Merge branch 'master' into parallel-s3-downloading

.github/workflows/master.yml

@@ -360,6 +360,52 @@ jobs:
           docker kill "$(docker ps -q)" ||:
           docker rm -f "$(docker ps -a -q)" ||:
           sudo rm -fr "$TEMP_PATH" "$CACHES_PATH"
+  BuilderBinGCC:
+    needs: [DockerHubPush]
+    if: ${{ !contains(github.event.pull_request.labels.*.name, 'pr-documentation') && !contains(github.event.pull_request.labels.*.name, 'pr-doc-fix') }}
+    runs-on: [self-hosted, builder]
+    steps:
+      - name: Set envs
+        run: |
+          cat >> "$GITHUB_ENV" << 'EOF'
+          TEMP_PATH=${{runner.temp}}/build_check
+          IMAGES_PATH=${{runner.temp}}/images_path
+          REPO_COPY=${{runner.temp}}/build_check/ClickHouse
+          CACHES_PATH=${{runner.temp}}/../ccaches
+          CHECK_NAME=ClickHouse build check (actions)
+          BUILD_NAME=binary_gcc
+          EOF
+      - name: Download changed images
+        uses: actions/download-artifact@v2
+        with:
+          name: changed_images
+          path: ${{ env.IMAGES_PATH }}
+      - name: Clear repository
+        run: |
+          sudo rm -fr "$GITHUB_WORKSPACE" && mkdir "$GITHUB_WORKSPACE"
+      - name: Check out repository code
+        uses: actions/checkout@v2
+        with:
+          submodules: 'true'
+          fetch-depth: 0 # otherwise we will have no info about contributors
+      - name: Build
+        run: |
+          sudo rm -fr "$TEMP_PATH"
+          mkdir -p "$TEMP_PATH"
+          cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
+          cd "$REPO_COPY/tests/ci" && python3 build_check.py "$CHECK_NAME" "$BUILD_NAME"
+      - name: Upload build URLs to artifacts
+        if: ${{ success() || failure() }}
+        uses: actions/upload-artifact@v2
+        with:
+          name: ${{ env.BUILD_NAME }}
+          path: ${{ env.TEMP_PATH }}/${{ env.BUILD_NAME }}.json
+      - name: Cleanup
+        if: always()
+        run: |
+          docker kill "$(docker ps -q)" ||:
+          docker rm -f "$(docker ps -a -q)" ||:
+          sudo rm -fr "$TEMP_PATH" "$CACHES_PATH"
   BuilderDebAsan:
     needs: [DockerHubPush]
     runs-on: [self-hosted, builder]
@@ -918,6 +964,7 @@ jobs:
       - BuilderDebRelease
       - BuilderDebAarch64
       - BuilderBinRelease
+      - BuilderBinGCC
       - BuilderDebAsan
       - BuilderDebTsan
       - BuilderDebUBsan
@@ -2608,6 +2655,40 @@ jobs:
           docker kill "$(docker ps -q)" ||:
           docker rm -f "$(docker ps -a -q)" ||:
           sudo rm -fr "$TEMP_PATH"
+  UnitTestsReleaseGCC:
+    needs: [BuilderBinGCC]
+    runs-on: [self-hosted, fuzzer-unit-tester]
+    steps:
+      - name: Set envs
+        run: |
+          cat >> "$GITHUB_ENV" << 'EOF'
+          TEMP_PATH=${{runner.temp}}/unit_tests_asan
+          REPORTS_PATH=${{runner.temp}}/reports_dir
+          CHECK_NAME=Unit tests (release-gcc, actions)
+          REPO_COPY=${{runner.temp}}/unit_tests_asan/ClickHouse
+          EOF
+      - name: Download json reports
+        uses: actions/download-artifact@v2
+        with:
+          path: ${{ env.REPORTS_PATH }}
+      - name: Clear repository
+        run: |
+          sudo rm -fr "$GITHUB_WORKSPACE" && mkdir "$GITHUB_WORKSPACE"
+      - name: Check out repository code
+        uses: actions/checkout@v2
+      - name: Unit test
+        run: |
+          sudo rm -fr "$TEMP_PATH"
+          mkdir -p "$TEMP_PATH"
+          cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
+          cd "$REPO_COPY/tests/ci"
+          python3 unit_tests_check.py "$CHECK_NAME"
+      - name: Cleanup
+        if: always()
+        run: |
+          docker kill "$(docker ps -q)" ||:
+          docker rm -f "$(docker ps -a -q)" ||:
+          sudo rm -fr "$TEMP_PATH"
   UnitTestsTsan:
     needs: [BuilderDebTsan]
     runs-on: [self-hosted, fuzzer-unit-tester]

.github/workflows/pull_request.yml

@@ -370,6 +370,48 @@ jobs:
           docker kill "$(docker ps -q)" ||:
           docker rm -f "$(docker ps -a -q)" ||:
           sudo rm -fr "$TEMP_PATH" "$CACHES_PATH"
+  BuilderBinGCC:
+    needs: [DockerHubPush, FastTest]
+    runs-on: [self-hosted, builder]
+    steps:
+      - name: Set envs
+        run: |
+          cat >> "$GITHUB_ENV" << 'EOF'
+          TEMP_PATH=${{runner.temp}}/build_check
+          IMAGES_PATH=${{runner.temp}}/images_path
+          REPO_COPY=${{runner.temp}}/build_check/ClickHouse
+          CACHES_PATH=${{runner.temp}}/../ccaches
+          CHECK_NAME=ClickHouse build check (actions)
+          BUILD_NAME=binary_gcc
+          EOF
+      - name: Download changed images
+        uses: actions/download-artifact@v2
+        with:
+          name: changed_images
+          path: ${{ runner.temp }}/images_path
+      - name: Check out repository code
+        uses: actions/checkout@v2
+        with:
+          submodules: 'true'
+          fetch-depth: 0 # otherwise we will have no info about contributors
+      - name: Build
+        run: |
+          sudo rm -fr "$TEMP_PATH"
+          mkdir -p "$TEMP_PATH"
+          cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
+          cd "$REPO_COPY/tests/ci" && python3 build_check.py "$CHECK_NAME" "$BUILD_NAME"
+      - name: Upload build URLs to artifacts
+        if: ${{ success() || failure() }}
+        uses: actions/upload-artifact@v2
+        with:
+          name: ${{ env.BUILD_NAME }}
+          path: ${{ runner.temp }}/build_check/${{ env.BUILD_NAME }}.json
+      - name: Cleanup
+        if: always()
+        run: |
+          docker kill "$(docker ps -q)" ||:
+          docker rm -f "$(docker ps -a -q)" ||:
+          sudo rm -fr "$TEMP_PATH" "$CACHES_PATH"
   BuilderDebAarch64:
     needs: [DockerHubPush, FastTest]
     runs-on: [self-hosted, builder]
@@ -963,6 +1005,7 @@ jobs:
       - BuilderDebRelease
       - BuilderDebAarch64
       - BuilderBinRelease
+      - BuilderBinGCC
       - BuilderDebAsan
       - BuilderDebTsan
       - BuilderDebUBsan
@@ -2808,6 +2851,40 @@ jobs:
           docker kill "$(docker ps -q)" ||:
           docker rm -f "$(docker ps -a -q)" ||:
           sudo rm -fr "$TEMP_PATH"
+  UnitTestsReleaseGCC:
+    needs: [BuilderBinGCC]
+    runs-on: [self-hosted, fuzzer-unit-tester]
+    steps:
+      - name: Set envs
+        run: |
+          cat >> "$GITHUB_ENV" << 'EOF'
+          TEMP_PATH=${{runner.temp}}/unit_tests_asan
+          REPORTS_PATH=${{runner.temp}}/reports_dir
+          CHECK_NAME=Unit tests (release-gcc, actions)
+          REPO_COPY=${{runner.temp}}/unit_tests_asan/ClickHouse
+          EOF
+      - name: Download json reports
+        uses: actions/download-artifact@v2
+        with:
+          path: ${{ env.REPORTS_PATH }}
+      - name: Clear repository
+        run: |
+          sudo rm -fr "$GITHUB_WORKSPACE" && mkdir "$GITHUB_WORKSPACE"
+      - name: Check out repository code
+        uses: actions/checkout@v2
+      - name: Unit test
+        run: |
+          sudo rm -fr "$TEMP_PATH"
+          mkdir -p "$TEMP_PATH"
+          cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
+          cd "$REPO_COPY/tests/ci"
+          python3 unit_tests_check.py "$CHECK_NAME"
+      - name: Cleanup
+        if: always()
+        run: |
+          docker kill "$(docker ps -q)" ||:
+          docker rm -f "$(docker ps -a -q)" ||:
+          sudo rm -fr "$TEMP_PATH"
   UnitTestsTsan:
     needs: [BuilderDebTsan]
     runs-on: [self-hosted, fuzzer-unit-tester]

CMakeLists.txt

@@ -261,7 +261,7 @@ endif ()
 # Add a section with the hash of the compiled machine code for integrity checks.
 # Only for official builds, because adding a section can be time consuming (rewrite of several GB).
 # And cross compiled binaries are not supported (since you cannot execute clickhouse hash-binary)
-if (OBJCOPY_PATH AND YANDEX_OFFICIAL_BUILD AND (NOT CMAKE_TOOLCHAIN_FILE))
+if (OBJCOPY_PATH AND CLICKHOUSE_OFFICIAL_BUILD AND (NOT CMAKE_TOOLCHAIN_FILE))
     set (USE_BINARY_HASH 1)
 endif ()
 

base/glibc-compatibility/CMakeLists.txt

@@ -51,6 +51,6 @@ if (GLIBC_COMPATIBILITY)
 
     message (STATUS "Some symbols from glibc will be replaced for compatibility")
 
-elseif (YANDEX_OFFICIAL_BUILD)
+elseif (CLICKHOUSE_OFFICIAL_BUILD)
     message (WARNING "Option GLIBC_COMPATIBILITY must be turned on for production builds.")
 endif ()

cmake/version.cmake

@@ -18,6 +18,6 @@ set (VERSION_STRING_SHORT "${VERSION_MAJOR}.${VERSION_MINOR}")
 
 math (EXPR VERSION_INTEGER "${VERSION_PATCH} + ${VERSION_MINOR}*1000 + ${VERSION_MAJOR}*1000000")
 
-if(YANDEX_OFFICIAL_BUILD)
+if(CLICKHOUSE_OFFICIAL_BUILD)
     set(VERSION_OFFICIAL " (official build)")
 endif()

contrib/avro-cmake/CMakeLists.txt

@@ -69,9 +69,10 @@ endif ()
 target_compile_options(_avrocpp PRIVATE ${SUPPRESS_WARNINGS})
 
 # create a symlink to include headers with <avro/...>
+set(AVRO_INCLUDE_DIR "${CMAKE_CURRENT_BINARY_DIR}/include")
 ADD_CUSTOM_TARGET(avro_symlink_headers ALL
-    COMMAND ${CMAKE_COMMAND} -E make_directory "${AVROCPP_ROOT_DIR}/include"
-    COMMAND ${CMAKE_COMMAND} -E create_symlink "${AVROCPP_ROOT_DIR}/api" "${AVROCPP_ROOT_DIR}/include/avro"
+    COMMAND ${CMAKE_COMMAND} -E make_directory "${AVRO_INCLUDE_DIR}"
+    COMMAND ${CMAKE_COMMAND} -E create_symlink "${AVROCPP_ROOT_DIR}/api" "${AVRO_INCLUDE_DIR}/avro"
 )
 add_dependencies(_avrocpp avro_symlink_headers)
-target_include_directories(_avrocpp SYSTEM BEFORE PUBLIC "${AVROCPP_ROOT_DIR}/include")
+target_include_directories(_avrocpp SYSTEM BEFORE PUBLIC "${AVRO_INCLUDE_DIR}")

contrib/boost-cmake/CMakeLists.txt

@@ -27,7 +27,11 @@ target_include_directories (_boost_headers_only SYSTEM BEFORE INTERFACE ${LIBRAR
 
 # asio
 
-target_compile_definitions (_boost_headers_only INTERFACE BOOST_ASIO_STANDALONE=1)
+target_compile_definitions (_boost_headers_only INTERFACE
+  BOOST_ASIO_STANDALONE=1
+  # Avoid using of deprecated in c++ > 17 std::result_of
+  BOOST_ASIO_HAS_STD_INVOKE_RESULT=1
+)
 
 # iostreams
 

docker/packager/packager

@@ -163,6 +163,7 @@ def parse_env_variables(
         cmake_flags.append("-DCMAKE_INSTALL_PREFIX=/usr")
         cmake_flags.append("-DCMAKE_INSTALL_SYSCONFDIR=/etc")
         cmake_flags.append("-DCMAKE_INSTALL_LOCALSTATEDIR=/var")
+        cmake_flags.append("-DBUILD_STANDALONE_KEEPER=ON")
         if is_release_build(build_type, package_type, sanitizer, split_binary):
             cmake_flags.append("-DINSTALL_STRIPPED_BINARIES=ON")
 
@@ -244,7 +245,7 @@ def parse_env_variables(
         result.append(f"AUTHOR='{author}'")
 
     if official:
-        cmake_flags.append("-DYANDEX_OFFICIAL_BUILD=1")
+        cmake_flags.append("-DCLICKHOUSE_OFFICIAL_BUILD=1")
 
     result.append('CMAKE_FLAGS="' + " ".join(cmake_flags) + '"')
 

docker/test/fasttest/run.sh

@@ -267,6 +267,7 @@ function run_tests
     local test_opts=(
         --hung-check
         --fast-tests-only
+        --no-random-settings
         --no-long
         --testname
         --shard

docker/test/fuzzer/run-fuzzer.sh

@@ -13,7 +13,7 @@ script_dir="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
 echo "$script_dir"
 repo_dir=ch
 BINARY_TO_DOWNLOAD=${BINARY_TO_DOWNLOAD:="clang-13_debug_none_bundled_unsplitted_disable_False_binary"}
-BINARY_URL_TO_DOWNLOAD=${BINARY_URL_TO_DOWNLOAD:="https://clickhouse-builds.s3.yandex.net/$PR_TO_TEST/$SHA_TO_TEST/clickhouse_build_check/$BINARY_TO_DOWNLOAD/clickhouse"}
+BINARY_URL_TO_DOWNLOAD=${BINARY_URL_TO_DOWNLOAD:="https://clickhouse-builds.s3.amazonaws.com/$PR_TO_TEST/$SHA_TO_TEST/clickhouse_build_check/$BINARY_TO_DOWNLOAD/clickhouse"}
 
 function clone
 {

docker/test/keeper-jepsen/run.sh

@@ -2,7 +2,7 @@
 set -euo pipefail
 
 
-CLICKHOUSE_PACKAGE=${CLICKHOUSE_PACKAGE:="https://clickhouse-builds.s3.yandex.net/$PR_TO_TEST/$SHA_TO_TEST/clickhouse_build_check/clang-13_relwithdebuginfo_none_bundled_unsplitted_disable_False_binary/clickhouse"}
+CLICKHOUSE_PACKAGE=${CLICKHOUSE_PACKAGE:="https://clickhouse-builds.s3.amazonaws.com/$PR_TO_TEST/$SHA_TO_TEST/clickhouse_build_check/clang-13_relwithdebuginfo_none_bundled_unsplitted_disable_False_binary/clickhouse"}
 CLICKHOUSE_REPO_PATH=${CLICKHOUSE_REPO_PATH:=""}
 
 
@@ -10,7 +10,7 @@ if [ -z "$CLICKHOUSE_REPO_PATH" ]; then
     CLICKHOUSE_REPO_PATH=ch
     rm -rf ch ||:
     mkdir ch ||:
-    wget -nv -nd -c "https://clickhouse-test-reports.s3.yandex.net/$PR_TO_TEST/$SHA_TO_TEST/repo/clickhouse_no_subs.tar.gz"
+    wget -nv -nd -c "https://clickhouse-test-reports.s3.amazonaws.com/$PR_TO_TEST/$SHA_TO_TEST/repo/clickhouse_no_subs.tar.gz"
     tar -C ch --strip-components=1 -xf clickhouse_no_subs.tar.gz
     ls -lath ||:
 fi

docker/test/performance-comparison/compare.sh

@@ -1294,15 +1294,15 @@ create table ci_checks engine File(TSVWithNamesAndTypes, 'ci-checks.tsv')
         select '' test_name,
             '$(sed -n 's/.*<!--message: \(.*\)-->/\1/p' report.html)' test_status,
             0 test_duration_ms,
-            'https://clickhouse-test-reports.s3.yandex.net/$PR_TO_TEST/$SHA_TO_TEST/performance_comparison/report.html#fail1' report_url
+            'https://clickhouse-test-reports.s3.amazonaws.com/$PR_TO_TEST/$SHA_TO_TEST/performance_comparison/report.html#fail1' report_url
         union all
             select test || ' #' || toString(query_index), 'slower' test_status, 0 test_duration_ms,
-                'https://clickhouse-test-reports.s3.yandex.net/$PR_TO_TEST/$SHA_TO_TEST/performance_comparison/report.html#changes-in-performance.'
+                'https://clickhouse-test-reports.s3.amazonaws.com/$PR_TO_TEST/$SHA_TO_TEST/performance_comparison/report.html#changes-in-performance.'
                     || test || '.' || toString(query_index) report_url
             from queries where changed_fail != 0 and diff > 0
         union all
             select test || ' #' || toString(query_index), 'unstable' test_status, 0 test_duration_ms,
-                'https://clickhouse-test-reports.s3.yandex.net/$PR_TO_TEST/$SHA_TO_TEST/performance_comparison/report.html#unstable-queries.'
+                'https://clickhouse-test-reports.s3.amazonaws.com/$PR_TO_TEST/$SHA_TO_TEST/performance_comparison/report.html#unstable-queries.'
                     || test || '.' || toString(query_index) report_url
             from queries where unstable_fail != 0
     )

docker/test/performance-comparison/download.sh

@@ -16,26 +16,17 @@ right_sha=$4
 datasets=${CHPC_DATASETS-"hits1 hits10 hits100 values"}
 
 declare -A dataset_paths
-if [[ $S3_URL == *"s3.amazonaws.com"* ]]; then
-    dataset_paths["hits10"]="https://clickhouse-private-datasets.s3.amazonaws.com/hits_10m_single/partitions/hits_10m_single.tar"
-    dataset_paths["hits100"]="https://clickhouse-private-datasets.s3.amazonaws.com/hits_100m_single/partitions/hits_100m_single.tar"
-    dataset_paths["hits1"]="https://clickhouse-datasets.s3.amazonaws.com/hits/partitions/hits_v1.tar"
-    dataset_paths["values"]="https://clickhouse-datasets.s3.amazonaws.com/values_with_expressions/partitions/test_values.tar"
-else
-    dataset_paths["hits10"]="https://s3.mds.yandex.net/clickhouse-private-datasets/hits_10m_single/partitions/hits_10m_single.tar"
-    dataset_paths["hits100"]="https://s3.mds.yandex.net/clickhouse-private-datasets/hits_100m_single/partitions/hits_100m_single.tar"
-    dataset_paths["hits1"]="https://clickhouse-datasets.s3.yandex.net/hits/partitions/hits_v1.tar"
-    dataset_paths["values"]="https://clickhouse-datasets.s3.yandex.net/values_with_expressions/partitions/test_values.tar"
-fi
+dataset_paths["hits10"]="https://clickhouse-private-datasets.s3.amazonaws.com/hits_10m_single/partitions/hits_10m_single.tar"
+dataset_paths["hits100"]="https://clickhouse-private-datasets.s3.amazonaws.com/hits_100m_single/partitions/hits_100m_single.tar"
+dataset_paths["hits1"]="https://clickhouse-datasets.s3.amazonaws.com/hits/partitions/hits_v1.tar"
+dataset_paths["values"]="https://clickhouse-datasets.s3.amazonaws.com/values_with_expressions/partitions/test_values.tar"
 
 
 function download
 {
     # Historically there were various paths for the performance test package.
     # Test all of them.
-    declare -a urls_to_try=("https://s3.amazonaws.com/clickhouse-builds/$left_pr/$left_sha/performance/performance.tgz"
-                            "https://clickhouse-builds.s3.yandex.net/$left_pr/$left_sha/clickhouse_build_check/performance/performance.tgz"
-                           )
+    declare -a urls_to_try=("https://s3.amazonaws.com/clickhouse-builds/$left_pr/$left_sha/performance/performance.tgz")
 
     for path in "${urls_to_try[@]}"
     do

docker/test/performance-comparison/entrypoint.sh

@@ -4,7 +4,7 @@ set -ex
 CHPC_CHECK_START_TIMESTAMP="$(date +%s)"
 export CHPC_CHECK_START_TIMESTAMP
 
-S3_URL=${S3_URL:="https://clickhouse-builds.s3.yandex.net"}
+S3_URL=${S3_URL:="https://clickhouse-builds.s3.amazonaws.com"}
 
 COMMON_BUILD_PREFIX="/clickhouse_build_check"
 if [[ $S3_URL == *"s3.amazonaws.com"* ]]; then
@@ -64,9 +64,7 @@ function find_reference_sha
         # Historically there were various path for the performance test package,
         # test all of them.
         unset found
-        declare -a urls_to_try=("https://s3.amazonaws.com/clickhouse-builds/0/$REF_SHA/performance/performance.tgz"
-                                "https://clickhouse-builds.s3.yandex.net/0/$REF_SHA/clickhouse_build_check/performance/performance.tgz"
-                               )
+        declare -a urls_to_try=("https://s3.amazonaws.com/clickhouse-builds/0/$REF_SHA/performance/performance.tgz")
         for path in "${urls_to_try[@]}"
         do
             if curl_with_retry "$path"

docker/test/stateful/Dockerfile

@@ -11,7 +11,7 @@ RUN apt-get update -y \
 
 COPY s3downloader /s3downloader
 
-ENV S3_URL="https://clickhouse-datasets.s3.yandex.net"
+ENV S3_URL="https://clickhouse-datasets.s3.amazonaws.com"
 ENV DATASETS="hits visits"
 ENV EXPORT_S3_STORAGE_POLICIES=1
 

docker/test/stateful/s3downloader

@@ -10,7 +10,7 @@ import requests
 import tempfile
 
 
-DEFAULT_URL = 'https://clickhouse-datasets.s3.yandex.net'
+DEFAULT_URL = 'https://clickhouse-datasets.s3.amazonaws.com'
 
 AVAILABLE_DATASETS = {
     'hits': 'hits_v1.tar',

docker/test/stateless/setup_minio.sh

@@ -41,6 +41,7 @@ sleep 5
 ./mc admin user add clickminio test testtest
 ./mc admin policy set clickminio readwrite user=test
 ./mc mb clickminio/test
+./mc policy set public clickminio/test
 
 
 # Upload data to Minio. By default after unpacking all tests will in

docker/test/stress/Dockerfile

@@ -29,7 +29,7 @@ COPY ./download_previous_release /download_previous_release
 COPY run.sh /
 
 ENV DATASETS="hits visits"
-ENV S3_URL="https://clickhouse-datasets.s3.yandex.net"
+ENV S3_URL="https://clickhouse-datasets.s3.amazonaws.com"
 ENV EXPORT_S3_STORAGE_POLICIES=1
 
 CMD ["/bin/bash", "/run.sh"]

packages/clickhouse-keeper-dbg.yaml

@@ -0,0 +1,28 @@
+# package sources should be placed in ${PWD}/root
+# nfpm should run from the same directory with a config
+name: "clickhouse-keeper-dbg"
+arch: "${DEB_ARCH}" # amd64, arm64
+platform: "linux"
+version: "${CLICKHOUSE_VERSION_STRING}"
+vendor: "ClickHouse Inc."
+homepage: "https://clickhouse.com"
+license: "Apache"
+section: "database"
+priority: "optional"
+maintainer: "ClickHouse Dev Team <packages+linux@clickhouse.com>"
+description: |
+  debugging symbols for clickhouse-keeper
+    This package contains the debugging symbols for clickhouse-keeper.
+
+contents:
+- src: root/usr/lib/debug/usr/bin/clickhouse-keeper.debug
+  dst: /usr/lib/debug/usr/bin/clickhouse-keeper.debug
+# docs
+- src: ../AUTHORS
+  dst: /usr/share/doc/clickhouse-keeper-dbg/AUTHORS
+- src: ../CHANGELOG.md
+  dst: /usr/share/doc/clickhouse-keeper-dbg/CHANGELOG.md
+- src: ../LICENSE
+  dst: /usr/share/doc/clickhouse-keeper-dbg/LICENSE
+- src: ../README.md
+  dst: /usr/share/doc/clickhouse-keeper-dbg/README.md

packages/clickhouse-keeper.yaml

@@ -0,0 +1,40 @@
+# package sources should be placed in ${PWD}/root
+# nfpm should run from the same directory with a config
+name: "clickhouse-keeper"
+arch: "${DEB_ARCH}" # amd64, arm64
+platform: "linux"
+version: "${CLICKHOUSE_VERSION_STRING}"
+vendor: "ClickHouse Inc."
+homepage: "https://clickhouse.com"
+license: "Apache"
+section: "database"
+priority: "optional"
+
+conflicts:
+- clickhouse-server
+depends:
+- adduser
+suggests:
+- clickhouse-keeper-dbg
+
+maintainer: "ClickHouse Dev Team <packages+linux@clickhouse.com>"
+description: |
+  Static clickhouse-keeper binary
+    A stand-alone clickhouse-keeper package
+
+
+contents:
+- src: root/etc/clickhouse-keeper
+  dst: /etc/clickhouse-keeper
+  type: config
+- src: root/usr/bin/clickhouse-keeper
+  dst: /usr/bin/clickhouse-keeper
+# docs
+- src: ../AUTHORS
+  dst: /usr/share/doc/clickhouse-keeper/AUTHORS
+- src: ../CHANGELOG.md
+  dst: /usr/share/doc/clickhouse-keeper/CHANGELOG.md
+- src: ../LICENSE
+  dst: /usr/share/doc/clickhouse-keeper/LICENSE
+- src: ../README.md
+  dst: /usr/share/doc/clickhouse-keeper/README.md

programs/keeper/CMakeLists.txt

@@ -71,17 +71,11 @@ if (BUILD_STANDALONE_KEEPER)
         ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressedReadBuffer.cpp
         ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressedReadBufferFromFile.cpp
         ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressedWriteBuffer.cpp
-        ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionCodecDelta.cpp
-        ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionCodecDoubleDelta.cpp
-        ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionCodecEncrypted.cpp
-        ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionCodecGorilla.cpp
         ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionCodecLZ4.cpp
         ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionCodecMultiple.cpp
         ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionCodecNone.cpp
-        ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionCodecT64.cpp
         ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionCodecZSTD.cpp
         ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/CompressionFactory.cpp
-        ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/getCompressionCodecForFile.cpp
         ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/ICompressionCodec.cpp
         ${CMAKE_CURRENT_SOURCE_DIR}/../../src/Compression/LZ4_decompress_faster.cpp
 

src/Access/Common/QuotaDefs.h

@@ -13,7 +13,7 @@ enum class QuotaType
 {
     QUERIES,        /// Number of queries.
     QUERY_SELECTS,  /// Number of select queries.
-    QUERY_INSERTS,  /// Number of inserts queries.
+    QUERY_INSERTS,  /// Number of insert queries.
     ERRORS,         /// Number of queries with exceptions.
     RESULT_ROWS,    /// Number of rows returned as result.
     RESULT_BYTES,   /// Number of bytes returned as result.

src/AggregateFunctions/AggregateFunctionSumMap.cpp

@@ -67,7 +67,7 @@ auto parseArguments(const std::string & name, const DataTypes & arguments)
         values_types.push_back(array_type->getNestedType());
     }
 
-    return std::tuple{std::move(keys_type), std::move(values_types), tuple_argument};
+    return std::tuple<DataTypePtr, DataTypes, bool>{std::move(keys_type), std::move(values_types), tuple_argument};
 }
 
 // This function instantiates a particular overload of the sumMap family of

src/CMakeLists.txt

@@ -573,10 +573,6 @@ if (ENABLE_TESTS)
         target_link_libraries(unit_tests_dbms PRIVATE ch_contrib::simdjson)
     endif()
 
-    if(TARGET ch_contrib::rapidjson)
-        target_include_directories(unit_tests_dbms PRIVATE ch_contrib::rapidjson)
-    endif()
-
     if (TARGET ch_contrib::yaml_cpp)
         target_link_libraries(unit_tests_dbms PRIVATE ch_contrib::yaml_cpp)
     endif()

src/Columns/ColumnObject.cpp

@@ -521,7 +521,7 @@ ColumnObject::ColumnObject(bool is_nullable_)
 {
 }
 
-ColumnObject::ColumnObject(SubcolumnsTree && subcolumns_, bool is_nullable_)
+ColumnObject::ColumnObject(Subcolumns && subcolumns_, bool is_nullable_)
     : is_nullable(is_nullable_)
     , subcolumns(std::move(subcolumns_))
     , num_rows(subcolumns.empty() ? 0 : (*subcolumns.begin())->data.size())
@@ -696,7 +696,7 @@ const ColumnObject::Subcolumn & ColumnObject::getSubcolumn(const PathInData & ke
 ColumnObject::Subcolumn & ColumnObject::getSubcolumn(const PathInData & key)
 {
     if (const auto * node = subcolumns.findLeaf(key))
-        return const_cast<SubcolumnsTree::Node *>(node)->data;
+        return const_cast<Subcolumns::Node *>(node)->data;
 
     throw Exception(ErrorCodes::ILLEGAL_COLUMN, "There is no subcolumn {} in ColumnObject", key.getPath());
 }
@@ -794,7 +794,7 @@ bool ColumnObject::isFinalized() const
 void ColumnObject::finalize()
 {
     size_t old_size = size();
-    SubcolumnsTree new_subcolumns;
+    Subcolumns new_subcolumns;
     for (auto && entry : subcolumns)
     {
         const auto & least_common_type = entry->data.getLeastCommonType();

src/Columns/ColumnObject.h

@@ -138,20 +138,20 @@ public:
         size_t num_of_defaults_in_prefix = 0;
     };
 
-    using SubcolumnsTree = SubcolumnsTree<Subcolumn>;
+    using Subcolumns = SubcolumnsTree<Subcolumn>;
 
 private:
     /// If true then all subcolumns are nullable.
     const bool is_nullable;
 
-    SubcolumnsTree subcolumns;
+    Subcolumns subcolumns;
     size_t num_rows;
 
 public:
     static constexpr auto COLUMN_NAME_DUMMY = "_dummy";
 
     explicit ColumnObject(bool is_nullable_);
-    ColumnObject(SubcolumnsTree && subcolumns_, bool is_nullable_);
+    ColumnObject(Subcolumns && subcolumns_, bool is_nullable_);
 
     /// Checks that all subcolumns have consistent sizes.
     void checkConsistency() const;
@@ -173,8 +173,8 @@ public:
     /// It cares about consistency of sizes of Nested arrays.
     void addNestedSubcolumn(const PathInData & key, const FieldInfo & field_info, size_t new_size);
 
-    const SubcolumnsTree & getSubcolumns() const { return subcolumns; }
-    SubcolumnsTree & getSubcolumns() { return subcolumns; }
+    const Subcolumns & getSubcolumns() const { return subcolumns; }
+    Subcolumns & getSubcolumns() { return subcolumns; }
     PathsInData getKeys() const;
 
     /// Finalizes all subcolumns.

src/Common/FileSegment.cpp

@@ -437,6 +437,7 @@ String FileSegment::stateToString(FileSegment::State state)
         case FileSegment::State::SKIP_CACHE:
             return "SKIP_CACHE";
     }
+    __builtin_unreachable();
 }
 
 String FileSegmentsHolder::toString()

src/Common/ProfileEvents.cpp

@@ -9,6 +9,7 @@
     M(SelectQuery, "Same as Query, but only for SELECT queries.") \
     M(InsertQuery, "Same as Query, but only for INSERT queries.") \
     M(AsyncInsertQuery, "Same as InsertQuery, but only for asynchronous INSERT queries.") \
+    M(AsyncInsertBytes, "Data size in bytes of asynchronous INSERT queries.") \
     M(FailedQuery, "Number of failed queries.") \
     M(FailedSelectQuery, "Same as FailedQuery, but only for SELECT queries.") \
     M(FailedInsertQuery, "Same as FailedQuery, but only for INSERT queries.") \

src/Compression/CompressionFactory.cpp

@@ -165,25 +165,36 @@ void registerCodecNone(CompressionCodecFactory & factory);
 void registerCodecLZ4(CompressionCodecFactory & factory);
 void registerCodecLZ4HC(CompressionCodecFactory & factory);
 void registerCodecZSTD(CompressionCodecFactory & factory);
+void registerCodecMultiple(CompressionCodecFactory & factory);
+
+
+/// Keeper use only general-purpose codecs, so we don't need these special codecs
+/// in standalone build
+#ifndef KEEPER_STANDALONE_BUILD
+
 void registerCodecDelta(CompressionCodecFactory & factory);
 void registerCodecT64(CompressionCodecFactory & factory);
 void registerCodecDoubleDelta(CompressionCodecFactory & factory);
 void registerCodecGorilla(CompressionCodecFactory & factory);
 void registerCodecEncrypted(CompressionCodecFactory & factory);
-void registerCodecMultiple(CompressionCodecFactory & factory);
+
+#endif
 
 CompressionCodecFactory::CompressionCodecFactory()
 {
-    registerCodecLZ4(*this);
     registerCodecNone(*this);
+    registerCodecLZ4(*this);
     registerCodecZSTD(*this);
     registerCodecLZ4HC(*this);
+    registerCodecMultiple(*this);
+
+#ifndef KEEPER_STANDALONE_BUILD
     registerCodecDelta(*this);
     registerCodecT64(*this);
     registerCodecDoubleDelta(*this);
     registerCodecGorilla(*this);
     registerCodecEncrypted(*this);
-    registerCodecMultiple(*this);
+#endif
 
     default_codec = get("LZ4", {});
 }

src/DataTypes/Serializations/SerializationObject.cpp

@@ -63,12 +63,12 @@ private:
     size_t num_dimensions_to_keep;
 };
 
-using Node = typename ColumnObject::SubcolumnsTree::Node;
+using Node = typename ColumnObject::Subcolumns::Node;
 
 /// Finds a subcolumn from the same Nested type as @entry and inserts
 /// an array with default values with consistent sizes as in Nested type.
 bool tryInsertDefaultFromNested(
-    const std::shared_ptr<Node> & entry, const ColumnObject::SubcolumnsTree & subcolumns)
+    const std::shared_ptr<Node> & entry, const ColumnObject::Subcolumns & subcolumns)
 {
     if (!entry->path.hasNested())
         return false;
@@ -198,7 +198,7 @@ void SerializationObject<Parser>::deserializeWholeText(IColumn & column, ReadBuf
 template <typename Parser>
 void SerializationObject<Parser>::deserializeTextEscaped(IColumn & column, ReadBuffer & istr, const FormatSettings &) const
 {
-    deserializeTextImpl(column, [&](String & s) { readEscapedStringInto(s, istr); });
+    deserializeTextImpl(column, [&](String & s) { readEscapedString(s, istr); });
 }
 
 template <typename Parser>

src/Disks/IO/CachedReadBufferFromRemoteFS.h

@@ -96,6 +96,7 @@ private:
             case ReadType::REMOTE_FS_READ_AND_PUT_IN_CACHE:
                 return "REMOTE_FS_READ_AND_PUT_IN_CACHE";
         }
+        __builtin_unreachable();
     }
     size_t first_offset = 0;
 };

src/Formats/registerFormats.cpp

@@ -13,6 +13,7 @@ void registerFileSegmentationEngineCSV(FormatFactory & factory);
 void registerFileSegmentationEngineJSONEachRow(FormatFactory & factory);
 void registerFileSegmentationEngineRegexp(FormatFactory & factory);
 void registerFileSegmentationEngineJSONAsString(FormatFactory & factory);
+void registerFileSegmentationEngineJSONAsObject(FormatFactory & factory);
 void registerFileSegmentationEngineJSONCompactEachRow(FormatFactory & factory);
 
 /// Formats for both input/output.
@@ -103,6 +104,7 @@ void registerProtobufSchemaReader(FormatFactory & factory);
 void registerProtobufListSchemaReader(FormatFactory & factory);
 void registerLineAsStringSchemaReader(FormatFactory & factory);
 void registerJSONAsStringSchemaReader(FormatFactory & factory);
+void registerJSONAsObjectSchemaReader(FormatFactory & factory);
 void registerRawBLOBSchemaReader(FormatFactory & factory);
 void registerMsgPackSchemaReader(FormatFactory & factory);
 void registerCapnProtoSchemaReader(FormatFactory & factory);
@@ -123,6 +125,7 @@ void registerFormats()
     registerFileSegmentationEngineJSONEachRow(factory);
     registerFileSegmentationEngineRegexp(factory);
     registerFileSegmentationEngineJSONAsString(factory);
+    registerFileSegmentationEngineJSONAsObject(factory);
     registerFileSegmentationEngineJSONCompactEachRow(factory);
 
     registerInputFormatNative(factory);
@@ -207,6 +210,7 @@ void registerFormats()
     registerProtobufListSchemaReader(factory);
     registerLineAsStringSchemaReader(factory);
     registerJSONAsStringSchemaReader(factory);
+    registerJSONAsObjectSchemaReader(factory);
     registerRawBLOBSchemaReader(factory);
     registerMsgPackSchemaReader(factory);
     registerCapnProtoSchemaReader(factory);

src/Interpreters/AsynchronousInsertQueue.cpp

@@ -32,6 +32,7 @@ namespace CurrentMetrics
 namespace ProfileEvents
 {
     extern const Event AsyncInsertQuery;
+    extern const Event AsyncInsertBytes;
 }
 
 namespace DB
@@ -222,7 +223,9 @@ void AsynchronousInsertQueue::pushImpl(InsertData::EntryPtr entry, QueueIterator
     if (!data)
         data = std::make_unique<InsertData>();
 
-    data->size += entry->bytes.size();
+    size_t entry_data_size = entry->bytes.size();
+
+    data->size += entry_data_size;
     data->last_update = std::chrono::steady_clock::now();
     data->entries.emplace_back(entry);
 
@@ -239,6 +242,7 @@ void AsynchronousInsertQueue::pushImpl(InsertData::EntryPtr entry, QueueIterator
 
     CurrentMetrics::add(CurrentMetrics::PendingAsyncInsert);
     ProfileEvents::increment(ProfileEvents::AsyncInsertQuery);
+    ProfileEvents::increment(ProfileEvents::AsyncInsertBytes, entry_data_size);
 }
 
 void AsynchronousInsertQueue::waitForProcessingQuery(const String & query_id, const Milliseconds & timeout)

src/Interpreters/ExpressionAnalyzer.cpp

@@ -100,20 +100,9 @@ bool checkPositionalArguments(ASTPtr & argument, const ASTSelectQuery * select_q
 {
     auto columns = select_query->select()->children;
 
-    const auto * group_by_expr_with_alias = dynamic_cast<const ASTWithAlias *>(argument.get());
-    if (group_by_expr_with_alias && !group_by_expr_with_alias->alias.empty())
-    {
-        for (const auto & column : columns)
-        {
-            const auto * col_with_alias = dynamic_cast<const ASTWithAlias *>(column.get());
-            if (col_with_alias)
-            {
-                const auto & alias = col_with_alias->alias;
-                if (!alias.empty() && alias == group_by_expr_with_alias->alias)
-                    return false;
-            }
-        }
-    }
+    const auto * expr_with_alias = dynamic_cast<const ASTWithAlias *>(argument.get());
+    if (expr_with_alias && !expr_with_alias->alias.empty())
+        return false;
 
     const auto * ast_literal = typeid_cast<const ASTLiteral *>(argument.get());
     if (!ast_literal)
@@ -130,7 +119,7 @@ bool checkPositionalArguments(ASTPtr & argument, const ASTSelectQuery * select_q
                         pos, columns.size());
 
     const auto & column = columns[--pos];
-    if (typeid_cast<const ASTIdentifier *>(column.get()))
+    if (typeid_cast<const ASTIdentifier *>(column.get()) || typeid_cast<const ASTLiteral *>(column.get()))
     {
         argument = column->clone();
     }
@@ -1324,7 +1313,9 @@ ActionsDAGPtr SelectQueryExpressionAnalyzer::appendOrderBy(ExpressionActionsChai
             throw Exception("Bad ORDER BY expression AST", ErrorCodes::UNKNOWN_TYPE_OF_AST_NODE);
 
         if (getContext()->getSettingsRef().enable_positional_arguments)
+        {
             replaceForPositionalArguments(ast->children.at(0), select_query, ASTSelectQuery::Expression::ORDER_BY);
+        }
     }
 
     getRootActions(select_query->orderBy(), only_types, step.actions());

src/Interpreters/MySQL/InterpretersMySQLDDLQuery.cpp

@@ -150,12 +150,12 @@ static ColumnsDescription createColumnsDescription(const NamesAndTypesList & col
 
     ColumnsDescription columns_description;
 
-    for (
-        auto [column_name_and_type, declare_column_ast] = std::tuple{columns_name_and_type.begin(), columns_definition->children.begin()};
-        column_name_and_type != columns_name_and_type.end();
-        column_name_and_type++,
-        declare_column_ast++
-    )
+    /// FIXME: we could write it like auto [a, b] = std::tuple(x, y),
+    /// but this produce endless recursion in gcc-11, and leads to SIGSEGV
+    /// (see git blame for details).
+    auto column_name_and_type = columns_name_and_type.begin();
+    auto declare_column_ast = columns_definition->children.begin();
+    for (; column_name_and_type != columns_name_and_type.end(); column_name_and_type++, declare_column_ast++)
     {
         const auto & declare_column = (*declare_column_ast)->as<MySQLParser::ASTDeclareColumn>();
         String comment;

src/Interpreters/RowRefs.cpp

@@ -192,7 +192,7 @@ private:
                             using Result = Element;
 
                             static TKey & extractKey(Element & elem) { return elem.value; }
-                            static Element extractResult(Element & elem) { return elem; }
+                            static Result extractResult(Element & elem) { return elem; }
                         };
 
                         if constexpr (is_descending)

src/Interpreters/TreeRewriter.cpp

@@ -345,7 +345,10 @@ void replaceWithSumCount(String column_name, ASTFunction & func)
     {
         /// Rewrite "avg" to sumCount().1 / sumCount().2
         auto new_arg1 = makeASTFunction("tupleElement", func_base, std::make_shared<ASTLiteral>(UInt8(1)));
-        auto new_arg2 = makeASTFunction("tupleElement", func_base, std::make_shared<ASTLiteral>(UInt8(2)));
+        auto new_arg2 = makeASTFunction("CAST",
+            makeASTFunction("tupleElement", func_base, std::make_shared<ASTLiteral>(UInt8(2))),
+            std::make_shared<ASTLiteral>("Float64"));
+
         func.name = "divide";
         exp_list->children.push_back(new_arg1);
         exp_list->children.push_back(new_arg2);

src/Interpreters/threadPoolCallbackRunner.cpp

@@ -25,7 +25,7 @@ CallbackRunner threadPoolCallbackRunner(ThreadPool & pool)
                 /// Usually it could be ok, because thread pool task is executed before user-level memory tracker is destroyed.
                 /// However, thread could stay alive inside the thread pool, and it's ThreadStatus as well.
                 /// When, finally, we destroy the thread (and the ThreadStatus),
-                /// it can use memory tracker in the ~ThreadStatus in order to alloc/free untracked_memory,\
+                /// it can use memory tracker in the ~ThreadStatus in order to alloc/free untracked_memory,
                 /// and by this time user-level memory tracker may be already destroyed.
                 ///
                 /// As a work-around, reset memory tracker to total, which is always alive.

src/Processors/Formats/Impl/JSONAsStringRowInputFormat.cpp

@@ -228,6 +228,14 @@ void registerNonTrivialPrefixAndSuffixCheckerJSONAsString(FormatFactory & factor
     factory.registerNonTrivialPrefixAndSuffixChecker("JSONAsString", nonTrivialPrefixAndSuffixCheckerJSONEachRowImpl);
 }
 
+void registerJSONAsStringSchemaReader(FormatFactory & factory)
+{
+    factory.registerExternalSchemaReader("JSONAsString", [](const FormatSettings &)
+    {
+        return std::make_shared<JSONAsStringExternalSchemaReader>();
+    });
+}
+
 void registerInputFormatJSONAsObject(FormatFactory & factory)
 {
     factory.registerInputFormat("JSONAsObject", [](
@@ -245,11 +253,16 @@ void registerNonTrivialPrefixAndSuffixCheckerJSONAsObject(FormatFactory & factor
     factory.registerNonTrivialPrefixAndSuffixChecker("JSONAsObject", nonTrivialPrefixAndSuffixCheckerJSONEachRowImpl);
 }
 
-void registerJSONAsStringSchemaReader(FormatFactory & factory)
+void registerFileSegmentationEngineJSONAsObject(FormatFactory & factory)
 {
-    factory.registerExternalSchemaReader("JSONAsString", [](const FormatSettings &)
+    factory.registerFileSegmentationEngine("JSONAsObject", &fileSegmentationEngineJSONEachRow);
+}
+
+void registerJSONAsObjectSchemaReader(FormatFactory & factory)
+{
+    factory.registerExternalSchemaReader("JSONAsObject", [](const FormatSettings &)
     {
-        return std::make_shared<JSONAsStringExternalSchemaReader>();
+        return std::make_shared<JSONAsObjectExternalSchemaReader>();
     });
 }
 

src/Processors/Formats/Impl/JSONAsStringRowInputFormat.h

@@ -5,6 +5,7 @@
 #include <Formats/FormatFactory.h>
 #include <IO/PeekableReadBuffer.h>
 #include <DataTypes/DataTypeString.h>
+#include <DataTypes/DataTypeObject.h>
 
 namespace DB
 {
@@ -73,4 +74,13 @@ public:
     }
 };
 
+class JSONAsObjectExternalSchemaReader : public IExternalSchemaReader
+{
+public:
+    NamesAndTypesList readSchema() override
+    {
+        return {{"json", std::make_shared<DataTypeObject>("json", false)}};
+    }
+};
+
 }

src/Processors/Transforms/WindowTransform.cpp

@@ -488,7 +488,7 @@ auto WindowTransform::moveRowNumberNoCheck(const RowNumber & _x, int64_t offset)
         }
     }
 
-    return std::tuple{x, offset};
+    return std::tuple<RowNumber, int64_t>{x, offset};
 }
 
 auto WindowTransform::moveRowNumber(const RowNumber & _x, int64_t offset) const
@@ -505,7 +505,7 @@ auto WindowTransform::moveRowNumber(const RowNumber & _x, int64_t offset) const
     assert(oo == 0);
 #endif
 
-    return std::tuple{x, o};
+    return std::tuple<RowNumber, int64_t>{x, o};
 }
 
 

src/Storages/ExternalDataSourceConfiguration.cpp

@@ -325,6 +325,7 @@ void URLBasedDataSourceConfiguration::set(const URLBasedDataSourceConfiguration
     compression_method = conf.compression_method;
     structure = conf.structure;
     http_method = conf.http_method;
+    headers = conf.headers;
 }
 
 
@@ -364,6 +365,10 @@ std::optional<URLBasedDataSourceConfig> getURLBasedDataSourceConfiguration(const
             {
                 configuration.structure = config.getString(config_prefix + ".structure", "");
             }
+            else if (key == "compression_method")
+            {
+                configuration.compression_method = config.getString(config_prefix + ".compression_method", "");
+            }
             else if (key == "headers")
             {
                 Poco::Util::AbstractConfiguration::Keys header_keys;

src/Storages/ExternalDataSourceConfiguration.h

@@ -114,6 +114,12 @@ struct StorageS3Configuration : URLBasedDataSourceConfiguration
     String secret_access_key;
 };
 
+
+struct StorageS3ClusterConfiguration : StorageS3Configuration
+{
+    String cluster_name;
+};
+
 struct URLBasedDataSourceConfig
 {
     URLBasedDataSourceConfiguration configuration;

src/Storages/MergeTree/MergeFromLogEntryTask.cpp

@@ -279,14 +279,17 @@ bool MergeFromLogEntryTask::finalize(ReplicatedMergeMutateTaskBase::PartLogWrite
             ProfileEvents::increment(ProfileEvents::DataAfterMergeDiffersFromReplica);
 
             LOG_ERROR(log,
-                "{}. Data after merge is not byte-identical to data on another replicas. There could be several"
-                " reasons: 1. Using newer version of compression library after server update. 2. Using another"
-                " compression method. 3. Non-deterministic compression algorithm (highly unlikely). 4."
-                " Non-deterministic merge algorithm due to logical error in code. 5. Data corruption in memory due"
-                " to bug in code. 6. Data corruption in memory due to hardware issue. 7. Manual modification of"
-                " source data after server startup. 8. Manual modification of checksums stored in ZooKeeper. 9."
-                " Part format related settings like 'enable_mixed_granularity_parts' are different on different"
-                " replicas. We will download merged part from replica to force byte-identical result.",
+                "{}. Data after merge is not byte-identical to data on another replicas. There could be several reasons:"
+                " 1. Using newer version of compression library after server update."
+                " 2. Using another compression method."
+                " 3. Non-deterministic compression algorithm (highly unlikely)."
+                " 4. Non-deterministic merge algorithm due to logical error in code."
+                " 5. Data corruption in memory due to bug in code."
+                " 6. Data corruption in memory due to hardware issue."
+                " 7. Manual modification of source data after server startup."
+                " 8. Manual modification of checksums stored in ZooKeeper."
+                " 9. Part format related settings like 'enable_mixed_granularity_parts' are different on different replicas."
+                " We will download merged part from replica to force byte-identical result.",
                 getCurrentExceptionMessage(false));
 
             write_part_log(ExecutionStatus::fromCurrentException());

src/Storages/MergeTree/MutateFromLogEntryTask.cpp

@@ -185,7 +185,8 @@ bool MutateFromLogEntryTask::finalize(ReplicatedMergeMutateTaskBase::PartLogWrit
 
             ProfileEvents::increment(ProfileEvents::DataAfterMutationDiffersFromReplica);
 
-            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));
+            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));
 
             write_part_log(ExecutionStatus::fromCurrentException());
 

src/Storages/PostgreSQL/MaterializedPostgreSQLConsumer.cpp

@@ -98,8 +98,24 @@ MaterializedPostgreSQLConsumer::StorageData::Buffer::Buffer(
 }
 
 
+void MaterializedPostgreSQLConsumer::assertCorrectInsertion(StorageData::Buffer & buffer, size_t column_idx)
+{
+    if (column_idx >= buffer.description.sample_block.columns()
+        || column_idx >= buffer.description.types.size()
+        || column_idx >= buffer.columns.size())
+        throw Exception(
+            ErrorCodes::LOGICAL_ERROR,
+            "Attempt to insert into buffer at position: {}, but block columns size is {}, types size: {}, columns size: {}, buffer structure: {}",
+            column_idx,
+            buffer.description.sample_block.columns(), buffer.description.types.size(), buffer.columns.size(),
+            buffer.description.sample_block.dumpStructure());
+}
+
+
 void MaterializedPostgreSQLConsumer::insertValue(StorageData::Buffer & buffer, const std::string & value, size_t column_idx)
 {
+    assertCorrectInsertion(buffer, column_idx);
+
     const auto & sample = buffer.description.sample_block.getByPosition(column_idx);
     bool is_nullable = buffer.description.types[column_idx].second;
 
@@ -134,6 +150,8 @@ void MaterializedPostgreSQLConsumer::insertValue(StorageData::Buffer & buffer, c
 
 void MaterializedPostgreSQLConsumer::insertDefaultValue(StorageData::Buffer & buffer, size_t column_idx)
 {
+    assertCorrectInsertion(buffer, column_idx);
+
     const auto & sample = buffer.description.sample_block.getByPosition(column_idx);
     insertDefaultPostgreSQLValue(*buffer.columns[column_idx], *sample.column);
 }
@@ -515,13 +533,14 @@ void MaterializedPostgreSQLConsumer::processReplicationMessage(const char * repl
 
 void MaterializedPostgreSQLConsumer::syncTables()
 {
-    try
+    for (const auto & table_name : tables_to_sync)
     {
-        for (const auto & table_name : tables_to_sync)
-        {
-            auto & storage_data = storages.find(table_name)->second;
-            Block result_rows = storage_data.buffer.description.sample_block.cloneWithColumns(std::move(storage_data.buffer.columns));
+        auto & storage_data = storages.find(table_name)->second;
+        Block result_rows = storage_data.buffer.description.sample_block.cloneWithColumns(std::move(storage_data.buffer.columns));
+        storage_data.buffer.columns = storage_data.buffer.description.sample_block.cloneEmptyColumns();
 
+        try
+        {
             if (result_rows.rows())
             {
                 auto storage = storage_data.storage;
@@ -543,13 +562,18 @@ void MaterializedPostgreSQLConsumer::syncTables()
 
                 CompletedPipelineExecutor executor(io.pipeline);
                 executor.execute();
-
-                storage_data.buffer.columns = storage_data.buffer.description.sample_block.cloneEmptyColumns();
             }
         }
+        catch (...)
+        {
+            tryLogCurrentException(__PRETTY_FUNCTION__);
+        }
+    }
 
-        LOG_DEBUG(log, "Table sync end for {} tables, last lsn: {} = {}, (attempted lsn {})", tables_to_sync.size(), current_lsn, getLSNValue(current_lsn), getLSNValue(final_lsn));
+    LOG_DEBUG(log, "Table sync end for {} tables, last lsn: {} = {}, (attempted lsn {})", tables_to_sync.size(), current_lsn, getLSNValue(current_lsn), getLSNValue(final_lsn));
 
+    try
+    {
         auto tx = std::make_shared<pqxx::nontransaction>(connection->getRef());
         current_lsn = advanceLSN(tx);
         tables_to_sync.clear();

src/Storages/PostgreSQL/MaterializedPostgreSQLConsumer.h

@@ -122,6 +122,8 @@ private:
 
     void markTableAsSkipped(Int32 relation_id, const String & relation_name);
 
+    static void assertCorrectInsertion(StorageData::Buffer & buffer, size_t column_idx);
+
     /// lsn - log sequnce nuumber, like wal offset (64 bit).
     static Int64 getLSNValue(const std::string & lsn)
     {

src/Storages/PostgreSQL/PostgreSQLReplicationHandler.cpp

@@ -64,8 +64,8 @@ PostgreSQLReplicationHandler::PostgreSQLReplicationHandler(
     bool is_attach_,
     const MaterializedPostgreSQLSettings & replication_settings,
     bool is_materialized_postgresql_database_)
-    : log(&Poco::Logger::get("PostgreSQLReplicationHandler"))
-    , context(context_)
+    : WithContext(context_->getGlobalContext())
+    , log(&Poco::Logger::get("PostgreSQLReplicationHandler"))
     , is_attach(is_attach_)
     , postgres_database(postgres_database_)
     , postgres_schema(replication_settings.materialized_postgresql_schema)
@@ -94,9 +94,9 @@ PostgreSQLReplicationHandler::PostgreSQLReplicationHandler(
     }
     publication_name = fmt::format("{}_ch_publication", replication_identifier);
 
-    startup_task = context->getSchedulePool().createTask("PostgreSQLReplicaStartup", [this]{ checkConnectionAndStart(); });
-    consumer_task = context->getSchedulePool().createTask("PostgreSQLReplicaStartup", [this]{ consumerFunc(); });
-    cleanup_task = context->getSchedulePool().createTask("PostgreSQLReplicaStartup", [this]{ cleanupFunc(); });
+    startup_task = getContext()->getSchedulePool().createTask("PostgreSQLReplicaStartup", [this]{ checkConnectionAndStart(); });
+    consumer_task = getContext()->getSchedulePool().createTask("PostgreSQLReplicaStartup", [this]{ consumerFunc(); });
+    cleanup_task = getContext()->getSchedulePool().createTask("PostgreSQLReplicaStartup", [this]{ cleanupFunc(); });
 }
 
 
@@ -296,7 +296,7 @@ void PostgreSQLReplicationHandler::startSynchronization(bool throw_on_error)
     /// (Apart from the case, when shutdownFinal is called).
     /// Handler uses it only for loadFromSnapshot and shutdown methods.
     consumer = std::make_shared<MaterializedPostgreSQLConsumer>(
-            context,
+            getContext(),
             std::move(tmp_connection),
             replication_slot,
             publication_name,
@@ -921,9 +921,9 @@ void PostgreSQLReplicationHandler::reloadFromSnapshot(const std::vector<std::pai
 
             for (const auto & [relation_id, table_name] : relation_data)
             {
-                auto storage = DatabaseCatalog::instance().getTable(StorageID(current_database_name, table_name), context);
+                auto storage = DatabaseCatalog::instance().getTable(StorageID(current_database_name, table_name), getContext());
                 auto * materialized_storage = storage->as <StorageMaterializedPostgreSQL>();
-                auto materialized_table_lock = materialized_storage->lockForShare(String(), context->getSettingsRef().lock_acquire_timeout);
+                auto materialized_table_lock = materialized_storage->lockForShare(String(), getContext()->getSettingsRef().lock_acquire_timeout);
 
                 /// If for some reason this temporary table already exists - also drop it.
                 auto temp_materialized_storage = materialized_storage->createTemporary();

src/Storages/PostgreSQL/PostgreSQLReplicationHandler.h

@@ -13,7 +13,7 @@ namespace DB
 class StorageMaterializedPostgreSQL;
 struct SettingChange;
 
-class PostgreSQLReplicationHandler
+class PostgreSQLReplicationHandler : WithContext
 {
 friend class TemporaryReplicationSlot;
 
@@ -98,7 +98,6 @@ private:
     std::pair<String, String> getSchemaAndTableName(const String & table_name) const;
 
     Poco::Logger * log;
-    ContextPtr context;
 
     /// If it is not attach, i.e. a create query, then if publication already exists - always drop it.
     bool is_attach;

src/Storages/StorageReplicatedMergeTree.cpp

@@ -1312,10 +1312,14 @@ void StorageReplicatedMergeTree::checkPartChecksumsAndAddCommitOps(const zkutil:
 
         if (replica_part_header.getColumnsHash() != local_part_header.getColumnsHash())
         {
-            /// Either it's a bug or ZooKeeper contains broken data.
-            /// TODO Fix KILL MUTATION and replace CHECKSUM_DOESNT_MATCH with LOGICAL_ERROR
-            /// (some replicas may skip killed mutation even if it was executed on other replicas)
-            throw Exception(ErrorCodes::CHECKSUM_DOESNT_MATCH, "Part {} from {} has different columns hash", part_name, replica);
+            /// Currently there are two (known) cases when it may happen:
+            ///  - KILL MUTATION query had removed mutation before all replicas have executed assigned MUTATE_PART entries.
+            ///    Some replicas may skip this mutation and update part version without actually applying any changes.
+            ///    It leads to mismatching checksum if changes were applied on other replicas.
+            ///  - ALTER_METADATA and MERGE_PARTS were reordered on some replicas.
+            ///    It may lead to different number of columns in merged parts on these replicas.
+            throw Exception(ErrorCodes::CHECKSUM_DOESNT_MATCH, "Part {} from {} has different columns hash "
+                            "(it may rarely happen on race condition with KILL MUTATION or ALTER COLUMN).", part_name, replica);
         }
 
         replica_part_header.getChecksums().checkEqual(local_part_header.getChecksums(), true);

src/Storages/StorageS3Cluster.cpp

@@ -51,6 +51,7 @@ StorageS3Cluster::StorageS3Cluster(
     const StorageID & table_id_,
     String cluster_name_,
     const String & format_name_,
+    UInt64 max_single_read_retries_,
     UInt64 max_connections_,
     const ColumnsDescription & columns_,
     const ConstraintsDescription & constraints_,
@@ -63,11 +64,26 @@ StorageS3Cluster::StorageS3Cluster(
     , format_name(format_name_)
     , compression_method(compression_method_)
 {
+    context_->getGlobalContext()->getRemoteHostFilter().checkURL(Poco::URI{filename});
     StorageInMemoryMetadata storage_metadata;
-    storage_metadata.setColumns(columns_);
+    StorageS3::updateClientAndAuthSettings(context_, client_auth);
+
+    if (columns_.empty())
+    {
+        const bool is_key_with_globs = filename.find_first_of("*?{") != std::string::npos;
+
+        /// `distributed_processing` is set to false, because this code is executed on the initiator, so there is no callback set
+        /// for asking for the next tasks.
+        /// `format_settings` is set to std::nullopt, because StorageS3Cluster is used only as table function
+        auto columns = StorageS3::getTableStructureFromDataImpl(format_name, client_auth, max_single_read_retries_, compression_method,
+            /*distributed_processing_*/false, is_key_with_globs, /*format_settings=*/std::nullopt, context_);
+        storage_metadata.setColumns(columns);
+    }
+    else
+        storage_metadata.setColumns(columns_);
+
     storage_metadata.setConstraints(constraints_);
     setInMemoryMetadata(storage_metadata);
-    StorageS3::updateClientAndAuthSettings(context_, client_auth);
 }
 
 /// The code executes on initiator
@@ -83,7 +99,6 @@ Pipe StorageS3Cluster::read(
     StorageS3::updateClientAndAuthSettings(context, client_auth);
 
     auto cluster = context->getCluster(cluster_name)->getClusterWithReplicasAsShards(context->getSettingsRef());
-    StorageS3::updateClientAndAuthSettings(context, client_auth);
 
     auto iterator = std::make_shared<StorageS3Source::DisclosedGlobIterator>(*client_auth.client, client_auth.uri);
     auto callback = std::make_shared<StorageS3Source::IteratorWrapper>([iterator]() mutable -> String

src/Storages/StorageS3Cluster.h

@@ -41,6 +41,7 @@ protected:
         const StorageID & table_id_,
         String cluster_name_,
         const String & format_name_,
+        UInt64 max_single_read_retries_,
         UInt64 max_connections_,
         const ColumnsDescription & columns_,
         const ConstraintsDescription & constraints_,

src/TableFunctions/TableFunctionS3.cpp

@@ -22,51 +22,29 @@ namespace ErrorCodes
     extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH;
 }
 
-void TableFunctionS3::parseArguments(const ASTPtr & ast_function, ContextPtr context)
+
+/// This is needed to avoid copy-pase. Because s3Cluster arguments only differ in additional argument (first) - cluster name
+void TableFunctionS3::parseArgumentsImpl(const String & error_message, ASTs & args, ContextPtr context, StorageS3Configuration & s3_configuration)
 {
-    /// Parse args
-    ASTs & args_func = ast_function->children;
-
-    const auto message = fmt::format(
-        "The signature of table function {} could be the following:\n" \
-        " - url\n"
-        " - url, format\n" \
-        " - url, format, structure\n" \
-        " - url, access_key_id, secret_access_key\n" \
-        " - url, format, structure, compression_method\n" \
-        " - url, access_key_id, secret_access_key, format\n"
-        " - url, access_key_id, secret_access_key, format, structure\n" \
-        " - url, access_key_id, secret_access_key, format, structure, compression_method",
-        getName());
-
-    if (args_func.size() != 1)
-        throw Exception("Table function '" + getName() + "' must have arguments.", ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH);
-
-    ASTs & args = args_func.at(0)->children;
-    StorageS3Configuration configuration;
-
     if (auto named_collection = getURLBasedDataSourceConfiguration(args, context))
     {
         auto [common_configuration, storage_specific_args] = named_collection.value();
-        configuration.set(common_configuration);
+        s3_configuration.set(common_configuration);
 
         for (const auto & [arg_name, arg_value] : storage_specific_args)
         {
             if (arg_name == "access_key_id")
-                configuration.access_key_id = arg_value->as<ASTLiteral>()->value.safeGet<String>();
+                s3_configuration.access_key_id = arg_value->as<ASTLiteral>()->value.safeGet<String>();
             else if (arg_name == "secret_access_key")
-                configuration.secret_access_key = arg_value->as<ASTLiteral>()->value.safeGet<String>();
+                s3_configuration.secret_access_key = arg_value->as<ASTLiteral>()->value.safeGet<String>();
             else
-                throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH,
-                                "Unknown key-value argument `{}` for StorageS3, expected: "
-                                "url, [access_key_id, secret_access_key], name of used format, structure and [compression_method].",
-                                arg_name);
+                throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, error_message);
         }
     }
     else
     {
         if (args.empty() || args.size() > 6)
-            throw Exception(message, ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH);
+            throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, error_message);
 
         for (auto & arg : args)
             arg = evaluateConstantExpressionOrIdentifierAsLiteral(arg, context);
@@ -110,53 +88,76 @@ void TableFunctionS3::parseArguments(const ASTPtr & ast_function, ContextPtr con
         }
 
         /// This argument is always the first
-        configuration.url = args[0]->as<ASTLiteral &>().value.safeGet<String>();
+        s3_configuration.url = args[0]->as<ASTLiteral &>().value.safeGet<String>();
 
         if (args_to_idx.contains("format"))
-            configuration.format = args[args_to_idx["format"]]->as<ASTLiteral &>().value.safeGet<String>();
+            s3_configuration.format = args[args_to_idx["format"]]->as<ASTLiteral &>().value.safeGet<String>();
 
         if (args_to_idx.contains("structure"))
-            configuration.structure = args[args_to_idx["structure"]]->as<ASTLiteral &>().value.safeGet<String>();
+            s3_configuration.structure = args[args_to_idx["structure"]]->as<ASTLiteral &>().value.safeGet<String>();
 
         if (args_to_idx.contains("compression_method"))
-            configuration.compression_method = args[args_to_idx["compression_method"]]->as<ASTLiteral &>().value.safeGet<String>();
+            s3_configuration.compression_method = args[args_to_idx["compression_method"]]->as<ASTLiteral &>().value.safeGet<String>();
 
         if (args_to_idx.contains("access_key_id"))
-            configuration.access_key_id = args[args_to_idx["access_key_id"]]->as<ASTLiteral &>().value.safeGet<String>();
+            s3_configuration.access_key_id = args[args_to_idx["access_key_id"]]->as<ASTLiteral &>().value.safeGet<String>();
 
         if (args_to_idx.contains("secret_access_key"))
-            configuration.secret_access_key = args[args_to_idx["secret_access_key"]]->as<ASTLiteral &>().value.safeGet<String>();
+            s3_configuration.secret_access_key = args[args_to_idx["secret_access_key"]]->as<ASTLiteral &>().value.safeGet<String>();
     }
 
-    if (configuration.format == "auto")
-        configuration.format = FormatFactory::instance().getFormatFromFileName(configuration.url, true);
+    if (s3_configuration.format == "auto")
+        s3_configuration.format = FormatFactory::instance().getFormatFromFileName(s3_configuration.url, true);
+}
 
-    s3_configuration = std::move(configuration);
+void TableFunctionS3::parseArguments(const ASTPtr & ast_function, ContextPtr context)
+{
+    /// Parse args
+    ASTs & args_func = ast_function->children;
+
+    const auto message = fmt::format(
+        "The signature of table function {} could be the following:\n" \
+        " - url\n" \
+        " - url, format\n" \
+        " - url, format, structure\n" \
+        " - url, access_key_id, secret_access_key\n" \
+        " - url, format, structure, compression_method\n" \
+        " - url, access_key_id, secret_access_key, format\n" \
+        " - url, access_key_id, secret_access_key, format, structure\n" \
+        " - url, access_key_id, secret_access_key, format, structure, compression_method",
+        getName());
+
+    if (args_func.size() != 1)
+        throw Exception("Table function '" + getName() + "' must have arguments.", ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH);
+
+    auto & args = args_func.at(0)->children;
+
+    parseArgumentsImpl(message, args, context, configuration);
 }
 
 ColumnsDescription TableFunctionS3::getActualTableStructure(ContextPtr context) const
 {
-    if (s3_configuration->structure == "auto")
+    if (configuration.structure == "auto")
     {
         return StorageS3::getTableStructureFromData(
-            s3_configuration->format,
-            S3::URI(Poco::URI(s3_configuration->url)),
-            s3_configuration->access_key_id,
-            s3_configuration->secret_access_key,
+            configuration.format,
+            S3::URI(Poco::URI(configuration.url)),
+            configuration.access_key_id,
+            configuration.secret_access_key,
             context->getSettingsRef().s3_max_connections,
             context->getSettingsRef().s3_max_single_read_retries,
-            s3_configuration->compression_method,
+            configuration.compression_method,
             false,
             std::nullopt,
             context);
     }
 
-    return parseColumnsListFromString(s3_configuration->structure, context);
+    return parseColumnsListFromString(configuration.structure, context);
 }
 
 StoragePtr TableFunctionS3::executeImpl(const ASTPtr & /*ast_function*/, ContextPtr context, const std::string & table_name, ColumnsDescription /*cached_columns*/) const
 {
-    Poco::URI uri (s3_configuration->url);
+    Poco::URI uri (configuration.url);
     S3::URI s3_uri (uri);
     UInt64 max_single_read_retries = context->getSettingsRef().s3_max_single_read_retries;
     UInt64 min_upload_part_size = context->getSettingsRef().s3_min_upload_part_size;
@@ -166,17 +167,17 @@ StoragePtr TableFunctionS3::executeImpl(const ASTPtr & /*ast_function*/, Context
     UInt64 max_connections = context->getSettingsRef().s3_max_connections;
 
     ColumnsDescription columns;
-    if (s3_configuration->structure != "auto")
-        columns = parseColumnsListFromString(s3_configuration->structure, context);
+    if (configuration.structure != "auto")
+        columns = parseColumnsListFromString(configuration.structure, context);
     else if (!structure_hint.empty())
         columns = structure_hint;
 
     StoragePtr storage = StorageS3::create(
         s3_uri,
-        s3_configuration->access_key_id,
-        s3_configuration->secret_access_key,
+        configuration.access_key_id,
+        configuration.secret_access_key,
         StorageID(getDatabaseName(), table_name),
-        s3_configuration->format,
+        configuration.format,
         max_single_read_retries,
         min_upload_part_size,
         upload_part_size_multiply_factor,
@@ -189,7 +190,7 @@ StoragePtr TableFunctionS3::executeImpl(const ASTPtr & /*ast_function*/, Context
         context,
         /// No format_settings for table function S3
         std::nullopt,
-        s3_configuration->compression_method);
+        configuration.compression_method);
 
     storage->startup();
 

src/TableFunctions/TableFunctionS3.h

@@ -12,6 +12,7 @@ namespace DB
 {
 
 class Context;
+class TableFunctionS3Cluster;
 
 /* s3(source, [access_key_id, secret_access_key,] format, structure[, compression]) - creates a temporary storage for a file in S3.
  */
@@ -23,13 +24,15 @@ public:
     {
         return name;
     }
-    bool hasStaticStructure() const override { return s3_configuration->structure != "auto"; }
+    bool hasStaticStructure() const override { return configuration.structure != "auto"; }
 
-    bool needStructureHint() const override { return s3_configuration->structure == "auto"; }
+    bool needStructureHint() const override { return configuration.structure == "auto"; }
 
     void setStructureHint(const ColumnsDescription & structure_hint_) override { structure_hint = structure_hint_; }
 
 protected:
+    friend class TableFunctionS3Cluster;
+
     StoragePtr executeImpl(
         const ASTPtr & ast_function,
         ContextPtr context,
@@ -41,7 +44,9 @@ protected:
     ColumnsDescription getActualTableStructure(ContextPtr context) const override;
     void parseArguments(const ASTPtr & ast_function, ContextPtr context) override;
 
-    std::optional<StorageS3Configuration> s3_configuration;
+    static void parseArgumentsImpl(const String & error_message, ASTs & args, ContextPtr context, StorageS3Configuration & configuration);
+
+    StorageS3Configuration configuration;
     ColumnsDescription structure_hint;
 };
 

src/TableFunctions/TableFunctionS3Cluster.cpp

@@ -32,6 +32,7 @@ namespace DB
 namespace ErrorCodes
 {
     extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH;
+    extern const int BAD_GET;
 }
 
 
@@ -45,55 +46,58 @@ void TableFunctionS3Cluster::parseArguments(const ASTPtr & ast_function, Context
 
     ASTs & args = args_func.at(0)->children;
 
+    for (auto & arg : args)
+        arg = evaluateConstantExpressionAsLiteral(arg, context);
+
     const auto message = fmt::format(
         "The signature of table function {} could be the following:\n" \
+        " - cluster, url\n"
+        " - cluster, url, format\n" \
         " - cluster, url, format, structure\n" \
+        " - cluster, url, access_key_id, secret_access_key\n" \
         " - cluster, url, format, structure, compression_method\n" \
+        " - cluster, url, access_key_id, secret_access_key, format\n"
         " - cluster, url, access_key_id, secret_access_key, format, structure\n" \
         " - cluster, url, access_key_id, secret_access_key, format, structure, compression_method",
         getName());
 
-    if (args.size() < 4 || args.size() > 7)
+    if (args.size() < 2 || args.size() > 7)
         throw Exception(message, ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH);
 
-    for (auto & arg : args)
-        arg = evaluateConstantExpressionOrIdentifierAsLiteral(arg, context);
-
     /// This arguments are always the first
-    cluster_name = args[0]->as<ASTLiteral &>().value.safeGet<String>();
-    filename = args[1]->as<ASTLiteral &>().value.safeGet<String>();
+    configuration.cluster_name = args[0]->as<ASTLiteral &>().value.safeGet<String>();
 
-    /// Size -> argument indexes
-    static auto size_to_args = std::map<size_t, std::map<String, size_t>>
-    {
-        {4, {{"format", 2}, {"structure", 3}}},
-        {5, {{"format", 2}, {"structure", 3}, {"compression_method", 4}}},
-        {6, {{"access_key_id", 2}, {"secret_access_key", 3}, {"format", 4}, {"structure", 5}}},
-        {7, {{"access_key_id", 2}, {"secret_access_key", 3}, {"format", 4}, {"structure", 5}, {"compression_method", 6}}}
-    };
+    if (!context->tryGetCluster(configuration.cluster_name))
+        throw Exception(ErrorCodes::BAD_GET, "Requested cluster '{}' not found", configuration.cluster_name);
 
-    auto & args_to_idx = size_to_args[args.size()];
+    /// Just cut the first arg (cluster_name) and try to parse s3 table function arguments as is
+    ASTs clipped_args;
+    clipped_args.reserve(args.size());
+    std::copy(args.begin() + 1, args.end(), std::back_inserter(clipped_args));
 
-    if (args_to_idx.contains("format"))
-        format = args[args_to_idx["format"]]->as<ASTLiteral &>().value.safeGet<String>();
-
-    if (args_to_idx.contains("structure"))
-        structure = args[args_to_idx["structure"]]->as<ASTLiteral &>().value.safeGet<String>();
-
-    if (args_to_idx.contains("compression_method"))
-        compression_method = args[args_to_idx["compression_method"]]->as<ASTLiteral &>().value.safeGet<String>();
-
-    if (args_to_idx.contains("access_key_id"))
-        access_key_id = args[args_to_idx["access_key_id"]]->as<ASTLiteral &>().value.safeGet<String>();
-
-    if (args_to_idx.contains("secret_access_key"))
-        secret_access_key = args[args_to_idx["secret_access_key"]]->as<ASTLiteral &>().value.safeGet<String>();
+    /// StorageS3ClusterConfiguration inherints from StorageS3Configuration, so it is safe to upcast it.
+    TableFunctionS3::parseArgumentsImpl(message, clipped_args, context, static_cast<StorageS3Configuration & >(configuration));
 }
 
 
 ColumnsDescription TableFunctionS3Cluster::getActualTableStructure(ContextPtr context) const
 {
-    return parseColumnsListFromString(structure, context);
+    if (configuration.structure == "auto")
+    {
+        return StorageS3::getTableStructureFromData(
+            configuration.format,
+            S3::URI(Poco::URI(configuration.url)),
+            configuration.access_key_id,
+            configuration.secret_access_key,
+            context->getSettingsRef().s3_max_connections,
+            context->getSettingsRef().s3_max_single_read_retries,
+            configuration.compression_method,
+            false,
+            std::nullopt,
+            context);
+    }
+
+    return parseColumnsListFromString(configuration.structure, context);
 }
 
 StoragePtr TableFunctionS3Cluster::executeImpl(
@@ -101,46 +105,60 @@ StoragePtr TableFunctionS3Cluster::executeImpl(
     const std::string & table_name, ColumnsDescription /*cached_columns*/) const
 {
     StoragePtr storage;
+
+    UInt64 max_single_read_retries = context->getSettingsRef().s3_max_single_read_retries;
+    UInt64 min_upload_part_size = context->getSettingsRef().s3_min_upload_part_size;
+    UInt64 upload_part_size_multiply_factor = context->getSettingsRef().s3_upload_part_size_multiply_factor;
+    UInt64 upload_part_size_multiply_parts_count_threshold = context->getSettingsRef().s3_upload_part_size_multiply_parts_count_threshold;
+    UInt64 max_single_part_upload_size = context->getSettingsRef().s3_max_single_part_upload_size;
+    UInt64 max_connections = context->getSettingsRef().s3_max_connections;
+
+    ColumnsDescription columns;
+    if (configuration.structure != "auto")
+        columns = parseColumnsListFromString(configuration.structure, context);
+    else if (!structure_hint.empty())
+        columns = structure_hint;
+
     if (context->getClientInfo().query_kind == ClientInfo::QueryKind::SECONDARY_QUERY)
     {
         /// On worker node this filename won't contains globs
-        Poco::URI uri (filename);
+        Poco::URI uri (configuration.url);
         S3::URI s3_uri (uri);
-        /// Actually this parameters are not used
-        UInt64 max_single_read_retries = context->getSettingsRef().s3_max_single_read_retries;
-        UInt64 min_upload_part_size = context->getSettingsRef().s3_min_upload_part_size;
-        UInt64 upload_part_size_multiply_factor = context->getSettingsRef().s3_upload_part_size_multiply_factor;
-        UInt64 upload_part_size_multiply_parts_count_threshold = context->getSettingsRef().s3_upload_part_size_multiply_parts_count_threshold;
-        UInt64 max_single_part_upload_size = context->getSettingsRef().s3_max_single_part_upload_size;
-        UInt64 max_connections = context->getSettingsRef().s3_max_connections;
         storage = StorageS3::create(
             s3_uri,
-            access_key_id,
-            secret_access_key,
+            configuration.access_key_id,
+            configuration.secret_access_key,
             StorageID(getDatabaseName(), table_name),
-            format,
+            configuration.format,
             max_single_read_retries,
             min_upload_part_size,
             upload_part_size_multiply_factor,
             upload_part_size_multiply_parts_count_threshold,
             max_single_part_upload_size,
             max_connections,
-            getActualTableStructure(context),
+            columns,
             ConstraintsDescription{},
             String{},
             context,
             // No format_settings for S3Cluster
             std::nullopt,
-            compression_method,
+            configuration.compression_method,
             /*distributed_processing=*/true);
     }
     else
     {
         storage = StorageS3Cluster::create(
-            filename, access_key_id, secret_access_key, StorageID(getDatabaseName(), table_name),
-            cluster_name, format, context->getSettingsRef().s3_max_connections,
-            getActualTableStructure(context), ConstraintsDescription{},
-            context, compression_method);
+            configuration.url,
+            configuration.access_key_id,
+            configuration.secret_access_key,
+            StorageID(getDatabaseName(), table_name),
+            configuration.cluster_name, configuration.format,
+            max_single_read_retries,
+            max_connections,
+            columns,
+            ConstraintsDescription{},
+            context,
+            configuration.compression_method);
     }
 
     storage->startup();

src/TableFunctions/TableFunctionS3Cluster.h

@@ -5,6 +5,7 @@
 #if USE_AWS_S3
 
 #include <TableFunctions/ITableFunction.h>
+#include <Storages/ExternalDataSourceConfiguration.h>
 
 
 namespace DB
@@ -28,7 +29,12 @@ public:
     {
         return name;
     }
-    bool hasStaticStructure() const override { return true; }
+
+    bool hasStaticStructure() const override { return configuration.structure != "auto"; }
+
+    bool needStructureHint() const override { return configuration.structure == "auto"; }
+
+    void setStructureHint(const ColumnsDescription & structure_hint_) override { structure_hint = structure_hint_; }
 
 protected:
     StoragePtr executeImpl(
@@ -42,13 +48,8 @@ protected:
     ColumnsDescription getActualTableStructure(ContextPtr) const override;
     void parseArguments(const ASTPtr &, ContextPtr) override;
 
-    String cluster_name;
-    String filename;
-    String format;
-    String structure;
-    String access_key_id;
-    String secret_access_key;
-    String compression_method = "auto";
+    StorageS3ClusterConfiguration configuration;
+    ColumnsDescription structure_hint;
 };
 
 }

tests/ci/build_report_check.py

@@ -148,6 +148,17 @@ if __name__ == "__main__":
                         build_name,
                     )
 
+    some_builds_are_missing = len(build_reports_map) < len(reports_order)
+
+    if some_builds_are_missing:
+        logging.info(
+            "Expected to get %s build results, got %s",
+            len(reports_order),
+            len(build_reports_map),
+        )
+    else:
+        logging.info("Got exactly %s builds", len(build_reports_map))
+
     build_reports = [
         build_reports_map[build_name]
         for build_name in reports_order
@@ -219,10 +230,10 @@ if __name__ == "__main__":
         if build_result.status == "success":
             ok_builds += 1
 
-    if ok_builds == 0:
+    if ok_builds == 0 or some_builds_are_missing:
         summary_status = "error"
 
-    description = "{}/{} builds are OK".format(ok_builds, total_builds)
+    description = f"{ok_builds}/{total_builds} builds are OK"
 
     print("::notice ::Report url: {}".format(url))
 

tests/ci/ci_config.py

@@ -206,6 +206,7 @@ CI_CONFIG = {
             "binary_freebsd",
             "binary_darwin_aarch64",
             "binary_ppc64le",
+            "binary_gcc",
         ],
     },
     "tests_config": {

tests/ci/workflow_approve_rerun_lambda/app.py

@@ -379,12 +379,16 @@ def check_need_to_rerun(workflow_description):
 
 def rerun_workflow(workflow_description, token):
     print("Going to rerun workflow")
-    _exec_post_with_retry(workflow_description.rerun_url, token)
+    try:
+        _exec_post_with_retry(f"{workflow_description.rerun_url}-failed-jobs", token)
+    except Exception:
+        _exec_post_with_retry(workflow_description.rerun_url, token)
 
 
 def main(event):
     token = get_token_from_aws()
     event_data = json.loads(event["body"])
+    print("The body received:", event_data)
     workflow_description = get_workflow_description_from_event(event_data)
 
     print("Got workflow description", workflow_description)

tests/clickhouse-test

@@ -374,7 +374,7 @@ class SettingsRandomizer:
         "output_format_parallel_formatting": lambda: random.randint(0, 1),
         "input_format_parallel_parsing": lambda: random.randint(0, 1),
         "min_chunk_bytes_for_parallel_parsing": lambda: max(1024, int(random.gauss(10 * 1024 * 1024, 5 * 1000 * 1000))),
-        "max_read_buffer_size": lambda: random.randint(1, 20) if random.random() < 0.1 else random.randint(500000, 1048576),
+        "max_read_buffer_size": lambda: random.randint(500000, 1048576),
         "prefer_localhost_replica": lambda: random.randint(0, 1),
         "max_block_size": lambda: random.randint(8000, 100000),
         "max_threads": lambda: random.randint(1, 64),
@@ -468,9 +468,11 @@ class TestCase:
 
         return testcase_args
 
-    def add_random_settings(self, client_options):
+    def add_random_settings(self, args, client_options):
         if self.tags and 'no-random-settings' in self.tags:
             return client_options
+        if args.no_random_settings:
+            return client_options
 
         if len(self.base_url_params) == 0:
             os.environ['CLICKHOUSE_URL_PARAMS'] = '&'.join(self.random_settings)
@@ -485,9 +487,11 @@ class TestCase:
         os.environ['CLICKHOUSE_URL_PARAMS'] = self.base_url_params
         os.environ['CLICKHOUSE_CLIENT_OPT'] = self.base_client_options
 
-    def add_info_about_settings(self, description):
+    def add_info_about_settings(self, args, description):
         if self.tags and 'no-random-settings' in self.tags:
             return description
+        if args.no_random_settings:
+            return description
 
         return description + "\n" + "Settings used in the test: " + "--" + " --".join(self.random_settings) + "\n"
 
@@ -788,13 +792,13 @@ class TestCase:
 
             self.runs_count += 1
             self.testcase_args = self.configure_testcase_args(args, self.case_file, suite.suite_tmp_path)
-            client_options = self.add_random_settings(client_options)
+            client_options = self.add_random_settings(args, client_options)
             proc, stdout, stderr, total_time = self.run_single_test(server_logs_level, client_options)
 
             result = self.process_result_impl(proc, stdout, stderr, total_time)
             result.check_if_need_retry(args, stdout, stderr, self.runs_count)
             if result.status == TestStatus.FAIL:
-                result.description = self.add_info_about_settings(result.description)
+                result.description = self.add_info_about_settings(args, result.description)
             return result
         except KeyboardInterrupt as e:
             raise e
@@ -802,12 +806,12 @@ class TestCase:
             return TestResult(self.name, TestStatus.FAIL,
                               FailureReason.INTERNAL_QUERY_FAIL,
                               0.,
-                              self.add_info_about_settings(self.get_description_from_exception_info(sys.exc_info())))
+                              self.add_info_about_settings(args, self.get_description_from_exception_info(sys.exc_info())))
         except (ConnectionRefusedError, ConnectionResetError):
             return TestResult(self.name, TestStatus.FAIL,
                               FailureReason.SERVER_DIED,
                               0.,
-                              self.add_info_about_settings(self.get_description_from_exception_info(sys.exc_info())))
+                              self.add_info_about_settings(args, self.get_description_from_exception_info(sys.exc_info())))
         except:
             return TestResult(self.name, TestStatus.UNKNOWN,
                               FailureReason.INTERNAL_ERROR,
@@ -1501,6 +1505,7 @@ if __name__ == '__main__':
     parser.add_argument('--print-time', action='store_true', dest='print_time', help='Print test time')
     parser.add_argument('--check-zookeeper-session', action='store_true', help='Check ZooKeeper session uptime to determine if failed test should be retried')
     parser.add_argument('--s3-storage', action='store_true', default=False, help='Run tests over s3 storage')
+    parser.add_argument('--no-random-settings', action='store_true', default=False, help='Disable settings randomization')
 
     parser.add_argument('--run-by-hash-num', type=int, help='Run tests matching crc32(test_name) % run_by_hash_total == run_by_hash_num')
     parser.add_argument('--run-by-hash-total', type=int, help='Total test groups for crc32(test_name) % run_by_hash_total == run_by_hash_num')

tests/config/config.d/named_collection.xml

@@ -14,5 +14,13 @@
             <user>default</user>
             <table>s</table>
         </clickhouse_dictionary>
+        <url_with_headers>
+            <headers>
+                <header>
+                    <name>X-ClickHouse-Format</name>
+                    <value>JSONEachRow</value>
+                </header>
+            </headers>
+        </url_with_headers>
     </named_collections>
 </clickhouse>

tests/integration/test_storage_s3/test.py

@@ -162,7 +162,7 @@ def test_put(started_cluster, maybe_auth, positive, compression):
     values_csv = "1,2,3\n3,2,1\n78,43,45\n"
     filename = "test.csv"
     put_query = f"""insert into table function s3('http://{started_cluster.minio_ip}:{started_cluster.minio_port}/{bucket}/{filename}',
-                    {maybe_auth}'CSV', '{table_format}', {compression}) values settings s3_truncate_on_insert=1 {values}"""
+                    {maybe_auth}'CSV', '{table_format}', '{compression}') values settings s3_truncate_on_insert=1 {values}"""
 
     try:
         run_query(instance, put_query)

tests/queries/0_stateless/00385_storage_file_and_clickhouse-local_app_long.sh

@@ -19,11 +19,11 @@ function pack_unpack_compare()
 
     ${CLICKHOUSE_CLIENT} --query "CREATE TABLE buf_00385 ENGINE = Memory AS $1"
     local res_orig
-    res_orig=$(${CLICKHOUSE_CLIENT} --max_threads=1 --query "SELECT $TABLE_HASH FROM buf_00385")
+    res_orig=$(${CLICKHOUSE_CLIENT} --max_block_size=65505 --max_threads=1 --query "SELECT $TABLE_HASH FROM buf_00385")
 
     ${CLICKHOUSE_CLIENT} --max_threads=1 --query "CREATE TABLE buf_file ENGINE = File($3) AS SELECT * FROM buf_00385"
     local res_db_file
-    res_db_file=$(${CLICKHOUSE_CLIENT} --max_threads=1 --query "SELECT $TABLE_HASH FROM buf_file")
+    res_db_file=$(${CLICKHOUSE_CLIENT} --max_block_size=65505 --max_threads=1 --query "SELECT $TABLE_HASH FROM buf_file")
 
     ${CLICKHOUSE_CLIENT} --max_threads=1 --query "SELECT * FROM buf_00385 FORMAT $3" > "$buf_file"
     local res_ch_local1

tests/queries/0_stateless/01037_polygon_dicts_correctness_all.sh

@@ -34,7 +34,7 @@ CREATE TABLE test_01037.polygons_array
 ENGINE = Memory;
 "
 
-$CLICKHOUSE_CLIENT --query="INSERT INTO test_01037.polygons_array FORMAT JSONEachRow" --max_insert_block_size=100000 < "${CURDIR}/01037_polygon_data"
+$CLICKHOUSE_CLIENT --query="INSERT INTO test_01037.polygons_array FORMAT JSONEachRow" --min_chunk_bytes_for_parallel_parsing=10485760 --max_insert_block_size=100000 < "${CURDIR}/01037_polygon_data"
 
 rm "${CURDIR}"/01037_polygon_data
 

tests/queries/0_stateless/01059_storage_file_compression.sh

@@ -16,10 +16,10 @@ do
     ${CLICKHOUSE_CLIENT} --query "DROP TABLE file"
 done
 
-${CLICKHOUSE_CLIENT} --query "SELECT count(), max(x) FROM file('${CLICKHOUSE_DATABASE}/{gz,br,xz,zst,lz4,bz2}.tsv.{gz,br,xz,zst,lz4,bz2}', TSV, 'x UInt64')"
+${CLICKHOUSE_CLIENT} --max_read_buffer_size=1048576 --query "SELECT count(), max(x) FROM file('${CLICKHOUSE_DATABASE}/{gz,br,xz,zst,lz4,bz2}.tsv.{gz,br,xz,zst,lz4,bz2}', TSV, 'x UInt64')"
 
 for m in gz br xz zst lz4 bz2
 do
-    ${CLICKHOUSE_CLIENT} --query "SELECT count() < 4000000, max(x) FROM file('${CLICKHOUSE_DATABASE}/${m}.tsv.${m}', RowBinary, 'x UInt8', 'none')"
+    ${CLICKHOUSE_CLIENT} --max_read_buffer_size=1048576 --query "SELECT count() < 4000000, max(x) FROM file('${CLICKHOUSE_DATABASE}/${m}.tsv.${m}', RowBinary, 'x UInt8', 'none')"
 done
 

tests/queries/0_stateless/01091_num_threads.sql

@@ -1,5 +1,6 @@
 set log_queries=1;
 set log_query_threads=1;
+set max_threads=0;
 
 WITH 01091 AS id SELECT 1;
 SYSTEM FLUSH LOGS;

tests/queries/0_stateless/01108_restart_replicas_rename_deadlock_zookeeper.sh

@@ -38,27 +38,17 @@ function restart_replicas_loop()
     done
     sleep 0.$RANDOM
 }
-function restart_thread_1()
-{
-    restart_replicas_loop
-}
-
-function restart_thread_2()
-{
-    restart_replicas_loop
-}
 
 export -f rename_thread_1
 export -f rename_thread_2
-export -f restart_thread_1
-export -f restart_thread_2
+export -f restart_replicas_loop
 
 TIMEOUT=10
 
 clickhouse_client_loop_timeout $TIMEOUT rename_thread_1 2> /dev/null &
 clickhouse_client_loop_timeout $TIMEOUT rename_thread_2 2> /dev/null &
-clickhouse_client_loop_timeout $TIMEOUT restart_thread_1 2> /dev/null &
-clickhouse_client_loop_timeout $TIMEOUT restart_thread_2 2> /dev/null &
+clickhouse_client_loop_timeout $TIMEOUT restart_replicas_loop 2> /dev/null &
+clickhouse_client_loop_timeout $TIMEOUT restart_replicas_loop 2> /dev/null &
 
 wait
 

tests/queries/0_stateless/01268_procfs_metrics.sh

@@ -41,4 +41,13 @@ timeout $TIMEOUT bash -c show_processes_func &
 
 wait
 
+# otherwise it can be alive after test
+query_alive=$($CLICKHOUSE_CLIENT --query "SELECT count() FROM system.processes WHERE query ILIKE 'SELECT * FROM numbers(600000000)%'")
+while [[ $query_alive != 0 ]]
+do
+    $CLICKHOUSE_CLIENT -q "KILL QUERY WHERE query ilike '%SELECT * FROM numbers(600000000)%'" 2> /dev/null 1> /dev/null
+    sleep 0.5
+    query_alive=$($CLICKHOUSE_CLIENT --query "SELECT count() FROM system.processes WHERE query ILIKE 'SELECT * FROM numbers(600000000)%'")
+done
+
 echo "Test OK"

tests/queries/0_stateless/01307_multiple_leaders_zookeeper.sh

@@ -20,7 +20,8 @@ function thread()
     REPLICA=$1
     ITERATIONS=$2
 
-    $CLICKHOUSE_CLIENT --max_block_size 1 --min_insert_block_size_rows 0 --min_insert_block_size_bytes 0 --query "INSERT INTO r$REPLICA SELECT number * $NUM_REPLICAS + $REPLICA FROM numbers($ITERATIONS)"
+    # It's legal to fetch something before insert finished
+    $CLICKHOUSE_CLIENT --max_block_size 1 --min_insert_block_size_rows 0 --min_insert_block_size_bytes 0 --query "INSERT INTO r$REPLICA SELECT number * $NUM_REPLICAS + $REPLICA FROM numbers($ITERATIONS)" 2>&1 | grep -v -F "Tried to commit obsolete part"
 }
 
 for REPLICA in $SEQ; do

tests/queries/0_stateless/01514_parallel_formatting.sql

@@ -1,5 +1,8 @@
 drop table if exists tsv;
 set output_format_parallel_formatting=1;
+set max_read_buffer_size=1048576;
+set max_block_size=65505;
+
 create table tsv(a int, b int default 7) engine File(TSV);
 
 insert into tsv(a) select number from numbers(10000000);

tests/queries/0_stateless/01529_bad_memory_tracking.sh

@@ -8,6 +8,6 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
 . "$CURDIR"/../shell_config.sh
 
 for _ in {1..10}; do
-    ${CLICKHOUSE_CLIENT} --max_memory_usage '10G' --query "SELECT i FROM generateRandom('i Array(Int8)', 1, 1, 1048577) LIMIT 65536" |& grep -v -e 'Received exception from server' -e 'Code: 241' -e '(query: '
+    ${CLICKHOUSE_CLIENT} --max_block_size=65505 --max_memory_usage '10G' --query "SELECT i FROM generateRandom('i Array(Int8)', 1, 1, 1048577) LIMIT 65536" |& grep -v -e 'Received exception from server' -e 'Code: 241' -e '(query: '
 done
 exit 0

tests/queries/0_stateless/01532_primary_key_without_order_by_zookeeper.reference

@@ -9,8 +9,8 @@ CREATE TABLE default.merge_tree_pk_sql\n(\n    `key` UInt64,\n    `value` String
 1	c
 2	b
 1	c	0
-2	e	555
 2	b	0
+2	e	555
 CREATE TABLE default.merge_tree_pk_sql\n(\n    `key` UInt64,\n    `value` String,\n    `key2` UInt64\n)\nENGINE = ReplacingMergeTree\nPRIMARY KEY key\nORDER BY (key, key2)\nSETTINGS index_granularity = 8192
 CREATE TABLE default.replicated_merge_tree_pk_sql\n(\n    `key` UInt64,\n    `value` String\n)\nENGINE = ReplicatedReplacingMergeTree(\'/clickhouse/test/01532_primary_key_without\', \'r1\')\nPRIMARY KEY key\nORDER BY key\nSETTINGS index_granularity = 8192
 1	a
@@ -18,6 +18,6 @@ CREATE TABLE default.replicated_merge_tree_pk_sql\n(\n    `key` UInt64,\n    `va
 1	c
 2	b
 1	c	0
-2	e	555
 2	b	0
+2	e	555
 CREATE TABLE default.replicated_merge_tree_pk_sql\n(\n    `key` UInt64,\n    `value` String,\n    `key2` UInt64\n)\nENGINE = ReplicatedReplacingMergeTree(\'/clickhouse/test/01532_primary_key_without\', \'r1\')\nPRIMARY KEY key\nORDER BY (key, key2)\nSETTINGS index_granularity = 8192

tests/queries/0_stateless/01532_primary_key_without_order_by_zookeeper.sql

@@ -1,4 +1,4 @@
--- Tags: zookeeper
+-- Tags: zookeeper, no-parallel
 
 DROP TABLE IF EXISTS merge_tree_pk;
 
@@ -15,14 +15,14 @@ SHOW CREATE TABLE merge_tree_pk;
 INSERT INTO merge_tree_pk VALUES (1, 'a');
 INSERT INTO merge_tree_pk VALUES (2, 'b');
 
-SELECT * FROM merge_tree_pk ORDER BY key;
+SELECT * FROM merge_tree_pk ORDER BY key, value;
 
 INSERT INTO merge_tree_pk VALUES (1, 'c');
 
 DETACH TABLE merge_tree_pk;
 ATTACH TABLE merge_tree_pk;
 
-SELECT * FROM merge_tree_pk FINAL ORDER BY key;
+SELECT * FROM merge_tree_pk FINAL ORDER BY key, value;
 
 DROP TABLE IF EXISTS merge_tree_pk;
 
@@ -41,14 +41,14 @@ SHOW CREATE TABLE merge_tree_pk_sql;
 INSERT INTO merge_tree_pk_sql VALUES (1, 'a');
 INSERT INTO merge_tree_pk_sql VALUES (2, 'b');
 
-SELECT * FROM merge_tree_pk_sql ORDER BY key;
+SELECT * FROM merge_tree_pk_sql ORDER BY key, value;
 
 INSERT INTO merge_tree_pk_sql VALUES (1, 'c');
 
 DETACH TABLE merge_tree_pk_sql;
 ATTACH TABLE merge_tree_pk_sql;
 
-SELECT * FROM merge_tree_pk_sql FINAL ORDER BY key;
+SELECT * FROM merge_tree_pk_sql FINAL ORDER BY key, value;
 
 ALTER TABLE merge_tree_pk_sql ADD COLUMN key2 UInt64, MODIFY ORDER BY (key, key2);
 
@@ -56,7 +56,7 @@ INSERT INTO merge_tree_pk_sql VALUES (2, 'd', 555);
 
 INSERT INTO merge_tree_pk_sql VALUES (2, 'e', 555);
 
-SELECT * FROM merge_tree_pk_sql FINAL ORDER BY key;
+SELECT * FROM merge_tree_pk_sql FINAL ORDER BY key, value;
 
 SHOW CREATE TABLE merge_tree_pk_sql;
 
@@ -77,14 +77,14 @@ SHOW CREATE TABLE replicated_merge_tree_pk_sql;
 INSERT INTO replicated_merge_tree_pk_sql VALUES (1, 'a');
 INSERT INTO replicated_merge_tree_pk_sql VALUES (2, 'b');
 
-SELECT * FROM replicated_merge_tree_pk_sql ORDER BY key;
+SELECT * FROM replicated_merge_tree_pk_sql ORDER BY key, value;
 
 INSERT INTO replicated_merge_tree_pk_sql VALUES (1, 'c');
 
 DETACH TABLE replicated_merge_tree_pk_sql;
 ATTACH TABLE replicated_merge_tree_pk_sql;
 
-SELECT * FROM replicated_merge_tree_pk_sql FINAL ORDER BY key;
+SELECT * FROM replicated_merge_tree_pk_sql FINAL ORDER BY key, value;
 
 ALTER TABLE replicated_merge_tree_pk_sql ADD COLUMN key2 UInt64, MODIFY ORDER BY (key, key2);
 
@@ -92,7 +92,7 @@ INSERT INTO replicated_merge_tree_pk_sql VALUES (2, 'd', 555);
 
 INSERT INTO replicated_merge_tree_pk_sql VALUES (2, 'e', 555);
 
-SELECT * FROM replicated_merge_tree_pk_sql FINAL ORDER BY key;
+SELECT * FROM replicated_merge_tree_pk_sql FINAL ORDER BY key, value;
 
 DETACH TABLE replicated_merge_tree_pk_sql;
 ATTACH TABLE replicated_merge_tree_pk_sql;

tests/queries/0_stateless/01801_s3_cluster.reference

@@ -2,6 +2,158 @@
 0	0	0
 0	0	0
 1	2	3
+10	11	12
+13	14	15
+16	17	18
+20	21	22
+23	24	25
+26	27	28
+4	5	6
+7	8	9
+0	0	0
+0	0	0
+0	0	0
+1	2	3
+10	11	12
+13	14	15
+16	17	18
+20	21	22
+23	24	25
+26	27	28
+4	5	6
+7	8	9
+0	0	0
+0	0	0
+0	0	0
+1	2	3
+4	5	6
+7	8	9
+10	11	12
+13	14	15
+16	17	18
+20	21	22
+23	24	25
+26	27	28
+0	0	0
+0	0	0
+0	0	0
+1	2	3
+10	11	12
+13	14	15
+16	17	18
+20	21	22
+23	24	25
+26	27	28
+4	5	6
+7	8	9
+0	0	0
+0	0	0
+0	0	0
+1	2	3
+4	5	6
+7	8	9
+10	11	12
+13	14	15
+16	17	18
+20	21	22
+23	24	25
+26	27	28
+0	0	0
+0	0	0
+0	0	0
+1	2	3
+10	11	12
+13	14	15
+16	17	18
+20	21	22
+23	24	25
+26	27	28
+4	5	6
+7	8	9
+0	0	0
+0	0	0
+0	0	0
+1	2	3
+4	5	6
+7	8	9
+10	11	12
+13	14	15
+16	17	18
+20	21	22
+23	24	25
+26	27	28
+0	0	0
+0	0	0
+0	0	0
+1	2	3
+4	5	6
+7	8	9
+10	11	12
+13	14	15
+16	17	18
+20	21	22
+23	24	25
+26	27	28
+0	0	0
+0	0	0
+10	11	12
+13	14	15
+16	17	18
+20	21	22
+23	24	25
+26	27	28
+0	0	0
+0	0	0
+10	11	12
+13	14	15
+16	17	18
+20	21	22
+23	24	25
+26	27	28
+0	0	0
+0	0	0
+0	0	0
+1	2	3
+4	5	6
+7	8	9
+10	11	12
+13	14	15
+16	17	18
+20	21	22
+23	24	25
+26	27	28
+0	0	0
+0	0	0
+10	11	12
+13	14	15
+16	17	18
+20	21	22
+23	24	25
+26	27	28
+0	0	0
+0	0	0
+0	0	0
+1	2	3
+4	5	6
+7	8	9
+10	11	12
+13	14	15
+16	17	18
+20	21	22
+23	24	25
+26	27	28
+0	0	0
+0	0	0
+10	11	12
+13	14	15
+16	17	18
+20	21	22
+23	24	25
+26	27	28
+0	0	0
+0	0	0
+0	0	0
+1	2	3
 4	5	6
 7	8	9
 10	11	12

tests/queries/0_stateless/01801_s3_cluster.sql

@@ -1,5 +1,22 @@
 -- Tags: no-fasttest
 -- Tag no-fasttest: Depends on AWS
 
-select * from s3('http://localhost:11111/test/{a,b,c}.tsv', 'test', 'testtest', 'TSV', 'a UInt64, b UInt64, c UInt64') ORDER BY a, b, c;
-select * from s3Cluster('test_cluster_two_shards', 'http://localhost:11111/test/{a,b,c}.tsv', 'test', 'testtest', 'TSV', 'a UInt64, b UInt64, c UInt64') ORDER BY a, b, c;
+select * from s3('http://localhost:11111/test/{a,b,c}.tsv') ORDER BY c1, c2, c3;
+select * from s3('http://localhost:11111/test/{a,b,c}.tsv', 'TSV') ORDER BY c1, c2, c3;
+select * from s3('http://localhost:11111/test/{a,b,c}.tsv', 'TSV', 'c1 UInt64, c2 UInt64, c3 UInt64') ORDER BY c1, c2, c3;
+select * from s3('http://localhost:11111/test/{a,b,c}.tsv', 'test', 'testtest') ORDER BY c1, c2, c3;
+select * from s3('http://localhost:11111/test/{a,b,c}.tsv', 'TSV', 'c1 UInt64, c2 UInt64, c3 UInt64', 'auto') ORDER BY c1, c2, c3;
+select * from s3('http://localhost:11111/test/{a,b,c}.tsv', 'test', 'testtest', 'TSV') ORDER BY c1, c2, c3;
+select * from s3('http://localhost:11111/test/{a,b,c}.tsv', 'test', 'testtest', 'TSV', 'c1 UInt64, c2 UInt64, c3 UInt64') ORDER BY c1, c2, c3;
+select * from s3('http://localhost:11111/test/{a,b,c}.tsv', 'test', 'testtest', 'TSV', 'c1 UInt64, c2 UInt64, c3 UInt64', 'auto') ORDER BY c1, c2, c3;
+
+
+select * from s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv') ORDER BY c1, c2, c3;
+select * from s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv', 'TSV') ORDER BY c1, c2, c3;
+select * from s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv', 'TSV', 'c1 UInt64, c2 UInt64, c3 UInt64') ORDER BY c1, c2, c3;
+select * from s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv', 'test', 'testtest') ORDER BY c1, c2, c3;
+select * from s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv', 'TSV', 'c1 UInt64, c2 UInt64, c3 UInt64', 'auto') ORDER BY c1, c2, c3;
+select * from s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv', 'test', 'testtest', 'TSV') ORDER BY c1, c2, c3;
+select * from s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv', 'test', 'testtest', 'TSV', 'c1 UInt64, c2 UInt64, c3 UInt64') ORDER BY c1, c2, c3;
+select * from s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv', 'test', 'testtest', 'TSV', 'c1 UInt64, c2 UInt64, c3 UInt64', 'auto') ORDER BY c1, c2, c3;
+

tests/queries/0_stateless/01825_type_json_btc.sh

@@ -5,11 +5,16 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
 # shellcheck source=../shell_config.sh
 . "$CUR_DIR"/../shell_config.sh
 
+user_files_path=$(clickhouse-client --query "select _path,_file from file('nonexist.txt', 'CSV', 'val1 char')" 2>&1 | grep Exception | awk '{gsub("/nonexist.txt","",$9); print $9}')
+mkdir -p ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/
+rm -rf ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME:?}/*
+cp $CUR_DIR/data_json/btc_transactions.json ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/
+
 ${CLICKHOUSE_CLIENT} -q "DROP TABLE IF EXISTS btc"
 
 ${CLICKHOUSE_CLIENT} -q "CREATE TABLE btc (data JSON) ENGINE = MergeTree ORDER BY tuple()" --allow_experimental_object_type 1
 
-cat $CUR_DIR/data_json/btc_transactions.json | ${CLICKHOUSE_CLIENT} -q "INSERT INTO btc FORMAT JSONAsObject"
+${CLICKHOUSE_CLIENT} -q "INSERT INTO btc SELECT * FROM file('${CLICKHOUSE_TEST_UNIQUE_NAME}/btc_transactions.json', 'JSONAsObject')"
 
 ${CLICKHOUSE_CLIENT} -q "SELECT count() FROM btc WHERE NOT ignore(*)"
 ${CLICKHOUSE_CLIENT} -q "DESC btc SETTINGS describe_extend_object_types = 1"
@@ -21,3 +26,5 @@ ${CLICKHOUSE_CLIENT} -q "SELECT avg(length(data.inputs.prev_out.spending_outpoin
 ${CLICKHOUSE_CLIENT} -q "SELECT data.out.spending_outpoints AS outpoints FROM btc WHERE arrayExists(x -> notEmpty(x), outpoints)"
 
 ${CLICKHOUSE_CLIENT} -q "DROP TABLE IF EXISTS btc"
+
+rm ${user_files_path}/${CLICKHOUSE_TEST_UNIQUE_NAME}/btc_transactions.json

tests/queries/0_stateless/02006_test_positional_arguments.reference

@@ -111,3 +111,9 @@ select substr('aaaaaaaaaaaaaa', 8) as a  group by a;
 aaaaaaa
 select substr('aaaaaaaaaaaaaa', 8) as a  group by substr('aaaaaaaaaaaaaa', 8);
 aaaaaaa
+select b from (select 5 as a, 'Hello' as b order by a);
+Hello
+select b from (select 5 as a, 'Hello' as b group by a);
+Hello
+select b from (select 5 as a, 'Hello' as b order by 1);
+Hello

tests/queries/0_stateless/02006_test_positional_arguments.sql

@@ -45,3 +45,6 @@ explain syntax select plus(1, 1) as a group by a;
 select substr('aaaaaaaaaaaaaa', 8) as a  group by a;
 select substr('aaaaaaaaaaaaaa', 8) as a  group by substr('aaaaaaaaaaaaaa', 8);
 
+select b from (select 5 as a, 'Hello' as b order by a);
+select b from (select 5 as a, 'Hello' as b group by a);
+select b from (select 5 as a, 'Hello' as b order by 1);

tests/queries/0_stateless/02149_read_in_order_fixed_prefix.sql

@@ -1,3 +1,5 @@
+SET max_threads=0;
+
 DROP TABLE IF EXISTS t_read_in_order;
 
 CREATE TABLE t_read_in_order(date Date, i UInt64, v UInt64)

tests/queries/0_stateless/02177_issue_31009.sql

@@ -1,5 +1,10 @@
 -- Tags: long
 
+SET max_threads=0;
+
+DROP TABLE IF EXISTS left;
+DROP TABLE IF EXISTS right;
+
 CREATE TABLE left ( key UInt32, value String ) ENGINE = MergeTree ORDER BY key;
 CREATE TABLE right (  key UInt32, value String ) ENGINE = MergeTree ORDER BY tuple();
 

tests/queries/0_stateless/02244_issue_35598_fuse.reference

@@ -0,0 +1,2 @@
+0	0	nan
+0	0	nan

tests/queries/0_stateless/02244_issue_35598_fuse.sql

@@ -0,0 +1,5 @@
+SELECT sum(x), count(x), avg(x) FROM (SELECT number :: Decimal32(0) AS x FROM numbers(0))
+SETTINGS optimize_syntax_fuse_functions = 0, optimize_fuse_sum_count_avg = 0;
+
+SELECT sum(x), count(x), avg(x) FROM (SELECT number :: Decimal32(0) AS x FROM numbers(0))
+SETTINGS optimize_syntax_fuse_functions = 1, optimize_fuse_sum_count_avg = 1;

tests/queries/0_stateless/02244_url_engine_headers_test.reference

@@ -0,0 +1 @@
+{"12":12}\n

tests/queries/0_stateless/02244_url_engine_headers_test.sql

@@ -0,0 +1 @@
+select * from url(url_with_headers, url='http://127.0.0.1:8123?query=select+12', format='RawBLOB');

tests/queries/0_stateless/02245_s3_schema_desc.reference

@@ -0,0 +1,24 @@
+c1	Nullable(String)					
+c2	Nullable(String)					
+c3	Nullable(String)					
+c1	Nullable(String)					
+c2	Nullable(String)					
+c3	Nullable(String)					
+c1	UInt64					
+c2	UInt64					
+c3	UInt64					
+c1	Nullable(String)					
+c2	Nullable(String)					
+c3	Nullable(String)					
+c1	UInt64					
+c2	UInt64					
+c3	UInt64					
+c1	Nullable(String)					
+c2	Nullable(String)					
+c3	Nullable(String)					
+c1	UInt64					
+c2	UInt64					
+c3	UInt64					
+c1	UInt64					
+c2	UInt64					
+c3	UInt64					

tests/queries/0_stateless/02245_s3_schema_desc.sql

@@ -0,0 +1,13 @@
+-- Tags: no-fasttest
+-- Tag no-fasttest: Depends on AWS
+
+desc s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv');
+desc s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv', 'TSV');
+desc s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv', 'TSV', 'c1 UInt64, c2 UInt64, c3 UInt64');
+desc s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv', 'test', 'testtest');
+desc s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv', 'TSV', 'c1 UInt64, c2 UInt64, c3 UInt64', 'auto');
+desc s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv', 'test', 'testtest', 'TSV');
+desc s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv', 'test', 'testtest', 'TSV', 'c1 UInt64, c2 UInt64, c3 UInt64');
+desc s3Cluster('test_cluster_two_shards_localhost', 'http://localhost:11111/test/{a,b,c}.tsv', 'test', 'testtest', 'TSV', 'c1 UInt64, c2 UInt64, c3 UInt64', 'auto');
+
+SELECT * FROM s3(decodeURLComponent(NULL), [NULL]);  --{serverError 170}

utils/check-style/check-black

@@ -6,7 +6,7 @@ set -e
 GIT_ROOT=$(git rev-parse --show-cdup)
 GIT_ROOT=${GIT_ROOT:-.}
 tmp=$(mktemp)
-if ! find "$GIT_ROOT" -name '*.py' -not -path "$GIT_ROOT/contrib/*" -exec black --check {} + 1>"$tmp" 2>&1; then
+if ! find "$GIT_ROOT" -name '*.py' -not -path "$GIT_ROOT/contrib/*" -exec black --check --diff {} + 1>"$tmp" 2>&1; then
   # Show the result only if some files need formatting
   cat "$tmp"
 fi