_table; + SELECT dictGetOrDefault('ssd_cache_dictionary', 'v3', id+1, intDiv(NULL, id)) FROM source_table; - DROP DICTIONARY 02950_database_for_ssd_cache_dictionary.ssd_cache_dictionary; - DROP TABLE 02950_database_for_ssd_cache_dictionary.source_table; - DROP DATABASE 02950_database_for_ssd_cache_dictionary;" + DROP DICTIONARY ssd_cache_dictionary; + DROP TABLE source_table;" diff --git a/tests/queries/0_stateless/02950_distributed_initial_query_event.sh b/tests/queries/0_stateless/02950_distributed_initial_query_event.sh index 7f690a681c4..737d5c6b41e 100755 --- a/tests/queries/0_stateless/02950_distributed_initial_query_event.sh +++ b/tests/queries/0_stateless/02950_distributed_initial_query_event.sh @@ -6,7 +6,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CUR_DIR"/../shell_config.sh # CREATE TABLE local (x UInt8) Engine=Memory; # CREATE TABLE distributed ON CLUSTER cluster (p Date, i Int32) ENGINE = Distributed(test_cluster_two_shards, currentDatabase(), x) -$CLICKHOUSE_CLIENT -n -q " +$CLICKHOUSE_CLIENT -q " DROP TABLE IF EXISTS local; DROP TABLE IF EXISTS distributed; CREATE TABLE local (x UInt8) Engine=Memory; diff --git a/tests/queries/0_stateless/02952_conjunction_optimization.reference b/tests/queries/0_stateless/02952_conjunction_optimization.reference index eeadfaae21d..8af0abefd3a 100644 --- a/tests/queries/0_stateless/02952_conjunction_optimization.reference +++ b/tests/queries/0_stateless/02952_conjunction_optimization.reference @@ -32,7 +32,7 @@ QUERY id: 0 FUNCTION id: 5, function_name: and, function_type: ordinary, result_type: Bool ARGUMENTS LIST id: 6, nodes: 2 - CONSTANT id: 7, constant_value: UInt64_1, constant_value_type: Bool + CONSTANT id: 7, constant_value: Bool_1, constant_value_type: Bool FUNCTION id: 8, function_name: notIn, function_type: ordinary, result_type: UInt8 ARGUMENTS LIST id: 9, nodes: 2 diff --git a/tests/queries/0_stateless/02952_conjunction_optimization.sql b/tests/queries/0_stateless/02952_conjunction_optimization.sql index 94bc352e4c5..fb6c26a2e1f 100644 --- a/tests/queries/0_stateless/02952_conjunction_optimization.sql +++ b/tests/queries/0_stateless/02952_conjunction_optimization.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS 02952_disjunction_optimization; diff --git a/tests/queries/0_stateless/02954_analyzer_fuzz_i57086.sql b/tests/queries/0_stateless/02954_analyzer_fuzz_i57086.sql index a8029fdd3d6..886944e30c0 100644 --- a/tests/queries/0_stateless/02954_analyzer_fuzz_i57086.sql +++ b/tests/queries/0_stateless/02954_analyzer_fuzz_i57086.sql @@ -12,4 +12,4 @@ WITH ROLLUP ORDER BY count() ASC, number DESC NULLS LAST - SETTINGS limit = 2, allow_experimental_analyzer = 1; + SETTINGS limit = 2, enable_analyzer = 1; diff --git a/tests/queries/0_stateless/02955_analyzer_using_functional_args.reference.j2 b/tests/queries/0_stateless/02955_analyzer_using_functional_args.reference.j2 index e0c6a439112..95a79c56194 100644 --- a/tests/queries/0_stateless/02955_analyzer_using_functional_args.reference.j2 +++ b/tests/queries/0_stateless/02955_analyzer_using_functional_args.reference.j2 @@ -10,7 +10,7 @@ a -- { echoOn } -- USING alias column contains default in old analyzer (but both queries below should have the same result) -SELECT y * 2, s || 'a' FROM t1 FULL JOIN t2 USING (y) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT y * 2, s || 'a' FROM t1 FULL JOIN t2 USING (y) ORDER BY ALL SETTINGS enable_analyzer = 1; 738 ba 7386 aa 13332 a @@ -18,11 +18,11 @@ SELECT y * 2, s || 'a' FROM (SELECT s, y FROM t1) t1 FULL JOIN (SELECT y FROM t2 738 ba 7386 aa 13332 a -SELECT (1, *) FROM t1 FULL JOIN t2 USING (y) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT (1, *) FROM t1 FULL JOIN t2 USING (y) ORDER BY ALL SETTINGS enable_analyzer = 1; (1,369,123,'b',124) (1,3693,1231,'a',0) (1,6666,0,'',48) -SELECT (1, *) FROM (SELECT s, y FROM t1) t1 FULL JOIN (SELECT y FROM t2) t2 USING (y) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT (1, *) FROM (SELECT s, y FROM t1) t1 FULL JOIN (SELECT y FROM t2) t2 USING (y) ORDER BY ALL SETTINGS enable_analyzer = 1; (1,369,'b') (1,3693,'a') (1,6666,'') @@ -42,27 +42,27 @@ SELECT (1, t1.*, t2.*) FROM (SELECT s, y FROM t1) t1 FULL JOIN (SELECT y FROM t2 (1,'',0,6666) (1,'a',3693,0) (1,'b',369,369) -SELECT t1.z, t2.z, t3.z FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1,2,3 SETTINGS allow_experimental_analyzer = 1; +SELECT t1.z, t2.z, t3.z FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1,2,3 SETTINGS enable_analyzer = 1; 0 0 43 0 48 0 124 124 0 1232 0 1232 -SELECT * FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1,2,3 SETTINGS allow_experimental_analyzer = 1; +SELECT * FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1,2,3 SETTINGS enable_analyzer = 1; 126 0 0 42 369 123 b 124 0 3693 1231 a 0 1231 6666 0 48 0 -SELECT t1.*, t2.*, t3.* FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1,2,3 SETTINGS allow_experimental_analyzer = 1; +SELECT t1.*, t2.*, t3.* FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1,2,3 SETTINGS enable_analyzer = 1; 0 126 0 42 0 6666 48 0 123 b 369 124 0 1231 a 3693 0 1231 -SELECT (1, t1.*, t2.*, t3.*) FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1 SETTINGS allow_experimental_analyzer = 1; +SELECT (1, t1.*, t2.*, t3.*) FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1 SETTINGS enable_analyzer = 1; (1,0,'',126,0,42) (1,0,'',6666,48,0) (1,123,'b',369,124,0) (1,1231,'a',3693,0,1231) -SELECT y FROM t1 FULL JOIN t2 USING (y) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT y FROM t1 FULL JOIN t2 USING (y) ORDER BY ALL SETTINGS enable_analyzer = 1; 369 3693 6666 @@ -78,11 +78,11 @@ SELECT s FROM (SELECT s, y FROM t1) t1 FULL JOIN (SELECT y FROM t2) t2 USING (y) a b -SELECT y FROM t1 FULL JOIN t2 USING (y) PREWHERE y * 2 > 2 ORDER BY ALL SETTINGS allow_experimental_analyzer = 1, join_use_nulls = 0; +SELECT y FROM t1 FULL JOIN t2 USING (y) PREWHERE y * 2 > 2 ORDER BY ALL SETTINGS enable_analyzer = 1, join_use_nulls = 0; 369 3693 6666 -SELECT y FROM t1 FULL JOIN t2 USING (y) PREWHERE y * 2 > 2 ORDER BY ALL SETTINGS allow_experimental_analyzer = 1, join_use_nulls = 1; +SELECT y FROM t1 FULL JOIN t2 USING (y) PREWHERE y * 2 > 2 ORDER BY ALL SETTINGS enable_analyzer = 1, join_use_nulls = 1; 369 3693 6666 diff --git a/tests/queries/0_stateless/02955_analyzer_using_functional_args.sql.j2 b/tests/queries/0_stateless/02955_analyzer_using_functional_args.sql.j2 index f5b81231afe..1dde83149b6 100644 --- a/tests/queries/0_stateless/02955_analyzer_using_functional_args.sql.j2 +++ b/tests/queries/0_stateless/02955_analyzer_using_functional_args.sql.j2 @@ -24,21 +24,21 @@ SELECT count() FROM t1 INNER JOIN t2 USING (y); SELECT count() FROM t2 INNER JOIN t1 USING (y); -- `SELECT *` works differently for ALIAS columns with analyzer -SELECT * FROM t1 INNER JOIN t2 USING (y, z) SETTINGS allow_experimental_analyzer = 1; -SELECT * FROM t2 INNER JOIN t1 USING (y, z) SETTINGS allow_experimental_analyzer = 1; +SELECT * FROM t1 INNER JOIN t2 USING (y, z) SETTINGS enable_analyzer = 1; +SELECT * FROM t2 INNER JOIN t1 USING (y, z) SETTINGS enable_analyzer = 1; SELECT t2.z FROM t1 INNER JOIN t2 USING (y); -SELECT * FROM t1 INNER JOIN t3 USING (y) SETTINGS allow_experimental_analyzer = 1; -SELECT * FROM t3 INNER JOIN t1 USING (y, z) SETTINGS allow_experimental_analyzer = 1; +SELECT * FROM t1 INNER JOIN t3 USING (y) SETTINGS enable_analyzer = 1; +SELECT * FROM t3 INNER JOIN t1 USING (y, z) SETTINGS enable_analyzer = 1; SELECT s FROM t1 INNER JOIN t3 USING (y); -- { echoOn } -- USING alias column contains default in old analyzer (but both queries below should have the same result) -SELECT y * 2, s || 'a' FROM t1 FULL JOIN t2 USING (y) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT y * 2, s || 'a' FROM t1 FULL JOIN t2 USING (y) ORDER BY ALL SETTINGS enable_analyzer = 1; SELECT y * 2, s || 'a' FROM (SELECT s, y FROM t1) t1 FULL JOIN (SELECT y FROM t2) t2 USING (y) ORDER BY ALL; -SELECT (1, *) FROM t1 FULL JOIN t2 USING (y) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; -SELECT (1, *) FROM (SELECT s, y FROM t1) t1 FULL JOIN (SELECT y FROM t2) t2 USING (y) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT (1, *) FROM t1 FULL JOIN t2 USING (y) ORDER BY ALL SETTINGS enable_analyzer = 1; +SELECT (1, *) FROM (SELECT s, y FROM t1) t1 FULL JOIN (SELECT y FROM t2) t2 USING (y) ORDER BY ALL SETTINGS enable_analyzer = 1; SELECT (1, t1.*) FROM t1 FULL JOIN t2 USING (y) ORDER BY ALL; SELECT (1, t1.*) FROM (SELECT s, y FROM t1) t1 FULL JOIN (SELECT y FROM t2) t2 USING (y) ORDER BY ALL; @@ -46,19 +46,19 @@ SELECT (1, t1.*) FROM (SELECT s, y FROM t1) t1 FULL JOIN (SELECT y FROM t2) t2 U SELECT (1, t1.*, t2.*) FROM t1 FULL JOIN t2 USING (y) ORDER BY ALL; SELECT (1, t1.*, t2.*) FROM (SELECT s, y FROM t1) t1 FULL JOIN (SELECT y FROM t2) t2 USING (y) ORDER BY ALL; -SELECT t1.z, t2.z, t3.z FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1,2,3 SETTINGS allow_experimental_analyzer = 1; -SELECT * FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1,2,3 SETTINGS allow_experimental_analyzer = 1; -SELECT t1.*, t2.*, t3.* FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1,2,3 SETTINGS allow_experimental_analyzer = 1; -SELECT (1, t1.*, t2.*, t3.*) FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1 SETTINGS allow_experimental_analyzer = 1; +SELECT t1.z, t2.z, t3.z FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1,2,3 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1,2,3 SETTINGS enable_analyzer = 1; +SELECT t1.*, t2.*, t3.* FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1,2,3 SETTINGS enable_analyzer = 1; +SELECT (1, t1.*, t2.*, t3.*) FROM t1 FULL JOIN t2 USING (y) FULL JOIN t3 USING (y) ORDER BY 1 SETTINGS enable_analyzer = 1; -SELECT y FROM t1 FULL JOIN t2 USING (y) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT y FROM t1 FULL JOIN t2 USING (y) ORDER BY ALL SETTINGS enable_analyzer = 1; SELECT y FROM (SELECT s, y FROM t1) t1 FULL JOIN (SELECT y FROM t2) t2 USING (y) ORDER BY ALL; SELECT s FROM t1 FULL JOIN t2 USING (y) ORDER BY ALL; SELECT s FROM (SELECT s, y FROM t1) t1 FULL JOIN (SELECT y FROM t2) t2 USING (y) ORDER BY ALL; -SELECT y FROM t1 FULL JOIN t2 USING (y) PREWHERE y * 2 > 2 ORDER BY ALL SETTINGS allow_experimental_analyzer = 1, join_use_nulls = 0; -SELECT y FROM t1 FULL JOIN t2 USING (y) PREWHERE y * 2 > 2 ORDER BY ALL SETTINGS allow_experimental_analyzer = 1, join_use_nulls = 1; +SELECT y FROM t1 FULL JOIN t2 USING (y) PREWHERE y * 2 > 2 ORDER BY ALL SETTINGS enable_analyzer = 1, join_use_nulls = 0; +SELECT y FROM t1 FULL JOIN t2 USING (y) PREWHERE y * 2 > 2 ORDER BY ALL SETTINGS enable_analyzer = 1, join_use_nulls = 1; DROP TABLE IF EXISTS t1; DROP TABLE IF EXISTS t2; diff --git a/tests/queries/0_stateless/02955_sparkBar_alias_sparkbar.sql b/tests/queries/0_stateless/02955_sparkBar_alias_sparkbar.sql index 98259fc8029..0f658379ff9 100644 --- a/tests/queries/0_stateless/02955_sparkBar_alias_sparkbar.sql +++ b/tests/queries/0_stateless/02955_sparkBar_alias_sparkbar.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS spark_bar_test; CREATE TABLE spark_bar_test (`value` Int64, `event_date` Date) ENGINE = MergeTree ORDER BY event_date; @@ -9,4 +9,3 @@ SELECT sparkbar(9)(event_date,cnt) FROM (SELECT sum(value) as cnt, event_date FR SELECT sparkBar(9)(event_date,cnt) FROM (SELECT sum(value) as cnt, event_date FROM spark_bar_test GROUP BY event_date); DROP TABLE IF EXISTS spark_bar_test; - diff --git a/tests/queries/0_stateless/02962_join_using_bug_57894.sql b/tests/queries/0_stateless/02962_join_using_bug_57894.sql index 87aef8b1a71..96190241da5 100644 --- a/tests/queries/0_stateless/02962_join_using_bug_57894.sql +++ b/tests/queries/0_stateless/02962_join_using_bug_57894.sql @@ -9,7 +9,7 @@ CREATE TABLE r (`x` LowCardinality(Nullable(UInt32)), `s` Nullable(String)) ENGI INSERT INTO r SELECT number, toString(number) FROM numbers(2, 8); INSERT INTO r VALUES (NULL, NULL); -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT x FROM t FULL JOIN r USING (x) ORDER BY ALL ; @@ -21,7 +21,7 @@ SETTINGS join_algorithm = 'partial_merge'; SELECT x FROM t FULL JOIN r USING (x) ORDER BY ALL SETTINGS join_algorithm = 'full_sorting_merge'; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT x FROM t FULL JOIN r USING (x) ORDER BY ALL ; 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 7ebef866360..6b285d423e7 100644 --- a/tests/queries/0_stateless/02963_test_flexible_disk_configuration.sql +++ b/tests/queries/0_stateless/02963_test_flexible_disk_configuration.sql @@ -2,13 +2,33 @@ drop table if exists test; create table test (a Int32) engine = MergeTree() order by tuple() -settings disk=disk(name='test1', type = object_storage, object_storage_type = local_blob_storage, path='./02963_test1/'); +settings disk=disk(name='02963_custom_disk', type = object_storage, object_storage_type = local_blob_storage, path='./02963_test1/'); -drop table test; +drop table if exists test; +create table test (a Int32) engine = MergeTree() order by tuple() +settings disk=disk(name='02963_custom_disk', type = object_storage, object_storage_type = local_blob_storage, path='./02963_test2/'); -- { serverError BAD_ARGUMENTS } + +drop table if exists test; +create table test (a Int32) engine = MergeTree() order by tuple() +settings disk=disk(name='02963_custom_disk'); -- { serverError BAD_ARGUMENTS } + +drop table if exists test; +create table test (a Int32) engine = MergeTree() order by tuple() +settings disk='02963_custom_disk'; -- { serverError BAD_ARGUMENTS } + +drop table if exists test; +create table test (a Int32) engine = MergeTree() order by tuple() +settings disk=disk(name='s3_disk_02963'); -- { serverError BAD_ARGUMENTS } + +drop table if exists test; create table test (a Int32) engine = MergeTree() order by tuple() settings disk='s3_disk_02963'; -drop table test; +drop table if exists test; +create table test (a Int32) engine = MergeTree() order by tuple() +settings disk=disk(name='s3_disk_02963', type = object_storage, object_storage_type = local_blob_storage, path='./02963_test2/'); -- { serverError BAD_ARGUMENTS } + +drop table if exists test; create table test (a Int32) engine = MergeTree() order by tuple() settings disk=disk(name='test1', type = object_storage, @@ -17,7 +37,7 @@ settings disk=disk(name='test1', access_key_id = clickhouse, secret_access_key = clickhouse); -drop table test; +drop table if exists test; create table test (a Int32) engine = MergeTree() order by tuple() settings disk=disk(name='test2', type = object_storage, @@ -27,7 +47,7 @@ settings disk=disk(name='test2', access_key_id = clickhouse, secret_access_key = clickhouse); -drop table test; +drop table if exists test; create table test (a Int32) engine = MergeTree() order by tuple() settings disk=disk(name='test3', type = object_storage, @@ -37,8 +57,8 @@ settings disk=disk(name='test3', endpoint = 'http://localhost:11111/test/common/', access_key_id = clickhouse, secret_access_key = clickhouse); -drop table test; +drop table if exists test; create table test (a Int32) engine = MergeTree() order by tuple() settings disk=disk(name='test4', type = object_storage, @@ -48,8 +68,8 @@ settings disk=disk(name='test4', endpoint = 'http://localhost:11111/test/common/', access_key_id = clickhouse, secret_access_key = clickhouse); -drop table test; +drop table if exists test; create table test (a Int32) engine = MergeTree() order by tuple() settings disk=disk(name='test5', type = object_storage, diff --git a/tests/queries/0_stateless/02967_analyzer_fuzz.sql b/tests/queries/0_stateless/02967_analyzer_fuzz.sql index 7f2d9afcc71..dab6ec2af22 100644 --- a/tests/queries/0_stateless/02967_analyzer_fuzz.sql +++ b/tests/queries/0_stateless/02967_analyzer_fuzz.sql @@ -7,7 +7,7 @@ GROUP BY makeDateTime64(NULL, NULL, pow(NULL, '257') - '-1', '0.2147483647', 257), makeDateTime64(pow(pow(NULL, '21474836.46') - '0.0000065535', 1048577), '922337203685477580.6', NULL, NULL, pow(NULL, 1.0001) - 65536, NULL) WITH CUBE - SETTINGS allow_experimental_analyzer = 1; + SETTINGS enable_analyzer = 1; CREATE TABLE data_01223 (`key` Int) ENGINE = Memory; @@ -17,4 +17,4 @@ SELECT count(round('92233720368547758.07', '-0.01', NULL, nan, '25.7', '-9223372 FROM dist_01223 WHERE round(NULL, 1025, 1.1754943508222875e-38, NULL) WITH TOTALS - SETTINGS allow_experimental_analyzer = 1; + SETTINGS enable_analyzer = 1; diff --git a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_3.sh b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_3.sh index e49a340ab67..e954cb0e78e 100755 --- a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_3.sh +++ b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_3.sh @@ -26,7 +26,7 @@ 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=0) r on l.key = r.key order by l.key limit 10 offset 10000 -SETTINGS allow_experimental_analyzer=1, +SETTINGS enable_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" @@ -35,7 +35,7 @@ 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=0) r on l.key = r.key order by l.key limit 10 offset 10000 -SETTINGS allow_experimental_analyzer=1, send_logs_level='trace', +SETTINGS enable_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=0" 2>&1 | grep "executeQuery\|.*Coordinator: Coordination done" | @@ -51,7 +51,7 @@ 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, +SETTINGS enable_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" @@ -60,7 +60,7 @@ 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, send_logs_level='trace', +SETTINGS enable_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=0" 2>&1 | grep "executeQuery\|.*Coordinator: Coordination done" | @@ -77,7 +77,7 @@ 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 join_algorithm='full_sorting_merge') r on l.key = r.key order by l.key limit 10 offset 10000 -SETTINGS allow_experimental_analyzer=1, parallel_replicas_prefer_local_join=0, +SETTINGS enable_analyzer=1, parallel_replicas_prefer_local_join=0, 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', join_algorithm='full_sorting_merge'" @@ -86,7 +86,7 @@ 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 join_algorithm='full_sorting_merge') r on l.key = r.key order by l.key limit 10 offset 10000 -SETTINGS allow_experimental_analyzer=1, parallel_replicas_prefer_local_join=0, send_logs_level='trace', +SETTINGS enable_analyzer=1, parallel_replicas_prefer_local_join=0, 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', join_algorithm='full_sorting_merge'" 2>&1 | grep "executeQuery\|.*Coordinator: Coordination done" | @@ -102,7 +102,7 @@ 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 join_algorithm='hash') r on l.key = r.key order by l.key limit 10 offset 10000 -SETTINGS allow_experimental_analyzer=1, parallel_replicas_prefer_local_join=0, +SETTINGS enable_analyzer=1, parallel_replicas_prefer_local_join=0, 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', join_algorithm='full_sorting_merge'" @@ -111,7 +111,7 @@ 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 join_algorithm='hash') r on l.key = r.key order by l.key limit 10 offset 10000 -SETTINGS allow_experimental_analyzer=1, parallel_replicas_prefer_local_join=0, send_logs_level='trace', +SETTINGS enable_analyzer=1, parallel_replicas_prefer_local_join=0, 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', join_algorithm='full_sorting_merge'" 2>&1 | grep "executeQuery\|.*Coordinator: Coordination done" | @@ -127,7 +127,7 @@ 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 join_algorithm='full_sorting_merge') r on l.key = r.key order by l.key limit 10 offset 10000 -SETTINGS allow_experimental_analyzer=1, parallel_replicas_prefer_local_join=0, +SETTINGS enable_analyzer=1, parallel_replicas_prefer_local_join=0, 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', join_algorithm='hash'" @@ -136,7 +136,7 @@ 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 join_algorithm='full_sorting_merge') r on l.key = r.key order by l.key limit 10 offset 10000 -SETTINGS allow_experimental_analyzer=1, parallel_replicas_prefer_local_join=0, send_logs_level='trace', +SETTINGS enable_analyzer=1, parallel_replicas_prefer_local_join=0, 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', join_algorithm='hash'" 2>&1 | grep "executeQuery\|.*Coordinator: Coordination done" | diff --git a/tests/queries/0_stateless/02967_parallel_replicas_joins_and_analyzer.reference b/tests/queries/0_stateless/02967_parallel_replicas_joins_and_analyzer.reference index 100e4e500cd..36f02b2f764 100644 --- a/tests/queries/0_stateless/02967_parallel_replicas_joins_and_analyzer.reference +++ b/tests/queries/0_stateless/02967_parallel_replicas_joins_and_analyzer.reference @@ -2,7 +2,7 @@ set parallel_replicas_prefer_local_join = 0; -- A query with only INNER/LEFT joins is fully send to replicas. JOIN is executed in GLOBAL mode. -select x, y, r.y, z, rr.z, a from (select l.x, l.y, r.y, r.z as z from (select x, y from tab1 where x != 2) l any left join (select y, z from tab2 where y != 4) r on l.y = r.y) ll any left join (select z, a from tab3 where z != 8) rr on ll.z = rr.z order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +select x, y, r.y, z, rr.z, a from (select l.x, l.y, r.y, r.z as z from (select x, y from tab1 where x != 2) l any left join (select y, z from tab2 where y != 4) r on l.y = r.y) ll any left join (select z, a from tab3 where z != 8) rr on ll.z = rr.z order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 0 0 0 0 0 0 1 1 0 0 0 0 3 3 0 0 0 0 @@ -18,7 +18,7 @@ select x, y, r.y, z, rr.z, a from (select l.x, l.y, r.y, r.z as z from (select x 13 13 0 0 0 0 14 14 14 14 0 0 15 15 0 0 0 0 -explain description=0 select x, y, r.y, z, rr.z, a from (select l.x, l.y, r.y, r.z as z from (select x, y from tab1 where x != 2) l any left join (select y, z from tab2 where y != 4) r on l.y = r.y) ll any left join (select z, a from tab3 where z != 8) rr on ll.z = rr.z SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +explain description=0 select x, y, r.y, z, rr.z, a from (select l.x, l.y, r.y, r.z as z from (select x, y from tab1 where x != 2) l any left join (select y, z from tab2 where y != 4) r on l.y = r.y) ll any left join (select z, a from tab3 where z != 8) rr on ll.z = rr.z SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression ReadFromRemoteParallelReplicas -- @@ -29,7 +29,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 0 0 0 0 0 0 1 1 0 0 0 0 3 3 0 0 0 0 @@ -52,7 +52,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression Sorting Expression @@ -65,7 +65,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select sum(x), sum(y), sum(r.y), sum(z), sum(rr.z), sum(a), key from sub3 ll any left join sub4 rr on ll.z = rr.z group by x % 2 as key) select * from sub5 order by key -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 54 54 50 50 12 12 0 64 64 0 0 0 0 1 explain description=0 @@ -75,7 +75,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select sum(x), sum(y), sum(r.y), sum(z), sum(rr.z), sum(a), key from sub3 ll any left join sub4 rr on ll.z = rr.z group by x % 2 as key) select * from sub5 order by key -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression Sorting Expression @@ -90,7 +90,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 0 0 0 0 0 0 1 1 0 0 0 0 3 3 0 0 0 0 @@ -113,7 +113,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression Sorting Expression @@ -130,7 +130,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 0 0 0 0 0 0 1 1 0 0 0 0 3 3 0 0 0 0 @@ -153,7 +153,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression Sorting Expression @@ -174,7 +174,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub2 r any right join sub1 l on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, l.y, y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 0 0 0 0 0 0 6 6 6 6 0 0 8 8 8 8 0 0 @@ -197,7 +197,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub2 r any right join sub1 l on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, l.y, y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression Join Expression @@ -215,7 +215,7 @@ sub2 as (select y, z from tab2 where y != 4), sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y = r.y), sub4 as (select z, a from tab3 where z != 8), sub5 as (select z, a, x, y, r.y, ll.z from sub4 rr any right join sub3 ll on ll.z = rr.z) -select * from sub5 order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +select * from sub5 order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 0 0 0 0 0 0 0 0 1 1 0 0 0 0 3 3 0 0 @@ -237,7 +237,7 @@ sub2 as (select y, z from tab2 where y != 4), sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y = r.y), sub4 as (select z, a from tab3 where z != 8), sub5 as (select z, a, x, y, r.y, ll.z from sub4 rr any right join sub3 ll on ll.z = rr.z) -select * from sub5 order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +select * from sub5 order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression Sorting Expression @@ -258,7 +258,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 0 0 0 0 0 0 1 1 0 0 0 0 3 3 0 0 0 0 @@ -281,7 +281,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression Sorting Expression @@ -294,7 +294,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 1, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1, parallel_replicas_allow_in_with_subquery=0; +SETTINGS allow_experimental_parallel_reading_from_replicas = 1, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1, parallel_replicas_allow_in_with_subquery=0; 0 0 0 0 0 0 1 1 0 0 0 0 3 3 0 0 0 0 @@ -317,7 +317,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 1, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1, parallel_replicas_allow_in_with_subquery=0;-- { echoOn } +SETTINGS allow_experimental_parallel_reading_from_replicas = 1, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1, parallel_replicas_allow_in_with_subquery=0;-- { echoOn } Expression Sorting Expression @@ -339,7 +339,7 @@ Expression ReadFromRemoteParallelReplicas set parallel_replicas_prefer_local_join = 1; -- A query with only INNER/LEFT joins is fully send to replicas. JOIN is executed in GLOBAL mode. -select x, y, r.y, z, rr.z, a from (select l.x, l.y, r.y, r.z as z from (select x, y from tab1 where x != 2) l any left join (select y, z from tab2 where y != 4) r on l.y = r.y) ll any left join (select z, a from tab3 where z != 8) rr on ll.z = rr.z order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +select x, y, r.y, z, rr.z, a from (select l.x, l.y, r.y, r.z as z from (select x, y from tab1 where x != 2) l any left join (select y, z from tab2 where y != 4) r on l.y = r.y) ll any left join (select z, a from tab3 where z != 8) rr on ll.z = rr.z order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 0 0 0 0 0 0 1 1 0 0 0 0 3 3 0 0 0 0 @@ -355,7 +355,7 @@ select x, y, r.y, z, rr.z, a from (select l.x, l.y, r.y, r.z as z from (select x 13 13 0 0 0 0 14 14 14 14 0 0 15 15 0 0 0 0 -explain description=0 select x, y, r.y, z, rr.z, a from (select l.x, l.y, r.y, r.z as z from (select x, y from tab1 where x != 2) l any left join (select y, z from tab2 where y != 4) r on l.y = r.y) ll any left join (select z, a from tab3 where z != 8) rr on ll.z = rr.z SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +explain description=0 select x, y, r.y, z, rr.z, a from (select l.x, l.y, r.y, r.z as z from (select x, y from tab1 where x != 2) l any left join (select y, z from tab2 where y != 4) r on l.y = r.y) ll any left join (select z, a from tab3 where z != 8) rr on ll.z = rr.z SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression ReadFromRemoteParallelReplicas -- @@ -366,7 +366,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 0 0 0 0 0 0 1 1 0 0 0 0 3 3 0 0 0 0 @@ -389,7 +389,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression Sorting Expression @@ -402,7 +402,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select sum(x), sum(y), sum(r.y), sum(z), sum(rr.z), sum(a), key from sub3 ll any left join sub4 rr on ll.z = rr.z group by x % 2 as key) select * from sub5 order by key -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 54 54 50 50 12 12 0 64 64 0 0 0 0 1 explain description=0 @@ -412,7 +412,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select sum(x), sum(y), sum(r.y), sum(z), sum(rr.z), sum(a), key from sub3 ll any left join sub4 rr on ll.z = rr.z group by x % 2 as key) select * from sub5 order by key -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression Sorting Expression @@ -427,7 +427,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 0 0 0 0 0 0 1 1 0 0 0 0 3 3 0 0 0 0 @@ -450,7 +450,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression Sorting Expression @@ -467,7 +467,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 0 0 0 0 0 0 1 1 0 0 0 0 3 3 0 0 0 0 @@ -490,7 +490,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression Sorting Expression @@ -511,7 +511,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub2 r any right join sub1 l on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, l.y, y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 0 0 0 0 0 0 6 6 6 6 0 0 8 8 8 8 0 0 @@ -534,7 +534,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub2 r any right join sub1 l on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, l.y, y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression Join Expression @@ -552,7 +552,7 @@ sub2 as (select y, z from tab2 where y != 4), sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y = r.y), sub4 as (select z, a from tab3 where z != 8), sub5 as (select z, a, x, y, r.y, ll.z from sub4 rr any right join sub3 ll on ll.z = rr.z) -select * from sub5 order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +select * from sub5 order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 0 0 0 0 0 0 0 0 1 1 0 0 0 0 3 3 0 0 @@ -574,7 +574,7 @@ sub2 as (select y, z from tab2 where y != 4), sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y = r.y), sub4 as (select z, a from tab3 where z != 8), sub5 as (select z, a, x, y, r.y, ll.z from sub4 rr any right join sub3 ll on ll.z = rr.z) -select * from sub5 order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +select * from sub5 order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression Sorting Expression @@ -595,7 +595,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; 0 0 0 0 0 0 1 1 0 0 0 0 3 3 0 0 0 0 @@ -618,7 +618,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; Expression Sorting Expression @@ -631,7 +631,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 1, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1, parallel_replicas_allow_in_with_subquery=0; +SETTINGS allow_experimental_parallel_reading_from_replicas = 1, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1, parallel_replicas_allow_in_with_subquery=0; 0 0 0 0 0 0 1 1 0 0 0 0 3 3 0 0 0 0 @@ -654,7 +654,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 1, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1, parallel_replicas_allow_in_with_subquery=0; +SETTINGS allow_experimental_parallel_reading_from_replicas = 1, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1, parallel_replicas_allow_in_with_subquery=0; Expression Sorting Expression diff --git a/tests/queries/0_stateless/02967_parallel_replicas_joins_and_analyzer.sql.j2 b/tests/queries/0_stateless/02967_parallel_replicas_joins_and_analyzer.sql.j2 index 54505b147a3..775663768fe 100644 --- a/tests/queries/0_stateless/02967_parallel_replicas_joins_and_analyzer.sql.j2 +++ b/tests/queries/0_stateless/02967_parallel_replicas_joins_and_analyzer.sql.j2 @@ -17,8 +17,8 @@ insert into tab3 select number * 4, number * 4 from numbers(4); set parallel_replicas_prefer_local_join = {{use_global_in}}; -- A query with only INNER/LEFT joins is fully send to replicas. JOIN is executed in GLOBAL mode. -select x, y, r.y, z, rr.z, a from (select l.x, l.y, r.y, r.z as z from (select x, y from tab1 where x != 2) l any left join (select y, z from tab2 where y != 4) r on l.y = r.y) ll any left join (select z, a from tab3 where z != 8) rr on ll.z = rr.z order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; -explain description=0 select x, y, r.y, z, rr.z, a from (select l.x, l.y, r.y, r.z as z from (select x, y from tab1 where x != 2) l any left join (select y, z from tab2 where y != 4) r on l.y = r.y) ll any left join (select z, a from tab3 where z != 8) rr on ll.z = rr.z SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +select x, y, r.y, z, rr.z, a from (select l.x, l.y, r.y, r.z as z from (select x, y from tab1 where x != 2) l any left join (select y, z from tab2 where y != 4) r on l.y = r.y) ll any left join (select z, a from tab3 where z != 8) rr on ll.z = rr.z order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; +explain description=0 select x, y, r.y, z, rr.z, a from (select l.x, l.y, r.y, r.z as z from (select x, y from tab1 where x != 2) l any left join (select y, z from tab2 where y != 4) r on l.y = r.y) ll any left join (select z, a from tab3 where z != 8) rr on ll.z = rr.z SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; -- -- The same query with cte; with sub1 as (select x, y from tab1 where x != 2), @@ -27,7 +27,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; explain description=0 with sub1 as (select x, y from tab1 where x != 2), @@ -36,7 +36,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; -- -- GROUP BY should work up to WithMergableStage with sub1 as (select x, y from tab1 where x != 2), @@ -45,7 +45,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select sum(x), sum(y), sum(r.y), sum(z), sum(rr.z), sum(a), key from sub3 ll any left join sub4 rr on ll.z = rr.z group by x % 2 as key) select * from sub5 order by key -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; explain description=0 with sub1 as (select x, y from tab1 where x != 2), @@ -54,7 +54,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select sum(x), sum(y), sum(r.y), sum(z), sum(rr.z), sum(a), key from sub3 ll any left join sub4 rr on ll.z = rr.z group by x % 2 as key) select * from sub5 order by key -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; -- -- ORDER BY in sub3 : sub1 is fully pushed, sub3 -> WithMergableStage with sub1 as (select x, y from tab1 where x != 2), @@ -63,7 +63,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; explain description=0 with sub1 as (select x, y from tab1 where x != 2), @@ -72,7 +72,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; -- -- ORDER BY in sub1 : sub1 -> WithMergableStage with sub1 as (select x, y from tab1 where x != 2 order by y), @@ -81,7 +81,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; explain description=0 with sub1 as (select x, y from tab1 where x != 2 order by y), @@ -90,7 +90,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; -- -- RIGHT JOIN in sub3: sub3 -> WithMergableStage with sub1 as (select x, y from tab1 where x != 2), @@ -99,7 +99,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub2 r any right join sub1 l on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, l.y, y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; explain description=0 with sub1 as (select x, y from tab1 where x != 2), @@ -108,7 +108,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub2 r any right join sub1 l on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, l.y, y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; -- -- RIGHT JOIN in sub5: sub5 -> WithMergableStage with sub1 as (select x, y from tab1 where x != 2), @@ -116,7 +116,7 @@ sub2 as (select y, z from tab2 where y != 4), sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y = r.y), sub4 as (select z, a from tab3 where z != 8), sub5 as (select z, a, x, y, r.y, ll.z from sub4 rr any right join sub3 ll on ll.z = rr.z) -select * from sub5 order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +select * from sub5 order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; explain description=0 with sub1 as (select x, y from tab1 where x != 2), @@ -124,7 +124,7 @@ sub2 as (select y, z from tab2 where y != 4), sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y = r.y), sub4 as (select z, a from tab3 where z != 8), sub5 as (select z, a, x, y, r.y, ll.z from sub4 rr any right join sub3 ll on ll.z = rr.z) -select * from sub5 order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +select * from sub5 order by x SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; -- -- Subqueries for IN allowed @@ -134,7 +134,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; explain description=0 with sub1 as (select x, y from tab1 where x in (select number from numbers(16) where number != 2)), @@ -143,7 +143,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1; +SETTINGS allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1; -- -- Subqueries for IN are not allowed @@ -153,7 +153,7 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 1, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1, parallel_replicas_allow_in_with_subquery=0; +SETTINGS allow_experimental_parallel_reading_from_replicas = 1, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1, parallel_replicas_allow_in_with_subquery=0; explain description=0 with sub1 as (select x, y from tab1 where x in (select number from numbers(16) where number != 2)), @@ -162,6 +162,6 @@ sub3 as (select l.x, l.y, r.y, r.z as z from sub1 l any left join sub2 r on l.y sub4 as (select z, a from tab3 where z != 8), sub5 as (select x, y, r.y, z, rr.z, a from sub3 ll any left join sub4 rr on ll.z = rr.z) select * from sub5 order by x -SETTINGS allow_experimental_parallel_reading_from_replicas = 1, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', allow_experimental_analyzer=1, parallel_replicas_allow_in_with_subquery=0; +SETTINGS allow_experimental_parallel_reading_from_replicas = 1, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, prefer_localhost_replica = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', enable_analyzer=1, parallel_replicas_allow_in_with_subquery=0; {%- endfor %} diff --git a/tests/queries/0_stateless/02968_full_sorting_join_fuzz.reference b/tests/queries/0_stateless/02968_full_sorting_join_fuzz.reference new file mode 100644 index 00000000000..0e9f5cc1db1 --- /dev/null +++ b/tests/queries/0_stateless/02968_full_sorting_join_fuzz.reference @@ -0,0 +1,3 @@ +[NULL] [] 100 0 + +[NULL] [] 100 0 diff --git a/tests/queries/0_stateless/02968_full_sorting_join_fuzz.sql b/tests/queries/0_stateless/02968_full_sorting_join_fuzz.sql new file mode 100644 index 00000000000..85ca740cef2 --- /dev/null +++ b/tests/queries/0_stateless/02968_full_sorting_join_fuzz.sql @@ -0,0 +1,15 @@ +SET max_bytes_in_join = 0, join_algorithm = 'full_sorting_merge', max_block_size = 10240; + +DROP TABLE IF EXISTS t1; +DROP TABLE IF EXISTS t2; + +CREATE TABLE t1 (`key` UInt32, `s` String) ENGINE = MergeTree ORDER BY key; +CREATE TABLE t2 (`key` UInt32, `s` String) ENGINE = MergeTree ORDER BY key; + +INSERT INTO t1 SELECT (sipHash64(number, 'x') % 10000000) + 1 AS key, concat('val', toString(number)) AS s FROM numbers_mt(10000000); +INSERT INTO t2 SELECT (sipHash64(number, 'y') % 1000000) + 1 AS key, concat('val', toString(number)) AS s FROM numbers_mt(1000000); + +SELECT materialize([NULL]), [], 100, count(materialize(NULL)) FROM t1 ALL INNER JOIN t2 ON t1.key = t2.key PREWHERE 10 WHERE t2.key != 0 WITH TOTALS; + +DROP TABLE t1; +DROP TABLE t2; diff --git a/tests/queries/0_stateless/02969_analyzer_eliminate_injective_functions.sql b/tests/queries/0_stateless/02969_analyzer_eliminate_injective_functions.sql index 15f2550a63e..a7d0c7793db 100644 --- a/tests/queries/0_stateless/02969_analyzer_eliminate_injective_functions.sql +++ b/tests/queries/0_stateless/02969_analyzer_eliminate_injective_functions.sql @@ -1,4 +1,4 @@ -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; EXPLAIN QUERY TREE SELECT toString(toString(number + 1)) as val, count() diff --git a/tests/queries/0_stateless/02969_functions_to_subcolumns_if_null.sql b/tests/queries/0_stateless/02969_functions_to_subcolumns_if_null.sql index 361fd7c7a4e..859d26af1a1 100644 --- a/tests/queries/0_stateless/02969_functions_to_subcolumns_if_null.sql +++ b/tests/queries/0_stateless/02969_functions_to_subcolumns_if_null.sql @@ -7,21 +7,21 @@ INSERT INTO t_subcolumns_if SELECT number::Nullable(Int64) as number FROM number SELECT sum(multiIf(id IS NOT NULL, 1, 0)) FROM t_subcolumns_if -SETTINGS allow_experimental_analyzer = 1, optimize_functions_to_subcolumns = 1; +SETTINGS enable_analyzer = 1, optimize_functions_to_subcolumns = 1; SELECT sum(multiIf(id IS NULL, 1, 0)) FROM t_subcolumns_if -SETTINGS allow_experimental_analyzer = 0, optimize_functions_to_subcolumns = 1; +SETTINGS enable_analyzer = 0, optimize_functions_to_subcolumns = 1; SELECT sum(multiIf(id IS NULL, 1, 0)) FROM t_subcolumns_if -SETTINGS allow_experimental_analyzer = 1, optimize_functions_to_subcolumns = 0; +SETTINGS enable_analyzer = 1, optimize_functions_to_subcolumns = 0; SELECT sum(multiIf(id IS NULL, 1, 0)) FROM t_subcolumns_if -SETTINGS allow_experimental_analyzer = 1, optimize_functions_to_subcolumns = 1; +SETTINGS enable_analyzer = 1, optimize_functions_to_subcolumns = 1; DROP TABLE IF EXISTS t_subcolumns_if; diff --git a/tests/queries/0_stateless/02971_analyzer_remote_id.sh b/tests/queries/0_stateless/02971_analyzer_remote_id.sh index ab3c5292529..6d504ce3568 100755 --- a/tests/queries/0_stateless/02971_analyzer_remote_id.sh +++ b/tests/queries/0_stateless/02971_analyzer_remote_id.sh @@ -5,5 +5,5 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh ${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}', '${CLICKHOUSE_DATABASE}.x') SETTINGS allow_experimental_analyzer = 1" 2>&1 \ +${CLICKHOUSE_LOCAL} --query="SELECT count() FROM remote('127.0.0.{2,3}', '${CLICKHOUSE_DATABASE}.x') SETTINGS enable_analyzer = 1" 2>&1 \ | grep -av "ASan doesn't fully support makecontext/swapcontext functions" 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 index 48e5232d18b..6df2f27dbbf 100644 --- a/tests/queries/0_stateless/02971_functions_to_subcolumns_column_names.sql +++ b/tests/queries/0_stateless/02971_functions_to_subcolumns_column_names.sql @@ -5,7 +5,7 @@ CREATE TABLE t_column_names (arr Array(UInt64), n Nullable(String)) ENGINE = Mem INSERT INTO t_column_names VALUES ([1, 2, 3], 'foo'); SET optimize_functions_to_subcolumns = 1; -SET allow_experimental_analyzer = 1; +SET enable_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; diff --git a/tests/queries/0_stateless/02971_functions_to_subcolumns_map.sql b/tests/queries/0_stateless/02971_functions_to_subcolumns_map.sql index e8a752a82d5..c53a03b8ccd 100644 --- a/tests/queries/0_stateless/02971_functions_to_subcolumns_map.sql +++ b/tests/queries/0_stateless/02971_functions_to_subcolumns_map.sql @@ -5,7 +5,7 @@ CREATE TABLE t_func_to_subcolumns_map (id UInt64, m Map(String, UInt64)) ENGINE 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; +SET enable_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; diff --git a/tests/queries/0_stateless/02971_functions_to_subcolumns_variant.sql b/tests/queries/0_stateless/02971_functions_to_subcolumns_variant.sql index 511bcc44514..2612664a7b2 100644 --- a/tests/queries/0_stateless/02971_functions_to_subcolumns_variant.sql +++ b/tests/queries/0_stateless/02971_functions_to_subcolumns_variant.sql @@ -7,7 +7,7 @@ CREATE TABLE t_func_to_subcolumns_variant (id UInt64, v Variant(String, UInt64)) INSERT INTO t_func_to_subcolumns_variant VALUES (1, 'foo') (2, 111); SET optimize_functions_to_subcolumns = 1; -SET allow_experimental_analyzer = 1; +SET enable_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; diff --git a/tests/queries/0_stateless/02972_parallel_replicas_cte.sql b/tests/queries/0_stateless/02972_parallel_replicas_cte.sql index 083b0ecc5c9..767cd61216b 100644 --- a/tests/queries/0_stateless/02972_parallel_replicas_cte.sql +++ b/tests/queries/0_stateless/02972_parallel_replicas_cte.sql @@ -15,15 +15,15 @@ WITH filtered_groups AS (SELECT a FROM pr_1 WHERE a >= 100) SELECT count() FROM pr_2 INNER JOIN filtered_groups ON pr_2.a = filtered_groups.a SETTINGS allow_experimental_parallel_reading_from_replicas = 1, parallel_replicas_for_non_replicated_merge_tree = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', max_parallel_replicas = 3; --- Testing that it is disabled for allow_experimental_analyzer=0. With analyzer it will be supported (with correct result) +-- Testing that it is disabled for enable_analyzer=0. With analyzer it will be supported (with correct result) WITH filtered_groups AS (SELECT a FROM pr_1 WHERE a >= 100) SELECT count() FROM pr_2 INNER JOIN filtered_groups ON pr_2.a = filtered_groups.a -SETTINGS allow_experimental_analyzer = 0, allow_experimental_parallel_reading_from_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', max_parallel_replicas = 3; -- { serverError SUPPORT_IS_DISABLED } +SETTINGS enable_analyzer = 0, allow_experimental_parallel_reading_from_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', max_parallel_replicas = 3; -- { serverError SUPPORT_IS_DISABLED } -- Disabled for any value of allow_experimental_parallel_reading_from_replicas != 1, not just 2 WITH filtered_groups AS (SELECT a FROM pr_1 WHERE a >= 100) SELECT count() FROM pr_2 INNER JOIN filtered_groups ON pr_2.a = filtered_groups.a -SETTINGS allow_experimental_analyzer = 0, allow_experimental_parallel_reading_from_replicas = 512, parallel_replicas_for_non_replicated_merge_tree = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', max_parallel_replicas = 3; -- { serverError SUPPORT_IS_DISABLED } +SETTINGS enable_analyzer = 0, allow_experimental_parallel_reading_from_replicas = 512, parallel_replicas_for_non_replicated_merge_tree = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', max_parallel_replicas = 3; -- { serverError SUPPORT_IS_DISABLED } -- Sanitizer SELECT count() FROM pr_2 JOIN numbers(10) as pr_1 ON pr_2.a = pr_1.number diff --git a/tests/queries/0_stateless/02974_analyzer_array_join_subcolumn.sql b/tests/queries/0_stateless/02974_analyzer_array_join_subcolumn.sql index 30fb3c76c1f..1fd103d0bb0 100644 --- a/tests/queries/0_stateless/02974_analyzer_array_join_subcolumn.sql +++ b/tests/queries/0_stateless/02974_analyzer_array_join_subcolumn.sql @@ -7,16 +7,16 @@ INSERT INTO t2 VALUES (1, {'a': (1, 2), 'b': (2, 3)}),; CREATE TABLE t3 (id Int32, c Tuple(v String, pe Map(String, Tuple(a UInt64, b UInt64)))) ENGINE = MergeTree ORDER BY id; INSERT INTO t3 VALUES (1, ('A', {'a':(1, 2),'b':(2, 3)})); -SELECT pe, pe.values.a FROM (SELECT * FROM t2) ARRAY JOIN pe SETTINGS allow_experimental_analyzer = 1; -SELECT p, p.values.a FROM (SELECT * FROM t2) ARRAY JOIN pe AS p SETTINGS allow_experimental_analyzer = 1; +SELECT pe, pe.values.a FROM (SELECT * FROM t2) ARRAY JOIN pe SETTINGS enable_analyzer = 1; +SELECT p, p.values.a FROM (SELECT * FROM t2) ARRAY JOIN pe AS p SETTINGS enable_analyzer = 1; SELECT pe, pe.values.a FROM t2 ARRAY JOIN pe; SELECT p, p.values.a FROM t2 ARRAY JOIN pe AS p; -SELECT c.pe, c.pe.values.a FROM (SELECT * FROM t3) ARRAY JOIN c.pe SETTINGS allow_experimental_analyzer = 1; -SELECT p, p.values.a FROM (SELECT * FROM t3) ARRAY JOIN c.pe as p SETTINGS allow_experimental_analyzer = 1; +SELECT c.pe, c.pe.values.a FROM (SELECT * FROM t3) ARRAY JOIN c.pe SETTINGS enable_analyzer = 1; +SELECT p, p.values.a FROM (SELECT * FROM t3) ARRAY JOIN c.pe as p SETTINGS enable_analyzer = 1; -SELECT c.pe, c.pe.values.a FROM t3 ARRAY JOIN c.pe SETTINGS allow_experimental_analyzer = 1; +SELECT c.pe, c.pe.values.a FROM t3 ARRAY JOIN c.pe SETTINGS enable_analyzer = 1; SELECT p, p.values.a FROM t3 ARRAY JOIN c.pe as p; diff --git a/tests/queries/0_stateless/02974_backup_query_format_null.sh b/tests/queries/0_stateless/02974_backup_query_format_null.sh index ddba2f6de16..345a4f47b20 100755 --- a/tests/queries/0_stateless/02974_backup_query_format_null.sh +++ b/tests/queries/0_stateless/02974_backup_query_format_null.sh @@ -4,7 +4,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " DROP TABLE IF EXISTS tbl; CREATE TABLE tbl (a Int32) ENGINE = MergeTree() ORDER BY tuple(); INSERT INTO tbl VALUES (2), (80), (-12345); @@ -14,7 +14,7 @@ backup_name="Disk('backups', '${CLICKHOUSE_TEST_UNIQUE_NAME}')" ${CLICKHOUSE_CLIENT} --query "BACKUP TABLE tbl TO ${backup_name} FORMAT Null" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " DROP TABLE tbl; RESTORE ALL FROM ${backup_name} FORMAT Null " diff --git a/tests/queries/0_stateless/02981_insert_select_resize_to_max_insert_threads.sh b/tests/queries/0_stateless/02981_insert_select_resize_to_max_insert_threads.sh index e65c9654c9c..7ad5a2179f9 100755 --- a/tests/queries/0_stateless/02981_insert_select_resize_to_max_insert_threads.sh +++ b/tests/queries/0_stateless/02981_insert_select_resize_to_max_insert_threads.sh @@ -6,7 +6,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} -nq """ +${CLICKHOUSE_CLIENT} -q """ CREATE TABLE t1_local ( n UInt64, diff --git a/tests/queries/0_stateless/02984_form_format.sh b/tests/queries/0_stateless/02984_form_format.sh index 471b48e0f68..814fe1a0f2c 100755 --- a/tests/queries/0_stateless/02984_form_format.sh +++ b/tests/queries/0_stateless/02984_form_format.sh @@ -1,6 +1,4 @@ #!/bin/bash -# Tags: no-parallel - CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh @@ -8,7 +6,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) FILE_NAME="data.tmp" FORM_DATA="${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/${FILE_NAME}" mkdir -p ${USER_FILES_PATH}/${CLICKHOUSE_TEST_UNIQUE_NAME}/ -touch $FORM_DATA +touch $FORM_DATA # Simple tests echo -ne "col1=42&col2=Hello%2C%20World%21" > $FORM_DATA diff --git a/tests/queries/0_stateless/02987_logical_optimizer_pass_lowcardinality.sql b/tests/queries/0_stateless/02987_logical_optimizer_pass_lowcardinality.sql index be7689025b2..266270562cc 100644 --- a/tests/queries/0_stateless/02987_logical_optimizer_pass_lowcardinality.sql +++ b/tests/queries/0_stateless/02987_logical_optimizer_pass_lowcardinality.sql @@ -2,4 +2,4 @@ CREATE TABLE 02987_logical_optimizer_table (key Int, value Int) ENGINE=Memory(); CREATE VIEW v1 AS SELECT * FROM 02987_logical_optimizer_table; CREATE TABLE 02987_logical_optimizer_merge AS v1 ENGINE=Merge(currentDatabase(), 'v1'); -SELECT _table, key FROM 02987_logical_optimizer_merge WHERE (_table = toFixedString(toFixedString(toFixedString('v1', toNullable(2)), 2), 2)) OR ((value = toLowCardinality(toNullable(10))) AND (_table = toFixedString(toNullable('v3'), 2))) OR ((value = 20) AND (_table = toFixedString(toFixedString(toFixedString('v1', 2), 2), 2)) AND (_table = toFixedString(toLowCardinality(toFixedString('v3', 2)), 2))) SETTINGS allow_experimental_analyzer = true, join_use_nulls = true, convert_query_to_cnf = true; +SELECT _table, key FROM 02987_logical_optimizer_merge WHERE (_table = toFixedString(toFixedString(toFixedString('v1', toNullable(2)), 2), 2)) OR ((value = toLowCardinality(toNullable(10))) AND (_table = toFixedString(toNullable('v3'), 2))) OR ((value = 20) AND (_table = toFixedString(toFixedString(toFixedString('v1', 2), 2), 2)) AND (_table = toFixedString(toLowCardinality(toFixedString('v3', 2)), 2))) SETTINGS enable_analyzer = true, join_use_nulls = true, convert_query_to_cnf = true; diff --git a/tests/queries/0_stateless/02989_join_using_parent_scope.reference b/tests/queries/0_stateless/02989_join_using_parent_scope.reference index 965bfaf7c43..72678a1db4d 100644 --- a/tests/queries/0_stateless/02989_join_using_parent_scope.reference +++ b/tests/queries/0_stateless/02989_join_using_parent_scope.reference @@ -23,7 +23,7 @@ SELECT b AS a, a FROM tb JOIN tabc USING (a) ORDER BY ALL; 3 3 SELECT 1 AS b FROM tb JOIN ta USING (b); -- { serverError UNKNOWN_IDENTIFIER } -- SELECT * returns all columns from both tables in new analyzer -SELECT 3 AS a, a, * FROM tb FULL JOIN tabc USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT 3 AS a, a, * FROM tb FULL JOIN tabc USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; 3 3 0 3 abc3 3 3 1 3 abc3 3 3 2 3 abc3 @@ -31,57 +31,57 @@ SELECT 3 AS a, a, * FROM tb FULL JOIN tabc USING (a) ORDER BY ALL SETTINGS allow 3 3 \N 0 abc0 3 3 \N 1 abc1 3 3 \N 2 abc2 -SELECT b + 1 AS a, * FROM tb JOIN tabc USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT b + 1 AS a, * FROM tb JOIN tabc USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; 1 0 1 abc1 2 1 2 abc2 3 2 3 abc3 -SELECT b + 1 AS a, * FROM tb JOIN tabc USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT b + 1 AS a, * FROM tb JOIN tabc USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; 1 0 1 abc1 2 1 2 abc2 3 2 3 abc3 -SELECT b + 1 AS a, * FROM tb LEFT JOIN tabc USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT b + 1 AS a, * FROM tb LEFT JOIN tabc USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; 1 0 1 abc1 2 1 2 abc2 3 2 3 abc3 4 3 \N \N -SELECT b + 1 AS a, * FROM tb RIGHT JOIN tabc USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT b + 1 AS a, * FROM tb RIGHT JOIN tabc USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; 1 0 1 abc1 2 1 2 abc2 3 2 3 abc3 \N \N 0 abc0 -SELECT b + 1 AS a, * FROM tb FULL JOIN tabc USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT b + 1 AS a, * FROM tb FULL JOIN tabc USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; 1 0 1 abc1 2 1 2 abc2 3 2 3 abc3 4 3 \N \N \N \N 0 abc0 -SELECT b + 1 AS a, * FROM tb FULL JOIN tabc USING (a) ORDER BY ALL SETTINGS asterisk_include_alias_columns = 1, allow_experimental_analyzer = 1; +SELECT b + 1 AS a, * FROM tb FULL JOIN tabc USING (a) ORDER BY ALL SETTINGS asterisk_include_alias_columns = 1, enable_analyzer = 1; 1 0 1 abc1 2 3 2 1 2 abc2 3 4 3 2 3 abc3 4 5 4 3 \N \N \N \N \N \N 0 abc0 1 2 -SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; 1 0 1 2 2 1 2 3 3 2 3 4 -SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 LEFT JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 LEFT JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; 1 0 1 2 2 1 2 3 3 2 3 4 4 3 \N \N -SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 RIGHT JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 RIGHT JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; 1 0 1 2 2 1 2 3 3 2 3 4 \N \N 0 1 -SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 FULL JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 FULL JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; 1 0 1 2 2 1 2 3 3 2 3 4 4 3 \N \N \N \N 0 1 -SELECT b + 1 AS a, s FROM tb FULL OUTER JOIN tabc USING (a) PREWHERE a > 2 ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT b + 1 AS a, s FROM tb FULL OUTER JOIN tabc USING (a) PREWHERE a > 2 ORDER BY ALL SETTINGS enable_analyzer = 1; 3 abc3 4 \N \N abc0 @@ -94,7 +94,7 @@ SELECT a + 2 AS b FROM tb JOIN tabc USING (b) ORDER BY ALL SETTINGS analyzer_compatibility_join_using_top_level_identifier = 1; -- { serverError UNKNOWN_IDENTIFIER } -- In new analyzer with `analyzer_compatibility_join_using_top_level_identifier = 0` we get `b` from left table SELECT a + 2 AS b FROM tb JOIN tabc USING (b) ORDER BY ALL -SETTINGS analyzer_compatibility_join_using_top_level_identifier = 0, allow_experimental_analyzer = 1; +SETTINGS analyzer_compatibility_join_using_top_level_identifier = 0, enable_analyzer = 1; 2 3 4 @@ -108,7 +108,7 @@ SELECT u1.uid, u1.spouse_name as name, u2.uid, u2.name FROM users u1 JOIN users u2 USING (name) ORDER BY u1.uid FORMAT TSVWithNamesAndTypes -SETTINGS allow_experimental_analyzer = 1, analyzer_compatibility_join_using_top_level_identifier = 1; +SETTINGS enable_analyzer = 1, analyzer_compatibility_join_using_top_level_identifier = 1; uid name u2.uid u2.name Int16 String Int16 String 1231 Ksenia 6666 Ksenia @@ -116,7 +116,7 @@ SELECT u1.uid, u1.spouse_name as name, u2.uid, u2.name FROM users u1 JOIN users u2 USING (name) ORDER BY u1.uid FORMAT TSVWithNamesAndTypes -SETTINGS allow_experimental_analyzer = 1, analyzer_compatibility_join_using_top_level_identifier = 0; +SETTINGS enable_analyzer = 1, analyzer_compatibility_join_using_top_level_identifier = 0; uid name u2.uid u2.name Int16 String Int16 String 1231 Ksenia 1231 John @@ -125,7 +125,7 @@ SELECT u1.uid, u1.spouse_name as name, u2.uid, u2.name FROM users u1 JOIN users u2 USING (name) ORDER BY u1.uid FORMAT TSVWithNamesAndTypes -SETTINGS allow_experimental_analyzer = 0; +SETTINGS enable_analyzer = 0; uid name u2.uid u2.name Int16 String Int16 String 1231 Ksenia 6666 Ksenia diff --git a/tests/queries/0_stateless/02989_join_using_parent_scope.sql b/tests/queries/0_stateless/02989_join_using_parent_scope.sql index 2e4d9f097f7..4283d8b6eca 100644 --- a/tests/queries/0_stateless/02989_join_using_parent_scope.sql +++ b/tests/queries/0_stateless/02989_join_using_parent_scope.sql @@ -24,21 +24,21 @@ SELECT b AS a, a FROM tb JOIN tabc USING (a) ORDER BY ALL; SELECT 1 AS b FROM tb JOIN ta USING (b); -- { serverError UNKNOWN_IDENTIFIER } -- SELECT * returns all columns from both tables in new analyzer -SELECT 3 AS a, a, * FROM tb FULL JOIN tabc USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; -SELECT b + 1 AS a, * FROM tb JOIN tabc USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT 3 AS a, a, * FROM tb FULL JOIN tabc USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; +SELECT b + 1 AS a, * FROM tb JOIN tabc USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; -SELECT b + 1 AS a, * FROM tb JOIN tabc USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; -SELECT b + 1 AS a, * FROM tb LEFT JOIN tabc USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; -SELECT b + 1 AS a, * FROM tb RIGHT JOIN tabc USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; -SELECT b + 1 AS a, * FROM tb FULL JOIN tabc USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; -SELECT b + 1 AS a, * FROM tb FULL JOIN tabc USING (a) ORDER BY ALL SETTINGS asterisk_include_alias_columns = 1, allow_experimental_analyzer = 1; +SELECT b + 1 AS a, * FROM tb JOIN tabc USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; +SELECT b + 1 AS a, * FROM tb LEFT JOIN tabc USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; +SELECT b + 1 AS a, * FROM tb RIGHT JOIN tabc USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; +SELECT b + 1 AS a, * FROM tb FULL JOIN tabc USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; +SELECT b + 1 AS a, * FROM tb FULL JOIN tabc USING (a) ORDER BY ALL SETTINGS asterisk_include_alias_columns = 1, enable_analyzer = 1; -SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; -SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 LEFT JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; -SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 RIGHT JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; -SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 FULL JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; +SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 LEFT JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; +SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 RIGHT JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; +SELECT b + 1 AS a, * FROM (SELECT b FROM tb) t1 FULL JOIN (SELECT a, b FROM tabc) t2 USING (a) ORDER BY ALL SETTINGS enable_analyzer = 1; -SELECT b + 1 AS a, s FROM tb FULL OUTER JOIN tabc USING (a) PREWHERE a > 2 ORDER BY ALL SETTINGS allow_experimental_analyzer = 1; +SELECT b + 1 AS a, s FROM tb FULL OUTER JOIN tabc USING (a) PREWHERE a > 2 ORDER BY ALL SETTINGS enable_analyzer = 1; -- It's a default behavior for old analyzer and new with analyzer_compatibility_join_using_top_level_identifier @@ -49,7 +49,7 @@ SETTINGS analyzer_compatibility_join_using_top_level_identifier = 1; -- { server -- In new analyzer with `analyzer_compatibility_join_using_top_level_identifier = 0` we get `b` from left table SELECT a + 2 AS b FROM tb JOIN tabc USING (b) ORDER BY ALL -SETTINGS analyzer_compatibility_join_using_top_level_identifier = 0, allow_experimental_analyzer = 1; +SETTINGS analyzer_compatibility_join_using_top_level_identifier = 0, enable_analyzer = 1; -- This is example where query may return different results with different `analyzer_compatibility_join_using_top_level_identifier` @@ -63,19 +63,19 @@ SELECT u1.uid, u1.spouse_name as name, u2.uid, u2.name FROM users u1 JOIN users u2 USING (name) ORDER BY u1.uid FORMAT TSVWithNamesAndTypes -SETTINGS allow_experimental_analyzer = 1, analyzer_compatibility_join_using_top_level_identifier = 1; +SETTINGS enable_analyzer = 1, analyzer_compatibility_join_using_top_level_identifier = 1; SELECT u1.uid, u1.spouse_name as name, u2.uid, u2.name FROM users u1 JOIN users u2 USING (name) ORDER BY u1.uid FORMAT TSVWithNamesAndTypes -SETTINGS allow_experimental_analyzer = 1, analyzer_compatibility_join_using_top_level_identifier = 0; +SETTINGS enable_analyzer = 1, analyzer_compatibility_join_using_top_level_identifier = 0; SELECT u1.uid, u1.spouse_name as name, u2.uid, u2.name FROM users u1 JOIN users u2 USING (name) ORDER BY u1.uid FORMAT TSVWithNamesAndTypes -SETTINGS allow_experimental_analyzer = 0; +SETTINGS enable_analyzer = 0; DROP TABLE IF EXISTS users; diff --git a/tests/queries/0_stateless/02991_count_rewrite_analyzer.sql b/tests/queries/0_stateless/02991_count_rewrite_analyzer.sql index b11aeedd225..bb0d3a1a9d1 100644 --- a/tests/queries/0_stateless/02991_count_rewrite_analyzer.sql +++ b/tests/queries/0_stateless/02991_count_rewrite_analyzer.sql @@ -1,5 +1,5 @@ -- Regression test for https://github.com/ClickHouse/ClickHouse/issues/59919 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT toTypeName(sum(toNullable('a') IN toNullable('a'))) AS x; SELECT toTypeName(count(toNullable('a') IN toNullable('a'))) AS x; 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 2a9e673d7bc..efe18918c93 100644 --- a/tests/queries/0_stateless/02992_analyzer_group_by_const.sql +++ b/tests/queries/0_stateless/02992_analyzer_group_by_const.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; -- Illegal column String of first argument of function concatWithSeparator. Must be a constant String. SELECT concatWithSeparator('a', 'b') GROUP BY 'a'; 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/02994_merge_tree_mutations_cleanup.sql.j2 b/tests/queries/0_stateless/02994_merge_tree_mutations_cleanup.sql.j2 index 1b9be79dbe4..65601cd39be 100644 --- a/tests/queries/0_stateless/02994_merge_tree_mutations_cleanup.sql.j2 +++ b/tests/queries/0_stateless/02994_merge_tree_mutations_cleanup.sql.j2 @@ -1,3 +1,5 @@ +-- Tags: no-fasttest +-- no-fasttest: Slow wait drop table if exists data_rmt; drop table if exists data_mt; diff --git a/tests/queries/0_stateless/02995_forget_partition.sh b/tests/queries/0_stateless/02995_forget_partition.sh index 6fa0b96e90d..e9d4590cce6 100755 --- a/tests/queries/0_stateless/02995_forget_partition.sh +++ b/tests/queries/0_stateless/02995_forget_partition.sh @@ -1,5 +1,6 @@ #!/usr/bin/env bash -# Tags: zookeeper, no-replicated-database +# Tags: zookeeper, no-replicated-database, no-fasttest +# no-fasttest: Slow wait CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02995_index_1.sh b/tests/queries/0_stateless/02995_index_1.sh index 128697fd0fe..76be6341d53 100755 --- a/tests/queries/0_stateless/02995_index_1.sh +++ b/tests/queries/0_stateless/02995_index_1.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage, no-distributed-cache CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -39,6 +39,6 @@ WHERE a >= (round(pow(sipHash64(1, try), 1 / (3 + sipHash64(2, try) % 8))) AS a1 AND c <= (c1 + round(pow(sipHash64(11, try), 1 / (3 + sipHash64(12, try) % 8))))::String HAVING count() > 0; " -done | ${CLICKHOUSE_CLIENT} +done | ${CLICKHOUSE_CLIENT} ${CLICKHOUSE_CLIENT} "DROP TABLE test" diff --git a/tests/queries/0_stateless/02995_index_10.sh b/tests/queries/0_stateless/02995_index_10.sh index c15ba00fd05..e7e7d3c3b42 100755 --- a/tests/queries/0_stateless/02995_index_10.sh +++ b/tests/queries/0_stateless/02995_index_10.sh @@ -1,11 +1,11 @@ #!/usr/bin/env bash -# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage, no-distributed-cache CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} " +${CLICKHOUSE_CLIENT} -q " 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; @@ -37,8 +37,9 @@ WHERE a >= (round(pow(sipHash64(1, try), 1 / (3 + sipHash64(2, try) % 8))) AS a1 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} +HAVING count() > 0 +SETTINGS trace_profile_events=0 -- test is too slow with profiling +;" +done | ${CLICKHOUSE_CLIENT} -${CLICKHOUSE_CLIENT} "DROP TABLE test" +${CLICKHOUSE_CLIENT} -q "DROP TABLE test" diff --git a/tests/queries/0_stateless/02995_index_5.sh b/tests/queries/0_stateless/02995_index_5.sh index 80f75a532e3..58e53a2c481 100755 --- a/tests/queries/0_stateless/02995_index_5.sh +++ b/tests/queries/0_stateless/02995_index_5.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage, no-distributed-cache CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -39,6 +39,6 @@ WHERE a >= (round(pow(sipHash64(1, try), 1 / (3 + sipHash64(2, try) % 8))) AS a1 AND c <= (c1 + round(pow(sipHash64(11, try), 1 / (3 + sipHash64(12, try) % 8))))::String HAVING count() > 0; " -done | ${CLICKHOUSE_CLIENT} +done | ${CLICKHOUSE_CLIENT} ${CLICKHOUSE_CLIENT} "DROP TABLE test" diff --git a/tests/queries/0_stateless/02995_index_6.sh b/tests/queries/0_stateless/02995_index_6.sh index e90387c7c0c..ef35f0e1126 100755 --- a/tests/queries/0_stateless/02995_index_6.sh +++ b/tests/queries/0_stateless/02995_index_6.sh @@ -1,5 +1,5 @@ #!/usr/bin/env bash -# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage +# Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-sanitize-coverage, no-distributed-cache CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh diff --git a/tests/queries/0_stateless/02996_analyzer_prewhere_projection.sql b/tests/queries/0_stateless/02996_analyzer_prewhere_projection.sql index 9d676001010..66f7c37a893 100644 --- a/tests/queries/0_stateless/02996_analyzer_prewhere_projection.sql +++ b/tests/queries/0_stateless/02996_analyzer_prewhere_projection.sql @@ -4,4 +4,4 @@ CREATE TABLE t__fuzz_0 (`i` LowCardinality(Int32), `j` Int32, `k` Int32, PROJECT INSERT INTO t__fuzz_0 Select number, number, number FROM numbers(100); SELECT * FROM t__fuzz_0 PREWHERE 7 AND (i < 2147483647) AND (j IN (2147483646, -2, 1)) -SETTINGS allow_experimental_analyzer = true; +SETTINGS enable_analyzer = true; diff --git a/tests/queries/0_stateless/02998_analyzer_prewhere_report.sql b/tests/queries/0_stateless/02998_analyzer_prewhere_report.sql index b3027181901..5a7cab854e9 100644 --- a/tests/queries/0_stateless/02998_analyzer_prewhere_report.sql +++ b/tests/queries/0_stateless/02998_analyzer_prewhere_report.sql @@ -15,4 +15,4 @@ SELECT arrayFilter(x -> (x IN (2, 3)), data) AS filtered FROM hits WHERE arrayExists(x -> (x IN (2, 3)), data) -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; diff --git a/tests/queries/0_stateless/02998_analyzer_secret_args_tree_node.reference b/tests/queries/0_stateless/02998_analyzer_secret_args_tree_node.reference index ffd2f68990b..67ef38093d4 100644 --- a/tests/queries/0_stateless/02998_analyzer_secret_args_tree_node.reference +++ b/tests/queries/0_stateless/02998_analyzer_secret_args_tree_node.reference @@ -1,5 +1,5 @@ -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; EXPLAIN QUERY TREE SELECT encrypt('aes-256-ofb', (SELECT 'qwerty'), '12345678901234567890123456789012'), encrypt('aes-256-ofb', (SELECT 'asdf'), '12345678901234567890123456789012'); QUERY id: 0 PROJECTION COLUMNS diff --git a/tests/queries/0_stateless/02998_analyzer_secret_args_tree_node.sql b/tests/queries/0_stateless/02998_analyzer_secret_args_tree_node.sql index f40b40b6c8c..a216f886f8a 100644 --- a/tests/queries/0_stateless/02998_analyzer_secret_args_tree_node.sql +++ b/tests/queries/0_stateless/02998_analyzer_secret_args_tree_node.sql @@ -2,7 +2,7 @@ -- encrypt function doesn't exist in the fastest build -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; EXPLAIN QUERY TREE SELECT encrypt('aes-256-ofb', (SELECT 'qwerty'), '12345678901234567890123456789012'), encrypt('aes-256-ofb', (SELECT 'asdf'), '12345678901234567890123456789012'); diff --git a/tests/queries/0_stateless/02998_primary_key_skip_columns.sql b/tests/queries/0_stateless/02998_primary_key_skip_columns.sql index ee558996b52..1abe692a7a4 100644 --- a/tests/queries/0_stateless/02998_primary_key_skip_columns.sql +++ b/tests/queries/0_stateless/02998_primary_key_skip_columns.sql @@ -1,4 +1,5 @@ --- Tags: no-asan, no-tsan, no-msan, no-ubsan, no-random-settings, no-debug +-- Tags: no-asan, no-tsan, no-msan, no-ubsan, no-random-settings, no-debug, no-fasttest +-- no-fasttest: Low index granularity and too many parts makes the test slow DROP TABLE IF EXISTS test; diff --git a/tests/queries/0_stateless/02998_projection_after_attach_partition.reference b/tests/queries/0_stateless/02998_projection_after_attach_partition.reference index 1cb984f0f34..93cae129842 100644 --- a/tests/queries/0_stateless/02998_projection_after_attach_partition.reference +++ b/tests/queries/0_stateless/02998_projection_after_attach_partition.reference @@ -19,12 +19,12 @@ INSERT INTO visits_order SELECT 2, 'user2', number from numbers(1, 10); INSERT INTO visits_order SELECT 2, 'another_user2', number*2 from numbers(1, 10); INSERT INTO visits_order SELECT 2, 'yet_another_user2', number*3 from numbers(1, 10); ALTER TABLE visits_order_dst ATTACH PARTITION ID '2' FROM visits_order; -SET allow_experimental_analyzer=0; +SET enable_analyzer=0; EXPLAIN SELECT * FROM visits_order_dst WHERE user_name='another_user2'; Expression ((Projection + Before ORDER BY)) Filter ReadFromMergeTree (user_name_projection) -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; EXPLAIN SELECT * FROM visits_order_dst WHERE user_name='another_user2'; Expression ((Project names + Projection)) Filter diff --git a/tests/queries/0_stateless/02998_projection_after_attach_partition.sql b/tests/queries/0_stateless/02998_projection_after_attach_partition.sql index 4e0121dafe9..72ee4ad81e8 100644 --- a/tests/queries/0_stateless/02998_projection_after_attach_partition.sql +++ b/tests/queries/0_stateless/02998_projection_after_attach_partition.sql @@ -25,10 +25,10 @@ INSERT INTO visits_order SELECT 2, 'yet_another_user2', number*3 from numbers(1, ALTER TABLE visits_order_dst ATTACH PARTITION ID '2' FROM visits_order; -SET allow_experimental_analyzer=0; +SET enable_analyzer=0; EXPLAIN SELECT * FROM visits_order_dst WHERE user_name='another_user2'; -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; EXPLAIN SELECT * FROM visits_order_dst WHERE user_name='another_user2'; diff --git a/tests/queries/0_stateless/02999_analyzer_preimage_null.sql b/tests/queries/0_stateless/02999_analyzer_preimage_null.sql index 07d3a0f69c1..0fc61cf0836 100644 --- a/tests/queries/0_stateless/02999_analyzer_preimage_null.sql +++ b/tests/queries/0_stateless/02999_analyzer_preimage_null.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SET optimize_time_filter_with_preimage=1; CREATE TABLE date_t__fuzz_0 (`id` UInt32, `value1` String, `date1` Date) ENGINE = ReplacingMergeTree ORDER BY id SETTINGS allow_nullable_key=1; diff --git a/tests/queries/0_stateless/03001_analyzer_nullable_nothing.sql b/tests/queries/0_stateless/03001_analyzer_nullable_nothing.sql index 32c378ebf0a..c1c7ca87b5f 100644 --- a/tests/queries/0_stateless/03001_analyzer_nullable_nothing.sql +++ b/tests/queries/0_stateless/03001_analyzer_nullable_nothing.sql @@ -3,4 +3,4 @@ SELECT count(_CAST(NULL, 'Nullable(Nothing)')), round(avg(_CAST(NULL, 'Nullable(Nothing)'))) AS k FROM numbers(256) - SETTINGS allow_experimental_analyzer = 1; + SETTINGS enable_analyzer = 1; diff --git a/tests/queries/0_stateless/03002_analyzer_prewhere.sql b/tests/queries/0_stateless/03002_analyzer_prewhere.sql index 0edf16f1cbe..976e7cab73d 100644 --- a/tests/queries/0_stateless/03002_analyzer_prewhere.sql +++ b/tests/queries/0_stateless/03002_analyzer_prewhere.sql @@ -1,4 +1,4 @@ -SET max_threads = 16, receive_timeout = 10., receive_data_timeout_ms = 10000, allow_suspicious_low_cardinality_types = true, enable_positional_arguments = false, log_queries = true, table_function_remote_max_addresses = 200, any_join_distinct_right_table_keys = true, joined_subquery_requires_alias = false, allow_experimental_analyzer = true, max_execution_time = 10., max_memory_usage = 10000000000, log_comment = '/workspace/ch/tests/queries/0_stateless/01710_projection_in_index.sql', send_logs_level = 'fatal', enable_optimize_predicate_expression = false, prefer_localhost_replica = true, allow_introspection_functions = true, optimize_functions_to_subcolumns = false, transform_null_in = true, optimize_use_projections = true, allow_deprecated_syntax_for_merge_tree = true, parallelize_output_from_storages = false; +SET max_threads = 16, receive_timeout = 10., receive_data_timeout_ms = 10000, allow_suspicious_low_cardinality_types = true, enable_positional_arguments = false, log_queries = true, table_function_remote_max_addresses = 200, any_join_distinct_right_table_keys = true, joined_subquery_requires_alias = false, enable_analyzer = true, max_execution_time = 10., max_memory_usage = 10000000000, log_comment = '/workspace/ch/tests/queries/0_stateless/01710_projection_in_index.sql', send_logs_level = 'fatal', enable_optimize_predicate_expression = false, prefer_localhost_replica = true, allow_introspection_functions = true, optimize_functions_to_subcolumns = false, transform_null_in = true, optimize_use_projections = true, allow_deprecated_syntax_for_merge_tree = true, parallelize_output_from_storages = false; CREATE TABLE t__fuzz_0 (`i` Int32, `j` Nullable(Int32), `k` Int32, PROJECTION p (SELECT * ORDER BY j)) ENGINE = MergeTree ORDER BY i SETTINGS index_granularity = 1, allow_nullable_key=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 new file mode 100644 index 00000000000..b19d389d8d0 --- /dev/null +++ b/tests/queries/0_stateless/03002_part_log_rmt_fetch_merge_error.reference @@ -0,0 +1,10 @@ +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.sh b/tests/queries/0_stateless/03002_part_log_rmt_fetch_merge_error.sh new file mode 100755 index 00000000000..e58c542b8ac --- /dev/null +++ b/tests/queries/0_stateless/03002_part_log_rmt_fetch_merge_error.sh @@ -0,0 +1,54 @@ +#!/usr/bin/env bash +# Tags: no-replicated-database, no-parallel, no-shared-merge-tree +# SMT: The merge process is completely different from RMT + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +set -e + +function wait_until() +{ + local q=$1 && shift + while [ "$($CLICKHOUSE_CLIENT -nm -q "$q")" != "1" ]; do + # too frequent FLUSH LOGS is too costly + sleep 2 + done +} + +$CLICKHOUSE_CLIENT -nm -q " + drop table if exists rmt_master; + drop table if exists rmt_slave; + + create table rmt_master (key Int) engine=ReplicatedMergeTree('/clickhouse/{database}', 'master') order by key settings always_fetch_merged_part=0, old_parts_lifetime=600; + -- always_fetch_merged_part=1, consider this table as a 'slave' + create table rmt_slave (key Int) engine=ReplicatedMergeTree('/clickhouse/{database}', 'slave') order by key settings always_fetch_merged_part=1, old_parts_lifetime=600; + + insert into rmt_master values (1); + + system sync replica rmt_master; + system sync replica rmt_slave; + system stop replicated sends rmt_master; + optimize table rmt_master final settings alter_sync=1, optimize_throw_if_noop=1; +" + +$CLICKHOUSE_CLIENT -nm -q " + 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; +" +# wait until rmt_slave will fetch the part and reflect this error in system.part_log +wait_until "system flush logs; select count()>0 from system.part_log where table = 'rmt_slave' and database = '$CLICKHOUSE_DATABASE' and error > 0" +$CLICKHOUSE_CLIENT -nm -q " + 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 new file mode 100644 index 00000000000..aac9e7527d1 --- /dev/null +++ b/tests/queries/0_stateless/03002_part_log_rmt_fetch_mutate_error.reference @@ -0,0 +1,10 @@ +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.sh b/tests/queries/0_stateless/03002_part_log_rmt_fetch_mutate_error.sh new file mode 100755 index 00000000000..cc8f53aafb9 --- /dev/null +++ b/tests/queries/0_stateless/03002_part_log_rmt_fetch_mutate_error.sh @@ -0,0 +1,64 @@ +#!/usr/bin/env bash +# Tags: no-replicated-database, no-parallel, no-shared-merge-tree +# SMT: The merge process is completely different from RMT + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +# shellcheck source=./mergetree_mutations.lib +. "$CUR_DIR"/mergetree_mutations.lib + +set -e + +function wait_until() +{ + local q=$1 && shift + while [ "$($CLICKHOUSE_CLIENT -nm -q "$q")" != "1" ]; do + # too frequent FLUSH LOGS is too costly + sleep 2 + done +} + +$CLICKHOUSE_CLIENT -nm -q " + drop table if exists rmt_master; + drop table if exists rmt_slave; + + create table rmt_master (key Int) engine=ReplicatedMergeTree('/clickhouse/{database}', 'master') order by tuple() settings always_fetch_merged_part=0, old_parts_lifetime=600; + -- prefer_fetch_merged_part_*_threshold=0, consider this table as a 'slave' + create table rmt_slave (key Int) engine=ReplicatedMergeTree('/clickhouse/{database}', 'slave') order by tuple() settings prefer_fetch_merged_part_time_threshold=0, prefer_fetch_merged_part_size_threshold=0, old_parts_lifetime=600; + + insert into rmt_master values (1); + + system sync replica rmt_master; + system sync replica rmt_slave; + system stop replicated sends rmt_master; + system stop pulling replication log rmt_slave; + alter table rmt_master update key=key+100 where 1 settings alter_sync=1; +" + +# first we need to make the rmt_master execute mutation so that it will have +# the part, and rmt_slave will consider it instead of performing mutation on +# it's own, otherwise prefer_fetch_merged_part_*_threshold will be simply ignored +wait_for_mutation rmt_master 0000000000 +$CLICKHOUSE_CLIENT -nm -q "system start pulling replication log rmt_slave" +# and wait until rmt_slave to fetch the part and reflect this error in system.part_log +wait_until "system flush logs; select count()>0 from system.part_log where table = 'rmt_slave' and database = '$CLICKHOUSE_DATABASE' and error > 0" +$CLICKHOUSE_CLIENT -nm -q " + 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; +" +wait_for_mutation rmt_slave 0000000000 +$CLICKHOUSE_CLIENT -nm -q " + 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_analyzer_setting.sql b/tests/queries/0_stateless/03003_analyzer_setting.sql index 2e5cab71277..3dbdaed4ad0 100644 --- a/tests/queries/0_stateless/03003_analyzer_setting.sql +++ b/tests/queries/0_stateless/03003_analyzer_setting.sql @@ -1,9 +1,9 @@ CREATE TABLE test (dummy Int8) ENGINE = Distributed(test_cluster_two_shards, 'system', 'one'); -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; -SELECT * FROM (SELECT * FROM test SETTINGS allow_experimental_analyzer = 1); -- { serverError INCORRECT_QUERY } +SELECT * FROM (SELECT * FROM test SETTINGS enable_analyzer = 1); -- { serverError INCORRECT_QUERY } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -SELECT * FROM (SELECT * FROM test SETTINGS allow_experimental_analyzer = 0); -- { serverError INCORRECT_QUERY } +SELECT * FROM (SELECT * FROM test SETTINGS enable_analyzer = 0); -- { serverError INCORRECT_QUERY } diff --git a/tests/queries/0_stateless/03003_functions_to_subcolumns_final.sql b/tests/queries/0_stateless/03003_functions_to_subcolumns_final.sql index 3fe29139c5f..b2ca478daa4 100644 --- a/tests/queries/0_stateless/03003_functions_to_subcolumns_final.sql +++ b/tests/queries/0_stateless/03003_functions_to_subcolumns_final.sql @@ -2,7 +2,7 @@ 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 enable_analyzer = 1; SET optimize_on_insert = 0; CREATE TABLE t_length_1 (id UInt64, arr Array(UInt64)) ENGINE = ReplacingMergeTree ORDER BY id; diff --git a/tests/queries/0_stateless/03006_join_on_inequal_expression_2.sql.j2 b/tests/queries/0_stateless/03006_join_on_inequal_expression_2.sql.j2 index f15fced161c..a09dc18739f 100644 --- a/tests/queries/0_stateless/03006_join_on_inequal_expression_2.sql.j2 +++ b/tests/queries/0_stateless/03006_join_on_inequal_expression_2.sql.j2 @@ -18,7 +18,7 @@ INSERT INTO t1 (key, a, attr) VALUES (1, 10, 'alpha'), (2, 15, 'beta'), (3, 20, INSERT INTO t2 (key, a, attr) VALUES (1, 5, 'ALPHA'), (2, 10, 'beta'), (4, 25, 'delta'); -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SET allow_experimental_join_condition=1; SET join_use_nulls=0; -- { echoOn } diff --git a/tests/queries/0_stateless/03006_join_on_inequal_expression_3.sql.j2 b/tests/queries/0_stateless/03006_join_on_inequal_expression_3.sql.j2 index a97153ce3aa..009ae10e4ff 100644 --- a/tests/queries/0_stateless/03006_join_on_inequal_expression_3.sql.j2 +++ b/tests/queries/0_stateless/03006_join_on_inequal_expression_3.sql.j2 @@ -18,7 +18,7 @@ INSERT INTO t1 (key, a, attr) VALUES (1, 10, 'alpha'), (2, 15, 'beta'), (3, 20, INSERT INTO t2 (key, a, attr) VALUES (1, 5, 'ALPHA'), (2, 10, 'beta'), (4, 25, 'delta'); -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SET allow_experimental_join_condition=1; SET join_use_nulls=0; -- { echoOn } diff --git a/tests/queries/0_stateless/03006_join_on_inequal_expression_4.sql.j2 b/tests/queries/0_stateless/03006_join_on_inequal_expression_4.sql.j2 index 3235019821b..37eaaa8ab5c 100644 --- a/tests/queries/0_stateless/03006_join_on_inequal_expression_4.sql.j2 +++ b/tests/queries/0_stateless/03006_join_on_inequal_expression_4.sql.j2 @@ -15,7 +15,7 @@ CREATE TABLE t2 ( INSERT INTO t1 (key, a) VALUES (1, 10), (2, 15), (3, 20); INSERT INTO t2 (key, a) VALUES (1, 5), (2, 10), (4, 25); -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SET allow_experimental_join_condition=1; SET join_algorithm='hash'; -- { echoOn } diff --git a/tests/queries/0_stateless/03006_join_on_inequal_expression_fast.reference b/tests/queries/0_stateless/03006_join_on_inequal_expression_fast.reference index 46f24f73356..a70e70ef7e9 100644 --- a/tests/queries/0_stateless/03006_join_on_inequal_expression_fast.reference +++ b/tests/queries/0_stateless/03006_join_on_inequal_expression_fast.reference @@ -382,6 +382,253 @@ key1 e 5 5 5 key1 C 3 4 5 key2 a2 1 1 1 0 0 \N key4 f 2 3 4 key4 F 1 1 1 SET join_algorithm='hash'; +SELECT t1.*, t2.* FROM t1 LEFT ANY JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 c 3 2 1 key1 C 3 4 5 +key1 d 4 7 2 key1 D 4 1 6 +key1 e 5 5 5 0 0 \N +key2 a2 1 1 1 0 0 \N +key4 f 2 3 4 key4 F 1 1 1 +SELECT t1.*, t2.* from t1 LEFT ANY JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 c 3 2 1 key1 B 2 1 2 +key1 d 4 7 2 key1 D 4 1 6 +key1 e 5 5 5 0 0 \N +key2 a2 1 1 1 0 0 \N +key4 f 2 3 4 0 0 \N +SELECT t1.*, t2.* from t1 LEFT ANY JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 B 2 1 2 +key1 b 2 3 2 key1 C 3 4 5 +key1 c 3 2 1 key1 D 4 1 6 +key1 d 4 7 2 0 0 \N +key1 e 5 5 5 0 0 \N +key2 a2 1 1 1 0 0 \N +key4 f 2 3 4 0 0 \N +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 LEFT ANY JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +1 1 1 1 1 1 +SELECT t1.*, t2.* FROM t1 LEFT SEMI JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 c 3 2 1 key1 C 3 4 5 +key1 d 4 7 2 key1 D 4 1 6 +key4 f 2 3 4 key4 F 1 1 1 +SELECT t1.*, t2.* from t1 LEFT SEMI JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 c 3 2 1 key1 B 2 1 2 +key1 d 4 7 2 key1 D 4 1 6 +SELECT t1.*, t2.* from t1 LEFT SEMI JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 B 2 1 2 +key1 b 2 3 2 key1 C 3 4 5 +key1 c 3 2 1 key1 D 4 1 6 +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 LEFT SEMI JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +1 1 1 1 1 1 +SELECT t1.*, t2.* FROM t1 LEFT ANTI JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 e 5 5 5 key1 0 0 \N +key2 a2 1 1 1 key2 0 0 \N +SELECT t1.*, t2.* from t1 LEFT ANTI JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 e 5 5 5 key1 0 0 \N +key2 a2 1 1 1 key2 0 0 \N +key4 f 2 3 4 key4 0 0 \N +SELECT t1.*, t2.* from t1 LEFT ANTI JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 d 4 7 2 key1 0 0 \N +key1 e 5 5 5 key1 0 0 \N +key2 a2 1 1 1 key2 0 0 \N +key4 f 2 3 4 key4 0 0 \N +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 LEFT ANTI JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +SELECT t1.*, t2.* FROM t1 RIGHT ANY JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); + 0 0 \N key3 a3 1 1 1 +key1 a 1 1 2 key1 A 1 2 1 +key1 a 1 1 2 key1 B 2 1 2 +key1 a 1 1 2 key1 C 3 4 5 +key1 a 1 1 2 key1 D 4 1 6 +key4 f 2 3 4 key4 F 1 1 1 +SELECT t1.*, t2.* from t1 RIGHT ANY JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); + 0 0 \N key3 a3 1 1 1 + 0 0 \N key4 F 1 1 1 +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 b 2 3 2 key1 C 3 4 5 +key1 d 4 7 2 key1 D 4 1 6 +SELECT t1.*, t2.* from t1 RIGHT ANY JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); + 0 0 \N key1 A 1 2 1 + 0 0 \N key3 a3 1 1 1 + 0 0 \N key4 F 1 1 1 +key1 a 1 1 2 key1 B 2 1 2 +key1 a 1 1 2 key1 C 3 4 5 +key1 a 1 1 2 key1 D 4 1 6 +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 RIGHT ANY JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +1 1 1 1 1 1 +SELECT t1.*, t2.* FROM t1 RIGHT SEMI JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 a 1 1 2 key1 B 2 1 2 +key1 a 1 1 2 key1 C 3 4 5 +key1 a 1 1 2 key1 D 4 1 6 +key4 f 2 3 4 key4 F 1 1 1 +SELECT t1.*, t2.* from t1 RIGHT SEMI JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 b 2 3 2 key1 C 3 4 5 +key1 d 4 7 2 key1 D 4 1 6 +SELECT t1.*, t2.* from t1 RIGHT SEMI JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 B 2 1 2 +key1 a 1 1 2 key1 C 3 4 5 +key1 a 1 1 2 key1 D 4 1 6 +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 RIGHT SEMI JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +1 1 1 1 1 1 +SELECT t1.*, t2.* FROM t1 RIGHT ANTI JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); + 0 0 \N key3 a3 1 1 1 +SELECT t1.*, t2.* from t1 RIGHT ANTI JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); + 0 0 \N key3 a3 1 1 1 + 0 0 \N key4 F 1 1 1 +SELECT t1.*, t2.* from t1 RIGHT ANTI JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); + 0 0 \N key1 A 1 2 1 + 0 0 \N key3 a3 1 1 1 + 0 0 \N key4 F 1 1 1 +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 RIGHT ANTI JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +SET join_algorithm='grace_hash'; +SELECT t1.*, t2.* FROM t1 LEFT ANY JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 c 3 2 1 key1 C 3 4 5 +key1 d 4 7 2 key1 D 4 1 6 +key1 e 5 5 5 0 0 \N +key2 a2 1 1 1 0 0 \N +key4 f 2 3 4 key4 F 1 1 1 +SELECT t1.*, t2.* from t1 LEFT ANY JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 c 3 2 1 key1 B 2 1 2 +key1 d 4 7 2 key1 D 4 1 6 +key1 e 5 5 5 0 0 \N +key2 a2 1 1 1 0 0 \N +key4 f 2 3 4 0 0 \N +SELECT t1.*, t2.* from t1 LEFT ANY JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 B 2 1 2 +key1 b 2 3 2 key1 C 3 4 5 +key1 c 3 2 1 key1 D 4 1 6 +key1 d 4 7 2 0 0 \N +key1 e 5 5 5 0 0 \N +key2 a2 1 1 1 0 0 \N +key4 f 2 3 4 0 0 \N +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 LEFT ANY JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +1 1 1 1 1 1 +SELECT t1.*, t2.* FROM t1 LEFT SEMI JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 c 3 2 1 key1 C 3 4 5 +key1 d 4 7 2 key1 D 4 1 6 +key4 f 2 3 4 key4 F 1 1 1 +SELECT t1.*, t2.* from t1 LEFT SEMI JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 c 3 2 1 key1 B 2 1 2 +key1 d 4 7 2 key1 D 4 1 6 +SELECT t1.*, t2.* from t1 LEFT SEMI JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 B 2 1 2 +key1 b 2 3 2 key1 C 3 4 5 +key1 c 3 2 1 key1 D 4 1 6 +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 LEFT SEMI JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +1 1 1 1 1 1 +SELECT t1.*, t2.* FROM t1 LEFT ANTI JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 e 5 5 5 key1 0 0 \N +key2 a2 1 1 1 key2 0 0 \N +SELECT t1.*, t2.* from t1 LEFT ANTI JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 e 5 5 5 key1 0 0 \N +key2 a2 1 1 1 key2 0 0 \N +key4 f 2 3 4 key4 0 0 \N +SELECT t1.*, t2.* from t1 LEFT ANTI JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 d 4 7 2 key1 0 0 \N +key1 e 5 5 5 key1 0 0 \N +key2 a2 1 1 1 key2 0 0 \N +key4 f 2 3 4 key4 0 0 \N +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 LEFT ANTI JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +SELECT t1.*, t2.* FROM t1 RIGHT ANY JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); + 0 0 \N key3 a3 1 1 1 +key1 a 1 1 2 key1 A 1 2 1 +key1 a 1 1 2 key1 B 2 1 2 +key1 a 1 1 2 key1 C 3 4 5 +key1 a 1 1 2 key1 D 4 1 6 +key4 f 2 3 4 key4 F 1 1 1 +SELECT t1.*, t2.* from t1 RIGHT ANY JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); + 0 0 \N key3 a3 1 1 1 + 0 0 \N key4 F 1 1 1 +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 b 2 3 2 key1 C 3 4 5 +key1 d 4 7 2 key1 D 4 1 6 +SELECT t1.*, t2.* from t1 RIGHT ANY JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); + 0 0 \N key1 A 1 2 1 + 0 0 \N key3 a3 1 1 1 + 0 0 \N key4 F 1 1 1 +key1 a 1 1 2 key1 B 2 1 2 +key1 a 1 1 2 key1 C 3 4 5 +key1 a 1 1 2 key1 D 4 1 6 +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 RIGHT ANY JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +1 1 1 1 1 1 +SELECT t1.*, t2.* FROM t1 RIGHT SEMI JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 a 1 1 2 key1 B 2 1 2 +key1 a 1 1 2 key1 C 3 4 5 +key1 a 1 1 2 key1 D 4 1 6 +key4 f 2 3 4 key4 F 1 1 1 +SELECT t1.*, t2.* from t1 RIGHT SEMI JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 b 2 3 2 key1 C 3 4 5 +key1 d 4 7 2 key1 D 4 1 6 +SELECT t1.*, t2.* from t1 RIGHT SEMI JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 B 2 1 2 +key1 a 1 1 2 key1 C 3 4 5 +key1 a 1 1 2 key1 D 4 1 6 +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 RIGHT SEMI JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +1 1 1 1 1 1 +SELECT t1.*, t2.* FROM t1 RIGHT ANTI JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); + 0 0 \N key3 a3 1 1 1 +SELECT t1.*, t2.* from t1 RIGHT ANTI JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); + 0 0 \N key3 a3 1 1 1 + 0 0 \N key4 F 1 1 1 +SELECT t1.*, t2.* from t1 RIGHT ANTI JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); + 0 0 \N key1 A 1 2 1 + 0 0 \N key3 a3 1 1 1 + 0 0 \N key4 F 1 1 1 +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 RIGHT ANTI JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +SET join_algorithm='hash'; +SELECT t1.* FROM t1 LEFT ANY JOIN t2 ON t1.key = t2.key AND t1.a < t2.a OR t1.a = t2.a ORDER BY ALL; +key1 a 1 1 2 +key1 b 2 3 2 +key1 c 3 2 1 +key1 d 4 7 2 +key1 e 5 5 5 +key2 a2 1 1 1 +key4 f 2 3 4 +SELECT t1.* FROM t1 LEFT SEMI JOIN t2 ON t1.key = t2.key AND t1.a < t2.a OR t1.a = t2.a ORDER BY ALL; +key1 a 1 1 2 +key1 b 2 3 2 +key1 c 3 2 1 +key1 d 4 7 2 +key2 a2 1 1 1 +key4 f 2 3 4 +SELECT t1.* FROM t1 LEFT ANTI JOIN t2 ON t1.key = t2.key AND t1.a < t2.a OR t1.a = t2.a ORDER BY ALL; +key1 e 5 5 5 +SELECT t1.* FROM t1 RIGHT ANY JOIN t2 ON t1.key = t2.key AND t1.a < t2.a OR t1.a = t2.a ORDER BY ALL; +key1 a 1 1 2 +key1 a 1 1 2 +key1 a 1 1 2 +key1 a 1 1 2 +key1 a 1 1 2 +key1 a 1 1 2 +SELECT t1.* FROM t1 RIGHT SEMI JOIN t2 ON t1.key = t2.key AND t1.a < t2.a OR t1.a = t2.a ORDER BY ALL; +key1 a 1 1 2 +key1 a 1 1 2 +key1 a 1 1 2 +key1 a 1 1 2 +key1 a 1 1 2 +key1 a 1 1 2 +SELECT t1.* FROM t1 RIGHT ANTI JOIN t2 ON t1.key = t2.key AND t1.a < t2.a OR t1.a = t2.a ORDER BY ALL; +SET join_algorithm='hash'; SELECT t1.*, t2.* FROM t1 LEFT JOIN t2 ON t1.key = t2.key AND t1.a < t2.a OR t1.a = t2.a ORDER BY (t1.key, t1.attr, t2.key, t2.attr); key1 a 1 1 2 key1 A 1 2 1 key1 a 1 1 2 key1 B 2 1 2 @@ -452,3 +699,46 @@ key2 a2 1 1 1 key1 A 1 2 1 key2 a2 1 1 1 key3 a3 1 1 1 key2 a2 1 1 1 key4 F 1 1 1 key4 f 2 3 4 key1 B 2 1 2 +SET join_algorithm='hash'; +SELECT t1.*, t2.* FROM t1 INNER ANY JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 c 3 2 1 key1 C 3 4 5 +key1 d 4 7 2 key1 D 4 1 6 +key4 f 2 3 4 key4 F 1 1 1 +SELECT t1.*, t2.* from t1 INNER ANY JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 c 3 2 1 key1 B 2 1 2 +key1 d 4 7 2 key1 D 4 1 6 +SELECT t1.*, t2.* from t1 INNER ANY JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 B 2 1 2 +key1 b 2 3 2 key1 C 3 4 5 +key1 c 3 2 1 key1 D 4 1 6 +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 INNER ANY JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +1 1 1 1 1 1 +SET join_algorithm='grace_hash'; +SELECT t1.*, t2.* FROM t1 INNER ANY JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 c 3 2 1 key1 C 3 4 5 +key1 d 4 7 2 key1 D 4 1 6 +key4 f 2 3 4 key4 F 1 1 1 +SELECT t1.*, t2.* from t1 INNER ANY JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 A 1 2 1 +key1 b 2 3 2 key1 B 2 1 2 +key1 c 3 2 1 key1 B 2 1 2 +key1 d 4 7 2 key1 D 4 1 6 +SELECT t1.*, t2.* from t1 INNER ANY JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +key1 a 1 1 2 key1 B 2 1 2 +key1 b 2 3 2 key1 C 3 4 5 +key1 c 3 2 1 key1 D 4 1 6 +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 INNER ANY JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +1 1 1 1 1 1 +SET join_algorithm='hash'; +SELECT t1.* FROM t1 INNER ANY JOIN t2 ON t1.key = t2.key AND t1.a < t2.a OR t1.a = t2.a ORDER BY ALL; +key1 a 1 1 2 +key1 b 2 3 2 +key1 c 3 2 1 +key1 d 4 7 2 +key2 a2 1 1 1 diff --git a/tests/queries/0_stateless/03006_join_on_inequal_expression_fast.sql.j2 b/tests/queries/0_stateless/03006_join_on_inequal_expression_fast.sql.j2 index a363101ca69..b300881c562 100644 --- a/tests/queries/0_stateless/03006_join_on_inequal_expression_fast.sql.j2 +++ b/tests/queries/0_stateless/03006_join_on_inequal_expression_fast.sql.j2 @@ -6,7 +6,7 @@ INSERT INTO t1 VALUES ('key1', 'a', 1, 1, 2), ('key1', 'b', 2, 3, 2), ('key1', ' CREATE TABLE t2 (key String, attr String, a UInt64, b UInt64, c Nullable(UInt64)) ENGINE = MergeTree ORDER BY key; INSERT INTO t2 VALUES ('key1', 'A', 1, 2, 1), ('key1', 'B', 2, 1, 2), ('key1', 'C', 3, 4, 5), ('key1', 'D', 4, 1, 6), ('key3', 'a3', 1, 1, 1), ('key4', 'F', 1,1,1); -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SET allow_experimental_join_condition=1; SET join_use_nulls=0; -- { echoOn } @@ -22,6 +22,26 @@ SELECT t1.*, t2.* FROM t1 {{ join_type }} JOIN t2 ON t1.key = t2.key AND (t1.a=2 {% endfor -%} {% endfor -%} +{% for algorithm in ['hash', 'grace_hash'] -%} +SET join_algorithm='{{ algorithm }}'; +{% for join_type in ['LEFT', 'RIGHT'] -%} +{% for join_strictness in ['ANY', 'SEMI', 'ANTI'] -%} +SELECT t1.*, t2.* FROM t1 {{ join_type }} {{ join_strictness }} JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +SELECT t1.*, t2.* from t1 {{ join_type }} {{ join_strictness }} JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +SELECT t1.*, t2.* from t1 {{ join_type }} {{ join_strictness }} JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 {{ join_type }} {{ join_strictness }} JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +{% endfor -%} +{% endfor -%} +{% endfor -%} + +{% for algorithm in ['hash'] -%} +SET join_algorithm='{{ algorithm }}'; +{% for join_type in ['LEFT', 'RIGHT'] -%} +{% for join_strictness in ['ANY', 'SEMI', 'ANTI'] -%} +SELECT t1.* FROM t1 {{ join_type }} {{ join_strictness }} JOIN t2 ON t1.key = t2.key AND t1.a < t2.a OR t1.a = t2.a ORDER BY ALL; +{% endfor -%} +{% endfor -%} +{% endfor -%} {% for algorithm in ['hash'] -%} SET join_algorithm='{{ algorithm }}'; @@ -29,6 +49,28 @@ SET join_algorithm='{{ algorithm }}'; SELECT t1.*, t2.* FROM t1 {{ join_type }} JOIN t2 ON t1.key = t2.key AND t1.a < t2.a OR t1.a = t2.a ORDER BY (t1.key, t1.attr, t2.key, t2.attr); {% endfor -%} {% endfor -%} + +{% for algorithm in ['hash', 'grace_hash'] -%} +SET join_algorithm='{{ algorithm }}'; +{% for join_type in ['INNER'] -%} +{% for join_strictness in ['ANY'] -%} +SELECT t1.*, t2.* FROM t1 {{ join_type }} {{ join_strictness }} JOIN t2 ON (t1.a < t2.a OR lower(t1.attr) == lower(t2.attr)) AND t1.key = t2.key ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +SELECT t1.*, t2.* from t1 {{ join_type }} {{ join_strictness }} JOIN t2 ON t1.key = t2.key and (t1.b + t2.b == t1.c + t2.c) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +SELECT t1.*, t2.* from t1 {{ join_type }} {{ join_strictness }} JOIN t2 ON t1.key = t2.key and (t1.a < t2.a) ORDER BY (t1.key, t1.attr, t2.key, t2.attr); +SELECT * FROM (SELECT 1 AS a, 1 AS b, 1 AS c) AS t1 {{ join_type }} {{ join_strictness }} JOIN (SELECT 1 AS a, 1 AS b, 1 AS c) AS t2 ON t1.a = t2.a AND (t1.b > 0 OR t2.b > 0); +{% endfor -%} +{% endfor -%} +{% endfor -%} + +{% for algorithm in ['hash'] -%} +SET join_algorithm='{{ algorithm }}'; +{% for join_type in ['INNER'] -%} +{% for join_strictness in ['ANY'] -%} +SELECT t1.* FROM t1 {{ join_type }} {{ join_strictness }} JOIN t2 ON t1.key = t2.key AND t1.a < t2.a OR t1.a = t2.a ORDER BY ALL; +{% endfor -%} +{% endfor -%} +{% endfor -%} + -- { echoOff } -- test error messages diff --git a/tests/queries/0_stateless/03006_parallel_replicas_cte_explain_syntax_crash.sql b/tests/queries/0_stateless/03006_parallel_replicas_cte_explain_syntax_crash.sql index df4ec9d26a3..7c8d6dd9aff 100644 --- a/tests/queries/0_stateless/03006_parallel_replicas_cte_explain_syntax_crash.sql +++ b/tests/queries/0_stateless/03006_parallel_replicas_cte_explain_syntax_crash.sql @@ -20,7 +20,7 @@ ORDER BY n AS SELECT * FROM numbers(10); -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET allow_experimental_parallel_reading_from_replicas = 1, parallel_replicas_for_non_replicated_merge_tree = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', max_parallel_replicas = 3, parallel_replicas_min_number_of_rows_per_replica=0; EXPLAIN SYNTAX diff --git a/tests/queries/0_stateless/03007_column_nullable_uninitialzed_value.sql b/tests/queries/0_stateless/03007_column_nullable_uninitialzed_value.sql index 9479044e0e0..44f6642d2a5 100644 --- a/tests/queries/0_stateless/03007_column_nullable_uninitialzed_value.sql +++ b/tests/queries/0_stateless/03007_column_nullable_uninitialzed_value.sql @@ -1 +1 @@ -SELECT count(NULL) IGNORE NULLS > avg(toDecimal32(NULL)) IGNORE NULLS, count() FROM numbers(1000) WITH TOTALS SETTINGS allow_experimental_analyzer = 1; +SELECT count(NULL) IGNORE NULLS > avg(toDecimal32(NULL)) IGNORE NULLS, count() FROM numbers(1000) WITH TOTALS SETTINGS enable_analyzer = 1; diff --git a/tests/queries/0_stateless/03008_deduplication_random_setttings.sh b/tests/queries/0_stateless/03008_deduplication_random_setttings.sh index e9f59138177..07b99eb4e99 100755 --- a/tests/queries/0_stateless/03008_deduplication_random_setttings.sh +++ b/tests/queries/0_stateless/03008_deduplication_random_setttings.sh @@ -35,7 +35,7 @@ THIS_RUN+=" deduplicate_src_table=$deduplicate_src_table" THIS_RUN+=" deduplicate_dst_table=$deduplicate_dst_table" THIS_RUN+=" insert_unique_blocks=$insert_unique_blocks" -$CLICKHOUSE_CLIENT --max_insert_block_size 1 -nmq " +$CLICKHOUSE_CLIENT --max_insert_block_size 1 -mq " $(python3 $CURDIR/03008_deduplication.python insert_several_blocks_into_table \ --insert-method $insert_method \ --table-engine $engine \ @@ -48,7 +48,7 @@ $CLICKHOUSE_CLIENT --max_insert_block_size 1 -nmq " ) " 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 " +$CLICKHOUSE_CLIENT --max_insert_block_size 1 -mq " $(python3 $CURDIR/03008_deduplication.python mv_generates_several_blocks \ --insert-method $insert_method \ --table-engine $engine \ @@ -61,7 +61,7 @@ $CLICKHOUSE_CLIENT --max_insert_block_size 1 -nmq " ) " 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 " +$CLICKHOUSE_CLIENT --max_insert_block_size 1 -mq " $(python3 $CURDIR/03008_deduplication.python several_mv_into_one_table \ --insert-method $insert_method \ --table-engine $engine \ diff --git a/tests/queries/0_stateless/03008_local_plain_rewritable.sh b/tests/queries/0_stateless/03008_local_plain_rewritable.sh index d51e180efc9..e61f9061297 100755 --- a/tests/queries/0_stateless/03008_local_plain_rewritable.sh +++ b/tests/queries/0_stateless/03008_local_plain_rewritable.sh @@ -8,7 +8,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) ${CLICKHOUSE_CLIENT} --query "drop table if exists 03008_test_local_mt sync" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " create table 03008_test_local_mt (a Int32, b Int64, c Int64) engine = MergeTree() partition by intDiv(a, 1000) order by tuple(a, b) settings disk = disk( @@ -19,35 +19,35 @@ settings disk = disk( path = '/var/lib/clickhouse/disks/local_plain_rewritable/') " -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " insert into 03008_test_local_mt (*) values (1, 2, 0), (2, 2, 2), (3, 1, 9), (4, 7, 7), (5, 10, 2), (6, 12, 5); insert into 03008_test_local_mt (*) select number, number, number from numbers_mt(10000); " -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " select count(*) from 03008_test_local_mt; select (*) from 03008_test_local_mt order by tuple(a, b) limit 10; " ${CLICKHOUSE_CLIENT} --query "optimize table 03008_test_local_mt final;" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " alter table 03008_test_local_mt modify setting disk = '03008_local_plain_rewritable', old_parts_lifetime = 3600; select engine_full from system.tables WHERE database = currentDatabase() AND name = '03008_test_local_mt'; " | grep -c "old_parts_lifetime = 3600" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " select count(*) from 03008_test_local_mt; select (*) from 03008_test_local_mt order by tuple(a, b) limit 10; " -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " alter table 03008_test_local_mt update c = 0 where a % 2 = 1; alter table 03008_test_local_mt add column d Int64 after c; alter table 03008_test_local_mt drop column c; " 2>&1 | grep -Fq "SUPPORT_IS_DISABLED" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " truncate table 03008_test_local_mt; select count(*) from 03008_test_local_mt; " diff --git a/tests/queries/0_stateless/03008_s3_plain_rewritable.sh b/tests/queries/0_stateless/03008_s3_plain_rewritable.sh index 4d5989f6f12..8eea7940774 100755 --- a/tests/queries/0_stateless/03008_s3_plain_rewritable.sh +++ b/tests/queries/0_stateless/03008_s3_plain_rewritable.sh @@ -46,7 +46,12 @@ ${CLICKHOUSE_CLIENT} --query "drop table if exists test_s3_mt_dst" ${CLICKHOUSE_CLIENT} -m --query " create table test_s3_mt_dst (a Int32, b Int64, c Int64) engine = MergeTree() partition by intDiv(a, 1000) order by tuple(a, b) -settings disk = '03008_s3_plain_rewritable' +settings disk = disk( + name = 03008_s3_plain_rewritable, + type = s3_plain_rewritable, + endpoint = 'http://localhost:11111/test/03008_test_s3_mt/', + access_key_id = clickhouse, + secret_access_key = clickhouse); " ${CLICKHOUSE_CLIENT} -m --query " diff --git a/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.sql b/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.sql index 71a175faac8..5ec6ee5a996 100644 --- a/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.sql +++ b/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.sql @@ -1,11 +1,11 @@ SET optimize_rewrite_sum_if_to_count_if = 1; -SET allow_experimental_analyzer = 0; +SET enable_analyzer = 0; SELECT (sumIf(toNullable(1), (number % 2) = 0), NULL) FROM numbers(10); SELECT (sum(if((number % 2) = 0, toNullable(1), 0)), NULL) FROM numbers(10); SELECT (tuple(sum(if((number % 2) = 0, toNullable(0), 123)) IGNORE NULLS), toUInt8(3)) FROM numbers(100); -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT (sumIf(toNullable(1), (number % 2) = 0), NULL) FROM numbers(10); EXPLAIN QUERY TREE SELECT (sumIf(toNullable(1), (number % 2) = 0), NULL) FROM numbers(10); SELECT (sum(if((number % 2) = 0, toNullable(1), 0)), NULL) FROM numbers(10); diff --git a/tests/queries/0_stateless/03013_group_by_use_nulls_with_materialize_and_analyzer.sql b/tests/queries/0_stateless/03013_group_by_use_nulls_with_materialize_and_analyzer.sql index b15593b2abd..7b57dbd807d 100644 --- a/tests/queries/0_stateless/03013_group_by_use_nulls_with_materialize_and_analyzer.sql +++ b/tests/queries/0_stateless/03013_group_by_use_nulls_with_materialize_and_analyzer.sql @@ -1,4 +1,4 @@ -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; set group_by_use_nulls = 1; set optimize_group_by_function_keys = 1; set optimize_injective_functions_in_group_by = 1; @@ -8,4 +8,3 @@ SELECT materialize(3) from numbers(10) GROUP BY GROUPING SETS (('str'), (materia SELECT ignore(3) from numbers(10) GROUP BY GROUPING SETS (('str'), (ignore(3))) order by all; SELECT materialize(ignore(3)) from numbers(10) GROUP BY GROUPING SETS (('str'), (materialize(ignore(3)))) order by all; SELECT ignore(materialize(3)) from numbers(10) GROUP BY GROUPING SETS (('str'), (ignore(materialize(3)))) order by all; - diff --git a/tests/queries/0_stateless/03013_json_key_ignore_case.reference b/tests/queries/0_stateless/03013_json_key_ignore_case.reference deleted file mode 100644 index 54683d8fbc5..00000000000 --- a/tests/queries/0_stateless/03013_json_key_ignore_case.reference +++ /dev/null @@ -1,3 +0,0 @@ -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 deleted file mode 100755 index 7882a29ae4f..00000000000 --- a/tests/queries/0_stateless/03013_json_key_ignore_case.sh +++ /dev/null @@ -1,18 +0,0 @@ -#!/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/03014_analyzer_groupby_fuzz_60317.sql b/tests/queries/0_stateless/03014_analyzer_groupby_fuzz_60317.sql index 094614cb78d..295f89c5a0a 100644 --- a/tests/queries/0_stateless/03014_analyzer_groupby_fuzz_60317.sql +++ b/tests/queries/0_stateless/03014_analyzer_groupby_fuzz_60317.sql @@ -6,7 +6,7 @@ SELECT FROM system.one GROUP BY _CAST(30, 'Nullable(UInt8)') -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; -- WITH CUBE (note that result is different with the analyzer (analyzer is correct including all combinations) SELECT @@ -24,4 +24,4 @@ GROUP BY _CAST(30, 'Nullable(UInt8)') WITH CUBE WITH TOTALS -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; diff --git a/tests/queries/0_stateless/03014_group_by_use_nulls_injective_functions_and_analyzer.sql b/tests/queries/0_stateless/03014_group_by_use_nulls_injective_functions_and_analyzer.sql index e7ea964b876..d700f9ba3b1 100644 --- a/tests/queries/0_stateless/03014_group_by_use_nulls_injective_functions_and_analyzer.sql +++ b/tests/queries/0_stateless/03014_group_by_use_nulls_injective_functions_and_analyzer.sql @@ -1,5 +1,4 @@ -set allow_experimental_analyzer=1, group_by_use_nulls=1, optimize_injective_functions_in_group_by=1; +set enable_analyzer=1, group_by_use_nulls=1, optimize_injective_functions_in_group_by=1; SELECT bitNot(bitNot(number)) + 3 FROM numbers(10) GROUP BY GROUPING SETS (('str', bitNot(bitNot(number))), ('str')) order by all; SELECT tuple(tuple(tuple(number))) FROM numbers(10) GROUP BY GROUPING SETS (('str', tuple(tuple(number))), ('str')) order by all; SELECT materialize(3) + 3 FROM numbers(10) GROUP BY GROUPING SETS (('str', materialize(materialize(3))), ('str')) order by all; - diff --git a/tests/queries/0_stateless/03015_analyzer_groupby_fuzz_60772.sql b/tests/queries/0_stateless/03015_analyzer_groupby_fuzz_60772.sql index d3bd9ef0ce3..5190b8635d6 100644 --- a/tests/queries/0_stateless/03015_analyzer_groupby_fuzz_60772.sql +++ b/tests/queries/0_stateless/03015_analyzer_groupby_fuzz_60772.sql @@ -7,7 +7,7 @@ GROUP BY toFixedString(toFixedString('2018-01-02 22:33:44', 19), 19), 'gr', '2018-01-02 22:33:44' -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; -- WITH CUBE (note that result is different with the analyzer (analyzer is correct including all combinations) SELECT @@ -20,4 +20,4 @@ GROUP BY 'gr', '2018-01-02 22:33:44' WITH CUBE -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; diff --git a/tests/queries/0_stateless/03016_analyzer_groupby_fuzz_59796.sql b/tests/queries/0_stateless/03016_analyzer_groupby_fuzz_59796.sql index c00a75a631c..6c926c32887 100644 --- a/tests/queries/0_stateless/03016_analyzer_groupby_fuzz_59796.sql +++ b/tests/queries/0_stateless/03016_analyzer_groupby_fuzz_59796.sql @@ -3,4 +3,4 @@ SELECT GROUP BY concat(unhex('00'), toFixedString(materialize(toFixedString(' key="v" ', 9)), 9), toFixedString(toFixedString('00', 2), toNullable(2)), toFixedString(toFixedString(toFixedString(' key="v" ', 9), 9), 9)), concat(' key="v" ') -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; diff --git a/tests/queries/0_stateless/03017_analyzer_groupby_fuzz_61600.sql b/tests/queries/0_stateless/03017_analyzer_groupby_fuzz_61600.sql index 53a5cfe9b1a..b22ea42b686 100644 --- a/tests/queries/0_stateless/03017_analyzer_groupby_fuzz_61600.sql +++ b/tests/queries/0_stateless/03017_analyzer_groupby_fuzz_61600.sql @@ -11,7 +11,7 @@ FROM set_index_not__fuzz_0 GROUP BY toNullable(3), concat(concat(NULLIF(1, 1), toNullable(toNullable(3)))) -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; -- WITH ROLLUP (note that result is different with the analyzer (analyzer is correct including all combinations) SELECT @@ -22,4 +22,4 @@ GROUP BY toNullable(3), concat(concat(NULLIF(1, 1), toNullable(toNullable(3)))) WITH ROLLUP -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; diff --git a/tests/queries/0_stateless/03023_analyzer_optimize_group_by_function_keys_with_nulls.sql b/tests/queries/0_stateless/03023_analyzer_optimize_group_by_function_keys_with_nulls.sql index f0b60caca36..0d0a81c9105 100644 --- a/tests/queries/0_stateless/03023_analyzer_optimize_group_by_function_keys_with_nulls.sql +++ b/tests/queries/0_stateless/03023_analyzer_optimize_group_by_function_keys_with_nulls.sql @@ -1,5 +1,4 @@ -set allow_experimental_analyzer=1; +set enable_analyzer=1; set group_by_use_nulls=1; set optimize_group_by_function_keys=1; SELECT ignore(toLowCardinality(number)) FROM numbers(10) GROUP BY GROUPING SETS ((ignore(toLowCardinality(number)), toLowCardinality(number))); - diff --git a/tests/queries/0_stateless/03023_group_by_use_nulls_analyzer_crashes.sql b/tests/queries/0_stateless/03023_group_by_use_nulls_analyzer_crashes.sql index b8c173520a9..d3d6ecaadaf 100644 --- a/tests/queries/0_stateless/03023_group_by_use_nulls_analyzer_crashes.sql +++ b/tests/queries/0_stateless/03023_group_by_use_nulls_analyzer_crashes.sql @@ -1,4 +1,4 @@ -set allow_experimental_analyzer = 1, group_by_use_nulls = 1; +set enable_analyzer = 1, group_by_use_nulls = 1; SELECT tuple(tuple(number)) as x FROM numbers(10) GROUP BY (number, tuple(number)) with cube order by x; @@ -56,4 +56,4 @@ SELECT arraySplit(number -> toUInt8(number), []) from numbers(1) GROUP BY toUInt SELECT count(arraySplit(number -> toUInt8(number), [arraySplit(x -> toUInt8(number), [])])) FROM numbers(10) GROUP BY number, [number] WITH ROLLUP settings group_by_use_nulls=1; -- {serverError ILLEGAL_TYPE_OF_ARGUMENT} -SELECT count(arraySplit(x -> toUInt8(number), [])) FROM numbers(10) GROUP BY number, [number] WITH ROLLUP settings group_by_use_nulls=1; \ No newline at end of file +SELECT count(arraySplit(x -> toUInt8(number), [])) FROM numbers(10) GROUP BY number, [number] WITH ROLLUP settings group_by_use_nulls=1; diff --git a/tests/queries/0_stateless/03023_remove_unused_column_distinct.sql b/tests/queries/0_stateless/03023_remove_unused_column_distinct.sql index c2f32bfe3c1..af8756f1fbc 100644 --- a/tests/queries/0_stateless/03023_remove_unused_column_distinct.sql +++ b/tests/queries/0_stateless/03023_remove_unused_column_distinct.sql @@ -12,4 +12,4 @@ FROM FROM numbers(10) ) ) -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; diff --git a/tests/queries/0_stateless/03031_clickhouse_local_input.sh b/tests/queries/0_stateless/03031_clickhouse_local_input.sh index 6f59e9b9703..e2f9cf48108 100755 --- a/tests/queries/0_stateless/03031_clickhouse_local_input.sh +++ b/tests/queries/0_stateless/03031_clickhouse_local_input.sh @@ -6,15 +6,15 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) tmp_file="$CUR_DIR/$CLICKHOUSE_DATABASE.txt" echo '# foo' -$CLICKHOUSE_LOCAL --engine_file_truncate_on_insert=1 -n -q "insert into function file('$tmp_file', 'LineAsString', 'x String') select * from input('x String') format LineAsString" << d2 -[1,2,3] [1,2,3] 0 Array(Int64) Array(Int64) -[1,2,3] [1,2,4] 0 Array(Int64) Array(Int64) -[1,2,3] 42 0 Array(Int64) Int64 -[1,2,3] abc 0 Array(Int64) String -[1,2,3] \N 0 Array(Int64) None -[1,2,4] [1,2,3] 1 Array(Int64) Array(Int64) -42 [1,2,3] 1 Int64 Array(Int64) -42 42 0 Int64 Int64 -42 43 0 Int64 Int64 -42 abc 0 Int64 String -42 \N 0 Int64 None -43 42 1 Int64 Int64 -abc [1,2,3] 1 String Array(Int64) -abc 42 1 String Int64 -abc abc 0 String String -abc abd 0 String String -abc \N 0 String None -abd abc 1 String String -\N [1,2,3] 1 None Array(Int64) -\N 42 1 None Int64 -\N abc 1 None String -\N \N 0 None None +[1,2,3] [1,2,3] 0 Array(Int64) true Array(Int64) true +[1,2,3] [1,2,4] 0 Array(Int64) true Array(Int64) true +[1,2,3] 2020-01-01 0 Array(Int64) true Date true +[1,2,3] 42 0 Array(Int64) true Int64 false +[1,2,3] abc 0 Array(Int64) true String false +[1,2,3] \N 0 Array(Int64) true None false +[1,2,4] [1,2,3] 1 Array(Int64) true Array(Int64) true +2020-01-01 [1,2,3] 1 Date true Array(Int64) true +2020-01-01 2020-01-01 0 Date true Date true +2020-01-01 2020-01-02 0 Date true Date true +2020-01-01 42 0 Date true Int64 false +2020-01-01 abc 0 Date true String false +2020-01-01 \N 0 Date true None false +2020-01-02 2020-01-01 1 Date true Date true +42 [1,2,3] 1 Int64 false Array(Int64) true +42 2020-01-01 1 Int64 false Date true +42 42 0 Int64 false Int64 false +42 43 0 Int64 false Int64 false +42 abc 0 Int64 false String false +42 \N 0 Int64 false None false +43 42 1 Int64 false Int64 false +abc [1,2,3] 1 String false Array(Int64) true +abc 2020-01-01 1 String false Date true +abc 42 1 String false Int64 false +abc abc 0 String false String false +abc abd 0 String false String false +abc \N 0 String false None false +abd abc 1 String false String false +\N [1,2,3] 1 None false Array(Int64) true +\N 2020-01-01 1 None false Date true +\N 42 1 None false Int64 false +\N abc 1 None false String false +\N \N 0 None false None false d1 >= d2 -[1,2,3] [1,2,3] 1 Array(Int64) Array(Int64) -[1,2,3] [1,2,4] 1 Array(Int64) Array(Int64) -[1,2,3] 42 1 Array(Int64) Int64 -[1,2,3] abc 1 Array(Int64) String -[1,2,3] \N 1 Array(Int64) None -[1,2,4] [1,2,3] 1 Array(Int64) Array(Int64) -42 [1,2,3] 1 Int64 Array(Int64) -42 42 1 Int64 Int64 -42 43 1 Int64 Int64 -42 abc 1 Int64 String -42 \N 1 Int64 None -43 42 1 Int64 Int64 -abc [1,2,3] 1 String Array(Int64) -abc 42 1 String Int64 -abc abc 1 String String -abc abd 1 String String -abc \N 1 String None -abd abc 1 String String -\N [1,2,3] 1 None Array(Int64) -\N 42 1 None Int64 -\N abc 1 None String -\N \N 1 None None +[1,2,3] [1,2,3] 1 Array(Int64) true Array(Int64) true +[1,2,3] [1,2,4] 1 Array(Int64) true Array(Int64) true +[1,2,3] 2020-01-01 1 Array(Int64) true Date true +[1,2,3] 42 1 Array(Int64) true Int64 false +[1,2,3] abc 1 Array(Int64) true String false +[1,2,3] \N 1 Array(Int64) true None false +[1,2,4] [1,2,3] 1 Array(Int64) true Array(Int64) true +2020-01-01 [1,2,3] 1 Date true Array(Int64) true +2020-01-01 2020-01-01 1 Date true Date true +2020-01-01 2020-01-02 1 Date true Date true +2020-01-01 42 1 Date true Int64 false +2020-01-01 abc 1 Date true String false +2020-01-01 \N 1 Date true None false +2020-01-02 2020-01-01 1 Date true Date true +42 [1,2,3] 1 Int64 false Array(Int64) true +42 2020-01-01 1 Int64 false Date true +42 42 1 Int64 false Int64 false +42 43 1 Int64 false Int64 false +42 abc 1 Int64 false String false +42 \N 1 Int64 false None false +43 42 1 Int64 false Int64 false +abc [1,2,3] 1 String false Array(Int64) true +abc 2020-01-01 1 String false Date true +abc 42 1 String false Int64 false +abc abc 1 String false String false +abc abd 1 String false String false +abc \N 1 String false None false +abd abc 1 String false String false +\N [1,2,3] 1 None false Array(Int64) true +\N 2020-01-01 1 None false Date true +\N 42 1 None false Int64 false +\N abc 1 None false String false +\N \N 1 None false None false diff --git a/tests/queries/0_stateless/03035_dynamic_sorting.sql b/tests/queries/0_stateless/03035_dynamic_sorting.sql index 0487fafc955..e0039a348c6 100644 --- a/tests/queries/0_stateless/03035_dynamic_sorting.sql +++ b/tests/queries/0_stateless/03035_dynamic_sorting.sql @@ -1,80 +1,55 @@ set allow_experimental_dynamic_type = 1; drop table if exists test; -create table test (d1 Dynamic, d2 Dynamic) engine=Memory; - -insert into test values (42, 42); -insert into test values (42, 43); -insert into test values (43, 42); - -insert into test values ('abc', 'abc'); -insert into test values ('abc', 'abd'); -insert into test values ('abd', 'abc'); - -insert into test values ([1,2,3], [1,2,3]); -insert into test values ([1,2,3], [1,2,4]); -insert into test values ([1,2,4], [1,2,3]); - -insert into test values (NULL, NULL); - -insert into test values (42, 'abc'); -insert into test values ('abc', 42); - -insert into test values (42, [1,2,3]); -insert into test values ([1,2,3], 42); - -insert into test values (42, NULL); -insert into test values (NULL, 42); - -insert into test values ('abc', [1,2,3]); -insert into test values ([1,2,3], 'abc'); - -insert into test values ('abc', NULL); -insert into test values (NULL, 'abc'); - -insert into test values ([1,2,3], NULL); -insert into test values (NULL, [1,2,3]); +create table test (d1 Dynamic(max_types=2), d2 Dynamic(max_types=2)) engine=Memory; +insert into test values (42, 42), (42, 43), (43, 42), ('abc', 'abc'), ('abc', 'abd'), ('abd', 'abc'), +([1,2,3], [1,2,3]), ([1,2,3], [1,2,4]), ([1,2,4], [1,2,3]), +('2020-01-01', '2020-01-01'), ('2020-01-01', '2020-01-02'), ('2020-01-02', '2020-01-01'), +(NULL, NULL), (42, 'abc'), ('abc', 42), (42, [1,2,3]), ([1,2,3], 42), (42, NULL), (NULL, 42), +('abc', [1,2,3]), ([1,2,3], 'abc'), ('abc', NULL), (NULL, 'abc'), ([1,2,3], NULL), (NULL, [1,2,3]), +(42, '2020-01-01'), ('2020-01-01', 42), ('2020-01-01', 'abc'), ('abc', '2020-01-01'), +('2020-01-01', [1,2,3]), ([1,2,3], '2020-01-01'), ('2020-01-01', NULL), (NULL, '2020-01-01'); select 'order by d1 nulls first'; -select d1, dynamicType(d1) from test order by d1 nulls first; +select d1, dynamicType(d1), isDynamicElementInSharedData(d1) from test order by d1 nulls first; select 'order by d1 nulls last'; -select d1, dynamicType(d1) from test order by d1 nulls last; +select d1, dynamicType(d1), isDynamicElementInSharedData(d1) from test order by d1 nulls last; select 'order by d2 nulls first'; -select d2, dynamicType(d2) from test order by d2 nulls first; +select d2, dynamicType(d2), isDynamicElementInSharedData(d2) from test order by d2 nulls first; select 'order by d2 nulls last'; -select d2, dynamicType(d2) from test order by d2 nulls last; +select d2, dynamicType(d2), isDynamicElementInSharedData(d2) from test order by d2 nulls last; select 'order by d1, d2 nulls first'; -select d1, d2, dynamicType(d1), dynamicType(d2) from test order by d1, d2 nulls first; +select d1, d2, dynamicType(d1), isDynamicElementInSharedData(d1), dynamicType(d2), isDynamicElementInSharedData(d2) from test order by d1, d2 nulls first; select 'order by d1, d2 nulls last'; -select d1, d2, dynamicType(d1), dynamicType(d2) from test order by d1, d2 nulls last; +select d1, d2, dynamicType(d1), isDynamicElementInSharedData(d1), dynamicType(d2), isDynamicElementInSharedData(d2) from test order by d1, d2 nulls last; select 'order by d2, d1 nulls first'; -select d1, d2, dynamicType(d1), dynamicType(d2) from test order by d2, d1 nulls first; +select d1, d2, dynamicType(d1), isDynamicElementInSharedData(d1), dynamicType(d2), isDynamicElementInSharedData(d2) from test order by d2, d1 nulls first; select 'order by d2, d1 nulls last'; -select d1, d2, dynamicType(d1), dynamicType(d2) from test order by d2, d1 nulls last; +select d1, d2, dynamicType(d1), isDynamicElementInSharedData(d1), dynamicType(d2), isDynamicElementInSharedData(d2) from test order by d2, d1 nulls last; select 'd1 = d2'; -select d1, d2, d1 = d2, dynamicType(d1), dynamicType(d2) from test order by d1, d2; +select d1, d2, d1 = d2, dynamicType(d1), isDynamicElementInSharedData(d1), dynamicType(d2), isDynamicElementInSharedData(d2) from test order by d1, d2; select 'd1 < d2'; -select d1, d2, d1 < d2, dynamicType(d1), dynamicType(d2) from test order by d1, d2; +select d1, d2, d1 < d2, dynamicType(d1), isDynamicElementInSharedData(d1), dynamicType(d2), isDynamicElementInSharedData(d2) from test order by d1, d2; select 'd1 <= d2'; -select d1, d2, d1 <= d2, dynamicType(d1), dynamicType(d2) from test order by d1, d2; +select d1, d2, d1 <= d2, dynamicType(d1), isDynamicElementInSharedData(d1), dynamicType(d2), isDynamicElementInSharedData(d2) from test order by d1, d2; select 'd1 > d2'; -select d1, d2, d1 > d2, dynamicType(d1), dynamicType(d2) from test order by d1, d2; +select d1, d2, d1 > d2, dynamicType(d1), isDynamicElementInSharedData(d1), dynamicType(d2), isDynamicElementInSharedData(d2) from test order by d1, d2; select 'd1 >= d2'; -select d1, d2, d2 >= d2, dynamicType(d1), dynamicType(d2) from test order by d1, d2; +select d1, d2, d2 >= d2, dynamicType(d1), isDynamicElementInSharedData(d1), dynamicType(d2), isDynamicElementInSharedData(d2) from test order by d1, d2; drop table test; diff --git a/tests/queries/0_stateless/03035_internal_functions_direct_call.sql b/tests/queries/0_stateless/03035_internal_functions_direct_call.sql index 951e0733dbb..e358e498343 100644 --- a/tests/queries/0_stateless/03035_internal_functions_direct_call.sql +++ b/tests/queries/0_stateless/03035_internal_functions_direct_call.sql @@ -4,7 +4,7 @@ SELECT __actionName(); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } SELECT __actionName('aaa', 'aaa', 'aaa'); -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH } -SELECT __actionName('aaa', '') SETTINGS allow_experimental_analyzer = 1; -- { serverError BAD_ARGUMENTS } +SELECT __actionName('aaa', '') SETTINGS enable_analyzer = 1; -- { serverError BAD_ARGUMENTS } SELECT __actionName('aaa', materialize('aaa')); -- { serverError BAD_ARGUMENTS,ILLEGAL_COLUMN } SELECT __actionName(materialize('aaa'), 'aaa'); -- { serverError ILLEGAL_COLUMN } SELECT __actionName('aaa', 'aaa'); diff --git a/tests/queries/0_stateless/03035_recursive_cte_postgres_1.reference b/tests/queries/0_stateless/03035_recursive_cte_postgres_1.reference index e6cbdc971eb..568421f3fba 100644 --- a/tests/queries/0_stateless/03035_recursive_cte_postgres_1.reference +++ b/tests/queries/0_stateless/03035_recursive_cte_postgres_1.reference @@ -32,7 +32,7 @@ -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- WITH RECURSIVE -- sum of 1..100 diff --git a/tests/queries/0_stateless/03035_recursive_cte_postgres_1.sql b/tests/queries/0_stateless/03035_recursive_cte_postgres_1.sql index 8026dadc331..9a4e313ce90 100644 --- a/tests/queries/0_stateless/03035_recursive_cte_postgres_1.sql +++ b/tests/queries/0_stateless/03035_recursive_cte_postgres_1.sql @@ -32,7 +32,7 @@ -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- WITH RECURSIVE diff --git a/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_compact_merge_tree.reference b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_compact_merge_tree.reference new file mode 100644 index 00000000000..ca6c5dbba82 --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_compact_merge_tree.reference @@ -0,0 +1,20 @@ +Array(Array(Dynamic)) +Array(Variant(String, UInt64)) +LowCardinality(String) +None +String +UInt64 +360000 +360000 +200000 +200000 +0 +0 +20000 +20000 +200000 +200000 +20000 +20000 +200000 +0 diff --git a/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_compact_merge_tree.sql b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_compact_merge_tree.sql new file mode 100644 index 00000000000..bff28fb5c90 --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_compact_merge_tree.sql @@ -0,0 +1,43 @@ +-- Tags: long, no-tsan, no-msan, no-ubsan, no-asan + +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(max_types=2)) 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, if (number % 5 == 1, [range((number % 10 + 1)::UInt64)]::Array(Array(Dynamic)), number) from numbers(100000, 100000) settings min_insert_block_size_rows=50000; +insert into test select number, if (number % 5 == 1, ('str_' || number)::LowCardinality(String)::Dynamic, number::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) == 'LowCardinality(String)'; +select count() from test where d.`LowCardinality(String)` 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.`LowCardinality(String)`, d.`Array(String)` from test format Null; +select d, d.UInt64, d.Date, d.`LowCardinality(String)`, 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.`LowCardinality(String)`, 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_shared_subcolumns_memory.reference b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_memory.reference new file mode 100644 index 00000000000..ca6c5dbba82 --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_memory.reference @@ -0,0 +1,20 @@ +Array(Array(Dynamic)) +Array(Variant(String, UInt64)) +LowCardinality(String) +None +String +UInt64 +360000 +360000 +200000 +200000 +0 +0 +20000 +20000 +200000 +200000 +20000 +20000 +200000 +0 diff --git a/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_memory.sql b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_memory.sql new file mode 100644 index 00000000000..4eed3d15529 --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_memory.sql @@ -0,0 +1,43 @@ +-- Tags: long, no-tsan, no-msan, no-ubsan, no-asan + +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(max_types=2)) 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, if (number % 5 == 1, [range((number % 10 + 1)::UInt64)]::Array(Array(Dynamic)), number) from numbers(100000, 100000) settings min_insert_block_size_rows=50000; +insert into test select number, if (number % 5 == 1, ('str_' || number)::LowCardinality(String)::Dynamic, number::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) == 'LowCardinality(String)'; +select count() from test where d.`LowCardinality(String)` 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_shared_subcolumns_small.reference.j2 b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_small.reference.j2 new file mode 100644 index 00000000000..de12c6b8737 --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_small.reference.j2 @@ -0,0 +1,2463 @@ +Memory +Array(Array(Dynamic)) +Array(Variant(String, UInt64)) +LowCardinality(String) +None +String +UInt64 +36 +36 +20 +20 +0 +0 +2 +2 +20 +20 +2 +2 +20 +0 +0 0 \N [] +1 1 \N [] +2 2 \N [] +3 3 \N [] +4 4 \N [] +5 5 \N [] +6 6 \N [] +7 7 \N [] +8 8 \N [] +9 9 \N [] +str_10 \N str_10 [] +10 10 \N [] +10 10 \N [] +[[0,1]] \N \N [] +str_11 \N \N [] +str_11 \N str_11 [] +str_12 \N str_12 [] +12 12 \N [] +12 12 \N [] +str_13 \N str_13 [] +13 13 \N [] +13 13 \N [] +str_14 \N str_14 [] +14 14 \N [] +14 14 \N [] +str_15 \N str_15 [] +15 15 \N [] +15 15 \N [] +[[0,1,2,3,4,5,6]] \N \N [] +str_16 \N \N [] +str_16 \N str_16 [] +str_17 \N str_17 [] +17 17 \N [] +17 17 \N [] +str_18 \N str_18 [] +18 18 \N [] +18 18 \N [] +str_19 \N str_19 [] +19 19 \N [] +19 19 \N [] +[20] \N \N [20] +['str_21','str_21'] \N \N ['str_21','str_21'] +[22,22,22] \N \N [22,22,22] +[23,23,23,23] \N \N [23,23,23,23] +[24,24,24,24,24] \N \N [24,24,24,24,24] +[25,25,25,25,25,25] \N \N [25,25,25,25,25,25] +[26,26,26,26,26,26,26] \N \N [26,26,26,26,26,26,26] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] \N \N [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +[28,28,28,28,28,28,28,28,28] \N \N [28,28,28,28,28,28,28,28,28] +[29,29,29,29,29,29,29,29,29,29] \N \N [29,29,29,29,29,29,29,29,29,29] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +[40] \N \N [40] +41 41 \N [] +\N \N \N [] +str_43 \N str_43 [] +[44,44,44,44,44] \N \N [44,44,44,44,44] +45 45 \N [] +\N \N \N [] +str_47 \N str_47 [] +['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] \N \N ['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] +49 49 \N [] +\N \N \N [] +str_51 \N str_51 [] +[52,52,52] \N \N [52,52,52] +53 53 \N [] +\N \N \N [] +str_55 \N str_55 [] +[56,56,56,56,56,56,56] \N \N [56,56,56,56,56,56,56] +57 57 \N [] +\N \N \N [] +str_59 \N str_59 [] +[60] \N \N [60] +61 61 \N [] +\N \N \N [] +str_63 \N str_63 [] +[64,64,64,64,64] \N \N [64,64,64,64,64] +65 65 \N [] +\N \N \N [] +str_67 \N str_67 [] +[68,68,68,68,68,68,68,68,68] \N \N [68,68,68,68,68,68,68,68,68] +69 69 \N [] +\N \N \N [] +str_71 \N str_71 [] +[NULL,NULL,NULL] \N \N [NULL,NULL,NULL] +73 73 \N [] +\N \N \N [] +str_75 \N str_75 [] +[76,76,76,76,76,76,76] \N \N [76,76,76,76,76,76,76] +77 77 \N [] +\N \N \N [] +str_79 \N str_79 [] +0 \N [] +1 \N [] +2 \N [] +3 \N [] +4 \N [] +5 \N [] +6 \N [] +7 \N [] +8 \N [] +9 \N [] +\N str_10 [] +10 \N [] +10 \N [] +\N \N [] +\N \N [] +\N str_11 [] +\N str_12 [] +12 \N [] +12 \N [] +\N str_13 [] +13 \N [] +13 \N [] +\N str_14 [] +14 \N [] +14 \N [] +\N str_15 [] +15 \N [] +15 \N [] +\N \N [] +\N \N [] +\N str_16 [] +\N str_17 [] +17 \N [] +17 \N [] +\N str_18 [] +18 \N [] +18 \N [] +\N str_19 [] +19 \N [] +19 \N [] +\N \N [20] +\N \N ['str_21','str_21'] +\N \N [22,22,22] +\N \N [23,23,23,23] +\N \N [24,24,24,24,24] +\N \N [25,25,25,25,25,25] +\N \N [26,26,26,26,26,26,26] +\N \N [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [28,28,28,28,28,28,28,28,28] +\N \N [29,29,29,29,29,29,29,29,29,29] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [40] +41 \N [] +\N \N [] +\N str_43 [] +\N \N [44,44,44,44,44] +45 \N [] +\N \N [] +\N str_47 [] +\N \N ['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] +49 \N [] +\N \N [] +\N str_51 [] +\N \N [52,52,52] +53 \N [] +\N \N [] +\N str_55 [] +\N \N [56,56,56,56,56,56,56] +57 \N [] +\N \N [] +\N str_59 [] +\N \N [60] +61 \N [] +\N \N [] +\N str_63 [] +\N \N [64,64,64,64,64] +65 \N [] +\N \N [] +\N str_67 [] +\N \N [68,68,68,68,68,68,68,68,68] +69 \N [] +\N \N [] +\N str_71 [] +\N \N [NULL,NULL,NULL] +73 \N [] +\N \N [] +\N str_75 [] +\N \N [76,76,76,76,76,76,76] +77 \N [] +\N \N [] +\N str_79 [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +0 0 \N [] 0 [] +1 1 \N [] 0 [] +2 2 \N [] 0 [] +3 3 \N [] 0 [] +4 4 \N [] 0 [] +5 5 \N [] 0 [] +6 6 \N [] 0 [] +7 7 \N [] 0 [] +8 8 \N [] 0 [] +9 9 \N [] 0 [] +str_10 \N \N [] 0 [] +10 10 \N [] 0 [] +10 10 \N [] 0 [] +[[0,1]] \N \N [] 0 [] +str_11 \N \N [] 0 [] +str_11 \N \N [] 0 [] +str_12 \N \N [] 0 [] +12 12 \N [] 0 [] +12 12 \N [] 0 [] +str_13 \N \N [] 0 [] +13 13 \N [] 0 [] +13 13 \N [] 0 [] +str_14 \N \N [] 0 [] +14 14 \N [] 0 [] +14 14 \N [] 0 [] +str_15 \N \N [] 0 [] +15 15 \N [] 0 [] +15 15 \N [] 0 [] +[[0,1,2,3,4,5,6]] \N \N [] 0 [] +str_16 \N \N [] 0 [] +str_16 \N \N [] 0 [] +str_17 \N \N [] 0 [] +17 17 \N [] 0 [] +17 17 \N [] 0 [] +str_18 \N \N [] 0 [] +18 18 \N [] 0 [] +18 18 \N [] 0 [] +str_19 \N \N [] 0 [] +19 19 \N [] 0 [] +19 19 \N [] 0 [] +[20] \N \N [20] 1 [20] +['str_21','str_21'] \N \N ['str_21','str_21'] 2 [NULL,NULL] +[22,22,22] \N \N [22,22,22] 3 [22,22,22] +[23,23,23,23] \N \N [23,23,23,23] 4 [23,23,23,23] +[24,24,24,24,24] \N \N [24,24,24,24,24] 5 [24,24,24,24,24] +[25,25,25,25,25,25] \N \N [25,25,25,25,25,25] 6 [25,25,25,25,25,25] +[26,26,26,26,26,26,26] \N \N [26,26,26,26,26,26,26] 7 [26,26,26,26,26,26,26] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] \N \N [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] 8 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +[28,28,28,28,28,28,28,28,28] \N \N [28,28,28,28,28,28,28,28,28] 9 [28,28,28,28,28,28,28,28,28] +[29,29,29,29,29,29,29,29,29,29] \N \N [29,29,29,29,29,29,29,29,29,29] 10 [29,29,29,29,29,29,29,29,29,29] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +[40] \N \N [40] 1 [40] +41 41 \N [] 0 [] +\N \N \N [] 0 [] +str_43 \N \N [] 0 [] +[44,44,44,44,44] \N \N [44,44,44,44,44] 5 [44,44,44,44,44] +45 45 \N [] 0 [] +\N \N \N [] 0 [] +str_47 \N \N [] 0 [] +['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] \N \N ['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] 9 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +49 49 \N [] 0 [] +\N \N \N [] 0 [] +str_51 \N \N [] 0 [] +[52,52,52] \N \N [52,52,52] 3 [52,52,52] +53 53 \N [] 0 [] +\N \N \N [] 0 [] +str_55 \N \N [] 0 [] +[56,56,56,56,56,56,56] \N \N [56,56,56,56,56,56,56] 7 [56,56,56,56,56,56,56] +57 57 \N [] 0 [] +\N \N \N [] 0 [] +str_59 \N \N [] 0 [] +[60] \N \N [60] 1 [60] +61 61 \N [] 0 [] +\N \N \N [] 0 [] +str_63 \N \N [] 0 [] +[64,64,64,64,64] \N \N [64,64,64,64,64] 5 [64,64,64,64,64] +65 65 \N [] 0 [] +\N \N \N [] 0 [] +str_67 \N \N [] 0 [] +[68,68,68,68,68,68,68,68,68] \N \N [68,68,68,68,68,68,68,68,68] 9 [68,68,68,68,68,68,68,68,68] +69 69 \N [] 0 [] +\N \N \N [] 0 [] +str_71 \N \N [] 0 [] +[NULL,NULL,NULL] \N \N [NULL,NULL,NULL] 3 [NULL,NULL,NULL] +73 73 \N [] 0 [] +\N \N \N [] 0 [] +str_75 \N \N [] 0 [] +[76,76,76,76,76,76,76] \N \N [76,76,76,76,76,76,76] 7 [76,76,76,76,76,76,76] +77 77 \N [] 0 [] +\N \N \N [] 0 [] +str_79 \N \N [] 0 [] +0 \N [] 0 [] [] +1 \N [] 0 [] [] +2 \N [] 0 [] [] +3 \N [] 0 [] [] +4 \N [] 0 [] [] +5 \N [] 0 [] [] +6 \N [] 0 [] [] +7 \N [] 0 [] [] +8 \N [] 0 [] [] +9 \N [] 0 [] [] +\N \N [] 0 [] [] +10 \N [] 0 [] [] +10 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +12 \N [] 0 [] [] +12 \N [] 0 [] [] +\N \N [] 0 [] [] +13 \N [] 0 [] [] +13 \N [] 0 [] [] +\N \N [] 0 [] [] +14 \N [] 0 [] [] +14 \N [] 0 [] [] +\N \N [] 0 [] [] +15 \N [] 0 [] [] +15 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +17 \N [] 0 [] [] +17 \N [] 0 [] [] +\N \N [] 0 [] [] +18 \N [] 0 [] [] +18 \N [] 0 [] [] +\N \N [] 0 [] [] +19 \N [] 0 [] [] +19 \N [] 0 [] [] +\N \N [20] 1 [20] [NULL] +\N \N ['str_21','str_21'] 2 [NULL,NULL] ['str_21','str_21'] +\N \N [22,22,22] 3 [22,22,22] [NULL,NULL,NULL] +\N \N [23,23,23,23] 4 [23,23,23,23] [NULL,NULL,NULL,NULL] +\N \N [24,24,24,24,24] 5 [24,24,24,24,24] [NULL,NULL,NULL,NULL,NULL] +\N \N [25,25,25,25,25,25] 6 [25,25,25,25,25,25] [NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [26,26,26,26,26,26,26] 7 [26,26,26,26,26,26,26] [NULL,NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] 8 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [28,28,28,28,28,28,28,28,28] 9 [28,28,28,28,28,28,28,28,28] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [29,29,29,29,29,29,29,29,29,29] 10 [29,29,29,29,29,29,29,29,29,29] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [40] 1 [40] [NULL] +41 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [44,44,44,44,44] 5 [44,44,44,44,44] [NULL,NULL,NULL,NULL,NULL] +45 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N ['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] 9 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] ['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] +49 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [52,52,52] 3 [52,52,52] [NULL,NULL,NULL] +53 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [56,56,56,56,56,56,56] 7 [56,56,56,56,56,56,56] [NULL,NULL,NULL,NULL,NULL,NULL,NULL] +57 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [60] 1 [60] [NULL] +61 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [64,64,64,64,64] 5 [64,64,64,64,64] [NULL,NULL,NULL,NULL,NULL] +65 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [68,68,68,68,68,68,68,68,68] 9 [68,68,68,68,68,68,68,68,68] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +69 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [NULL,NULL,NULL] 3 [NULL,NULL,NULL] [NULL,NULL,NULL] +73 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [76,76,76,76,76,76,76] 7 [76,76,76,76,76,76,76] [NULL,NULL,NULL,NULL,NULL,NULL,NULL] +77 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +0 0 [] [] +1 0 [] [] +2 0 [] [] +3 0 [] [] +4 0 [] [] +5 0 [] [] +6 0 [] [] +7 0 [] [] +8 0 [] [] +9 0 [] [] +str_10 0 [] [] +10 0 [] [] +10 0 [] [] +[[0,1]] 0 [] [] +str_11 0 [] [] +str_11 0 [] [] +str_12 0 [] [] +12 0 [] [] +12 0 [] [] +str_13 0 [] [] +13 0 [] [] +13 0 [] [] +str_14 0 [] [] +14 0 [] [] +14 0 [] [] +str_15 0 [] [] +15 0 [] [] +15 0 [] [] +[[0,1,2,3,4,5,6]] 0 [] [] +str_16 0 [] [] +str_16 0 [] [] +str_17 0 [] [] +17 0 [] [] +17 0 [] [] +str_18 0 [] [] +18 0 [] [] +18 0 [] [] +str_19 0 [] [] +19 0 [] [] +19 0 [] [] +[20] 0 [] [20] +['str_21','str_21'] 0 [] [NULL,NULL] +[22,22,22] 0 [] [22,22,22] +[23,23,23,23] 0 [] [23,23,23,23] +[24,24,24,24,24] 0 [] [24,24,24,24,24] +[25,25,25,25,25,25] 0 [] [25,25,25,25,25,25] +[26,26,26,26,26,26,26] 0 [] [26,26,26,26,26,26,26] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] 0 [] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +[28,28,28,28,28,28,28,28,28] 0 [] [28,28,28,28,28,28,28,28,28] +[29,29,29,29,29,29,29,29,29,29] 0 [] [29,29,29,29,29,29,29,29,29,29] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +[40] 0 [] [40] +41 0 [] [] +\N 0 [] [] +str_43 0 [] [] +[44,44,44,44,44] 0 [] [44,44,44,44,44] +45 0 [] [] +\N 0 [] [] +str_47 0 [] [] +['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] 0 [] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +49 0 [] [] +\N 0 [] [] +str_51 0 [] [] +[52,52,52] 0 [] [52,52,52] +53 0 [] [] +\N 0 [] [] +str_55 0 [] [] +[56,56,56,56,56,56,56] 0 [] [56,56,56,56,56,56,56] +57 0 [] [] +\N 0 [] [] +str_59 0 [] [] +[60] 0 [] [60] +61 0 [] [] +\N 0 [] [] +str_63 0 [] [] +[64,64,64,64,64] 0 [] [64,64,64,64,64] +65 0 [] [] +\N 0 [] [] +str_67 0 [] [] +[68,68,68,68,68,68,68,68,68] 0 [] [68,68,68,68,68,68,68,68,68] +69 0 [] [] +\N 0 [] [] +str_71 0 [] [] +[NULL,NULL,NULL] 0 [] [NULL,NULL,NULL] +73 0 [] [] +\N 0 [] [] +str_75 0 [] [] +[76,76,76,76,76,76,76] 0 [] [76,76,76,76,76,76,76] +77 0 [] [] +\N 0 [] [] +str_79 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 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [20] +[] 0 [NULL,NULL] +[] 0 [22,22,22] +[] 0 [23,23,23,23] +[] 0 [24,24,24,24,24] +[] 0 [25,25,25,25,25,25] +[] 0 [26,26,26,26,26,26,26] +[] 0 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +[] 0 [28,28,28,28,28,28,28,28,28] +[] 0 [29,29,29,29,29,29,29,29,29,29] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [40] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [44,44,44,44,44] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [52,52,52] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [56,56,56,56,56,56,56] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [60] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [64,64,64,64,64] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [68,68,68,68,68,68,68,68,68] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [NULL,NULL,NULL] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [76,76,76,76,76,76,76] +[] 0 [] +[] 0 [] +[] 0 [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[2] [[0,1]] [[[],[]]] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[7] [[0,1,2,3,4,5,6]] [[[],[],[],[],[],[],[]]] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000 +Array(Array(Dynamic)) +Array(Variant(String, UInt64)) +LowCardinality(String) +None +String +UInt64 +36 +36 +20 +20 +0 +0 +2 +2 +20 +20 +2 +2 +20 +0 +0 0 \N [] +1 1 \N [] +2 2 \N [] +3 3 \N [] +4 4 \N [] +5 5 \N [] +6 6 \N [] +7 7 \N [] +8 8 \N [] +9 9 \N [] +str_10 \N str_10 [] +10 10 \N [] +10 10 \N [] +[[0,1]] \N \N [] +str_11 \N \N [] +str_11 \N str_11 [] +str_12 \N str_12 [] +12 12 \N [] +12 12 \N [] +str_13 \N str_13 [] +13 13 \N [] +13 13 \N [] +str_14 \N str_14 [] +14 14 \N [] +14 14 \N [] +str_15 \N str_15 [] +15 15 \N [] +15 15 \N [] +[[0,1,2,3,4,5,6]] \N \N [] +str_16 \N \N [] +str_16 \N str_16 [] +str_17 \N str_17 [] +17 17 \N [] +17 17 \N [] +str_18 \N str_18 [] +18 18 \N [] +18 18 \N [] +str_19 \N str_19 [] +19 19 \N [] +19 19 \N [] +[20] \N \N [20] +['str_21','str_21'] \N \N ['str_21','str_21'] +[22,22,22] \N \N [22,22,22] +[23,23,23,23] \N \N [23,23,23,23] +[24,24,24,24,24] \N \N [24,24,24,24,24] +[25,25,25,25,25,25] \N \N [25,25,25,25,25,25] +[26,26,26,26,26,26,26] \N \N [26,26,26,26,26,26,26] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] \N \N [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +[28,28,28,28,28,28,28,28,28] \N \N [28,28,28,28,28,28,28,28,28] +[29,29,29,29,29,29,29,29,29,29] \N \N [29,29,29,29,29,29,29,29,29,29] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +[40] \N \N [40] +41 41 \N [] +\N \N \N [] +str_43 \N str_43 [] +[44,44,44,44,44] \N \N [44,44,44,44,44] +45 45 \N [] +\N \N \N [] +str_47 \N str_47 [] +['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] \N \N ['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] +49 49 \N [] +\N \N \N [] +str_51 \N str_51 [] +[52,52,52] \N \N [52,52,52] +53 53 \N [] +\N \N \N [] +str_55 \N str_55 [] +[56,56,56,56,56,56,56] \N \N [56,56,56,56,56,56,56] +57 57 \N [] +\N \N \N [] +str_59 \N str_59 [] +[60] \N \N [60] +61 61 \N [] +\N \N \N [] +str_63 \N str_63 [] +[64,64,64,64,64] \N \N [64,64,64,64,64] +65 65 \N [] +\N \N \N [] +str_67 \N str_67 [] +[68,68,68,68,68,68,68,68,68] \N \N [68,68,68,68,68,68,68,68,68] +69 69 \N [] +\N \N \N [] +str_71 \N str_71 [] +[NULL,NULL,NULL] \N \N [NULL,NULL,NULL] +73 73 \N [] +\N \N \N [] +str_75 \N str_75 [] +[76,76,76,76,76,76,76] \N \N [76,76,76,76,76,76,76] +77 77 \N [] +\N \N \N [] +str_79 \N str_79 [] +0 \N [] +1 \N [] +2 \N [] +3 \N [] +4 \N [] +5 \N [] +6 \N [] +7 \N [] +8 \N [] +9 \N [] +\N str_10 [] +10 \N [] +10 \N [] +\N \N [] +\N \N [] +\N str_11 [] +\N str_12 [] +12 \N [] +12 \N [] +\N str_13 [] +13 \N [] +13 \N [] +\N str_14 [] +14 \N [] +14 \N [] +\N str_15 [] +15 \N [] +15 \N [] +\N \N [] +\N \N [] +\N str_16 [] +\N str_17 [] +17 \N [] +17 \N [] +\N str_18 [] +18 \N [] +18 \N [] +\N str_19 [] +19 \N [] +19 \N [] +\N \N [20] +\N \N ['str_21','str_21'] +\N \N [22,22,22] +\N \N [23,23,23,23] +\N \N [24,24,24,24,24] +\N \N [25,25,25,25,25,25] +\N \N [26,26,26,26,26,26,26] +\N \N [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [28,28,28,28,28,28,28,28,28] +\N \N [29,29,29,29,29,29,29,29,29,29] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [40] +41 \N [] +\N \N [] +\N str_43 [] +\N \N [44,44,44,44,44] +45 \N [] +\N \N [] +\N str_47 [] +\N \N ['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] +49 \N [] +\N \N [] +\N str_51 [] +\N \N [52,52,52] +53 \N [] +\N \N [] +\N str_55 [] +\N \N [56,56,56,56,56,56,56] +57 \N [] +\N \N [] +\N str_59 [] +\N \N [60] +61 \N [] +\N \N [] +\N str_63 [] +\N \N [64,64,64,64,64] +65 \N [] +\N \N [] +\N str_67 [] +\N \N [68,68,68,68,68,68,68,68,68] +69 \N [] +\N \N [] +\N str_71 [] +\N \N [NULL,NULL,NULL] +73 \N [] +\N \N [] +\N str_75 [] +\N \N [76,76,76,76,76,76,76] +77 \N [] +\N \N [] +\N str_79 [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +0 0 \N [] 0 [] +1 1 \N [] 0 [] +2 2 \N [] 0 [] +3 3 \N [] 0 [] +4 4 \N [] 0 [] +5 5 \N [] 0 [] +6 6 \N [] 0 [] +7 7 \N [] 0 [] +8 8 \N [] 0 [] +9 9 \N [] 0 [] +str_10 \N \N [] 0 [] +10 10 \N [] 0 [] +10 10 \N [] 0 [] +[[0,1]] \N \N [] 0 [] +str_11 \N \N [] 0 [] +str_11 \N \N [] 0 [] +str_12 \N \N [] 0 [] +12 12 \N [] 0 [] +12 12 \N [] 0 [] +str_13 \N \N [] 0 [] +13 13 \N [] 0 [] +13 13 \N [] 0 [] +str_14 \N \N [] 0 [] +14 14 \N [] 0 [] +14 14 \N [] 0 [] +str_15 \N \N [] 0 [] +15 15 \N [] 0 [] +15 15 \N [] 0 [] +[[0,1,2,3,4,5,6]] \N \N [] 0 [] +str_16 \N \N [] 0 [] +str_16 \N \N [] 0 [] +str_17 \N \N [] 0 [] +17 17 \N [] 0 [] +17 17 \N [] 0 [] +str_18 \N \N [] 0 [] +18 18 \N [] 0 [] +18 18 \N [] 0 [] +str_19 \N \N [] 0 [] +19 19 \N [] 0 [] +19 19 \N [] 0 [] +[20] \N \N [20] 1 [20] +['str_21','str_21'] \N \N ['str_21','str_21'] 2 [NULL,NULL] +[22,22,22] \N \N [22,22,22] 3 [22,22,22] +[23,23,23,23] \N \N [23,23,23,23] 4 [23,23,23,23] +[24,24,24,24,24] \N \N [24,24,24,24,24] 5 [24,24,24,24,24] +[25,25,25,25,25,25] \N \N [25,25,25,25,25,25] 6 [25,25,25,25,25,25] +[26,26,26,26,26,26,26] \N \N [26,26,26,26,26,26,26] 7 [26,26,26,26,26,26,26] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] \N \N [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] 8 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +[28,28,28,28,28,28,28,28,28] \N \N [28,28,28,28,28,28,28,28,28] 9 [28,28,28,28,28,28,28,28,28] +[29,29,29,29,29,29,29,29,29,29] \N \N [29,29,29,29,29,29,29,29,29,29] 10 [29,29,29,29,29,29,29,29,29,29] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +[40] \N \N [40] 1 [40] +41 41 \N [] 0 [] +\N \N \N [] 0 [] +str_43 \N \N [] 0 [] +[44,44,44,44,44] \N \N [44,44,44,44,44] 5 [44,44,44,44,44] +45 45 \N [] 0 [] +\N \N \N [] 0 [] +str_47 \N \N [] 0 [] +['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] \N \N ['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] 9 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +49 49 \N [] 0 [] +\N \N \N [] 0 [] +str_51 \N \N [] 0 [] +[52,52,52] \N \N [52,52,52] 3 [52,52,52] +53 53 \N [] 0 [] +\N \N \N [] 0 [] +str_55 \N \N [] 0 [] +[56,56,56,56,56,56,56] \N \N [56,56,56,56,56,56,56] 7 [56,56,56,56,56,56,56] +57 57 \N [] 0 [] +\N \N \N [] 0 [] +str_59 \N \N [] 0 [] +[60] \N \N [60] 1 [60] +61 61 \N [] 0 [] +\N \N \N [] 0 [] +str_63 \N \N [] 0 [] +[64,64,64,64,64] \N \N [64,64,64,64,64] 5 [64,64,64,64,64] +65 65 \N [] 0 [] +\N \N \N [] 0 [] +str_67 \N \N [] 0 [] +[68,68,68,68,68,68,68,68,68] \N \N [68,68,68,68,68,68,68,68,68] 9 [68,68,68,68,68,68,68,68,68] +69 69 \N [] 0 [] +\N \N \N [] 0 [] +str_71 \N \N [] 0 [] +[NULL,NULL,NULL] \N \N [NULL,NULL,NULL] 3 [NULL,NULL,NULL] +73 73 \N [] 0 [] +\N \N \N [] 0 [] +str_75 \N \N [] 0 [] +[76,76,76,76,76,76,76] \N \N [76,76,76,76,76,76,76] 7 [76,76,76,76,76,76,76] +77 77 \N [] 0 [] +\N \N \N [] 0 [] +str_79 \N \N [] 0 [] +0 \N [] 0 [] [] +1 \N [] 0 [] [] +2 \N [] 0 [] [] +3 \N [] 0 [] [] +4 \N [] 0 [] [] +5 \N [] 0 [] [] +6 \N [] 0 [] [] +7 \N [] 0 [] [] +8 \N [] 0 [] [] +9 \N [] 0 [] [] +\N \N [] 0 [] [] +10 \N [] 0 [] [] +10 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +12 \N [] 0 [] [] +12 \N [] 0 [] [] +\N \N [] 0 [] [] +13 \N [] 0 [] [] +13 \N [] 0 [] [] +\N \N [] 0 [] [] +14 \N [] 0 [] [] +14 \N [] 0 [] [] +\N \N [] 0 [] [] +15 \N [] 0 [] [] +15 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +17 \N [] 0 [] [] +17 \N [] 0 [] [] +\N \N [] 0 [] [] +18 \N [] 0 [] [] +18 \N [] 0 [] [] +\N \N [] 0 [] [] +19 \N [] 0 [] [] +19 \N [] 0 [] [] +\N \N [20] 1 [20] [NULL] +\N \N ['str_21','str_21'] 2 [NULL,NULL] ['str_21','str_21'] +\N \N [22,22,22] 3 [22,22,22] [NULL,NULL,NULL] +\N \N [23,23,23,23] 4 [23,23,23,23] [NULL,NULL,NULL,NULL] +\N \N [24,24,24,24,24] 5 [24,24,24,24,24] [NULL,NULL,NULL,NULL,NULL] +\N \N [25,25,25,25,25,25] 6 [25,25,25,25,25,25] [NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [26,26,26,26,26,26,26] 7 [26,26,26,26,26,26,26] [NULL,NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] 8 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [28,28,28,28,28,28,28,28,28] 9 [28,28,28,28,28,28,28,28,28] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [29,29,29,29,29,29,29,29,29,29] 10 [29,29,29,29,29,29,29,29,29,29] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [40] 1 [40] [NULL] +41 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [44,44,44,44,44] 5 [44,44,44,44,44] [NULL,NULL,NULL,NULL,NULL] +45 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N ['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] 9 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] ['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] +49 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [52,52,52] 3 [52,52,52] [NULL,NULL,NULL] +53 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [56,56,56,56,56,56,56] 7 [56,56,56,56,56,56,56] [NULL,NULL,NULL,NULL,NULL,NULL,NULL] +57 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [60] 1 [60] [NULL] +61 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [64,64,64,64,64] 5 [64,64,64,64,64] [NULL,NULL,NULL,NULL,NULL] +65 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [68,68,68,68,68,68,68,68,68] 9 [68,68,68,68,68,68,68,68,68] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +69 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [NULL,NULL,NULL] 3 [NULL,NULL,NULL] [NULL,NULL,NULL] +73 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [76,76,76,76,76,76,76] 7 [76,76,76,76,76,76,76] [NULL,NULL,NULL,NULL,NULL,NULL,NULL] +77 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +0 0 [] [] +1 0 [] [] +2 0 [] [] +3 0 [] [] +4 0 [] [] +5 0 [] [] +6 0 [] [] +7 0 [] [] +8 0 [] [] +9 0 [] [] +str_10 0 [] [] +10 0 [] [] +10 0 [] [] +[[0,1]] 0 [] [] +str_11 0 [] [] +str_11 0 [] [] +str_12 0 [] [] +12 0 [] [] +12 0 [] [] +str_13 0 [] [] +13 0 [] [] +13 0 [] [] +str_14 0 [] [] +14 0 [] [] +14 0 [] [] +str_15 0 [] [] +15 0 [] [] +15 0 [] [] +[[0,1,2,3,4,5,6]] 0 [] [] +str_16 0 [] [] +str_16 0 [] [] +str_17 0 [] [] +17 0 [] [] +17 0 [] [] +str_18 0 [] [] +18 0 [] [] +18 0 [] [] +str_19 0 [] [] +19 0 [] [] +19 0 [] [] +[20] 0 [] [20] +['str_21','str_21'] 0 [] [NULL,NULL] +[22,22,22] 0 [] [22,22,22] +[23,23,23,23] 0 [] [23,23,23,23] +[24,24,24,24,24] 0 [] [24,24,24,24,24] +[25,25,25,25,25,25] 0 [] [25,25,25,25,25,25] +[26,26,26,26,26,26,26] 0 [] [26,26,26,26,26,26,26] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] 0 [] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +[28,28,28,28,28,28,28,28,28] 0 [] [28,28,28,28,28,28,28,28,28] +[29,29,29,29,29,29,29,29,29,29] 0 [] [29,29,29,29,29,29,29,29,29,29] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +[40] 0 [] [40] +41 0 [] [] +\N 0 [] [] +str_43 0 [] [] +[44,44,44,44,44] 0 [] [44,44,44,44,44] +45 0 [] [] +\N 0 [] [] +str_47 0 [] [] +['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] 0 [] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +49 0 [] [] +\N 0 [] [] +str_51 0 [] [] +[52,52,52] 0 [] [52,52,52] +53 0 [] [] +\N 0 [] [] +str_55 0 [] [] +[56,56,56,56,56,56,56] 0 [] [56,56,56,56,56,56,56] +57 0 [] [] +\N 0 [] [] +str_59 0 [] [] +[60] 0 [] [60] +61 0 [] [] +\N 0 [] [] +str_63 0 [] [] +[64,64,64,64,64] 0 [] [64,64,64,64,64] +65 0 [] [] +\N 0 [] [] +str_67 0 [] [] +[68,68,68,68,68,68,68,68,68] 0 [] [68,68,68,68,68,68,68,68,68] +69 0 [] [] +\N 0 [] [] +str_71 0 [] [] +[NULL,NULL,NULL] 0 [] [NULL,NULL,NULL] +73 0 [] [] +\N 0 [] [] +str_75 0 [] [] +[76,76,76,76,76,76,76] 0 [] [76,76,76,76,76,76,76] +77 0 [] [] +\N 0 [] [] +str_79 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 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [20] +[] 0 [NULL,NULL] +[] 0 [22,22,22] +[] 0 [23,23,23,23] +[] 0 [24,24,24,24,24] +[] 0 [25,25,25,25,25,25] +[] 0 [26,26,26,26,26,26,26] +[] 0 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +[] 0 [28,28,28,28,28,28,28,28,28] +[] 0 [29,29,29,29,29,29,29,29,29,29] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [40] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [44,44,44,44,44] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [52,52,52] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [56,56,56,56,56,56,56] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [60] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [64,64,64,64,64] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [68,68,68,68,68,68,68,68,68] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [NULL,NULL,NULL] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [76,76,76,76,76,76,76] +[] 0 [] +[] 0 [] +[] 0 [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[2] [[0,1]] [[[],[]]] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[7] [[0,1,2,3,4,5,6]] [[[],[],[],[],[],[],[]]] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1 +Array(Array(Dynamic)) +Array(Variant(String, UInt64)) +LowCardinality(String) +None +String +UInt64 +36 +36 +20 +20 +0 +0 +2 +2 +20 +20 +2 +2 +20 +0 +0 0 \N [] +1 1 \N [] +2 2 \N [] +3 3 \N [] +4 4 \N [] +5 5 \N [] +6 6 \N [] +7 7 \N [] +8 8 \N [] +9 9 \N [] +str_10 \N str_10 [] +10 10 \N [] +10 10 \N [] +[[0,1]] \N \N [] +str_11 \N \N [] +str_11 \N str_11 [] +str_12 \N str_12 [] +12 12 \N [] +12 12 \N [] +str_13 \N str_13 [] +13 13 \N [] +13 13 \N [] +str_14 \N str_14 [] +14 14 \N [] +14 14 \N [] +str_15 \N str_15 [] +15 15 \N [] +15 15 \N [] +[[0,1,2,3,4,5,6]] \N \N [] +str_16 \N \N [] +str_16 \N str_16 [] +str_17 \N str_17 [] +17 17 \N [] +17 17 \N [] +str_18 \N str_18 [] +18 18 \N [] +18 18 \N [] +str_19 \N str_19 [] +19 19 \N [] +19 19 \N [] +[20] \N \N [20] +['str_21','str_21'] \N \N ['str_21','str_21'] +[22,22,22] \N \N [22,22,22] +[23,23,23,23] \N \N [23,23,23,23] +[24,24,24,24,24] \N \N [24,24,24,24,24] +[25,25,25,25,25,25] \N \N [25,25,25,25,25,25] +[26,26,26,26,26,26,26] \N \N [26,26,26,26,26,26,26] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] \N \N [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +[28,28,28,28,28,28,28,28,28] \N \N [28,28,28,28,28,28,28,28,28] +[29,29,29,29,29,29,29,29,29,29] \N \N [29,29,29,29,29,29,29,29,29,29] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +\N \N \N [] +[40] \N \N [40] +41 41 \N [] +\N \N \N [] +str_43 \N str_43 [] +[44,44,44,44,44] \N \N [44,44,44,44,44] +45 45 \N [] +\N \N \N [] +str_47 \N str_47 [] +['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] \N \N ['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] +49 49 \N [] +\N \N \N [] +str_51 \N str_51 [] +[52,52,52] \N \N [52,52,52] +53 53 \N [] +\N \N \N [] +str_55 \N str_55 [] +[56,56,56,56,56,56,56] \N \N [56,56,56,56,56,56,56] +57 57 \N [] +\N \N \N [] +str_59 \N str_59 [] +[60] \N \N [60] +61 61 \N [] +\N \N \N [] +str_63 \N str_63 [] +[64,64,64,64,64] \N \N [64,64,64,64,64] +65 65 \N [] +\N \N \N [] +str_67 \N str_67 [] +[68,68,68,68,68,68,68,68,68] \N \N [68,68,68,68,68,68,68,68,68] +69 69 \N [] +\N \N \N [] +str_71 \N str_71 [] +[NULL,NULL,NULL] \N \N [NULL,NULL,NULL] +73 73 \N [] +\N \N \N [] +str_75 \N str_75 [] +[76,76,76,76,76,76,76] \N \N [76,76,76,76,76,76,76] +77 77 \N [] +\N \N \N [] +str_79 \N str_79 [] +0 \N [] +1 \N [] +2 \N [] +3 \N [] +4 \N [] +5 \N [] +6 \N [] +7 \N [] +8 \N [] +9 \N [] +\N str_10 [] +10 \N [] +10 \N [] +\N \N [] +\N \N [] +\N str_11 [] +\N str_12 [] +12 \N [] +12 \N [] +\N str_13 [] +13 \N [] +13 \N [] +\N str_14 [] +14 \N [] +14 \N [] +\N str_15 [] +15 \N [] +15 \N [] +\N \N [] +\N \N [] +\N str_16 [] +\N str_17 [] +17 \N [] +17 \N [] +\N str_18 [] +18 \N [] +18 \N [] +\N str_19 [] +19 \N [] +19 \N [] +\N \N [20] +\N \N ['str_21','str_21'] +\N \N [22,22,22] +\N \N [23,23,23,23] +\N \N [24,24,24,24,24] +\N \N [25,25,25,25,25,25] +\N \N [26,26,26,26,26,26,26] +\N \N [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [28,28,28,28,28,28,28,28,28] +\N \N [29,29,29,29,29,29,29,29,29,29] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [40] +41 \N [] +\N \N [] +\N str_43 [] +\N \N [44,44,44,44,44] +45 \N [] +\N \N [] +\N str_47 [] +\N \N ['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] +49 \N [] +\N \N [] +\N str_51 [] +\N \N [52,52,52] +53 \N [] +\N \N [] +\N str_55 [] +\N \N [56,56,56,56,56,56,56] +57 \N [] +\N \N [] +\N str_59 [] +\N \N [60] +61 \N [] +\N \N [] +\N str_63 [] +\N \N [64,64,64,64,64] +65 \N [] +\N \N [] +\N str_67 [] +\N \N [68,68,68,68,68,68,68,68,68] +69 \N [] +\N \N [] +\N str_71 [] +\N \N [NULL,NULL,NULL] +73 \N [] +\N \N [] +\N str_75 [] +\N \N [76,76,76,76,76,76,76] +77 \N [] +\N \N [] +\N str_79 [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +\N \N [] +0 0 \N [] 0 [] +1 1 \N [] 0 [] +2 2 \N [] 0 [] +3 3 \N [] 0 [] +4 4 \N [] 0 [] +5 5 \N [] 0 [] +6 6 \N [] 0 [] +7 7 \N [] 0 [] +8 8 \N [] 0 [] +9 9 \N [] 0 [] +str_10 \N \N [] 0 [] +10 10 \N [] 0 [] +10 10 \N [] 0 [] +[[0,1]] \N \N [] 0 [] +str_11 \N \N [] 0 [] +str_11 \N \N [] 0 [] +str_12 \N \N [] 0 [] +12 12 \N [] 0 [] +12 12 \N [] 0 [] +str_13 \N \N [] 0 [] +13 13 \N [] 0 [] +13 13 \N [] 0 [] +str_14 \N \N [] 0 [] +14 14 \N [] 0 [] +14 14 \N [] 0 [] +str_15 \N \N [] 0 [] +15 15 \N [] 0 [] +15 15 \N [] 0 [] +[[0,1,2,3,4,5,6]] \N \N [] 0 [] +str_16 \N \N [] 0 [] +str_16 \N \N [] 0 [] +str_17 \N \N [] 0 [] +17 17 \N [] 0 [] +17 17 \N [] 0 [] +str_18 \N \N [] 0 [] +18 18 \N [] 0 [] +18 18 \N [] 0 [] +str_19 \N \N [] 0 [] +19 19 \N [] 0 [] +19 19 \N [] 0 [] +[20] \N \N [20] 1 [20] +['str_21','str_21'] \N \N ['str_21','str_21'] 2 [NULL,NULL] +[22,22,22] \N \N [22,22,22] 3 [22,22,22] +[23,23,23,23] \N \N [23,23,23,23] 4 [23,23,23,23] +[24,24,24,24,24] \N \N [24,24,24,24,24] 5 [24,24,24,24,24] +[25,25,25,25,25,25] \N \N [25,25,25,25,25,25] 6 [25,25,25,25,25,25] +[26,26,26,26,26,26,26] \N \N [26,26,26,26,26,26,26] 7 [26,26,26,26,26,26,26] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] \N \N [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] 8 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +[28,28,28,28,28,28,28,28,28] \N \N [28,28,28,28,28,28,28,28,28] 9 [28,28,28,28,28,28,28,28,28] +[29,29,29,29,29,29,29,29,29,29] \N \N [29,29,29,29,29,29,29,29,29,29] 10 [29,29,29,29,29,29,29,29,29,29] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +\N \N \N [] 0 [] +[40] \N \N [40] 1 [40] +41 41 \N [] 0 [] +\N \N \N [] 0 [] +str_43 \N \N [] 0 [] +[44,44,44,44,44] \N \N [44,44,44,44,44] 5 [44,44,44,44,44] +45 45 \N [] 0 [] +\N \N \N [] 0 [] +str_47 \N \N [] 0 [] +['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] \N \N ['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] 9 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +49 49 \N [] 0 [] +\N \N \N [] 0 [] +str_51 \N \N [] 0 [] +[52,52,52] \N \N [52,52,52] 3 [52,52,52] +53 53 \N [] 0 [] +\N \N \N [] 0 [] +str_55 \N \N [] 0 [] +[56,56,56,56,56,56,56] \N \N [56,56,56,56,56,56,56] 7 [56,56,56,56,56,56,56] +57 57 \N [] 0 [] +\N \N \N [] 0 [] +str_59 \N \N [] 0 [] +[60] \N \N [60] 1 [60] +61 61 \N [] 0 [] +\N \N \N [] 0 [] +str_63 \N \N [] 0 [] +[64,64,64,64,64] \N \N [64,64,64,64,64] 5 [64,64,64,64,64] +65 65 \N [] 0 [] +\N \N \N [] 0 [] +str_67 \N \N [] 0 [] +[68,68,68,68,68,68,68,68,68] \N \N [68,68,68,68,68,68,68,68,68] 9 [68,68,68,68,68,68,68,68,68] +69 69 \N [] 0 [] +\N \N \N [] 0 [] +str_71 \N \N [] 0 [] +[NULL,NULL,NULL] \N \N [NULL,NULL,NULL] 3 [NULL,NULL,NULL] +73 73 \N [] 0 [] +\N \N \N [] 0 [] +str_75 \N \N [] 0 [] +[76,76,76,76,76,76,76] \N \N [76,76,76,76,76,76,76] 7 [76,76,76,76,76,76,76] +77 77 \N [] 0 [] +\N \N \N [] 0 [] +str_79 \N \N [] 0 [] +0 \N [] 0 [] [] +1 \N [] 0 [] [] +2 \N [] 0 [] [] +3 \N [] 0 [] [] +4 \N [] 0 [] [] +5 \N [] 0 [] [] +6 \N [] 0 [] [] +7 \N [] 0 [] [] +8 \N [] 0 [] [] +9 \N [] 0 [] [] +\N \N [] 0 [] [] +10 \N [] 0 [] [] +10 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +12 \N [] 0 [] [] +12 \N [] 0 [] [] +\N \N [] 0 [] [] +13 \N [] 0 [] [] +13 \N [] 0 [] [] +\N \N [] 0 [] [] +14 \N [] 0 [] [] +14 \N [] 0 [] [] +\N \N [] 0 [] [] +15 \N [] 0 [] [] +15 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +17 \N [] 0 [] [] +17 \N [] 0 [] [] +\N \N [] 0 [] [] +18 \N [] 0 [] [] +18 \N [] 0 [] [] +\N \N [] 0 [] [] +19 \N [] 0 [] [] +19 \N [] 0 [] [] +\N \N [20] 1 [20] [NULL] +\N \N ['str_21','str_21'] 2 [NULL,NULL] ['str_21','str_21'] +\N \N [22,22,22] 3 [22,22,22] [NULL,NULL,NULL] +\N \N [23,23,23,23] 4 [23,23,23,23] [NULL,NULL,NULL,NULL] +\N \N [24,24,24,24,24] 5 [24,24,24,24,24] [NULL,NULL,NULL,NULL,NULL] +\N \N [25,25,25,25,25,25] 6 [25,25,25,25,25,25] [NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [26,26,26,26,26,26,26] 7 [26,26,26,26,26,26,26] [NULL,NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] 8 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [28,28,28,28,28,28,28,28,28] 9 [28,28,28,28,28,28,28,28,28] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [29,29,29,29,29,29,29,29,29,29] 10 [29,29,29,29,29,29,29,29,29,29] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [40] 1 [40] [NULL] +41 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [44,44,44,44,44] 5 [44,44,44,44,44] [NULL,NULL,NULL,NULL,NULL] +45 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N ['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] 9 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] ['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] +49 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [52,52,52] 3 [52,52,52] [NULL,NULL,NULL] +53 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [56,56,56,56,56,56,56] 7 [56,56,56,56,56,56,56] [NULL,NULL,NULL,NULL,NULL,NULL,NULL] +57 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [60] 1 [60] [NULL] +61 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [64,64,64,64,64] 5 [64,64,64,64,64] [NULL,NULL,NULL,NULL,NULL] +65 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [68,68,68,68,68,68,68,68,68] 9 [68,68,68,68,68,68,68,68,68] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +69 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [NULL,NULL,NULL] 3 [NULL,NULL,NULL] [NULL,NULL,NULL] +73 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [76,76,76,76,76,76,76] 7 [76,76,76,76,76,76,76] [NULL,NULL,NULL,NULL,NULL,NULL,NULL] +77 \N [] 0 [] [] +\N \N [] 0 [] [] +\N \N [] 0 [] [] +0 0 [] [] +1 0 [] [] +2 0 [] [] +3 0 [] [] +4 0 [] [] +5 0 [] [] +6 0 [] [] +7 0 [] [] +8 0 [] [] +9 0 [] [] +str_10 0 [] [] +10 0 [] [] +10 0 [] [] +[[0,1]] 0 [] [] +str_11 0 [] [] +str_11 0 [] [] +str_12 0 [] [] +12 0 [] [] +12 0 [] [] +str_13 0 [] [] +13 0 [] [] +13 0 [] [] +str_14 0 [] [] +14 0 [] [] +14 0 [] [] +str_15 0 [] [] +15 0 [] [] +15 0 [] [] +[[0,1,2,3,4,5,6]] 0 [] [] +str_16 0 [] [] +str_16 0 [] [] +str_17 0 [] [] +17 0 [] [] +17 0 [] [] +str_18 0 [] [] +18 0 [] [] +18 0 [] [] +str_19 0 [] [] +19 0 [] [] +19 0 [] [] +[20] 0 [] [20] +['str_21','str_21'] 0 [] [NULL,NULL] +[22,22,22] 0 [] [22,22,22] +[23,23,23,23] 0 [] [23,23,23,23] +[24,24,24,24,24] 0 [] [24,24,24,24,24] +[25,25,25,25,25,25] 0 [] [25,25,25,25,25,25] +[26,26,26,26,26,26,26] 0 [] [26,26,26,26,26,26,26] +[NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] 0 [] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +[28,28,28,28,28,28,28,28,28] 0 [] [28,28,28,28,28,28,28,28,28] +[29,29,29,29,29,29,29,29,29,29] 0 [] [29,29,29,29,29,29,29,29,29,29] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +\N 0 [] [] +[40] 0 [] [40] +41 0 [] [] +\N 0 [] [] +str_43 0 [] [] +[44,44,44,44,44] 0 [] [44,44,44,44,44] +45 0 [] [] +\N 0 [] [] +str_47 0 [] [] +['str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48','str_48'] 0 [] [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +49 0 [] [] +\N 0 [] [] +str_51 0 [] [] +[52,52,52] 0 [] [52,52,52] +53 0 [] [] +\N 0 [] [] +str_55 0 [] [] +[56,56,56,56,56,56,56] 0 [] [56,56,56,56,56,56,56] +57 0 [] [] +\N 0 [] [] +str_59 0 [] [] +[60] 0 [] [60] +61 0 [] [] +\N 0 [] [] +str_63 0 [] [] +[64,64,64,64,64] 0 [] [64,64,64,64,64] +65 0 [] [] +\N 0 [] [] +str_67 0 [] [] +[68,68,68,68,68,68,68,68,68] 0 [] [68,68,68,68,68,68,68,68,68] +69 0 [] [] +\N 0 [] [] +str_71 0 [] [] +[NULL,NULL,NULL] 0 [] [NULL,NULL,NULL] +73 0 [] [] +\N 0 [] [] +str_75 0 [] [] +[76,76,76,76,76,76,76] 0 [] [76,76,76,76,76,76,76] +77 0 [] [] +\N 0 [] [] +str_79 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 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [20] +[] 0 [NULL,NULL] +[] 0 [22,22,22] +[] 0 [23,23,23,23] +[] 0 [24,24,24,24,24] +[] 0 [25,25,25,25,25,25] +[] 0 [26,26,26,26,26,26,26] +[] 0 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +[] 0 [28,28,28,28,28,28,28,28,28] +[] 0 [29,29,29,29,29,29,29,29,29,29] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [40] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [44,44,44,44,44] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [52,52,52] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [56,56,56,56,56,56,56] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [60] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [64,64,64,64,64] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [68,68,68,68,68,68,68,68,68] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [NULL,NULL,NULL] +[] 0 [] +[] 0 [] +[] 0 [] +[] 0 [76,76,76,76,76,76,76] +[] 0 [] +[] 0 [] +[] 0 [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[2] [[0,1]] [[[],[]]] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[7] [[0,1,2,3,4,5,6]] [[[],[],[],[],[],[],[]]] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] +[] [] [] diff --git a/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_small.sql.j2 b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_small.sql.j2 new file mode 100644 index 00000000000..dde4f3f53c3 --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_small.sql.j2 @@ -0,0 +1,47 @@ +set allow_experimental_variant_type = 1; +set use_variant_as_common_type = 1; +set allow_experimental_dynamic_type = 1; + +drop table if exists test; + +{% for engine in ['Memory', 'MergeTree order by id settings min_rows_for_wide_part=1000000000, min_bytes_for_wide_part=10000000000', 'MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1'] -%} + +select '{{ engine }}'; +create table test (id UInt64, d Dynamic(max_types=2)) engine={{ engine }}; + +insert into test select number, number from numbers(10); +insert into test select number, 'str_' || toString(number) from numbers(10, 10); +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); +insert into test select number, NULL from numbers(30, 10); +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); +insert into test select number, if(number % 5 == 1, [range((number % 10 + 1)::UInt64)]::Array(Array(Dynamic)), number) from numbers(10, 10); +insert into test select number, if(number % 5 == 1, ('str_' || number)::LowCardinality(String)::Dynamic, number::Dynamic) from numbers(10, 10); + +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) == 'LowCardinality(String)'; +select count() from test where d.`LowCardinality(String)` 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 order by id, d; +select d.UInt64, d.String, d.`Array(Variant(String, UInt64))` from test order by id, d; +select d.Int8, d.Date, d.`Array(String)` from test order by id, d; +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 order by id, d; +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 order by id, d; +select d, d.`Tuple(a UInt64, b String)`.a, d.`Array(Dynamic)`.`Variant(String, UInt64)`.UInt64, d.`Array(Variant(String, UInt64))`.UInt64 from test order by id, d; +select d.`Array(Dynamic)`.`Variant(String, UInt64)`.UInt64, d.`Array(Dynamic)`.size0, d.`Array(Variant(String, UInt64))`.UInt64 from test order by id, d; +select d.`Array(Array(Dynamic))`.size1, d.`Array(Array(Dynamic))`.UInt64, d.`Array(Array(Dynamic))`.`Map(String, Tuple(a UInt64))`.values.a from test order by id, d; + +drop table test; + +{% endfor -%} diff --git a/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_wide_merge_tree.reference b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_wide_merge_tree.reference new file mode 100644 index 00000000000..ca6c5dbba82 --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_wide_merge_tree.reference @@ -0,0 +1,20 @@ +Array(Array(Dynamic)) +Array(Variant(String, UInt64)) +LowCardinality(String) +None +String +UInt64 +360000 +360000 +200000 +200000 +0 +0 +20000 +20000 +200000 +200000 +20000 +20000 +200000 +0 diff --git a/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_wide_merge_tree.sql b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_wide_merge_tree.sql new file mode 100644 index 00000000000..61dc8fca01a --- /dev/null +++ b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_wide_merge_tree.sql @@ -0,0 +1,43 @@ +-- Tags: long, no-tsan, no-msan, no-ubsan, no-asan + +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(max_types=2)) 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, if (number % 5 == 1, [range((number % 10 + 1)::UInt64)]::Array(Array(Dynamic)), number) from numbers(100000, 100000) settings min_insert_block_size_rows=50000; +insert into test select number, if (number % 5 == 1, ('str_' || number)::LowCardinality(String)::Dynamic, number::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) == 'LowCardinality(String)'; +select count() from test where d.`LowCardinality(String)` 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 d94a68aa4d8..e1a13d1ce71 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 @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET optimize_move_to_prewhere = 0; SET query_plan_convert_outer_join_to_inner_join = 0; diff --git a/tests/queries/0_stateless/03036_recursive_cte_postgres_2.reference b/tests/queries/0_stateless/03036_recursive_cte_postgres_2.reference index adc06d2edd2..a10e1da6907 100644 --- a/tests/queries/0_stateless/03036_recursive_cte_postgres_2.reference +++ b/tests/queries/0_stateless/03036_recursive_cte_postgres_2.reference @@ -32,7 +32,7 @@ -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- -- Some examples with a tree -- diff --git a/tests/queries/0_stateless/03036_recursive_cte_postgres_2.sql b/tests/queries/0_stateless/03036_recursive_cte_postgres_2.sql index f06b1c62426..b8e850b4721 100644 --- a/tests/queries/0_stateless/03036_recursive_cte_postgres_2.sql +++ b/tests/queries/0_stateless/03036_recursive_cte_postgres_2.sql @@ -32,7 +32,7 @@ -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- -- Some examples with a tree diff --git a/tests/queries/0_stateless/03036_with_numbers.sql b/tests/queries/0_stateless/03036_with_numbers.sql index 3463ce826e2..bd0f6b6179c 100644 --- a/tests/queries/0_stateless/03036_with_numbers.sql +++ b/tests/queries/0_stateless/03036_with_numbers.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/13843 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; WITH 10 AS n SELECT * FROM numbers(n); 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 index d0d777a5a38..b0be05f07a2 100644 --- 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 @@ -1,28 +1,66 @@ -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 +50000 DateTime false +60000 Date false +70000 Array(UInt16) false +80000 String false +100000 None false +100000 UInt64 false +--------------------- +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String false +100000 None false +100000 UInt64 false +--------------------- +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String false +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) false +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) true +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) true +30000 String false +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) true +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +100000 None false +100000 UInt64 true +110000 String false +200000 Map(UInt64, UInt64) false 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 index d2c787040e5..fb23e15738e 100644 --- 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 @@ -2,7 +2,7 @@ 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; +create table test (id UInt64, d Dynamic(max_types=2)) 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); @@ -12,22 +12,37 @@ 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); +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); + +system stop merges test; +insert into test select number, 'str_' || number from numbers(30000); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); +system start merges test; +optimize table test final; +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); + drop table test; 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 index d0d777a5a38..b0be05f07a2 100644 --- 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 @@ -1,28 +1,66 @@ -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 +50000 DateTime false +60000 Date false +70000 Array(UInt16) false +80000 String false +100000 None false +100000 UInt64 false +--------------------- +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String false +100000 None false +100000 UInt64 false +--------------------- +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String false +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) false +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) true +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) true +30000 String false +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) true +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +100000 None false +100000 UInt64 true +110000 String false +200000 Map(UInt64, UInt64) false 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 index f99bf771608..c098a3191e0 100644 --- 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 @@ -2,7 +2,7 @@ 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; +create table test (id UInt64, d Dynamic(max_types=2)) 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); @@ -12,22 +12,36 @@ 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); +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); + +system stop merges test; +insert into test select number, 'str_' || number from numbers(30000); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); +system start merges test; +optimize table test final; +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); 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 index d0d777a5a38..b0be05f07a2 100644 --- 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 @@ -1,28 +1,66 @@ -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 +50000 DateTime false +60000 Date false +70000 Array(UInt16) false +80000 String false +100000 None false +100000 UInt64 false +--------------------- +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String false +100000 None false +100000 UInt64 false +--------------------- +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String false +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) false +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) true +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) true +30000 String false +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) true +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +100000 None false +100000 UInt64 true +110000 String false +200000 Map(UInt64, UInt64) false 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 index be81596d043..17b1e451143 100644 --- 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 @@ -2,7 +2,7 @@ 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; +create table test (id UInt64, d Dynamic(max_types=2)) 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); @@ -12,23 +12,36 @@ 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); +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); +system start merges test; optimize table test final;; +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); + +system stop merges test; +insert into test select number, 'str_' || number from numbers(30000); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); +system start merges test; +optimize table test final; +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); drop table test; 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 index d0d777a5a38..b0be05f07a2 100644 --- 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 @@ -1,28 +1,66 @@ -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 +50000 DateTime false +60000 Date false +70000 Array(UInt16) false +80000 String false +100000 None false +100000 UInt64 false +--------------------- +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String false +100000 None false +100000 UInt64 false +--------------------- +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String false +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) false +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) true +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) true +30000 String false +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +80000 String true +100000 None false +100000 UInt64 false +200000 Map(UInt64, UInt64) false +--------------------- +10000 Tuple(UInt64, UInt64) true +50000 DateTime true +60000 Date true +70000 Array(UInt16) true +100000 None false +100000 UInt64 true +110000 String false +200000 Map(UInt64, UInt64) false 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 index f6396af42a8..fd6c0109263 100644 --- 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 @@ -2,7 +2,7 @@ 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; +create table test (id UInt64, d Dynamic(max_types=2)) 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); @@ -12,22 +12,36 @@ 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); +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); + +system stop merges test; +insert into test select number, 'str_' || number from numbers(30000); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); +system start merges test; +optimize table test final; +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); drop table test; diff --git a/tests/queries/0_stateless/03037_dynamic_merges_small.reference.j2 b/tests/queries/0_stateless/03037_dynamic_merges_small.reference.j2 index 96a854630ed..7d3bc371e36 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_small.reference.j2 +++ b/tests/queries/0_stateless/03037_dynamic_merges_small.reference.j2 @@ -1,112 +1,264 @@ -5 DateTime -6 Date -7 Array(UInt16) -8 String -10 None -10 UInt64 -7 Array(UInt16) -10 None -10 UInt64 -19 String -7 Array(UInt16) -10 None -10 UInt64 -19 String -20 Map(UInt64, UInt64) -10 None -10 UInt64 -20 Map(UInt64, UInt64) -26 String -1 Tuple(UInt64, UInt64) -10 None -10 UInt64 -20 Map(UInt64, UInt64) -26 String -10 None -10 UInt64 -20 Map(UInt64, UInt64) -27 String -5 DateTime -6 Date -7 Array(UInt16) -8 String -10 None -10 UInt64 -7 Array(UInt16) -10 None -10 UInt64 -19 String -7 Array(UInt16) -10 None -10 UInt64 -19 String -20 Map(UInt64, UInt64) -10 None -10 UInt64 -20 Map(UInt64, UInt64) -26 String -1 Tuple(UInt64, UInt64) -10 None -10 UInt64 -20 Map(UInt64, UInt64) -26 String -10 None -10 UInt64 -20 Map(UInt64, UInt64) -27 String -5 DateTime -6 Date -7 Array(UInt16) -8 String -10 None -10 UInt64 -7 Array(UInt16) -10 None -10 UInt64 -19 String -7 Array(UInt16) -10 None -10 UInt64 -19 String -20 Map(UInt64, UInt64) -10 None -10 UInt64 -20 Map(UInt64, UInt64) -26 String -1 Tuple(UInt64, UInt64) -10 None -10 UInt64 -20 Map(UInt64, UInt64) -26 String -10 None -10 UInt64 -20 Map(UInt64, UInt64) -27 String -5 DateTime -6 Date -7 Array(UInt16) -8 String -10 None -10 UInt64 -7 Array(UInt16) -10 None -10 UInt64 -19 String -7 Array(UInt16) -10 None -10 UInt64 -19 String -20 Map(UInt64, UInt64) -10 None -10 UInt64 -20 Map(UInt64, UInt64) -26 String -1 Tuple(UInt64, UInt64) -10 None -10 UInt64 -20 Map(UInt64, UInt64) -26 String -10 None -10 UInt64 -20 Map(UInt64, UInt64) -27 String +5 DateTime false +6 Date false +7 Array(UInt16) false +8 String false +10 None false +10 UInt64 false +--------------------- +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String false +10 None false +10 UInt64 false +--------------------- +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String false +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) false +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) true +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) true +3 String false +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) true +5 DateTime true +6 Date true +7 Array(UInt16) true +10 None false +10 UInt64 true +11 String false +20 Map(UInt64, UInt64) false +5 DateTime false +6 Date false +7 Array(UInt16) false +8 String false +10 None false +10 UInt64 false +--------------------- +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String false +10 None false +10 UInt64 false +--------------------- +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String false +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) false +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) true +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) true +3 String false +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) true +5 DateTime true +6 Date true +7 Array(UInt16) true +10 None false +10 UInt64 true +11 String false +20 Map(UInt64, UInt64) false +5 DateTime false +6 Date false +7 Array(UInt16) false +8 String false +10 None false +10 UInt64 false +--------------------- +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String false +10 None false +10 UInt64 false +--------------------- +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String false +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) false +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) true +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) true +3 String false +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) true +5 DateTime true +6 Date true +7 Array(UInt16) true +10 None false +10 UInt64 true +11 String false +20 Map(UInt64, UInt64) false +5 DateTime false +6 Date false +7 Array(UInt16) false +8 String false +10 None false +10 UInt64 false +--------------------- +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String false +10 None false +10 UInt64 false +--------------------- +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String false +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) false +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) true +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) true +3 String false +5 DateTime true +6 Date true +7 Array(UInt16) true +8 String true +10 None false +10 UInt64 false +20 Map(UInt64, UInt64) false +--------------------- +1 Tuple(UInt64, UInt64) true +5 DateTime true +6 Date true +7 Array(UInt16) true +10 None false +10 UInt64 true +11 String false +20 Map(UInt64, UInt64) false diff --git a/tests/queries/0_stateless/03037_dynamic_merges_small.sql.j2 b/tests/queries/0_stateless/03037_dynamic_merges_small.sql.j2 index 263e92be403..3778399d0a4 100644 --- a/tests/queries/0_stateless/03037_dynamic_merges_small.sql.j2 +++ b/tests/queries/0_stateless/03037_dynamic_merges_small.sql.j2 @@ -9,7 +9,7 @@ drop table if exists test; 'MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1', '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'] -%} -create table test (id UInt64, d Dynamic(max_types=3)) engine={{ engine }}; +create table test (id UInt64, d Dynamic(max_types=2)) engine={{ engine }}; system stop merges test; insert into test select number, number from numbers(10); @@ -19,23 +19,37 @@ insert into test select number, toDate(number) from numbers(6); insert into test select number, toDateTime(number) from numbers(5); insert into test select number, NULL from numbers(10); -select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); system stop merges test; insert into test select number, map(number, number) from numbers(20); -select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); system stop merges test; insert into test select number, tuple(number, number) from numbers(1); -select count(), dynamicType(d) from test group by dynamicType(d) order by count(), dynamicType(d); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); + +system stop merges test; +insert into test select number, 'str_' || number from numbers(3); +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); +system start merges test; +optimize table test final; +select '---------------------'; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(d) order by count(), dynamicType(d); drop table test; diff --git a/tests/queries/0_stateless/03037_precent_rank.reference b/tests/queries/0_stateless/03037_precent_rank.reference new file mode 100644 index 00000000000..6a23f3884cd --- /dev/null +++ b/tests/queries/0_stateless/03037_precent_rank.reference @@ -0,0 +1,22 @@ +Lenovo Thinkpad Laptop 700 1 0 +Sony VAIO Laptop 700 1 0 +Dell Vostro Laptop 800 3 0.6666666666666666 +HP Elite Laptop 1200 4 1 +Microsoft Lumia Smartphone 200 1 0 +HTC One Smartphone 400 2 0.3333333333333333 +Nexus Smartphone 500 3 0.6666666666666666 +iPhone Smartphone 900 4 1 +Kindle Fire Tablet 150 1 0 +Samsung Galaxy Tab Tablet 200 2 0.5 +iPad Tablet 700 3 1 +Others Unknow 200 1 0 +0 1 0 +1 2 1 +2 3 2 +3 4 3 +4 5 4 +5 6 5 +6 7 6 +7 8 7 +8 9 8 +9 10 9 diff --git a/tests/queries/0_stateless/03037_precent_rank.sql b/tests/queries/0_stateless/03037_precent_rank.sql new file mode 100644 index 00000000000..b0f83fa3340 --- /dev/null +++ b/tests/queries/0_stateless/03037_precent_rank.sql @@ -0,0 +1,52 @@ +drop table if exists product_groups; +drop table if exists products; + +CREATE TABLE product_groups ( + group_id Int64, + group_name String +) Engine = Memory; + + +CREATE TABLE products ( + product_id Int64, + product_name String, + price DECIMAL(11, 2), + group_id Int64 +) Engine = Memory; + +INSERT INTO product_groups VALUES (1, 'Smartphone'),(2, 'Laptop'),(3, 'Tablet'); + +INSERT INTO products (product_id,product_name, group_id,price) VALUES (1, 'Microsoft Lumia', 1, 200), (2, 'HTC One', 1, 400), (3, 'Nexus', 1, 500), (4, 'iPhone', 1, 900),(5, 'HP Elite', 2, 1200),(6, 'Lenovo Thinkpad', 2, 700),(7, 'Sony VAIO', 2, 700),(8, 'Dell Vostro', 2, 800),(9, 'iPad', 3, 700),(10, 'Kindle Fire', 3, 150),(11, 'Samsung Galaxy Tab', 3, 200); + +INSERT INTO product_groups VALUES (4, 'Unknow'); +INSERT INTO products (product_id,product_name, group_id,price) VALUES (12, 'Others', 4, 200); + +SELECT * +FROM +( + SELECT + product_name, + group_name, + price, + rank() OVER (PARTITION BY group_name ORDER BY price ASC) AS rank, + percent_rank() OVER (PARTITION BY group_name ORDER BY price ASC) AS percent + FROM products + INNER JOIN product_groups USING (group_id) +) AS t +ORDER BY + group_name ASC, + price ASC, + product_name ASC; + +drop table product_groups; +drop table products; + +select number, row_number, cast(percent_rank * 10000 as Int32) as percent_rank +from ( + select number, row_number() over () as row_number, percent_rank() over (order by number) as percent_rank + from numbers(10000) + order by number + limit 10 +) +settings max_block_size=100; + diff --git a/tests/queries/0_stateless/03037_recursive_cte_postgres_3.reference b/tests/queries/0_stateless/03037_recursive_cte_postgres_3.reference index 3f8f47ac23e..f0ff70a7c60 100644 --- a/tests/queries/0_stateless/03037_recursive_cte_postgres_3.reference +++ b/tests/queries/0_stateless/03037_recursive_cte_postgres_3.reference @@ -32,7 +32,7 @@ -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- -- different tree example -- diff --git a/tests/queries/0_stateless/03037_recursive_cte_postgres_3.sql b/tests/queries/0_stateless/03037_recursive_cte_postgres_3.sql index 3c74ee889c4..213e8bc995d 100644 --- a/tests/queries/0_stateless/03037_recursive_cte_postgres_3.sql +++ b/tests/queries/0_stateless/03037_recursive_cte_postgres_3.sql @@ -32,7 +32,7 @@ -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- -- different tree example diff --git a/tests/queries/0_stateless/03037_union_view.sql b/tests/queries/0_stateless/03037_union_view.sql index 3ea81b829ba..d963444fd91 100644 --- a/tests/queries/0_stateless/03037_union_view.sql +++ b/tests/queries/0_stateless/03037_union_view.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/55803 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; DROP TABLE IF EXISTS broken_table; DROP TABLE IF EXISTS broken_view; @@ -28,4 +28,4 @@ FROM broken_view v FINAL WHERE v.start IS NOT NULL; DROP TABLE IF EXISTS broken_table; -DROP TABLE IF EXISTS broken_view; \ No newline at end of file +DROP TABLE IF EXISTS broken_view; diff --git a/tests/queries/0_stateless/03038_ambiguous_column.sql b/tests/queries/0_stateless/03038_ambiguous_column.sql index 9df3cd9bc9b..131bc552f56 100644 --- a/tests/queries/0_stateless/03038_ambiguous_column.sql +++ b/tests/queries/0_stateless/03038_ambiguous_column.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/48308 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; DROP TABLE IF EXISTS 03038_table; CREATE TABLE 03038_table 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 index 4be740f6050..a4c2df74a74 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.reference +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.reference @@ -1,21 +1,63 @@ -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 +6667 Tuple(a Dynamic(max_types=2)):DateTime false +13333 Tuple(a Dynamic(max_types=2)):IPv4 false +16667 Tuple(a Dynamic(max_types=2)):Date false +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) false +33334 Tuple(a Dynamic(max_types=2)):UInt64 false +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +140000 UInt64:None false +--------------------- +6667 Tuple(a Dynamic(max_types=2)):DateTime true +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 false +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +140000 UInt64:None false +--------------------- +6667 Tuple(a Dynamic(max_types=2)):DateTime true +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +16667 Tuple(a Dynamic(max_types=2)):DateTime false +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 false +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false +--------------------- +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +23334 Tuple(a Dynamic(max_types=2)):DateTime true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 true +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false +--------------------- +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +23334 Tuple(a Dynamic(max_types=2)):DateTime true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 true +40000 String:None false +40000 Tuple(a Dynamic(max_types=2)):DateTime false +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false +--------------------- +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 true +40000 String:None false +63334 Tuple(a Dynamic(max_types=2)):DateTime true +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false 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 index 81888946681..8ba192cb5db 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_horizontal.sql @@ -6,24 +6,39 @@ 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; +create table test (id UInt64, d Dynamic(max_types=2)) 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); +insert into test select number, tuple(if(number % 3 == 0, number, 'str_' || toString(number)))::Tuple(a Dynamic(max_types=2)) from numbers(100000); +insert into test select number, tuple(if(number % 3 == 0, toDate(number), range(number % 10)))::Tuple(a Dynamic(max_types=2)) from numbers(50000); +insert into test select number, multiIf(number % 5 == 0, tuple(if(number % 3 == 0, toDateTime(number), toIPv4(number)))::Tuple(a Dynamic(max_types=2)), number % 5 == 1 or number % 5 == 2, number, 'str_' || number) from numbers(100000); -select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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); +system stop merges test; +insert into test select number, tuple(if(number % 3 == 0, toDateTime(number), NULL))::Tuple(a Dynamic(max_types=2)) from numbers(50000); +insert into test select number, tuple(if(number % 2 == 0, tuple(number), NULL))::Tuple(a Dynamic(max_types=2)) 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; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag order by count(), type; + +system stop merges test; +insert into test select number, tuple(toDateTime(number))::Tuple(a Dynamic(max_types=2)) from numbers(40000); + +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag order by count(), type; +system start merges test; +optimize table test final; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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 index 4be740f6050..a4c2df74a74 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.reference +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.reference @@ -1,21 +1,63 @@ -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 +6667 Tuple(a Dynamic(max_types=2)):DateTime false +13333 Tuple(a Dynamic(max_types=2)):IPv4 false +16667 Tuple(a Dynamic(max_types=2)):Date false +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) false +33334 Tuple(a Dynamic(max_types=2)):UInt64 false +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +140000 UInt64:None false +--------------------- +6667 Tuple(a Dynamic(max_types=2)):DateTime true +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 false +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +140000 UInt64:None false +--------------------- +6667 Tuple(a Dynamic(max_types=2)):DateTime true +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +16667 Tuple(a Dynamic(max_types=2)):DateTime false +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 false +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false +--------------------- +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +23334 Tuple(a Dynamic(max_types=2)):DateTime true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 true +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false +--------------------- +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +23334 Tuple(a Dynamic(max_types=2)):DateTime true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 true +40000 String:None false +40000 Tuple(a Dynamic(max_types=2)):DateTime false +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false +--------------------- +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 true +40000 String:None false +63334 Tuple(a Dynamic(max_types=2)):DateTime true +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false 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 index ba58ca471a2..1ea7eefdd53 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_compact_vertical.sql @@ -6,24 +6,39 @@ 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; +create table test (id UInt64, d Dynamic(max_types=2)) 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); +insert into test select number, tuple(if(number % 3 == 0, number, 'str_' || toString(number)))::Tuple(a Dynamic(max_types=2)) from numbers(100000); +insert into test select number, tuple(if(number % 3 == 0, toDate(number), range(number % 10)))::Tuple(a Dynamic(max_types=2)) from numbers(50000); +insert into test select number, multiIf(number % 5 == 0, tuple(if(number % 3 == 0, toDateTime(number), toIPv4(number)))::Tuple(a Dynamic(max_types=2)), number % 5 == 1 or number % 5 == 2, number, 'str_' || number) from numbers(100000); -select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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); +system stop merges test; +insert into test select number, tuple(if(number % 3 == 0, toDateTime(number), NULL))::Tuple(a Dynamic(max_types=2)) from numbers(50000); +insert into test select number, tuple(if(number % 2 == 0, tuple(number), NULL))::Tuple(a Dynamic(max_types=2)) 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; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag order by count(), type; + +system stop merges test; +insert into test select number, tuple(toDateTime(number))::Tuple(a Dynamic(max_types=2)) from numbers(40000); + +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag order by count(), type; +system start merges test; +optimize table test final; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag order by count(), type; drop table test; diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_small.reference.j2 b/tests/queries/0_stateless/03038_nested_dynamic_merges_small.reference.j2 index ae07c164074..3d7e8b60f73 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_small.reference.j2 +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_small.reference.j2 @@ -1,84 +1,256 @@ -2 Tuple(a Dynamic(max_types=3)):Date -3 Tuple(a Dynamic(max_types=3)):Array(UInt8) -5 Tuple(a Dynamic(max_types=3)):String -5 Tuple(a Dynamic(max_types=3)):UInt64 -10 UInt64:None -3 Tuple(a Dynamic(max_types=3)):Array(UInt8) -5 Tuple(a Dynamic(max_types=3)):UInt64 -7 Tuple(a Dynamic(max_types=3)):String -10 UInt64:None -2 Tuple(a Dynamic(max_types=3)):DateTime -3 Tuple(a Dynamic(max_types=3)):Array(UInt8) -5 Tuple(a Dynamic(max_types=3)):UInt64 -7 Tuple(a Dynamic(max_types=3)):String -10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -10 UInt64:None -13 Tuple(a Dynamic(max_types=3)):None -5 Tuple(a Dynamic(max_types=3)):UInt64 -10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -10 UInt64:None -12 Tuple(a Dynamic(max_types=3)):String -13 Tuple(a Dynamic(max_types=3)):None -2 Tuple(a Dynamic(max_types=3)):Date -3 Tuple(a Dynamic(max_types=3)):Array(UInt8) -5 Tuple(a Dynamic(max_types=3)):String -5 Tuple(a Dynamic(max_types=3)):UInt64 -10 UInt64:None -3 Tuple(a Dynamic(max_types=3)):Array(UInt8) -5 Tuple(a Dynamic(max_types=3)):UInt64 -7 Tuple(a Dynamic(max_types=3)):String -10 UInt64:None -2 Tuple(a Dynamic(max_types=3)):DateTime -3 Tuple(a Dynamic(max_types=3)):Array(UInt8) -5 Tuple(a Dynamic(max_types=3)):UInt64 -7 Tuple(a Dynamic(max_types=3)):String -10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -10 UInt64:None -13 Tuple(a Dynamic(max_types=3)):None -5 Tuple(a Dynamic(max_types=3)):UInt64 -10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -10 UInt64:None -12 Tuple(a Dynamic(max_types=3)):String -13 Tuple(a Dynamic(max_types=3)):None -2 Tuple(a Dynamic(max_types=3)):Date -3 Tuple(a Dynamic(max_types=3)):Array(UInt8) -5 Tuple(a Dynamic(max_types=3)):String -5 Tuple(a Dynamic(max_types=3)):UInt64 -10 UInt64:None -3 Tuple(a Dynamic(max_types=3)):Array(UInt8) -5 Tuple(a Dynamic(max_types=3)):UInt64 -7 Tuple(a Dynamic(max_types=3)):String -10 UInt64:None -2 Tuple(a Dynamic(max_types=3)):DateTime -3 Tuple(a Dynamic(max_types=3)):Array(UInt8) -5 Tuple(a Dynamic(max_types=3)):UInt64 -7 Tuple(a Dynamic(max_types=3)):String -10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -10 UInt64:None -13 Tuple(a Dynamic(max_types=3)):None -5 Tuple(a Dynamic(max_types=3)):UInt64 -10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -10 UInt64:None -12 Tuple(a Dynamic(max_types=3)):String -13 Tuple(a Dynamic(max_types=3)):None -2 Tuple(a Dynamic(max_types=3)):Date -3 Tuple(a Dynamic(max_types=3)):Array(UInt8) -5 Tuple(a Dynamic(max_types=3)):String -5 Tuple(a Dynamic(max_types=3)):UInt64 -10 UInt64:None -3 Tuple(a Dynamic(max_types=3)):Array(UInt8) -5 Tuple(a Dynamic(max_types=3)):UInt64 -7 Tuple(a Dynamic(max_types=3)):String -10 UInt64:None -2 Tuple(a Dynamic(max_types=3)):DateTime -3 Tuple(a Dynamic(max_types=3)):Array(UInt8) -5 Tuple(a Dynamic(max_types=3)):UInt64 -7 Tuple(a Dynamic(max_types=3)):String -10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -10 UInt64:None -13 Tuple(a Dynamic(max_types=3)):None -5 Tuple(a Dynamic(max_types=3)):UInt64 -10 Tuple(a Dynamic(max_types=3)):Tuple(UInt64) -10 UInt64:None -12 Tuple(a Dynamic(max_types=3)):String -13 Tuple(a Dynamic(max_types=3)):None +test +2 Tuple(a Dynamic(max_types=2)):DateTime false +2 Tuple(a Dynamic(max_types=2)):IPv4 false +4 Tuple(a Dynamic(max_types=2)):Date false +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) false +7 Tuple(a Dynamic(max_types=2)):UInt64 false +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):DateTime true +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +7 Tuple(a Dynamic(max_types=2)):UInt64 false +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):DateTime true +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +4 Tuple(a Dynamic(max_types=2)):DateTime false +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +7 Tuple(a Dynamic(max_types=2)):UInt64 false +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +6 Tuple(a Dynamic(max_types=2)):DateTime true +7 Tuple(a Dynamic(max_types=2)):UInt64 true +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +6 Tuple(a Dynamic(max_types=2)):DateTime true +7 Tuple(a Dynamic(max_types=2)):UInt64 true +8 String:None false +8 Tuple(a Dynamic(max_types=2)):DateTime false +13 Tuple(a Dynamic(max_types=2)):String false +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +7 Tuple(a Dynamic(max_types=2)):UInt64 true +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +14 Tuple(a Dynamic(max_types=2)):DateTime true +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false +test +2 Tuple(a Dynamic(max_types=2)):DateTime false +2 Tuple(a Dynamic(max_types=2)):IPv4 false +4 Tuple(a Dynamic(max_types=2)):Date false +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) false +7 Tuple(a Dynamic(max_types=2)):UInt64 false +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):DateTime true +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +7 Tuple(a Dynamic(max_types=2)):UInt64 false +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):DateTime true +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +4 Tuple(a Dynamic(max_types=2)):DateTime false +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +7 Tuple(a Dynamic(max_types=2)):UInt64 false +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +6 Tuple(a Dynamic(max_types=2)):DateTime true +7 Tuple(a Dynamic(max_types=2)):UInt64 true +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +6 Tuple(a Dynamic(max_types=2)):DateTime true +7 Tuple(a Dynamic(max_types=2)):UInt64 true +8 String:None false +8 Tuple(a Dynamic(max_types=2)):DateTime false +13 Tuple(a Dynamic(max_types=2)):String false +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +7 Tuple(a Dynamic(max_types=2)):UInt64 true +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +14 Tuple(a Dynamic(max_types=2)):DateTime true +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false +test +2 Tuple(a Dynamic(max_types=2)):DateTime false +2 Tuple(a Dynamic(max_types=2)):IPv4 false +4 Tuple(a Dynamic(max_types=2)):Date false +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) false +7 Tuple(a Dynamic(max_types=2)):UInt64 false +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):DateTime true +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +7 Tuple(a Dynamic(max_types=2)):UInt64 false +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):DateTime true +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +4 Tuple(a Dynamic(max_types=2)):DateTime false +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +7 Tuple(a Dynamic(max_types=2)):UInt64 false +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +6 Tuple(a Dynamic(max_types=2)):DateTime true +7 Tuple(a Dynamic(max_types=2)):UInt64 true +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +6 Tuple(a Dynamic(max_types=2)):DateTime true +7 Tuple(a Dynamic(max_types=2)):UInt64 true +8 String:None false +8 Tuple(a Dynamic(max_types=2)):DateTime false +13 Tuple(a Dynamic(max_types=2)):String false +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +7 Tuple(a Dynamic(max_types=2)):UInt64 true +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String true +14 Tuple(a Dynamic(max_types=2)):DateTime false +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false +test +2 Tuple(a Dynamic(max_types=2)):DateTime false +2 Tuple(a Dynamic(max_types=2)):IPv4 false +4 Tuple(a Dynamic(max_types=2)):Date false +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) false +7 Tuple(a Dynamic(max_types=2)):UInt64 false +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):DateTime true +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +7 Tuple(a Dynamic(max_types=2)):UInt64 false +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):DateTime true +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +4 Tuple(a Dynamic(max_types=2)):DateTime false +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +7 Tuple(a Dynamic(max_types=2)):UInt64 false +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +6 Tuple(a Dynamic(max_types=2)):DateTime true +7 Tuple(a Dynamic(max_types=2)):UInt64 true +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String false +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +6 Tuple(a Dynamic(max_types=2)):DateTime true +7 Tuple(a Dynamic(max_types=2)):UInt64 true +8 String:None false +8 Tuple(a Dynamic(max_types=2)):DateTime false +13 Tuple(a Dynamic(max_types=2)):String false +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false +--------------------- +2 Tuple(a Dynamic(max_types=2)):IPv4 true +4 Tuple(a Dynamic(max_types=2)):Date true +6 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +7 Tuple(a Dynamic(max_types=2)):UInt64 true +8 String:None false +13 Tuple(a Dynamic(max_types=2)):String true +14 Tuple(a Dynamic(max_types=2)):DateTime false +20 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +26 Tuple(a Dynamic(max_types=2)):None false +28 UInt64:None false diff --git a/tests/queries/0_stateless/03038_nested_dynamic_merges_small.sql.j2 b/tests/queries/0_stateless/03038_nested_dynamic_merges_small.sql.j2 index 7828c2af49c..8682b6cef81 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_small.sql.j2 +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_small.sql.j2 @@ -10,25 +10,41 @@ drop table if exists test; 'MergeTree order by id settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1', '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'] -%} -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; +select 'test'; +create table test (id UInt64, d Dynamic(max_types=2)) engine={{ engine }}; system stop merges test; -insert into test select number, number from numbers(10); -insert into test select number, tuple(if(number % 2 == 0, number, 'str_' || toString(number)))::Tuple(a Dynamic(max_types=3)) from numbers(10); -insert into test select number, tuple(if(number % 3 == 0, toDate(number), range(number % 10)))::Tuple(a Dynamic(max_types=3)) from numbers(5); +insert into test select number, number from numbers(20); +insert into test select number, tuple(if(number % 3 == 0, number, 'str_' || toString(number)))::Tuple(a Dynamic(max_types=2)) from numbers(20); +insert into test select number, tuple(if(number % 3 == 0, toDate(number), range(number % 10)))::Tuple(a Dynamic(max_types=2)) from numbers(10); +insert into test select number, multiIf(number % 5 == 0, tuple(if(number % 3 == 0, toDateTime(number), toIPv4(number)))::Tuple(a Dynamic(max_types=2)), number % 5 == 1 or number % 5 == 2, number, 'str_' || number) from numbers(20); -select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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(5); -insert into test select number, tuple(if(number % 2 == 0, tuple(number), NULL))::Tuple(a Dynamic(max_types=3)) from numbers(20); +system stop merges test; +insert into test select number, tuple(if(number % 3 == 0, toDateTime(number), NULL))::Tuple(a Dynamic(max_types=2)) from numbers(10); +insert into test select number, tuple(if(number % 2 == 0, tuple(number), NULL))::Tuple(a Dynamic(max_types=2)) from numbers(40); -select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag order by count(), type; + +system stop merges test; +insert into test select number, tuple(toDateTime(number))::Tuple(a Dynamic(max_types=2)) from numbers(8); + +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag order by count(), type; +system start merges test; +optimize table test final; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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 index 4be740f6050..a4c2df74a74 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.reference +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.reference @@ -1,21 +1,63 @@ -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 +6667 Tuple(a Dynamic(max_types=2)):DateTime false +13333 Tuple(a Dynamic(max_types=2)):IPv4 false +16667 Tuple(a Dynamic(max_types=2)):Date false +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) false +33334 Tuple(a Dynamic(max_types=2)):UInt64 false +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +140000 UInt64:None false +--------------------- +6667 Tuple(a Dynamic(max_types=2)):DateTime true +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 false +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +140000 UInt64:None false +--------------------- +6667 Tuple(a Dynamic(max_types=2)):DateTime true +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +16667 Tuple(a Dynamic(max_types=2)):DateTime false +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 false +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false +--------------------- +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +23334 Tuple(a Dynamic(max_types=2)):DateTime true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 true +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false +--------------------- +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +23334 Tuple(a Dynamic(max_types=2)):DateTime true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 true +40000 String:None false +40000 Tuple(a Dynamic(max_types=2)):DateTime false +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false +--------------------- +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 true +40000 String:None false +63334 Tuple(a Dynamic(max_types=2)):DateTime true +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false 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 index a53c5b0b2a5..c6a09036c30 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_horizontal.sql @@ -6,24 +6,39 @@ 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; +create table test (id UInt64, d Dynamic(max_types=2)) 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); +insert into test select number, tuple(if(number % 3 == 0, number, 'str_' || toString(number)))::Tuple(a Dynamic(max_types=2)) from numbers(100000); +insert into test select number, tuple(if(number % 3 == 0, toDate(number), range(number % 10)))::Tuple(a Dynamic(max_types=2)) from numbers(50000); +insert into test select number, multiIf(number % 5 == 0, tuple(if(number % 3 == 0, toDateTime(number), toIPv4(number)))::Tuple(a Dynamic(max_types=2)), number % 5 == 1 or number % 5 == 2, number, 'str_' || number) from numbers(100000); -select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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); +system stop merges test; +insert into test select number, tuple(if(number % 3 == 0, toDateTime(number), NULL))::Tuple(a Dynamic(max_types=2)) from numbers(50000); +insert into test select number, tuple(if(number % 2 == 0, tuple(number), NULL))::Tuple(a Dynamic(max_types=2)) 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; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag order by count(), type; + +system stop merges test; +insert into test select number, tuple(toDateTime(number))::Tuple(a Dynamic(max_types=2)) from numbers(40000); + +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag order by count(), type; +system start merges test; +optimize table test final; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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 index 4be740f6050..a4c2df74a74 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.reference +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.reference @@ -1,21 +1,63 @@ -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 +6667 Tuple(a Dynamic(max_types=2)):DateTime false +13333 Tuple(a Dynamic(max_types=2)):IPv4 false +16667 Tuple(a Dynamic(max_types=2)):Date false +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) false +33334 Tuple(a Dynamic(max_types=2)):UInt64 false +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +140000 UInt64:None false +--------------------- +6667 Tuple(a Dynamic(max_types=2)):DateTime true +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 false +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +140000 UInt64:None false +--------------------- +6667 Tuple(a Dynamic(max_types=2)):DateTime true +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +16667 Tuple(a Dynamic(max_types=2)):DateTime false +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 false +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false +--------------------- +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +23334 Tuple(a Dynamic(max_types=2)):DateTime true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 true +40000 String:None false +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false +--------------------- +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +23334 Tuple(a Dynamic(max_types=2)):DateTime true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 true +40000 String:None false +40000 Tuple(a Dynamic(max_types=2)):DateTime false +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false +--------------------- +13333 Tuple(a Dynamic(max_types=2)):IPv4 true +16667 Tuple(a Dynamic(max_types=2)):Date true +33333 Tuple(a Dynamic(max_types=2)):Array(UInt8) true +33334 Tuple(a Dynamic(max_types=2)):UInt64 true +40000 String:None false +63334 Tuple(a Dynamic(max_types=2)):DateTime true +66666 Tuple(a Dynamic(max_types=2)):String false +100000 Tuple(a Dynamic(max_types=2)):Tuple(UInt64) false +133333 Tuple(a Dynamic(max_types=2)):None false +140000 UInt64:None false 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 index 4256b010ec0..c1964c45d98 100644 --- a/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql +++ b/tests/queries/0_stateless/03038_nested_dynamic_merges_wide_vertical.sql @@ -6,24 +6,39 @@ 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; +create table test (id UInt64, d Dynamic(max_types=2)) 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); +insert into test select number, tuple(if(number % 3 == 0, number, 'str_' || toString(number)))::Tuple(a Dynamic(max_types=2)) from numbers(100000); +insert into test select number, tuple(if(number % 3 == 0, toDate(number), range(number % 10)))::Tuple(a Dynamic(max_types=2)) from numbers(50000); +insert into test select number, multiIf(number % 5 == 0, tuple(if(number % 3 == 0, toDateTime(number), toIPv4(number)))::Tuple(a Dynamic(max_types=2)), number % 5 == 1 or number % 5 == 2, number, 'str_' || number) from numbers(100000); -select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=3))`.a) as type from test group by type order by count(), type; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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); +system stop merges test; +insert into test select number, tuple(if(number % 3 == 0, toDateTime(number), NULL))::Tuple(a Dynamic(max_types=2)) from numbers(50000); +insert into test select number, tuple(if(number % 2 == 0, tuple(number), NULL))::Tuple(a Dynamic(max_types=2)) 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; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag 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; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag order by count(), type; + +system stop merges test; +insert into test select number, tuple(toDateTime(number))::Tuple(a Dynamic(max_types=2)) from numbers(40000); + +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag order by count(), type; +system start merges test; +optimize table test final; +select '---------------------'; +select count(), dynamicType(d) || ':' || dynamicType(d.`Tuple(a Dynamic(max_types=2))`.a) as type, isDynamicElementInSharedData(d.`Tuple(a Dynamic(max_types=2))`.a) as flag from test group by type, flag order by count(), type; drop table test; diff --git a/tests/queries/0_stateless/03038_recursive_cte_postgres_4.reference b/tests/queries/0_stateless/03038_recursive_cte_postgres_4.reference index b920fc298b3..cf070eebc38 100644 --- a/tests/queries/0_stateless/03038_recursive_cte_postgres_4.reference +++ b/tests/queries/0_stateless/03038_recursive_cte_postgres_4.reference @@ -32,7 +32,7 @@ -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- -- test cycle detection -- diff --git a/tests/queries/0_stateless/03038_recursive_cte_postgres_4.sql b/tests/queries/0_stateless/03038_recursive_cte_postgres_4.sql index a6fe3b1e55c..7dad74893b9 100644 --- a/tests/queries/0_stateless/03038_recursive_cte_postgres_4.sql +++ b/tests/queries/0_stateless/03038_recursive_cte_postgres_4.sql @@ -32,7 +32,7 @@ -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- -- test cycle detection diff --git a/tests/queries/0_stateless/03039_recursive_cte_postgres_5.reference b/tests/queries/0_stateless/03039_recursive_cte_postgres_5.reference index 2910ac174e7..43399a0ab95 100644 --- a/tests/queries/0_stateless/03039_recursive_cte_postgres_5.reference +++ b/tests/queries/0_stateless/03039_recursive_cte_postgres_5.reference @@ -32,7 +32,7 @@ -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- -- test multiple WITH queries -- diff --git a/tests/queries/0_stateless/03039_recursive_cte_postgres_5.sql b/tests/queries/0_stateless/03039_recursive_cte_postgres_5.sql index e188a455434..eb4043cca42 100644 --- a/tests/queries/0_stateless/03039_recursive_cte_postgres_5.sql +++ b/tests/queries/0_stateless/03039_recursive_cte_postgres_5.sql @@ -33,7 +33,7 @@ -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- -- test multiple WITH queries diff --git a/tests/queries/0_stateless/03039_unknown_identifier_window_function.sql b/tests/queries/0_stateless/03039_unknown_identifier_window_function.sql index 640d217d2f9..652085d9f5a 100644 --- a/tests/queries/0_stateless/03039_unknown_identifier_window_function.sql +++ b/tests/queries/0_stateless/03039_unknown_identifier_window_function.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/45535 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT *, diff --git a/tests/queries/0_stateless/03040_alias_column_join.sql b/tests/queries/0_stateless/03040_alias_column_join.sql index 54f579c0feb..6ffd749a6c4 100644 --- a/tests/queries/0_stateless/03040_alias_column_join.sql +++ b/tests/queries/0_stateless/03040_alias_column_join.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/44365 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; DROP TABLE IF EXISTS 03040_test; CREATE TABLE 03040_test diff --git a/tests/queries/0_stateless/03040_array_sum_and_join.sql b/tests/queries/0_stateless/03040_array_sum_and_join.sql index 9aeddc9f765..90d3d83c9a2 100644 --- a/tests/queries/0_stateless/03040_array_sum_and_join.sql +++ b/tests/queries/0_stateless/03040_array_sum_and_join.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; select t.1 as cnt, t.2 as name, 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 index 2ec301b747b..9386548c74d 100644 --- 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 @@ -1,13 +1,13 @@ initial insert alter add column 1 -3 None +3 None false 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 +4 String false +4 UInt64 false +7 None false 0 0 \N \N \N \N 0 1 1 \N \N \N \N 0 2 2 \N \N \N \N 0 @@ -24,147 +24,158 @@ insert after alter add column 1 13 13 str_13 str_13 \N \N 0 14 14 \N \N \N \N 0 alter modify column 1 -7 None -8 String +4 String true +4 UInt64 true +7 None false 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 +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 12 \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 after alter modify column 1 -8 None -11 String +1 Date true +5 String true +5 UInt64 true +8 None false 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 +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 12 \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 +16 16 16 \N 16 \N 0 17 17 str_17 str_17 \N \N 0 -18 18 1970-01-19 1970-01-19 \N \N 0 +18 18 1970-01-19 \N \N 1970-01-19 0 alter modify column 2 -8 None -11 String +1 Date true +5 String true +5 UInt64 true +8 None false 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 +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 12 \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 +16 16 16 \N 16 \N 0 17 17 str_17 str_17 \N \N 0 -18 18 1970-01-19 1970-01-19 \N \N 0 +18 18 1970-01-19 \N \N 1970-01-19 0 insert after alter modify column 2 -1 Date -1 UInt64 -9 None -12 String +1 String false +1 UInt64 false +2 Date true +5 String true +5 UInt64 true +9 None false 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 +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 12 \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 +16 16 16 \N 16 \N 0 17 17 str_17 str_17 \N \N 0 -18 18 1970-01-19 1970-01-19 \N \N 0 +18 18 1970-01-19 \N \N 1970-01-19 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 +1 String false +1 UInt64 false +2 Date true +5 String true +5 UInt64 true +9 None false 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 +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 12 \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 +16 16 16 \N 0 \N 16 \N 0 17 17 17 \N 0 str_17 \N \N 0 -18 18 18 \N 0 1970-01-19 \N \N 0 +18 18 18 \N 0 \N \N 1970-01-19 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 +1 String false +1 UInt64 false +2 Date true +5 String true +5 UInt64 true +12 None false 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 +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 12 \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 +16 16 16 \N 0 \N 16 \N 0 17 17 17 \N 0 str_17 \N \N 0 -18 18 18 \N 0 1970-01-19 \N \N 0 +18 18 18 \N 0 \N \N 1970-01-19 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 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 index 4ab700306d4..de05ba36177 100644 --- 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 @@ -9,7 +9,7 @@ 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 count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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'; @@ -17,37 +17,37 @@ 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 count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +alter table test modify column d Dynamic(max_types=0) settings mutations_sync=1; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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 count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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); +alter table test modify column d Dynamic(max_types=2) settings mutations_sync=1; +select count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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 count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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 count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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 count(), dynamicType(d), isDynamicElementInSharedData(d) from test group by dynamicType(d), isDynamicElementInSharedData(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 index c592528c3cd..d7123288280 100644 --- a/tests/queries/0_stateless/03040_dynamic_type_alters_1_memory.reference +++ b/tests/queries/0_stateless/03040_dynamic_type_alters_1_memory.reference @@ -24,48 +24,28 @@ insert after alter add column 1 13 13 str_13 str_13 \N \N 0 14 14 \N \N \N \N 0 alter modify column 1 +4 String +4 UInt64 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 +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 12 \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 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 +1 Date +5 String +5 UInt64 8 None 0 0 \N \N \N \N 0 1 1 \N \N \N \N 0 @@ -83,13 +63,37 @@ alter modify column 2 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 +16 16 16 \N 16 \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 +18 18 1970-01-19 \N \N 1970-01-19 0 +alter modify column 2 1 Date +5 String 5 UInt64 -8 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 \N 16 \N 0 +17 17 str_17 str_17 \N \N 0 +18 18 1970-01-19 \N \N 1970-01-19 0 +insert after alter modify column 2 +2 Date +6 String +6 UInt64 9 None 0 0 \N \N \N \N 0 1 1 \N \N \N \N 0 @@ -107,17 +111,17 @@ insert after alter modify column 2 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 +16 16 16 \N 16 \N 0 17 17 str_17 str_17 \N \N 0 -18 18 1970-01-19 1970-01-19 \N \N 0 +18 18 1970-01-19 \N \N 1970-01-19 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 +2 Date +6 String +6 UInt64 9 None 0 0 0 \N 0 \N \N \N 0 1 1 1 \N 0 \N \N \N 0 @@ -135,17 +139,17 @@ alter modify column 3 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 +16 16 16 \N 0 \N 16 \N 0 17 17 17 \N 0 str_17 \N \N 0 -18 18 18 \N 0 1970-01-19 \N \N 0 +18 18 18 \N 0 \N \N 1970-01-19 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 +2 Date +6 String +6 UInt64 12 None 0 0 0 \N 0 \N \N \N 0 1 1 1 \N 0 \N \N \N 0 @@ -163,9 +167,9 @@ insert after alter modify column 3 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 +16 16 16 \N 0 \N 16 \N 0 17 17 17 \N 0 str_17 \N \N 0 -18 18 18 \N 0 1970-01-19 \N \N 0 +18 18 18 \N 0 \N \N 1970-01-19 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 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 index 2ec301b747b..d7123288280 100644 --- 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 @@ -24,147 +24,152 @@ insert after alter add column 1 13 13 str_13 str_13 \N \N 0 14 14 \N \N \N \N 0 alter modify column 1 +4 String +4 UInt64 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 +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 12 \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 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 +5 String +5 UInt64 +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 3 \N \N 0 -4 4 4 4 \N \N 0 -5 5 5 5 \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 12 \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 +16 16 16 \N 16 \N 0 17 17 str_17 str_17 \N \N 0 -18 18 1970-01-19 1970-01-19 \N \N 0 +18 18 1970-01-19 \N \N 1970-01-19 0 +alter modify column 2 +1 Date +5 String +5 UInt64 +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 \N 16 \N 0 +17 17 str_17 str_17 \N \N 0 +18 18 1970-01-19 \N \N 1970-01-19 0 +insert after alter modify column 2 +2 Date +6 String +6 UInt64 +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 \N 16 \N 0 +17 17 str_17 str_17 \N \N 0 +18 18 1970-01-19 \N \N 1970-01-19 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 +2 Date +6 String +6 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 +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 12 \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 +16 16 16 \N 0 \N 16 \N 0 17 17 17 \N 0 str_17 \N \N 0 -18 18 18 \N 0 1970-01-19 \N \N 0 +18 18 18 \N 0 \N \N 1970-01-19 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 +2 Date +6 String +6 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 +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 12 \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 +16 16 16 \N 0 \N 16 \N 0 17 17 17 \N 0 str_17 \N \N 0 -18 18 18 \N 0 1970-01-19 \N \N 0 +18 18 18 \N 0 \N \N 1970-01-19 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 diff --git a/tests/queries/0_stateless/03040_recursive_cte_postgres_6.reference b/tests/queries/0_stateless/03040_recursive_cte_postgres_6.reference index 2d9d7bb9f6c..7ccc3df9f0d 100644 --- a/tests/queries/0_stateless/03040_recursive_cte_postgres_6.reference +++ b/tests/queries/0_stateless/03040_recursive_cte_postgres_6.reference @@ -32,7 +32,7 @@ -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- -- error cases -- diff --git a/tests/queries/0_stateless/03040_recursive_cte_postgres_6.sql b/tests/queries/0_stateless/03040_recursive_cte_postgres_6.sql index bff2ece6ece..6a6044198c8 100644 --- a/tests/queries/0_stateless/03040_recursive_cte_postgres_6.sql +++ b/tests/queries/0_stateless/03040_recursive_cte_postgres_6.sql @@ -32,7 +32,7 @@ -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- -- error cases diff --git a/tests/queries/0_stateless/03041_analyzer_gigachad_join.sql b/tests/queries/0_stateless/03041_analyzer_gigachad_join.sql index 7906e65f8b8..88f7fc562b1 100644 --- a/tests/queries/0_stateless/03041_analyzer_gigachad_join.sql +++ b/tests/queries/0_stateless/03041_analyzer_gigachad_join.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE IF NOT EXISTS first engine = MergeTree PARTITION BY (inn, toYYYYMM(received)) ORDER BY (inn, sessionId) AS SELECT now() AS received, '123456789' AS inn, '42' AS sessionId; diff --git a/tests/queries/0_stateless/03041_dynamic_type_check_table.sh b/tests/queries/0_stateless/03041_dynamic_type_check_table.sh index c8bd533e253..da24b892cbd 100755 --- a/tests/queries/0_stateless/03041_dynamic_type_check_table.sh +++ b/tests/queries/0_stateless/03041_dynamic_type_check_table.sh @@ -13,7 +13,7 @@ function run() $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(max_types=3) settings mutations_sync=1" + $CH_CLIENT -q "alter table test add column d Dynamic(max_types=2) 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" diff --git a/tests/queries/0_stateless/03041_recursive_cte_postgres_7.reference b/tests/queries/0_stateless/03041_recursive_cte_postgres_7.reference index 6bcc3f89dcb..b98b2ec7e89 100644 --- a/tests/queries/0_stateless/03041_recursive_cte_postgres_7.reference +++ b/tests/queries/0_stateless/03041_recursive_cte_postgres_7.reference @@ -32,7 +32,7 @@ -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; WITH RECURSIVE foo AS (SELECT 1 AS i UNION ALL diff --git a/tests/queries/0_stateless/03041_recursive_cte_postgres_7.sql b/tests/queries/0_stateless/03041_recursive_cte_postgres_7.sql index 71c34b7361f..5f4455efcc6 100644 --- a/tests/queries/0_stateless/03041_recursive_cte_postgres_7.sql +++ b/tests/queries/0_stateless/03041_recursive_cte_postgres_7.sql @@ -32,7 +32,7 @@ -- { echoOn } -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; WITH RECURSIVE foo AS (SELECT 1 AS i diff --git a/tests/queries/0_stateless/03041_select_with_query_result.sql b/tests/queries/0_stateless/03041_select_with_query_result.sql index 061223b43e1..e5897ea12cf 100644 --- a/tests/queries/0_stateless/03041_select_with_query_result.sql +++ b/tests/queries/0_stateless/03041_select_with_query_result.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/44153 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; DROP TABLE IF EXISTS parent; DROP TABLE IF EXISTS join_table_1; DROP TABLE IF EXISTS join_table_2; diff --git a/tests/queries/0_stateless/03042_analyzer_alias_join.sql b/tests/queries/0_stateless/03042_analyzer_alias_join.sql index dac3b6a4983..d9a8d8b4c7b 100644 --- a/tests/queries/0_stateless/03042_analyzer_alias_join.sql +++ b/tests/queries/0_stateless/03042_analyzer_alias_join.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/14978 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE test1(id UInt64, t1value UInt64) ENGINE=MergeTree ORDER BY tuple(); CREATE TABLE test2(id UInt64, t2value String) ENGINE=MergeTree ORDER BY tuple(); diff --git a/tests/queries/0_stateless/03042_not_found_column_c1.sql b/tests/queries/0_stateless/03042_not_found_column_c1.sql index b4dce2af489..08202dc0dca 100644 --- a/tests/queries/0_stateless/03042_not_found_column_c1.sql +++ b/tests/queries/0_stateless/03042_not_found_column_c1.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/42399 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE IF NOT EXISTS t0 (c0 Int32) ENGINE = Memory() ; CREATE TABLE t1 (c0 Int32, c1 Int32, c2 Int32) ENGINE = Memory() ; diff --git a/tests/queries/0_stateless/03043_group_array_result_is_expected.sql b/tests/queries/0_stateless/03043_group_array_result_is_expected.sql index 5311927ae3c..e2c79e5c41e 100644 --- a/tests/queries/0_stateless/03043_group_array_result_is_expected.sql +++ b/tests/queries/0_stateless/03043_group_array_result_is_expected.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/27115 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; drop table if exists fill_ex; create table fill_ex ( diff --git a/tests/queries/0_stateless/03044_analyzer_alias_join.sql b/tests/queries/0_stateless/03044_analyzer_alias_join.sql index 3ab8edb005f..7636edbb411 100644 --- a/tests/queries/0_stateless/03044_analyzer_alias_join.sql +++ b/tests/queries/0_stateless/03044_analyzer_alias_join.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/17319 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TEMPORARY TABLE hits (date Date, data Float64) engine=Memory(); SELECT diff --git a/tests/queries/0_stateless/03044_array_join_columns_in_nested_table.sql b/tests/queries/0_stateless/03044_array_join_columns_in_nested_table.sql index 0cf05763202..4885b7e3f81 100644 --- a/tests/queries/0_stateless/03044_array_join_columns_in_nested_table.sql +++ b/tests/queries/0_stateless/03044_array_join_columns_in_nested_table.sql @@ -1,3 +1,3 @@ -- https://github.com/ClickHouse/ClickHouse/issues/11813 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; select 1 from (select 1 x) l join (select 1 y, [1] a) r on l.x = r.y array join r.a; diff --git a/tests/queries/0_stateless/03045_analyzer_alias_join_with_if.sql b/tests/queries/0_stateless/03045_analyzer_alias_join_with_if.sql index ee8756b9460..cbc46726467 100644 --- a/tests/queries/0_stateless/03045_analyzer_alias_join_with_if.sql +++ b/tests/queries/0_stateless/03045_analyzer_alias_join_with_if.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/13210 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE test_a_table ( name String, a_col String diff --git a/tests/queries/0_stateless/03045_unknown_identifier_alias_substitution.sql b/tests/queries/0_stateless/03045_unknown_identifier_alias_substitution.sql index d97dfc880b3..967b7b24787 100644 --- a/tests/queries/0_stateless/03045_unknown_identifier_alias_substitution.sql +++ b/tests/queries/0_stateless/03045_unknown_identifier_alias_substitution.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/23053 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; DROP TABLE IF EXISTS repl_tbl; CREATE TEMPORARY TABLE repl_tbl diff --git a/tests/queries/0_stateless/03046_column_in_block_array_join.sql b/tests/queries/0_stateless/03046_column_in_block_array_join.sql index c6b4613af3f..f91a18da831 100644 --- a/tests/queries/0_stateless/03046_column_in_block_array_join.sql +++ b/tests/queries/0_stateless/03046_column_in_block_array_join.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/37729 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; DROP TABLE IF EXISTS nested_test; DROP TABLE IF EXISTS join_test; diff --git a/tests/queries/0_stateless/03047_analyzer_alias_join.sql b/tests/queries/0_stateless/03047_analyzer_alias_join.sql index 7d44c92b6f1..29fc711aaf4 100644 --- a/tests/queries/0_stateless/03047_analyzer_alias_join.sql +++ b/tests/queries/0_stateless/03047_analyzer_alias_join.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT 1 AS value, * diff --git a/tests/queries/0_stateless/03047_group_by_field_identified_aggregation.sql b/tests/queries/0_stateless/03047_group_by_field_identified_aggregation.sql index cfaf1df44bd..d12e85ca03d 100644 --- a/tests/queries/0_stateless/03047_group_by_field_identified_aggregation.sql +++ b/tests/queries/0_stateless/03047_group_by_field_identified_aggregation.sql @@ -1,4 +1,4 @@ -- https://github.com/ClickHouse/ClickHouse/issues/32639 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT 0 AND id ? 1 : 2 AS a, sum(id) FROM (SELECT 1 AS id) GROUP BY a; diff --git a/tests/queries/0_stateless/03048_not_found_column_xxx_in_block.sql b/tests/queries/0_stateless/03048_not_found_column_xxx_in_block.sql index 42fd581e142..f511ea81e26 100644 --- a/tests/queries/0_stateless/03048_not_found_column_xxx_in_block.sql +++ b/tests/queries/0_stateless/03048_not_found_column_xxx_in_block.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/41964 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; DROP TABLE IF EXISTS ab_12_aaa; DROP TABLE IF EXISTS ab_12_bbb; diff --git a/tests/queries/0_stateless/03049_analyzer_group_by_alias.sql b/tests/queries/0_stateless/03049_analyzer_group_by_alias.sql index d25babe6788..712d1c27e6e 100644 --- a/tests/queries/0_stateless/03049_analyzer_group_by_alias.sql +++ b/tests/queries/0_stateless/03049_analyzer_group_by_alias.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/7520 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE test (`a` UInt32, `b` UInt32) ENGINE = Memory; INSERT INTO test VALUES (1,2), (1,3), (2,4); diff --git a/tests/queries/0_stateless/03049_unknown_identifier_materialized_column.sql b/tests/queries/0_stateless/03049_unknown_identifier_materialized_column.sql index 938f270b9e4..0efe59a1f1c 100644 --- a/tests/queries/0_stateless/03049_unknown_identifier_materialized_column.sql +++ b/tests/queries/0_stateless/03049_unknown_identifier_materialized_column.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/54317 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; DROP DATABASE IF EXISTS {CLICKHOUSE_DATABASE:Identifier}; CREATE DATABASE {CLICKHOUSE_DATABASE:Identifier}; diff --git a/tests/queries/0_stateless/03050_select_one_one_one.sql b/tests/queries/0_stateless/03050_select_one_one_one.sql index 28a55e0c471..09f3f20c35d 100644 --- a/tests/queries/0_stateless/03050_select_one_one_one.sql +++ b/tests/queries/0_stateless/03050_select_one_one_one.sql @@ -1,4 +1,4 @@ -- https://github.com/ClickHouse/ClickHouse/issues/36973 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT 1, 1, 1; SELECT * FROM (SELECT 1, 1, 1); diff --git a/tests/queries/0_stateless/03051_many_ctes.sql b/tests/queries/0_stateless/03051_many_ctes.sql index d4e613bd279..e442813b6a2 100644 --- a/tests/queries/0_stateless/03051_many_ctes.sql +++ b/tests/queries/0_stateless/03051_many_ctes.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/40955 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; WITH toInt64(2) AS new_x SELECT new_x AS x FROM (SELECT 1 AS x) t; WITH toInt64(2) AS new_x SELECT * replace(new_x as x) FROM (SELECT 1 AS x) t; SELECT 2 AS x FROM (SELECT 1 AS x) t; diff --git a/tests/queries/0_stateless/03052_query_hash_includes_aliases.sql b/tests/queries/0_stateless/03052_query_hash_includes_aliases.sql index 24e9ab0f36e..55993175bf3 100644 --- a/tests/queries/0_stateless/03052_query_hash_includes_aliases.sql +++ b/tests/queries/0_stateless/03052_query_hash_includes_aliases.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/pull/40065 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT ( diff --git a/tests/queries/0_stateless/03053_analyzer_join_alias.sql b/tests/queries/0_stateless/03053_analyzer_join_alias.sql index 894b8af7c6f..677cf9d4d5e 100644 --- a/tests/queries/0_stateless/03053_analyzer_join_alias.sql +++ b/tests/queries/0_stateless/03053_analyzer_join_alias.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/23104 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; DROP DATABASE IF EXISTS {CLICKHOUSE_DATABASE:Identifier}; CREATE DATABASE {CLICKHOUSE_DATABASE:Identifier}; diff --git a/tests/queries/0_stateless/03054_analyzer_join_alias.sql b/tests/queries/0_stateless/03054_analyzer_join_alias.sql index e124aa33a9b..f018f57cc6f 100644 --- a/tests/queries/0_stateless/03054_analyzer_join_alias.sql +++ b/tests/queries/0_stateless/03054_analyzer_join_alias.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/21584 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT count() FROM ( diff --git a/tests/queries/0_stateless/03055_analyzer_subquery_group_array.sql b/tests/queries/0_stateless/03055_analyzer_subquery_group_array.sql index 25b6dcb3564..29ba1dd7c5b 100644 --- a/tests/queries/0_stateless/03055_analyzer_subquery_group_array.sql +++ b/tests/queries/0_stateless/03055_analyzer_subquery_group_array.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/23344 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT logTrace(repeat('Hello', 100)), ignore(*) FROM ( SELECT ignore((SELECT groupArrayState(([number], [number])) FROM numbers(19000))) diff --git a/tests/queries/0_stateless/03057_analyzer_subquery_alias_join.sql b/tests/queries/0_stateless/03057_analyzer_subquery_alias_join.sql index 2217af327fa..92f603ed595 100644 --- a/tests/queries/0_stateless/03057_analyzer_subquery_alias_join.sql +++ b/tests/queries/0_stateless/03057_analyzer_subquery_alias_join.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/10276 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT sum(x.n) as n, sum(z.n) as n2 diff --git a/tests/queries/0_stateless/03058_analyzer_ambiguous_columns.sql b/tests/queries/0_stateless/03058_analyzer_ambiguous_columns.sql index 3cce77f0240..ef3c0e5f63d 100644 --- a/tests/queries/0_stateless/03058_analyzer_ambiguous_columns.sql +++ b/tests/queries/0_stateless/03058_analyzer_ambiguous_columns.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/4567 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; DROP TABLE IF EXISTS fact; DROP TABLE IF EXISTS animals; DROP TABLE IF EXISTS colors; @@ -23,4 +23,3 @@ select id, animal_name, animal_key, color_name, color_key from fact a left join (select toInt64(animal_key) animal_key, animal_name from animals) b on (a.animal_key = b.animal_key) left join (select toInt64(color_key) color_key, color_name from colors) c on (a.color_key = c.color_key); -- { serverError AMBIGUOUS_IDENTIFIER } - diff --git a/tests/queries/0_stateless/03059_analyzer_join_engine_missing_column.sql b/tests/queries/0_stateless/03059_analyzer_join_engine_missing_column.sql index 27782462075..164a42e5ba3 100644 --- a/tests/queries/0_stateless/03059_analyzer_join_engine_missing_column.sql +++ b/tests/queries/0_stateless/03059_analyzer_join_engine_missing_column.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/17710 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE id_val(id UInt32, val UInt32) ENGINE = Memory; CREATE TABLE id_val_join0(id UInt32, val UInt8) ENGINE = Join(ANY, LEFT, id) SETTINGS join_use_nulls = 0; CREATE TABLE id_val_join1(id UInt32, val UInt8) ENGINE = Join(ANY, LEFT, id) SETTINGS join_use_nulls = 1; diff --git a/tests/queries/0_stateless/03060_analyzer_regular_view_alias.sql b/tests/queries/0_stateless/03060_analyzer_regular_view_alias.sql index f8cd8690ee5..0556683b97a 100644 --- a/tests/queries/0_stateless/03060_analyzer_regular_view_alias.sql +++ b/tests/queries/0_stateless/03060_analyzer_regular_view_alias.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/11068 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; create table vt(datetime_value DateTime, value Float64) Engine=Memory; create view computed_datum_hours as diff --git a/tests/queries/0_stateless/03061_analyzer_alias_as_right_key_in_join.sql b/tests/queries/0_stateless/03061_analyzer_alias_as_right_key_in_join.sql index 6fee6d1f73d..a1b50967a46 100644 --- a/tests/queries/0_stateless/03061_analyzer_alias_as_right_key_in_join.sql +++ b/tests/queries/0_stateless/03061_analyzer_alias_as_right_key_in_join.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/24395 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE xxxx_yyy (key UInt32, key_b ALIAS key) ENGINE=MergeTree() ORDER BY key; INSERT INTO xxxx_yyy SELECT number FROM numbers(10); diff --git a/tests/queries/0_stateless/03062_analyzer_join_engine_missing_column.sql b/tests/queries/0_stateless/03062_analyzer_join_engine_missing_column.sql index 9748175e4d4..487d74b3317 100644 --- a/tests/queries/0_stateless/03062_analyzer_join_engine_missing_column.sql +++ b/tests/queries/0_stateless/03062_analyzer_join_engine_missing_column.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/23416 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; create table test (TOPIC String, PARTITION UInt64, OFFSET UInt64, ID UInt64) ENGINE ReplicatedMergeTree('/clickhouse/tables/{database}/test_03062', 'r2') ORDER BY (TOPIC, PARTITION, OFFSET); create table test_join (TOPIC String, PARTITION UInt64, OFFSET UInt64) ENGINE = Join(ANY, LEFT, `TOPIC`, `PARTITION`) SETTINGS join_any_take_last_row = 1; diff --git a/tests/queries/0_stateless/03063_analyzer_multi_join_wrong_table_specifier.sql b/tests/queries/0_stateless/03063_analyzer_multi_join_wrong_table_specifier.sql index 7eab1fa846a..5655d4a0110 100644 --- a/tests/queries/0_stateless/03063_analyzer_multi_join_wrong_table_specifier.sql +++ b/tests/queries/0_stateless/03063_analyzer_multi_join_wrong_table_specifier.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/23162 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE t1 ( k Int64, x Int64) ENGINE = Memory; CREATE TABLE t2( x Int64 ) ENGINE = Memory; @@ -14,4 +14,3 @@ WHERE (t1.d >= now()); -- { serverError UNKNOWN_IDENTIFIER } SELECT * FROM t1 INNER JOIN s ON t1.k = s.k WHERE (t1.d >= now()); -- { serverError UNKNOWN_IDENTIFIER } - diff --git a/tests/queries/0_stateless/03064_analyzer_named_subqueries.sql b/tests/queries/0_stateless/03064_analyzer_named_subqueries.sql index 59ebb9d9af3..d5696411091 100644 --- a/tests/queries/0_stateless/03064_analyzer_named_subqueries.sql +++ b/tests/queries/0_stateless/03064_analyzer_named_subqueries.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/25655 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT sum(t.b) / 1 a, sum(t.a) diff --git a/tests/queries/0_stateless/03065_analyzer_cross_join_and_array_join.sql b/tests/queries/0_stateless/03065_analyzer_cross_join_and_array_join.sql index 7e6befe181e..5034e2eed8f 100644 --- a/tests/queries/0_stateless/03065_analyzer_cross_join_and_array_join.sql +++ b/tests/queries/0_stateless/03065_analyzer_cross_join_and_array_join.sql @@ -1,3 +1,3 @@ -- https://github.com/ClickHouse/ClickHouse/issues/11757 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; select * from (select [1, 2] a) aa cross join (select [3, 4] b) bb array join aa.a, bb.b; diff --git a/tests/queries/0_stateless/03066_analyzer_global_with_statement.sql b/tests/queries/0_stateless/03066_analyzer_global_with_statement.sql index 8983be242c3..2b879ed73da 100644 --- a/tests/queries/0_stateless/03066_analyzer_global_with_statement.sql +++ b/tests/queries/0_stateless/03066_analyzer_global_with_statement.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; WITH 0 AS test SELECT * FROM diff --git a/tests/queries/0_stateless/03067_analyzer_complex_alias_join.sql b/tests/queries/0_stateless/03067_analyzer_complex_alias_join.sql index 052a9eaf734..58845b93771 100644 --- a/tests/queries/0_stateless/03067_analyzer_complex_alias_join.sql +++ b/tests/queries/0_stateless/03067_analyzer_complex_alias_join.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; with d as (select 'key'::Varchar(255) c, 'x'::Varchar(255) s) SELECT r1, c as r2 FROM ( diff --git a/tests/queries/0_stateless/03068_analyzer_distributed_join.sql b/tests/queries/0_stateless/03068_analyzer_distributed_join.sql index 542380feb7c..459c8f5e8ac 100644 --- a/tests/queries/0_stateless/03068_analyzer_distributed_join.sql +++ b/tests/queries/0_stateless/03068_analyzer_distributed_join.sql @@ -1,7 +1,7 @@ -- Tags: no-replicated-database -- Closes: https://github.com/ClickHouse/ClickHouse/issues/6571 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE LINEITEM_shard ON CLUSTER test_shard_localhost ( L_ORDERKEY UInt64, diff --git a/tests/queries/0_stateless/03069_analyzer_with_alias_in_array_join.sql b/tests/queries/0_stateless/03069_analyzer_with_alias_in_array_join.sql index 09d2985fe60..5ec04cbc0f5 100644 --- a/tests/queries/0_stateless/03069_analyzer_with_alias_in_array_join.sql +++ b/tests/queries/0_stateless/03069_analyzer_with_alias_in_array_join.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/4432 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; WITH [1, 2] AS zz SELECT x FROM system.one diff --git a/tests/queries/0_stateless/03070_analyzer_CTE_scalar_as_numbers.sql b/tests/queries/0_stateless/03070_analyzer_CTE_scalar_as_numbers.sql index 7aadab2ca73..a94ae811476 100644 --- a/tests/queries/0_stateless/03070_analyzer_CTE_scalar_as_numbers.sql +++ b/tests/queries/0_stateless/03070_analyzer_CTE_scalar_as_numbers.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/8259 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; with (select 25) as something select *, something diff --git a/tests/queries/0_stateless/03071_analyzer_array_join_forbid_non_existing_columns.sql b/tests/queries/0_stateless/03071_analyzer_array_join_forbid_non_existing_columns.sql index e2eb758d649..211fa2a3119 100644 --- a/tests/queries/0_stateless/03071_analyzer_array_join_forbid_non_existing_columns.sql +++ b/tests/queries/0_stateless/03071_analyzer_array_join_forbid_non_existing_columns.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/9233 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT * FROM ( diff --git a/tests/queries/0_stateless/03071_fix_short_circuit_logic.sql b/tests/queries/0_stateless/03071_fix_short_circuit_logic.sql index 7745bceca0b..dc5fb5085fb 100644 --- a/tests/queries/0_stateless/03071_fix_short_circuit_logic.sql +++ b/tests/queries/0_stateless/03071_fix_short_circuit_logic.sql @@ -59,4 +59,4 @@ FROM ) 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 +SETTINGS max_threads = 1, short_circuit_function_evaluation = 'enable', enable_analyzer = 0; diff --git a/tests/queries/0_stateless/03072_analyzer_missing_columns_from_subquery.sql b/tests/queries/0_stateless/03072_analyzer_missing_columns_from_subquery.sql index e2846033913..ec3b067cbdf 100644 --- a/tests/queries/0_stateless/03072_analyzer_missing_columns_from_subquery.sql +++ b/tests/queries/0_stateless/03072_analyzer_missing_columns_from_subquery.sql @@ -1,3 +1,3 @@ -- https://github.com/ClickHouse/ClickHouse/issues/14699 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; select * from (select number from numbers(1)) where not ignore(*); diff --git a/tests/queries/0_stateless/03073_analyzer_alias_as_column_name.sql b/tests/queries/0_stateless/03073_analyzer_alias_as_column_name.sql index 5599324c62b..bba51e28ba3 100644 --- a/tests/queries/0_stateless/03073_analyzer_alias_as_column_name.sql +++ b/tests/queries/0_stateless/03073_analyzer_alias_as_column_name.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/27068 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE test ( id String, create_time DateTime ) ENGINE = MergeTree ORDER BY id; insert into test values(1,'1970-02-01 00:00:00'); diff --git a/tests/queries/0_stateless/03074_analyzer_alias_column_in_view.sql b/tests/queries/0_stateless/03074_analyzer_alias_column_in_view.sql index 4df5f6f48e6..314b6c0e8d6 100644 --- a/tests/queries/0_stateless/03074_analyzer_alias_column_in_view.sql +++ b/tests/queries/0_stateless/03074_analyzer_alias_column_in_view.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/28687 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; create view alias (dummy int, n alias dummy) as select * from system.one; select n from alias; diff --git a/tests/queries/0_stateless/03075_analyzer_subquery_alias.sql b/tests/queries/0_stateless/03075_analyzer_subquery_alias.sql index 416815e761b..4f097350da6 100644 --- a/tests/queries/0_stateless/03075_analyzer_subquery_alias.sql +++ b/tests/queries/0_stateless/03075_analyzer_subquery_alias.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/28777 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT sum(q0.a2) AS a1, sum(q0.a1) AS a9 diff --git a/tests/queries/0_stateless/03076_analyzer_multiple_joins_alias.sql b/tests/queries/0_stateless/03076_analyzer_multiple_joins_alias.sql index 7ac9fe6b446..894e3bc56ee 100644 --- a/tests/queries/0_stateless/03076_analyzer_multiple_joins_alias.sql +++ b/tests/queries/0_stateless/03076_analyzer_multiple_joins_alias.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/29734 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT * FROM ( @@ -49,4 +49,3 @@ INNER JOIN SELECT number AS x FROM numbers(10) ) AS c ON a.x = c.x; -- { serverError UNKNOWN_IDENTIFIER } - diff --git a/tests/queries/0_stateless/03077_analyzer_multi_scalar_subquery_aliases.sql b/tests/queries/0_stateless/03077_analyzer_multi_scalar_subquery_aliases.sql index 5a181023c57..d4335d35e51 100644 --- a/tests/queries/0_stateless/03077_analyzer_multi_scalar_subquery_aliases.sql +++ b/tests/queries/0_stateless/03077_analyzer_multi_scalar_subquery_aliases.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/33825 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE t1 (i Int64, j Int64) ENGINE = Memory; INSERT INTO t1 SELECT number, number FROM system.numbers LIMIT 10; SELECT diff --git a/tests/queries/0_stateless/03078_analyzer_multi_scalar_subquery_aliases.sql b/tests/queries/0_stateless/03078_analyzer_multi_scalar_subquery_aliases.sql index d91a9ed106d..b9b850619ea 100644 --- a/tests/queries/0_stateless/03078_analyzer_multi_scalar_subquery_aliases.sql +++ b/tests/queries/0_stateless/03078_analyzer_multi_scalar_subquery_aliases.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/33825 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE t2 (first_column Int64, second_column Int64) ENGINE = Memory; INSERT INTO t2 SELECT number, number FROM system.numbers LIMIT 10; diff --git a/tests/queries/0_stateless/03079_analyzer_numeric_literals_as_column_names.sql b/tests/queries/0_stateless/03079_analyzer_numeric_literals_as_column_names.sql index 955d3b49a00..80e681c0776 100644 --- a/tests/queries/0_stateless/03079_analyzer_numeric_literals_as_column_names.sql +++ b/tests/queries/0_stateless/03079_analyzer_numeric_literals_as_column_names.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE testdata (`1` String) ENGINE=MergeTree ORDER BY tuple(); INSERT INTO testdata VALUES ('testdata'); diff --git a/tests/queries/0_stateless/03080_analyzer_prefer_column_name_to_alias__virtual_columns.sql b/tests/queries/0_stateless/03080_analyzer_prefer_column_name_to_alias__virtual_columns.sql index 01ab868f9ea..2138828cd27 100644 --- a/tests/queries/0_stateless/03080_analyzer_prefer_column_name_to_alias__virtual_columns.sql +++ b/tests/queries/0_stateless/03080_analyzer_prefer_column_name_to_alias__virtual_columns.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/35652 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE test ( id UInt64 ) diff --git a/tests/queries/0_stateless/03080_incorrect_join_with_merge.sql b/tests/queries/0_stateless/03080_incorrect_join_with_merge.sql index 7682e6ce866..a34c71a44e2 100644 --- a/tests/queries/0_stateless/03080_incorrect_join_with_merge.sql +++ b/tests/queries/0_stateless/03080_incorrect_join_with_merge.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/29838 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SET distributed_foreground_insert=1; CREATE TABLE first_table_lr diff --git a/tests/queries/0_stateless/03081_analyzer_agg_func_CTE.sql b/tests/queries/0_stateless/03081_analyzer_agg_func_CTE.sql index e6a540dc5df..3cb02512a7f 100644 --- a/tests/queries/0_stateless/03081_analyzer_agg_func_CTE.sql +++ b/tests/queries/0_stateless/03081_analyzer_agg_func_CTE.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/36189 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE test ( `dt` Date, diff --git a/tests/queries/0_stateless/03082_analyzer_left_join_correct_column.sql b/tests/queries/0_stateless/03082_analyzer_left_join_correct_column.sql index 8f17248ed0d..3b83f978326 100644 --- a/tests/queries/0_stateless/03082_analyzer_left_join_correct_column.sql +++ b/tests/queries/0_stateless/03082_analyzer_left_join_correct_column.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/39634 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE test1 ( `pk` String, diff --git a/tests/queries/0_stateless/03084_analyzer_join_column_alias.sql b/tests/queries/0_stateless/03084_analyzer_join_column_alias.sql index 930726898b5..8a7258f5838 100644 --- a/tests/queries/0_stateless/03084_analyzer_join_column_alias.sql +++ b/tests/queries/0_stateless/03084_analyzer_join_column_alias.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/47432 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; create table t1 engine = MergeTree() order by tuple() diff --git a/tests/queries/0_stateless/03085_analyzer_alias_column_group_by.sql b/tests/queries/0_stateless/03085_analyzer_alias_column_group_by.sql index fd67194b08b..c360e86197f 100644 --- a/tests/queries/0_stateless/03085_analyzer_alias_column_group_by.sql +++ b/tests/queries/0_stateless/03085_analyzer_alias_column_group_by.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/54910 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT toTypeName(stat_standard_id) AS stat_standard_id_1, count(1) AS value FROM ( SELECT 'string value' AS stat_standard_id ) GROUP BY stat_standard_id_1 LIMIT 1 diff --git a/tests/queries/0_stateless/03086_analyzer_window_func_part_of_group_by.sql b/tests/queries/0_stateless/03086_analyzer_window_func_part_of_group_by.sql index 31747328d1f..7e44b37f865 100644 --- a/tests/queries/0_stateless/03086_analyzer_window_func_part_of_group_by.sql +++ b/tests/queries/0_stateless/03086_analyzer_window_func_part_of_group_by.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/57321 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT ver, max(ver) OVER () AS ver_max diff --git a/tests/queries/0_stateless/03087_analyzer_subquery_with_alias.sql b/tests/queries/0_stateless/03087_analyzer_subquery_with_alias.sql index 6546e50c99e..a00ca4960d7 100644 --- a/tests/queries/0_stateless/03087_analyzer_subquery_with_alias.sql +++ b/tests/queries/0_stateless/03087_analyzer_subquery_with_alias.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/59154 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT * FROM ( diff --git a/tests/queries/0_stateless/03088_analyzer_ambiguous_column_multi_call.sql b/tests/queries/0_stateless/03088_analyzer_ambiguous_column_multi_call.sql index e6f1ed81f91..3670404d124 100644 --- a/tests/queries/0_stateless/03088_analyzer_ambiguous_column_multi_call.sql +++ b/tests/queries/0_stateless/03088_analyzer_ambiguous_column_multi_call.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/61014 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; DROP DATABASE IF EXISTS {CLICKHOUSE_DATABASE:Identifier}; create database {CLICKHOUSE_DATABASE:Identifier}; diff --git a/tests/queries/0_stateless/03089_analyzer_alias_replacement.sql b/tests/queries/0_stateless/03089_analyzer_alias_replacement.sql index 069da5fdd65..5526e1aaf7d 100644 --- a/tests/queries/0_stateless/03089_analyzer_alias_replacement.sql +++ b/tests/queries/0_stateless/03089_analyzer_alias_replacement.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/61950 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; with dummy + 1 as dummy select dummy from system.one; diff --git a/tests/queries/0_stateless/03090_analyzer_multiple_using_statements.sql b/tests/queries/0_stateless/03090_analyzer_multiple_using_statements.sql index c35f33782ff..08ea103d3c9 100644 --- a/tests/queries/0_stateless/03090_analyzer_multiple_using_statements.sql +++ b/tests/queries/0_stateless/03090_analyzer_multiple_using_statements.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/55647 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT * diff --git a/tests/queries/0_stateless/03091_analyzer_same_table_name_in_different_databases.sql b/tests/queries/0_stateless/03091_analyzer_same_table_name_in_different_databases.sql index 599275c66e8..11984aec496 100644 --- a/tests/queries/0_stateless/03091_analyzer_same_table_name_in_different_databases.sql +++ b/tests/queries/0_stateless/03091_analyzer_same_table_name_in_different_databases.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/61947 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; DROP DATABASE IF EXISTS {CLICKHOUSE_DATABASE:Identifier}; DROP DATABASE IF EXISTS {CLICKHOUSE_DATABASE_1:Identifier}; diff --git a/tests/queries/0_stateless/03092_analyzer_same_table_name_in_different_databases.sql b/tests/queries/0_stateless/03092_analyzer_same_table_name_in_different_databases.sql index 10d18324c3c..83b1a902721 100644 --- a/tests/queries/0_stateless/03092_analyzer_same_table_name_in_different_databases.sql +++ b/tests/queries/0_stateless/03092_analyzer_same_table_name_in_different_databases.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/61947 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; DROP DATABASE IF EXISTS {CLICKHOUSE_DATABASE:Identifier}; DROP DATABASE IF EXISTS {CLICKHOUSE_DATABASE_1:Identifier}; diff --git a/tests/queries/0_stateless/03093_analyzer_column_alias.sql b/tests/queries/0_stateless/03093_analyzer_column_alias.sql index 9ff0f78ba24..edf89108b56 100644 --- a/tests/queries/0_stateless/03093_analyzer_column_alias.sql +++ b/tests/queries/0_stateless/03093_analyzer_column_alias.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/26674 -SET allow_experimental_analyzer = true; +SET enable_analyzer = true; SELECT Carrier, diff --git a/tests/queries/0_stateless/03093_analyzer_miel_test.sql b/tests/queries/0_stateless/03093_analyzer_miel_test.sql index f408882dcd5..4915864bb51 100644 --- a/tests/queries/0_stateless/03093_analyzer_miel_test.sql +++ b/tests/queries/0_stateless/03093_analyzer_miel_test.sql @@ -8,7 +8,7 @@ INSERT INTO test_03093 VALUES ('x1', 123, {'k1': ''}); INSERT INTO test_03093 VALUES ('x1', 123, {'k1': '', 'k11': ''}); INSERT INTO test_03093 VALUES ('x1', 12, {'k1': ''}); -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; select app, arrayZip(untuple(sumMap(k.keys, replicate(1, k.keys)))) from test_03093 PREWHERE c > 1 group by app; select app, arrayZip(untuple(sumMap(k.keys, replicate(1, k.keys)))) from test_03093 WHERE c > 1 group by app; diff --git a/tests/queries/0_stateless/03093_bug37909_query_does_not_finish.sql b/tests/queries/0_stateless/03093_bug37909_query_does_not_finish.sql index 62fa3f437af..90f0a8a3b8e 100644 --- a/tests/queries/0_stateless/03093_bug37909_query_does_not_finish.sql +++ b/tests/queries/0_stateless/03093_bug37909_query_does_not_finish.sql @@ -75,4 +75,4 @@ FROM /* WHERE (v_date >= '2022-05-08') AND (v_date <= '2022-06-07') placing condition has same effect */ GROUP BY vDate ORDER BY vDate ASC -SETTINGS allow_experimental_analyzer = 1; -- the query times out if allow_experimental_analyzer = 0 +SETTINGS enable_analyzer = 1; -- the query times out if enable_analyzer = 0 diff --git a/tests/queries/0_stateless/03094_analyzer_fiddle_multiif.sql b/tests/queries/0_stateless/03094_analyzer_fiddle_multiif.sql index 1b1603be18e..8426749917a 100644 --- a/tests/queries/0_stateless/03094_analyzer_fiddle_multiif.sql +++ b/tests/queries/0_stateless/03094_analyzer_fiddle_multiif.sql @@ -5,7 +5,7 @@ INSERT INTO users_03094 VALUES ('John', 33); INSERT INTO users_03094 VALUES ('Ksenia', 48); INSERT INTO users_03094 VALUES ('Alice', 50); -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT multiIf((age > 30) or (true), '1', '2') AS a, diff --git a/tests/queries/0_stateless/03094_named_tuple_bug24607.sql b/tests/queries/0_stateless/03094_named_tuple_bug24607.sql index e3c97f3fe41..698c339e53e 100644 --- a/tests/queries/0_stateless/03094_named_tuple_bug24607.sql +++ b/tests/queries/0_stateless/03094_named_tuple_bug24607.sql @@ -1,4 +1,4 @@ SELECT JSONExtract('{"a":1, "b":"test"}', 'Tuple(a UInt8, b String)') AS x, x.a -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; diff --git a/tests/queries/0_stateless/03094_one_thousand_joins.sql b/tests/queries/0_stateless/03094_one_thousand_joins.sql index 1f6bd99df7f..6ae4e4d4d3c 100644 --- a/tests/queries/0_stateless/03094_one_thousand_joins.sql +++ b/tests/queries/0_stateless/03094_one_thousand_joins.sql @@ -2,7 +2,7 @@ -- (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 +SET enable_analyzer = 1; -- old analyzer returns TOO_DEEP_SUBQUERIES -- Bug 33446, marked as 'long' because it still runs around 10 sec SELECT * FROM (SELECT 1 AS x) t1 JOIN (SELECT 1 AS x) t2 ON t1.x = t2.x JOIN (SELECT 1 AS x) t3 ON t1.x = t3.x JOIN (SELECT 1 AS x) t4 ON t1.x = t4.x JOIN (SELECT 1 AS x) t5 ON t1.x = t5.x JOIN (SELECT 1 AS x) t6 ON t1.x = t6.x JOIN (SELECT 1 AS x) t7 ON t1.x = t7.x JOIN (SELECT 1 AS x) t8 ON t1.x = t8.x JOIN (SELECT 1 AS x) t9 ON t1.x = t9.x JOIN (SELECT 1 AS x) t10 ON t1.x = t10.x JOIN (SELECT 1 AS x) t11 ON t1.x = t11.x JOIN (SELECT 1 AS x) t12 ON t1.x = t12.x JOIN (SELECT 1 AS x) t13 ON t1.x = t13.x JOIN (SELECT 1 AS x) t14 ON t1.x = t14.x JOIN (SELECT 1 AS x) t15 ON t1.x = t15.x JOIN (SELECT 1 AS x) t16 ON t1.x = t16.x JOIN (SELECT 1 AS x) t17 ON t1.x = t17.x JOIN (SELECT 1 AS x) t18 ON t1.x = t18.x JOIN (SELECT 1 AS x) t19 ON t1.x = t19.x JOIN (SELECT 1 AS x) t20 ON t1.x = t20.x JOIN (SELECT 1 AS x) t21 ON t1.x = t21.x JOIN (SELECT 1 AS x) t22 ON t1.x = t22.x JOIN (SELECT 1 AS x) t23 ON t1.x = t23.x JOIN (SELECT 1 AS x) t24 ON t1.x = t24.x JOIN (SELECT 1 AS x) t25 ON t1.x = t25.x JOIN (SELECT 1 AS x) t26 ON t1.x = t26.x JOIN (SELECT 1 AS x) t27 ON t1.x = t27.x JOIN (SELECT 1 AS x) t28 ON t1.x = t28.x JOIN (SELECT 1 AS x) t29 ON t1.x = t29.x JOIN (SELECT 1 AS x) t30 ON t1.x = t30.x JOIN (SELECT 1 AS x) t31 ON t1.x = t31.x JOIN (SELECT 1 AS x) t32 ON t1.x = t32.x JOIN (SELECT 1 AS x) t33 ON t1.x = t33.x JOIN (SELECT 1 AS x) t34 ON t1.x = t34.x JOIN (SELECT 1 AS x) t35 ON t1.x = t35.x JOIN (SELECT 1 AS x) t36 ON t1.x = t36.x JOIN (SELECT 1 AS x) t37 ON t1.x = t37.x JOIN (SELECT 1 AS x) t38 ON t1.x = t38.x JOIN (SELECT 1 AS x) t39 ON t1.x = t39.x JOIN (SELECT 1 AS x) t40 ON t1.x = t40.x JOIN (SELECT 1 AS x) t41 ON t1.x = t41.x JOIN (SELECT 1 AS x) t42 ON t1.x = t42.x JOIN (SELECT 1 AS x) t43 ON t1.x = t43.x JOIN (SELECT 1 AS x) t44 ON t1.x = t44.x JOIN (SELECT 1 AS x) t45 ON t1.x = t45.x JOIN (SELECT 1 AS x) t46 ON t1.x = t46.x JOIN (SELECT 1 AS x) t47 ON t1.x = t47.x JOIN (SELECT 1 AS x) t48 ON t1.x = t48.x JOIN (SELECT 1 AS x) t49 ON t1.x = t49.x JOIN (SELECT 1 AS x) t50 ON t1.x = t50.x JOIN (SELECT 1 AS x) t51 ON t1.x = t51.x JOIN (SELECT 1 AS x) t52 ON t1.x = t52.x JOIN (SELECT 1 AS x) t53 ON t1.x = t53.x JOIN (SELECT 1 AS x) t54 ON t1.x = t54.x JOIN (SELECT 1 AS x) t55 ON t1.x = t55.x JOIN (SELECT 1 AS x) t56 ON t1.x = t56.x JOIN (SELECT 1 AS x) t57 ON t1.x = t57.x JOIN (SELECT 1 AS x) t58 ON t1.x = t58.x JOIN (SELECT 1 AS x) t59 ON t1.x = t59.x JOIN (SELECT 1 AS x) t60 ON t1.x = t60.x JOIN (SELECT 1 AS x) t61 ON t1.x = t61.x JOIN (SELECT 1 AS x) t62 ON t1.x = t62.x JOIN (SELECT 1 AS x) t63 ON t1.x = t63.x JOIN (SELECT 1 AS x) t64 ON t1.x = t64.x JOIN (SELECT 1 AS x) t65 ON t1.x = t65.x JOIN (SELECT 1 AS x) t66 ON t1.x = t66.x JOIN (SELECT 1 AS x) t67 ON t1.x = t67.x JOIN (SELECT 1 AS x) t68 ON t1.x = t68.x JOIN (SELECT 1 AS x) t69 ON t1.x = t69.x JOIN (SELECT 1 AS x) t70 ON t1.x = t70.x JOIN (SELECT 1 AS x) t71 ON t1.x = t71.x JOIN (SELECT 1 AS x) t72 ON t1.x = t72.x JOIN (SELECT 1 AS x) t73 ON t1.x = t73.x JOIN (SELECT 1 AS x) t74 ON t1.x = t74.x JOIN (SELECT 1 AS x) t75 ON t1.x = t75.x JOIN (SELECT 1 AS x) t76 ON t1.x = t76.x JOIN (SELECT 1 AS x) t77 ON t1.x = t77.x JOIN (SELECT 1 AS x) t78 ON t1.x = t78.x JOIN (SELECT 1 AS x) t79 ON t1.x = t79.x JOIN (SELECT 1 AS x) t80 ON t1.x = t80.x JOIN (SELECT 1 AS x) t81 ON t1.x = t81.x JOIN (SELECT 1 AS x) t82 ON t1.x = t82.x JOIN (SELECT 1 AS x) t83 ON t1.x = t83.x JOIN (SELECT 1 AS x) t84 ON t1.x = t84.x JOIN (SELECT 1 AS x) t85 ON t1.x = t85.x JOIN (SELECT 1 AS x) t86 ON t1.x = t86.x JOIN (SELECT 1 AS x) t87 ON t1.x = t87.x JOIN (SELECT 1 AS x) t88 ON t1.x = t88.x JOIN (SELECT 1 AS x) t89 ON t1.x = t89.x JOIN (SELECT 1 AS x) t90 ON t1.x = t90.x JOIN (SELECT 1 AS x) t91 ON t1.x = t91.x JOIN (SELECT 1 AS x) t92 ON t1.x = t92.x JOIN (SELECT 1 AS x) t93 ON t1.x = t93.x JOIN (SELECT 1 AS x) t94 ON t1.x = t94.x JOIN (SELECT 1 AS x) t95 ON t1.x = t95.x JOIN (SELECT 1 AS x) t96 ON t1.x = t96.x JOIN (SELECT 1 AS x) t97 ON t1.x = t97.x JOIN (SELECT 1 AS x) t98 ON t1.x = t98.x JOIN (SELECT 1 AS x) t99 ON t1.x = t99.x JOIN (SELECT 1 AS x) t100 ON t1.x = t100.x JOIN (SELECT 1 AS x) t101 ON t1.x = t101.x JOIN (SELECT 1 AS x) t102 ON t1.x = t102.x JOIN (SELECT 1 AS x) t103 ON t1.x = t103.x JOIN (SELECT 1 AS x) t104 ON t1.x = t104.x JOIN (SELECT 1 AS x) t105 ON t1.x = t105.x JOIN (SELECT 1 AS x) t106 ON t1.x = t106.x JOIN (SELECT 1 AS x) t107 ON t1.x = t107.x JOIN (SELECT 1 AS x) t108 ON t1.x = t108.x JOIN (SELECT 1 AS x) t109 ON t1.x = t109.x JOIN (SELECT 1 AS x) t110 ON t1.x = t110.x JOIN (SELECT 1 AS x) t111 ON t1.x = t111.x JOIN (SELECT 1 AS x) t112 ON t1.x = t112.x JOIN (SELECT 1 AS x) t113 ON t1.x = t113.x JOIN (SELECT 1 AS x) t114 ON t1.x = t114.x JOIN (SELECT 1 AS x) t115 ON t1.x = t115.x JOIN (SELECT 1 AS x) t116 ON t1.x = t116.x JOIN (SELECT 1 AS x) t117 ON t1.x = t117.x JOIN (SELECT 1 AS x) t118 ON t1.x = t118.x JOIN (SELECT 1 AS x) t119 ON t1.x = t119.x JOIN (SELECT 1 AS x) t120 ON t1.x = t120.x JOIN (SELECT 1 AS x) t121 ON t1.x = t121.x JOIN (SELECT 1 AS x) t122 ON t1.x = t122.x JOIN (SELECT 1 AS x) t123 ON t1.x = t123.x JOIN (SELECT 1 AS x) t124 ON t1.x = t124.x JOIN (SELECT 1 AS x) t125 ON t1.x = t125.x JOIN (SELECT 1 AS x) t126 ON t1.x = t126.x JOIN (SELECT 1 AS x) t127 ON t1.x = t127.x JOIN (SELECT 1 AS x) t128 ON t1.x = t128.x JOIN (SELECT 1 AS x) t129 ON t1.x = t129.x JOIN (SELECT 1 AS x) t130 ON t1.x = t130.x JOIN (SELECT 1 AS x) t131 ON t1.x = t131.x JOIN (SELECT 1 AS x) t132 ON t1.x = t132.x JOIN (SELECT 1 AS x) t133 ON t1.x = t133.x JOIN (SELECT 1 AS x) t134 ON t1.x = t134.x JOIN (SELECT 1 AS x) t135 ON t1.x = t135.x JOIN (SELECT 1 AS x) t136 ON t1.x = t136.x JOIN (SELECT 1 AS x) t137 ON t1.x = t137.x JOIN (SELECT 1 AS x) t138 ON t1.x = t138.x JOIN (SELECT 1 AS x) t139 ON t1.x = t139.x JOIN (SELECT 1 AS x) t140 ON t1.x = t140.x JOIN (SELECT 1 AS x) t141 ON t1.x = t141.x JOIN (SELECT 1 AS x) t142 ON t1.x = t142.x JOIN (SELECT 1 AS x) t143 ON t1.x = t143.x JOIN (SELECT 1 AS x) t144 ON t1.x = t144.x JOIN (SELECT 1 AS x) t145 ON t1.x = t145.x JOIN (SELECT 1 AS x) t146 ON t1.x = t146.x JOIN (SELECT 1 AS x) t147 ON t1.x = t147.x JOIN (SELECT 1 AS x) t148 ON t1.x = t148.x JOIN (SELECT 1 AS x) t149 ON t1.x = t149.x JOIN (SELECT 1 AS x) t150 ON t1.x = t150.x JOIN (SELECT 1 AS x) t151 ON t1.x = t151.x JOIN (SELECT 1 AS x) t152 ON t1.x = t152.x JOIN (SELECT 1 AS x) t153 ON t1.x = t153.x JOIN (SELECT 1 AS x) t154 ON t1.x = t154.x JOIN (SELECT 1 AS x) t155 ON t1.x = t155.x JOIN (SELECT 1 AS x) t156 ON t1.x = t156.x JOIN (SELECT 1 AS x) t157 ON t1.x = t157.x JOIN (SELECT 1 AS x) t158 ON t1.x = t158.x JOIN (SELECT 1 AS x) t159 ON t1.x = t159.x JOIN (SELECT 1 AS x) t160 ON t1.x = t160.x JOIN (SELECT 1 AS x) t161 ON t1.x = t161.x JOIN (SELECT 1 AS x) t162 ON t1.x = t162.x JOIN (SELECT 1 AS x) t163 ON t1.x = t163.x JOIN (SELECT 1 AS x) t164 ON t1.x = t164.x JOIN (SELECT 1 AS x) t165 ON t1.x = t165.x JOIN (SELECT 1 AS x) t166 ON t1.x = t166.x JOIN (SELECT 1 AS x) t167 ON t1.x = t167.x JOIN (SELECT 1 AS x) t168 ON t1.x = t168.x JOIN (SELECT 1 AS x) t169 ON t1.x = t169.x JOIN (SELECT 1 AS x) t170 ON t1.x = t170.x JOIN (SELECT 1 AS x) t171 ON t1.x = t171.x JOIN (SELECT 1 AS x) t172 ON t1.x = t172.x JOIN (SELECT 1 AS x) t173 ON t1.x = t173.x JOIN (SELECT 1 AS x) t174 ON t1.x = t174.x JOIN (SELECT 1 AS x) t175 ON t1.x = t175.x JOIN (SELECT 1 AS x) t176 ON t1.x = t176.x JOIN (SELECT 1 AS x) t177 ON t1.x = t177.x JOIN (SELECT 1 AS x) t178 ON t1.x = t178.x JOIN (SELECT 1 AS x) t179 ON t1.x = t179.x JOIN (SELECT 1 AS x) t180 ON t1.x = t180.x JOIN (SELECT 1 AS x) t181 ON t1.x = t181.x JOIN (SELECT 1 AS x) t182 ON t1.x = t182.x JOIN (SELECT 1 AS x) t183 ON t1.x = t183.x JOIN (SELECT 1 AS x) t184 ON t1.x = t184.x JOIN (SELECT 1 AS x) t185 ON t1.x = t185.x JOIN (SELECT 1 AS x) t186 ON t1.x = t186.x JOIN (SELECT 1 AS x) t187 ON t1.x = t187.x JOIN (SELECT 1 AS x) t188 ON t1.x = t188.x JOIN (SELECT 1 AS x) t189 ON t1.x = t189.x JOIN (SELECT 1 AS x) t190 ON t1.x = t190.x JOIN (SELECT 1 AS x) t191 ON t1.x = t191.x JOIN (SELECT 1 AS x) t192 ON t1.x = t192.x JOIN (SELECT 1 AS x) t193 ON t1.x = t193.x JOIN (SELECT 1 AS x) t194 ON t1.x = t194.x JOIN (SELECT 1 AS x) t195 ON t1.x = t195.x JOIN (SELECT 1 AS x) t196 ON t1.x = t196.x JOIN (SELECT 1 AS x) t197 ON t1.x = t197.x JOIN (SELECT 1 AS x) t198 ON t1.x = t198.x JOIN (SELECT 1 AS x) t199 ON t1.x = t199.x JOIN (SELECT 1 AS x) t200 ON t1.x = t200.x JOIN (SELECT 1 AS x) t201 ON t1.x = t201.x JOIN (SELECT 1 AS x) t202 ON t1.x = t202.x JOIN (SELECT 1 AS x) t203 ON t1.x = t203.x JOIN (SELECT 1 AS x) t204 ON t1.x = t204.x JOIN (SELECT 1 AS x) t205 ON t1.x = t205.x JOIN (SELECT 1 AS x) t206 ON t1.x = t206.x JOIN (SELECT 1 AS x) t207 ON t1.x = t207.x JOIN (SELECT 1 AS x) t208 ON t1.x = t208.x JOIN (SELECT 1 AS x) t209 ON t1.x = t209.x JOIN (SELECT 1 AS x) t210 ON t1.x = t210.x JOIN (SELECT 1 AS x) t211 ON t1.x = t211.x JOIN (SELECT 1 AS x) t212 ON t1.x = t212.x JOIN (SELECT 1 AS x) t213 ON t1.x = t213.x JOIN (SELECT 1 AS x) t214 ON t1.x = t214.x JOIN (SELECT 1 AS x) t215 ON t1.x = t215.x JOIN (SELECT 1 AS x) t216 ON t1.x = t216.x JOIN (SELECT 1 AS x) t217 ON t1.x = t217.x JOIN (SELECT 1 AS x) t218 ON t1.x = t218.x JOIN (SELECT 1 AS x) t219 ON t1.x = t219.x JOIN (SELECT 1 AS x) t220 ON t1.x = t220.x JOIN (SELECT 1 AS x) t221 ON t1.x = t221.x JOIN (SELECT 1 AS x) t222 ON t1.x = t222.x JOIN (SELECT 1 AS x) t223 ON t1.x = t223.x JOIN (SELECT 1 AS x) t224 ON t1.x = t224.x JOIN (SELECT 1 AS x) t225 ON t1.x = t225.x JOIN (SELECT 1 AS x) t226 ON t1.x = t226.x JOIN (SELECT 1 AS x) t227 ON t1.x = t227.x JOIN (SELECT 1 AS x) t228 ON t1.x = t228.x JOIN (SELECT 1 AS x) t229 ON t1.x = t229.x JOIN (SELECT 1 AS x) t230 ON t1.x = t230.x JOIN (SELECT 1 AS x) t231 ON t1.x = t231.x JOIN (SELECT 1 AS x) t232 ON t1.x = t232.x JOIN (SELECT 1 AS x) t233 ON t1.x = t233.x JOIN (SELECT 1 AS x) t234 ON t1.x = t234.x JOIN (SELECT 1 AS x) t235 ON t1.x = t235.x JOIN (SELECT 1 AS x) t236 ON t1.x = t236.x JOIN (SELECT 1 AS x) t237 ON t1.x = t237.x JOIN (SELECT 1 AS x) t238 ON t1.x = t238.x JOIN (SELECT 1 AS x) t239 ON t1.x = t239.x JOIN (SELECT 1 AS x) t240 ON t1.x = t240.x JOIN (SELECT 1 AS x) t241 ON t1.x = t241.x JOIN (SELECT 1 AS x) t242 ON t1.x = t242.x JOIN (SELECT 1 AS x) t243 ON t1.x = t243.x JOIN (SELECT 1 AS x) t244 ON t1.x = t244.x JOIN (SELECT 1 AS x) t245 ON t1.x = t245.x JOIN (SELECT 1 AS x) t246 ON t1.x = t246.x JOIN (SELECT 1 AS x) t247 ON t1.x = t247.x JOIN (SELECT 1 AS x) t248 ON t1.x = t248.x JOIN (SELECT 1 AS x) t249 ON t1.x = t249.x JOIN (SELECT 1 AS x) t250 ON t1.x = t250.x JOIN (SELECT 1 AS x) t251 ON t1.x = t251.x JOIN (SELECT 1 AS x) t252 ON t1.x = t252.x JOIN (SELECT 1 AS x) t253 ON t1.x = t253.x JOIN (SELECT 1 AS x) t254 ON t1.x = t254.x JOIN (SELECT 1 AS x) t255 ON t1.x = t255.x JOIN (SELECT 1 AS x) t256 ON t1.x = t256.x JOIN (SELECT 1 AS x) t257 ON t1.x = t257.x JOIN (SELECT 1 AS x) t258 ON t1.x = t258.x JOIN (SELECT 1 AS x) t259 ON t1.x = t259.x JOIN (SELECT 1 AS x) t260 ON t1.x = t260.x JOIN (SELECT 1 AS x) t261 ON t1.x = t261.x JOIN (SELECT 1 AS x) t262 ON t1.x = t262.x JOIN (SELECT 1 AS x) t263 ON t1.x = t263.x JOIN (SELECT 1 AS x) t264 ON t1.x = t264.x JOIN (SELECT 1 AS x) t265 ON t1.x = t265.x JOIN (SELECT 1 AS x) t266 ON t1.x = t266.x JOIN (SELECT 1 AS x) t267 ON t1.x = t267.x JOIN (SELECT 1 AS x) t268 ON t1.x = t268.x JOIN (SELECT 1 AS x) t269 ON t1.x = t269.x JOIN (SELECT 1 AS x) t270 ON t1.x = t270.x JOIN (SELECT 1 AS x) t271 ON t1.x = t271.x JOIN (SELECT 1 AS x) t272 ON t1.x = t272.x JOIN (SELECT 1 AS x) t273 ON t1.x = t273.x JOIN (SELECT 1 AS x) t274 ON t1.x = t274.x JOIN (SELECT 1 AS x) t275 ON t1.x = t275.x JOIN (SELECT 1 AS x) t276 ON t1.x = t276.x JOIN (SELECT 1 AS x) t277 ON t1.x = t277.x JOIN (SELECT 1 AS x) t278 ON t1.x = t278.x JOIN (SELECT 1 AS x) t279 ON t1.x = t279.x JOIN (SELECT 1 AS x) t280 ON t1.x = t280.x JOIN (SELECT 1 AS x) t281 ON t1.x = t281.x JOIN (SELECT 1 AS x) t282 ON t1.x = t282.x JOIN (SELECT 1 AS x) t283 ON t1.x = t283.x JOIN (SELECT 1 AS x) t284 ON t1.x = t284.x JOIN (SELECT 1 AS x) t285 ON t1.x = t285.x JOIN (SELECT 1 AS x) t286 ON t1.x = t286.x JOIN (SELECT 1 AS x) t287 ON t1.x = t287.x JOIN (SELECT 1 AS x) t288 ON t1.x = t288.x JOIN (SELECT 1 AS x) t289 ON t1.x = t289.x JOIN (SELECT 1 AS x) t290 ON t1.x = t290.x JOIN (SELECT 1 AS x) t291 ON t1.x = t291.x JOIN (SELECT 1 AS x) t292 ON t1.x = t292.x JOIN (SELECT 1 AS x) t293 ON t1.x = t293.x JOIN (SELECT 1 AS x) t294 ON t1.x = t294.x JOIN (SELECT 1 AS x) t295 ON t1.x = t295.x JOIN (SELECT 1 AS x) t296 ON t1.x = t296.x JOIN (SELECT 1 AS x) t297 ON t1.x = t297.x JOIN (SELECT 1 AS x) t298 ON t1.x = t298.x JOIN (SELECT 1 AS x) t299 ON t1.x = t299.x JOIN (SELECT 1 AS x) t300 ON t1.x = t300.x JOIN (SELECT 1 AS x) t301 ON t1.x = t301.x JOIN (SELECT 1 AS x) t302 ON t1.x = t302.x JOIN (SELECT 1 AS x) t303 ON t1.x = t303.x JOIN (SELECT 1 AS x) t304 ON t1.x = t304.x JOIN (SELECT 1 AS x) t305 ON t1.x = t305.x JOIN (SELECT 1 AS x) t306 ON t1.x = t306.x JOIN (SELECT 1 AS x) t307 ON t1.x = t307.x JOIN (SELECT 1 AS x) t308 ON t1.x = t308.x JOIN (SELECT 1 AS x) t309 ON t1.x = t309.x JOIN (SELECT 1 AS x) t310 ON t1.x = t310.x JOIN (SELECT 1 AS x) t311 ON t1.x = t311.x JOIN (SELECT 1 AS x) t312 ON t1.x = t312.x JOIN (SELECT 1 AS x) t313 ON t1.x = t313.x JOIN (SELECT 1 AS x) t314 ON t1.x = t314.x JOIN (SELECT 1 AS x) t315 ON t1.x = t315.x JOIN (SELECT 1 AS x) t316 ON t1.x = t316.x JOIN (SELECT 1 AS x) t317 ON t1.x = t317.x JOIN (SELECT 1 AS x) t318 ON t1.x = t318.x JOIN (SELECT 1 AS x) t319 ON t1.x = t319.x JOIN (SELECT 1 AS x) t320 ON t1.x = t320.x JOIN (SELECT 1 AS x) t321 ON t1.x = t321.x JOIN (SELECT 1 AS x) t322 ON t1.x = t322.x JOIN (SELECT 1 AS x) t323 ON t1.x = t323.x JOIN (SELECT 1 AS x) t324 ON t1.x = t324.x JOIN (SELECT 1 AS x) t325 ON t1.x = t325.x JOIN (SELECT 1 AS x) t326 ON t1.x = t326.x JOIN (SELECT 1 AS x) t327 ON t1.x = t327.x JOIN (SELECT 1 AS x) t328 ON t1.x = t328.x JOIN (SELECT 1 AS x) t329 ON t1.x = t329.x JOIN (SELECT 1 AS x) t330 ON t1.x = t330.x JOIN (SELECT 1 AS x) t331 ON t1.x = t331.x JOIN (SELECT 1 AS x) t332 ON t1.x = t332.x JOIN (SELECT 1 AS x) t333 ON t1.x = t333.x JOIN (SELECT 1 AS x) t334 ON t1.x = t334.x JOIN (SELECT 1 AS x) t335 ON t1.x = t335.x JOIN (SELECT 1 AS x) t336 ON t1.x = t336.x JOIN (SELECT 1 AS x) t337 ON t1.x = t337.x JOIN (SELECT 1 AS x) t338 ON t1.x = t338.x JOIN (SELECT 1 AS x) t339 ON t1.x = t339.x JOIN (SELECT 1 AS x) t340 ON t1.x = t340.x JOIN (SELECT 1 AS x) t341 ON t1.x = t341.x JOIN (SELECT 1 AS x) t342 ON t1.x = t342.x JOIN (SELECT 1 AS x) t343 ON t1.x = t343.x JOIN (SELECT 1 AS x) t344 ON t1.x = t344.x JOIN (SELECT 1 AS x) t345 ON t1.x = t345.x JOIN (SELECT 1 AS x) t346 ON t1.x = t346.x JOIN (SELECT 1 AS x) t347 ON t1.x = t347.x JOIN (SELECT 1 AS x) t348 ON t1.x = t348.x JOIN (SELECT 1 AS x) t349 ON t1.x = t349.x JOIN (SELECT 1 AS x) t350 ON t1.x = t350.x JOIN (SELECT 1 AS x) t351 ON t1.x = t351.x JOIN (SELECT 1 AS x) t352 ON t1.x = t352.x JOIN (SELECT 1 AS x) t353 ON t1.x = t353.x JOIN (SELECT 1 AS x) t354 ON t1.x = t354.x JOIN (SELECT 1 AS x) t355 ON t1.x = t355.x JOIN (SELECT 1 AS x) t356 ON t1.x = t356.x JOIN (SELECT 1 AS x) t357 ON t1.x = t357.x JOIN (SELECT 1 AS x) t358 ON t1.x = t358.x JOIN (SELECT 1 AS x) t359 ON t1.x = t359.x JOIN (SELECT 1 AS x) t360 ON t1.x = t360.x JOIN (SELECT 1 AS x) t361 ON t1.x = t361.x JOIN (SELECT 1 AS x) t362 ON t1.x = t362.x JOIN (SELECT 1 AS x) t363 ON t1.x = t363.x JOIN (SELECT 1 AS x) t364 ON t1.x = t364.x JOIN (SELECT 1 AS x) t365 ON t1.x = t365.x JOIN (SELECT 1 AS x) t366 ON t1.x = t366.x JOIN (SELECT 1 AS x) t367 ON t1.x = t367.x JOIN (SELECT 1 AS x) t368 ON t1.x = t368.x JOIN (SELECT 1 AS x) t369 ON t1.x = t369.x JOIN (SELECT 1 AS x) t370 ON t1.x = t370.x JOIN (SELECT 1 AS x) t371 ON t1.x = t371.x JOIN (SELECT 1 AS x) t372 ON t1.x = t372.x JOIN (SELECT 1 AS x) t373 ON t1.x = t373.x JOIN (SELECT 1 AS x) t374 ON t1.x = t374.x JOIN (SELECT 1 AS x) t375 ON t1.x = t375.x JOIN (SELECT 1 AS x) t376 ON t1.x = t376.x JOIN (SELECT 1 AS x) t377 ON t1.x = t377.x JOIN (SELECT 1 AS x) t378 ON t1.x = t378.x JOIN (SELECT 1 AS x) t379 ON t1.x = t379.x JOIN (SELECT 1 AS x) t380 ON t1.x = t380.x JOIN (SELECT 1 AS x) t381 ON t1.x = t381.x JOIN (SELECT 1 AS x) t382 ON t1.x = t382.x JOIN (SELECT 1 AS x) t383 ON t1.x = t383.x JOIN (SELECT 1 AS x) t384 ON t1.x = t384.x JOIN (SELECT 1 AS x) t385 ON t1.x = t385.x JOIN (SELECT 1 AS x) t386 ON t1.x = t386.x JOIN (SELECT 1 AS x) t387 ON t1.x = t387.x JOIN (SELECT 1 AS x) t388 ON t1.x = t388.x JOIN (SELECT 1 AS x) t389 ON t1.x = t389.x JOIN (SELECT 1 AS x) t390 ON t1.x = t390.x JOIN (SELECT 1 AS x) t391 ON t1.x = t391.x JOIN (SELECT 1 AS x) t392 ON t1.x = t392.x JOIN (SELECT 1 AS x) t393 ON t1.x = t393.x JOIN (SELECT 1 AS x) t394 ON t1.x = t394.x JOIN (SELECT 1 AS x) t395 ON t1.x = t395.x JOIN (SELECT 1 AS x) t396 ON t1.x = t396.x JOIN (SELECT 1 AS x) t397 ON t1.x = t397.x JOIN (SELECT 1 AS x) t398 ON t1.x = t398.x JOIN (SELECT 1 AS x) t399 ON t1.x = t399.x JOIN (SELECT 1 AS x) t400 ON t1.x = t400.x JOIN (SELECT 1 AS x) t401 ON t1.x = t401.x JOIN (SELECT 1 AS x) t402 ON t1.x = t402.x JOIN (SELECT 1 AS x) t403 ON t1.x = t403.x JOIN (SELECT 1 AS x) t404 ON t1.x = t404.x JOIN (SELECT 1 AS x) t405 ON t1.x = t405.x JOIN (SELECT 1 AS x) t406 ON t1.x = t406.x JOIN (SELECT 1 AS x) t407 ON t1.x = t407.x JOIN (SELECT 1 AS x) t408 ON t1.x = t408.x JOIN (SELECT 1 AS x) t409 ON t1.x = t409.x JOIN (SELECT 1 AS x) t410 ON t1.x = t410.x JOIN (SELECT 1 AS x) t411 ON t1.x = t411.x JOIN (SELECT 1 AS x) t412 ON t1.x = t412.x JOIN (SELECT 1 AS x) t413 ON t1.x = t413.x JOIN (SELECT 1 AS x) t414 ON t1.x = t414.x JOIN (SELECT 1 AS x) t415 ON t1.x = t415.x JOIN (SELECT 1 AS x) t416 ON t1.x = t416.x JOIN (SELECT 1 AS x) t417 ON t1.x = t417.x JOIN (SELECT 1 AS x) t418 ON t1.x = t418.x JOIN (SELECT 1 AS x) t419 ON t1.x = t419.x JOIN (SELECT 1 AS x) t420 ON t1.x = t420.x JOIN (SELECT 1 AS x) t421 ON t1.x = t421.x JOIN (SELECT 1 AS x) t422 ON t1.x = t422.x JOIN (SELECT 1 AS x) t423 ON t1.x = t423.x JOIN (SELECT 1 AS x) t424 ON t1.x = t424.x JOIN (SELECT 1 AS x) t425 ON t1.x = t425.x JOIN (SELECT 1 AS x) t426 ON t1.x = t426.x JOIN (SELECT 1 AS x) t427 ON t1.x = t427.x JOIN (SELECT 1 AS x) t428 ON t1.x = t428.x JOIN (SELECT 1 AS x) t429 ON t1.x = t429.x JOIN (SELECT 1 AS x) t430 ON t1.x = t430.x JOIN (SELECT 1 AS x) t431 ON t1.x = t431.x JOIN (SELECT 1 AS x) t432 ON t1.x = t432.x JOIN (SELECT 1 AS x) t433 ON t1.x = t433.x JOIN (SELECT 1 AS x) t434 ON t1.x = t434.x JOIN (SELECT 1 AS x) t435 ON t1.x = t435.x JOIN (SELECT 1 AS x) t436 ON t1.x = t436.x JOIN (SELECT 1 AS x) t437 ON t1.x = t437.x JOIN (SELECT 1 AS x) t438 ON t1.x = t438.x JOIN (SELECT 1 AS x) t439 ON t1.x = t439.x JOIN (SELECT 1 AS x) t440 ON t1.x = t440.x JOIN (SELECT 1 AS x) t441 ON t1.x = t441.x JOIN (SELECT 1 AS x) t442 ON t1.x = t442.x JOIN (SELECT 1 AS x) t443 ON t1.x = t443.x JOIN (SELECT 1 AS x) t444 ON t1.x = t444.x JOIN (SELECT 1 AS x) t445 ON t1.x = t445.x JOIN (SELECT 1 AS x) t446 ON t1.x = t446.x JOIN (SELECT 1 AS x) t447 ON t1.x = t447.x JOIN (SELECT 1 AS x) t448 ON t1.x = t448.x JOIN (SELECT 1 AS x) t449 ON t1.x = t449.x JOIN (SELECT 1 AS x) t450 ON t1.x = t450.x JOIN (SELECT 1 AS x) t451 ON t1.x = t451.x JOIN (SELECT 1 AS x) t452 ON t1.x = t452.x JOIN (SELECT 1 AS x) t453 ON t1.x = t453.x JOIN (SELECT 1 AS x) t454 ON t1.x = t454.x JOIN (SELECT 1 AS x) t455 ON t1.x = t455.x JOIN (SELECT 1 AS x) t456 ON t1.x = t456.x JOIN (SELECT 1 AS x) t457 ON t1.x = t457.x JOIN (SELECT 1 AS x) t458 ON t1.x = t458.x JOIN (SELECT 1 AS x) t459 ON t1.x = t459.x JOIN (SELECT 1 AS x) t460 ON t1.x = t460.x JOIN (SELECT 1 AS x) t461 ON t1.x = t461.x JOIN (SELECT 1 AS x) t462 ON t1.x = t462.x JOIN (SELECT 1 AS x) t463 ON t1.x = t463.x JOIN (SELECT 1 AS x) t464 ON t1.x = t464.x JOIN (SELECT 1 AS x) t465 ON t1.x = t465.x JOIN (SELECT 1 AS x) t466 ON t1.x = t466.x JOIN (SELECT 1 AS x) t467 ON t1.x = t467.x JOIN (SELECT 1 AS x) t468 ON t1.x = t468.x JOIN (SELECT 1 AS x) t469 ON t1.x = t469.x JOIN (SELECT 1 AS x) t470 ON t1.x = t470.x JOIN (SELECT 1 AS x) t471 ON t1.x = t471.x JOIN (SELECT 1 AS x) t472 ON t1.x = t472.x JOIN (SELECT 1 AS x) t473 ON t1.x = t473.x JOIN (SELECT 1 AS x) t474 ON t1.x = t474.x JOIN (SELECT 1 AS x) t475 ON t1.x = t475.x JOIN (SELECT 1 AS x) t476 ON t1.x = t476.x JOIN (SELECT 1 AS x) t477 ON t1.x = t477.x JOIN (SELECT 1 AS x) t478 ON t1.x = t478.x JOIN (SELECT 1 AS x) t479 ON t1.x = t479.x JOIN (SELECT 1 AS x) t480 ON t1.x = t480.x JOIN (SELECT 1 AS x) t481 ON t1.x = t481.x JOIN (SELECT 1 AS x) t482 ON t1.x = t482.x JOIN (SELECT 1 AS x) t483 ON t1.x = t483.x JOIN (SELECT 1 AS x) t484 ON t1.x = t484.x JOIN (SELECT 1 AS x) t485 ON t1.x = t485.x JOIN (SELECT 1 AS x) t486 ON t1.x = t486.x JOIN (SELECT 1 AS x) t487 ON t1.x = t487.x JOIN (SELECT 1 AS x) t488 ON t1.x = t488.x JOIN (SELECT 1 AS x) t489 ON t1.x = t489.x JOIN (SELECT 1 AS x) t490 ON t1.x = t490.x JOIN (SELECT 1 AS x) t491 ON t1.x = t491.x JOIN (SELECT 1 AS x) t492 ON t1.x = t492.x JOIN (SELECT 1 AS x) t493 ON t1.x = t493.x JOIN (SELECT 1 AS x) t494 ON t1.x = t494.x JOIN (SELECT 1 AS x) t495 ON t1.x = t495.x JOIN (SELECT 1 AS x) t496 ON t1.x = t496.x JOIN (SELECT 1 AS x) t497 ON t1.x = t497.x JOIN (SELECT 1 AS x) t498 ON t1.x = t498.x JOIN (SELECT 1 AS x) t499 ON t1.x = t499.x JOIN (SELECT 1 AS x) t500 ON t1.x = t500.x JOIN (SELECT 1 AS x) t501 ON t1.x = t501.x JOIN (SELECT 1 AS x) t502 ON t1.x = t502.x JOIN (SELECT 1 AS x) t503 ON t1.x = t503.x JOIN (SELECT 1 AS x) t504 ON t1.x = t504.x JOIN (SELECT 1 AS x) t505 ON t1.x = t505.x JOIN (SELECT 1 AS x) t506 ON t1.x = t506.x JOIN (SELECT 1 AS x) t507 ON t1.x = t507.x JOIN (SELECT 1 AS x) t508 ON t1.x = t508.x JOIN (SELECT 1 AS x) t509 ON t1.x = t509.x JOIN (SELECT 1 AS x) t510 ON t1.x = t510.x JOIN (SELECT 1 AS x) t511 ON t1.x = t511.x JOIN (SELECT 1 AS x) t512 ON t1.x = t512.x JOIN (SELECT 1 AS x) t513 ON t1.x = t513.x JOIN (SELECT 1 AS x) t514 ON t1.x = t514.x JOIN (SELECT 1 AS x) t515 ON t1.x = t515.x JOIN (SELECT 1 AS x) t516 ON t1.x = t516.x JOIN (SELECT 1 AS x) t517 ON t1.x = t517.x JOIN (SELECT 1 AS x) t518 ON t1.x = t518.x JOIN (SELECT 1 AS x) t519 ON t1.x = t519.x JOIN (SELECT 1 AS x) t520 ON t1.x = t520.x JOIN (SELECT 1 AS x) t521 ON t1.x = t521.x JOIN (SELECT 1 AS x) t522 ON t1.x = t522.x JOIN (SELECT 1 AS x) t523 ON t1.x = t523.x JOIN (SELECT 1 AS x) t524 ON t1.x = t524.x JOIN (SELECT 1 AS x) t525 ON t1.x = t525.x JOIN (SELECT 1 AS x) t526 ON t1.x = t526.x JOIN (SELECT 1 AS x) t527 ON t1.x = t527.x JOIN (SELECT 1 AS x) t528 ON t1.x = t528.x JOIN (SELECT 1 AS x) t529 ON t1.x = t529.x JOIN (SELECT 1 AS x) t530 ON t1.x = t530.x JOIN (SELECT 1 AS x) t531 ON t1.x = t531.x JOIN (SELECT 1 AS x) t532 ON t1.x = t532.x JOIN (SELECT 1 AS x) t533 ON t1.x = t533.x JOIN (SELECT 1 AS x) t534 ON t1.x = t534.x JOIN (SELECT 1 AS x) t535 ON t1.x = t535.x JOIN (SELECT 1 AS x) t536 ON t1.x = t536.x JOIN (SELECT 1 AS x) t537 ON t1.x = t537.x JOIN (SELECT 1 AS x) t538 ON t1.x = t538.x JOIN (SELECT 1 AS x) t539 ON t1.x = t539.x JOIN (SELECT 1 AS x) t540 ON t1.x = t540.x JOIN (SELECT 1 AS x) t541 ON t1.x = t541.x JOIN (SELECT 1 AS x) t542 ON t1.x = t542.x JOIN (SELECT 1 AS x) t543 ON t1.x = t543.x JOIN (SELECT 1 AS x) t544 ON t1.x = t544.x JOIN (SELECT 1 AS x) t545 ON t1.x = t545.x JOIN (SELECT 1 AS x) t546 ON t1.x = t546.x JOIN (SELECT 1 AS x) t547 ON t1.x = t547.x JOIN (SELECT 1 AS x) t548 ON t1.x = t548.x JOIN (SELECT 1 AS x) t549 ON t1.x = t549.x JOIN (SELECT 1 AS x) t550 ON t1.x = t550.x JOIN (SELECT 1 AS x) t551 ON t1.x = t551.x JOIN (SELECT 1 AS x) t552 ON t1.x = t552.x JOIN (SELECT 1 AS x) t553 ON t1.x = t553.x JOIN (SELECT 1 AS x) t554 ON t1.x = t554.x JOIN (SELECT 1 AS x) t555 ON t1.x = t555.x JOIN (SELECT 1 AS x) t556 ON t1.x = t556.x JOIN (SELECT 1 AS x) t557 ON t1.x = t557.x JOIN (SELECT 1 AS x) t558 ON t1.x = t558.x JOIN (SELECT 1 AS x) t559 ON t1.x = t559.x JOIN (SELECT 1 AS x) t560 ON t1.x = t560.x JOIN (SELECT 1 AS x) t561 ON t1.x = t561.x JOIN (SELECT 1 AS x) t562 ON t1.x = t562.x JOIN (SELECT 1 AS x) t563 ON t1.x = t563.x JOIN (SELECT 1 AS x) t564 ON t1.x = t564.x JOIN (SELECT 1 AS x) t565 ON t1.x = t565.x JOIN (SELECT 1 AS x) t566 ON t1.x = t566.x JOIN (SELECT 1 AS x) t567 ON t1.x = t567.x JOIN (SELECT 1 AS x) t568 ON t1.x = t568.x JOIN (SELECT 1 AS x) t569 ON t1.x = t569.x JOIN (SELECT 1 AS x) t570 ON t1.x = t570.x JOIN (SELECT 1 AS x) t571 ON t1.x = t571.x JOIN (SELECT 1 AS x) t572 ON t1.x = t572.x JOIN (SELECT 1 AS x) t573 ON t1.x = t573.x JOIN (SELECT 1 AS x) t574 ON t1.x = t574.x JOIN (SELECT 1 AS x) t575 ON t1.x = t575.x JOIN (SELECT 1 AS x) t576 ON t1.x = t576.x JOIN (SELECT 1 AS x) t577 ON t1.x = t577.x JOIN (SELECT 1 AS x) t578 ON t1.x = t578.x JOIN (SELECT 1 AS x) t579 ON t1.x = t579.x JOIN (SELECT 1 AS x) t580 ON t1.x = t580.x JOIN (SELECT 1 AS x) t581 ON t1.x = t581.x JOIN (SELECT 1 AS x) t582 ON t1.x = t582.x JOIN (SELECT 1 AS x) t583 ON t1.x = t583.x JOIN (SELECT 1 AS x) t584 ON t1.x = t584.x JOIN (SELECT 1 AS x) t585 ON t1.x = t585.x JOIN (SELECT 1 AS x) t586 ON t1.x = t586.x JOIN (SELECT 1 AS x) t587 ON t1.x = t587.x JOIN (SELECT 1 AS x) t588 ON t1.x = t588.x JOIN (SELECT 1 AS x) t589 ON t1.x = t589.x JOIN (SELECT 1 AS x) t590 ON t1.x = t590.x JOIN (SELECT 1 AS x) t591 ON t1.x = t591.x JOIN (SELECT 1 AS x) t592 ON t1.x = t592.x JOIN (SELECT 1 AS x) t593 ON t1.x = t593.x JOIN (SELECT 1 AS x) t594 ON t1.x = t594.x JOIN (SELECT 1 AS x) t595 ON t1.x = t595.x JOIN (SELECT 1 AS x) t596 ON t1.x = t596.x JOIN (SELECT 1 AS x) t597 ON t1.x = t597.x JOIN (SELECT 1 AS x) t598 ON t1.x = t598.x JOIN (SELECT 1 AS x) t599 ON t1.x = t599.x JOIN (SELECT 1 AS x) t600 ON t1.x = t600.x JOIN (SELECT 1 AS x) t601 ON t1.x = t601.x JOIN (SELECT 1 AS x) t602 ON t1.x = t602.x JOIN (SELECT 1 AS x) t603 ON t1.x = t603.x JOIN (SELECT 1 AS x) t604 ON t1.x = t604.x JOIN (SELECT 1 AS x) t605 ON t1.x = t605.x JOIN (SELECT 1 AS x) t606 ON t1.x = t606.x JOIN (SELECT 1 AS x) t607 ON t1.x = t607.x JOIN (SELECT 1 AS x) t608 ON t1.x = t608.x JOIN (SELECT 1 AS x) t609 ON t1.x = t609.x JOIN (SELECT 1 AS x) t610 ON t1.x = t610.x JOIN (SELECT 1 AS x) t611 ON t1.x = t611.x JOIN (SELECT 1 AS x) t612 ON t1.x = t612.x JOIN (SELECT 1 AS x) t613 ON t1.x = t613.x JOIN (SELECT 1 AS x) t614 ON t1.x = t614.x JOIN (SELECT 1 AS x) t615 ON t1.x = t615.x JOIN (SELECT 1 AS x) t616 ON t1.x = t616.x JOIN (SELECT 1 AS x) t617 ON t1.x = t617.x JOIN (SELECT 1 AS x) t618 ON t1.x = t618.x JOIN (SELECT 1 AS x) t619 ON t1.x = t619.x JOIN (SELECT 1 AS x) t620 ON t1.x = t620.x JOIN (SELECT 1 AS x) t621 ON t1.x = t621.x JOIN (SELECT 1 AS x) t622 ON t1.x = t622.x JOIN (SELECT 1 AS x) t623 ON t1.x = t623.x JOIN (SELECT 1 AS x) t624 ON t1.x = t624.x JOIN (SELECT 1 AS x) t625 ON t1.x = t625.x JOIN (SELECT 1 AS x) t626 ON t1.x = t626.x JOIN (SELECT 1 AS x) t627 ON t1.x = t627.x JOIN (SELECT 1 AS x) t628 ON t1.x = t628.x JOIN (SELECT 1 AS x) t629 ON t1.x = t629.x JOIN (SELECT 1 AS x) t630 ON t1.x = t630.x JOIN (SELECT 1 AS x) t631 ON t1.x = t631.x JOIN (SELECT 1 AS x) t632 ON t1.x = t632.x JOIN (SELECT 1 AS x) t633 ON t1.x = t633.x JOIN (SELECT 1 AS x) t634 ON t1.x = t634.x JOIN (SELECT 1 AS x) t635 ON t1.x = t635.x JOIN (SELECT 1 AS x) t636 ON t1.x = t636.x JOIN (SELECT 1 AS x) t637 ON t1.x = t637.x JOIN (SELECT 1 AS x) t638 ON t1.x = t638.x JOIN (SELECT 1 AS x) t639 ON t1.x = t639.x JOIN (SELECT 1 AS x) t640 ON t1.x = t640.x JOIN (SELECT 1 AS x) t641 ON t1.x = t641.x JOIN (SELECT 1 AS x) t642 ON t1.x = t642.x JOIN (SELECT 1 AS x) t643 ON t1.x = t643.x JOIN (SELECT 1 AS x) t644 ON t1.x = t644.x JOIN (SELECT 1 AS x) t645 ON t1.x = t645.x JOIN (SELECT 1 AS x) t646 ON t1.x = t646.x JOIN (SELECT 1 AS x) t647 ON t1.x = t647.x JOIN (SELECT 1 AS x) t648 ON t1.x = t648.x JOIN (SELECT 1 AS x) t649 ON t1.x = t649.x JOIN (SELECT 1 AS x) t650 ON t1.x = t650.x JOIN (SELECT 1 AS x) t651 ON t1.x = t651.x JOIN (SELECT 1 AS x) t652 ON t1.x = t652.x JOIN (SELECT 1 AS x) t653 ON t1.x = t653.x JOIN (SELECT 1 AS x) t654 ON t1.x = t654.x JOIN (SELECT 1 AS x) t655 ON t1.x = t655.x JOIN (SELECT 1 AS x) t656 ON t1.x = t656.x JOIN (SELECT 1 AS x) t657 ON t1.x = t657.x JOIN (SELECT 1 AS x) t658 ON t1.x = t658.x JOIN (SELECT 1 AS x) t659 ON t1.x = t659.x JOIN (SELECT 1 AS x) t660 ON t1.x = t660.x JOIN (SELECT 1 AS x) t661 ON t1.x = t661.x JOIN (SELECT 1 AS x) t662 ON t1.x = t662.x JOIN (SELECT 1 AS x) t663 ON t1.x = t663.x JOIN (SELECT 1 AS x) t664 ON t1.x = t664.x JOIN (SELECT 1 AS x) t665 ON t1.x = t665.x JOIN (SELECT 1 AS x) t666 ON t1.x = t666.x diff --git a/tests/queries/0_stateless/03095_window_functions_qualify.sql b/tests/queries/0_stateless/03095_window_functions_qualify.sql index adedff2e2cf..72903992745 100644 --- a/tests/queries/0_stateless/03095_window_functions_qualify.sql +++ b/tests/queries/0_stateless/03095_window_functions_qualify.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT number, COUNT() OVER (PARTITION BY number % 3) AS partition_count FROM numbers(10) QUALIFY partition_count = 4 ORDER BY number; diff --git a/tests/queries/0_stateless/03096_http_interface_role_query_param.sh b/tests/queries/0_stateless/03096_http_interface_role_query_param.sh index 5ae2d5f2f18..db42ee6bb41 100755 --- a/tests/queries/0_stateless/03096_http_interface_role_query_param.sh +++ b/tests/queries/0_stateless/03096_http_interface_role_query_param.sh @@ -21,7 +21,7 @@ CHANGED_SETTING_VALUE="42" SHOW_CURRENT_ROLES_QUERY="SELECT role_name FROM system.current_roles ORDER BY role_name ASC" SHOW_CHANGED_SETTINGS_QUERY="SELECT name, value FROM system.settings WHERE changed = 1 AND name = '$CHANGED_SETTING_NAME' ORDER BY name ASC" -$CLICKHOUSE_CLIENT -n --query " +$CLICKHOUSE_CLIENT --query " DROP USER IF EXISTS $TEST_USER; DROP ROLE IF EXISTS $TEST_ROLE1; DROP ROLE IF EXISTS $TEST_ROLE2; @@ -94,7 +94,7 @@ OUT=$($CLICKHOUSE_CURL -u $TEST_USER_AUTH -sS "$CLICKHOUSE_URL&role=$TEST_ROLE1& echo -ne $OUT | grep -o "Code: 512" || echo "expected code 512, got: $OUT" echo -ne $OUT | grep -o "SET_NON_GRANTED_ROLE" || echo "expected SET_NON_GRANTED_ROLE error, got: $OUT" -$CLICKHOUSE_CLIENT -n --query " +$CLICKHOUSE_CLIENT --query " DROP USER $TEST_USER; DROP ROLE $TEST_ROLE1; DROP ROLE $TEST_ROLE2; diff --git a/tests/queries/0_stateless/03096_text_log_format_string_args_not_empty.sql b/tests/queries/0_stateless/03096_text_log_format_string_args_not_empty.sql index b1ddd141e04..0e60ee77fe5 100644 --- a/tests/queries/0_stateless/03096_text_log_format_string_args_not_empty.sql +++ b/tests/queries/0_stateless/03096_text_log_format_string_args_not_empty.sql @@ -1,4 +1,4 @@ -set allow_experimental_analyzer = true; +set enable_analyzer = true; select count; -- { serverError UNKNOWN_IDENTIFIER } diff --git a/tests/queries/0_stateless/03097_query_log_join_processes.sql b/tests/queries/0_stateless/03097_query_log_join_processes.sql index daf3136e3fe..135160f4709 100644 --- a/tests/queries/0_stateless/03097_query_log_join_processes.sql +++ b/tests/queries/0_stateless/03097_query_log_join_processes.sql @@ -2,6 +2,6 @@ SYSTEM FLUSH LOGS; -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT count(1) as num, hostName() as hostName FROM system.query_log as a INNER JOIN system.processes as b on a.query_id = b.query_id and type = 'QueryStart' and dateDiff('second', event_time, now()) > 5 and current_database = currentDatabase() FORMAT Null; diff --git a/tests/queries/0_stateless/03098_prefer_column_to_alias_subquery.sql b/tests/queries/0_stateless/03098_prefer_column_to_alias_subquery.sql index cf3768e2f29..cb41151b9c9 100644 --- a/tests/queries/0_stateless/03098_prefer_column_to_alias_subquery.sql +++ b/tests/queries/0_stateless/03098_prefer_column_to_alias_subquery.sql @@ -13,7 +13,7 @@ CREATE TABLE clickhouse_alias_issue_2 ( column_2 Nullable(Float32) ) Engine=Memory; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; INSERT INTO `clickhouse_alias_issue_1` VALUES (1, 100), (2, 200), (3, 300); @@ -28,7 +28,7 @@ VALUES (1, 10), (2, 20), (3, 30); -- \N 30 3 -- \N 20 2 -- \N 10 1 -SELECT * +SELECT * FROM ( SELECT diff --git a/tests/queries/0_stateless/03099_analyzer_multi_join.sql b/tests/queries/0_stateless/03099_analyzer_multi_join.sql index 67985962ba8..dfdaeeea770 100644 --- a/tests/queries/0_stateless/03099_analyzer_multi_join.sql +++ b/tests/queries/0_stateless/03099_analyzer_multi_join.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/56503 -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT tb1.owner_id AS owner_id, diff --git a/tests/queries/0_stateless/03100_analyzer_constants_in_multiif.sql b/tests/queries/0_stateless/03100_analyzer_constants_in_multiif.sql index 04666411760..c9a4000d25e 100644 --- a/tests/queries/0_stateless/03100_analyzer_constants_in_multiif.sql +++ b/tests/queries/0_stateless/03100_analyzer_constants_in_multiif.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/59101 -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; CREATE TABLE users (name String, age Int16) ENGINE=Memory; INSERT INTO users VALUES ('John', 33); diff --git a/tests/queries/0_stateless/03101_analyzer_identifiers_1.sql b/tests/queries/0_stateless/03101_analyzer_identifiers_1.sql index 2e0ad7d8a5b..499f712e57a 100644 --- a/tests/queries/0_stateless/03101_analyzer_identifiers_1.sql +++ b/tests/queries/0_stateless/03101_analyzer_identifiers_1.sql @@ -1,6 +1,6 @@ -- https://github.com/ClickHouse/ClickHouse/issues/23194 -- This test add query-templates for fuzzer -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP DATABASE IF EXISTS {CLICKHOUSE_DATABASE:Identifier}; CREATE DATABASE {CLICKHOUSE_DATABASE:Identifier}; diff --git a/tests/queries/0_stateless/03101_analyzer_identifiers_2.sql b/tests/queries/0_stateless/03101_analyzer_identifiers_2.sql index 35f34e33689..92c3e98265f 100644 --- a/tests/queries/0_stateless/03101_analyzer_identifiers_2.sql +++ b/tests/queries/0_stateless/03101_analyzer_identifiers_2.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/23194 -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; CREATE TEMPORARY TABLE test1 (a String, nest Nested(x String, y String)); diff --git a/tests/queries/0_stateless/03101_analyzer_identifiers_3.sql b/tests/queries/0_stateless/03101_analyzer_identifiers_3.sql index 77a0f040e88..997fee91930 100644 --- a/tests/queries/0_stateless/03101_analyzer_identifiers_3.sql +++ b/tests/queries/0_stateless/03101_analyzer_identifiers_3.sql @@ -1,7 +1,7 @@ -- Tags: no-parallel -- Looks like you cannot use the query parameter as a column name. -- https://github.com/ClickHouse/ClickHouse/issues/23194 -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP DATABASE IF EXISTS db1_03101; DROP DATABASE IF EXISTS db2_03101; diff --git a/tests/queries/0_stateless/03101_analyzer_identifiers_4.sql b/tests/queries/0_stateless/03101_analyzer_identifiers_4.sql index eba6ad09a3f..869310aa181 100644 --- a/tests/queries/0_stateless/03101_analyzer_identifiers_4.sql +++ b/tests/queries/0_stateless/03101_analyzer_identifiers_4.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/23194 -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP DATABASE IF EXISTS {CLICKHOUSE_DATABASE:Identifier}; CREATE DATABASE {CLICKHOUSE_DATABASE:Identifier}; diff --git a/tests/queries/0_stateless/03101_analyzer_invalid_join_on.sql b/tests/queries/0_stateless/03101_analyzer_invalid_join_on.sql index 8539376e136..d7e26862d36 100644 --- a/tests/queries/0_stateless/03101_analyzer_invalid_join_on.sql +++ b/tests/queries/0_stateless/03101_analyzer_invalid_join_on.sql @@ -3,23 +3,23 @@ drop table if exists t1; drop table if exists t2; -set allow_experimental_analyzer=1; +set enable_analyzer=1; create table t1 (c3 String, primary key(c3)) engine = MergeTree; create table t2 (c11 String, primary key(c11)) engine = MergeTree; insert into t1 values ('succeed'); insert into t2 values ('succeed'); -select +select ref_0.c11 as c_2_c30_0 - from + from t2 as ref_0 - cross join (select - ref_1.c3 as c_6_c28_15 - from + cross join (select + ref_1.c3 as c_6_c28_15 + from t1 as ref_1 ) as subq_0 where subq_0.c_6_c28_15 = (select c11 from t2 order by c11 limit 1); drop table if exists t1; -drop table if exists t2; \ No newline at end of file +drop table if exists t2; diff --git a/tests/queries/0_stateless/03102_prefer_column_name_to_alias.sql b/tests/queries/0_stateless/03102_prefer_column_name_to_alias.sql index 48e97fd0841..fafdb660e44 100644 --- a/tests/queries/0_stateless/03102_prefer_column_name_to_alias.sql +++ b/tests/queries/0_stateless/03102_prefer_column_name_to_alias.sql @@ -4,7 +4,7 @@ DROP TABLE IF EXISTS loans; CREATE TABLE loans (loan_number int, security_id text) ENGINE=Memory; -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; INSERT INTO loans VALUES (1, 'AAA'); INSERT INTO loans VALUES (1, 'AAA'); diff --git a/tests/queries/0_stateless/03103_positional_arguments.sql b/tests/queries/0_stateless/03103_positional_arguments.sql index ad30719dc3e..eecaa3f4ea6 100644 --- a/tests/queries/0_stateless/03103_positional_arguments.sql +++ b/tests/queries/0_stateless/03103_positional_arguments.sql @@ -1,6 +1,6 @@ -- https://github.com/ClickHouse/ClickHouse/issues/56466 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; DROP TABLE IF EXISTS users; diff --git a/tests/queries/0_stateless/03104_create_view_join.sql b/tests/queries/0_stateless/03104_create_view_join.sql index bed3d81c9a8..a39be92b1f1 100644 --- a/tests/queries/0_stateless/03104_create_view_join.sql +++ b/tests/queries/0_stateless/03104_create_view_join.sql @@ -4,7 +4,7 @@ DROP TABLE IF EXISTS test_table_01; DROP TABLE IF EXISTS test_table_02; DROP TABLE IF EXISTS test_view_01; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; CREATE TABLE test_table_01 ( column Int32 @@ -15,12 +15,12 @@ CREATE TABLE test_table_02 ( ) ENGINE = Memory(); CREATE VIEW test_view_01 AS -SELECT +SELECT t1.column, t2.column -FROM test_table_01 AS t1 +FROM test_table_01 AS t1 INNER JOIN test_table_02 AS t2 ON t1.column = t2.column; DROP TABLE IF EXISTS test_table_01; DROP TABLE IF EXISTS test_table_02; -DROP TABLE IF EXISTS test_view_01; \ No newline at end of file +DROP TABLE IF EXISTS test_view_01; diff --git a/tests/queries/0_stateless/03107_ill_formed_select_in_materialized_view.sql b/tests/queries/0_stateless/03107_ill_formed_select_in_materialized_view.sql index d142bf1662e..5f6ec74bdeb 100644 --- a/tests/queries/0_stateless/03107_ill_formed_select_in_materialized_view.sql +++ b/tests/queries/0_stateless/03107_ill_formed_select_in_materialized_view.sql @@ -6,10 +6,10 @@ DROP TABLE iF EXISTS b; CREATE TABLE a ( a UInt64, b UInt64) ENGINE = Memory; CREATE TABLE b ( b UInt64) ENGINE = Memory; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET joined_subquery_requires_alias = 0; CREATE MATERIALIZED VIEW view_4 ( bb UInt64, cnt UInt64) Engine=MergeTree ORDER BY bb POPULATE AS SELECT bb, count() AS cnt FROM (SELECT a, b AS j, b AS bb FROM a INNER JOIN (SELECT b AS j, b AS bb FROM b ) USING (j)) GROUP BY bb; -- { serverError UNKNOWN_IDENTIFIER } DROP TABLE IF EXISTS a; -DROP TABLE iF EXISTS b; \ No newline at end of file +DROP TABLE iF EXISTS b; diff --git a/tests/queries/0_stateless/03108_describe_union_all.sql b/tests/queries/0_stateless/03108_describe_union_all.sql index c5172902a2b..7e207ae2854 100644 --- a/tests/queries/0_stateless/03108_describe_union_all.sql +++ b/tests/queries/0_stateless/03108_describe_union_all.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/8030 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; -DESCRIBE (SELECT 1, 1 UNION ALL SELECT 1, 2); \ No newline at end of file +DESCRIBE (SELECT 1, 1 UNION ALL SELECT 1, 2); diff --git a/tests/queries/0_stateless/03109_ast_too_big.sql b/tests/queries/0_stateless/03109_ast_too_big.sql index 3eedf305f44..1464f90fe83 100644 --- a/tests/queries/0_stateless/03109_ast_too_big.sql +++ b/tests/queries/0_stateless/03109_ast_too_big.sql @@ -1,6 +1,6 @@ -- https://github.com/ClickHouse/ClickHouse/issues/32139 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; WITH data AS ( @@ -50,4 +50,4 @@ SELECT value12 AS v12, value13 AS v13, value14 AS v14 -FORMAT Null; \ No newline at end of file +FORMAT Null; diff --git a/tests/queries/0_stateless/03110_unicode_alias.sql b/tests/queries/0_stateless/03110_unicode_alias.sql index b8cbe7390fa..aa33195ea51 100644 --- a/tests/queries/0_stateless/03110_unicode_alias.sql +++ b/tests/queries/0_stateless/03110_unicode_alias.sql @@ -1,6 +1,6 @@ -- https://github.com/ClickHouse/ClickHouse/issues/47288 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; select 1 as `c0` from ( diff --git a/tests/queries/0_stateless/03111_inner_join_group_by.sql b/tests/queries/0_stateless/03111_inner_join_group_by.sql index fcc8c55a566..6ebaacfc3fe 100644 --- a/tests/queries/0_stateless/03111_inner_join_group_by.sql +++ b/tests/queries/0_stateless/03111_inner_join_group_by.sql @@ -1,6 +1,6 @@ -- https://github.com/ClickHouse/ClickHouse/issues/50705 -set allow_experimental_analyzer=1; +set enable_analyzer=1; SELECT count(s0.number), @@ -15,4 +15,4 @@ INNER JOIN LIMIT 10 ) AS s1 ON s0.number = s1.number GROUP BY s0.number > 5 -LIMIT 10 -- {serverError NOT_AN_AGGREGATE} \ No newline at end of file +LIMIT 10 -- {serverError NOT_AN_AGGREGATE} diff --git a/tests/queries/0_stateless/03112_analyzer_not_found_column_in_block.sql b/tests/queries/0_stateless/03112_analyzer_not_found_column_in_block.sql index eb07ff1d837..cc734b2c49a 100644 --- a/tests/queries/0_stateless/03112_analyzer_not_found_column_in_block.sql +++ b/tests/queries/0_stateless/03112_analyzer_not_found_column_in_block.sql @@ -15,7 +15,7 @@ PRIMARY KEY (user_id, timestamp); INSERT INTO my_first_table (user_id, message, timestamp, metric) VALUES (101, 'Hello, ClickHouse!', now(), -1.0 ), (102, 'Insert a lot of rows per batch', yesterday(), 1.41421 ), (102, 'Sort your data based on your commonly-used queries', today(), 2.718 ), (101, 'Granules are the smallest chunks of data read', now() + 5, 3.14159 ); -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT user_id @@ -24,4 +24,4 @@ FROM my_first_table WHERE timestamp > 0 and user_id IN (101) LIMIT 2 BY user_id; -DROP TABLE IF EXISTS my_first_table; \ No newline at end of file +DROP TABLE IF EXISTS my_first_table; diff --git a/tests/queries/0_stateless/03113_analyzer_not_found_column_in_block_2.sql b/tests/queries/0_stateless/03113_analyzer_not_found_column_in_block_2.sql index 7bcc6565ea0..4389bdf83fa 100644 --- a/tests/queries/0_stateless/03113_analyzer_not_found_column_in_block_2.sql +++ b/tests/queries/0_stateless/03113_analyzer_not_found_column_in_block_2.sql @@ -4,11 +4,11 @@ drop table if exists t; create table t (ID String) Engine= Memory() ; insert into t values('a'),('b'),('c'); - + -- This optimization is disabled by default and even its description says that it could lead to -- inconsistencies for distributed queries. set optimize_if_transform_strings_to_enum=0; -set allow_experimental_analyzer=1; +set enable_analyzer=1; SELECT multiIf( ((multiIf(ID='' AND (ID = 'a' OR ID = 'c' OR ID = 'b'),'a','x') as y) = 'c') OR (multiIf(ID='' AND (ID = 'a' OR ID = 'c' OR ID = 'b'),'a','x') = 'b') OR @@ -18,4 +18,4 @@ SELECT multiIf( ((multiIf(ID='' AND (ID = 'a' OR ID = 'c' OR ID = 'b'),'a','x') FROM remote('127.0.0.{1,2}', currentDatabase(), t) GROUP BY alias; -drop table if exists t; \ No newline at end of file +drop table if exists t; diff --git a/tests/queries/0_stateless/03114_analyzer_cte_with_join.sql b/tests/queries/0_stateless/03114_analyzer_cte_with_join.sql index 140197c7104..65dd3cb66b6 100644 --- a/tests/queries/0_stateless/03114_analyzer_cte_with_join.sql +++ b/tests/queries/0_stateless/03114_analyzer_cte_with_join.sql @@ -1,7 +1,7 @@ -- Tags: no-replicated-database -- https://github.com/ClickHouse/ClickHouse/issues/58500 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; drop table if exists t; diff --git a/tests/queries/0_stateless/03115_alias_exists_column.sql b/tests/queries/0_stateless/03115_alias_exists_column.sql index 654cdd71175..65fc0d8f9a9 100644 --- a/tests/queries/0_stateless/03115_alias_exists_column.sql +++ b/tests/queries/0_stateless/03115_alias_exists_column.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/44412 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT EXISTS(SELECT 1) AS mycheck FORMAT TSVWithNames; diff --git a/tests/queries/0_stateless/03116_analyzer_explicit_alias_as_column_name.sql b/tests/queries/0_stateless/03116_analyzer_explicit_alias_as_column_name.sql index 16035b8e72a..d3e3a29077a 100644 --- a/tests/queries/0_stateless/03116_analyzer_explicit_alias_as_column_name.sql +++ b/tests/queries/0_stateless/03116_analyzer_explicit_alias_as_column_name.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/39923 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT errors.name AS labels, diff --git a/tests/queries/0_stateless/03117_analyzer_same_column_name_as_func.sql b/tests/queries/0_stateless/03117_analyzer_same_column_name_as_func.sql index 43f428a1bf9..a3f4da89525 100644 --- a/tests/queries/0_stateless/03117_analyzer_same_column_name_as_func.sql +++ b/tests/queries/0_stateless/03117_analyzer_same_column_name_as_func.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/39855 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; create table x( a UInt64, diff --git a/tests/queries/0_stateless/03118_analyzer_multi_join_prewhere.sql b/tests/queries/0_stateless/03118_analyzer_multi_join_prewhere.sql index 8680e9215c3..84f89c2c647 100644 --- a/tests/queries/0_stateless/03118_analyzer_multi_join_prewhere.sql +++ b/tests/queries/0_stateless/03118_analyzer_multi_join_prewhere.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/4596 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE a1 ( ANIMAL Nullable(String) ) engine = MergeTree order by tuple(); insert into a1 values('CROCO'); diff --git a/tests/queries/0_stateless/03119_analyzer_window_function_in_CTE_alias.sql b/tests/queries/0_stateless/03119_analyzer_window_function_in_CTE_alias.sql index 2eb2c66b551..edbb324bda2 100644 --- a/tests/queries/0_stateless/03119_analyzer_window_function_in_CTE_alias.sql +++ b/tests/queries/0_stateless/03119_analyzer_window_function_in_CTE_alias.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/47422 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; DROP TEMPORARY TABLE IF EXISTS test; CREATE TEMPORARY TABLE test (a Float32, id UInt64); diff --git a/tests/queries/0_stateless/03120_analyzer_dist_join.sql b/tests/queries/0_stateless/03120_analyzer_dist_join.sql index 624da39c69b..e40df56c5ac 100644 --- a/tests/queries/0_stateless/03120_analyzer_dist_join.sql +++ b/tests/queries/0_stateless/03120_analyzer_dist_join.sql @@ -1,6 +1,6 @@ -- Tags: no-replicated-database -- https://github.com/ClickHouse/ClickHouse/issues/8547 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SET distributed_foreground_insert=1; CREATE TABLE a1_replicated ON CLUSTER test_shard_localhost ( diff --git a/tests/queries/0_stateless/03120_analyzer_param_in_CTE_alias.sql b/tests/queries/0_stateless/03120_analyzer_param_in_CTE_alias.sql index 16fffae4737..d1e3d5a2ffa 100644 --- a/tests/queries/0_stateless/03120_analyzer_param_in_CTE_alias.sql +++ b/tests/queries/0_stateless/03120_analyzer_param_in_CTE_alias.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/33000 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SET param_test_a=30; diff --git a/tests/queries/0_stateless/03121_analyzer_filed_redefenition_in_subquery.sql b/tests/queries/0_stateless/03121_analyzer_filed_redefenition_in_subquery.sql index fe05259a320..891eb7aac1c 100644 --- a/tests/queries/0_stateless/03121_analyzer_filed_redefenition_in_subquery.sql +++ b/tests/queries/0_stateless/03121_analyzer_filed_redefenition_in_subquery.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/14739 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; drop table if exists test_subquery; diff --git a/tests/queries/0_stateless/03122_analyzer_collate_in_window_function.sql b/tests/queries/0_stateless/03122_analyzer_collate_in_window_function.sql index 707c5d889f7..efd2e2fc873 100644 --- a/tests/queries/0_stateless/03122_analyzer_collate_in_window_function.sql +++ b/tests/queries/0_stateless/03122_analyzer_collate_in_window_function.sql @@ -1,6 +1,6 @@ -- Tags: no-fasttest -- https://github.com/ClickHouse/ClickHouse/issues/44039 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; create table test_window_collate(c1 String, c2 String) engine=MergeTree order by c1; insert into test_window_collate values('1', 'äøŠęµ·'); diff --git a/tests/queries/0_stateless/03123_analyzer_dist_join_CTE.sql b/tests/queries/0_stateless/03123_analyzer_dist_join_CTE.sql index 4fb8e0b91c4..4d9f5e2971a 100644 --- a/tests/queries/0_stateless/03123_analyzer_dist_join_CTE.sql +++ b/tests/queries/0_stateless/03123_analyzer_dist_join_CTE.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/22923 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SET prefer_localhost_replica=0; create table "t0" (a Int64, b Int64) engine = MergeTree() partition by a order by a; diff --git a/tests/queries/0_stateless/03124_analyzer_nested_CTE_dist_in.sql b/tests/queries/0_stateless/03124_analyzer_nested_CTE_dist_in.sql index 406a50c6d16..be5346efa1c 100644 --- a/tests/queries/0_stateless/03124_analyzer_nested_CTE_dist_in.sql +++ b/tests/queries/0_stateless/03124_analyzer_nested_CTE_dist_in.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/23865 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; create table table_local engine = Memory AS select * from numbers(10); diff --git a/tests/queries/0_stateless/03125_analyzer_CTE_two_joins.sql b/tests/queries/0_stateless/03125_analyzer_CTE_two_joins.sql index f7d5bb5f195..934e2bc3656 100644 --- a/tests/queries/0_stateless/03125_analyzer_CTE_two_joins.sql +++ b/tests/queries/0_stateless/03125_analyzer_CTE_two_joins.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/29748 -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; create table events ( distinct_id String ) engine = Memory; diff --git a/tests/queries/0_stateless/03126_column_not_under_group_by.sql b/tests/queries/0_stateless/03126_column_not_under_group_by.sql index 890a4e823a9..516126c899c 100644 --- a/tests/queries/0_stateless/03126_column_not_under_group_by.sql +++ b/tests/queries/0_stateless/03126_column_not_under_group_by.sql @@ -1,7 +1,6 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT v.x, r.a, sum(c) FROM (select 1 x, 2 c) AS v ANY LEFT JOIN (SELECT 1 x, 2 a) AS r ON v.x = r.x GROUP BY v.x; -- { serverError NOT_AN_AGGREGATE} - 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/03129_cte_with_final.sql b/tests/queries/0_stateless/03129_cte_with_final.sql index 01e1ca6dfb0..2a0714ec571 100644 --- a/tests/queries/0_stateless/03129_cte_with_final.sql +++ b/tests/queries/0_stateless/03129_cte_with_final.sql @@ -12,7 +12,7 @@ ORDER BY key; INSERT INTO t Values (1, 'first', '2024-04-19 01:01:01'); INSERT INTO t Values (1, 'first', '2024-04-19 01:01:01'); -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; EXPLAIN QUERY TREE passes=1 WITH merged_test AS( diff --git a/tests/queries/0_stateless/03130_analyzer_self_join_group_by.sql b/tests/queries/0_stateless/03130_analyzer_self_join_group_by.sql index 66b6b99981b..81af10c4a64 100644 --- a/tests/queries/0_stateless/03130_analyzer_self_join_group_by.sql +++ b/tests/queries/0_stateless/03130_analyzer_self_join_group_by.sql @@ -2,7 +2,7 @@ DROP TABLE IF EXISTS t1; CREATE TABLE t1 (x Int32) ENGINE = MergeTree ORDER BY x; INSERT INTO t1 VALUES (1), (2), (3); -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT t2.x FROM t1 JOIN t1 as t2 ON t1.x = t2.x GROUP BY t1.x; -- { serverError NOT_AN_AGGREGATE } SELECT t2.number FROM numbers(10) as t1 JOIN numbers(10) as t2 ON t1.number = t2.number GROUP BY t1.number; -- { serverError NOT_AN_AGGREGATE } diff --git a/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.sql b/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.sql index 28362f1f469..b3d1827d98f 100644 --- a/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.sql +++ b/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET join_algorithm = 'hash'; DROP TABLE IF EXISTS test_table_1; @@ -6,14 +6,18 @@ CREATE TABLE test_table_1 ( id UInt64, value String -) ENGINE=MergeTree ORDER BY id; +) ENGINE=MergeTree ORDER BY id +SETTINGS index_granularity = 16 # We have number of granules in the `EXPLAIN` output in reference file +; DROP TABLE IF EXISTS test_table_2; CREATE TABLE test_table_2 ( id UInt64, value String -) ENGINE=MergeTree ORDER BY id; +) ENGINE=MergeTree ORDER BY id +SETTINGS index_granularity = 16 +; INSERT INTO test_table_1 VALUES (1, 'Value_1'), (2, 'Value_2'); INSERT INTO test_table_2 VALUES (2, 'Value_2'), (3, 'Value_3'); diff --git a/tests/queries/0_stateless/03132_rewrite_aggregate_function_with_if_implicit_cast.sql b/tests/queries/0_stateless/03132_rewrite_aggregate_function_with_if_implicit_cast.sql index 4e8096fbed1..7b7237ea7d1 100644 --- a/tests/queries/0_stateless/03132_rewrite_aggregate_function_with_if_implicit_cast.sql +++ b/tests/queries/0_stateless/03132_rewrite_aggregate_function_with_if_implicit_cast.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- { echoOn } SELECT concat(1, sum(if(0, toUInt128(concat('%', toLowCardinality(toNullable(1)), toUInt256(1))), materialize(0)))); SELECT any(if((number % 10) = 5, number, CAST(NULL, 'Nullable(Int128)'))) AS a, toTypeName(a) FROM numbers(100) AS a; diff --git a/tests/queries/0_stateless/03132_sqlancer_union_all.sql b/tests/queries/0_stateless/03132_sqlancer_union_all.sql index 576da9e53d5..2502ce31ec3 100644 --- a/tests/queries/0_stateless/03132_sqlancer_union_all.sql +++ b/tests/queries/0_stateless/03132_sqlancer_union_all.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS t0; DROP TABLE IF EXISTS t1; diff --git a/tests/queries/0_stateless/03140_client_subsequent_external_tables.sh b/tests/queries/0_stateless/03140_client_subsequent_external_tables.sh index af75bf42172..64398a515f5 100755 --- a/tests/queries/0_stateless/03140_client_subsequent_external_tables.sh +++ b/tests/queries/0_stateless/03140_client_subsequent_external_tables.sh @@ -7,7 +7,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) INPUT_FILE=$CUR_DIR/$CLICKHOUSE_DATABASE.tsv echo "foo" > "$INPUT_FILE" -$CLICKHOUSE_CLIENT --external --file="$INPUT_FILE" --name=t --structure='x String' -nm -q " +$CLICKHOUSE_CLIENT --external --file="$INPUT_FILE" --name=t --structure='x String' -m -q " select * from t; select * from t; " diff --git a/tests/queries/0_stateless/03142_untuple_crash.sql b/tests/queries/0_stateless/03142_untuple_crash.sql index ac5dbba0de1..45106973271 100644 --- a/tests/queries/0_stateless/03142_untuple_crash.sql +++ b/tests/queries/0_stateless/03142_untuple_crash.sql @@ -1,2 +1,2 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT untuple(x -> 0) -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } diff --git a/tests/queries/0_stateless/03142_window_function_limit_by.sql b/tests/queries/0_stateless/03142_window_function_limit_by.sql index 44dd890db41..a3f6b56aee5 100644 --- a/tests/queries/0_stateless/03142_window_function_limit_by.sql +++ b/tests/queries/0_stateless/03142_window_function_limit_by.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; -- https://github.com/ClickHouse/ClickHouse/issues/55965 @@ -8,7 +8,7 @@ CREATE TABLE error_win_func `k` String, `in` UInt64, `out` UInt64 -) +) ENGINE = MergeTree ORDER BY k AS SELECT * from VALUES (('a', 2, 4), ('a', 4, 2), ('a', 6, 3), ('a', 8, 4)); @@ -16,7 +16,7 @@ SELECT * from VALUES (('a', 2, 4), ('a', 4, 2), ('a', 6, 3), ('a', 8, 4)); SELECT k, in / out AS ratio, - count(*) OVER w AS count_rows_w + count(*) OVER w AS count_rows_w FROM error_win_func WINDOW w AS (ROWS BETWEEN CURRENT ROW AND 3 FOLLOWING) @@ -38,4 +38,3 @@ WHERE st IN ('x', 'y') LIMIT 1 BY m; DROP TABLE t; - diff --git a/tests/queries/0_stateless/03143_cte_scope.sql b/tests/queries/0_stateless/03143_cte_scope.sql index 1b1d9444651..cf4478e2ff9 100644 --- a/tests/queries/0_stateless/03143_cte_scope.sql +++ b/tests/queries/0_stateless/03143_cte_scope.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/56287 -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS tmp_a; DROP TABLE IF EXISTS tmp_b; diff --git a/tests/queries/0_stateless/03143_group_by_constant_secondary.sql b/tests/queries/0_stateless/03143_group_by_constant_secondary.sql index 030e3212748..099160dd522 100644 --- a/tests/queries/0_stateless/03143_group_by_constant_secondary.sql +++ b/tests/queries/0_stateless/03143_group_by_constant_secondary.sql @@ -3,6 +3,6 @@ SELECT count() FROM remote(test_cluster_two_shards, system, one) GROUP BY 'hi' SETTINGS - allow_experimental_analyzer = 1, + enable_analyzer = 1, group_by_two_level_threshold = 1, group_by_two_level_threshold_bytes = 33950592; diff --git a/tests/queries/0_stateless/03143_parallel_replicas_mat_view_bug.sql b/tests/queries/0_stateless/03143_parallel_replicas_mat_view_bug.sql index 97ed29802c7..02a8a2f3ce0 100644 --- a/tests/queries/0_stateless/03143_parallel_replicas_mat_view_bug.sql +++ b/tests/queries/0_stateless/03143_parallel_replicas_mat_view_bug.sql @@ -2,7 +2,7 @@ DROP TABLE IF EXISTS mv_table; DROP TABLE IF EXISTS null_table; SET cluster_for_parallel_replicas='parallel_replicas', max_parallel_replicas=4, allow_experimental_parallel_reading_from_replicas=1; -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE TABLE null_table (str String) ENGINE = Null; CREATE MATERIALIZED VIEW mv_table (str String) ENGINE = ReplicatedMergeTree('/clickhouse/tables/{database}/03143_parallel_replicas_mat_view_bug', '{replica}') ORDER BY str AS SELECT str AS str FROM null_table; diff --git a/tests/queries/0_stateless/03143_prewhere_profile_events.sh b/tests/queries/0_stateless/03143_prewhere_profile_events.sh index 00daa0fe7cc..6a6b993e5f8 100755 --- a/tests/queries/0_stateless/03143_prewhere_profile_events.sh +++ b/tests/queries/0_stateless/03143_prewhere_profile_events.sh @@ -5,7 +5,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} -nq " +${CLICKHOUSE_CLIENT} -q " DROP TABLE IF EXISTS t; CREATE TABLE t(a UInt32, b UInt32, c UInt32, d UInt32) ENGINE=MergeTree ORDER BY a SETTINGS min_bytes_for_wide_part=0, min_rows_for_wide_part=0; @@ -25,7 +25,7 @@ client_opts=( --max_threads 8 ) -${CLICKHOUSE_CLIENT} "${client_opts[@]}" --query_id "$query_id_1" -nq " +${CLICKHOUSE_CLIENT} "${client_opts[@]}" --query_id "$query_id_1" -q " SELECT * FROM t PREWHERE (b % 8192) = 42 @@ -33,7 +33,7 @@ PREWHERE (b % 8192) = 42 FORMAT Null " -${CLICKHOUSE_CLIENT} "${client_opts[@]}" --query_id "$query_id_2" -nq " +${CLICKHOUSE_CLIENT} "${client_opts[@]}" --query_id "$query_id_2" -q " SELECT * FROM t PREWHERE (b % 8192) = 42 AND (c % 8192) = 42 @@ -42,7 +42,7 @@ PREWHERE (b % 8192) = 42 AND (c % 8192) = 42 settings enable_multiple_prewhere_read_steps=1; " -${CLICKHOUSE_CLIENT} "${client_opts[@]}" --query_id "$query_id_3" -nq " +${CLICKHOUSE_CLIENT} "${client_opts[@]}" --query_id "$query_id_3" -q " SELECT * FROM t PREWHERE (b % 8192) = 42 AND (c % 16384) = 42 @@ -51,7 +51,7 @@ PREWHERE (b % 8192) = 42 AND (c % 16384) = 42 settings enable_multiple_prewhere_read_steps=0; " -${CLICKHOUSE_CLIENT} "${client_opts[@]}" --query_id "$query_id_4" -nq " +${CLICKHOUSE_CLIENT} "${client_opts[@]}" --query_id "$query_id_4" -q " SELECT b, c FROM t PREWHERE (b % 8192) = 42 AND (c % 8192) = 42 @@ -59,7 +59,7 @@ PREWHERE (b % 8192) = 42 AND (c % 8192) = 42 settings enable_multiple_prewhere_read_steps=1; " -${CLICKHOUSE_CLIENT} -nq " +${CLICKHOUSE_CLIENT} -q " SYSTEM FLUSH LOGS; -- 52503 which is 43 * number of granules, 10000000 diff --git a/tests/queries/0_stateless/03144_aggregate_states_with_different_types.sql b/tests/queries/0_stateless/03144_aggregate_states_with_different_types.sql index 5bb4a8c9c9c..4445d5b8449 100644 --- a/tests/queries/0_stateless/03144_aggregate_states_with_different_types.sql +++ b/tests/queries/0_stateless/03144_aggregate_states_with_different_types.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; select * APPLY hex from ( diff --git a/tests/queries/0_stateless/03144_invalid_filter.sql b/tests/queries/0_stateless/03144_invalid_filter.sql index deb8d7b96b3..5b434972cc7 100644 --- a/tests/queries/0_stateless/03144_invalid_filter.sql +++ b/tests/queries/0_stateless/03144_invalid_filter.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/48049 -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; CREATE TABLE test_table (`id` UInt64, `value` String) ENGINE = TinyLog() AS Select number, number::String from numbers(10); diff --git a/tests/queries/0_stateless/03145_non_loaded_projection_backup.sh b/tests/queries/0_stateless/03145_non_loaded_projection_backup.sh index 95aef9bbc5b..4e7b318e202 100755 --- a/tests/queries/0_stateless/03145_non_loaded_projection_backup.sh +++ b/tests/queries/0_stateless/03145_non_loaded_projection_backup.sh @@ -4,7 +4,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " drop table if exists tp_1; create table tp_1 (x Int32, y Int32, projection p (select x, y order by x)) engine = MergeTree order by y partition by intDiv(y, 100) settings max_parts_to_merge_at_once=1; insert into tp_1 select number, number from numbers(3); @@ -25,7 +25,7 @@ alter table tp_1 drop projection pp; alter table tp_1 attach partition '0'; " -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " set send_logs_level='fatal'; check table tp_1 settings check_query_single_value_result = 0;" | grep -o "Found unexpected projection directories: pp.proj" @@ -34,19 +34,19 @@ $CLICKHOUSE_CLIENT -q " backup table tp_1 to Disk('backups', '$backup_id'); " | grep -o "BACKUP_CREATED" -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " set send_logs_level='fatal'; drop table tp_1; restore table tp_1 from Disk('backups', '$backup_id'); " | grep -o "RESTORED" $CLICKHOUSE_CLIENT -q "select count() from tp_1;" -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " set send_logs_level='fatal'; check table tp_1 settings check_query_single_value_result = 0;" | grep -o "Found unexpected projection directories: pp.proj" -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " set send_logs_level='fatal'; check table tp_1" -$CLICKHOUSE_CLIENT -nm -q " +$CLICKHOUSE_CLIENT -m -q " set send_logs_level='fatal'; drop table tp_1" diff --git a/tests/queries/0_stateless/03146_bug47862.sql b/tests/queries/0_stateless/03146_bug47862.sql index 918f2316bea..0f411400618 100644 --- a/tests/queries/0_stateless/03146_bug47862.sql +++ b/tests/queries/0_stateless/03146_bug47862.sql @@ -9,4 +9,4 @@ FROM ( WHERE val_idx != 0 ) WHERE cast_res > 0 -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; diff --git a/tests/queries/0_stateless/03146_tpc_ds_grouping.sql b/tests/queries/0_stateless/03146_tpc_ds_grouping.sql index f48c40e9bc4..cb290086b51 100644 --- a/tests/queries/0_stateless/03146_tpc_ds_grouping.sql +++ b/tests/queries/0_stateless/03146_tpc_ds_grouping.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/46335 -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT key_a + key_b AS d, rank() OVER () AS f diff --git a/tests/queries/0_stateless/03147_table_function_loop.sql b/tests/queries/0_stateless/03147_table_function_loop.sql index aa3c8e2def5..e10155fe8b3 100644 --- a/tests/queries/0_stateless/03147_table_function_loop.sql +++ b/tests/queries/0_stateless/03147_table_function_loop.sql @@ -3,14 +3,11 @@ SELECT * FROM loop(numbers(3)) LIMIT 10; SELECT * FROM loop (numbers(3)) LIMIT 10 settings max_block_size = 1; -DROP DATABASE IF EXISTS 03147_db; -CREATE DATABASE IF NOT EXISTS 03147_db; -CREATE TABLE 03147_db.t (n Int8) ENGINE=MergeTree ORDER BY n; -INSERT INTO 03147_db.t SELECT * FROM numbers(10); -USE 03147_db; +CREATE TABLE t (n Int8) ENGINE=MergeTree ORDER BY n; +INSERT INTO t SELECT * FROM numbers(10); -SELECT * FROM loop(03147_db.t) LIMIT 15; +SELECT * FROM loop({CLICKHOUSE_DATABASE:Identifier}.t) LIMIT 15; SELECT * FROM loop(t) LIMIT 15; -SELECT * FROM loop(03147_db, t) LIMIT 15; +SELECT * FROM loop({CLICKHOUSE_DATABASE:Identifier}, t) LIMIT 15; SELECT * FROM loop('', '') -- { serverError UNKNOWN_TABLE } 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 2b4b96a9cbf..9c290133bf9 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 @@ -1,5 +1,4 @@ #!/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/03148_query_log_used_dictionaries.sql b/tests/queries/0_stateless/03148_query_log_used_dictionaries.sql index 1b647a7ee62..75b7489ae5b 100644 --- a/tests/queries/0_stateless/03148_query_log_used_dictionaries.sql +++ b/tests/queries/0_stateless/03148_query_log_used_dictionaries.sql @@ -15,7 +15,7 @@ SELECT dictGet('03148_dictionary', 'name', number) as dict_value FROM numbers(1) SETTINGS - allow_experimental_analyzer = 1, + enable_analyzer = 1, log_comment = 'simple_with_analyzer' FORMAT Null; @@ -34,7 +34,7 @@ FROM ( FROM numbers(1) ) t SETTINGS - allow_experimental_analyzer = 1, + enable_analyzer = 1, log_comment = 'nested_with_analyzer' FORMAT Null; @@ -50,7 +50,7 @@ SELECT dictGet('03148_dictionary', 'name', number) as dict_value FROM numbers(1) SETTINGS - allow_experimental_analyzer = 0, + enable_analyzer = 0, log_comment = 'simple_without_analyzer' FORMAT Null; @@ -69,7 +69,7 @@ FROM ( FROM numbers(1) ) t SETTINGS - allow_experimental_analyzer = 0, + enable_analyzer = 0, log_comment = 'nested_without_analyzer' FORMAT Null; diff --git a/tests/queries/0_stateless/03148_setting_max_streams_to_max_threads_ratio_overflow.sql b/tests/queries/0_stateless/03148_setting_max_streams_to_max_threads_ratio_overflow.sql index af326c15bd8..38f25f60e09 100644 --- a/tests/queries/0_stateless/03148_setting_max_streams_to_max_threads_ratio_overflow.sql +++ b/tests/queries/0_stateless/03148_setting_max_streams_to_max_threads_ratio_overflow.sql @@ -7,8 +7,8 @@ CREATE TABLE test_table INSERT INTO test_table VALUES (0, 'Value_0'); -SELECT * FROM test_table SETTINGS max_threads = 1025, max_streams_to_max_threads_ratio = -9223372036854775808, allow_experimental_analyzer = 1; -- { serverError PARAMETER_OUT_OF_BOUND } +SELECT * FROM test_table SETTINGS max_threads = 1025, max_streams_to_max_threads_ratio = -9223372036854775808, enable_analyzer = 1; -- { serverError PARAMETER_OUT_OF_BOUND } -SELECT * FROM test_table SETTINGS max_threads = 1025, max_streams_to_max_threads_ratio = -9223372036854775808, allow_experimental_analyzer = 0; -- { serverError PARAMETER_OUT_OF_BOUND } +SELECT * FROM test_table SETTINGS max_threads = 1025, max_streams_to_max_threads_ratio = -9223372036854775808, enable_analyzer = 0; -- { serverError PARAMETER_OUT_OF_BOUND } DROP TABLE test_table; diff --git a/tests/queries/0_stateless/03150_dynamic_type_mv_insert.reference b/tests/queries/0_stateless/03150_dynamic_type_mv_insert.reference index 0b76d30953e..370e6352657 100644 --- a/tests/queries/0_stateless/03150_dynamic_type_mv_insert.reference +++ b/tests/queries/0_stateless/03150_dynamic_type_mv_insert.reference @@ -1,35 +1,35 @@ 1 2024-01-01 Date 2 1704056400 Decimal(18, 3) -3 1 String -4 2 String +3 1 Float32 +4 2 Float64 1 2024-01-01 Date 1 2024-01-01 Date 2 1704056400 Decimal(18, 3) 2 1704056400 Decimal(18, 3) -3 1 String -3 1 String -4 2 String -4 2 String - -1 2024-01-01 String -1 2024-01-01 String -2 1704056400 String -2 1704056400 String -3 1 String -3 1 String -4 2 String -4 2 String +3 1 Float32 +3 1 Float32 +4 2 Float64 +4 2 Float64 1 2024-01-01 Date -1 2024-01-01 String -1 2024-01-01 String +1 2024-01-01 Date 2 1704056400 Decimal(18, 3) -2 1704056400 String -2 1704056400 String -3 1 String -3 1 String -3 1 String -4 2 String -4 2 String -4 2 String +2 1704056400 Decimal(18, 3) +3 1 Float32 +3 1 Float32 +4 2 Float64 +4 2 Float64 + +1 2024-01-01 Date +1 2024-01-01 Date +1 2024-01-01 Date +2 1704056400 Decimal(18, 3) +2 1704056400 Decimal(18, 3) +2 1704056400 Decimal(18, 3) +3 1 Float32 +3 1 Float32 +3 1 Float32 +4 2 Float64 +4 2 Float64 +4 2 Float64 diff --git a/tests/queries/0_stateless/03150_dynamic_type_mv_insert.sql b/tests/queries/0_stateless/03150_dynamic_type_mv_insert.sql index ad5ea9512c6..71d5dd4abd1 100644 --- a/tests/queries/0_stateless/03150_dynamic_type_mv_insert.sql +++ b/tests/queries/0_stateless/03150_dynamic_type_mv_insert.sql @@ -1,5 +1,6 @@ SET allow_experimental_dynamic_type=1; +DROP TABLE IF EXISTS null_table; CREATE TABLE null_table ( n1 UInt8, @@ -7,9 +8,11 @@ CREATE TABLE null_table ) ENGINE = Null; +DROP VIEW IF EXISTS dummy_rmv; CREATE MATERIALIZED VIEW dummy_rmv TO to_table AS SELECT * FROM null_table; +DROP TABLE IF EXISTS to_table; CREATE TABLE to_table ( n1 UInt8, @@ -32,3 +35,7 @@ select ''; ALTER TABLE to_table MODIFY COLUMN n2 Dynamic(max_types=10); INSERT INTO null_table ( n1, n2 ) VALUES (1, '2024-01-01'), (2, toDateTime64('2024-01-01', 3, 'Asia/Istanbul')), (3, toFloat32(1)), (4, toFloat64(2)); SELECT *, dynamicType(n2) FROM to_table ORDER BY ALL; + +DROP TABLE null_table; +DROP VIEW dummy_rmv; +DROP TABLE to_table; diff --git a/tests/queries/0_stateless/03150_grouping_sets_use_nulls_pushdown.sql b/tests/queries/0_stateless/03150_grouping_sets_use_nulls_pushdown.sql index a0bd1381351..c39143216d4 100644 --- a/tests/queries/0_stateless/03150_grouping_sets_use_nulls_pushdown.sql +++ b/tests/queries/0_stateless/03150_grouping_sets_use_nulls_pushdown.sql @@ -21,21 +21,21 @@ SELECT * FROM ( SELECT day_ FROM test_grouping_sets_predicate GROUP BY GROUPING SETS ( (day_, type_1), (day_) ) ) WHERE day_ = '2023-01-05' ORDER BY * -SETTINGS allow_experimental_analyzer=1; +SETTINGS enable_analyzer=1; SELECT * FROM ( SELECT * FROM test_grouping_sets_predicate GROUP BY GROUPING SETS ( (day_, type_1), (day_) ) ) WHERE day_ = '2023-01-05' GROUP BY * ORDER BY ALL -SETTINGS allow_experimental_analyzer=1; +SETTINGS enable_analyzer=1; SELECT * FROM ( SELECT * FROM test_grouping_sets_predicate GROUP BY GROUPING SETS ( (*), (day_) ) ) WHERE day_ = '2023-01-05' GROUP BY GROUPING SETS (*) ORDER BY type_1 -SETTINGS allow_experimental_analyzer=1; +SETTINGS enable_analyzer=1; SELECT * FROM ( SELECT day_, COUNT(*) FROM test_grouping_sets_predicate GROUP BY GROUPING SETS ( (day_, type_1), (day_) ) ) diff --git a/tests/queries/0_stateless/03151_analyzer_view_read_only_necessary_columns.sql b/tests/queries/0_stateless/03151_analyzer_view_read_only_necessary_columns.sql index 40204b5cd03..ac86a8705d5 100644 --- a/tests/queries/0_stateless/03151_analyzer_view_read_only_necessary_columns.sql +++ b/tests/queries/0_stateless/03151_analyzer_view_read_only_necessary_columns.sql @@ -8,7 +8,7 @@ CREATE TABLE test_table DROP VIEW IF EXISTS test_view; CREATE VIEW test_view AS SELECT id, value FROM test_table; -EXPLAIN header = 1 SELECT sum(id) FROM test_view settings allow_experimental_analyzer=1; +EXPLAIN header = 1 SELECT sum(id) FROM test_view settings enable_analyzer=1; DROP VIEW test_view; DROP TABLE test_table; diff --git a/tests/queries/0_stateless/03151_dynamic_type_scale_max_types.reference b/tests/queries/0_stateless/03151_dynamic_type_scale_max_types.reference index d96fbf658d8..2d3b2f118f6 100644 --- a/tests/queries/0_stateless/03151_dynamic_type_scale_max_types.reference +++ b/tests/queries/0_stateless/03151_dynamic_type_scale_max_types.reference @@ -1,26 +1,26 @@ -1 2024-01-01 Date -2 1704056400 String -3 1 String -4 2 String +1 2024-01-01 Date false +2 1704056400 Decimal(18, 3) false +3 1 Float32 true +4 2 Float64 true -1 2024-01-01 Date -1 2024-01-01 Date -2 1704056400 Decimal(18, 3) -2 1704056400 String -3 1 Float32 -3 1 String -4 2 Float64 -4 2 String +1 2024-01-01 Date false +1 2024-01-01 Date false +2 1704056400 Decimal(18, 3) false +2 1704056400 Decimal(18, 3) false +3 1 Float32 false +3 1 Float32 false +4 2 Float64 false +4 2 Float64 false -1 2024-01-01 String -1 2024-01-01 String -1 2024-01-01 String -2 1704056400 String -2 1704056400 String -2 1704056400 String -3 1 String -3 1 String -3 1 String -4 2 String -4 2 String -4 2 String +1 2024-01-01 Date true +1 2024-01-01 Date true +1 2024-01-01 Date true +2 1704056400 Decimal(18, 3) true +2 1704056400 Decimal(18, 3) true +2 1704056400 Decimal(18, 3) true +3 1 Float32 true +3 1 Float32 true +3 1 Float32 true +4 2 Float64 true +4 2 Float64 true +4 2 Float64 true diff --git a/tests/queries/0_stateless/03151_dynamic_type_scale_max_types.sql b/tests/queries/0_stateless/03151_dynamic_type_scale_max_types.sql index 632f3504fdb..e476d34a1db 100644 --- a/tests/queries/0_stateless/03151_dynamic_type_scale_max_types.sql +++ b/tests/queries/0_stateless/03151_dynamic_type_scale_max_types.sql @@ -11,16 +11,16 @@ CREATE TABLE to_table ENGINE = MergeTree ORDER BY n1; INSERT INTO to_table ( n1, n2 ) VALUES (1, '2024-01-01'), (2, toDateTime64('2024-01-01', 3, 'Asia/Istanbul')), (3, toFloat32(1)), (4, toFloat64(2)); -SELECT *, dynamicType(n2) FROM to_table ORDER BY ALL; +SELECT *, dynamicType(n2), isDynamicElementInSharedData(n2) FROM to_table ORDER BY ALL; select ''; ALTER TABLE to_table MODIFY COLUMN n2 Dynamic(max_types=5); INSERT INTO to_table ( n1, n2 ) VALUES (1, '2024-01-01'), (2, toDateTime64('2024-01-01', 3, 'Asia/Istanbul')), (3, toFloat32(1)), (4, toFloat64(2)); -SELECT *, dynamicType(n2) FROM to_table ORDER BY ALL; +SELECT *, dynamicType(n2), isDynamicElementInSharedData(n2) FROM to_table ORDER BY ALL; select ''; -ALTER TABLE to_table MODIFY COLUMN n2 Dynamic(max_types=1); +ALTER TABLE to_table MODIFY COLUMN n2 Dynamic(max_types=0); INSERT INTO to_table ( n1, n2 ) VALUES (1, '2024-01-01'), (2, toDateTime64('2024-01-01', 3, 'Asia/Istanbul')), (3, toFloat32(1)), (4, toFloat64(2)); -SELECT *, dynamicType(n2) FROM to_table ORDER BY ALL; +SELECT *, dynamicType(n2), isDynamicElementInSharedData(n2) FROM to_table ORDER BY ALL; ALTER TABLE to_table MODIFY COLUMN n2 Dynamic(max_types=500); -- { serverError UNEXPECTED_AST_STRUCTURE } diff --git a/tests/queries/0_stateless/03152_dynamic_type_simple.reference b/tests/queries/0_stateless/03152_dynamic_type_simple.reference index 5f243209ff3..e508bdd1990 100644 --- a/tests/queries/0_stateless/03152_dynamic_type_simple.reference +++ b/tests/queries/0_stateless/03152_dynamic_type_simple.reference @@ -7,7 +7,7 @@ string2 String \N None 42 Int64 42 Int64 string String -string String [1, 2] String +string String [1,2] Array(Int64) [1,2] Array(Int64) \N None ā”Œā”€dā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”¬ā”€dynamicType(d)ā”€ā”¬ā”€d.Int64ā”€ā”¬ā”€d.Stringā”€ā”¬ā”€ā”€ā”€ā”€ā”€d.Dateā”€ā”¬ā”€d.Float64ā”€ā”¬ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€d.DateTimeā”€ā”¬ā”€d.Array(Int64)ā”€ā”¬ā”€d.Array(String)ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā”€ā” 1. ā”‚ 42 ā”‚ Int64 ā”‚ 42 ā”‚ į“ŗįµį“øį“ø ā”‚ į“ŗįµį“øį“ø ā”‚ į“ŗįµį“øį“ø ā”‚ į“ŗįµį“øį“ø ā”‚ [] ā”‚ [] ā”‚ diff --git a/tests/queries/0_stateless/03152_dynamic_type_simple.sql b/tests/queries/0_stateless/03152_dynamic_type_simple.sql index fd5328faf15..ed24b213b1c 100644 --- a/tests/queries/0_stateless/03152_dynamic_type_simple.sql +++ b/tests/queries/0_stateless/03152_dynamic_type_simple.sql @@ -1,14 +1,17 @@ SET allow_experimental_dynamic_type=1; +DROP TABLE IF EXISTS test_max_types; CREATE TABLE test_max_types (d Dynamic(max_types=5)) ENGINE = Memory; INSERT INTO test_max_types VALUES ('string1'), (42), (3.14), ([1, 2]), (toDate('2021-01-01')), ('string2'); SELECT d, dynamicType(d) FROM test_max_types; SELECT ''; +DROP TABLE IF EXISTS test_nested_dynamic; CREATE TABLE test_nested_dynamic (d1 Dynamic, d2 Dynamic(max_types=2)) ENGINE = Memory; INSERT INTO test_nested_dynamic VALUES (NULL, 42), (42, 'string'), ('string', [1, 2]), ([1, 2], NULL); SELECT d1, dynamicType(d1), d2, dynamicType(d2) FROM test_nested_dynamic; +DROP TABLE IF EXISTS test_rapid_schema; CREATE TABLE test_rapid_schema (d Dynamic) ENGINE = Memory; INSERT INTO test_rapid_schema VALUES (42), ('string1'), (toDate('2021-01-01')), ([1, 2, 3]), (3.14), ('string2'), (toDateTime('2021-01-01 12:00:00')), (['array', 'of', 'strings']), (NULL), (toFloat64(42.42)); @@ -27,3 +30,8 @@ FROM FROM numbers(10000) ) ); + +DROP TABLE test_max_types; +DROP TABLE test_nested_dynamic; +DROP TABLE test_rapid_schema; + diff --git a/tests/queries/0_stateless/03152_join_filter_push_down_equivalent_columns.sql b/tests/queries/0_stateless/03152_join_filter_push_down_equivalent_columns.sql index 645e89034d7..ea8a9e1f8d7 100644 --- a/tests/queries/0_stateless/03152_join_filter_push_down_equivalent_columns.sql +++ b/tests/queries/0_stateless/03152_join_filter_push_down_equivalent_columns.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS users; CREATE TABLE users (uid Int16, name String, age Int16) ENGINE=MergeTree order by (uid, name); diff --git a/tests/queries/0_stateless/03153_dynamic_type_empty.sql b/tests/queries/0_stateless/03153_dynamic_type_empty.sql index 8e942fe6f6e..3a0c98e63ee 100644 --- a/tests/queries/0_stateless/03153_dynamic_type_empty.sql +++ b/tests/queries/0_stateless/03153_dynamic_type_empty.sql @@ -1,5 +1,7 @@ SET allow_experimental_dynamic_type=1; +DROP TABLE IF EXISTS test_null_empty; CREATE TABLE test_null_empty (d Dynamic) ENGINE = Memory; INSERT INTO test_null_empty VALUES ([]), ([1]), ([]), (['1']), ([]), (()),((1)), (()), (('1')), (()), ({}), ({1:2}), ({}), ({'1':'2'}), ({}); SELECT d, dynamicType(d) FROM test_null_empty; +DROP TABLE test_null_empty; diff --git a/tests/queries/0_stateless/03154_recursive_cte_distributed.sql b/tests/queries/0_stateless/03154_recursive_cte_distributed.sql index b8c3356d5d6..47e0b9aad0b 100644 --- a/tests/queries/0_stateless/03154_recursive_cte_distributed.sql +++ b/tests/queries/0_stateless/03154_recursive_cte_distributed.sql @@ -1,6 +1,6 @@ -- Tags: shard -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_table; CREATE TABLE test_table diff --git a/tests/queries/0_stateless/03155_analyzer_interpolate.sql b/tests/queries/0_stateless/03155_analyzer_interpolate.sql index 30423cb86ff..42c5f5ef65f 100644 --- a/tests/queries/0_stateless/03155_analyzer_interpolate.sql +++ b/tests/queries/0_stateless/03155_analyzer_interpolate.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/62464 -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT n, [number] AS inter FROM ( SELECT toFloat32(number % 10) AS n, number diff --git a/tests/queries/0_stateless/03155_in_nested_subselects.sql b/tests/queries/0_stateless/03155_in_nested_subselects.sql index 4f5ccd30aa3..faecb73040d 100644 --- a/tests/queries/0_stateless/03155_in_nested_subselects.sql +++ b/tests/queries/0_stateless/03155_in_nested_subselects.sql @@ -1,5 +1,5 @@ -- https://github.com/ClickHouse/ClickHouse/issues/63833 -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; create table Example (id Int32) engine = MergeTree ORDER BY id; INSERT INTO Example SELECT number AS id FROM numbers(2); @@ -8,10 +8,10 @@ create table Null engine=Null as Example ; --create table Null engine=MergeTree order by id as Example ; create materialized view Transform to Example as -select * from Null -join ( select * FROM Example +select * from Null +join ( select * FROM Example WHERE id IN (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM (SELECT * FROM Null))))) - ) as old + ) as old using id; INSERT INTO Null SELECT number AS id FROM numbers(2); diff --git a/tests/queries/0_stateless/03155_test_move_to_prewhere.sh b/tests/queries/0_stateless/03155_test_move_to_prewhere.sh index b6980b3a23a..f33a6b3ef27 100755 --- a/tests/queries/0_stateless/03155_test_move_to_prewhere.sh +++ b/tests/queries/0_stateless/03155_test_move_to_prewhere.sh @@ -4,7 +4,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -${CLICKHOUSE_CLIENT} -nq " +${CLICKHOUSE_CLIENT} -q " CREATE TABLE event_envoy ( timestamp_interval DateTime CODEC(DoubleDelta), @@ -18,7 +18,7 @@ ${CLICKHOUSE_CLIENT} -nq " INSERT INTO event_envoy SELECT now() - number, 'us-east-1', 'ch_super_fast' FROM numbers_mt(1e5); " -${CLICKHOUSE_CLIENT} -nq " +${CLICKHOUSE_CLIENT} -q " CREATE TABLE event_envoy_remote ( timestamp_interval DateTime CODEC(DoubleDelta), diff --git a/tests/queries/0_stateless/03156_default_multiquery_split.sh b/tests/queries/0_stateless/03156_default_multiquery_split.sh index ac64c2d093d..d849fb5a162 100755 --- a/tests/queries/0_stateless/03156_default_multiquery_split.sh +++ b/tests/queries/0_stateless/03156_default_multiquery_split.sh @@ -1,11 +1,10 @@ #!/usr/bin/env bash -# Tags: no-parallel, no-ordinary-database CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -SQL_FILE_NAME=$"03156_default_multiquery_split_$$.sql" +SQL_FILE_NAME=$"03156_default_multiquery_split_${CLICKHOUSE_DATABASE}.sql" # The old multiquery implementation uses '\n' to split INSERT query segmentation # this case is mainly to test the following situations @@ -54,6 +53,6 @@ SELECT * FROM TEST2 ORDER BY value; DROP TABLE TEST1; DROP TABLE TEST2; EOF -$CLICKHOUSE_CLIENT -m -n < "$SQL_FILE_NAME" +$CLICKHOUSE_CLIENT -m < "$SQL_FILE_NAME" rm "$SQL_FILE_NAME" diff --git a/tests/queries/0_stateless/03159_dynamic_type_all_types.sql b/tests/queries/0_stateless/03159_dynamic_type_all_types.sql index d302205ca23..fffea1bd0f5 100644 --- a/tests/queries/0_stateless/03159_dynamic_type_all_types.sql +++ b/tests/queries/0_stateless/03159_dynamic_type_all_types.sql @@ -6,7 +6,7 @@ SET allow_experimental_variant_type=1; SET allow_suspicious_low_cardinality_types=1; -CREATE TABLE t (d Dynamic(max_types=255)) ENGINE = Memory; +CREATE TABLE t (d Dynamic(max_types=254)) ENGINE = Memory; -- Integer types: signed and unsigned integers (UInt8, UInt16, UInt32, UInt64, UInt128, UInt256, Int8, Int16, Int32, Int64, Int128, Int256) INSERT INTO t VALUES (-128::Int8), (-127::Int8), (-1::Int8), (0::Int8), (1::Int8), (126::Int8), (127::Int8); INSERT INTO t VALUES (-128::Int8), (-127::Int8), (-1::Int8), (0::Int8), (1::Int8), (126::Int8), (127::Int8); @@ -84,7 +84,7 @@ INSERT INTO t VALUES ([(1, (2, ['aa', 'bb']), [(3, 'cc'), (4, 'dd')]), (5, (6, [ SELECT dynamicType(d), d FROM t ORDER BY substring(dynamicType(d),1,1), length(dynamicType(d)), d; -CREATE TABLE t2 (d Dynamic(max_types=255)) ENGINE = Memory; +CREATE TABLE t2 (d Dynamic(max_types=254)) ENGINE = Memory; INSERT INTO t2 SELECT * FROM t; SELECT ''; diff --git a/tests/queries/0_stateless/03161_cnf_reduction.reference b/tests/queries/0_stateless/03161_cnf_reduction.reference index 5e39c0f3223..41051af362f 100644 --- a/tests/queries/0_stateless/03161_cnf_reduction.reference +++ b/tests/queries/0_stateless/03161_cnf_reduction.reference @@ -2,7 +2,7 @@ SELECT id FROM `03161_table` WHERE f -SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, allow_experimental_analyzer = 1 +SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, enable_analyzer = 1 -- Expected result with analyzer: 1 @@ -11,7 +11,7 @@ SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, allow_experim SELECT id FROM `03161_table` WHERE f -SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, allow_experimental_analyzer = 0 +SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, enable_analyzer = 0 -- Expected result w/o analyzer: 1 diff --git a/tests/queries/0_stateless/03161_cnf_reduction.sql b/tests/queries/0_stateless/03161_cnf_reduction.sql index b34e9171d45..c232823e9cd 100644 --- a/tests/queries/0_stateless/03161_cnf_reduction.sql +++ b/tests/queries/0_stateless/03161_cnf_reduction.sql @@ -10,7 +10,7 @@ EXPLAIN SYNTAX SELECT id FROM 03161_table WHERE f AND (NOT(f) OR f) -SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, allow_experimental_analyzer = 1; +SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, enable_analyzer = 1; SELECT ''; @@ -19,7 +19,7 @@ SELECT '-- Expected result with analyzer:'; SELECT id FROM 03161_table WHERE f AND (NOT(f) OR f) -SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, allow_experimental_analyzer = 1; +SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, enable_analyzer = 1; SELECT ''; @@ -29,7 +29,7 @@ EXPLAIN SYNTAX SELECT id FROM 03161_table WHERE f AND (NOT(f) OR f) -SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, allow_experimental_analyzer = 0; +SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, enable_analyzer = 0; SELECT ''; @@ -38,7 +38,7 @@ SELECT '-- Expected result w/o analyzer:'; SELECT id FROM 03161_table WHERE f AND (NOT(f) OR f) -SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, allow_experimental_analyzer = 0; +SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, enable_analyzer = 0; DROP TABLE IF EXISTS 03161_table; @@ -58,7 +58,7 @@ SELECT '-- Reproducer from the issue with analyzer'; SELECT count() FROM 03161_reproducer WHERE ((NOT c2) AND c2 AND (NOT c1)) OR ((NOT c2) AND c3 AND (NOT c5)) OR ((NOT c7) AND (NOT c8)) OR (c9 AND c6 AND c8 AND (NOT c8) AND (NOT c7)) -SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, allow_experimental_analyzer = 1; +SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, enable_analyzer = 1; SELECT ''; @@ -67,6 +67,6 @@ SELECT '-- Reproducer from the issue w/o analyzer'; SELECT count() FROM 03161_reproducer WHERE ((NOT c2) AND c2 AND (NOT c1)) OR ((NOT c2) AND c3 AND (NOT c5)) OR ((NOT c7) AND (NOT c8)) OR (c9 AND c6 AND c8 AND (NOT c8) AND (NOT c7)) -SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, allow_experimental_analyzer = 0; +SETTINGS convert_query_to_cnf = 1, optimize_using_constraints = 1, enable_analyzer = 0; DROP TABLE IF EXISTS 03161_reproducer; diff --git a/tests/queries/0_stateless/03161_lightweight_delete_projection.reference b/tests/queries/0_stateless/03161_lightweight_delete_projection.reference index c5a6cbab0bc..8edf541c2a0 100644 --- a/tests/queries/0_stateless/03161_lightweight_delete_projection.reference +++ b/tests/queries/0_stateless/03161_lightweight_delete_projection.reference @@ -1,2 +1,90 @@ -1231 John 33 +compact part +testing throw default mode +-- { echoOn } + +ALTER TABLE users_compact MODIFY SETTING lightweight_mutation_projection_mode = 'throw'; +DELETE FROM users_compact WHERE uid = 1231; -- { serverError SUPPORT_IS_DISABLED } +SELECT 'testing drop mode'; +testing drop mode +ALTER TABLE users_compact MODIFY SETTING lightweight_mutation_projection_mode = 'drop'; +DELETE FROM users_compact WHERE uid = 1231; +SELECT * FROM users_compact ORDER BY uid; +SYSTEM FLUSH LOGS; +-- all_1_1_0_2 +SELECT + name +FROM system.parts +WHERE (database = currentDatabase()) AND (`table` = 'users_compact') AND (active = 1); +all_1_1_0_2 +-- expecting no projection +SELECT + name, parent_name +FROM system.projection_parts +WHERE (database = currentDatabase()) AND (`table` = 'users_compact') AND (active = 1); +SELECT 'testing rebuild mode'; +testing rebuild mode +INSERT INTO users_compact VALUES (6666, 'Ksenia', 48), (8888, 'Alice', 50); +ALTER TABLE users_compact MODIFY SETTING lightweight_mutation_projection_mode = 'rebuild'; +DELETE FROM users_compact WHERE uid = 6666; +SELECT * FROM users_compact ORDER BY uid; 8888 Alice 50 +SYSTEM FLUSH LOGS; +-- all_1_1_0_4, all_3_3_0_4 +SELECT + name +FROM system.parts +WHERE (database = currentDatabase()) AND (`table` = 'users_compact') AND (active = 1); +all_1_1_0_4 +all_3_3_0_4 +-- expecting projection p1, p2 +SELECT + name, parent_name +FROM system.projection_parts +WHERE (database = currentDatabase()) AND (`table` = 'users_compact') AND (active = 1) AND parent_name like 'all_3_3%'; +p1 all_3_3_0_4 +p2 all_3_3_0_4 +wide part +testing throw default mode +-- { echoOn } + +ALTER TABLE users_wide MODIFY SETTING lightweight_mutation_projection_mode = 'throw'; +DELETE FROM users_wide WHERE uid = 1231; -- { serverError SUPPORT_IS_DISABLED } +SELECT 'testing drop mode'; +testing drop mode +ALTER TABLE users_wide MODIFY SETTING lightweight_mutation_projection_mode = 'drop'; +DELETE FROM users_wide WHERE uid = 1231; +SELECT * FROM users_wide ORDER BY uid; +SYSTEM FLUSH LOGS; +-- all_1_1_0_2 +SELECT + name +FROM system.parts +WHERE (database = currentDatabase()) AND (`table` = 'users_wide') AND (active = 1); +all_1_1_0_2 +-- expecting no projection +SELECT + name, parent_name +FROM system.projection_parts +WHERE (database = currentDatabase()) AND (`table` = 'users_wide') AND (active = 1); +SELECT 'testing rebuild mode'; +testing rebuild mode +INSERT INTO users_wide VALUES (6666, 'Ksenia', 48), (8888, 'Alice', 50); +ALTER TABLE users_wide MODIFY SETTING lightweight_mutation_projection_mode = 'rebuild'; +DELETE FROM users_wide WHERE uid = 6666; +SELECT * FROM users_wide ORDER BY uid; +8888 Alice 50 +SYSTEM FLUSH LOGS; +-- all_1_1_0_4, all_3_3_0_4 +SELECT + name +FROM system.parts +WHERE (database = currentDatabase()) AND (`table` = 'users_wide') AND (active = 1); +all_1_1_0_4 +all_3_3_0_4 +-- expecting projection p1, p2 +SELECT + name, parent_name +FROM system.projection_parts +WHERE (database = currentDatabase()) AND (`table` = 'users_wide') AND (active = 1) AND parent_name like 'all_3_3%'; +p1 all_3_3_0_4 +p2 all_3_3_0_4 diff --git a/tests/queries/0_stateless/03161_lightweight_delete_projection.sql b/tests/queries/0_stateless/03161_lightweight_delete_projection.sql index b189388e356..da6427cbf22 100644 --- a/tests/queries/0_stateless/03161_lightweight_delete_projection.sql +++ b/tests/queries/0_stateless/03161_lightweight_delete_projection.sql @@ -1,31 +1,145 @@ +-- For cloud version, should also consider min_bytes_for_full_part_storage since packed storage exists, +-- but for less redundancy, just let CI test the parameter. -DROP TABLE IF EXISTS users; +SET lightweight_deletes_sync = 2, alter_sync = 2; -CREATE TABLE users ( +DROP TABLE IF EXISTS users_compact; + + +SELECT 'compact part'; + +CREATE TABLE users_compact ( uid Int16, name String, age Int16, projection p1 (select count(), age group by age), projection p2 (select age, name group by age, name) -) ENGINE = MergeTree order by uid; +) ENGINE = MergeTree order by uid +SETTINGS min_bytes_for_wide_part = 10485760; -INSERT INTO users VALUES (1231, 'John', 33); -INSERT INTO users VALUES (6666, 'Ksenia', 48); -INSERT INTO users VALUES (8888, 'Alice', 50); +INSERT INTO users_compact VALUES (1231, 'John', 33); -DELETE FROM users WHERE 1; -- { serverError NOT_IMPLEMENTED } +SELECT 'testing throw default mode'; -DELETE FROM users WHERE uid = 8888 SETTINGS lightweight_mutation_projection_mode = 'throw'; -- { serverError NOT_IMPLEMENTED } +-- { echoOn } -DELETE FROM users WHERE uid = 6666 SETTINGS lightweight_mutation_projection_mode = 'drop'; +ALTER TABLE users_compact MODIFY SETTING lightweight_mutation_projection_mode = 'throw'; + +DELETE FROM users_compact WHERE uid = 1231; -- { serverError SUPPORT_IS_DISABLED } + +SELECT 'testing drop mode'; +ALTER TABLE users_compact MODIFY SETTING lightweight_mutation_projection_mode = 'drop'; + +DELETE FROM users_compact WHERE uid = 1231; + +SELECT * FROM users_compact ORDER BY uid; + +SYSTEM FLUSH LOGS; + +-- all_1_1_0_2 +SELECT + name +FROM system.parts +WHERE (database = currentDatabase()) AND (`table` = 'users_compact') AND (active = 1); -- expecting no projection SELECT - name, - `table` + name, parent_name FROM system.projection_parts -WHERE (database = currentDatabase()) AND (`table` = 'users'); +WHERE (database = currentDatabase()) AND (`table` = 'users_compact') AND (active = 1); -SELECT * FROM users ORDER BY uid; +SELECT 'testing rebuild mode'; +INSERT INTO users_compact VALUES (6666, 'Ksenia', 48), (8888, 'Alice', 50); -DROP TABLE users; +ALTER TABLE users_compact MODIFY SETTING lightweight_mutation_projection_mode = 'rebuild'; + +DELETE FROM users_compact WHERE uid = 6666; + +SELECT * FROM users_compact ORDER BY uid; + +SYSTEM FLUSH LOGS; + +-- all_1_1_0_4, all_3_3_0_4 +SELECT + name +FROM system.parts +WHERE (database = currentDatabase()) AND (`table` = 'users_compact') AND (active = 1); + +-- expecting projection p1, p2 +SELECT + name, parent_name +FROM system.projection_parts +WHERE (database = currentDatabase()) AND (`table` = 'users_compact') AND (active = 1) AND parent_name like 'all_3_3%'; + +-- { echoOff } + +DROP TABLE users_compact; + + +SELECT 'wide part'; +CREATE TABLE users_wide ( + uid Int16, + name String, + age Int16, + projection p1 (select count(), age group by age), + projection p2 (select age, name group by age, name) +) ENGINE = MergeTree order by uid +SETTINGS min_bytes_for_wide_part = 0; + +INSERT INTO users_wide VALUES (1231, 'John', 33); + +SELECT 'testing throw default mode'; + +-- { echoOn } + +ALTER TABLE users_wide MODIFY SETTING lightweight_mutation_projection_mode = 'throw'; + +DELETE FROM users_wide WHERE uid = 1231; -- { serverError SUPPORT_IS_DISABLED } + +SELECT 'testing drop mode'; +ALTER TABLE users_wide MODIFY SETTING lightweight_mutation_projection_mode = 'drop'; + +DELETE FROM users_wide WHERE uid = 1231; + +SELECT * FROM users_wide ORDER BY uid; + +SYSTEM FLUSH LOGS; + +-- all_1_1_0_2 +SELECT + name +FROM system.parts +WHERE (database = currentDatabase()) AND (`table` = 'users_wide') AND (active = 1); + +-- expecting no projection +SELECT + name, parent_name +FROM system.projection_parts +WHERE (database = currentDatabase()) AND (`table` = 'users_wide') AND (active = 1); + +SELECT 'testing rebuild mode'; +INSERT INTO users_wide VALUES (6666, 'Ksenia', 48), (8888, 'Alice', 50); + +ALTER TABLE users_wide MODIFY SETTING lightweight_mutation_projection_mode = 'rebuild'; + +DELETE FROM users_wide WHERE uid = 6666; + +SELECT * FROM users_wide ORDER BY uid; + +SYSTEM FLUSH LOGS; + +-- all_1_1_0_4, all_3_3_0_4 +SELECT + name +FROM system.parts +WHERE (database = currentDatabase()) AND (`table` = 'users_wide') AND (active = 1); + +-- expecting projection p1, p2 +SELECT + name, parent_name +FROM system.projection_parts +WHERE (database = currentDatabase()) AND (`table` = 'users_wide') AND (active = 1) AND parent_name like 'all_3_3%'; + +-- { echoOff } + +DROP TABLE users_wide; diff --git a/tests/queries/0_stateless/03164_adapting_parquet_reader_output_size.sql b/tests/queries/0_stateless/03164_adapting_parquet_reader_output_size.sql index fa098b64702..e6b13510301 100644 --- a/tests/queries/0_stateless/03164_adapting_parquet_reader_output_size.sql +++ b/tests/queries/0_stateless/03164_adapting_parquet_reader_output_size.sql @@ -1,4 +1,4 @@ --- Tags: no-fasttest, no-parallel, no-random-settings +-- Tags: no-fasttest, no-random-settings set max_insert_threads=1; @@ -22,4 +22,4 @@ CREATE TABLE test_parquet (col1 String, col2 String, col3 String, col4 String, c INSERT INTO test_parquet SELECT rand(),rand(),rand(),rand(),rand(),rand(),rand() FROM numbers(100000); SELECT max(blockSize()) FROM test_parquet; -DROP TABLE IF EXISTS test_parquet; \ No newline at end of file +DROP TABLE IF EXISTS test_parquet; diff --git a/tests/queries/0_stateless/03164_analyzer_global_in_alias.sql b/tests/queries/0_stateless/03164_analyzer_global_in_alias.sql index 00c293334ee..ccfacd12d98 100644 --- a/tests/queries/0_stateless/03164_analyzer_global_in_alias.sql +++ b/tests/queries/0_stateless/03164_analyzer_global_in_alias.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT 1 GLOBAL IN (SELECT 1) AS s, s FROM remote('127.0.0.{2,3}', system.one) GROUP BY 1; SELECT 1 GLOBAL IN (SELECT 1) AS s FROM remote('127.0.0.{2,3}', system.one) GROUP BY 1; diff --git a/tests/queries/0_stateless/03164_early_constant_folding_analyzer.sql b/tests/queries/0_stateless/03164_early_constant_folding_analyzer.sql index dbffbc1af71..b1018d00082 100644 --- a/tests/queries/0_stateless/03164_early_constant_folding_analyzer.sql +++ b/tests/queries/0_stateless/03164_early_constant_folding_analyzer.sql @@ -27,4 +27,4 @@ ENGINE = MergeTree ORDER BY (date, pull_request_number, commit_sha, check_name, insert into checks select * from generateRandom() limit 1; -select trimLeft(explain) from (explain SELECT count(1) FROM checks WHERE test_name IS NOT NULL) where explain like '%ReadFromPreparedSource%' SETTINGS allow_experimental_analyzer = 1, allow_experimental_parallel_reading_from_replicas = 0; +select trimLeft(explain) from (explain SELECT count(1) FROM checks WHERE test_name IS NOT NULL) where explain like '%ReadFromPreparedSource%' SETTINGS enable_analyzer = 1, allow_experimental_parallel_reading_from_replicas = 0; diff --git a/tests/queries/0_stateless/03164_materialize_skip_index.sql b/tests/queries/0_stateless/03164_materialize_skip_index.sql index 4e59ef6b6cd..0443872596d 100644 --- a/tests/queries/0_stateless/03164_materialize_skip_index.sql +++ b/tests/queries/0_stateless/03164_materialize_skip_index.sql @@ -9,7 +9,7 @@ CREATE TABLE t_skip_index_insert ) ENGINE = MergeTree ORDER BY tuple() SETTINGS index_granularity = 4; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SET materialize_skip_indexes_on_insert = 0; SYSTEM STOP MERGES t_skip_index_insert; diff --git a/tests/queries/0_stateless/03165_order_by_duplicate.reference b/tests/queries/0_stateless/03165_order_by_duplicate.reference index 5d5e7a33f4a..93d312d5d02 100644 --- a/tests/queries/0_stateless/03165_order_by_duplicate.reference +++ b/tests/queries/0_stateless/03165_order_by_duplicate.reference @@ -36,4 +36,4 @@ QUERY id: 0 LIMIT BY LIST id: 16, nodes: 1 COLUMN id: 2, column_name: id, result_type: UInt64, source_id: 3 - SETTINGS allow_experimental_analyzer=1 + SETTINGS enable_analyzer=1 diff --git a/tests/queries/0_stateless/03165_order_by_duplicate.sql b/tests/queries/0_stateless/03165_order_by_duplicate.sql index 0054cbc36a6..b8bcc10e119 100644 --- a/tests/queries/0_stateless/03165_order_by_duplicate.sql +++ b/tests/queries/0_stateless/03165_order_by_duplicate.sql @@ -13,4 +13,4 @@ WHERE id IN ( ) ORDER BY id ASC LIMIT 1 BY id -SETTINGS allow_experimental_analyzer = 1; +SETTINGS enable_analyzer = 1; diff --git a/tests/queries/0_stateless/03166_mv_prewhere_duplicating_name_bug.sql b/tests/queries/0_stateless/03166_mv_prewhere_duplicating_name_bug.sql index e32d23920dd..e27e8645466 100644 --- a/tests/queries/0_stateless/03166_mv_prewhere_duplicating_name_bug.sql +++ b/tests/queries/0_stateless/03166_mv_prewhere_duplicating_name_bug.sql @@ -3,5 +3,5 @@ create table dst (s String, lc LowCardinality(String)) engine MergeTree order by create materialized view mv to dst (s String, lc String) as select 'a' as s, toLowCardinality('b') as lc from src; insert into src values (1); -select s, lc from mv where not ignore(lc) settings allow_experimental_analyzer=0; -select s, lc from mv where not ignore(lc) settings allow_experimental_analyzer=1; +select s, lc from mv where not ignore(lc) settings enable_analyzer=0; +select s, lc from mv where not ignore(lc) settings enable_analyzer=1; diff --git a/tests/queries/0_stateless/03166_skip_indexes_vertical_merge_1.sql b/tests/queries/0_stateless/03166_skip_indexes_vertical_merge_1.sql index d3e3b38a3cb..96221f27e73 100644 --- a/tests/queries/0_stateless/03166_skip_indexes_vertical_merge_1.sql +++ b/tests/queries/0_stateless/03166_skip_indexes_vertical_merge_1.sql @@ -1,6 +1,6 @@ DROP TABLE IF EXISTS t_ind_merge_1; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; CREATE TABLE t_ind_merge_1 (a UInt64, b UInt64, c UInt64, d UInt64, INDEX idx_b b TYPE minmax) ENGINE = MergeTree diff --git a/tests/queries/0_stateless/03167_parametrized_view_with_cte.sql b/tests/queries/0_stateless/03167_parametrized_view_with_cte.sql index 1ac5540047a..ae6ab586415 100644 --- a/tests/queries/0_stateless/03167_parametrized_view_with_cte.sql +++ b/tests/queries/0_stateless/03167_parametrized_view_with_cte.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; CREATE OR REPLACE VIEW param_test AS SELECT {test_str:String} as s_result; WITH 'OK' AS s SELECT * FROM param_test(test_str=s); WITH (SELECT 123) AS s SELECT * FROM param_test(test_str=s); diff --git a/tests/queries/0_stateless/03168_query_log_privileges_not_empty.reference b/tests/queries/0_stateless/03168_query_log_privileges_not_empty.reference index e3ac97f9945..f4ada41d77d 100644 --- a/tests/queries/0_stateless/03168_query_log_privileges_not_empty.reference +++ b/tests/queries/0_stateless/03168_query_log_privileges_not_empty.reference @@ -1,4 +1,3 @@ -1 3168 8613 [] ['SELECT(a, b) ON default.d_03168_query_log'] [] [] diff --git a/tests/queries/0_stateless/03168_query_log_privileges_not_empty.sh b/tests/queries/0_stateless/03168_query_log_privileges_not_empty.sh index 9abc635a874..b817052e355 100755 --- a/tests/queries/0_stateless/03168_query_log_privileges_not_empty.sh +++ b/tests/queries/0_stateless/03168_query_log_privileges_not_empty.sh @@ -1,32 +1,28 @@ #!/usr/bin/env bash -# Tags: no-parallel CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -user_name="u_03168_query_log" -table_name="default.d_03168_query_log" +user_name="u_03168_query_log_${CLICKHOUSE_DATABASE}" +table_name="d_03168_query_log" test_query="select a, b from ${table_name}" -${CLICKHOUSE_CLIENT_BINARY} --query "drop user if exists ${user_name}" -${CLICKHOUSE_CLIENT_BINARY} --query "create user ${user_name}" -${CLICKHOUSE_CLIENT_BINARY} --query "drop table if exists ${table_name}" -${CLICKHOUSE_CLIENT_BINARY} --query "create table ${table_name} (a UInt64, b UInt64) order by a" +${CLICKHOUSE_CLIENT} --query "drop user if exists ${user_name}" +${CLICKHOUSE_CLIENT} --query "create user ${user_name}" +${CLICKHOUSE_CLIENT} --query "drop table if exists ${table_name}" +${CLICKHOUSE_CLIENT} --query "create table ${table_name} (a UInt64, b UInt64) order by a" +${CLICKHOUSE_CLIENT} --query "insert into table ${table_name} values (3168, 8613)" -${CLICKHOUSE_CLIENT_BINARY} --query "insert into table ${table_name} values (3168, 8613)" +${CLICKHOUSE_CLIENT} --user ${user_name} --query "${test_query}" 2>&1 >/dev/null | (grep -q "ACCESS_DENIED" || echo "Expected ACCESS_DENIED error not found") -error="$(${CLICKHOUSE_CLIENT_BINARY} --user ${user_name} --query "${test_query}" 2>&1 >/dev/null)" -echo "${error}" | grep -Fc "ACCESS_DENIED" +${CLICKHOUSE_CLIENT} --query "grant select(a, b) on ${table_name} to ${user_name}" +${CLICKHOUSE_CLIENT} --user ${user_name} --query "${test_query}" -${CLICKHOUSE_CLIENT_BINARY} --query "grant select(a, b) on ${table_name} to ${user_name}" +${CLICKHOUSE_CLIENT} --query "system flush logs" +${CLICKHOUSE_CLIENT} --query "select used_privileges, missing_privileges from system.query_log where query = '${test_query}' and type = 'ExceptionBeforeStart' and current_database = currentDatabase() order by event_time desc limit 1" +${CLICKHOUSE_CLIENT} --query "select used_privileges, missing_privileges from system.query_log where query = '${test_query}' and type = 'QueryStart' and current_database = currentDatabase() order by event_time desc limit 1" +${CLICKHOUSE_CLIENT} --query "select used_privileges, missing_privileges from system.query_log where query = '${test_query}' and type = 'QueryFinish' and current_database = currentDatabase() order by event_time desc limit 1" -${CLICKHOUSE_CLIENT_BINARY} --user ${user_name} --query "${test_query}" - -${CLICKHOUSE_CLIENT_BINARY} --query "system flush logs" -${CLICKHOUSE_CLIENT_BINARY} --query "select used_privileges, missing_privileges from system.query_log where query = '${test_query}' and type = 'ExceptionBeforeStart' and current_database = currentDatabase() order by event_time desc limit 1" -${CLICKHOUSE_CLIENT_BINARY} --query "select used_privileges, missing_privileges from system.query_log where query = '${test_query}' and type = 'QueryStart' and current_database = currentDatabase() order by event_time desc limit 1" -${CLICKHOUSE_CLIENT_BINARY} --query "select used_privileges, missing_privileges from system.query_log where query = '${test_query}' and type = 'QueryFinish' and current_database = currentDatabase() order by event_time desc limit 1" - -${CLICKHOUSE_CLIENT_BINARY} --query "drop table ${table_name}" -${CLICKHOUSE_CLIENT_BINARY} --query "drop user ${user_name}" +${CLICKHOUSE_CLIENT} --query "drop table ${table_name}" +${CLICKHOUSE_CLIENT} --query "drop user ${user_name}" diff --git a/tests/queries/0_stateless/03169_optimize_injective_functions_inside_uniq_crash.sql b/tests/queries/0_stateless/03169_optimize_injective_functions_inside_uniq_crash.sql index 50d99b851a6..5ab32415f1c 100644 --- a/tests/queries/0_stateless/03169_optimize_injective_functions_inside_uniq_crash.sql +++ b/tests/queries/0_stateless/03169_optimize_injective_functions_inside_uniq_crash.sql @@ -7,7 +7,7 @@ FROM FROM numbers(4096 * 100) GROUP BY k ) -SETTINGS allow_experimental_analyzer = 1, optimize_injective_functions_inside_uniq=0; +SETTINGS enable_analyzer = 1, optimize_injective_functions_inside_uniq=0; SELECT sum(u) FROM @@ -18,4 +18,4 @@ FROM FROM numbers(4096 * 100) GROUP BY k ) -SETTINGS allow_experimental_analyzer = 1, optimize_injective_functions_inside_uniq=1; +SETTINGS enable_analyzer = 1, optimize_injective_functions_inside_uniq=1; diff --git a/tests/queries/0_stateless/03169_time_virtual_column.sh b/tests/queries/0_stateless/03169_time_virtual_column.sh index fef1de8c6f2..b289f39accb 100755 --- a/tests/queries/0_stateless/03169_time_virtual_column.sh +++ b/tests/queries/0_stateless/03169_time_virtual_column.sh @@ -6,7 +6,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) echo "1,2" > $CLICKHOUSE_TEST_UNIQUE_NAME.csv sleep 1 -$CLICKHOUSE_LOCAL -nm -q " +$CLICKHOUSE_LOCAL -m -q " select _size, (dateDiff('millisecond', _time, now()) < 600000 AND dateDiff('millisecond', _time, now()) > 0) from file('$CLICKHOUSE_TEST_UNIQUE_NAME.csv'); " rm $CLICKHOUSE_TEST_UNIQUE_NAME.csv diff --git a/tests/queries/0_stateless/03170_part_offset_as_table_column.sql b/tests/queries/0_stateless/03170_part_offset_as_table_column.sql index 36cbc156744..7711457f23f 100644 --- a/tests/queries/0_stateless/03170_part_offset_as_table_column.sql +++ b/tests/queries/0_stateless/03170_part_offset_as_table_column.sql @@ -9,12 +9,12 @@ ORDER BY key; INSERT INTO test_table (key) SELECT number FROM numbers(10); -set allow_experimental_analyzer=0; +set enable_analyzer=0; SELECT * FROM test_table; -set allow_experimental_analyzer=1; +set enable_analyzer=1; SELECT * FROM test_table; diff --git a/tests/queries/0_stateless/03171_condition_pushdown.sql b/tests/queries/0_stateless/03171_condition_pushdown.sql index 9cfe41ce921..fcf5db886e9 100644 --- a/tests/queries/0_stateless/03171_condition_pushdown.sql +++ b/tests/queries/0_stateless/03171_condition_pushdown.sql @@ -1,5 +1,5 @@ -- This query succeeds only if it is correctly optimized. -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT * FROM (SELECT * FROM numbers(1e19)) AS t1, (SELECT * FROM numbers(1e19)) AS t2 WHERE t1.number IN (123, 456) AND t2.number = t1.number ORDER BY ALL; -- Still TODO: diff --git a/tests/queries/0_stateless/03171_function_to_subcolumns_fuzzer.sql b/tests/queries/0_stateless/03171_function_to_subcolumns_fuzzer.sql index f10019a78dd..53476c5bdd1 100644 --- a/tests/queries/0_stateless/03171_function_to_subcolumns_fuzzer.sql +++ b/tests/queries/0_stateless/03171_function_to_subcolumns_fuzzer.sql @@ -1,5 +1,5 @@ SET optimize_functions_to_subcolumns = 1; -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS t_func_to_subcolumns_map_2; diff --git a/tests/queries/0_stateless/03171_hashed_dictionary_short_circuit_bug_fix.sql b/tests/queries/0_stateless/03171_hashed_dictionary_short_circuit_bug_fix.sql index e1b5531a442..6d3a63dbadb 100644 --- a/tests/queries/0_stateless/03171_hashed_dictionary_short_circuit_bug_fix.sql +++ b/tests/queries/0_stateless/03171_hashed_dictionary_short_circuit_bug_fix.sql @@ -1,5 +1,3 @@ --- Tags: no-parallel - CREATE TABLE x ( hash_id UInt64, user_result Decimal(3, 2) ) ENGINE = Memory(); CREATE TABLE y ( hash_id UInt64, user_result DECIMAL(18, 6) ) ENGINE = Memory(); diff --git a/tests/queries/0_stateless/03172_dynamic_binary_serialization.sh b/tests/queries/0_stateless/03172_dynamic_binary_serialization.sh index 9b57e5c8718..b9bab2bd70b 100755 --- a/tests/queries/0_stateless/03172_dynamic_binary_serialization.sh +++ b/tests/queries/0_stateless/03172_dynamic_binary_serialization.sh @@ -5,7 +5,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$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 --allow_experimental_dynamic_type=1 -q "create table test (id UInt64, d Dynamic(max_types=254)) engine=Memory" $CLICKHOUSE_CLIENT -q "insert into test select 0, NULL" $CLICKHOUSE_CLIENT -q "insert into test select 1, materialize(42)::UInt8" @@ -58,6 +58,6 @@ $CLICKHOUSE_CLIENT -q "insert into test select 47, materialize([[(20, 20), (50, $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 "select * from test format RowBinary" | $CLICKHOUSE_LOCAL --allow_experimental_dynamic_type=1 --input-format RowBinary --structure 'id UInt64, d Dynamic(max_types=254)' -q "select d, dynamicType(d) from table order by id" $CLICKHOUSE_CLIENT -q "drop table test" diff --git a/tests/queries/0_stateless/03172_system_detached_tables.reference b/tests/queries/0_stateless/03172_system_detached_tables.reference index 83d1ff13942..2fce89e5980 100644 --- a/tests/queries/0_stateless/03172_system_detached_tables.reference +++ b/tests/queries/0_stateless/03172_system_detached_tables.reference @@ -1,11 +1,11 @@ database atomic tests -test03172_system_detached_tables test_table 0 -test03172_system_detached_tables test_table_perm 1 -test03172_system_detached_tables test_table 0 -test03172_system_detached_tables test_table_perm 1 -test03172_system_detached_tables test_table 0 +default_atomic test_table 0 +default_atomic test_table_perm 1 +default_atomic test_table 0 +default_atomic test_table_perm 1 +default_atomic test_table 0 ----------------------- database lazy tests -before attach test03172_system_detached_tables_lazy test_table 0 -before attach test03172_system_detached_tables_lazy test_table_perm 1 +before attach default_lazy test_table 0 +before attach default_lazy test_table_perm 1 DROP TABLE diff --git a/tests/queries/0_stateless/03172_system_detached_tables.sh b/tests/queries/0_stateless/03172_system_detached_tables.sh new file mode 100755 index 00000000000..47775abcc45 --- /dev/null +++ b/tests/queries/0_stateless/03172_system_detached_tables.sh @@ -0,0 +1,68 @@ +#!/bin/bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +DATABASE_ATOMIC="${CLICKHOUSE_DATABASE}_atomic" +DATABASE_LAZY="${CLICKHOUSE_DATABASE}_lazy" + +$CLICKHOUSE_CLIENT --multiquery " + +SELECT 'database atomic tests'; +DROP DATABASE IF EXISTS ${DATABASE_ATOMIC}; +CREATE DATABASE IF NOT EXISTS ${DATABASE_ATOMIC} ENGINE=Atomic; + +CREATE TABLE ${DATABASE_ATOMIC}.test_table (n Int64) ENGINE=MergeTree ORDER BY n; +SELECT * FROM system.detached_tables WHERE database='${DATABASE_ATOMIC}'; + +DETACH TABLE ${DATABASE_ATOMIC}.test_table; +SELECT database, table, is_permanently FROM system.detached_tables WHERE database='${DATABASE_ATOMIC}'; + +ATTACH TABLE ${DATABASE_ATOMIC}.test_table; + +CREATE TABLE ${DATABASE_ATOMIC}.test_table_perm (n Int64) ENGINE=MergeTree ORDER BY n; +SELECT * FROM system.detached_tables WHERE database='${DATABASE_ATOMIC}'; + +DETACH TABLE ${DATABASE_ATOMIC}.test_table_perm PERMANENTLY; +SELECT database, table, is_permanently FROM system.detached_tables WHERE database='${DATABASE_ATOMIC}'; + +DETACH TABLE ${DATABASE_ATOMIC}.test_table SYNC; +SELECT database, table, is_permanently FROM system.detached_tables WHERE database='${DATABASE_ATOMIC}'; + +SELECT database, table, is_permanently FROM system.detached_tables WHERE database='${DATABASE_ATOMIC}' AND table='test_table'; + +DROP DATABASE ${DATABASE_ATOMIC} SYNC; + +" + +$CLICKHOUSE_CLIENT --multiquery " + +SELECT '-----------------------'; +SELECT 'database lazy tests'; + +DROP DATABASE IF EXISTS ${DATABASE_LAZY}; +CREATE DATABASE ${DATABASE_LAZY} Engine=Lazy(10); + +CREATE TABLE ${DATABASE_LAZY}.test_table (number UInt64) engine=Log; +INSERT INTO ${DATABASE_LAZY}.test_table SELECT * FROM numbers(100); +DETACH TABLE ${DATABASE_LAZY}.test_table; + +CREATE TABLE ${DATABASE_LAZY}.test_table_perm (number UInt64) engine=Log; +INSERT INTO ${DATABASE_LAZY}.test_table_perm SELECT * FROM numbers(100); +DETACH table ${DATABASE_LAZY}.test_table_perm PERMANENTLY; + +SELECT 'before attach', database, table, is_permanently FROM system.detached_tables WHERE database='${DATABASE_LAZY}'; + +ATTACH TABLE ${DATABASE_LAZY}.test_table; +ATTACH TABLE ${DATABASE_LAZY}.test_table_perm; + +SELECT 'after attach', database, table, is_permanently FROM system.detached_tables WHERE database='${DATABASE_LAZY}'; + +SELECT 'DROP TABLE'; +DROP TABLE ${DATABASE_LAZY}.test_table SYNC; +DROP TABLE ${DATABASE_LAZY}.test_table_perm SYNC; + +DROP DATABASE ${DATABASE_LAZY} SYNC; + +" diff --git a/tests/queries/0_stateless/03172_system_detached_tables.sql b/tests/queries/0_stateless/03172_system_detached_tables.sql deleted file mode 100644 index 1a3c2d7cc0f..00000000000 --- a/tests/queries/0_stateless/03172_system_detached_tables.sql +++ /dev/null @@ -1,53 +0,0 @@ --- 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_forbid_qualify.sql b/tests/queries/0_stateless/03173_forbid_qualify.sql index d8cb2bad2ea..0a41385c52f 100644 --- a/tests/queries/0_stateless/03173_forbid_qualify.sql +++ b/tests/queries/0_stateless/03173_forbid_qualify.sql @@ -4,8 +4,8 @@ create table test_qualify (number Int64) ENGINE = MergeTree ORDER BY (number); insert into test_qualify SELECT * FROM numbers(100); select count() from test_qualify; -- 100 -select * from test_qualify qualify row_number() over (order by number) = 50 SETTINGS allow_experimental_analyzer = 1; -- 49 -select * from test_qualify qualify row_number() over (order by number) = 50 SETTINGS allow_experimental_analyzer = 0; -- { serverError NOT_IMPLEMENTED } +select * from test_qualify qualify row_number() over (order by number) = 50 SETTINGS enable_analyzer = 1; -- 49 +select * from test_qualify qualify row_number() over (order by number) = 50 SETTINGS enable_analyzer = 0; -- { serverError NOT_IMPLEMENTED } delete from test_qualify where number in (select number from test_qualify qualify row_number() over (order by number) = 50); -- { serverError UNFINISHED } select count() from test_qualify; -- 100 diff --git a/tests/queries/0_stateless/03173_parallel_replicas_join_bug.sh b/tests/queries/0_stateless/03173_parallel_replicas_join_bug.sh index 20a29e2734e..af702569794 100755 --- a/tests/queries/0_stateless/03173_parallel_replicas_join_bug.sh +++ b/tests/queries/0_stateless/03173_parallel_replicas_join_bug.sh @@ -5,7 +5,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -nq " +$CLICKHOUSE_CLIENT -q " CREATE TABLE ids (id UUID, whatever String) Engine=MergeTree ORDER BY tuple(); INSERT INTO ids VALUES ('a1451105-722e-4fe7-bfaa-65ad2ae249c2', 'whatever'); @@ -17,7 +17,7 @@ $CLICKHOUSE_CLIENT -nq " " $CLICKHOUSE_CLIENT -nq " -SET allow_experimental_analyzer = 1, cluster_for_parallel_replicas = 'parallel_replicas', max_parallel_replicas = 10, allow_experimental_parallel_reading_from_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, max_threads = 1; +SET enable_analyzer = 1, cluster_for_parallel_replicas = 'parallel_replicas', max_parallel_replicas = 10, allow_experimental_parallel_reading_from_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, max_threads = 1; SELECT id, diff --git a/tests/queries/0_stateless/03173_row_binary_and_native_with_binary_encoded_types.reference b/tests/queries/0_stateless/03173_row_binary_and_native_with_binary_encoded_types.reference index 1ba147f9627..7de0804e0f2 100644 --- a/tests/queries/0_stateless/03173_row_binary_and_native_with_binary_encoded_types.reference +++ b/tests/queries/0_stateless/03173_row_binary_and_native_with_binary_encoded_types.reference @@ -110,5 +110,5 @@ true Bool [{42:(1,[(2,{1:2})])}] Dynamic [{42:(1,[(2,{1:2})])}] Dynamic(max_types=10) [{42:(1,[(2,{1:2})])}] Dynamic(max_types=10) -[{42:(1,[(2,{1:2})])}] Dynamic(max_types=255) -[{42:(1,[(2,{1:2})])}] Dynamic(max_types=255) +[{42:(1,[(2,{1:2})])}] Dynamic(max_types=254) +[{42:(1,[(2,{1:2})])}] Dynamic(max_types=254) diff --git a/tests/queries/0_stateless/03173_row_binary_and_native_with_binary_encoded_types.sh b/tests/queries/0_stateless/03173_row_binary_and_native_with_binary_encoded_types.sh index 0c585d36348..1e674a29072 100755 --- a/tests/queries/0_stateless/03173_row_binary_and_native_with_binary_encoded_types.sh +++ b/tests/queries/0_stateless/03173_row_binary_and_native_with_binary_encoded_types.sh @@ -6,8 +6,8 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) 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" + $CLICKHOUSE_LOCAL --stacktrace --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 --stacktrace --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" @@ -66,4 +66,4 @@ 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)" +test "materialize([map(42, tuple(1, [tuple(2, map(1, 2))]))])::Dynamic(max_types=254)" diff --git a/tests/queries/0_stateless/03174_exact_rows_before_aggregation.reference b/tests/queries/0_stateless/03174_exact_rows_before_aggregation.reference new file mode 100644 index 00000000000..5f7fedbbcb3 --- /dev/null +++ b/tests/queries/0_stateless/03174_exact_rows_before_aggregation.reference @@ -0,0 +1,548 @@ +{ + "meta": + [ + { + "name": "i", + "type": "Int32" + } + ], + + "data": + [ + [0], + [1], + [2], + [3], + [4], + [5], + [6], + [7], + [8], + [9] + ], + + "rows": 10, + + "rows_before_aggregation": 10 +} + + + + + + i + Int32 + + + + + + 0 + + + 1 + + + 2 + + + 3 + + + 4 + + + 5 + + + 6 + + + 7 + + + 8 + + + 9 + + + 10 + 10 + +{ + "meta": + [ + { + "name": "i", + "type": "Int32" + } + ], + + "data": + [ + [10], + [11], + [12] + ], + + "rows": 3, + + "rows_before_aggregation": 3 +} +{ + "meta": + [ + { + "name": "i", + "type": "Int32" + } + ], + + "data": + [ + [0], + [1], + [2], + [3], + [4], + [5], + [6], + [7], + [8], + [9], + [10], + [11], + [12], + [13], + [14], + [15], + [16], + [17], + [18], + [19] + ], + + "rows": 20, + + "rows_before_aggregation": 20 +} +{ + "meta": + [ + { + "name": "max(i)", + "type": "Int32" + } + ], + + "data": + [ + [19] + ], + + "rows": 1, + + "rows_before_limit_at_least": 1, + + "rows_before_aggregation": 20 +} +{ + "meta": + [ + { + "name": "i", + "type": "Int32" + } + ], + + "data": + [ + [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] + ], + + "rows": 30, + + "rows_before_limit_at_least": 60, + + "rows_before_aggregation": 60 +} +{ + "meta": + [ + { + "name": "i", + "type": "Int32" + } + ], + + "data": + [ + [0], + [1], + [2], + [3], + [4], + [5], + [6], + [7], + [8], + [9], + [10], + [11], + [12], + [13], + [14], + [15], + [16], + [17], + [18], + [19] + ], + + "rows": 20, + + "rows_before_limit_at_least": 40, + + "rows_before_aggregation": 40 +} +{ + "meta": + [ + { + "name": "i", + "type": "Int32" + } + ], + + "data": + [ + [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] + ], + + "rows": 30, + + "rows_before_limit_at_least": 30, + + "rows_before_aggregation": 60 +} +{ + "meta": + [ + { + "name": "i", + "type": "Int32" + } + ], + + "data": + [ + [0], + [1], + [2], + [3], + [4], + [5], + [6], + [7], + [8], + [9], + [10], + [11], + [12], + [13], + [14], + [15], + [16], + [17], + [18], + [19] + ], + + "rows": 20, + + "rows_before_limit_at_least": 20, + + "rows_before_aggregation": 40 +} +{ + "meta": + [ + { + "name": "max(i)", + "type": "Int32" + } + ], + + "data": + [ + [19] + ], + + "rows": 1, + + "rows_before_limit_at_least": 1, + + "rows_before_aggregation": 40 +} +{ + "meta": + [ + { + "name": "i", + "type": "Int32" + } + ], + + "data": + [ + [0], + [1], + [2], + [3], + [4], + [5], + [6], + [7], + [8], + [9] + ], + + "rows": 10, + + "rows_before_limit_at_least": 10, + + "rows_before_aggregation": 20 +} +{ + "meta": + [ + { + "name": "i", + "type": "Int32" + } + ], + + "data": + [ + [0], + [1], + [2], + [3], + [4], + [5], + [6], + [7], + [8], + [9] + ], + + "rows": 10, + + "rows_before_limit_at_least": 10, + + "rows_before_aggregation": 20 +} +{ + "meta": + [ + { + "name": "i", + "type": "Int32" + } + ], + + "data": + [ + [0], + [1], + [2], + [3], + [4], + [5], + [6], + [7], + [8], + [9] + ], + + "rows": 10, + + "rows_before_aggregation": 10 +} +{ + "meta": + [ + { + "name": "max(i)", + "type": "Int32" + } + ], + + "data": + [ + [19] + ], + + "rows": 1, + + "rows_before_limit_at_least": 1, + + "rows_before_aggregation": 20 +} +{ + "meta": + [ + { + "name": "i", + "type": "Int32" + } + ], + + "data": + [ + [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] + ], + + "rows": 30, + + "rows_before_limit_at_least": 60, + + "rows_before_aggregation": 60 +} +{ + "meta": + [ + { + "name": "i", + "type": "Int32" + } + ], + + "data": + [ + [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] + ], + + "rows": 30, + + "rows_before_limit_at_least": 30, + + "rows_before_aggregation": 60 +} diff --git a/tests/queries/0_stateless/03174_exact_rows_before_aggregation.sql b/tests/queries/0_stateless/03174_exact_rows_before_aggregation.sql new file mode 100644 index 00000000000..f9fd4ef5a7b --- /dev/null +++ b/tests/queries/0_stateless/03174_exact_rows_before_aggregation.sql @@ -0,0 +1,47 @@ +-- Tags: no-parallel, no-random-merge-tree-settings + +set rows_before_aggregation = 1, exact_rows_before_limit = 1, output_format_write_statistics = 0, max_block_size = 100; + +drop table if exists test; + +create table test (i int) engine MergeTree order by tuple(); +insert into test select arrayJoin(range(10000)); + +select * from test where i < 10 group by i order by i FORMAT JSONCompact; +select * from test where i < 10 group by i order by i FORMAT XML; +select * from test group by i having i in (10, 11, 12) order by i FORMAT JSONCompact; +select * from test where i < 20 group by i order by i FORMAT JSONCompact; +select max(i) from test where i < 20 limit 1 FORMAT JSONCompact; + +set prefer_localhost_replica = 0; +select * from cluster(test_cluster_two_shards, currentDatabase(), test) where i < 30 group by i order by i FORMAT JSONCompact; +select * from cluster(test_cluster_two_shards, currentDatabase(), test) where i < 20 group by i order by i FORMAT JSONCompact; + +set prefer_localhost_replica = 1; +select * from cluster(test_cluster_two_shards, currentDatabase(), test) where i < 30 group by i order by i FORMAT JSONCompact; +select * from cluster(test_cluster_two_shards, currentDatabase(), test) where i < 20 group by i order by i FORMAT JSONCompact; + +select max(i) from cluster(test_cluster_two_shards, currentDatabase(), test) where i < 20 FORMAT JSONCompact; + +select * from (select * from cluster(test_cluster_two_shards, currentDatabase(), test) where i < 10) group by i order by i limit 10 FORMAT JSONCompact; +set prefer_localhost_replica = 0; +select * from (select * from cluster(test_cluster_two_shards, currentDatabase(), test) where i < 10) group by i order by i limit 10 FORMAT JSONCompact; + +drop table if exists test; + +create table test (i int) engine MergeTree order by i; + +insert into test select arrayJoin(range(10000)); + +set optimize_aggregation_in_order=1; + +select * from test where i < 10 group by i order by i FORMAT JSONCompact; +select max(i) from test where i < 20 limit 1 FORMAT JSONCompact; + +set prefer_localhost_replica = 0; +select * from cluster(test_cluster_two_shards, currentDatabase(), test) where i < 30 group by i order by i FORMAT JSONCompact; + +set prefer_localhost_replica = 1; +select * from cluster(test_cluster_two_shards, currentDatabase(), test) where i < 30 group by i order by i FORMAT JSONCompact; + +drop table if exists test; diff --git a/tests/queries/0_stateless/03174_merge_join_bug.sql b/tests/queries/0_stateless/03174_merge_join_bug.sql index ab4cb6cd4a9..ab3c384765d 100644 --- a/tests/queries/0_stateless/03174_merge_join_bug.sql +++ b/tests/queries/0_stateless/03174_merge_join_bug.sql @@ -1,6 +1,6 @@ -- Tags: no-random-settings -SET allow_experimental_analyzer=1, join_algorithm = 'full_sorting_merge'; +SET enable_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); diff --git a/tests/queries/0_stateless/03174_projection_deduplicate.sql b/tests/queries/0_stateless/03174_projection_deduplicate.sql index 46222b69dc7..f43f0a1f236 100644 --- a/tests/queries/0_stateless/03174_projection_deduplicate.sql +++ b/tests/queries/0_stateless/03174_projection_deduplicate.sql @@ -17,7 +17,7 @@ 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 } +OPTIMIZE TABLE test_projection_deduplicate DEDUPLICATE; -- { serverError SUPPORT_IS_DISABLED } SELECT * FROM test_projection_deduplicate; diff --git a/tests/queries/0_stateless/03198_dynamic_read_subcolumns.sql b/tests/queries/0_stateless/03198_dynamic_read_subcolumns.sql index bbe701f022b..7bbe2d3c533 100644 --- a/tests/queries/0_stateless/03198_dynamic_read_subcolumns.sql +++ b/tests/queries/0_stateless/03198_dynamic_read_subcolumns.sql @@ -1,4 +1,5 @@ -- Tags: no-random-settings, no-object-storage, no-parallel +-- no-parallel: Running `DROP MARK CACHE` can have a big impact on other concurrent tests -- 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; 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 index ce53f467823..583257d8fd3 100755 --- 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 @@ -5,25 +5,25 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CUR_DIR"/../shell_config.sh echo -e 'a,b,c\n1,2,3' > $CLICKHOUSE_TEST_UNIQUE_NAME.csv -$CLICKHOUSE_LOCAL -nm -q " +$CLICKHOUSE_LOCAL -m -q " DESC file('$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 " +$CLICKHOUSE_LOCAL -m -q " DESC file('$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 " +$CLICKHOUSE_LOCAL -m -q " DESC file('$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 " +$CLICKHOUSE_LOCAL -m -q " DESC file('$CLICKHOUSE_TEST_UNIQUE_NAME.tsv') SETTINGS input_format_tsv_skip_first_lines=1; DESC file('$CLICKHOUSE_TEST_UNIQUE_NAME.tsv') SETTINGS input_format_tsv_skip_first_lines=0; SELECT count() from system.schema_inference_cache where format = 'TSV' and additional_format_info like '%skip_first_lines%';" diff --git a/tests/queries/0_stateless/03198_unload_primary_key_outdated.sh b/tests/queries/0_stateless/03198_unload_primary_key_outdated.sh index 4f217935123..c759cc34425 100755 --- a/tests/queries/0_stateless/03198_unload_primary_key_outdated.sh +++ b/tests/queries/0_stateless/03198_unload_primary_key_outdated.sh @@ -4,7 +4,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_CLIENT -n " +$CLICKHOUSE_CLIENT " DROP TABLE IF EXISTS t_unload_primary_key; CREATE TABLE t_unload_primary_key (a UInt64, b UInt64) @@ -26,7 +26,7 @@ for _ in {1..100}; do sleep 0.3 done -$CLICKHOUSE_CLIENT -n " +$CLICKHOUSE_CLIENT " SELECT name, active, primary_key_bytes_in_memory FROM system.parts WHERE database = '$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.sh b/tests/queries/0_stateless/03199_dictionary_table_access.sh index 952b466b5da..14f017c7fbc 100755 --- a/tests/queries/0_stateless/03199_dictionary_table_access.sh +++ b/tests/queries/0_stateless/03199_dictionary_table_access.sh @@ -8,7 +8,7 @@ username="user_${CLICKHOUSE_TEST_UNIQUE_NAME}" dictname="dict_${CLICKHOUSE_TEST_UNIQUE_NAME}" dicttablename="dict_table_${CLICKHOUSE_TEST_UNIQUE_NAME}" -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --query " CREATE DICTIONARY IF NOT EXISTS ${dictname} ( id UInt64, @@ -26,15 +26,15 @@ ${CLICKHOUSE_CLIENT} -nm --query " SELECT * FROM ${dicttablename}; " -$CLICKHOUSE_CLIENT -nm --user="${username}" --query " +$CLICKHOUSE_CLIENT -m --user="${username}" --query " SELECT * FROM ${dictname}; " 2>&1 | grep -o ACCESS_DENIED | uniq -$CLICKHOUSE_CLIENT -nm --user="${username}" --query " +$CLICKHOUSE_CLIENT -m --user="${username}" --query " SELECT * FROM ${dicttablename}; " 2>&1 | grep -o ACCESS_DENIED | uniq -${CLICKHOUSE_CLIENT} -nm --query " +${CLICKHOUSE_CLIENT} -m --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_join_with_materialized_column.sql b/tests/queries/0_stateless/03199_join_with_materialized_column.sql index 8c53c5b3e66..fef171cb9d1 100644 --- a/tests/queries/0_stateless/03199_join_with_materialized_column.sql +++ b/tests/queries/0_stateless/03199_join_with_materialized_column.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; DROP TABLE IF EXISTS table_with_materialized; CREATE TABLE table_with_materialized (col String MATERIALIZED 'A') ENGINE = Memory; diff --git a/tests/queries/0_stateless/03199_json_extract_dynamic.reference b/tests/queries/0_stateless/03199_json_extract_dynamic.reference index 759b7763cd1..955106946ea 100644 --- a/tests/queries/0_stateless/03199_json_extract_dynamic.reference +++ b/tests/queries/0_stateless/03199_json_extract_dynamic.reference @@ -12,7 +12,7 @@ Hello String [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 00:00:00','2020-01-01 00:00:00'] Array(Nullable(DateTime)) ['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)) diff --git a/tests/queries/0_stateless/03199_queries_with_new_analyzer.sql b/tests/queries/0_stateless/03199_queries_with_new_analyzer.sql index c32d7524492..d400a025f76 100644 --- a/tests/queries/0_stateless/03199_queries_with_new_analyzer.sql +++ b/tests/queries/0_stateless/03199_queries_with_new_analyzer.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer=1; +SET enable_analyzer=1; SELECT *, ngramMinHash(*) AS minhash, mortonEncode(untuple(ngramMinHash(*))) AS z FROM (SELECT toString(number) FROM numbers(10)) @@ -38,4 +38,3 @@ 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/03200_memory_engine_alter_dynamic.reference b/tests/queries/0_stateless/03200_memory_engine_alter_dynamic.reference index 6d2c1334d6e..8d2470dea44 100644 --- a/tests/queries/0_stateless/03200_memory_engine_alter_dynamic.reference +++ b/tests/queries/0_stateless/03200_memory_engine_alter_dynamic.reference @@ -1,10 +1,10 @@ -\N -\N -\N -\N -\N -\N -\N -\N -\N -\N +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 diff --git a/tests/queries/0_stateless/03200_memory_engine_alter_dynamic.sql b/tests/queries/0_stateless/03200_memory_engine_alter_dynamic.sql index a01a595dbb5..939b49e1599 100644 --- a/tests/queries/0_stateless/03200_memory_engine_alter_dynamic.sql +++ b/tests/queries/0_stateless/03200_memory_engine_alter_dynamic.sql @@ -1,7 +1,8 @@ set allow_experimental_dynamic_type=1; +drop table if exists test; 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; - +alter table test modify column d Dynamic(max_types=0); +select d.UInt64 from test settings enable_analyzer=1; +select d.UInt64 from test settings enable_analyzer=0; +drop table test; diff --git a/tests/queries/0_stateless/03200_subcolumns_join_use_nulls.sql b/tests/queries/0_stateless/03200_subcolumns_join_use_nulls.sql index 2dd0a37657d..6777a753490 100644 --- a/tests/queries/0_stateless/03200_subcolumns_join_use_nulls.sql +++ b/tests/queries/0_stateless/03200_subcolumns_join_use_nulls.sql @@ -8,6 +8,6 @@ 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; +SETTINGS enable_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.sh b/tests/queries/0_stateless/03201_avro_negative_block_size_arrays.sh index dcecd7b3bea..f7101989377 100755 --- a/tests/queries/0_stateless/03201_avro_negative_block_size_arrays.sh +++ b/tests/queries/0_stateless/03201_avro_negative_block_size_arrays.sh @@ -1,5 +1,6 @@ #!/usr/bin/env bash -# Tags: no-parallel, no-fasttest +# Tags: no-fasttest +# no-fasttest: Requires libraries set -e 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 index 24369fd6497..b791f328da4 100644 --- a/tests/queries/0_stateless/03201_sumIf_to_countIf_return_type.sql +++ b/tests/queries/0_stateless/03201_sumIf_to_countIf_return_type.sql @@ -1,2 +1,2 @@ -SET allow_experimental_analyzer = 1; +SET enable_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/03203_drop_detached_partition_all.reference b/tests/queries/0_stateless/03203_drop_detached_partition_all.reference new file mode 100644 index 00000000000..c0f52d1d898 --- /dev/null +++ b/tests/queries/0_stateless/03203_drop_detached_partition_all.reference @@ -0,0 +1,5 @@ +1 1 +2 2 +3 3 +3 +0 diff --git a/tests/queries/0_stateless/03203_drop_detached_partition_all.sql b/tests/queries/0_stateless/03203_drop_detached_partition_all.sql new file mode 100644 index 00000000000..e29eb4ae36b --- /dev/null +++ b/tests/queries/0_stateless/03203_drop_detached_partition_all.sql @@ -0,0 +1,8 @@ +DROP TABLE IF EXISTS t_03203; +CREATE TABLE t_03203 (p UInt64, v UInt64) ENGINE = MergeTree PARTITION BY p ORDER BY v; +INSERT INTO t_03203 VALUES (1, 1), (2, 2), (3, 3); +SELECT * FROM t_03203 ORDER BY p, v; +ALTER TABLE t_03203 DETACH PARTITION ALL; +SELECT count() FROM system.detached_parts WHERE database = currentDatabase() AND table = 't_03203'; +ALTER TABLE t_03203 DROP DETACHED PARTITION ALL SETTINGS allow_drop_detached = 1; +SELECT count() FROM system.detached_parts WHERE database = currentDatabase() AND table = 't_03203'; diff --git a/tests/queries/0_stateless/03203_fill_missed_subcolumns.reference b/tests/queries/0_stateless/03203_fill_missed_subcolumns.reference new file mode 100644 index 00000000000..872eb7da3c8 --- /dev/null +++ b/tests/queries/0_stateless/03203_fill_missed_subcolumns.reference @@ -0,0 +1,31 @@ +0 +2 +4 +6 +8 +0 +2 +4 +6 +8 +1 ['aaa',NULL] [NULL,NULL] +2 ['ccc'] [NULL] +3 [NULL] [NULL] +4 [NULL,'bbb'] ['ddd',NULL] +5 [NULL] [NULL] +1 2 2 +2 1 1 +3 1 1 +4 2 2 +5 1 1 +1 [0,1] [1,1] +2 [0] [1] +3 [1] [1] +4 [1,0] [0,1] +5 [1] [1] +1 ('foo','bar') [1,NULL,3] +2 ('aaa','bbb') [1,NULL,3] +3 ('ccc','ddd') [4,5,6] +1 foo bar 3 [0,1,0] +2 foo bar 3 [0,1,0] +3 ccc ddd 3 [0,0,0] diff --git a/tests/queries/0_stateless/03203_fill_missed_subcolumns.sql b/tests/queries/0_stateless/03203_fill_missed_subcolumns.sql new file mode 100644 index 00000000000..2789c9de35c --- /dev/null +++ b/tests/queries/0_stateless/03203_fill_missed_subcolumns.sql @@ -0,0 +1,47 @@ +DROP TABLE IF EXISTS t_missed_subcolumns; + +-- .null subcolumn + +CREATE TABLE t_missed_subcolumns (x UInt32) ENGINE = MergeTree ORDER BY tuple(); +INSERT INTO t_missed_subcolumns SELECT * FROM numbers(10); + +ALTER TABLE t_missed_subcolumns ADD COLUMN `y` Nullable(UInt32); + +INSERT INTO t_missed_subcolumns SELECT number, if(number % 2, NULL, number) FROM numbers(10); + +SELECT x FROM t_missed_subcolumns WHERE y IS NOT NULL SETTINGS optimize_functions_to_subcolumns = 1; +SELECT x FROM t_missed_subcolumns WHERE y IS NOT NULL SETTINGS optimize_functions_to_subcolumns = 0; + +DROP TABLE IF EXISTS t_missed_subcolumns; + +-- .null and .size0 subcolumn in array + +CREATE TABLE t_missed_subcolumns (id UInt64, `n.a` Array(Nullable(String))) ENGINE = MergeTree ORDER BY id; + +INSERT INTO t_missed_subcolumns VALUES (1, ['aaa', NULL]) (2, ['ccc']) (3, [NULL]); +ALTER TABLE t_missed_subcolumns ADD COLUMN `n.b` Array(Nullable(String)); +INSERT INTO t_missed_subcolumns VALUES (4, [NULL, 'bbb'], ['ddd', NULL]), (5, [NULL], [NULL]); + +SELECT id, n.a, n.b FROM t_missed_subcolumns ORDER BY id; +SELECT id, n.a.size0, n.b.size0 FROM t_missed_subcolumns ORDER BY id; +SELECT id, n.a.null, n.b.null FROM t_missed_subcolumns ORDER BY id; + +DROP TABLE IF EXISTS t_missed_subcolumns; + +-- subcolumns and custom defaults + +CREATE TABLE t_missed_subcolumns (id UInt64) ENGINE = MergeTree ORDER BY id; +SYSTEM STOP MERGES t_missed_subcolumns; + +INSERT INTO t_missed_subcolumns VALUES (1); + +ALTER TABLE t_missed_subcolumns ADD COLUMN t Tuple(a String, b String) DEFAULT ('foo', 'bar'); +INSERT INTO t_missed_subcolumns VALUES (2, ('aaa', 'bbb')); + +ALTER TABLE t_missed_subcolumns ADD COLUMN arr Array(Nullable(UInt64)) DEFAULT [1, NULL, 3]; +INSERT INTO t_missed_subcolumns VALUES (3, ('ccc', 'ddd'), [4, 5, 6]); + +SELECT id, t, arr FROM t_missed_subcolumns ORDER BY id; +SELECT id, t.a, t.b, arr.size0, arr.null FROM t_missed_subcolumns ORDER BY id; + +DROP TABLE t_missed_subcolumns; diff --git a/tests/queries/0_stateless/03203_hive_style_partitioning.reference b/tests/queries/0_stateless/03203_hive_style_partitioning.reference new file mode 100644 index 00000000000..a4a2e48e046 --- /dev/null +++ b/tests/queries/0_stateless/03203_hive_style_partitioning.reference @@ -0,0 +1,120 @@ +TESTING THE FILE HIVE PARTITIONING +first last Elizabeth +Jorge Frank Elizabeth +Hunter Moreno Elizabeth +Esther Guzman Elizabeth +Dennis Stephens Elizabeth +Nettie Franklin Elizabeth +Stanley Gibson Elizabeth +Eugenia Greer Elizabeth +Jeffery Delgado Elizabeth +Clara Cross Elizabeth +Elizabeth Gordon Elizabeth +Eva Schmidt Elizabeth Schmidt +Samuel Schmidt Elizabeth Schmidt +Eva Schmidt Elizabeth +Samuel Schmidt Elizabeth +Elizabeth Gordon Elizabeth Gordon +Elizabeth Gordon Elizabeth +Elizabeth Gordon Elizabeth Gordon +Elizabeth Gordon Elizabeth +first last Elizabeth +Jorge Frank Elizabeth +Hunter Moreno Elizabeth +Esther Guzman Elizabeth +Dennis Stephens Elizabeth +Nettie Franklin Elizabeth +Stanley Gibson Elizabeth +Eugenia Greer Elizabeth +Jeffery Delgado Elizabeth +Clara Cross Elizabeth +Elizabeth Gordon Elizabeth +42 2020-01-01 +[1,2,3] 42.42 +Array(Int64) LowCardinality(Float64) +101 +2070 +4081 +2070 +2070 +1 +1 +TESTING THE URL PARTITIONING +first last Elizabeth +Jorge Frank Elizabeth +Hunter Moreno Elizabeth +Esther Guzman Elizabeth +Dennis Stephens Elizabeth +Nettie Franklin Elizabeth +Stanley Gibson Elizabeth +Eugenia Greer Elizabeth +Jeffery Delgado Elizabeth +Clara Cross Elizabeth +Elizabeth Gordon Elizabeth +Eva Schmidt Elizabeth Schmidt +Samuel Schmidt Elizabeth Schmidt +Eva Schmidt Elizabeth +Samuel Schmidt Elizabeth +Elizabeth Gordon Elizabeth Gordon +Elizabeth Gordon Elizabeth +Elizabeth Gordon Elizabeth Gordon +Elizabeth Gordon Elizabeth +first last Elizabeth +Jorge Frank Elizabeth +Hunter Moreno Elizabeth +Esther Guzman Elizabeth +Dennis Stephens Elizabeth +Nettie Franklin Elizabeth +Stanley Gibson Elizabeth +Eugenia Greer Elizabeth +Jeffery Delgado Elizabeth +Clara Cross Elizabeth +1 +TESTING THE S3 PARTITIONING +first last Elizabeth +Jorge Frank Elizabeth +Hunter Moreno Elizabeth +Esther Guzman Elizabeth +Dennis Stephens Elizabeth +Nettie Franklin Elizabeth +Stanley Gibson Elizabeth +Eugenia Greer Elizabeth +Jeffery Delgado Elizabeth +Clara Cross Elizabeth +Elizabeth Gordon Elizabeth +Eva Schmidt Elizabeth Schmidt +Samuel Schmidt Elizabeth Schmidt +Eva Schmidt Elizabeth +Samuel Schmidt Elizabeth +Elizabeth Gordon Elizabeth Gordon +Elizabeth Gordon Elizabeth +Elizabeth Gordon Elizabeth Gordon +Elizabeth Gordon Elizabeth +first last Elizabeth +Jorge Frank Elizabeth +Hunter Moreno Elizabeth +Esther Guzman Elizabeth +Dennis Stephens Elizabeth +Nettie Franklin Elizabeth +Stanley Gibson Elizabeth +Eugenia Greer Elizabeth +Jeffery Delgado Elizabeth +Clara Cross Elizabeth +Elizabeth Gordon Elizabeth +OK +TESTING THE S3CLUSTER PARTITIONING +first last Elizabeth +Jorge Frank Elizabeth +Hunter Moreno Elizabeth +Esther Guzman Elizabeth +Dennis Stephens Elizabeth +Nettie Franklin Elizabeth +Stanley Gibson Elizabeth +Eugenia Greer Elizabeth +Jeffery Delgado Elizabeth +Clara Cross Elizabeth +Elizabeth Gordon Elizabeth +Eva Schmidt Elizabeth Schmidt +Samuel Schmidt Elizabeth Schmidt +Eva Schmidt Elizabeth +Samuel Schmidt Elizabeth diff --git a/tests/queries/0_stateless/03203_hive_style_partitioning.sh b/tests/queries/0_stateless/03203_hive_style_partitioning.sh new file mode 100755 index 00000000000..db1f073d736 --- /dev/null +++ b/tests/queries/0_stateless/03203_hive_style_partitioning.sh @@ -0,0 +1,136 @@ +#!/usr/bin/env bash +# Tags: no-fasttest + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +$CLICKHOUSE_LOCAL -q "SELECT 'TESTING THE FILE HIVE PARTITIONING'" + + +$CLICKHOUSE_LOCAL -n -q """ +set use_hive_partitioning = 1; + +SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; + +SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/sample.parquet') WHERE column0 = _column0; + +SELECT *, _column0, _column1 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; +SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; + +SELECT *, _column0, _column1 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; +SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; + +SELECT *, _column0, _column1 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column1 = _column1; +SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column1 = _column1; + +SELECT *, _column0, _column1 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; +SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; + +SELECT *, _non_existing_column FROM file('$CURDIR/data_hive/partitioning/non_existing_column=Elizabeth/sample.parquet') LIMIT 10; +SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=*/sample.parquet') WHERE column0 = _column0; + +SELECT _number, _date FROM file('$CURDIR/data_hive/partitioning/number=42/date=2020-01-01/sample.parquet') LIMIT 1; +SELECT _array, _float FROM file('$CURDIR/data_hive/partitioning/array=[1,2,3]/float=42.42/sample.parquet') LIMIT 1; +SELECT toTypeName(_array), toTypeName(_float) FROM file('$CURDIR/data_hive/partitioning/array=[1,2,3]/float=42.42/sample.parquet') LIMIT 1; +SELECT count(*) FROM file('$CURDIR/data_hive/partitioning/number=42/date=2020-01-01/sample.parquet') WHERE _number = 42; +""" + +$CLICKHOUSE_LOCAL -n -q """ +set use_hive_partitioning = 1; + +SELECT _identifier FROM file('$CURDIR/data_hive/partitioning/identifier=*/email.csv') LIMIT 2; +SELECT __identifier FROM file('$CURDIR/data_hive/partitioning/identifier=*/email.csv') LIMIT 2; +""" + +$CLICKHOUSE_LOCAL -n -q """ +set use_hive_partitioning = 1; + +SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/column0=Elizabeth/sample.parquet') LIMIT 10; +""" 2>&1 | grep -c "INCORRECT_DATA" + +$CLICKHOUSE_LOCAL -n -q """ +set use_hive_partitioning = 0; + +SELECT *, _column0 FROM file('$CURDIR/data_hive/partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; +""" 2>&1 | grep -c "UNKNOWN_IDENTIFIER" + + +$CLICKHOUSE_LOCAL -q "SELECT 'TESTING THE URL PARTITIONING'" + + +$CLICKHOUSE_LOCAL -n -q """ +set use_hive_partitioning = 1; + +SELECT *, _column0 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; + +SELECT *, _column0 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') WHERE column0 = _column0; + +SELECT *, _column0, _column1 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; +SELECT *, _column0 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; + +SELECT *, _column0, _column1 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; +SELECT *, _column0 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; + +SELECT *, _column0, _column1 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column1 = _column1; +SELECT *, _column0 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column1 = _column1; + +SELECT *, _column0, _column1 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; +SELECT *, _column0 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; + +SELECT *, _non_existing_column FROM url('http://localhost:11111/test/hive_partitioning/non_existing_column=Elizabeth/sample.parquet') LIMIT 10;""" + +$CLICKHOUSE_LOCAL -n -q """ +set use_hive_partitioning = 0; + +SELECT *, _column0 FROM url('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; +""" 2>&1 | grep -c "UNKNOWN_IDENTIFIER" + + +$CLICKHOUSE_LOCAL -q "SELECT 'TESTING THE S3 PARTITIONING'" + + +$CLICKHOUSE_CLIENT -n -q """ +set use_hive_partitioning = 1; + +SELECT *, _column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; + +SELECT *, _column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') WHERE column0 = _column0; + +SELECT *, _column0, _column1 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; +SELECT *, _column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; + +SELECT *, _column0, _column1 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; +SELECT *, _column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; + +SELECT *, _column0, _column1 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column1 = _column1; +SELECT *, _column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column1 = _column1; + +SELECT *, _column0, _column1 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; +SELECT *, _column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; + +SELECT *, _non_existing_column FROM s3('http://localhost:11111/test/hive_partitioning/non_existing_column=Elizabeth/sample.parquet') LIMIT 10; +SELECT *, _column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=*/sample.parquet') WHERE column0 = _column0; +""" + +$CLICKHOUSE_CLIENT -n -q """ +set use_hive_partitioning = 0; + +SELECT *, _column0 FROM s3('http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; +""" 2>&1 | grep -F -q "UNKNOWN_IDENTIFIER" && echo "OK" || echo "FAIL"; + +$CLICKHOUSE_LOCAL -q "SELECT 'TESTING THE S3CLUSTER PARTITIONING'" + +$CLICKHOUSE_CLIENT -n -q """ +set use_hive_partitioning = 1; + +SELECT *, _column0 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') LIMIT 10; + +SELECT *, _column0 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/sample.parquet') WHERE column0 = _column0; + +SELECT *, _column0, _column1 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; +SELECT *, _column0 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column1 = _column1; + +SELECT *, _column0, _column1 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; +SELECT *, _column0 FROM s3Cluster(test_cluster_one_shard_three_replicas_localhost, 'http://localhost:11111/test/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet') WHERE column0 = _column0 AND column1 = _column1; +""" diff --git a/tests/queries/0_stateless/03204_storage_join_optimize.reference b/tests/queries/0_stateless/03204_storage_join_optimize.reference new file mode 100644 index 00000000000..af98bcd6397 --- /dev/null +++ b/tests/queries/0_stateless/03204_storage_join_optimize.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/03204_storage_join_optimize.sql b/tests/queries/0_stateless/03204_storage_join_optimize.sql new file mode 100644 index 00000000000..03a4658ba6c --- /dev/null +++ b/tests/queries/0_stateless/03204_storage_join_optimize.sql @@ -0,0 +1,5 @@ +CREATE TABLE dict_03204 (k UInt64, v UInt64) ENGINE = Join(ANY, LEFT, k); +INSERT INTO dict_03204 SELECT number, number FROM numbers(10); +OPTIMIZE TABLE dict_03204; +SELECT * FROM dict_03204 ORDER BY k; +DROP TABLE dict_03204; diff --git a/tests/queries/0_stateless/03205_system_sync_replica_format.reference b/tests/queries/0_stateless/03205_system_sync_replica_format.reference new file mode 100644 index 00000000000..aad51dd90b0 --- /dev/null +++ b/tests/queries/0_stateless/03205_system_sync_replica_format.reference @@ -0,0 +1 @@ +SYSTEM SYNC REPLICA db.`table` LIGHTWEIGHT diff --git a/tests/queries/0_stateless/03205_system_sync_replica_format.sql b/tests/queries/0_stateless/03205_system_sync_replica_format.sql new file mode 100644 index 00000000000..329bce80afc --- /dev/null +++ b/tests/queries/0_stateless/03205_system_sync_replica_format.sql @@ -0,0 +1 @@ +SELECT formatQuery('SYSTEM SYNC REPLICA db.table LIGHTWEIGHT'); diff --git a/tests/queries/0_stateless/03206_projection_merge_special_mergetree.reference b/tests/queries/0_stateless/03206_projection_merge_special_mergetree.reference new file mode 100644 index 00000000000..1a9cc2b7fbf --- /dev/null +++ b/tests/queries/0_stateless/03206_projection_merge_special_mergetree.reference @@ -0,0 +1 @@ +p diff --git a/tests/queries/0_stateless/03206_projection_merge_special_mergetree.sql b/tests/queries/0_stateless/03206_projection_merge_special_mergetree.sql new file mode 100644 index 00000000000..82684f754b6 --- /dev/null +++ b/tests/queries/0_stateless/03206_projection_merge_special_mergetree.sql @@ -0,0 +1,106 @@ +DROP TABLE IF EXISTS tp; + +-- test regular merge tree +CREATE TABLE tp ( + type Int32, + eventcnt UInt64, + PROJECTION p (select sum(eventcnt), type group by type) +) engine = MergeTree order by type; + +INSERT INTO tp SELECT number%3, 1 FROM numbers(3); + +OPTIMIZE TABLE tp DEDUPLICATE; -- { serverError SUPPORT_IS_DISABLED } + +DROP TABLE tp; + +CREATE TABLE tp ( + type Int32, + eventcnt UInt64, + PROJECTION p (select sum(eventcnt), type group by type) +) engine = MergeTree order by type +SETTINGS deduplicate_merge_projection_mode = 'drop'; + +INSERT INTO tp SELECT number%3, 1 FROM numbers(3); + +OPTIMIZE TABLE tp DEDUPLICATE; + +ALTER TABLE tp MODIFY SETTING deduplicate_merge_projection_mode = 'throw'; + +OPTIMIZE TABLE tp DEDUPLICATE; -- { serverError SUPPORT_IS_DISABLED } + +DROP TABLE tp; + + +-- test irregular merge tree +CREATE TABLE tp ( + type Int32, + eventcnt UInt64, + PROJECTION p (select sum(eventcnt), type group by type) +) engine = ReplacingMergeTree order by type; -- { serverError SUPPORT_IS_DISABLED } + +CREATE TABLE tp ( + type Int32, + eventcnt UInt64, + PROJECTION p (select sum(eventcnt), type group by type) +) engine = ReplacingMergeTree order by type +SETTINGS deduplicate_merge_projection_mode = 'throw'; -- { serverError SUPPORT_IS_DISABLED } + +CREATE TABLE tp ( + type Int32, + eventcnt UInt64, + PROJECTION p (select sum(eventcnt), type group by type) +) engine = ReplacingMergeTree order by type +SETTINGS deduplicate_merge_projection_mode = 'drop'; + +INSERT INTO tp SELECT number%3, 1 FROM numbers(3); + +OPTIMIZE TABLE tp FINAL; + +-- expecting no projection +SYSTEM FLUSH LOGS; +SELECT + name +FROM system.projection_parts +WHERE (database = currentDatabase()) AND (`table` = 'tp') AND (active = 1); + +ALTER TABLE tp MODIFY SETTING deduplicate_merge_projection_mode = 'throw'; + +OPTIMIZE TABLE tp DEDUPLICATE; -- { serverError SUPPORT_IS_DISABLED } + +DROP TABLE tp; + +CREATE TABLE tp ( + type Int32, + eventcnt UInt64, + PROJECTION p (select sum(eventcnt), type group by type) +) engine = ReplacingMergeTree order by type +SETTINGS deduplicate_merge_projection_mode = 'rebuild'; + +ALTER TABLE tp MODIFY SETTING deduplicate_merge_projection_mode = 'throw'; + +OPTIMIZE TABLE tp DEDUPLICATE; -- { serverError SUPPORT_IS_DISABLED } + +DROP TABLE tp; + +-- test alter add projection case +CREATE TABLE tp ( + type Int32, + eventcnt UInt64 +) engine = ReplacingMergeTree order by type; + +ALTER TABLE tp ADD PROJECTION p (SELECT sum(eventcnt), type GROUP BY type); -- { serverError SUPPORT_IS_DISABLED } + +ALTER TABLE tp MODIFY SETTING deduplicate_merge_projection_mode = 'drop'; + +ALTER TABLE tp ADD PROJECTION p (SELECT sum(eventcnt), type GROUP BY type); + +INSERT INTO tp SELECT number%3, 1 FROM numbers(3); + +SYSTEM FLUSH LOGS; +-- expecting projection p +SELECT + name +FROM system.projection_parts +WHERE (database = currentDatabase()) AND (`table` = 'tp') AND (active = 1); + +DROP TABLE tp; diff --git a/tests/queries/0_stateless/03206_replication_lag_metric.reference b/tests/queries/0_stateless/03206_replication_lag_metric.reference new file mode 100644 index 00000000000..02f4a7264b1 --- /dev/null +++ b/tests/queries/0_stateless/03206_replication_lag_metric.reference @@ -0,0 +1,4 @@ +0 +2 +0 +2 diff --git a/tests/queries/0_stateless/03206_replication_lag_metric.sql b/tests/queries/0_stateless/03206_replication_lag_metric.sql new file mode 100644 index 00000000000..998c332a11c --- /dev/null +++ b/tests/queries/0_stateless/03206_replication_lag_metric.sql @@ -0,0 +1,11 @@ +-- Tags: no-parallel + +CREATE DATABASE rdb1 ENGINE = Replicated('/test/test_replication_lag_metric', 'shard1', 'replica1'); +CREATE DATABASE rdb2 ENGINE = Replicated('/test/test_replication_lag_metric', 'shard1', 'replica2'); + +SET distributed_ddl_task_timeout = 0; +CREATE TABLE rdb1.t (id UInt32) ENGINE = ReplicatedMergeTree ORDER BY id; +SELECT replication_lag FROM system.clusters WHERE cluster IN ('rdb1', 'rdb2') ORDER BY cluster ASC, replica_num ASC; + +DROP DATABASE rdb1; +DROP DATABASE rdb2; diff --git a/tests/queries/0_stateless/03208_buffer_over_distributed_type_mismatch.reference b/tests/queries/0_stateless/03208_buffer_over_distributed_type_mismatch.reference new file mode 100644 index 00000000000..1dc3acfeccb --- /dev/null +++ b/tests/queries/0_stateless/03208_buffer_over_distributed_type_mismatch.reference @@ -0,0 +1,18 @@ +100 +100 +101 +101 +101 +102 +101 +101 +102 +100 +100 +101 +101 +101 +102 +101 +101 +102 diff --git a/tests/queries/0_stateless/03208_buffer_over_distributed_type_mismatch.sql b/tests/queries/0_stateless/03208_buffer_over_distributed_type_mismatch.sql new file mode 100644 index 00000000000..333c445403d --- /dev/null +++ b/tests/queries/0_stateless/03208_buffer_over_distributed_type_mismatch.sql @@ -0,0 +1,76 @@ + +DROP TABLE IF EXISTS realtimedrep; +CREATE TABLE realtimedrep (`amount` Int32) ENGINE = MergeTree() ORDER BY tuple(); +INSERT INTO realtimedrep FORMAT Values (100); + +DROP TABLE IF EXISTS realtimedistributed; +CREATE TABLE realtimedistributed (`amount` Int32) ENGINE = Distributed(test_cluster_two_shards, currentDatabase(), realtimedrep, rand()); + +DROP TABLE IF EXISTS realtimebuff__fuzz_19; +CREATE TABLE realtimebuff__fuzz_19 (`amount` UInt32) ENGINE = Buffer(currentDatabase(), 'realtimedistributed', 16, 3600, 36000, 10000, 1000000, 10000000, 100000000); +INSERT INTO realtimebuff__fuzz_19 FORMAT Values (101); + +DROP TABLE IF EXISTS realtimebuff__fuzz_20; +CREATE TABLE realtimebuff__fuzz_20 (`amount` Nullable(Int32)) ENGINE = Buffer(currentDatabase(), 'realtimedistributed', 16, 3600, 36000, 10000, 1000000, 10000000, 100000000); +INSERT INTO realtimebuff__fuzz_20 FORMAT Values (101); + +SELECT amount FROM realtimebuff__fuzz_19 t1 ORDER BY ALL; +SELECT amount + 1 FROM realtimebuff__fuzz_19 t1 ORDER BY ALL; +SELECT amount + 1 FROM realtimebuff__fuzz_20 t1 ORDER BY ALL; +SELECT sum(amount) = 100 FROM realtimebuff__fuzz_19 ORDER BY ALL; -- { serverError CANNOT_CONVERT_TYPE } +SELECT sum(amount) = 100 FROM realtimebuff__fuzz_20 ORDER BY ALL; -- { serverError CANNOT_CONVERT_TYPE } + +SELECT amount FROM realtimebuff__fuzz_19 t1 +JOIN (SELECT number :: UInt32 AS amount FROM numbers(3) ) t2 ON t1.amount = t2.amount +ORDER BY ALL +SETTINGS allow_experimental_analyzer = 0; -- { serverError UNKNOWN_IDENTIFIER } + +SELECT amount FROM realtimebuff__fuzz_19 t1 +JOIN (SELECT number :: UInt32 AS amount FROM numbers(3) ) t2 ON t1.amount = t2.amount +ORDER BY ALL +SETTINGS allow_experimental_analyzer = 1; + +SELECT amount FROM realtimebuff__fuzz_19 t1 +JOIN (SELECT number :: UInt32 AS amount FROM numbers(300) ) t2 ON t1.amount = t2.amount +ORDER BY ALL +SETTINGS allow_experimental_analyzer = 0; -- { serverError UNKNOWN_IDENTIFIER } + +SELECT amount FROM realtimebuff__fuzz_19 t1 +JOIN (SELECT number :: UInt32 AS amount FROM numbers(300) ) t2 ON t1.amount = t2.amount +ORDER BY ALL +SETTINGS allow_experimental_analyzer = 1; + +SELECT t2.amount + 1 FROM (SELECT number :: UInt32 AS amount FROM numbers(300) ) t1 +JOIN realtimebuff__fuzz_19 t2 USING (amount) +ORDER BY ALL +; + +SELECT t2.amount + 1 FROM (SELECT number :: UInt32 AS amount FROM numbers(300) ) t1 +JOIN realtimebuff__fuzz_19 t2 ON t1.amount = t2.amount +ORDER BY ALL +; + +SELECT amount FROM realtimebuff__fuzz_19 t1 +JOIN realtimebuff__fuzz_19 t2 ON t1.amount = t2.amount +; -- { serverError NOT_IMPLEMENTED,UNKNOWN_IDENTIFIER } + +SELECT amount FROM realtimebuff__fuzz_19 t1 +JOIN realtimebuff__fuzz_19 t2 ON t1.amount = t2.amount +JOIN realtimebuff__fuzz_19 t3 ON t1.amount = t3.amount +; -- { serverError NOT_IMPLEMENTED,AMBIGUOUS_COLUMN_NAME } + + +-- fuzzers: + +SELECT + toLowCardinality(1) + materialize(toLowCardinality(2)) +FROM realtimebuff__fuzz_19 +GROUP BY toLowCardinality(1) +FORMAT Null +; + +SELECT intDivOrZero(intDivOrZero(toLowCardinality(-128), toLowCardinality(-1)) = 0, materialize(toLowCardinality(4))) +FROM realtimebuff__fuzz_19 GROUP BY materialize(toLowCardinality(-127)), intDivOrZero(0, 0) = toLowCardinality(toLowCardinality(0)) +WITH TOTALS ORDER BY ALL DESC NULLS FIRST +FORMAT Null +; diff --git a/tests/queries/0_stateless/03209_parallel_replicas_lost_decimal_conversion.reference b/tests/queries/0_stateless/03209_parallel_replicas_lost_decimal_conversion.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03209_parallel_replicas_lost_decimal_conversion.sql b/tests/queries/0_stateless/03209_parallel_replicas_lost_decimal_conversion.sql new file mode 100644 index 00000000000..bcc9dec306b --- /dev/null +++ b/tests/queries/0_stateless/03209_parallel_replicas_lost_decimal_conversion.sql @@ -0,0 +1,11 @@ +DROP TABLE IF EXISTS t_03209 SYNC; + +CREATE TABLE t_03209 ( `a` Decimal(18, 0), `b` Decimal(18, 1), `c` Decimal(36, 0) ) ENGINE = ReplicatedMergeTree('/clickhouse/{database}/test_03209', 'r1') ORDER BY tuple(); +INSERT INTO t_03209 VALUES ('33', '44.4', '35'); + +SET max_parallel_replicas = 2, cluster_for_parallel_replicas='parallel_replicas'; + +SELECT * FROM t_03209 WHERE a IN toDecimal32('33.3000', 4) SETTINGS allow_experimental_parallel_reading_from_replicas=0; +SELECT * FROM t_03209 WHERE a IN toDecimal32('33.3000', 4) SETTINGS allow_experimental_parallel_reading_from_replicas=1; + +DROP TABLE t_03209 SYNC; diff --git a/tests/queries/0_stateless/03210_dynamic_squashing.reference b/tests/queries/0_stateless/03210_dynamic_squashing.reference new file mode 100644 index 00000000000..1c23c22f550 --- /dev/null +++ b/tests/queries/0_stateless/03210_dynamic_squashing.reference @@ -0,0 +1,12 @@ +1 +Array(UInt8) true +None false +UInt64 false +2 +Array(UInt8) true +None false +UInt64 false +3 +Array(UInt8) true +String false +UInt64 true diff --git a/tests/queries/0_stateless/03210_dynamic_squashing.sql b/tests/queries/0_stateless/03210_dynamic_squashing.sql new file mode 100644 index 00000000000..71d09263fda --- /dev/null +++ b/tests/queries/0_stateless/03210_dynamic_squashing.sql @@ -0,0 +1,25 @@ +-- Tags: long + +set allow_experimental_dynamic_type = 1; +set max_block_size = 1000; + +drop table if exists test; + +create table test (d Dynamic) engine=MergeTree order by tuple(); +insert into test select multiIf(number < 1000, NULL::Dynamic(max_types=1), number < 3000, range(number % 5)::Dynamic(max_types=1), number::Dynamic(max_types=1)) from numbers(100000); +select '1'; +select distinct dynamicType(d) as type, isDynamicElementInSharedData(d) as flag from test order by type; + +drop table test; +create table test (d Dynamic(max_types=1)) engine=MergeTree order by tuple(); +insert into test select multiIf(number < 1000, NULL::Dynamic(max_types=1), number < 3000, range(number % 5)::Dynamic(max_types=1), number::Dynamic(max_types=1)) from numbers(100000); +select '2'; +select distinct dynamicType(d) as type, isDynamicElementInSharedData(d) as flag from test order by type; + +truncate table test; +insert into test select multiIf(number < 1000, 'Str'::Dynamic(max_types=1), number < 3000, range(number % 5)::Dynamic(max_types=1), number::Dynamic(max_types=1)) from numbers(100000); +select '3'; +select distinct dynamicType(d) as type, isDynamicElementInSharedData(d) as flag from test order by type; + +drop table test; + diff --git a/tests/queries/0_stateless/03210_lag_lead_inframe_types.reference b/tests/queries/0_stateless/03210_lag_lead_inframe_types.reference index d4734a85e72..4ecf7f56b07 100644 --- a/tests/queries/0_stateless/03210_lag_lead_inframe_types.reference +++ b/tests/queries/0_stateless/03210_lag_lead_inframe_types.reference @@ -38,3 +38,19 @@ 7 8 9 +15 \N 3 15 15 15 15 +14 \N 2 10 10 10 154 +13 \N 2 10 10 10 143 +12 \N 2 10 10 10 14 +11 \N 2 10 10 10 12 +10 \N 2 10 10 10 10 +9 \N 1 5 5 5 99 +8 \N 1 5 5 5 88 +7 \N 1 5 5 5 9 +6 \N 1 5 5 5 7 +5 \N 1 5 5 5 5 +4 \N 0 0 0 0 44 +3 \N 0 0 0 0 33 +2 \N 0 0 0 0 4 +1 \N 0 0 0 0 2 +0 \N 0 0 0 0 0 diff --git a/tests/queries/0_stateless/03210_lag_lead_inframe_types.sql b/tests/queries/0_stateless/03210_lag_lead_inframe_types.sql index f6017ee6690..cc6746e428f 100644 --- a/tests/queries/0_stateless/03210_lag_lead_inframe_types.sql +++ b/tests/queries/0_stateless/03210_lag_lead_inframe_types.sql @@ -2,3 +2,23 @@ SELECT lagInFrame(2::UInt128, 2, number) OVER w FROM numbers(10) WINDOW w AS (OR SELECT leadInFrame(2::UInt128, 2, number) OVER w FROM numbers(10) WINDOW w AS (ORDER BY number); SELECT lagInFrame(2::UInt64, 2, number) OVER w FROM numbers(10) WINDOW w AS (ORDER BY number); SELECT leadInFrame(2::UInt64, 2, number) OVER w FROM numbers(10) WINDOW w AS (ORDER BY number); + +SELECT + number, + YYYYMMDDToDate(1, toLowCardinality(11), max(YYYYMMDDToDate(YYYYMMDDToDate(toLowCardinality(1), 11, materialize(NULL), 19700101.1, 1, 27, 7, materialize(toUInt256(37)), 9, 19, 9), 1, toUInt128(11), NULL, 19700101.1, 1, 27, 7, 37, 9, 19, 9), toUInt256(30)) IGNORE NULLS OVER w, NULL, 19700101.1, toNullable(1), 27, materialize(7), 37, 9, 19, 9), + p, + pp, + lagInFrame(number, number - pp) OVER w AS lag2, + lagInFrame(number, number - pp, number * 11) OVER w AS lag, + leadInFrame(number, number - pp, number * 11) OVER w AS lead +FROM +( + SELECT + number, + intDiv(number, 5) AS p, + p * 5 AS pp + FROM numbers(16) +) +WHERE toLowCardinality(1) +WINDOW w AS (PARTITION BY p ORDER BY number ASC NULLS FIRST ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) +ORDER BY number DESC NULLS LAST; diff --git a/tests/queries/0_stateless/03212_max_bytes_to_read_for_schema_inference_in_cache.reference b/tests/queries/0_stateless/03212_max_bytes_to_read_for_schema_inference_in_cache.reference index cd109daac52..13b1138d1c4 100644 --- a/tests/queries/0_stateless/03212_max_bytes_to_read_for_schema_inference_in_cache.reference +++ b/tests/queries/0_stateless/03212_max_bytes_to_read_for_schema_inference_in_cache.reference @@ -1,2 +1,2 @@ x Nullable(Int64) -schema_inference_hints=, max_rows_to_read_for_schema_inference=25000, max_bytes_to_read_for_schema_inference=1000, schema_inference_make_columns_nullable=true, try_infer_integers=true, try_infer_dates=true, try_infer_datetimes=true, try_infer_numbers_from_strings=false, read_bools_as_numbers=true, read_bools_as_strings=true, read_objects_as_strings=true, read_numbers_as_strings=true, read_arrays_as_strings=true, try_infer_objects_as_tuples=true, infer_incomplete_types_as_strings=true, try_infer_objects=false, use_string_type_for_ambiguous_paths_in_named_tuples_inference_from_objects=false +schema_inference_hints=, max_rows_to_read_for_schema_inference=25000, max_bytes_to_read_for_schema_inference=1000, schema_inference_make_columns_nullable=true, try_infer_integers=true, try_infer_dates=true, try_infer_datetimes=true, try_infer_datetimes_only_datetime64=false, try_infer_numbers_from_strings=false, read_bools_as_numbers=true, read_bools_as_strings=true, read_objects_as_strings=true, read_numbers_as_strings=true, read_arrays_as_strings=true, try_infer_objects_as_tuples=true, infer_incomplete_types_as_strings=true, try_infer_objects=false, use_string_type_for_ambiguous_paths_in_named_tuples_inference_from_objects=false diff --git a/tests/queries/0_stateless/03212_max_bytes_to_read_for_schema_inference_in_cache.sh b/tests/queries/0_stateless/03212_max_bytes_to_read_for_schema_inference_in_cache.sh index 8a77538f592..adbb0cb6de0 100755 --- a/tests/queries/0_stateless/03212_max_bytes_to_read_for_schema_inference_in_cache.sh +++ b/tests/queries/0_stateless/03212_max_bytes_to_read_for_schema_inference_in_cache.sh @@ -5,7 +5,7 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) . "$CUR_DIR"/../shell_config.sh echo '{"x" : 42}' > $CLICKHOUSE_TEST_UNIQUE_NAME.json -$CLICKHOUSE_LOCAL -nm -q " +$CLICKHOUSE_LOCAL -m -q " DESC file('$CLICKHOUSE_TEST_UNIQUE_NAME.json') SETTINGS input_format_max_bytes_to_read_for_schema_inference=1000; SELECT additional_format_info from system.schema_inference_cache" diff --git a/tests/queries/0_stateless/03214_count_distinct_null_key_memory_leak.reference b/tests/queries/0_stateless/03214_count_distinct_null_key_memory_leak.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03214_count_distinct_null_key_memory_leak.sql b/tests/queries/0_stateless/03214_count_distinct_null_key_memory_leak.sql new file mode 100644 index 00000000000..84804e4e016 --- /dev/null +++ b/tests/queries/0_stateless/03214_count_distinct_null_key_memory_leak.sql @@ -0,0 +1,18 @@ +-- Tags: no-fasttest + +DROP TABLE IF EXISTS testnull; +CREATE TABLE testnull +( + `a` Nullable(String), + `b` Nullable(String), + `c` Nullable(String) +) +ENGINE = MergeTree +PARTITION BY tuple() +ORDER BY c +SETTINGS index_granularity = 8192, allow_nullable_key=1; + +INSERT INTO testnull(b,c) SELECT toString(rand64()) AS b, toString(rand64()) AS c FROM numbers(1000000); +SELECT count(distinct b) FROM testnull GROUP BY a SETTINGS max_memory_usage = 10000000; -- {serverError MEMORY_LIMIT_EXCEEDED} + +DROP TABLE testnull; \ No newline at end of file diff --git a/tests/queries/0_stateless/03215_grant_current_grants.reference b/tests/queries/0_stateless/03215_grant_current_grants.reference new file mode 100644 index 00000000000..e4f6850b806 --- /dev/null +++ b/tests/queries/0_stateless/03215_grant_current_grants.reference @@ -0,0 +1,2 @@ +GRANT SELECT, CREATE TABLE, CREATE VIEW ON default.* +GRANT SELECT ON default.* diff --git a/tests/queries/0_stateless/03215_grant_current_grants.sh b/tests/queries/0_stateless/03215_grant_current_grants.sh new file mode 100755 index 00000000000..68af4a62bba --- /dev/null +++ b/tests/queries/0_stateless/03215_grant_current_grants.sh @@ -0,0 +1,26 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + + +user1="user03215_1_${CLICKHOUSE_DATABASE}_$RANDOM" +user2="user03215_2_${CLICKHOUSE_DATABASE}_$RANDOM" +user3="user03215_3_${CLICKHOUSE_DATABASE}_$RANDOM" +db=${CLICKHOUSE_DATABASE} + + +${CLICKHOUSE_CLIENT} --query "CREATE USER $user1, $user2, $user3;"; +${CLICKHOUSE_CLIENT} --query "GRANT SELECT, CREATE TABLE, CREATE VIEW ON $db.* TO $user1 WITH GRANT OPTION;"; + +${CLICKHOUSE_CLIENT} --query "GRANT CURRENT GRANTS ON $db.* TO $user2" --user $user1; +${CLICKHOUSE_CLIENT} --query "GRANT CURRENT GRANTS ON $db.* TO $user3" --user $user2; + +${CLICKHOUSE_CLIENT} --query "SHOW GRANTS FOR $user2" | sed 's/ TO.*//'; +${CLICKHOUSE_CLIENT} --query "SHOW GRANTS FOR $user3" | sed 's/ TO.*//'; + +${CLICKHOUSE_CLIENT} --query "GRANT CURRENT GRANTS(SELECT ON $db.*) TO $user3" --user $user1; +${CLICKHOUSE_CLIENT} --query "SHOW GRANTS FOR $user3" | sed 's/ TO.*//'; + +${CLICKHOUSE_CLIENT} --query "DROP USER IF EXISTS $user1, $user2, $user3"; diff --git a/tests/queries/0_stateless/03215_multilinestring_geometry.reference b/tests/queries/0_stateless/03215_multilinestring_geometry.reference new file mode 100644 index 00000000000..9702dd6d6f8 --- /dev/null +++ b/tests/queries/0_stateless/03215_multilinestring_geometry.reference @@ -0,0 +1,30 @@ +-- { echoOn } +SELECT readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))'); +[[(1,1),(2,2),(3,3),(1,1)]] +SELECT toTypeName(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))')); +MultiLineString +SELECT wkt(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))')); +MULTILINESTRING((1 1,2 2,3 3,1 1)) +SELECT readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1), (1 0, 2 0, 3 0))'); +[[(1,1),(2,2),(3,3),(1,1)],[(1,0),(2,0),(3,0)]] +SELECT toTypeName(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1), (1 0, 2 0, 3 0))')); +MultiLineString +SELECT wkt(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1), (1 0, 2 0, 3 0))')); +MULTILINESTRING((1 1,2 2,3 3,1 1),(1 0,2 0,3 0)) +-- Native Array(Array(Tuple(Float64, Float64))) is treated as Polygon, not as MultiLineString. +WITH wkt(CAST([[(1, 1), (2, 2), (3, 3), (1, 1)]], 'Array(Array(Tuple(Float64, Float64)))')) as x +SELECT x, toTypeName(x), readWKTPolygon(x) as y, toTypeName(y); +POLYGON((1 1,2 2,3 3,1 1)) String [[(1,1),(2,2),(3,3),(1,1)]] Polygon +-- Non constant tests + +DROP TABLE IF EXISTS t; +CREATE TABLE IF NOT EXISTS t (shape Array(Array(Tuple(Float64, Float64))), wkt_string String, ord Float64) Engine = Memory; +INSERT INTO t (ord, shape, wkt_string) VALUES (1, [[(1, 1), (2, 2), (3, 3), (1, 1)]], 'MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))'); +INSERT INTO t (ord, shape, wkt_string) VALUES (2, [[(1, 1), (2, 2), (3, 3), (1, 1)], [(1, 0), (2, 0), (3, 0)]], 'MULTILINESTRING ((1 1, 2 2, 3 3, 1 1), (1 0, 2 0, 3 0))'); +INSERT INTO t (ord, shape, wkt_string) VALUES (3, [[(1, 0), (2, 1), (3, 0), (4, 1), (5, 0), (6, 1), (7, 0), (8, 1), (9, 0), (10, 1)]], 'MULTILINESTRING ((1 0, 2 1, 3 0, 4 1, 5 0, 6 1, 7 0, 8 1, 9 0, 10 1))'); +-- Native Array(Array(Tuple(Float64, Float64))) is treated as Polygon, not as MultiLineString. +-- but reading MultiLineString should still return an Array(Array(Tuple(Float64, Float64))) +select wkt(shape), readWKTMultiLineString(wkt_string), readWKTMultiLineString(wkt_string) = shape from t order by ord; +POLYGON((1 1,2 2,3 3,1 1)) [[(1,1),(2,2),(3,3),(1,1)]] 1 +POLYGON((1 1,2 2,3 3,1 1),(1 0,2 0,3 0,1 0)) [[(1,1),(2,2),(3,3),(1,1)],[(1,0),(2,0),(3,0)]] 1 +POLYGON((1 0,2 1,3 0,4 1,5 0,6 1,7 0,8 1,9 0,10 1,1 0)) [[(1,0),(2,1),(3,0),(4,1),(5,0),(6,1),(7,0),(8,1),(9,0),(10,1)]] 1 diff --git a/tests/queries/0_stateless/03215_multilinestring_geometry.sql b/tests/queries/0_stateless/03215_multilinestring_geometry.sql new file mode 100644 index 00000000000..cf4ef15f63d --- /dev/null +++ b/tests/queries/0_stateless/03215_multilinestring_geometry.sql @@ -0,0 +1,26 @@ +-- { echoOn } +SELECT readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))'); +SELECT toTypeName(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))')); +SELECT wkt(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))')); + +SELECT readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1), (1 0, 2 0, 3 0))'); +SELECT toTypeName(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1), (1 0, 2 0, 3 0))')); +SELECT wkt(readWKTMultiLineString('MULTILINESTRING ((1 1, 2 2, 3 3, 1 1), (1 0, 2 0, 3 0))')); + +-- Native Array(Array(Tuple(Float64, Float64))) is treated as Polygon, not as MultiLineString. +WITH wkt(CAST([[(1, 1), (2, 2), (3, 3), (1, 1)]], 'Array(Array(Tuple(Float64, Float64)))')) as x +SELECT x, toTypeName(x), readWKTPolygon(x) as y, toTypeName(y); + +-- Non constant tests + +DROP TABLE IF EXISTS t; +CREATE TABLE IF NOT EXISTS t (shape Array(Array(Tuple(Float64, Float64))), wkt_string String, ord Float64) Engine = Memory; +INSERT INTO t (ord, shape, wkt_string) VALUES (1, [[(1, 1), (2, 2), (3, 3), (1, 1)]], 'MULTILINESTRING ((1 1, 2 2, 3 3, 1 1))'); +INSERT INTO t (ord, shape, wkt_string) VALUES (2, [[(1, 1), (2, 2), (3, 3), (1, 1)], [(1, 0), (2, 0), (3, 0)]], 'MULTILINESTRING ((1 1, 2 2, 3 3, 1 1), (1 0, 2 0, 3 0))'); +INSERT INTO t (ord, shape, wkt_string) VALUES (3, [[(1, 0), (2, 1), (3, 0), (4, 1), (5, 0), (6, 1), (7, 0), (8, 1), (9, 0), (10, 1)]], 'MULTILINESTRING ((1 0, 2 1, 3 0, 4 1, 5 0, 6 1, 7 0, 8 1, 9 0, 10 1))'); + +-- Native Array(Array(Tuple(Float64, Float64))) is treated as Polygon, not as MultiLineString. +-- but reading MultiLineString should still return an Array(Array(Tuple(Float64, Float64))) +select wkt(shape), readWKTMultiLineString(wkt_string), readWKTMultiLineString(wkt_string) = shape from t order by ord; + + diff --git a/tests/queries/0_stateless/03215_parquet_index.reference b/tests/queries/0_stateless/03215_parquet_index.reference new file mode 100644 index 00000000000..334f2f3824e --- /dev/null +++ b/tests/queries/0_stateless/03215_parquet_index.reference @@ -0,0 +1,300 @@ +0 0 +1 1 +2 2 +3 3 +4 4 +5 5 +6 6 +7 7 +8 8 +9 9 +10 10 +11 11 +12 12 +13 13 +14 14 +15 15 +16 16 +17 17 +18 18 +19 19 +20 20 +21 21 +22 22 +23 23 +24 24 +25 25 +26 26 +27 27 +28 28 +29 29 +30 30 +31 31 +32 32 +33 33 +34 34 +35 35 +36 36 +37 37 +38 38 +39 39 +40 40 +41 41 +42 42 +43 43 +44 44 +45 45 +46 46 +47 47 +48 48 +49 49 +50 50 +51 51 +52 52 +53 53 +54 54 +55 55 +56 56 +57 57 +58 58 +59 59 +60 60 +61 61 +62 62 +63 63 +64 64 +65 65 +66 66 +67 67 +68 68 +69 69 +70 70 +71 71 +72 72 +73 73 +74 74 +75 75 +76 76 +77 77 +78 78 +79 79 +80 80 +81 81 +82 82 +83 83 +84 84 +85 85 +86 86 +87 87 +88 88 +89 89 +90 90 +91 91 +92 92 +93 93 +94 94 +95 95 +96 96 +97 97 +98 98 +99 99 +0 0 +1 1 +2 2 +3 3 +4 4 +5 5 +6 6 +7 7 +8 8 +9 9 +10 10 +11 11 +12 12 +13 13 +14 14 +15 15 +16 16 +17 17 +18 18 +19 19 +20 20 +21 21 +22 22 +23 23 +24 24 +25 25 +26 26 +27 27 +28 28 +29 29 +30 30 +31 31 +32 32 +33 33 +34 34 +35 35 +36 36 +37 37 +38 38 +39 39 +40 40 +41 41 +42 42 +43 43 +44 44 +45 45 +46 46 +47 47 +48 48 +49 49 +50 50 +51 51 +52 52 +53 53 +54 54 +55 55 +56 56 +57 57 +58 58 +59 59 +60 60 +61 61 +62 62 +63 63 +64 64 +65 65 +66 66 +67 67 +68 68 +69 69 +70 70 +71 71 +72 72 +73 73 +74 74 +75 75 +76 76 +77 77 +78 78 +79 79 +80 80 +81 81 +82 82 +83 83 +84 84 +85 85 +86 86 +87 87 +88 88 +89 89 +90 90 +91 91 +92 92 +93 93 +94 94 +95 95 +96 96 +97 97 +98 98 +99 99 +0 0 +1 1 +2 2 +3 3 +4 4 +5 5 +6 6 +7 7 +8 8 +9 9 +10 10 +11 11 +12 12 +13 13 +14 14 +15 15 +16 16 +17 17 +18 18 +19 19 +20 20 +21 21 +22 22 +23 23 +24 24 +25 25 +26 26 +27 27 +28 28 +29 29 +30 30 +31 31 +32 32 +33 33 +34 34 +35 35 +36 36 +37 37 +38 38 +39 39 +40 40 +41 41 +42 42 +43 43 +44 44 +45 45 +46 46 +47 47 +48 48 +49 49 +50 50 +51 51 +52 52 +53 53 +54 54 +55 55 +56 56 +57 57 +58 58 +59 59 +60 60 +61 61 +62 62 +63 63 +64 64 +65 65 +66 66 +67 67 +68 68 +69 69 +70 70 +71 71 +72 72 +73 73 +74 74 +75 75 +76 76 +77 77 +78 78 +79 79 +80 80 +81 81 +82 82 +83 83 +84 84 +85 85 +86 86 +87 87 +88 88 +89 89 +90 90 +91 91 +92 92 +93 93 +94 94 +95 95 +96 96 +97 97 +98 98 +99 99 diff --git a/tests/queries/0_stateless/03215_parquet_index.sql b/tests/queries/0_stateless/03215_parquet_index.sql new file mode 100644 index 00000000000..5766f40f021 --- /dev/null +++ b/tests/queries/0_stateless/03215_parquet_index.sql @@ -0,0 +1,21 @@ +-- Tags: no-fasttest + +-- default settings. +DROP TABLE IF EXISTS test_parquet; +CREATE TABLE test_parquet (col1 int, col2 String) ENGINE=File(Parquet); +INSERT INTO test_parquet SELECT number, toString(number) FROM numbers(100); +SELECT col1, col2 FROM test_parquet; + + +-- Parquet will have indexes in columns. We are not checking that indexes exist here, there is an integration test test_parquet_page_index for that. We just check that a setting doesn't break the SELECT +DROP TABLE IF EXISTS test_parquet; +CREATE TABLE test_parquet (col1 int, col2 String) ENGINE=File(Parquet) SETTINGS output_format_parquet_use_custom_encoder=false, output_format_parquet_write_page_index=true; +INSERT INTO test_parquet SELECT number, toString(number) FROM numbers(100); +SELECT col1, col2 FROM test_parquet; + + +-- Parquet will not have indexes in columns. +DROP TABLE IF EXISTS test_parquet; +CREATE TABLE test_parquet (col1 int, col2 String) ENGINE=File(Parquet) SETTINGS output_format_parquet_use_custom_encoder=false, output_format_parquet_write_page_index=false; +INSERT INTO test_parquet SELECT number, toString(number) FROM numbers(100); +SELECT col1, col2 FROM test_parquet; diff --git a/tests/queries/0_stateless/03215_parsing_archive_name_s3.sql b/tests/queries/0_stateless/03215_parsing_archive_name_s3.sql index e34be475c5a..e0d63d313fa 100644 --- a/tests/queries/0_stateless/03215_parsing_archive_name_s3.sql +++ b/tests/queries/0_stateless/03215_parsing_archive_name_s3.sql @@ -1,7 +1,15 @@ -- Tags: no-fasttest -- Tag no-fasttest: Depends on AWS +SET s3_truncate_on_insert=1; + +INSERT INTO FUNCTION s3(s3_conn, filename='::03215_archive.csv') SELECT 1; SELECT _file, _path FROM s3(s3_conn, filename='::03215_archive.csv') ORDER BY (_file, _path); + SELECT _file, _path FROM s3(s3_conn, filename='test :: 03215_archive.csv') ORDER BY (_file, _path); -- { serverError S3_ERROR } + +INSERT INTO FUNCTION s3(s3_conn, filename='test::03215_archive.csv') SELECT 1; SELECT _file, _path FROM s3(s3_conn, filename='test::03215_archive.csv') ORDER BY (_file, _path); + +INSERT INTO FUNCTION s3(s3_conn, filename='test.zip::03215_archive.csv') SETTINGS allow_archive_path_syntax=0 SELECT 1; SELECT _file, _path FROM s3(s3_conn, filename='test.zip::03215_archive.csv') ORDER BY (_file, _path) SETTINGS allow_archive_path_syntax=0; diff --git a/tests/queries/0_stateless/03215_validate_type_in_alter_add_modify_column.reference b/tests/queries/0_stateless/03215_validate_type_in_alter_add_modify_column.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03215_validate_type_in_alter_add_modify_column.sql b/tests/queries/0_stateless/03215_validate_type_in_alter_add_modify_column.sql new file mode 100644 index 00000000000..267bc7111f4 --- /dev/null +++ b/tests/queries/0_stateless/03215_validate_type_in_alter_add_modify_column.sql @@ -0,0 +1,19 @@ +set allow_experimental_variant_type = 0; +set allow_experimental_dynamic_type = 0; +set allow_suspicious_low_cardinality_types = 0; +set allow_suspicious_fixed_string_types = 0; + +drop table if exists test; +create table test (id UInt64) engine=MergeTree order by id; +alter table test add column bad Variant(UInt32, String); -- {serverError ILLEGAL_COLUMN} +alter table test add column bad Dynamic; -- {serverError ILLEGAL_COLUMN} +alter table test add column bad LowCardinality(UInt8); -- {serverError SUSPICIOUS_TYPE_FOR_LOW_CARDINALITY} +alter table test add column bad FixedString(10000); -- {serverError ILLEGAL_COLUMN} + +alter table test modify column id Variant(UInt32, String); -- {serverError ILLEGAL_COLUMN} +alter table test modify column id Dynamic; -- {serverError ILLEGAL_COLUMN} +alter table test modify column id LowCardinality(UInt8); -- {serverError SUSPICIOUS_TYPE_FOR_LOW_CARDINALITY} +alter table test modify column id FixedString(10000); -- {serverError ILLEGAL_COLUMN} + +drop table test; + diff --git a/tests/queries/0_stateless/03215_view_with_recursive.reference b/tests/queries/0_stateless/03215_view_with_recursive.reference new file mode 100644 index 00000000000..c3ca8065a70 --- /dev/null +++ b/tests/queries/0_stateless/03215_view_with_recursive.reference @@ -0,0 +1,2 @@ +5050 +8 diff --git a/tests/queries/0_stateless/03215_view_with_recursive.sql b/tests/queries/0_stateless/03215_view_with_recursive.sql new file mode 100644 index 00000000000..5d93ccc5438 --- /dev/null +++ b/tests/queries/0_stateless/03215_view_with_recursive.sql @@ -0,0 +1,43 @@ +SET allow_experimental_analyzer = 1; + +CREATE VIEW 03215_test_v +AS WITH RECURSIVE test_table AS + ( + SELECT 1 AS number + UNION ALL + SELECT number + 1 + FROM test_table + WHERE number < 100 + ) +SELECT sum(number) +FROM test_table; + +SELECT * FROM 03215_test_v; + +CREATE VIEW 03215_multi_v +AS WITH RECURSIVE + task AS + ( + SELECT + number AS task_id, + number - 1 AS parent_id + FROM numbers(10) + ), + rtq AS + ( + SELECT + task_id, + parent_id + FROM task AS t + WHERE t.parent_id = 1 + UNION ALL + SELECT + t.task_id, + t.parent_id + FROM task AS t, rtq AS r + WHERE t.parent_id = r.task_id + ) +SELECT count() +FROM rtq; + +SELECT * FROM 03215_multi_v; diff --git a/tests/queries/0_stateless/03217_datetime64_constant_to_ast.reference b/tests/queries/0_stateless/03217_datetime64_constant_to_ast.reference new file mode 100644 index 00000000000..c20baa0d261 --- /dev/null +++ b/tests/queries/0_stateless/03217_datetime64_constant_to_ast.reference @@ -0,0 +1,2 @@ +1970-01-01 00:00:01.000 +1970-01-01 00:00:01.000 diff --git a/tests/queries/0_stateless/03217_datetime64_constant_to_ast.sql b/tests/queries/0_stateless/03217_datetime64_constant_to_ast.sql new file mode 100644 index 00000000000..63334a511c7 --- /dev/null +++ b/tests/queries/0_stateless/03217_datetime64_constant_to_ast.sql @@ -0,0 +1,6 @@ + +SET session_timezone = 'UTC'; + +SELECT toDateTime64('1970-01-01 00:00:01', 3) +FROM remote('127.0.0.{1,2}', system, one) +; diff --git a/tests/queries/0_stateless/03217_filtering_in_storage_merge.reference b/tests/queries/0_stateless/03217_filtering_in_storage_merge.reference new file mode 100644 index 00000000000..d366ad04c39 --- /dev/null +++ b/tests/queries/0_stateless/03217_filtering_in_storage_merge.reference @@ -0,0 +1,6 @@ +Expression ((Project names + Projection)) + Aggregating + Expression (Before GROUP BY) + ReadFromMerge + Filter (( + ( + ))) + ReadFromMergeTree (default.test_03217_merge_replica_1) diff --git a/tests/queries/0_stateless/03217_filtering_in_storage_merge.sql b/tests/queries/0_stateless/03217_filtering_in_storage_merge.sql new file mode 100644 index 00000000000..42d31e95f9c --- /dev/null +++ b/tests/queries/0_stateless/03217_filtering_in_storage_merge.sql @@ -0,0 +1,16 @@ +CREATE TABLE test_03217_merge_replica_1(x UInt32) + ENGINE ReplicatedMergeTree('/clickhouse/tables/{database}/test_03217_merge_replica', 'r1') + ORDER BY x; +CREATE TABLE test_03217_merge_replica_2(x UInt32) + ENGINE ReplicatedMergeTree('/clickhouse/tables/{database}/test_03217_merge_replica', 'r2') + ORDER BY x; + + +CREATE TABLE test_03217_all_replicas (x UInt32) + ENGINE = Merge(currentDatabase(), 'test_03217_merge_replica_*'); + +INSERT INTO test_03217_merge_replica_1 SELECT number AS x FROM numbers(10); +SYSTEM SYNC REPLICA test_03217_merge_replica_2; + +-- If the filter on _table is not applied, then the plan will show both replicas +EXPLAIN SELECT _table, count() FROM test_03217_all_replicas WHERE _table = 'test_03217_merge_replica_1' AND x >= 0 GROUP BY _table SETTINGS allow_experimental_analyzer=1; diff --git a/tests/queries/0_stateless/03217_filtering_in_system_tables.reference b/tests/queries/0_stateless/03217_filtering_in_system_tables.reference new file mode 100644 index 00000000000..c0761c3f689 --- /dev/null +++ b/tests/queries/0_stateless/03217_filtering_in_system_tables.reference @@ -0,0 +1,6 @@ +information_schema tables +both default test_03217_system_tables_replica_1 r1 +both default test_03217_system_tables_replica_2 r2 +default test_03217_system_tables_replica_1 r1 +1 +1 diff --git a/tests/queries/0_stateless/03217_filtering_in_system_tables.sql b/tests/queries/0_stateless/03217_filtering_in_system_tables.sql new file mode 100644 index 00000000000..eb506dfe39a --- /dev/null +++ b/tests/queries/0_stateless/03217_filtering_in_system_tables.sql @@ -0,0 +1,30 @@ +-- If filtering is not done correctly on databases, then this query report to read 3 rows, which are: `system.tables`, `information_schema.tables` and `INFORMATION_SCHEMA.tables` +SELECT database, table FROM system.tables WHERE database = 'information_schema' AND table = 'tables'; + +CREATE TABLE test_03217_system_tables_replica_1(x UInt32) + ENGINE ReplicatedMergeTree('/clickhouse/tables/{database}/test_03217_system_tables_replica', 'r1') + ORDER BY x; +CREATE TABLE test_03217_system_tables_replica_2(x UInt32) + ENGINE ReplicatedMergeTree('/clickhouse/tables/{database}/test_03217_system_tables_replica', 'r2') + ORDER BY x; + +-- Make sure we can read both replicas +-- The replica name might be altered because of `_functional_tests_helper_database_replicated_replace_args_macros`, +-- thus we need to use `left` +SELECT 'both', database, table, left(replica_name, 2) FROM system.replicas WHERE database = currentDatabase(); +-- If filtering is not done correctly on database-table column, then this query report to read 2 rows, which are the above tables +SELECT database, table, left(replica_name, 2) FROM system.replicas WHERE database = currentDatabase() AND table = 'test_03217_system_tables_replica_1' AND replica_name LIKE 'r1%'; +SYSTEM FLUSH LOGS; +-- argMax is necessary to make the test repeatable + +-- StorageSystemTables +SELECT argMax(read_rows, event_time_microseconds) FROM system.query_log WHERE 1 + AND current_database = currentDatabase() + AND query LIKE '%SELECT database, table FROM system.tables WHERE database = \'information_schema\' AND table = \'tables\';' + AND type = 'QueryFinish'; + +-- StorageSystemReplicas +SELECT argMax(read_rows, event_time_microseconds) FROM system.query_log WHERE 1 + AND current_database = currentDatabase() + AND query LIKE '%SELECT database, table, left(replica_name, 2) FROM system.replicas WHERE database = currentDatabase() AND table = \'test_03217_system_tables_replica_1\' AND replica_name LIKE \'r1\%\';' + AND type = 'QueryFinish'; diff --git a/tests/queries/0_stateless/03217_fliter_pushdown_no_keys.reference b/tests/queries/0_stateless/03217_fliter_pushdown_no_keys.reference new file mode 100644 index 00000000000..9838dd1b936 --- /dev/null +++ b/tests/queries/0_stateless/03217_fliter_pushdown_no_keys.reference @@ -0,0 +1,6 @@ +--- +1 1 +--- +3 3 +--- +--- diff --git a/tests/queries/0_stateless/03217_fliter_pushdown_no_keys.sql b/tests/queries/0_stateless/03217_fliter_pushdown_no_keys.sql new file mode 100644 index 00000000000..cb8bf59e790 --- /dev/null +++ b/tests/queries/0_stateless/03217_fliter_pushdown_no_keys.sql @@ -0,0 +1,26 @@ + + + +select * from ( select sum(last_seen) as dates_seen, materialize(1) as last_seen ) where last_seen > 2; +select * from ( select sum(last_seen) as dates_seen, materialize(2) as last_seen ) where last_seen < 2; +select * from ( select sum(last_seen) as dates_seen, materialize(2) as last_seen GROUP BY 'a' ) where last_seen < 2; + +select '---'; +select * from ( select sum(last_seen) as dates_seen, 1 as last_seen UNION ALL select sum(last_seen) as dates_seen, 3 as last_seen ) where last_seen < 2; + +select '---'; +select * from ( select sum(last_seen) as dates_seen, 1 as last_seen UNION ALL select sum(last_seen) as dates_seen, 3 as last_seen ) where last_seen > 2; + +select '---'; +with activity as ( + select + groupUniqArrayState(toDate('2025-01-01 01:00:00')) as dates_seen, + toDateTime('2025-01-01 01:00:00') as last_seen + union all + select + groupUniqArrayState(toDate('2023-11-11 11:11:11')) as dates_seen, + toDateTime('2023-11-11 11:11:11') as last_seen +) +select last_seen from activity +where last_seen < toDateTime('2020-01-01 00:00:00'); +select '---'; diff --git a/tests/queries/0_stateless/03217_primary_index_memory_leak.reference b/tests/queries/0_stateless/03217_primary_index_memory_leak.reference new file mode 100644 index 00000000000..4913dd5e690 --- /dev/null +++ b/tests/queries/0_stateless/03217_primary_index_memory_leak.reference @@ -0,0 +1 @@ +150000 diff --git a/tests/queries/0_stateless/03217_primary_index_memory_leak.sql b/tests/queries/0_stateless/03217_primary_index_memory_leak.sql new file mode 100644 index 00000000000..d5a553c7d72 --- /dev/null +++ b/tests/queries/0_stateless/03217_primary_index_memory_leak.sql @@ -0,0 +1,15 @@ +-- Tags: no-debug, no-tsan, no-msan, no-asan, no-random-settings, no-random-merge-tree-settings + +DROP TABLE IF EXISTS t_primary_index_memory; +CREATE TABLE t_primary_index_memory (s String) ENGINE = MergeTree +ORDER BY s SETTINGS index_granularity = 1; + +INSERT INTO t_primary_index_memory SELECT repeat('a', 10000) FROM numbers(150000) +SETTINGS + max_block_size = 32, + max_memory_usage = '100M', + max_insert_block_size = 1024, + min_insert_block_size_rows = 1024; + +SELECT count() FROM t_primary_index_memory; +DROP TABLE t_primary_index_memory; diff --git a/tests/queries/0_stateless/03218_materialize_msan.reference b/tests/queries/0_stateless/03218_materialize_msan.reference new file mode 100644 index 00000000000..eb84f35f9f4 --- /dev/null +++ b/tests/queries/0_stateless/03218_materialize_msan.reference @@ -0,0 +1 @@ +[(NULL,'11\01111111\011111','1111')] -2147483648 \N diff --git a/tests/queries/0_stateless/03218_materialize_msan.sql b/tests/queries/0_stateless/03218_materialize_msan.sql new file mode 100644 index 00000000000..7e7043e687b --- /dev/null +++ b/tests/queries/0_stateless/03218_materialize_msan.sql @@ -0,0 +1,23 @@ +SET enable_analyzer = 1; + +SELECT + materialize([(NULL, '11\01111111\011111', '1111')]) AS t, + (t[1048576]).2, + materialize(-2147483648), + (t[-2147483648]).1 +GROUP BY + materialize([(NULL, '1')]), + '', + (materialize((t[1023]).2), (materialize(''), (t[2147483647]).1, materialize(9223372036854775807)), (materialize(''), materialize(NULL, 2147483647, t[65535], 256)), materialize(NULL)) +; -- { serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH} + +SELECT + materialize([(NULL, '11\01111111\011111', '1111')]) AS t, + (t[1048576]).2, + materialize(-2147483648), + (t[-2147483648]).1 +GROUP BY + materialize([(NULL, '1')]), + '', + (materialize((t[1023]).2), (materialize(''), (t[2147483647]).1, materialize(9223372036854775807)), (materialize(''), materialize(NULL), materialize(2147483647), materialize(t[65535]), materialize(256)), materialize(NULL)) +; diff --git a/tests/queries/0_stateless/03221_create_if_not_exists_setting.reference b/tests/queries/0_stateless/03221_create_if_not_exists_setting.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03221_create_if_not_exists_setting.sql b/tests/queries/0_stateless/03221_create_if_not_exists_setting.sql new file mode 100644 index 00000000000..18b3ed7bcec --- /dev/null +++ b/tests/queries/0_stateless/03221_create_if_not_exists_setting.sql @@ -0,0 +1,24 @@ +-- Tags: no-parallel + +SET create_if_not_exists=0; -- Default + +DROP TABLE IF EXISTS example_table; +CREATE TABLE example_table (id UInt32) ENGINE=MergeTree() ORDER BY id; +CREATE TABLE example_table (id UInt32) ENGINE=MergeTree() ORDER BY id; -- { serverError TABLE_ALREADY_EXISTS } + +DROP DATABASE IF EXISTS example_database; +CREATE DATABASE example_database; +CREATE DATABASE example_database; -- { serverError DATABASE_ALREADY_EXISTS } + +SET create_if_not_exists=1; + +DROP TABLE IF EXISTS example_table; +CREATE TABLE example_table (id UInt32) ENGINE=MergeTree() ORDER BY id; +CREATE TABLE example_table (id UInt32) ENGINE=MergeTree() ORDER BY id; + +DROP DATABASE IF EXISTS example_database; +CREATE DATABASE example_database; +CREATE DATABASE example_database; + +DROP DATABASE IF EXISTS example_database; +DROP TABLE IF EXISTS example_table; \ No newline at end of file diff --git a/tests/queries/0_stateless/03221_insert_timeout_overflow_mode.reference b/tests/queries/0_stateless/03221_insert_timeout_overflow_mode.reference new file mode 100644 index 00000000000..68538c3f75b --- /dev/null +++ b/tests/queries/0_stateless/03221_insert_timeout_overflow_mode.reference @@ -0,0 +1,2 @@ +QUERY_WAS_CANCELLED +QUERY_WAS_CANCELLED diff --git a/tests/queries/0_stateless/03221_insert_timeout_overflow_mode.sh b/tests/queries/0_stateless/03221_insert_timeout_overflow_mode.sh new file mode 100755 index 00000000000..db943a665cb --- /dev/null +++ b/tests/queries/0_stateless/03221_insert_timeout_overflow_mode.sh @@ -0,0 +1,8 @@ +#!/usr/bin/env bash + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +${CLICKHOUSE_CLIENT} --query "create table null_t (number UInt64) engine = Null;" +${CLICKHOUSE_CLIENT} --query "select sleep(0.1) from system.numbers settings max_block_size = 1 format Native" 2>/dev/null | ${CLICKHOUSE_CLIENT} --max_execution_time 0.3 --timeout_overflow_mode break --query "insert into null_t format Native" 2>&1 | grep -o "QUERY_WAS_CANCELLED" diff --git a/tests/queries/0_stateless/03221_merge_profile_events.reference b/tests/queries/0_stateless/03221_merge_profile_events.reference new file mode 100644 index 00000000000..d969717336b --- /dev/null +++ b/tests/queries/0_stateless/03221_merge_profile_events.reference @@ -0,0 +1,3 @@ +Horizontal 1 20000 3 0 480000 1 1 1 1 +Vertical 1 20000 1 2 480000 1 1 1 1 1 1 +Vertical 2 400000 2 6 12800000 1 1 1 1 1 1 1 1 1 1 diff --git a/tests/queries/0_stateless/03221_merge_profile_events.sql b/tests/queries/0_stateless/03221_merge_profile_events.sql new file mode 100644 index 00000000000..1aa3dd266f8 --- /dev/null +++ b/tests/queries/0_stateless/03221_merge_profile_events.sql @@ -0,0 +1,90 @@ +-- Tags: no-random-settings, no-random-merge-tree-settings + +DROP TABLE IF EXISTS t_merge_profile_events_1; + +CREATE TABLE t_merge_profile_events_1 (id UInt64, v1 UInt64, v2 UInt64) +ENGINE = MergeTree ORDER BY id +SETTINGS min_bytes_for_wide_part = 0; + +INSERT INTO t_merge_profile_events_1 SELECT number, number, number FROM numbers(10000); +INSERT INTO t_merge_profile_events_1 SELECT number, number, number FROM numbers(10000); + +OPTIMIZE TABLE t_merge_profile_events_1 FINAL; +SYSTEM FLUSH LOGS; + +SELECT + merge_algorithm, + ProfileEvents['Merge'], + ProfileEvents['MergedRows'], + ProfileEvents['MergedColumns'], + ProfileEvents['GatheredColumns'], + ProfileEvents['MergedUncompressedBytes'], + ProfileEvents['MergeTotalMilliseconds'] > 0, + ProfileEvents['MergeExecuteMilliseconds'] > 0, + ProfileEvents['MergeHorizontalStageTotalMilliseconds'] > 0, + ProfileEvents['MergeHorizontalStageExecuteMilliseconds'] > 0 +FROM system.part_log WHERE database = currentDatabase() AND table = 't_merge_profile_events_1' AND event_type = 'MergeParts' AND part_name = 'all_1_2_1'; + +DROP TABLE IF EXISTS t_merge_profile_events_1; + +DROP TABLE IF EXISTS t_merge_profile_events_2; + +CREATE TABLE t_merge_profile_events_2 (id UInt64, v1 UInt64, v2 UInt64) +ENGINE = MergeTree ORDER BY id +SETTINGS min_bytes_for_wide_part = 0, vertical_merge_algorithm_min_rows_to_activate = 1, vertical_merge_algorithm_min_columns_to_activate = 1; + +INSERT INTO t_merge_profile_events_2 SELECT number, number, number FROM numbers(10000); +INSERT INTO t_merge_profile_events_2 SELECT number, number, number FROM numbers(10000); + +OPTIMIZE TABLE t_merge_profile_events_2 FINAL; +SYSTEM FLUSH LOGS; + +SELECT + merge_algorithm, + ProfileEvents['Merge'], + ProfileEvents['MergedRows'], + ProfileEvents['MergedColumns'], + ProfileEvents['GatheredColumns'], + ProfileEvents['MergedUncompressedBytes'], + ProfileEvents['MergeTotalMilliseconds'] > 0, + ProfileEvents['MergeExecuteMilliseconds'] > 0, + ProfileEvents['MergeHorizontalStageTotalMilliseconds'] > 0, + ProfileEvents['MergeHorizontalStageExecuteMilliseconds'] > 0, + ProfileEvents['MergeVerticalStageTotalMilliseconds'] > 0, + ProfileEvents['MergeVerticalStageExecuteMilliseconds'] > 0, +FROM system.part_log WHERE database = currentDatabase() AND table = 't_merge_profile_events_2' AND event_type = 'MergeParts' AND part_name = 'all_1_2_1'; + +DROP TABLE IF EXISTS t_merge_profile_events_2; + +DROP TABLE IF EXISTS t_merge_profile_events_3; + +CREATE TABLE t_merge_profile_events_3 (id UInt64, v1 UInt64, v2 UInt64, PROJECTION p (SELECT v2, v2 * v2, v2 * 2, v2 * 10, v1 ORDER BY v1)) +ENGINE = MergeTree ORDER BY id +SETTINGS min_bytes_for_wide_part = 0, vertical_merge_algorithm_min_rows_to_activate = 1, vertical_merge_algorithm_min_columns_to_activate = 1; + +INSERT INTO t_merge_profile_events_3 SELECT number, number, number FROM numbers(100000); +INSERT INTO t_merge_profile_events_3 SELECT number, number, number FROM numbers(100000); + +OPTIMIZE TABLE t_merge_profile_events_3 FINAL; +SYSTEM FLUSH LOGS; + +SELECT + merge_algorithm, + ProfileEvents['Merge'], + ProfileEvents['MergedRows'], + ProfileEvents['MergedColumns'], + ProfileEvents['GatheredColumns'], + ProfileEvents['MergedUncompressedBytes'], + ProfileEvents['MergeTotalMilliseconds'] > 0, + ProfileEvents['MergeExecuteMilliseconds'] > 0, + ProfileEvents['MergeHorizontalStageTotalMilliseconds'] > 0, + ProfileEvents['MergeHorizontalStageExecuteMilliseconds'] > 0, + ProfileEvents['MergeVerticalStageTotalMilliseconds'] > 0, + ProfileEvents['MergeVerticalStageExecuteMilliseconds'] > 0, + ProfileEvents['MergeProjectionStageTotalMilliseconds'] > 0, + ProfileEvents['MergeProjectionStageExecuteMilliseconds'] > 0, + ProfileEvents['MergeExecuteMilliseconds'] <= duration_ms, + ProfileEvents['MergeTotalMilliseconds'] <= duration_ms +FROM system.part_log WHERE database = currentDatabase() AND table = 't_merge_profile_events_3' AND event_type = 'MergeParts' AND part_name = 'all_1_2_1'; + +DROP TABLE IF EXISTS t_merge_profile_events_3; diff --git a/tests/queries/0_stateless/03221_mutate_profile_events.reference b/tests/queries/0_stateless/03221_mutate_profile_events.reference new file mode 100644 index 00000000000..d094c37ff88 --- /dev/null +++ b/tests/queries/0_stateless/03221_mutate_profile_events.reference @@ -0,0 +1,2 @@ +3 2 1 10000 160000 0 1 1 1 +4 2 1 10000 320000 1 0 1 1 diff --git a/tests/queries/0_stateless/03221_mutate_profile_events.sql b/tests/queries/0_stateless/03221_mutate_profile_events.sql new file mode 100644 index 00000000000..e9f7f9670bd --- /dev/null +++ b/tests/queries/0_stateless/03221_mutate_profile_events.sql @@ -0,0 +1,33 @@ +-- Tags: no-random-settings, no-random-merge-tree-settings + +DROP TABLE IF EXISTS t_mutate_profile_events; + +CREATE TABLE t_mutate_profile_events (key UInt64, id UInt64, v1 UInt64, v2 UInt64) +ENGINE = MergeTree ORDER BY id PARTITION BY key +SETTINGS min_bytes_for_wide_part = 0; + +INSERT INTO t_mutate_profile_events SELECT 1, number, number, number FROM numbers(10000); +INSERT INTO t_mutate_profile_events SELECT 2, number, number, number FROM numbers(10000); + +SET mutations_sync = 2; + +ALTER TABLE t_mutate_profile_events UPDATE v1 = 1000 WHERE key = 1; +ALTER TABLE t_mutate_profile_events DELETE WHERE key = 2 AND v2 % 10 = 0; + +SYSTEM FLUSH LOGS; + +SELECT + splitByChar('_', part_name)[-1] AS version, + sum(ProfileEvents['MutationTotalParts']), + sum(ProfileEvents['MutationUntouchedParts']), + sum(ProfileEvents['MutatedRows']), + sum(ProfileEvents['MutatedUncompressedBytes']), + sum(ProfileEvents['MutationAllPartColumns']), + sum(ProfileEvents['MutationSomePartColumns']), + sum(ProfileEvents['MutationTotalMilliseconds']) > 0, + sum(ProfileEvents['MutationExecuteMilliseconds']) > 0, +FROM system.part_log +WHERE database = currentDatabase() AND table = 't_mutate_profile_events' AND event_type = 'MutatePart' +GROUP BY version ORDER BY version; + +DROP TABLE IF EXISTS t_mutate_profile_events diff --git a/tests/queries/0_stateless/03221_mutation_analyzer_skip_part.reference b/tests/queries/0_stateless/03221_mutation_analyzer_skip_part.reference new file mode 100644 index 00000000000..68f8708dbaf --- /dev/null +++ b/tests/queries/0_stateless/03221_mutation_analyzer_skip_part.reference @@ -0,0 +1,4 @@ +1_1_1_0_3 10000 +1_1_1_0_4 0 +2_2_2_0_3 0 +2_2_2_0_4 10000 diff --git a/tests/queries/0_stateless/03221_mutation_analyzer_skip_part.sql b/tests/queries/0_stateless/03221_mutation_analyzer_skip_part.sql new file mode 100644 index 00000000000..bf9a10e2af4 --- /dev/null +++ b/tests/queries/0_stateless/03221_mutation_analyzer_skip_part.sql @@ -0,0 +1,21 @@ +DROP TABLE IF EXISTS t_mutate_skip_part; + +CREATE TABLE t_mutate_skip_part (key UInt64, id UInt64, v1 UInt64, v2 UInt64) ENGINE = MergeTree ORDER BY id PARTITION BY key; + +INSERT INTO t_mutate_skip_part SELECT 1, number, number, number FROM numbers(10000); +INSERT INTO t_mutate_skip_part SELECT 2, number, number, number FROM numbers(10000); + +SET mutations_sync = 2; + +ALTER TABLE t_mutate_skip_part UPDATE v1 = 1000 WHERE key = 1; +ALTER TABLE t_mutate_skip_part DELETE WHERE key = 2 AND v2 % 10 = 0; + +SYSTEM FLUSH LOGS; + +-- If part is skipped in mutation and hardlinked then read_rows must be 0. +SELECT part_name, read_rows +FROM system.part_log +WHERE database = currentDatabase() AND table = 't_mutate_skip_part' AND event_type = 'MutatePart' +ORDER BY part_name; + +DROP TABLE IF EXISTS t_mutate_skip_part; diff --git a/tests/queries/0_stateless/data_minio/::03215_archive.csv b/tests/queries/0_stateless/03221_s3_imds_decent_timeout.reference similarity index 100% rename from tests/queries/0_stateless/data_minio/::03215_archive.csv rename to tests/queries/0_stateless/03221_s3_imds_decent_timeout.reference diff --git a/tests/queries/0_stateless/03221_s3_imds_decent_timeout.sh b/tests/queries/0_stateless/03221_s3_imds_decent_timeout.sh new file mode 100755 index 00000000000..021278955cd --- /dev/null +++ b/tests/queries/0_stateless/03221_s3_imds_decent_timeout.sh @@ -0,0 +1,16 @@ +#!/usr/bin/env bash +# Tags: no-fasttest, no-asan, no-msan, no-tsan +# ^ requires S3 + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +# Inaccessible IMDS should not introduce large delays, so this query should reply quickly at least sometimes: +while true +do + # This host (likely) drops packets sent to it (does not reply), so it is good for testing timeouts. + # At the same time, we expect that the clickhouse host does not drop packets and quickly replies with 4xx, which is a non-retriable error for S3. + AWS_EC2_METADATA_SERVICE_ENDPOINT='https://10.255.255.255/' ${CLICKHOUSE_LOCAL} --time --query "SELECT * FROM s3('${CLICKHOUSE_PORT_HTTP_PROTO}://${CLICKHOUSE_HOST}:${CLICKHOUSE_PORT_HTTP}/nonexistent')" |& grep -v -F 404 | + ${CLICKHOUSE_LOCAL} --input-format TSV "SELECT c1::Float64 < 1 FROM table" | grep 1 && break +done diff --git a/tests/queries/0_stateless/03222_create_timeseries_table.reference b/tests/queries/0_stateless/03222_create_timeseries_table.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03222_create_timeseries_table.sql b/tests/queries/0_stateless/03222_create_timeseries_table.sql new file mode 100644 index 00000000000..bdb29e7d366 --- /dev/null +++ b/tests/queries/0_stateless/03222_create_timeseries_table.sql @@ -0,0 +1,7 @@ +SET allow_experimental_time_series_table = 1; + +CREATE TABLE 03222_timeseries_table1 ENGINE = TimeSeries FORMAT Null; +CREATE TABLE 03222_timeseries_table2 ENGINE = TimeSeries SETTINGS store_min_time_and_max_time = 1, aggregate_min_time_and_max_time = 1 FORMAT Null; +--- This doesn't work because allow_nullable_key cannot be set in query for the internal MergeTree tables +--- CREATE TABLE 03222_timeseries_table3 ENGINE = TimeSeries SETTINGS store_min_time_and_max_time = 1, aggregate_min_time_and_max_time = 0; +CREATE TABLE 03222_timeseries_table4 ENGINE = TimeSeries SETTINGS store_min_time_and_max_time = 0 FORMAT Null; diff --git a/tests/queries/0_stateless/03222_date_time_inference.reference b/tests/queries/0_stateless/03222_date_time_inference.reference new file mode 100644 index 00000000000..3288308a1d0 --- /dev/null +++ b/tests/queries/0_stateless/03222_date_time_inference.reference @@ -0,0 +1,253 @@ +Date +2020-01-01 Date +2020-01-01 Date +2020-01-01 Date +2020-01-01 Date +2020-01-01 Date +2020-01-01 Date +2020-01-01 Date +2020-01-01 Date +2020-01-01 Date +2020-01-01 Date +2020-01-01 Date +2020-01-01 Date +String +2020_01_01 String +2020_1_01 String +2020_01_1 String +2020_1_1 String +2020a01a01 String +2020a1a01 String +2020a01a1 String +2020a1a1 String +20200101 String +DateTime +2020-01-02 18:42:42 DateTime +2020-01-02 18:42:42 DateTime +2020-01-02 18:42:42 DateTime +String +2020_01_01 42:42:42 String +2020a01a01 42:42:42 String +2020-01-01 42.42.42 String +2020-01-01 42 42 42 String +2020-01-01 42a42a42 String +DateTime64 +2020-01-02 18:42:42.424200000 DateTime64(9) +2020-01-02 18:42:42.424200000 DateTime64(9) +2020-01-02 18:42:42.424200000 DateTime64(9) +String +2020_01_01 42:42:42.4242 String +2020a01a01 42:42:42.4242 String +2020-01-01 42.42.42.4242 String +2020-01-01 42 42 42.4242 String +2020-01-01 42a42a42.4242 String +DateTime/DateTime64 best effort +2000-01-01 00:00:00 DateTime +2000-01-01 01:00:00 DateTime +2000-01-01 01:00:00.000000000 DateTime64(9) +2017-01-01 22:02:03 DateTime +2017-01-01 22:02:03.000000000 DateTime64(9) +2017-01-01 21:02:03 DateTime +2017-01-01 21:02:03.000000000 DateTime64(9) +2017-01-01 22:02:03 DateTime +2017-01-01 22:02:03.000000000 DateTime64(9) +2017-01-02 01:02:03 DateTime +2017-01-02 01:02:03.000000000 DateTime64(9) +1970-01-02 01:02:03 DateTime +1970-01-02 01:02:03.000000000 DateTime64(9) +1970-01-02 01:02:03 DateTime +1970-01-02 01:02:03.000000000 DateTime64(9) +2018-02-11 03:40:50 DateTime +2018-02-11 03:40:50.000000000 DateTime64(9) +2000-04-17 01:02:03 DateTime +2000-04-17 01:02:03.000000000 DateTime64(9) +1970-01-02 01:00:00 DateTime +1970-01-02 01:00:00.000000000 DateTime64(9) +1970-01-02 01:02:03 DateTime +1970-01-02 01:02:03.000000000 DateTime64(9) +1970-01-02 01:02:03 DateTime +1970-01-02 01:02:03.000000000 DateTime64(9) +2015-12-31 20:00:00 DateTime +2015-12-31 20:00:00 DateTime +2016-01-01 00:00:00 DateTime +2016-01-01 00:00:00 DateTime +2017-01-01 22:02:03 DateTime +2017-01-01 22:02:03.000000000 DateTime64(9) +2017-01-02 03:04:05 DateTime +2017-01-02 03:04:05.000000000 DateTime64(9) +2017-01-02 03:04:05 DateTime +2017-01-02 03:04:05.000000000 DateTime64(9) +2017-01-02 03:04:05 DateTime +2017-01-02 03:04:05.000000000 DateTime64(9) +2017-01-02 03:04:05 DateTime +2017-01-02 03:04:05.000000000 DateTime64(9) +2017-01-02 04:04:05 DateTime +2017-01-02 04:04:05.000000000 DateTime64(9) +2017-01-02 02:34:05 DateTime +2017-01-02 02:34:05.000000000 DateTime64(9) +2017-01-02 00:04:05 DateTime +2017-01-02 00:04:05.000000000 DateTime64(9) +2017-01-02 02:04:05 DateTime +2017-01-02 02:04:05.000000000 DateTime64(9) +2017-01-02 00:04:05 DateTime +2017-01-02 00:04:05.000000000 DateTime64(9) +2017-01-01 18:04:05 DateTime +2017-01-01 18:04:05.000000000 DateTime64(9) +2017-01-02 03:04:05 DateTime +2017-01-02 03:04:05.000000000 DateTime64(9) +2017-01-01 23:04:05 DateTime +2017-01-01 23:04:05.000000000 DateTime64(9) +2017-02-01 23:04:05 DateTime +2017-02-01 23:04:05.000000000 DateTime64(9) +2017-06-01 23:04:05 DateTime +2017-06-01 23:04:05.000000000 DateTime64(9) +2017-01-02 00:04:05 DateTime +2017-01-02 00:04:05.000000000 DateTime64(9) +2017-01-02 03:04:05 DateTime +2017-01-02 03:04:05.000000000 DateTime64(9) +2017-01-02 03:04:05 DateTime +2017-01-02 03:04:05.000000000 DateTime64(9) +2017-01-02 04:04:05 DateTime +2017-01-02 04:04:05.000000000 DateTime64(9) +2017-01-02 04:04:05 DateTime +2017-01-02 04:04:05.000000000 DateTime64(9) +2017-01-02 02:04:05 DateTime +2017-01-02 02:04:05.000000000 DateTime64(9) +2017-01-02 03:04:05 DateTime +2017-01-02 03:04:05.000000000 DateTime64(9) +2017-04-01 11:22:33 DateTime +2017-04-01 11:22:33.000000000 DateTime64(9) +2017-04-01 22:02:03 DateTime +2017-04-01 22:02:03.000000000 DateTime64(9) +2017-04-01 22:02:03 DateTime +2017-04-01 22:02:03.000000000 DateTime64(9) +2017-04-02 01:02:03 DateTime +2017-04-02 01:02:03.000000000 DateTime64(9) +2017-04-02 11:22:33 DateTime +2017-04-02 11:22:33.000000000 DateTime64(9) +2017-04-02 01:02:03 DateTime +2017-04-02 01:02:03.000000000 DateTime64(9) +2017-04-02 01:22:33 DateTime +2017-04-02 01:22:33.000000000 DateTime64(9) +2017-04-02 01:02:03 DateTime +2017-04-02 01:02:03.000000000 DateTime64(9) +2017-04-02 01:02:33 DateTime +2017-04-02 01:02:33.000000000 DateTime64(9) +2017-04-01 22:02:03 DateTime +2017-04-01 22:02:03.000000000 DateTime64(9) +2017-04-02 01:02:03 DateTime +2017-04-02 01:02:03.000000000 DateTime64(9) +2017-04-01 22:02:03 DateTime +2017-04-01 22:02:03.000000000 DateTime64(9) +2017-04-01 21:02:03 DateTime +2017-04-01 21:02:03.000000000 DateTime64(9) +2017-04-02 01:02:03 DateTime +2017-04-02 01:02:03.000000000 DateTime64(9) +2017-01-01 22:02:03 DateTime +2017-01-01 22:02:03.000000000 DateTime64(9) +2017-04-25 01:02:03 DateTime +2017-04-25 01:02:03.000000000 DateTime64(9) +2017-04-25 01:02:03 DateTime +2017-04-25 01:02:03.000000000 DateTime64(9) +2017-01-25 01:02:03 DateTime +2017-01-25 01:02:03.000000000 DateTime64(9) +2017-01-24 22:02:03 DateTime +2017-01-24 22:02:03.000000000 DateTime64(9) +2017-01-25 13:02:03 DateTime +2017-01-25 13:02:03.000000000 DateTime64(9) +2017-01-25 01:02:03 DateTime +2017-01-25 01:02:03.000000000 DateTime64(9) +2017-01-25 01:02:03 DateTime +2017-01-25 01:02:03.000000000 DateTime64(9) +2017-01-24 22:02:03 DateTime +2017-01-24 22:02:03.000000000 DateTime64(9) +2017-01-24 22:02:03 DateTime +2017-01-24 22:02:03.000000000 DateTime64(9) +2017-01-25 10:02:03 DateTime +2017-01-25 10:02:03.000000000 DateTime64(9) +2017-01-25 10:02:03 DateTime +2017-01-25 10:02:03.000000000 DateTime64(9) +2017-01-25 10:02:03 DateTime +2017-01-25 10:02:03.000000000 DateTime64(9) +2017-01-25 09:32:03 DateTime +2017-01-25 09:32:03.000000000 DateTime64(9) +2017-01-25 01:02:03 DateTime +2017-01-25 01:02:03.000000000 DateTime64(9) +2017-01-25 13:02:03 DateTime +2017-01-25 13:02:03.000000000 DateTime64(9) +2017-01-25 13:02:03 DateTime +2017-01-25 13:02:03.000000000 DateTime64(9) +2017-01-25 10:02:03 DateTime +2017-01-25 10:02:03.000000000 DateTime64(9) +2018-02-11 03:40:50 DateTime +2018-02-11 03:40:50.000000000 DateTime64(9) +2018-02-11 03:40:50 DateTime +2018-02-11 03:40:50.000000000 DateTime64(9) +String +2 String +20 String +200 String +2000 String +20000 String +200001 String +2000010 String +20000101 String +200001010 String +2000010101 String +20000101010 String +200001010101 String +2000010101010 String +20000101010101 String +2.1 String +20.1 String +200.1 String +2000.1 String +20000.1 String +200001.1 String +2000010.1 String +20000101.1 String +200001010.1 String +2000010101.1 String +20000101010.1 String +200001010101.1 String +2000010101010.1 String +20000101010101.1 String +Mar String +Mar1 String +Mar 1 String +Mar01 String +Mar 01 String +Mar2020 String +Mar 2020 String +Mar012020 String +Mar 012020 String +Mar01012020 String +Mar 01012020 String +Mar0101202001 String +Mar 0101202001 String +Mar010120200101 String +Mar 010120200101 String +Mar01012020010101 String +Mar 01012020010101 String +Mar01012020010101.000 String +Mar 0101202001010101.000 String +2000 01 01 01:00:00 String +2000 01 01 01:00:00.000 String +2000a01a01 01:00:00 String +2000a01a01 01:00:00.000 String +2000-01-01 01 00 00 String +2000-01-01 01 00 00.000 String +2000-01-01 01-00-00 String +2000-01-01 01-00-00.000 String +2000-01-01 01a00a00 String +2000-01-01 01a00a00.000 String +2000-01 01:00:00 String +2000-01 01:00:00.000 String +2000 01 String +2000-01 String +Mar 2000 00:00:00 String +Mar 2000 00:00:00.000 String +2000 00:00:00 String +2000 00:00:00.000 String +Mar 2000-01-01 00:00:00 String +Mar 2000-01-01 00:00:00.000 String diff --git a/tests/queries/0_stateless/03222_date_time_inference.sql b/tests/queries/0_stateless/03222_date_time_inference.sql new file mode 100644 index 00000000000..ebd472294be --- /dev/null +++ b/tests/queries/0_stateless/03222_date_time_inference.sql @@ -0,0 +1,269 @@ +set input_format_try_infer_datetimes = 1; +set input_format_try_infer_dates = 1; +set schema_inference_make_columns_nullable = 0; +set input_format_json_try_infer_numbers_from_strings = 0; +set session_timezone = 'UTC'; + +select 'Date'; +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020:01:01"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020:1:01"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020:01:1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020:1:1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020-01-01"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020-1-01"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020-01-1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020-1-1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020/01/01"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020/1/01"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020/01/1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020/1/1"}'); + +select 'String'; +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020_01_01"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020_1_01"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020_01_1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020_1_1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020a01a01"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020a1a01"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020a01a1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020a1a1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "20200101"}'); + +select 'DateTime'; +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020:01:01 42:42:42"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020/01/01 42:42:42"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020-01-01 42:42:42"}'); + +select 'String'; +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020_01_01 42:42:42"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020a01a01 42:42:42"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020-01-01 42.42.42"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020-01-01 42 42 42"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020-01-01 42a42a42"}'); + +select 'DateTime64'; +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020:01:01 42:42:42.4242"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020/01/01 42:42:42.4242"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020-01-01 42:42:42.4242"}'); + +select 'String'; +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020_01_01 42:42:42.4242"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020a01a01 42:42:42.4242"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020-01-01 42.42.42.4242"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020-01-01 42 42 42.4242"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2020-01-01 42a42a42.4242"}'); + +set date_time_input_format='best_effort'; +select 'DateTime/DateTime64 best effort'; +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000-01-01 00:00:00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000-01-01 01:00:00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000-01-01 01:00:00.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "02/01/17 010203 MSK"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "02/01/17 010203.000 MSK"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "02/01/17 010203 MSK+0100"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "02/01/17 010203.000 MSK+0100"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "02/01/17 010203 UTC+0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "02/01/17 010203.000 UTC+0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "02/01/17 010203Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "02/01/17 010203.000Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "02/01/1970 010203Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "02/01/1970 010203.000Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "02/01/70 010203Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "02/01/70 010203.000Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "11 Feb 2018 06:40:50 +0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "11 Feb 2018 06:40:50.000 +0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "17 Apr 2000 2 1:2:3"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "17 Apr 2000 2 1:2:3.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "19700102 01:00:00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "19700102 01:00:00.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "19700102010203Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "19700102010203Z.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "1970/01/02 010203Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "1970/01/02 010203.000Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2016-01-01MSD"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2016-01-01 MSD"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2016-01-01UTC"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2016-01-01Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "201701 02 010203 UTC+0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "201701 02 010203.000 UTC+0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05+0"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05.000+0"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05+00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05.000+00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05+0000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05.000+0000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05 -0100"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05.000 -0100"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05+030"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05.000+030"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05+0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05.000+0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05+1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05.000+1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05+300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05.000+300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05+900"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05.000+900"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05GMT"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05.000GMT"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05 MSD"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05.000 MSD"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05 MSD Feb"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05.000 MSD Feb"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05 MSD Jun"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05.000 MSD Jun"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05 MSK"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02 03:04:05.000 MSK"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02T03:04:05"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02T03:04:05.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02T03:04:05+00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02T03:04:05.000+00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02T03:04:05 -0100"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02T03:04:05.000 -0100"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02T03:04:05-0100"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02T03:04:05.000-0100"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02T03:04:05+0100"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02T03:04:05.000+0100"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02T03:04:05Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017-01-02T03:04:05.000Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 01 11:22:33"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 01 11:22:33.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 010203 UTC+0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 010203.000 UTC+0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 01:2:3 UTC+0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 01:2:3.000 UTC+0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:02:3"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:02:3.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 11:22:33"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 11:22:33.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:2:03"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:2:03.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:22:33"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:22:33.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:2:3"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:2:3.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:2:33"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:2:33.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:2:3 MSK"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:2:3.000 MSK"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:2:3 UTC+0000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:2:3.000 UTC+0000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:2:3 UTC+0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:2:3.000 UTC+0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:2:3 UTC+0400"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 02 1:2:3.000 UTC+0400"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 2 1:2:3"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Apr 2 1:2:3.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Jan 02 010203 UTC+0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2017 Jan 02 010203.000 UTC+0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Apr 2017 01:02:03"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Apr 2017 01:02:03.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Apr 2017 1:2:3"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Apr 2017 1:2:3.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3 MSK"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3.000 MSK"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3 PM"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3.000 PM"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3.000Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3 Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3.000 Z"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3 Z +0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3.000 Z +0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3 Z+03:00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3.000 Z+03:00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3 Z +03:00 PM"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3.000 Z +03:00 PM"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3 Z +0300 PM"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3.000 Z +0300 PM"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3 Z+03:00 PM"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3.000 Z+03:00 PM"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3 Z +03:30 PM"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3.000 Z +03:30 PM"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3Z Mon"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3.000Z Mon"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3 Z PM"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3.000 Z PM"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3Z PM"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3.000Z PM"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3 Z PM +03:00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "25 Jan 2017 1:2:3.000 Z PM +03:00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Sun 11 Feb 2018 06:40:50 +0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Sun 11 Feb 2018 06:40:50.000 +0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Sun, 11 Feb 2018 06:40:50 +0300"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Sun, 11 Feb 2018 06:40:50.000 +0300"}'); + +select 'String'; +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "20"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "200"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "20000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "200001"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000010"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "20000101"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "200001010"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000010101"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "20000101010"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "200001010101"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000010101010"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "20000101010101"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2.1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "20.1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "200.1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000.1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "20000.1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "200001.1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000010.1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "20000101.1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "200001010.1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000010101.1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "20000101010.1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "200001010101.1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000010101010.1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "20000101010101.1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar 1"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar01"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar 01"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar2020"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar 2020"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar012020"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar 012020"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar01012020"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar 01012020"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar0101202001"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar 0101202001"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar010120200101"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar 010120200101"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar01012020010101"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar 01012020010101"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar01012020010101.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar 0101202001010101.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000 01 01 01:00:00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000 01 01 01:00:00.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000a01a01 01:00:00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000a01a01 01:00:00.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000-01-01 01 00 00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000-01-01 01 00 00.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000-01-01 01-00-00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000-01-01 01-00-00.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000-01-01 01a00a00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000-01-01 01a00a00.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000-01 01:00:00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000-01 01:00:00.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000 01"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000-01"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar 2000 00:00:00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar 2000 00:00:00.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000 00:00:00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "2000 00:00:00.000"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar 2000-01-01 00:00:00"}'); +select x, toTypeName(x) from format(JSONEachRow, '{"x" : "Mar 2000-01-01 00:00:00.000"}'); + + diff --git a/tests/queries/0_stateless/03222_parallel_replicas_min_marks_to_read_overflow.reference b/tests/queries/0_stateless/03222_parallel_replicas_min_marks_to_read_overflow.reference new file mode 100644 index 00000000000..b6c452ba328 --- /dev/null +++ b/tests/queries/0_stateless/03222_parallel_replicas_min_marks_to_read_overflow.reference @@ -0,0 +1,21 @@ +1006 +1007 +1008 +1009 +101 +1010 +1011 +1012 +1013 +1014 +--- +100 100 +101 101 +102 102 +103 103 +104 104 +105 105 +106 106 +107 107 +108 108 +109 109 diff --git a/tests/queries/0_stateless/03222_parallel_replicas_min_marks_to_read_overflow.sql b/tests/queries/0_stateless/03222_parallel_replicas_min_marks_to_read_overflow.sql new file mode 100644 index 00000000000..6f486f8f0fe --- /dev/null +++ b/tests/queries/0_stateless/03222_parallel_replicas_min_marks_to_read_overflow.sql @@ -0,0 +1,23 @@ +DROP TABLE IF EXISTS test__fuzz_22 SYNC; + +CREATE TABLE test__fuzz_22 (k Float32, v String) ENGINE = ReplicatedMergeTree('/clickhouse/03222/{database}/test__fuzz_22', 'r1') ORDER BY k SETTINGS index_granularity = 1; + +INSERT INTO test__fuzz_22 SELECT number, toString(number) FROM numbers(10_000); + +SET allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 3, cluster_for_parallel_replicas='test_cluster_one_shard_three_replicas_localhost'; + +SELECT v +FROM test__fuzz_22 +ORDER BY v +LIMIT 10, 10 +SETTINGS merge_tree_min_rows_for_concurrent_read = 9223372036854775806; + +SELECT '---'; + +SELECT k, v +FROM test__fuzz_22 +ORDER BY k +LIMIT 100, 10 +SETTINGS optimize_read_in_order=1, merge_tree_min_rows_for_concurrent_read = 9223372036854775806; + +DROP TABLE test__fuzz_22 SYNC; diff --git a/tests/queries/0_stateless/03223_analyzer_with_cube_fuzz.reference b/tests/queries/0_stateless/03223_analyzer_with_cube_fuzz.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03223_analyzer_with_cube_fuzz.sql b/tests/queries/0_stateless/03223_analyzer_with_cube_fuzz.sql new file mode 100644 index 00000000000..f3bccc79b3f --- /dev/null +++ b/tests/queries/0_stateless/03223_analyzer_with_cube_fuzz.sql @@ -0,0 +1,29 @@ +SET enable_analyzer = 1; + +DROP TABLE IF EXISTS t1; +DROP TABLE IF EXISTS t2; + +CREATE TABLE t1 (`a` Int64, `b` Int64) ENGINE = MergeTree ORDER BY a; +CREATE TABLE t2 (`key` Int32, `val` Int64) ENGINE = MergeTree ORDER BY key; +insert into t1 Select number, number from numbers(100000); +insert into t2 Select number, number from numbers(100000); + + +SELECT + 1 * 1000.0001, + (count(1.) = -2147483647) AND (count(a) = 1.1920928955078125e-7) AND (count(val) = 1048577) AND (sum(val) = ((NULL * 1048576) / -9223372036854775807)) AND (sum(a) = ((9223372036854775806 * 10000000000.) / 1048575)) +FROM +( + SELECT + a, + val + FROM t1 + FULL OUTER JOIN t2 ON (t1.a = t2.key) OR (1 * inf) OR (t1.b = t2.key) +) +GROUP BY '65537' + WITH CUBE +FORMAT Null +SETTINGS max_block_size = 100, join_use_nulls = 1, max_execution_time = 1., max_result_rows = 0, max_result_bytes = 0; -- { serverError TIMEOUT_EXCEEDED } + +DROP TABLE t1; +DROP TABLE t2; diff --git a/tests/queries/0_stateless/data_hive/partitioning/array=[1,2,3]/float=42.42/sample.parquet b/tests/queries/0_stateless/data_hive/partitioning/array=[1,2,3]/float=42.42/sample.parquet new file mode 100644 index 00000000000..9b6a78cf8cc Binary files /dev/null and b/tests/queries/0_stateless/data_hive/partitioning/array=[1,2,3]/float=42.42/sample.parquet differ diff --git a/tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/column0=Elizabeth/sample.parquet b/tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/column0=Elizabeth/sample.parquet new file mode 100644 index 00000000000..9b6a78cf8cc Binary files /dev/null and b/tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/column0=Elizabeth/sample.parquet differ diff --git a/tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/column1=Gordon/sample.parquet b/tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/column1=Gordon/sample.parquet new file mode 100644 index 00000000000..9b6a78cf8cc Binary files /dev/null and b/tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/column1=Gordon/sample.parquet differ diff --git a/tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet b/tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet new file mode 100644 index 00000000000..9b6a78cf8cc Binary files /dev/null and b/tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet differ diff --git a/tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/sample.parquet b/tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/sample.parquet new file mode 100644 index 00000000000..9b6a78cf8cc Binary files /dev/null and b/tests/queries/0_stateless/data_hive/partitioning/column0=Elizabeth/sample.parquet differ diff --git a/tests/queries/0_stateless/data_hive/partitioning/identifier=2070/email.csv b/tests/queries/0_stateless/data_hive/partitioning/identifier=2070/email.csv new file mode 100644 index 00000000000..936d995cc64 --- /dev/null +++ b/tests/queries/0_stateless/data_hive/partitioning/identifier=2070/email.csv @@ -0,0 +1,5 @@ +_login_email,_identifier,_first_name,_last_name +laura@example.com,2070,Laura,Grey +craig@example.com,4081,Craig,Johnson +mary@example.com,9346,Mary,Jenkins +jamie@example.com,5079,Jamie,Smith diff --git a/tests/queries/0_stateless/data_hive/partitioning/non_existing_column=Elizabeth/sample.parquet b/tests/queries/0_stateless/data_hive/partitioning/non_existing_column=Elizabeth/sample.parquet new file mode 100644 index 00000000000..9b6a78cf8cc Binary files /dev/null and b/tests/queries/0_stateless/data_hive/partitioning/non_existing_column=Elizabeth/sample.parquet differ diff --git a/tests/queries/0_stateless/data_hive/partitioning/number=42/date=2020-01-01/sample.parquet b/tests/queries/0_stateless/data_hive/partitioning/number=42/date=2020-01-01/sample.parquet new file mode 100644 index 00000000000..9b6a78cf8cc Binary files /dev/null and b/tests/queries/0_stateless/data_hive/partitioning/number=42/date=2020-01-01/sample.parquet 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 deleted file mode 100644 index ad8f7cb4507..00000000000 Binary files a/tests/queries/0_stateless/data_json/key_ignore_case.json and /dev/null differ diff --git a/tests/queries/0_stateless/data_minio/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet b/tests/queries/0_stateless/data_minio/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet new file mode 100644 index 00000000000..9b6a78cf8cc Binary files /dev/null and b/tests/queries/0_stateless/data_minio/hive_partitioning/column0=Elizabeth/column1=Gordon/sample.parquet differ diff --git a/tests/queries/0_stateless/data_minio/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet b/tests/queries/0_stateless/data_minio/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet new file mode 100644 index 00000000000..9b6a78cf8cc Binary files /dev/null and b/tests/queries/0_stateless/data_minio/hive_partitioning/column0=Elizabeth/column1=Schmidt/sample.parquet differ diff --git a/tests/queries/0_stateless/data_minio/hive_partitioning/column0=Elizabeth/sample.parquet b/tests/queries/0_stateless/data_minio/hive_partitioning/column0=Elizabeth/sample.parquet new file mode 100644 index 00000000000..9b6a78cf8cc Binary files /dev/null and b/tests/queries/0_stateless/data_minio/hive_partitioning/column0=Elizabeth/sample.parquet differ diff --git a/tests/queries/0_stateless/data_minio/hive_partitioning/non_existing_column=Elizabeth/sample.parquet b/tests/queries/0_stateless/data_minio/hive_partitioning/non_existing_column=Elizabeth/sample.parquet new file mode 100644 index 00000000000..9b6a78cf8cc Binary files /dev/null and b/tests/queries/0_stateless/data_minio/hive_partitioning/non_existing_column=Elizabeth/sample.parquet differ diff --git a/tests/queries/0_stateless/data_minio/test.zip::03215_archive.csv b/tests/queries/0_stateless/data_minio/test.zip::03215_archive.csv deleted file mode 100644 index d00491fd7e5..00000000000 --- a/tests/queries/0_stateless/data_minio/test.zip::03215_archive.csv +++ /dev/null @@ -1 +0,0 @@ -1 diff --git a/tests/queries/0_stateless/data_minio/test::03215_archive.csv b/tests/queries/0_stateless/data_minio/test::03215_archive.csv deleted file mode 100644 index d00491fd7e5..00000000000 --- a/tests/queries/0_stateless/data_minio/test::03215_archive.csv +++ /dev/null @@ -1 +0,0 @@ -1 diff --git a/tests/queries/0_stateless/helpers/pure_http_client.py b/tests/queries/0_stateless/helpers/pure_http_client.py index a31a91e0550..c3c4109ce5b 100644 --- a/tests/queries/0_stateless/helpers/pure_http_client.py +++ b/tests/queries/0_stateless/helpers/pure_http_client.py @@ -19,9 +19,14 @@ class ClickHouseClient: self.host = host def query( - self, query, connection_timeout=500, settings=dict(), binary_result=False + self, + query, + connection_timeout=500, + settings=dict(), + binary_result=False, + with_retries=True, ): - NUMBER_OF_TRIES = 30 + NUMBER_OF_TRIES = 30 if with_retries else 1 DELAY = 10 params = { @@ -40,7 +45,8 @@ class ClickHouseClient: if r.status_code == 200: return r.content if binary_result else r.text else: - print("ATTENTION: try #%d failed" % i) + if with_retries: + print("ATTENTION: try #%d failed" % i) if i != (NUMBER_OF_TRIES - 1): print(query) print(r.text) diff --git a/tests/queries/1_stateful/00081_group_by_without_key_and_totals.sql b/tests/queries/1_stateful/00081_group_by_without_key_and_totals.sql index 03e9e0feb40..1d10a2b9212 100644 --- a/tests/queries/1_stateful/00081_group_by_without_key_and_totals.sql +++ b/tests/queries/1_stateful/00081_group_by_without_key_and_totals.sql @@ -1,4 +1,4 @@ -SET allow_experimental_analyzer = 1; +SET enable_analyzer = 1; SELECT count() AS c FROM test.hits WHERE CounterID = 1704509 WITH TOTALS SETTINGS totals_mode = 'before_having', max_rows_to_group_by = 100000, group_by_overflow_mode = 'any'; SELECT count() AS c FROM test.hits WHERE CounterID = 1704509 WITH TOTALS SETTINGS totals_mode = 'after_having_inclusive', max_rows_to_group_by = 100000, group_by_overflow_mode = 'any'; @@ -14,4 +14,3 @@ SELECT TraficSourceID AS k, count() AS c FROM test.hits WHERE CounterID = 170450 SELECT TraficSourceID AS k, count() AS c FROM test.hits WHERE CounterID = 1704509 GROUP BY k WITH TOTALS ORDER BY k SETTINGS totals_mode = 'after_having_inclusive', max_rows_to_group_by = 100000, group_by_overflow_mode = 'any'; SELECT TraficSourceID AS k, count() AS c FROM test.hits WHERE CounterID = 1704509 GROUP BY k WITH TOTALS ORDER BY k SETTINGS totals_mode = 'after_having_exclusive', max_rows_to_group_by = 100000, group_by_overflow_mode = 'any'; SELECT TraficSourceID AS k, count() AS c FROM test.hits WHERE CounterID = 1704509 GROUP BY k WITH TOTALS ORDER BY k SETTINGS totals_mode = 'after_having_auto', max_rows_to_group_by = 100000, group_by_overflow_mode = 'any'; - diff --git a/tests/queries/1_stateful/00172_early_constant_folding.sql b/tests/queries/1_stateful/00172_early_constant_folding.sql index b4b58ba3cb0..343e87a26d1 100644 --- a/tests/queries/1_stateful/00172_early_constant_folding.sql +++ b/tests/queries/1_stateful/00172_early_constant_folding.sql @@ -2,5 +2,5 @@ set max_threads=10; set optimize_use_implicit_projections=1; -EXPLAIN PIPELINE SELECT count(JavaEnable) FROM test.hits WHERE WatchID = 1 OR Title = 'next' OR URL = 'prev' OR URL = '???' OR 1 SETTINGS allow_experimental_analyzer = 0; -EXPLAIN PIPELINE SELECT count(JavaEnable) FROM test.hits WHERE WatchID = 1 OR Title = 'next' OR URL = 'prev' OR URL = '???' OR 1 SETTINGS allow_experimental_analyzer = 1; +EXPLAIN PIPELINE SELECT count(JavaEnable) FROM test.hits WHERE WatchID = 1 OR Title = 'next' OR URL = 'prev' OR URL = '???' OR 1 SETTINGS enable_analyzer = 0; +EXPLAIN PIPELINE SELECT count(JavaEnable) FROM test.hits WHERE WatchID = 1 OR Title = 'next' OR URL = 'prev' OR URL = '???' OR 1 SETTINGS enable_analyzer = 1; diff --git a/tests/queries/1_stateful/00173_group_by_use_nulls.reference b/tests/queries/1_stateful/00173_group_by_use_nulls.reference index e82b996ad3c..4f8d10a9221 100644 --- a/tests/queries/1_stateful/00173_group_by_use_nulls.reference +++ b/tests/queries/1_stateful/00173_group_by_use_nulls.reference @@ -9,7 +9,7 @@ 33010362 1336 800784 1336 -- { echoOn } -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; SELECT CounterID AS k, quantileBFloat16(0.5)(ResolutionWidth) diff --git a/tests/queries/1_stateful/00173_group_by_use_nulls.sql b/tests/queries/1_stateful/00173_group_by_use_nulls.sql index 8531e9efaf8..ed537bb289a 100644 --- a/tests/queries/1_stateful/00173_group_by_use_nulls.sql +++ b/tests/queries/1_stateful/00173_group_by_use_nulls.sql @@ -21,7 +21,7 @@ LIMIT 10 SETTINGS group_by_use_nulls = 1 FORMAT Null; -- { echoOn } -set allow_experimental_analyzer = 1; +set enable_analyzer = 1; SELECT CounterID AS k, diff --git a/tests/queries/shell_config.sh b/tests/queries/shell_config.sh index f7017958635..9e5bf75d335 100644 --- a/tests/queries/shell_config.sh +++ b/tests/queries/shell_config.sh @@ -1,9 +1,6 @@ #!/usr/bin/env bash # shellcheck disable=SC2120 -# Don't check for ODR violation, since we may test shared build with ASAN -export ASAN_OPTIONS=detect_odr_violation=0 - # If ClickHouse was built with coverage - dump the coverage information at exit # (in other cases this environment variable has no effect) export CLICKHOUSE_WRITE_COVERAGE="coverage" diff --git a/tests/tsan_ignorelist.txt b/tests/tsan_ignorelist.txt index 96bf6e4251f..2a31fc9bc15 100644 --- a/tests/tsan_ignorelist.txt +++ b/tests/tsan_ignorelist.txt @@ -5,11 +5,9 @@ # # Caveats for generic entry "fun": # - does not work for __attribute__((__always_inline__)) +# - and may not work for functions that had been inlined # - requires asterisk at the beginning *and* end for static functions # [thread] # https://github.com/ClickHouse/ClickHouse/issues/55629 fun:rd_kafka_broker_set_nodename -# https://github.com/ClickHouse/ClickHouse/issues/60443 -fun:*rd_avg_calc* -fun:*rd_avg_rollover* diff --git a/tests/ubsan_ignorelist.txt b/tests/ubsan_ignorelist.txt index 57d6598afa6..b75819b3f4b 100644 --- a/tests/ubsan_ignorelist.txt +++ b/tests/ubsan_ignorelist.txt @@ -9,6 +9,7 @@ # # Caveats for generic entry "fun": # - does not work for __attribute__((__always_inline__)) +# - and may not work for functions that had been inlined # - requires asterisk at the beginning *and* end for static functions # [undefined] diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt index 8a9a8d2e76c..ce7c666912e 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 2942 +personal_ws-1.1 en 2942 AArch ACLs ALTERs @@ -247,6 +247,7 @@ DoubleDelta Doxygen Durre ECMA +ETag Ecto EdgeAngle EdgeLengthKm @@ -561,6 +562,7 @@ MindsDB Mongodb Monotonicity MsgPack +MultiLineString MultiPolygon Multiline Multiqueries @@ -1587,6 +1589,7 @@ enum's enums erfc errorCodeToName +etag evalMLMethod exFAT expiryMsec @@ -1672,6 +1675,7 @@ fuzzQuery fuzzer fuzzers gRPC +gaugehistogram gccMurmurHash gcem generateRandom @@ -1698,6 +1702,8 @@ getOSKernelVersion getServerPort getSetting getSizeOfEnumType +getSubcolumn +getTypeSerializationStreams getblockinfo getevents ghcnd @@ -2087,6 +2093,7 @@ multiSearchFirstPositionUTF multibyte multidirectory multiline +multilinestring multiplyDecimal multipolygon multisearchany @@ -2110,6 +2117,7 @@ namenode namepassword nameprofile namequota +namespace namespaces natively nats @@ -2364,6 +2372,7 @@ rankCorr rapidjson rawblob readWKTLineString +readWKTMultiLineString readWKTMultiPolygon readWKTPoint readWKTPolygon @@ -2556,6 +2565,7 @@ startsWithUTF startswith statbox stateful +stateset stddev stddevPop stddevPopStable @@ -2687,6 +2697,10 @@ themself threadpool throwIf timeDiff +TimeSeries +timeSeriesData +timeSeriesMetrics +timeSeriesTags timeSlot timeSlots timeZone diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv index 75c10fa67b8..71a4a722a36 100644 --- a/utils/list-versions/version_date.tsv +++ b/utils/list-versions/version_date.tsv @@ -1,5 +1,7 @@ +v24.7.3.42-stable 2024-08-08 v24.7.2.13-stable 2024-08-01 v24.7.1.2915-stable 2024-07-30 +v24.6.3.95-stable 2024-08-06 v24.6.2.17-stable 2024-07-05 v24.6.1.4423-stable 2024-07-01 v24.5.5.78-stable 2024-08-05 @@ -11,6 +13,7 @@ v24.4.4.113-stable 2024-08-02 v24.4.3.25-stable 2024-06-14 v24.4.2.141-stable 2024-06-07 v24.4.1.2088-stable 2024-05-01 +v24.3.7.30-lts 2024-08-14 v24.3.6.48-lts 2024-08-02 v24.3.5.46-lts 2024-07-03 v24.3.4.147-lts 2024-06-13