diff --git a/.github/workflows/backport_branches.yml b/.github/workflows/backport_branches.yml index d69168b01ee..1df8cea4ad5 100644 --- a/.github/workflows/backport_branches.yml +++ b/.github/workflows/backport_branches.yml @@ -3,6 +3,9 @@ name: BackportPR env: # Force the stdout and stderr streams to be unbuffered PYTHONUNBUFFERED: 1 + # Export system tables to ClickHouse Cloud + CLICKHOUSE_CI_LOGS_HOST: ${{ secrets.CLICKHOUSE_CI_LOGS_HOST }} + CLICKHOUSE_CI_LOGS_PASSWORD: ${{ secrets.CLICKHOUSE_CI_LOGS_PASSWORD }} on: # yamllint disable-line rule:truthy push: diff --git a/.github/workflows/master.yml b/.github/workflows/master.yml index ae1862e327f..f25611e8666 100644 --- a/.github/workflows/master.yml +++ b/.github/workflows/master.yml @@ -3,6 +3,9 @@ name: MasterCI env: # Force the stdout and stderr streams to be unbuffered PYTHONUNBUFFERED: 1 + # Export system tables to ClickHouse Cloud + CLICKHOUSE_CI_LOGS_HOST: ${{ secrets.CLICKHOUSE_CI_LOGS_HOST }} + CLICKHOUSE_CI_LOGS_PASSWORD: ${{ secrets.CLICKHOUSE_CI_LOGS_PASSWORD }} on: # yamllint disable-line rule:truthy push: diff --git a/.github/workflows/pull_request.yml b/.github/workflows/pull_request.yml index d97b9975c3c..25927954809 100644 --- a/.github/workflows/pull_request.yml +++ b/.github/workflows/pull_request.yml @@ -3,6 +3,9 @@ name: PullRequestCI env: # Force the stdout and stderr streams to be unbuffered PYTHONUNBUFFERED: 1 + # Export system tables to ClickHouse Cloud + CLICKHOUSE_CI_LOGS_HOST: ${{ secrets.CLICKHOUSE_CI_LOGS_HOST }} + CLICKHOUSE_CI_LOGS_PASSWORD: ${{ secrets.CLICKHOUSE_CI_LOGS_PASSWORD }} on: # yamllint disable-line rule:truthy pull_request: diff --git a/.github/workflows/release_branches.yml b/.github/workflows/release_branches.yml index 21284815583..e5527177aa5 100644 --- a/.github/workflows/release_branches.yml +++ b/.github/workflows/release_branches.yml @@ -3,6 +3,9 @@ name: ReleaseBranchCI env: # Force the stdout and stderr streams to be unbuffered PYTHONUNBUFFERED: 1 + # Export system tables to ClickHouse Cloud + CLICKHOUSE_CI_LOGS_HOST: ${{ secrets.CLICKHOUSE_CI_LOGS_HOST }} + CLICKHOUSE_CI_LOGS_PASSWORD: ${{ secrets.CLICKHOUSE_CI_LOGS_PASSWORD }} on: # yamllint disable-line rule:truthy push: diff --git a/CMakeLists.txt b/CMakeLists.txt index 157f8c9fac0..55bcf5fbf3c 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -208,9 +208,6 @@ option(OMIT_HEAVY_DEBUG_SYMBOLS "Do not generate debugger info for heavy modules (ClickHouse functions and dictionaries, some contrib)" ${OMIT_HEAVY_DEBUG_SYMBOLS_DEFAULT}) -if (CMAKE_BUILD_TYPE_UC STREQUAL "DEBUG") - set(USE_DEBUG_HELPERS ON) -endif() option(USE_DEBUG_HELPERS "Enable debug helpers" ${USE_DEBUG_HELPERS}) option(BUILD_STANDALONE_KEEPER "Build keeper as small standalone binary" OFF) diff --git a/base/base/JSON.cpp b/base/base/JSON.cpp index 4c6d97b4444..0b43be38149 100644 --- a/base/base/JSON.cpp +++ b/base/base/JSON.cpp @@ -7,8 +7,6 @@ #include #include -#include - #define JSON_MAX_DEPTH 100 diff --git a/base/base/wide_integer_impl.h b/base/base/wide_integer_impl.h index 411841e6d9f..fc4e9e551ca 100644 --- a/base/base/wide_integer_impl.h +++ b/base/base/wide_integer_impl.h @@ -12,7 +12,6 @@ #include #include -#include #include // NOLINTBEGIN(*) @@ -22,6 +21,7 @@ #define CONSTEXPR_FROM_DOUBLE constexpr using FromDoubleIntermediateType = long double; #else +#include /// `wide_integer_from_builtin` can't be constexpr with non-literal `cpp_bin_float_double_extended` #define CONSTEXPR_FROM_DOUBLE using FromDoubleIntermediateType = boost::multiprecision::cpp_bin_float_double_extended; diff --git a/base/poco/Data/ODBC/src/Unicode_UNIXODBC.cpp b/base/poco/Data/ODBC/src/Unicode_UNIXODBC.cpp index 4caf097c28a..1c5555f8cf3 100644 --- a/base/poco/Data/ODBC/src/Unicode_UNIXODBC.cpp +++ b/base/poco/Data/ODBC/src/Unicode_UNIXODBC.cpp @@ -19,7 +19,6 @@ #include "Poco/UTF16Encoding.h" #include "Poco/Buffer.h" #include "Poco/Exception.h" -#include using Poco::Buffer; diff --git a/base/poco/Foundation/src/Task.cpp b/base/poco/Foundation/src/Task.cpp index a850ae37eff..4303d50d6e8 100644 --- a/base/poco/Foundation/src/Task.cpp +++ b/base/poco/Foundation/src/Task.cpp @@ -16,7 +16,6 @@ #include "Poco/TaskManager.h" #include "Poco/Exception.h" -#include #include diff --git a/base/poco/JSON/src/Object.cpp b/base/poco/JSON/src/Object.cpp index 7fca65c5b01..b041f570934 100644 --- a/base/poco/JSON/src/Object.cpp +++ b/base/poco/JSON/src/Object.cpp @@ -14,7 +14,6 @@ #include "Poco/JSON/Object.h" #include -#include using Poco::Dynamic::Var; diff --git a/base/poco/Net/src/HTTPClientSession.cpp b/base/poco/Net/src/HTTPClientSession.cpp index c5697b556d1..2712c0c452e 100644 --- a/base/poco/Net/src/HTTPClientSession.cpp +++ b/base/poco/Net/src/HTTPClientSession.cpp @@ -26,7 +26,6 @@ #include "Poco/CountingStream.h" #include "Poco/RegularExpression.h" #include -#include using Poco::NumberFormatter; diff --git a/docker/README.md b/docker/README.md index ec52ddd143e..bae02201add 100644 --- a/docker/README.md +++ b/docker/README.md @@ -1,5 +1,5 @@ ## ClickHouse Dockerfiles -This directory contain Dockerfiles for `clickhouse-client` and `clickhouse-server`. They are updated in each release. +This directory contain Dockerfiles for `clickhouse-server`. They are updated in each release. Also there is bunch of images for testing and CI. They are listed in `images.json` file and updated on each commit to master. If you need to add another image, place information about it into `images.json`. diff --git a/docker/client/Dockerfile b/docker/client/Dockerfile deleted file mode 100644 index 1c185daec75..00000000000 --- a/docker/client/Dockerfile +++ /dev/null @@ -1,34 +0,0 @@ -FROM ubuntu:18.04 - -# ARG for quick switch to a given ubuntu mirror -ARG apt_archive="http://archive.ubuntu.com" -RUN sed -i "s|http://archive.ubuntu.com|$apt_archive|g" /etc/apt/sources.list - -ARG repository="deb https://repo.clickhouse.com/deb/stable/ main/" -ARG version=22.1.1.* - -RUN apt-get update \ - && apt-get install --yes --no-install-recommends \ - apt-transport-https \ - ca-certificates \ - dirmngr \ - gnupg \ - && mkdir -p /etc/apt/sources.list.d \ - && apt-key adv --keyserver keyserver.ubuntu.com --recv E0C56BD4 \ - && echo $repository > /etc/apt/sources.list.d/clickhouse.list \ - && apt-get update \ - && env DEBIAN_FRONTEND=noninteractive \ - apt-get install --allow-unauthenticated --yes --no-install-recommends \ - clickhouse-client=$version \ - clickhouse-common-static=$version \ - locales \ - tzdata \ - && rm -rf /var/lib/apt/lists/* /var/cache/debconf \ - && apt-get clean - -RUN locale-gen en_US.UTF-8 -ENV LANG en_US.UTF-8 -ENV LANGUAGE en_US:en -ENV LC_ALL en_US.UTF-8 - -ENTRYPOINT ["/usr/bin/clickhouse-client"] diff --git a/docker/client/README.md b/docker/client/README.md deleted file mode 100644 index bbcc7d60794..00000000000 --- a/docker/client/README.md +++ /dev/null @@ -1,7 +0,0 @@ -# ClickHouse Client Docker Image - -For more information see [ClickHouse Server Docker Image](https://hub.docker.com/r/clickhouse/clickhouse-server/). - -## License - -View [license information](https://github.com/ClickHouse/ClickHouse/blob/master/LICENSE) for the software contained in this image. diff --git a/docker/packager/packager b/docker/packager/packager index b6b9682f418..c0baacccef2 100755 --- a/docker/packager/packager +++ b/docker/packager/packager @@ -22,7 +22,7 @@ def check_image_exists_locally(image_name: str) -> bool: output = subprocess.check_output( f"docker images -q {image_name} 2> /dev/null", shell=True ) - return output != "" + return output != b"" except subprocess.CalledProcessError: return False @@ -46,7 +46,7 @@ def build_image(image_name: str, filepath: Path) -> None: ) -def pre_build(repo_path: Path, env_variables: List[str]): +def pre_build(repo_path: Path, env_variables: List[str]) -> None: if "WITH_PERFORMANCE=1" in env_variables: current_branch = subprocess.check_output( "git branch --show-current", shell=True, encoding="utf-8" @@ -81,8 +81,9 @@ def run_docker_image_with_env( env_variables: List[str], ch_root: Path, ccache_dir: Optional[Path], -): +) -> None: output_dir.mkdir(parents=True, exist_ok=True) + env_part = " -e ".join(env_variables) if env_part: env_part = " -e " + env_part @@ -129,9 +130,10 @@ def parse_env_variables( version: str, official: bool, additional_pkgs: bool, + with_profiler: bool, with_coverage: bool, with_binaries: str, -): +) -> List[str]: DARWIN_SUFFIX = "-darwin" DARWIN_ARM_SUFFIX = "-darwin-aarch64" ARM_SUFFIX = "-aarch64" @@ -322,6 +324,9 @@ def parse_env_variables( # utils are not included into clickhouse-bundle, so build everything build_target = "all" + if with_profiler: + cmake_flags.append("-DENABLE_BUILD_PROFILING=1") + if with_coverage: cmake_flags.append("-DWITH_COVERAGE=1") @@ -416,6 +421,7 @@ def parse_args() -> argparse.Namespace: parser.add_argument("--version") parser.add_argument("--official", action="store_true") parser.add_argument("--additional-pkgs", action="store_true") + parser.add_argument("--with-profiler", action="store_true") parser.add_argument("--with-coverage", action="store_true") parser.add_argument( "--with-binaries", choices=("programs", "tests", ""), default="" @@ -451,7 +457,7 @@ def parse_args() -> argparse.Namespace: return args -def main(): +def main() -> None: logging.basicConfig(level=logging.INFO, format="%(asctime)s %(message)s") args = parse_args() @@ -479,6 +485,7 @@ def main(): args.version, args.official, args.additional_pkgs, + args.with_profiler, args.with_coverage, args.with_binaries, ) diff --git a/docker/test/performance-comparison/compare.sh b/docker/test/performance-comparison/compare.sh index 798d2a40b12..f949e66ab17 100755 --- a/docker/test/performance-comparison/compare.sh +++ b/docker/test/performance-comparison/compare.sh @@ -665,9 +665,8 @@ create view partial_query_times as select * from -- Report for backward-incompatible ('partial') queries that we could only run on the new server (e.g. -- queries with new functions added in the tested PR). create table partial_queries_report engine File(TSV, 'report/partial-queries-report.tsv') - settings output_format_decimal_trailing_zeros = 1 - as select toDecimal64(time_median, 3) time, - toDecimal64(time_stddev / time_median, 3) relative_time_stddev, + as select round(time_median, 3) time, + round(time_stddev / time_median, 3) relative_time_stddev, test, query_index, query_display_name from partial_query_times join query_display_names using (test, query_index) @@ -739,28 +738,26 @@ create table queries engine File(TSVWithNamesAndTypes, 'report/queries.tsv') ; create table changed_perf_report engine File(TSV, 'report/changed-perf.tsv') - settings output_format_decimal_trailing_zeros = 1 as with -- server_time is sometimes reported as zero (if it's less than 1 ms), -- so we have to work around this to not get an error about conversion -- of NaN to decimal. (left > right ? left / right : right / left) as times_change_float, isFinite(times_change_float) as times_change_finite, - toDecimal64(times_change_finite ? times_change_float : 1., 3) as times_change_decimal, + round(times_change_finite ? times_change_float : 1., 3) as times_change_decimal, times_change_finite ? (left > right ? '-' : '+') || toString(times_change_decimal) || 'x' : '--' as times_change_str select - toDecimal64(left, 3), toDecimal64(right, 3), times_change_str, - toDecimal64(diff, 3), toDecimal64(stat_threshold, 3), + round(left, 3), round(right, 3), times_change_str, + round(diff, 3), round(stat_threshold, 3), changed_fail, test, query_index, query_display_name from queries where changed_show order by abs(diff) desc; create table unstable_queries_report engine File(TSV, 'report/unstable-queries.tsv') - settings output_format_decimal_trailing_zeros = 1 as select - toDecimal64(left, 3), toDecimal64(right, 3), toDecimal64(diff, 3), - toDecimal64(stat_threshold, 3), unstable_fail, test, query_index, query_display_name + round(left, 3), round(right, 3), round(diff, 3), + round(stat_threshold, 3), unstable_fail, test, query_index, query_display_name from queries where unstable_show order by stat_threshold desc; @@ -789,11 +786,10 @@ create view total_speedup as ; create table test_perf_changes_report engine File(TSV, 'report/test-perf-changes.tsv') - settings output_format_decimal_trailing_zeros = 1 as with (times_speedup >= 1 - ? '-' || toString(toDecimal64(times_speedup, 3)) || 'x' - : '+' || toString(toDecimal64(1 / times_speedup, 3)) || 'x') + ? '-' || toString(round(times_speedup, 3)) || 'x' + : '+' || toString(round(1 / times_speedup, 3)) || 'x') as times_speedup_str select test, times_speedup_str, queries, bad, changed, unstable -- Not sure what's the precedence of UNION ALL vs WHERE & ORDER BY, hence all @@ -817,11 +813,10 @@ create view total_client_time_per_query as select * 'test text, query_index int, client float, server float'); create table slow_on_client_report engine File(TSV, 'report/slow-on-client.tsv') - settings output_format_decimal_trailing_zeros = 1 - as select client, server, toDecimal64(client/server, 3) p, + as select client, server, round(client/server, 3) p, test, query_display_name from total_client_time_per_query left join query_display_names using (test, query_index) - where p > toDecimal64(1.02, 3) order by p desc; + where p > round(1.02, 3) order by p desc; create table wall_clock_time_per_test engine Memory as select * from file('wall-clock-times.tsv', TSV, 'test text, real float, user float, system float'); @@ -899,15 +894,14 @@ create view test_times_view_total as ; create table test_times_report engine File(TSV, 'report/test-times.tsv') - settings output_format_decimal_trailing_zeros = 1 as select test, - toDecimal64(real, 3), - toDecimal64(total_client_time, 3), + round(real, 3), + round(total_client_time, 3), queries, - toDecimal64(query_max, 3), - toDecimal64(avg_real_per_query, 3), - toDecimal64(query_min, 3), + round(query_max, 3), + round(avg_real_per_query, 3), + round(query_min, 3), runs from ( select * from test_times_view @@ -919,21 +913,20 @@ create table test_times_report engine File(TSV, 'report/test-times.tsv') -- report for all queries page, only main metric create table all_tests_report engine File(TSV, 'report/all-queries.tsv') - settings output_format_decimal_trailing_zeros = 1 as with -- server_time is sometimes reported as zero (if it's less than 1 ms), -- so we have to work around this to not get an error about conversion -- of NaN to decimal. (left > right ? left / right : right / left) as times_change_float, isFinite(times_change_float) as times_change_finite, - toDecimal64(times_change_finite ? times_change_float : 1., 3) as times_change_decimal, + round(times_change_finite ? times_change_float : 1., 3) as times_change_decimal, times_change_finite ? (left > right ? '-' : '+') || toString(times_change_decimal) || 'x' : '--' as times_change_str select changed_fail, unstable_fail, - toDecimal64(left, 3), toDecimal64(right, 3), times_change_str, - toDecimal64(isFinite(diff) ? diff : 0, 3), - toDecimal64(isFinite(stat_threshold) ? stat_threshold : 0, 3), + round(left, 3), round(right, 3), times_change_str, + round(isFinite(diff) ? diff : 0, 3), + round(isFinite(stat_threshold) ? stat_threshold : 0, 3), test, query_index, query_display_name from queries order by test, query_index; @@ -1044,27 +1037,6 @@ create table unstable_run_traces engine File(TSVWithNamesAndTypes, order by count() desc ; -create table metric_devation engine File(TSVWithNamesAndTypes, - 'report/metric-deviation.$version.tsv') - settings output_format_decimal_trailing_zeros = 1 - -- first goes the key used to split the file with grep - as select test, query_index, query_display_name, - toDecimal64(d, 3) d, q, metric - from ( - select - test, query_index, - (q[3] - q[1])/q[2] d, - quantilesExact(0, 0.5, 1)(value) q, metric - from (select * from unstable_run_metrics - union all select * from unstable_run_traces - union all select * from unstable_run_metrics_2) mm - group by test, query_index, metric - having isFinite(d) and d > 0.5 and q[3] > 5 - ) metrics - left join query_display_names using (test, query_index) - order by test, query_index, d desc - ; - create table stacks engine File(TSV, 'report/stacks.$version.tsv') as select -- first goes the key used to split the file with grep @@ -1173,9 +1145,8 @@ create table metrics engine File(TSV, 'metrics/metrics.tsv') as -- Show metrics that have changed create table changes engine File(TSV, 'metrics/changes.tsv') - settings output_format_decimal_trailing_zeros = 1 as select metric, left, right, - toDecimal64(diff, 3), toDecimal64(times_diff, 3) + round(diff, 3), round(times_diff, 3) from ( select metric, median(left) as left, median(right) as right, (right - left) / left diff, @@ -1226,7 +1197,6 @@ create table ci_checks engine File(TSVWithNamesAndTypes, 'ci-checks.tsv') '$SHA_TO_TEST' :: LowCardinality(String) AS commit_sha, '${CLICKHOUSE_PERFORMANCE_COMPARISON_CHECK_NAME:-Performance}' :: LowCardinality(String) AS check_name, '$(sed -n 's/.*/\1/p' report.html)' :: LowCardinality(String) AS check_status, - -- TODO toDateTime() can't parse output of 'date', so no time for now. (($(date +%s) - $CHPC_CHECK_START_TIMESTAMP) * 1000) :: UInt64 AS check_duration_ms, fromUnixTimestamp($CHPC_CHECK_START_TIMESTAMP) check_start_time, test_name :: LowCardinality(String) AS test_name , diff --git a/docs/en/operations/utilities/clickhouse-keeper-client.md b/docs/en/operations/utilities/clickhouse-keeper-client.md index 8dd5f01cd1c..2140d22b620 100644 --- a/docs/en/operations/utilities/clickhouse-keeper-client.md +++ b/docs/en/operations/utilities/clickhouse-keeper-client.md @@ -45,9 +45,14 @@ keeper foo bar - `ls [path]` -- Lists the nodes for the given path (default: cwd) - `cd [path]` -- Change the working path (default `.`) - `set [version]` -- Updates the node's value. Only update if version matches (default: -1) -- `create ` -- Creates new node +- `create [mode]` -- Creates new node with the set value +- `touch ` -- Creates new node with an empty string as value. Doesn't throw an exception if the node already exists - `get ` -- Returns the node's value - `remove ` -- Remove the node - `rmr ` -- Recursively deletes path. Confirmation required - `flwc ` -- Executes four-letter-word command - `help` -- Prints this message +- `get_stat [path]` -- Returns the node's stat (default `.`) +- `find_super_nodes [path]` -- Finds nodes with number of children larger than some threshold for the given path (default `.`) +- `delete_stale_backups` -- Deletes ClickHouse nodes used for backups that are now inactive +- `find_big_family [path] [n]` -- Returns the top n nodes with the biggest family in the subtree (default path = `.` and n = 10) diff --git a/docs/en/sql-reference/functions/string-functions.md b/docs/en/sql-reference/functions/string-functions.md index 527ce2434c0..e39bdbae06e 100644 --- a/docs/en/sql-reference/functions/string-functions.md +++ b/docs/en/sql-reference/functions/string-functions.md @@ -729,6 +729,30 @@ Returns whether string `str` ends with `suffix`. endsWith(str, suffix) ``` +## endsWithUTF8 + +Returns whether string `str` ends with `suffix`, the difference between `endsWithUTF8` and `endsWith` is that `endsWithUTF8` match `str` and `suffix` by UTF-8 characters. + +**Syntax** + +```sql +endsWithUTF8(str, suffix) +``` + +**Example** + +``` sql +SELECT endsWithUTF8('中国', '\xbd'), endsWith('中国', '\xbd') +``` + +Result: + +```result +┌─endsWithUTF8('中国', '½')─┬─endsWith('中国', '½')─┐ +│ 0 │ 1 │ +└──────────────────────────┴──────────────────────┘ +``` + ## startsWith Returns whether string `str` starts with `prefix`. @@ -745,6 +769,25 @@ startsWith(str, prefix) SELECT startsWith('Spider-Man', 'Spi'); ``` +## startsWithUTF8 + +Returns whether string `str` starts with `prefix`, the difference between `startsWithUTF8` and `startsWith` is that `startsWithUTF8` match `str` and `suffix` by UTF-8 characters. + + +**Example** + +``` sql +SELECT startsWithUTF8('中国', '\xe4'), startsWith('中国', '\xe4') +``` + +Result: + +```result +┌─startsWithUTF8('中国', '⥩─┬─startsWith('中国', '⥩─┐ +│ 0 │ 1 │ +└────────────────────────────┴────────────────────────┘ +``` + ## trim Removes the specified characters from the start or end of a string. If not specified otherwise, the function removes whitespace (ASCII-character 32). diff --git a/docs/en/sql-reference/statements/create/quota.md b/docs/en/sql-reference/statements/create/quota.md index c69285171ab..a6ced870c18 100644 --- a/docs/en/sql-reference/statements/create/quota.md +++ b/docs/en/sql-reference/statements/create/quota.md @@ -11,6 +11,7 @@ Syntax: ``` sql CREATE QUOTA [IF NOT EXISTS | OR REPLACE] name [ON CLUSTER cluster_name] + [IN access_storage_type] [KEYED BY {user_name | ip_address | client_key | client_key,user_name | client_key,ip_address} | NOT KEYED] [FOR [RANDOMIZED] INTERVAL number {second | minute | hour | day | week | month | quarter | year} {MAX { {queries | query_selects | query_inserts | errors | result_rows | result_bytes | read_rows | read_bytes | execution_time} = number } [,...] | diff --git a/docs/en/sql-reference/statements/create/role.md b/docs/en/sql-reference/statements/create/role.md index 9b14e220e1f..4b6fffe4f60 100644 --- a/docs/en/sql-reference/statements/create/role.md +++ b/docs/en/sql-reference/statements/create/role.md @@ -11,6 +11,7 @@ Syntax: ``` sql CREATE ROLE [IF NOT EXISTS | OR REPLACE] name1 [ON CLUSTER cluster_name1] [, name2 [ON CLUSTER cluster_name2] ...] + [IN access_storage_type] [SETTINGS variable [= value] [MIN [=] min_value] [MAX [=] max_value] [CONST|READONLY|WRITABLE|CHANGEABLE_IN_READONLY] | PROFILE 'profile_name'] [,...] ``` diff --git a/docs/en/sql-reference/statements/create/row-policy.md b/docs/en/sql-reference/statements/create/row-policy.md index 83bb2e6bb9a..cd7718793bd 100644 --- a/docs/en/sql-reference/statements/create/row-policy.md +++ b/docs/en/sql-reference/statements/create/row-policy.md @@ -16,6 +16,7 @@ Syntax: ``` sql CREATE [ROW] POLICY [IF NOT EXISTS | OR REPLACE] policy_name1 [ON CLUSTER cluster_name1] ON [db1.]table1|db1.* [, policy_name2 [ON CLUSTER cluster_name2] ON [db2.]table2|db2.* ...] + [IN access_storage_type] [FOR SELECT] USING condition [AS {PERMISSIVE | RESTRICTIVE}] [TO {role1 [, role2 ...] | ALL | ALL EXCEPT role1 [, role2 ...]}] diff --git a/docs/en/sql-reference/statements/create/settings-profile.md b/docs/en/sql-reference/statements/create/settings-profile.md index 8e221a4d82f..d8afce9d6de 100644 --- a/docs/en/sql-reference/statements/create/settings-profile.md +++ b/docs/en/sql-reference/statements/create/settings-profile.md @@ -12,6 +12,7 @@ Syntax: ``` sql CREATE SETTINGS PROFILE [IF NOT EXISTS | OR REPLACE] name1 [ON CLUSTER cluster_name1] [, name2 [ON CLUSTER cluster_name2] ...] + [IN access_storage_type] [SETTINGS variable [= value] [MIN [=] min_value] [MAX [=] max_value] [CONST|READONLY|WRITABLE|CHANGEABLE_IN_READONLY] | INHERIT 'profile_name'] [,...] ``` diff --git a/docs/en/sql-reference/statements/create/user.md b/docs/en/sql-reference/statements/create/user.md index d168be63c36..11d4eae7bc8 100644 --- a/docs/en/sql-reference/statements/create/user.md +++ b/docs/en/sql-reference/statements/create/user.md @@ -14,6 +14,7 @@ CREATE USER [IF NOT EXISTS | OR REPLACE] name1 [ON CLUSTER cluster_name1] [, name2 [ON CLUSTER cluster_name2] ...] [NOT IDENTIFIED | IDENTIFIED {[WITH {no_password | plaintext_password | sha256_password | sha256_hash | double_sha1_password | double_sha1_hash}] BY {'password' | 'hash'}} | {WITH ldap SERVER 'server_name'} | {WITH kerberos [REALM 'realm']} | {WITH ssl_certificate CN 'common_name'}] [HOST {LOCAL | NAME 'name' | REGEXP 'name_regexp' | IP 'address' | LIKE 'pattern'} [,...] | ANY | NONE] + [IN access_storage_type] [DEFAULT ROLE role [,...]] [DEFAULT DATABASE database | NONE] [GRANTEES {user | role | ANY | NONE} [,...] [EXCEPT {user | role} [,...]]] diff --git a/docs/en/sql-reference/statements/drop.md b/docs/en/sql-reference/statements/drop.md index c91457993c4..eed76dbcd5c 100644 --- a/docs/en/sql-reference/statements/drop.md +++ b/docs/en/sql-reference/statements/drop.md @@ -49,7 +49,7 @@ Deletes a user. Syntax: ``` sql -DROP USER [IF EXISTS] name [,...] [ON CLUSTER cluster_name] +DROP USER [IF EXISTS] name [,...] [ON CLUSTER cluster_name] [FROM access_storage_type] ``` ## DROP ROLE @@ -59,7 +59,7 @@ Deletes a role. The deleted role is revoked from all the entities where it was a Syntax: ``` sql -DROP ROLE [IF EXISTS] name [,...] [ON CLUSTER cluster_name] +DROP ROLE [IF EXISTS] name [,...] [ON CLUSTER cluster_name] [FROM access_storage_type] ``` ## DROP ROW POLICY @@ -69,7 +69,7 @@ Deletes a row policy. Deleted row policy is revoked from all the entities where Syntax: ``` sql -DROP [ROW] POLICY [IF EXISTS] name [,...] ON [database.]table [,...] [ON CLUSTER cluster_name] +DROP [ROW] POLICY [IF EXISTS] name [,...] ON [database.]table [,...] [ON CLUSTER cluster_name] [FROM access_storage_type] ``` ## DROP QUOTA @@ -79,7 +79,7 @@ Deletes a quota. The deleted quota is revoked from all the entities where it was Syntax: ``` sql -DROP QUOTA [IF EXISTS] name [,...] [ON CLUSTER cluster_name] +DROP QUOTA [IF EXISTS] name [,...] [ON CLUSTER cluster_name] [FROM access_storage_type] ``` ## DROP SETTINGS PROFILE @@ -89,7 +89,7 @@ Deletes a settings profile. The deleted settings profile is revoked from all the Syntax: ``` sql -DROP [SETTINGS] PROFILE [IF EXISTS] name [,...] [ON CLUSTER cluster_name] +DROP [SETTINGS] PROFILE [IF EXISTS] name [,...] [ON CLUSTER cluster_name] [FROM access_storage_type] ``` ## DROP VIEW diff --git a/docs/en/sql-reference/statements/move.md b/docs/en/sql-reference/statements/move.md new file mode 100644 index 00000000000..fac738ff711 --- /dev/null +++ b/docs/en/sql-reference/statements/move.md @@ -0,0 +1,32 @@ +--- +slug: /en/sql-reference/statements/move +sidebar_position: 54 +sidebar_label: MOVE +--- + +# MOVE access entity statement + +This statement allows to move an access entity from one access storage to another. + +Syntax: + +```sql +MOVE {USER, ROLE, QUOTA, SETTINGS PROFILE, ROW POLICY} name1 [, name2, ...] TO access_storage_type +``` + +Currently, there are five access storages in ClickHouse: + - `local_directory` + - `memory` + - `replicated` + - `users_xml` (ro) + - `ldap` (ro) + +Examples: + +```sql +MOVE USER test TO local_directory +``` + +```sql +MOVE ROLE test TO memory +``` diff --git a/docs/ru/getting-started/example-datasets/wikistat.md b/docs/ru/getting-started/example-datasets/wikistat.md deleted file mode 100644 index 479616d667b..00000000000 --- a/docs/ru/getting-started/example-datasets/wikistat.md +++ /dev/null @@ -1,32 +0,0 @@ ---- -slug: /ru/getting-started/example-datasets/wikistat -sidebar_position: 17 -sidebar_label: WikiStat ---- - -# WikiStat {#wikistat} - -См: http://dumps.wikimedia.org/other/pagecounts-raw/ - -Создание таблицы: - -``` sql -CREATE TABLE wikistat -( - date Date, - time DateTime, - project String, - subproject String, - path String, - hits UInt64, - size UInt64 -) ENGINE = MergeTree(date, (path, time), 8192); -``` - -Загрузка данных: - -``` bash -$ for i in {2007..2016}; do for j in {01..12}; do echo $i-$j >&2; curl -sSL "http://dumps.wikimedia.org/other/pagecounts-raw/$i/$i-$j/" | grep -oE 'pagecounts-[0-9]+-[0-9]+\.gz'; done; done | sort | uniq | tee links.txt -$ cat links.txt | while read link; do wget http://dumps.wikimedia.org/other/pagecounts-raw/$(echo $link | sed -r 's/pagecounts-([0-9]{4})([0-9]{2})[0-9]{2}-[0-9]+\.gz/\1/')/$(echo $link | sed -r 's/pagecounts-([0-9]{4})([0-9]{2})[0-9]{2}-[0-9]+\.gz/\1-\2/')/$link; done -$ ls -1 /opt/wikistat/ | grep gz | while read i; do echo $i; gzip -cd /opt/wikistat/$i | ./wikistat-loader --time="$(echo -n $i | sed -r 's/pagecounts-([0-9]{4})([0-9]{2})([0-9]{2})-([0-9]{2})([0-9]{2})([0-9]{2})\.gz/\1-\2-\3 \4-00-00/')" | clickhouse-client --query="INSERT INTO wikistat FORMAT TabSeparated"; done -``` diff --git a/docs/ru/getting-started/example-datasets/wikistat.md b/docs/ru/getting-started/example-datasets/wikistat.md new file mode 120000 index 00000000000..2d429d00984 --- /dev/null +++ b/docs/ru/getting-started/example-datasets/wikistat.md @@ -0,0 +1 @@ +../../../en/getting-started/example-datasets/wikistat.md \ No newline at end of file diff --git a/docs/zh/getting-started/example-datasets/wikistat.md b/docs/zh/getting-started/example-datasets/wikistat.md deleted file mode 100644 index 4ce13b0f1d3..00000000000 --- a/docs/zh/getting-started/example-datasets/wikistat.md +++ /dev/null @@ -1,32 +0,0 @@ ---- -slug: /zh/getting-started/example-datasets/wikistat -sidebar_position: 17 -sidebar_label: WikiStat ---- - -# WikiStat {#wikistat} - -参考: http://dumps.wikimedia.org/other/pagecounts-raw/ - -创建表结构： - -``` sql -CREATE TABLE wikistat -( - date Date, - time DateTime, - project String, - subproject String, - path String, - hits UInt64, - size UInt64 -) ENGINE = MergeTree(date, (path, time), 8192); -``` - -加载数据： - -``` bash -$ for i in {2007..2016}; do for j in {01..12}; do echo $i-$j >&2; curl -sSL "http://dumps.wikimedia.org/other/pagecounts-raw/$i/$i-$j/" | grep -oE 'pagecounts-[0-9]+-[0-9]+\.gz'; done; done | sort | uniq | tee links.txt -$ cat links.txt | while read link; do wget http://dumps.wikimedia.org/other/pagecounts-raw/$(echo $link | sed -r 's/pagecounts-([0-9]{4})([0-9]{2})[0-9]{2}-[0-9]+\.gz/\1/')/$(echo $link | sed -r 's/pagecounts-([0-9]{4})([0-9]{2})[0-9]{2}-[0-9]+\.gz/\1-\2/')/$link; done -$ ls -1 /opt/wikistat/ | grep gz | while read i; do echo $i; gzip -cd /opt/wikistat/$i | ./wikistat-loader --time="$(echo -n $i | sed -r 's/pagecounts-([0-9]{4})([0-9]{2})([0-9]{2})-([0-9]{2})([0-9]{2})([0-9]{2})\.gz/\1-\2-\3 \4-00-00/')" | clickhouse-client --query="INSERT INTO wikistat FORMAT TabSeparated"; done -``` diff --git a/docs/zh/getting-started/example-datasets/wikistat.md b/docs/zh/getting-started/example-datasets/wikistat.md new file mode 120000 index 00000000000..2d429d00984 --- /dev/null +++ b/docs/zh/getting-started/example-datasets/wikistat.md @@ -0,0 +1 @@ +../../../en/getting-started/example-datasets/wikistat.md \ No newline at end of file diff --git a/programs/disks/ICommand.cpp b/programs/disks/ICommand.cpp index 52d1a2196a9..86188fb6db1 100644 --- a/programs/disks/ICommand.cpp +++ b/programs/disks/ICommand.cpp @@ -1,4 +1,6 @@ #include "ICommand.h" +#include + namespace DB { diff --git a/programs/keeper-client/Commands.cpp b/programs/keeper-client/Commands.cpp index 3afd49e9855..c822a631798 100644 --- a/programs/keeper-client/Commands.cpp +++ b/programs/keeper-client/Commands.cpp @@ -1,5 +1,6 @@ #include "Commands.h" +#include #include "KeeperClient.h" @@ -24,8 +25,18 @@ void LSCommand::execute(const ASTKeeperQuery * query, KeeperClient * client) con else path = client->cwd; - for (const auto & child : client->zookeeper->getChildren(path)) - std::cout << child << " "; + auto children = client->zookeeper->getChildren(path); + std::sort(children.begin(), children.end()); + + bool need_space = false; + for (const auto & child : children) + { + if (std::exchange(need_space, true)) + std::cout << " "; + + std::cout << child; + } + std::cout << "\n"; } @@ -115,6 +126,21 @@ void CreateCommand::execute(const ASTKeeperQuery * query, KeeperClient * client) static_cast(query->args[2].safeGet())); } +bool TouchCommand::parse(IParser::Pos & pos, std::shared_ptr & node, Expected & expected) const +{ + String arg; + if (!parseKeeperPath(pos, expected, arg)) + return false; + node->args.push_back(std::move(arg)); + + return true; +} + +void TouchCommand::execute(const ASTKeeperQuery * query, KeeperClient * client) const +{ + client->zookeeper->createIfNotExists(client->getAbsolutePath(query->args[0].safeGet()), ""); +} + bool GetCommand::parse(IParser::Pos & pos, std::shared_ptr & node, Expected & expected) const { String arg; @@ -130,6 +156,173 @@ void GetCommand::execute(const ASTKeeperQuery * query, KeeperClient * client) co std::cout << client->zookeeper->get(client->getAbsolutePath(query->args[0].safeGet())) << "\n"; } +bool GetStatCommand::parse(IParser::Pos & pos, std::shared_ptr & node, Expected & expected) const +{ + String arg; + if (!parseKeeperPath(pos, expected, arg)) + return true; + + node->args.push_back(std::move(arg)); + return true; +} + +void GetStatCommand::execute(const ASTKeeperQuery * query, KeeperClient * client) const +{ + Coordination::Stat stat; + String path; + if (!query->args.empty()) + path = client->getAbsolutePath(query->args[0].safeGet()); + else + path = client->cwd; + + client->zookeeper->get(path, &stat); + + std::cout << "cZxid = " << stat.czxid << "\n"; + std::cout << "mZxid = " << stat.mzxid << "\n"; + std::cout << "pZxid = " << stat.pzxid << "\n"; + std::cout << "ctime = " << stat.ctime << "\n"; + std::cout << "mtime = " << stat.mtime << "\n"; + std::cout << "version = " << stat.version << "\n"; + std::cout << "cversion = " << stat.cversion << "\n"; + std::cout << "aversion = " << stat.aversion << "\n"; + std::cout << "ephemeralOwner = " << stat.ephemeralOwner << "\n"; + std::cout << "dataLength = " << stat.dataLength << "\n"; + std::cout << "numChildren = " << stat.numChildren << "\n"; +} + +bool FindSuperNodes::parse(IParser::Pos & pos, std::shared_ptr & node, Expected & expected) const +{ + ASTPtr threshold; + if (!ParserUnsignedInteger{}.parse(pos, threshold, expected)) + return false; + + node->args.push_back(threshold->as().value); + + String path; + if (!parseKeeperPath(pos, expected, path)) + path = "."; + + node->args.push_back(std::move(path)); + return true; +} + +void FindSuperNodes::execute(const ASTKeeperQuery * query, KeeperClient * client) const +{ + auto threshold = query->args[0].safeGet(); + auto path = client->getAbsolutePath(query->args[1].safeGet()); + + Coordination::Stat stat; + client->zookeeper->get(path, &stat); + + if (stat.numChildren >= static_cast(threshold)) + { + std::cout << static_cast(path) << "\t" << stat.numChildren << "\n"; + return; + } + + auto children = client->zookeeper->getChildren(path); + std::sort(children.begin(), children.end()); + for (const auto & child : children) + { + auto next_query = *query; + next_query.args[1] = DB::Field(path / child); + execute(&next_query, client); + } +} + +bool DeleteStaleBackups::parse(IParser::Pos & /* pos */, std::shared_ptr & /* node */, Expected & /* expected */) const +{ + return true; +} + +void DeleteStaleBackups::execute(const ASTKeeperQuery * /* query */, KeeperClient * client) const +{ + client->askConfirmation( + "You are going to delete all inactive backups in /clickhouse/backups.", + [client] + { + fs::path backup_root = "/clickhouse/backups"; + auto backups = client->zookeeper->getChildren(backup_root); + std::sort(backups.begin(), backups.end()); + + for (const auto & child : backups) + { + auto backup_path = backup_root / child; + std::cout << "Found backup " << backup_path << ", checking if it's active\n"; + + String stage_path = backup_path / "stage"; + auto stages = client->zookeeper->getChildren(stage_path); + + bool is_active = false; + for (const auto & stage : stages) + { + if (startsWith(stage, "alive")) + { + is_active = true; + break; + } + } + + if (is_active) + { + std::cout << "Backup " << backup_path << " is active, not going to delete\n"; + continue; + } + + std::cout << "Backup " << backup_path << " is not active, deleting it\n"; + client->zookeeper->removeRecursive(backup_path); + } + }); +} + +bool FindBigFamily::parse(IParser::Pos & pos, std::shared_ptr & node, Expected & expected) const +{ + String path; + if (!parseKeeperPath(pos, expected, path)) + path = "."; + + node->args.push_back(std::move(path)); + + ASTPtr count; + if (ParserUnsignedInteger{}.parse(pos, count, expected)) + node->args.push_back(count->as().value); + else + node->args.push_back(UInt64(10)); + + return true; +} + +void FindBigFamily::execute(const ASTKeeperQuery * query, KeeperClient * client) const +{ + auto path = client->getAbsolutePath(query->args[0].safeGet()); + auto n = query->args[1].safeGet(); + + std::vector> result; + + std::queue queue; + queue.push(path); + while (!queue.empty()) + { + auto next_path = queue.front(); + queue.pop(); + + auto children = client->zookeeper->getChildren(next_path); + std::transform(children.cbegin(), children.cend(), children.begin(), [&](const String & child) { return next_path / child; }); + + auto response = client->zookeeper->get(children); + + for (size_t i = 0; i < response.size(); ++i) + { + result.emplace_back(response[i].stat.numChildren, children[i]); + queue.push(children[i]); + } + } + + std::sort(result.begin(), result.end(), std::greater()); + for (UInt64 i = 0; i < std::min(result.size(), static_cast(n)); ++i) + std::cout << std::get<1>(result[i]) << "\t" << std::get<0>(result[i]) << "\n"; +} + bool RMCommand::parse(IParser::Pos & pos, std::shared_ptr & node, Expected & expected) const { String arg; @@ -170,7 +363,7 @@ bool HelpCommand::parse(IParser::Pos & /* pos */, std::shared_ptrgetHelpMessage() << "\n"; + std::cout << pair.second->generateHelpString() << "\n"; } bool FourLetterWordCommand::parse(IParser::Pos & pos, std::shared_ptr & node, Expected & expected) const diff --git a/programs/keeper-client/Commands.h b/programs/keeper-client/Commands.h index e4debd53e42..7d64fef1738 100644 --- a/programs/keeper-client/Commands.h +++ b/programs/keeper-client/Commands.h @@ -21,6 +21,12 @@ public: virtual String getName() const = 0; virtual ~IKeeperClientCommand() = default; + + String generateHelpString() const + { + return fmt::vformat(getHelpMessage(), fmt::make_format_args(getName())); + } + }; using Command = std::shared_ptr; @@ -34,7 +40,7 @@ class LSCommand : public IKeeperClientCommand void execute(const ASTKeeperQuery * query, KeeperClient * client) const override; - String getHelpMessage() const override { return "ls [path] -- Lists the nodes for the given path (default: cwd)"; } + String getHelpMessage() const override { return "{} [path] -- Lists the nodes for the given path (default: cwd)"; } }; class CDCommand : public IKeeperClientCommand @@ -45,7 +51,7 @@ class CDCommand : public IKeeperClientCommand void execute(const ASTKeeperQuery * query, KeeperClient * client) const override; - String getHelpMessage() const override { return "cd [path] -- Change the working path (default `.`)"; } + String getHelpMessage() const override { return "{} [path] -- Change the working path (default `.`)"; } }; class SetCommand : public IKeeperClientCommand @@ -58,7 +64,7 @@ class SetCommand : public IKeeperClientCommand String getHelpMessage() const override { - return "set [version] -- Updates the node's value. Only update if version matches (default: -1)"; + return "{} [version] -- Updates the node's value. Only update if version matches (default: -1)"; } }; @@ -70,7 +76,18 @@ class CreateCommand : public IKeeperClientCommand void execute(const ASTKeeperQuery * query, KeeperClient * client) const override; - String getHelpMessage() const override { return "create -- Creates new node"; } + String getHelpMessage() const override { return "{} [mode] -- Creates new node with the set value"; } +}; + +class TouchCommand : public IKeeperClientCommand +{ + String getName() const override { return "touch"; } + + bool parse(IParser::Pos & pos, std::shared_ptr & node, Expected & expected) const override; + + void execute(const ASTKeeperQuery * query, KeeperClient * client) const override; + + String getHelpMessage() const override { return "{} -- Creates new node with an empty string as value. Doesn't throw an exception if the node already exists"; } }; class GetCommand : public IKeeperClientCommand @@ -81,9 +98,63 @@ class GetCommand : public IKeeperClientCommand void execute(const ASTKeeperQuery * query, KeeperClient * client) const override; - String getHelpMessage() const override { return "get -- Returns the node's value"; } + String getHelpMessage() const override { return "{} -- Returns the node's value"; } }; +class GetStatCommand : public IKeeperClientCommand +{ + String getName() const override { return "get_stat"; } + + bool parse(IParser::Pos & pos, std::shared_ptr & node, Expected & expected) const override; + + void execute(const ASTKeeperQuery * query, KeeperClient * client) const override; + + String getHelpMessage() const override { return "{} [path] -- Returns the node's stat (default `.`)"; } +}; + +class FindSuperNodes : public IKeeperClientCommand +{ + String getName() const override { return "find_super_nodes"; } + + bool parse(IParser::Pos & pos, std::shared_ptr & node, Expected & expected) const override; + + void execute(const ASTKeeperQuery * query, KeeperClient * client) const override; + + String getHelpMessage() const override + { + return "{} [path] -- Finds nodes with number of children larger than some threshold for the given path (default `.`)"; + } +}; + +class DeleteStaleBackups : public IKeeperClientCommand +{ + String getName() const override { return "delete_stale_backups"; } + + bool parse(IParser::Pos & pos, std::shared_ptr & node, Expected & expected) const override; + + void execute(const ASTKeeperQuery * query, KeeperClient * client) const override; + + String getHelpMessage() const override + { + return "{} -- Deletes ClickHouse nodes used for backups that are now inactive"; + } +}; + +class FindBigFamily : public IKeeperClientCommand +{ + String getName() const override { return "find_big_family"; } + + bool parse(IParser::Pos & pos, std::shared_ptr & node, Expected & expected) const override; + + void execute(const ASTKeeperQuery * query, KeeperClient * client) const override; + + String getHelpMessage() const override + { + return "{} [path] [n] -- Returns the top n nodes with the biggest family in the subtree (default path = `.` and n = 10)"; + } +}; + + class RMCommand : public IKeeperClientCommand { String getName() const override { return "rm"; } @@ -92,7 +163,7 @@ class RMCommand : public IKeeperClientCommand void execute(const ASTKeeperQuery * query, KeeperClient * client) const override; - String getHelpMessage() const override { return "remove -- Remove the node"; } + String getHelpMessage() const override { return "{} -- Remove the node"; } }; class RMRCommand : public IKeeperClientCommand @@ -103,7 +174,7 @@ class RMRCommand : public IKeeperClientCommand void execute(const ASTKeeperQuery * query, KeeperClient * client) const override; - String getHelpMessage() const override { return "rmr -- Recursively deletes path. Confirmation required"; } + String getHelpMessage() const override { return "{} -- Recursively deletes path. Confirmation required"; } }; class HelpCommand : public IKeeperClientCommand @@ -114,7 +185,7 @@ class HelpCommand : public IKeeperClientCommand void execute(const ASTKeeperQuery * query, KeeperClient * client) const override; - String getHelpMessage() const override { return "help -- Prints this message"; } + String getHelpMessage() const override { return "{} -- Prints this message"; } }; class FourLetterWordCommand : public IKeeperClientCommand @@ -125,7 +196,7 @@ class FourLetterWordCommand : public IKeeperClientCommand void execute(const ASTKeeperQuery * query, KeeperClient * client) const override; - String getHelpMessage() const override { return "flwc -- Executes four-letter-word command"; } + String getHelpMessage() const override { return "{} -- Executes four-letter-word command"; } }; } diff --git a/programs/keeper-client/KeeperClient.cpp b/programs/keeper-client/KeeperClient.cpp index 4123a86515c..893be4d25d1 100644 --- a/programs/keeper-client/KeeperClient.cpp +++ b/programs/keeper-client/KeeperClient.cpp @@ -176,7 +176,12 @@ void KeeperClient::initialize(Poco::Util::Application & /* self */) std::make_shared(), std::make_shared(), std::make_shared(), + std::make_shared(), std::make_shared(), + std::make_shared(), + std::make_shared(), + std::make_shared(), + std::make_shared(), std::make_shared(), std::make_shared(), std::make_shared(), diff --git a/src/Access/AccessControl.cpp b/src/Access/AccessControl.cpp index 05cba7f8510..b8bc3c337e3 100644 --- a/src/Access/AccessControl.cpp +++ b/src/Access/AccessControl.cpp @@ -418,7 +418,7 @@ void AccessControl::addStoragesFromUserDirectoriesConfig( String type = key_in_user_directories; if (size_t bracket_pos = type.find('['); bracket_pos != String::npos) type.resize(bracket_pos); - if ((type == "users_xml") || (type == "users_config")) + if ((type == "users.xml") || (type == "users_config")) type = UsersConfigAccessStorage::STORAGE_TYPE; else if ((type == "local") || (type == "local_directory")) type = DiskAccessStorage::STORAGE_TYPE; @@ -528,12 +528,14 @@ scope_guard AccessControl::subscribeForChanges(const std::vector & ids, co return changes_notifier->subscribeForChanges(ids, handler); } -std::optional AccessControl::insertImpl(const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists) +bool AccessControl::insertImpl(const UUID & id, const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists) { - auto id = MultipleAccessStorage::insertImpl(entity, replace_if_exists, throw_if_exists); - if (id) + if (MultipleAccessStorage::insertImpl(id, entity, replace_if_exists, throw_if_exists)) + { changes_notifier->sendNotifications(); - return id; + return true; + } + return false; } bool AccessControl::removeImpl(const UUID & id, bool throw_if_not_exists) diff --git a/src/Access/AccessControl.h b/src/Access/AccessControl.h index c7b94955a47..74bcc119184 100644 --- a/src/Access/AccessControl.h +++ b/src/Access/AccessControl.h @@ -232,7 +232,7 @@ private: class CustomSettingsPrefixes; class PasswordComplexityRules; - std::optional insertImpl(const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists) override; + bool insertImpl(const UUID & id, const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists) override; bool removeImpl(const UUID & id, bool throw_if_not_exists) override; bool updateImpl(const UUID & id, const UpdateFunc & update_func, bool throw_if_not_exists) override; diff --git a/src/Access/DiskAccessStorage.cpp b/src/Access/DiskAccessStorage.cpp index 938881fafff..190c7567b85 100644 --- a/src/Access/DiskAccessStorage.cpp +++ b/src/Access/DiskAccessStorage.cpp @@ -498,20 +498,10 @@ std::optional> DiskAccessStorage::readNameWi } -std::optional DiskAccessStorage::insertImpl(const AccessEntityPtr & new_entity, bool replace_if_exists, bool throw_if_exists) -{ - UUID id = generateRandomID(); - if (insertWithID(id, new_entity, replace_if_exists, throw_if_exists, /* write_on_disk= */ true)) - return id; - - return std::nullopt; -} - - -bool DiskAccessStorage::insertWithID(const UUID & id, const AccessEntityPtr & new_entity, bool replace_if_exists, bool throw_if_exists, bool write_on_disk) +bool DiskAccessStorage::insertImpl(const UUID & id, const AccessEntityPtr & new_entity, bool replace_if_exists, bool throw_if_exists) { std::lock_guard lock{mutex}; - return insertNoLock(id, new_entity, replace_if_exists, throw_if_exists, write_on_disk); + return insertNoLock(id, new_entity, replace_if_exists, throw_if_exists, /* write_on_disk = */ true); } @@ -745,7 +735,7 @@ void DiskAccessStorage::restoreFromBackup(RestorerFromBackup & restorer) restorer.addDataRestoreTask([this, my_entities = std::move(entities), replace_if_exists, throw_if_exists] { for (const auto & [id, entity] : my_entities) - insertWithID(id, entity, replace_if_exists, throw_if_exists, /* write_on_disk= */ true); + insert(id, entity, replace_if_exists, throw_if_exists); }); } diff --git a/src/Access/DiskAccessStorage.h b/src/Access/DiskAccessStorage.h index 069a966c8e9..5d94008b34f 100644 --- a/src/Access/DiskAccessStorage.h +++ b/src/Access/DiskAccessStorage.h @@ -13,7 +13,7 @@ class AccessChangesNotifier; class DiskAccessStorage : public IAccessStorage { public: - static constexpr char STORAGE_TYPE[] = "local directory"; + static constexpr char STORAGE_TYPE[] = "local_directory"; DiskAccessStorage(const String & storage_name_, const String & directory_path_, AccessChangesNotifier & changes_notifier_, bool readonly_, bool allow_backup_); ~DiskAccessStorage() override; @@ -39,7 +39,7 @@ private: std::vector findAllImpl(AccessEntityType type) const override; AccessEntityPtr readImpl(const UUID & id, bool throw_if_not_exists) const override; std::optional> readNameWithTypeImpl(const UUID & id, bool throw_if_not_exists) const override; - std::optional insertImpl(const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists) override; + bool insertImpl(const UUID & id, const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists) override; bool removeImpl(const UUID & id, bool throw_if_not_exists) override; bool updateImpl(const UUID & id, const UpdateFunc & update_func, bool throw_if_not_exists) override; @@ -53,7 +53,6 @@ private: void listsWritingThreadFunc() TSA_NO_THREAD_SAFETY_ANALYSIS; void stopListsWritingThread(); - bool insertWithID(const UUID & id, const AccessEntityPtr & new_entity, bool replace_if_exists, bool throw_if_exists, bool write_on_disk); bool insertNoLock(const UUID & id, const AccessEntityPtr & new_entity, bool replace_if_exists, bool throw_if_exists, bool write_on_disk) TSA_REQUIRES(mutex); bool updateNoLock(const UUID & id, const UpdateFunc & update_func, bool throw_if_not_exists, bool write_on_disk) TSA_REQUIRES(mutex); bool removeNoLock(const UUID & id, bool throw_if_not_exists, bool write_on_disk) TSA_REQUIRES(mutex); diff --git a/src/Access/IAccessStorage.cpp b/src/Access/IAccessStorage.cpp index 9468e8d220a..ec25fd505f0 100644 --- a/src/Access/IAccessStorage.cpp +++ b/src/Access/IAccessStorage.cpp @@ -93,6 +93,17 @@ String IAccessStorage::readName(const UUID & id) const } +bool IAccessStorage::exists(const std::vector & ids) const +{ + for (const auto & id : ids) + { + if (!exists(id)) + return false; + } + + return true; +} + std::optional IAccessStorage::readName(const UUID & id, bool throw_if_not_exists) const { if (auto name_and_type = readNameWithType(id, throw_if_not_exists)) @@ -167,38 +178,69 @@ UUID IAccessStorage::insert(const AccessEntityPtr & entity) return *insert(entity, /* replace_if_exists = */ false, /* throw_if_exists = */ true); } - std::optional IAccessStorage::insert(const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists) { - return insertImpl(entity, replace_if_exists, throw_if_exists); + auto id = generateRandomID(); + + if (insert(id, entity, replace_if_exists, throw_if_exists)) + return id; + + return std::nullopt; +} + + +bool IAccessStorage::insert(const DB::UUID & id, const DB::AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists) +{ + return insertImpl(id, entity, replace_if_exists, throw_if_exists); } std::vector IAccessStorage::insert(const std::vector & multiple_entities, bool replace_if_exists, bool throw_if_exists) { + return insert(multiple_entities, /* ids = */ {}, replace_if_exists, throw_if_exists); +} + +std::vector IAccessStorage::insert(const std::vector & multiple_entities, const std::vector & ids, bool replace_if_exists, bool throw_if_exists) +{ + assert(ids.empty() || (multiple_entities.size() == ids.size())); + if (multiple_entities.empty()) return {}; if (multiple_entities.size() == 1) { - if (auto id = insert(multiple_entities[0], replace_if_exists, throw_if_exists)) - return {*id}; + UUID id; + if (!ids.empty()) + id = ids[0]; + else + id = generateRandomID(); + + if (insert(id, multiple_entities[0], replace_if_exists, throw_if_exists)) + return {id}; return {}; } std::vector successfully_inserted; try { - std::vector ids; - for (const auto & entity : multiple_entities) + std::vector new_ids; + for (size_t i = 0; i < multiple_entities.size(); ++i) { - if (auto id = insertImpl(entity, replace_if_exists, throw_if_exists)) + const auto & entity = multiple_entities[i]; + + UUID id; + if (!ids.empty()) + id = ids[i]; + else + id = generateRandomID(); + + if (insert(id, entity, replace_if_exists, throw_if_exists)) { successfully_inserted.push_back(entity); - ids.push_back(*id); + new_ids.push_back(id); } } - return ids; + return new_ids; } catch (Exception & e) { @@ -244,7 +286,7 @@ std::vector IAccessStorage::insertOrReplace(const std::vector IAccessStorage::insertImpl(const AccessEntityPtr & entity, bool, bool) +bool IAccessStorage::insertImpl(const UUID &, const AccessEntityPtr & entity, bool, bool) { if (isReadOnly()) throwReadonlyCannotInsert(entity->getType(), entity->getName()); diff --git a/src/Access/IAccessStorage.h b/src/Access/IAccessStorage.h index aa3947201e7..b82537df639 100644 --- a/src/Access/IAccessStorage.h +++ b/src/Access/IAccessStorage.h @@ -3,6 +3,8 @@ #include #include #include +#include +#include #include #include #include @@ -92,6 +94,7 @@ public: /// Returns whether there is an entity with such identifier in the storage. virtual bool exists(const UUID & id) const = 0; + bool exists(const std::vector & ids) const; /// Reads an entity. Throws an exception if not found. template @@ -100,6 +103,9 @@ public: template std::shared_ptr read(const String & name, bool throw_if_not_exists = true) const; + template + std::vector read(const std::vector & ids, bool throw_if_not_exists = true) const; + /// Reads an entity. Returns nullptr if not found. template std::shared_ptr tryRead(const UUID & id) const; @@ -128,7 +134,9 @@ public: /// Throws an exception if the specified name already exists. UUID insert(const AccessEntityPtr & entity); std::optional insert(const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists); + bool insert(const UUID & id, const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists); std::vector insert(const std::vector & multiple_entities, bool replace_if_exists = false, bool throw_if_exists = true); + std::vector insert(const std::vector & multiple_entities, const std::vector & ids, bool replace_if_exists = false, bool throw_if_exists = true); /// Inserts an entity to the storage. Returns ID of a new entry in the storage. std::optional tryInsert(const AccessEntityPtr & entity); @@ -179,7 +187,7 @@ protected: virtual std::vector findAllImpl(AccessEntityType type) const = 0; virtual AccessEntityPtr readImpl(const UUID & id, bool throw_if_not_exists) const = 0; virtual std::optional> readNameWithTypeImpl(const UUID & id, bool throw_if_not_exists) const; - virtual std::optional insertImpl(const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists); + virtual bool insertImpl(const UUID & id, const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists); virtual bool removeImpl(const UUID & id, bool throw_if_not_exists); virtual bool updateImpl(const UUID & id, const UpdateFunc & update_func, bool throw_if_not_exists); virtual std::optional authenticateImpl(const Credentials & credentials, const Poco::Net::IPAddress & address, const ExternalAuthenticators & external_authenticators, bool throw_if_user_not_exists, bool allow_no_password, bool allow_plaintext_password) const; @@ -240,6 +248,19 @@ std::shared_ptr IAccessStorage::read(const String & name, bo } +template +std::vector IAccessStorage::read(const std::vector & ids, bool throw_if_not_exists) const +{ + std::vector result; + result.reserve(ids.size()); + + for (const auto & id : ids) + result.push_back(read(id, throw_if_not_exists)); + + return result; +} + + template std::shared_ptr IAccessStorage::tryRead(const UUID & id) const { @@ -265,4 +286,9 @@ std::vector>> IAccessStorage return entities; } +inline bool parseAccessStorageName(IParser::Pos & pos, Expected & expected, String & storage_name) +{ + return parseIdentifierOrStringLiteral(pos, expected, storage_name); +} + } diff --git a/src/Access/MemoryAccessStorage.cpp b/src/Access/MemoryAccessStorage.cpp index c4192ee4552..999e2c41d0b 100644 --- a/src/Access/MemoryAccessStorage.cpp +++ b/src/Access/MemoryAccessStorage.cpp @@ -63,17 +63,7 @@ AccessEntityPtr MemoryAccessStorage::readImpl(const UUID & id, bool throw_if_not } -std::optional MemoryAccessStorage::insertImpl(const AccessEntityPtr & new_entity, bool replace_if_exists, bool throw_if_exists) -{ - UUID id = generateRandomID(); - if (insertWithID(id, new_entity, replace_if_exists, throw_if_exists)) - return id; - - return std::nullopt; -} - - -bool MemoryAccessStorage::insertWithID(const UUID & id, const AccessEntityPtr & new_entity, bool replace_if_exists, bool throw_if_exists) +bool MemoryAccessStorage::insertImpl(const UUID & id, const AccessEntityPtr & new_entity, bool replace_if_exists, bool throw_if_exists) { std::lock_guard lock{mutex}; return insertNoLock(id, new_entity, replace_if_exists, throw_if_exists); @@ -300,7 +290,7 @@ void MemoryAccessStorage::restoreFromBackup(RestorerFromBackup & restorer) restorer.addDataRestoreTask([this, my_entities = std::move(entities), replace_if_exists, throw_if_exists] { for (const auto & [id, entity] : my_entities) - insertWithID(id, entity, replace_if_exists, throw_if_exists); + insert(id, entity, replace_if_exists, throw_if_exists); }); } diff --git a/src/Access/MemoryAccessStorage.h b/src/Access/MemoryAccessStorage.h index b63132147da..9d3fdb0a388 100644 --- a/src/Access/MemoryAccessStorage.h +++ b/src/Access/MemoryAccessStorage.h @@ -6,6 +6,7 @@ #include #include #include +#include namespace DB @@ -22,11 +23,6 @@ public: const char * getStorageType() const override { return STORAGE_TYPE; } - /// Inserts an entity with a specified ID. - /// If `replace_if_exists == true` it can replace an existing entry with such ID and also remove an existing entry - /// with such name & type. - bool insertWithID(const UUID & id, const AccessEntityPtr & new_entity, bool replace_if_exists, bool throw_if_exists); - /// Removes all entities except the specified list `ids_to_keep`. /// The function skips IDs not contained in the storage. void removeAllExcept(const std::vector & ids_to_keep); @@ -44,7 +40,7 @@ private: std::optional findImpl(AccessEntityType type, const String & name) const override; std::vector findAllImpl(AccessEntityType type) const override; AccessEntityPtr readImpl(const UUID & id, bool throw_if_not_exists) const override; - std::optional insertImpl(const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists) override; + bool insertImpl(const UUID & id, const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists) override; bool removeImpl(const UUID & id, bool throw_if_not_exists) override; bool updateImpl(const UUID & id, const UpdateFunc & update_func, bool throw_if_not_exists) override; diff --git a/src/Access/MultipleAccessStorage.cpp b/src/Access/MultipleAccessStorage.cpp index f36a670ddbf..0550c140c17 100644 --- a/src/Access/MultipleAccessStorage.cpp +++ b/src/Access/MultipleAccessStorage.cpp @@ -16,6 +16,7 @@ namespace ErrorCodes { extern const int ACCESS_ENTITY_ALREADY_EXISTS; extern const int ACCESS_STORAGE_FOR_INSERTION_NOT_FOUND; + extern const int ACCESS_ENTITY_NOT_FOUND; } using Storage = IAccessStorage; @@ -178,6 +179,91 @@ ConstStoragePtr MultipleAccessStorage::getStorage(const UUID & id) const return const_cast(this)->getStorage(id); } +StoragePtr MultipleAccessStorage::findStorageByName(const DB::String & storage_name) +{ + auto storages = getStoragesInternal(); + for (const auto & storage : *storages) + { + if (storage->getStorageName() == storage_name) + return storage; + } + + return nullptr; +} + + +ConstStoragePtr MultipleAccessStorage::findStorageByName(const DB::String & storage_name) const +{ + return const_cast(this)->findStorageByName(storage_name); +} + + +StoragePtr MultipleAccessStorage::getStorageByName(const DB::String & storage_name) +{ + auto storage = findStorageByName(storage_name); + if (storage) + return storage; + + throw Exception(ErrorCodes::ACCESS_ENTITY_NOT_FOUND, "Access storage with name {} is not found", storage_name); +} + + +ConstStoragePtr MultipleAccessStorage::getStorageByName(const DB::String & storage_name) const +{ + return const_cast(this)->getStorageByName(storage_name); +} + +StoragePtr MultipleAccessStorage::findExcludingStorage(AccessEntityType type, const DB::String & name, DB::MultipleAccessStorage::StoragePtr exclude) const +{ + auto storages = getStoragesInternal(); + for (const auto & storage : *storages) + { + if (storage == exclude) + continue; + + if (storage->find(type, name)) + return storage; + } + + return nullptr; +} + +void MultipleAccessStorage::moveAccessEntities(const std::vector & ids, const String & source_storage_name, const String & destination_storage_name) +{ + auto source_storage = getStorageByName(source_storage_name); + auto destination_storage = getStorageByName(destination_storage_name); + + auto to_move = source_storage->read(ids); + bool need_rollback = false; + + try + { + source_storage->remove(ids); + need_rollback = true; + destination_storage->insert(to_move, ids); + } + catch (Exception & e) + { + String message; + + bool need_comma = false; + for (const auto & entity : to_move) + { + if (std::exchange(need_comma, true)) + message += ", "; + + message += entity->formatTypeWithName(); + } + + e.addMessage("while moving {} from {} to {}", message, source_storage_name, destination_storage_name); + + if (need_rollback) + source_storage->insert(to_move, ids); + + throw; + } +} + AccessEntityPtr MultipleAccessStorage::readImpl(const UUID & id, bool throw_if_not_exists) const { if (auto storage = findStorage(id)) @@ -245,7 +331,7 @@ void MultipleAccessStorage::reload(ReloadMode reload_mode) } -std::optional MultipleAccessStorage::insertImpl(const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists) +bool MultipleAccessStorage::insertImpl(const UUID & id, const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists) { std::shared_ptr storage_for_insertion; @@ -268,13 +354,14 @@ std::optional MultipleAccessStorage::insertImpl(const AccessEntityPtr & en getStorageName()); } - auto id = storage_for_insertion->insert(entity, replace_if_exists, throw_if_exists); - if (id) + if (storage_for_insertion->insert(id, entity, replace_if_exists, throw_if_exists)) { std::lock_guard lock{mutex}; - ids_cache.set(*id, storage_for_insertion); + ids_cache.set(id, storage_for_insertion); + return true; } - return id; + + return false; } diff --git a/src/Access/MultipleAccessStorage.h b/src/Access/MultipleAccessStorage.h index 6a0c1bdfc02..069d414f601 100644 --- a/src/Access/MultipleAccessStorage.h +++ b/src/Access/MultipleAccessStorage.h @@ -41,6 +41,16 @@ public: ConstStoragePtr getStorage(const UUID & id) const; StoragePtr getStorage(const UUID & id); + ConstStoragePtr findStorageByName(const String & storage_name) const; + StoragePtr findStorageByName(const String & storage_name); + ConstStoragePtr getStorageByName(const String & storage_name) const; + StoragePtr getStorageByName(const String & storage_name); + + /// Search for an access entity storage, excluding one. Returns nullptr if not found. + StoragePtr findExcludingStorage(AccessEntityType type, const String & name, StoragePtr exclude) const; + + void moveAccessEntities(const std::vector & ids, const String & source_storage_name, const String & destination_storage_name); + bool exists(const UUID & id) const override; bool isBackupAllowed() const override; @@ -53,7 +63,7 @@ protected: std::vector findAllImpl(AccessEntityType type) const override; AccessEntityPtr readImpl(const UUID & id, bool throw_if_not_exists) const override; std::optional> readNameWithTypeImpl(const UUID & id, bool throw_if_not_exists) const override; - std::optional insertImpl(const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists) override; + bool insertImpl(const UUID & id, const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists) override; bool removeImpl(const UUID & id, bool throw_if_not_exists) override; bool updateImpl(const UUID & id, const UpdateFunc & update_func, bool throw_if_not_exists) override; std::optional authenticateImpl(const Credentials & credentials, const Poco::Net::IPAddress & address, const ExternalAuthenticators & external_authenticators, bool throw_if_user_not_exists, bool allow_no_password, bool allow_plaintext_password) const override; @@ -65,6 +75,8 @@ private: std::shared_ptr nested_storages TSA_GUARDED_BY(mutex); mutable CacheBase ids_cache TSA_GUARDED_BY(mutex); mutable std::mutex mutex; + + mutable std::mutex move_mutex; }; } diff --git a/src/Access/ReplicatedAccessStorage.cpp b/src/Access/ReplicatedAccessStorage.cpp index 56d68be9268..275672da115 100644 --- a/src/Access/ReplicatedAccessStorage.cpp +++ b/src/Access/ReplicatedAccessStorage.cpp @@ -108,17 +108,7 @@ static void retryOnZooKeeperUserError(size_t attempts, Func && function) } } -std::optional ReplicatedAccessStorage::insertImpl(const AccessEntityPtr & new_entity, bool replace_if_exists, bool throw_if_exists) -{ - const UUID id = generateRandomID(); - if (insertWithID(id, new_entity, replace_if_exists, throw_if_exists)) - return id; - - return std::nullopt; -} - - -bool ReplicatedAccessStorage::insertWithID(const UUID & id, const AccessEntityPtr & new_entity, bool replace_if_exists, bool throw_if_exists) +bool ReplicatedAccessStorage::insertImpl(const UUID & id, const AccessEntityPtr & new_entity, bool replace_if_exists, bool throw_if_exists) { const AccessEntityTypeInfo type_info = AccessEntityTypeInfo::get(new_entity->getType()); const String & name = new_entity->getName(); @@ -619,7 +609,7 @@ AccessEntityPtr ReplicatedAccessStorage::tryReadEntityFromZooKeeper(const zkutil void ReplicatedAccessStorage::setEntityNoLock(const UUID & id, const AccessEntityPtr & entity) { LOG_DEBUG(getLogger(), "Setting id {} to entity named {}", toString(id), entity->getName()); - memory_storage.insertWithID(id, entity, /* replace_if_exists= */ true, /* throw_if_exists= */ false); + memory_storage.insert(id, entity, /* replace_if_exists= */ true, /* throw_if_exists= */ false); } @@ -711,7 +701,7 @@ void ReplicatedAccessStorage::restoreFromBackup(RestorerFromBackup & restorer) restorer.addDataRestoreTask([this, my_entities = std::move(entities), replace_if_exists, throw_if_exists] { for (const auto & [id, entity] : my_entities) - insertWithID(id, entity, replace_if_exists, throw_if_exists); + insert(id, entity, replace_if_exists, throw_if_exists); }); } diff --git a/src/Access/ReplicatedAccessStorage.h b/src/Access/ReplicatedAccessStorage.h index 555d58e6b04..cddb20860f7 100644 --- a/src/Access/ReplicatedAccessStorage.h +++ b/src/Access/ReplicatedAccessStorage.h @@ -46,11 +46,10 @@ private: std::unique_ptr watching_thread; std::shared_ptr> watched_queue; - std::optional insertImpl(const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists) override; + bool insertImpl(const UUID & id, const AccessEntityPtr & new_entity, bool replace_if_exists, bool throw_if_exists) override; bool removeImpl(const UUID & id, bool throw_if_not_exists) override; bool updateImpl(const UUID & id, const UpdateFunc & update_func, bool throw_if_not_exists) override; - bool insertWithID(const UUID & id, const AccessEntityPtr & new_entity, bool replace_if_exists, bool throw_if_exists); bool insertZooKeeper(const zkutil::ZooKeeperPtr & zookeeper, const UUID & id, const AccessEntityPtr & entity, bool replace_if_exists, bool throw_if_exists); bool removeZooKeeper(const zkutil::ZooKeeperPtr & zookeeper, const UUID & id, bool throw_if_not_exists); bool updateZooKeeper(const zkutil::ZooKeeperPtr & zookeeper, const UUID & id, const UpdateFunc & update_func, bool throw_if_not_exists); diff --git a/src/Access/UsersConfigAccessStorage.h b/src/Access/UsersConfigAccessStorage.h index b533ccbf200..d8ddf07ff40 100644 --- a/src/Access/UsersConfigAccessStorage.h +++ b/src/Access/UsersConfigAccessStorage.h @@ -20,7 +20,7 @@ class UsersConfigAccessStorage : public IAccessStorage { public: - static constexpr char STORAGE_TYPE[] = "users.xml"; + static constexpr char STORAGE_TYPE[] = "users_xml"; UsersConfigAccessStorage(const String & storage_name_, AccessControl & access_control_, bool allow_backup_); ~UsersConfigAccessStorage() override; diff --git a/src/AggregateFunctions/tests/gtest_ranks.cpp b/src/AggregateFunctions/tests/gtest_ranks.cpp index b29271cbec7..99c7fbd26d7 100644 --- a/src/AggregateFunctions/tests/gtest_ranks.cpp +++ b/src/AggregateFunctions/tests/gtest_ranks.cpp @@ -2,7 +2,6 @@ #include #include #include -#include #include diff --git a/src/Analyzer/Passes/FuseFunctionsPass.cpp b/src/Analyzer/Passes/FuseFunctionsPass.cpp index 2cb7afa4ad6..7b27bd70ce3 100644 --- a/src/Analyzer/Passes/FuseFunctionsPass.cpp +++ b/src/Analyzer/Passes/FuseFunctionsPass.cpp @@ -14,6 +14,9 @@ #include #include +#include + + namespace DB { diff --git a/src/Analyzer/examples/query_analyzer.cpp b/src/Analyzer/examples/query_analyzer.cpp index 5a20b46b346..869113b10f8 100644 --- a/src/Analyzer/examples/query_analyzer.cpp +++ b/src/Analyzer/examples/query_analyzer.cpp @@ -1,4 +1,3 @@ -#include int main(int argc, char ** argv) { diff --git a/src/Client/ConnectionParameters.cpp b/src/Client/ConnectionParameters.cpp index 8c7a4e830fe..1f90e757b73 100644 --- a/src/Client/ConnectionParameters.cpp +++ b/src/Client/ConnectionParameters.cpp @@ -1,6 +1,5 @@ #include "ConnectionParameters.h" #include -#include #include #include #include diff --git a/src/Client/ConnectionString.cpp b/src/Client/ConnectionString.cpp index 8150ae98c8b..0893ef1af00 100644 --- a/src/Client/ConnectionString.cpp +++ b/src/Client/ConnectionString.cpp @@ -6,7 +6,6 @@ #include #include -#include #include #include diff --git a/src/Columns/ColumnObject.cpp b/src/Columns/ColumnObject.cpp index 07872774559..2fb6d2c3028 100644 --- a/src/Columns/ColumnObject.cpp +++ b/src/Columns/ColumnObject.cpp @@ -2,17 +2,17 @@ #include #include #include -#include #include #include #include #include #include -#include #include #include #include #include +#include + namespace DB { diff --git a/src/Columns/tests/gtest_column_sparse.cpp b/src/Columns/tests/gtest_column_sparse.cpp index 6062ea51941..c3450ff91b4 100644 --- a/src/Columns/tests/gtest_column_sparse.cpp +++ b/src/Columns/tests/gtest_column_sparse.cpp @@ -10,6 +10,7 @@ #include + using namespace DB; static pcg64 rng(randomSeed()); diff --git a/src/Common/Config/configReadClient.cpp b/src/Common/Config/configReadClient.cpp index 44d338c07af..9f137124be0 100644 --- a/src/Common/Config/configReadClient.cpp +++ b/src/Common/Config/configReadClient.cpp @@ -3,7 +3,6 @@ #include #include "ConfigProcessor.h" #include -#include #include namespace fs = std::filesystem; diff --git a/src/Common/CounterInFile.h b/src/Common/CounterInFile.h index d52f4a10050..fe3b74173f6 100644 --- a/src/Common/CounterInFile.h +++ b/src/Common/CounterInFile.h @@ -4,7 +4,6 @@ #include #include -#include #include #include diff --git a/src/Common/DateLUTImpl.cpp b/src/Common/DateLUTImpl.cpp index d5e04238ef9..4c21d9c9783 100644 --- a/src/Common/DateLUTImpl.cpp +++ b/src/Common/DateLUTImpl.cpp @@ -10,7 +10,6 @@ #include #include #include -#include /// Embedded timezones. diff --git a/src/Common/EventNotifier.h b/src/Common/EventNotifier.h index edf7622512d..1d34f7ad9b1 100644 --- a/src/Common/EventNotifier.h +++ b/src/Common/EventNotifier.h @@ -7,11 +7,11 @@ #include #include #include -#include #include #include + namespace DB { diff --git a/src/Common/FST.cpp b/src/Common/FST.cpp index 30e10610eab..af2acc859e7 100644 --- a/src/Common/FST.cpp +++ b/src/Common/FST.cpp @@ -1,7 +1,6 @@ #include "FST.h" #include #include -#include #include #include #include diff --git a/src/Common/SpaceSaving.h b/src/Common/SpaceSaving.h index 3452402ba82..5d3e8fa321b 100644 --- a/src/Common/SpaceSaving.h +++ b/src/Common/SpaceSaving.h @@ -1,6 +1,5 @@ #pragma once -#include #include #include @@ -19,6 +18,7 @@ #include #include + /* * Implementation of the Filtered Space-Saving for TopK streaming analysis. * http://www.l2f.inesc-id.pt/~fmmb/wiki/uploads/Work/misnis.ref0a.pdf diff --git a/src/Common/StudentTTest.cpp b/src/Common/StudentTTest.cpp index 59842488fd0..03159dca2ca 100644 --- a/src/Common/StudentTTest.cpp +++ b/src/Common/StudentTTest.cpp @@ -1,7 +1,6 @@ #include "StudentTTest.h" #include -#include #include #include #include diff --git a/src/Common/ThreadPool.cpp b/src/Common/ThreadPool.cpp index cc88594d84f..4a5bdeffcee 100644 --- a/src/Common/ThreadPool.cpp +++ b/src/Common/ThreadPool.cpp @@ -6,7 +6,6 @@ #include #include -#include #include #include diff --git a/src/Common/UnicodeBar.cpp b/src/Common/UnicodeBar.cpp index bad39d8080c..253d720e8a1 100644 --- a/src/Common/UnicodeBar.cpp +++ b/src/Common/UnicodeBar.cpp @@ -7,7 +7,6 @@ #include #include -#include namespace DB { diff --git a/src/Common/VersionNumber.h b/src/Common/VersionNumber.h index 94bf234c1da..050c94b9574 100644 --- a/src/Common/VersionNumber.h +++ b/src/Common/VersionNumber.h @@ -2,9 +2,9 @@ #include #include -#include #include + namespace DB { @@ -27,11 +27,6 @@ struct VersionNumber std::string toString() const; - friend std::ostream & operator<<(std::ostream & os, const VersionNumber & v) - { - return os << v.toString(); - } - private: using Components = std::vector; Components components; diff --git a/src/Common/benchmarks/integer_hash_tables_and_hashes.cpp b/src/Common/benchmarks/integer_hash_tables_and_hashes.cpp index c245fc471cc..2a4e366f1c5 100644 --- a/src/Common/benchmarks/integer_hash_tables_and_hashes.cpp +++ b/src/Common/benchmarks/integer_hash_tables_and_hashes.cpp @@ -1,7 +1,6 @@ #include #include -#include #include #include diff --git a/src/Common/examples/average.cpp b/src/Common/examples/average.cpp index f281abdced2..cd4f9321cf0 100644 --- a/src/Common/examples/average.cpp +++ b/src/Common/examples/average.cpp @@ -1,4 +1,3 @@ -#include #include #include diff --git a/src/Common/examples/chaos_sanitizer.cpp b/src/Common/examples/chaos_sanitizer.cpp index 0b7d21aa33e..76e22411a2e 100644 --- a/src/Common/examples/chaos_sanitizer.cpp +++ b/src/Common/examples/chaos_sanitizer.cpp @@ -6,7 +6,6 @@ #include -#include #include diff --git a/src/Common/examples/compact_array.cpp b/src/Common/examples/compact_array.cpp index 58c4ea3be1e..9e16372deff 100644 --- a/src/Common/examples/compact_array.cpp +++ b/src/Common/examples/compact_array.cpp @@ -4,11 +4,11 @@ #include #include #include -#include #include #include #include + namespace fs = std::filesystem; static std::string createTmpPath(const std::string & filename) diff --git a/src/Common/tests/gtest_hash_table.cpp b/src/Common/tests/gtest_hash_table.cpp index 0221a682577..72941126cfd 100644 --- a/src/Common/tests/gtest_hash_table.cpp +++ b/src/Common/tests/gtest_hash_table.cpp @@ -1,5 +1,5 @@ #include -#include +#include #include diff --git a/src/Common/tests/gtest_lru_cache.cpp b/src/Common/tests/gtest_lru_cache.cpp index 1185dd58e5e..0360b5b3324 100644 --- a/src/Common/tests/gtest_lru_cache.cpp +++ b/src/Common/tests/gtest_lru_cache.cpp @@ -1,5 +1,4 @@ #include -#include #include #include diff --git a/src/Common/tests/gtest_lru_hash_map.cpp b/src/Common/tests/gtest_lru_hash_map.cpp index f45a503be43..cbb3d302efc 100644 --- a/src/Common/tests/gtest_lru_hash_map.cpp +++ b/src/Common/tests/gtest_lru_hash_map.cpp @@ -1,5 +1,4 @@ #include -#include #include diff --git a/src/Common/tests/gtest_lru_resource_cache.cpp b/src/Common/tests/gtest_lru_resource_cache.cpp index f88eded531e..bc037824ff8 100644 --- a/src/Common/tests/gtest_lru_resource_cache.cpp +++ b/src/Common/tests/gtest_lru_resource_cache.cpp @@ -1,5 +1,4 @@ #include -#include #include #include diff --git a/src/Common/tests/gtest_slru_cache.cpp b/src/Common/tests/gtest_slru_cache.cpp index 52549592f0e..ed04f427d9d 100644 --- a/src/Common/tests/gtest_slru_cache.cpp +++ b/src/Common/tests/gtest_slru_cache.cpp @@ -1,5 +1,4 @@ #include -#include #include #include diff --git a/src/Common/tests/gtest_thread_pool_limit.cpp b/src/Common/tests/gtest_thread_pool_limit.cpp index 17f79d17894..b47c8cdad18 100644 --- a/src/Common/tests/gtest_thread_pool_limit.cpp +++ b/src/Common/tests/gtest_thread_pool_limit.cpp @@ -1,5 +1,4 @@ #include -#include #include #include diff --git a/src/Common/tests/gtest_thread_pool_loop.cpp b/src/Common/tests/gtest_thread_pool_loop.cpp index 556c39df949..170a888ff72 100644 --- a/src/Common/tests/gtest_thread_pool_loop.cpp +++ b/src/Common/tests/gtest_thread_pool_loop.cpp @@ -1,5 +1,4 @@ #include -#include #include #include diff --git a/src/Common/tests/gtest_thread_pool_schedule_exception.cpp b/src/Common/tests/gtest_thread_pool_schedule_exception.cpp index 5dbad00848d..d8e00b5314c 100644 --- a/src/Common/tests/gtest_thread_pool_schedule_exception.cpp +++ b/src/Common/tests/gtest_thread_pool_schedule_exception.cpp @@ -1,4 +1,3 @@ -#include #include #include #include diff --git a/src/Compression/CompressionCodecDeflateQpl.cpp b/src/Compression/CompressionCodecDeflateQpl.cpp index 5dcd6008b51..0737e523ba0 100644 --- a/src/Compression/CompressionCodecDeflateQpl.cpp +++ b/src/Compression/CompressionCodecDeflateQpl.cpp @@ -1,15 +1,17 @@ #ifdef ENABLE_QPL_COMPRESSION + #include #include #include #include #include -#include #include #include #include "libaccel_config.h" #include #include +#include + namespace DB { diff --git a/src/Compression/fuzzers/compressed_buffer_fuzzer.cpp b/src/Compression/fuzzers/compressed_buffer_fuzzer.cpp index 1f669696fb9..bdab11ef8ab 100644 --- a/src/Compression/fuzzers/compressed_buffer_fuzzer.cpp +++ b/src/Compression/fuzzers/compressed_buffer_fuzzer.cpp @@ -1,4 +1,3 @@ -#include #include #include #include diff --git a/src/Compression/fuzzers/delta_decompress_fuzzer.cpp b/src/Compression/fuzzers/delta_decompress_fuzzer.cpp index b039777da15..eaef1d1896c 100644 --- a/src/Compression/fuzzers/delta_decompress_fuzzer.cpp +++ b/src/Compression/fuzzers/delta_decompress_fuzzer.cpp @@ -1,4 +1,3 @@ -#include #include #include diff --git a/src/Compression/fuzzers/double_delta_decompress_fuzzer.cpp b/src/Compression/fuzzers/double_delta_decompress_fuzzer.cpp index f9822daa3bd..c32120bacbf 100644 --- a/src/Compression/fuzzers/double_delta_decompress_fuzzer.cpp +++ b/src/Compression/fuzzers/double_delta_decompress_fuzzer.cpp @@ -1,4 +1,3 @@ -#include #include #include diff --git a/src/Compression/fuzzers/encrypted_decompress_fuzzer.cpp b/src/Compression/fuzzers/encrypted_decompress_fuzzer.cpp index 3e3d0e164fe..eb95c83e1d7 100644 --- a/src/Compression/fuzzers/encrypted_decompress_fuzzer.cpp +++ b/src/Compression/fuzzers/encrypted_decompress_fuzzer.cpp @@ -1,6 +1,5 @@ #include #include -#include #include #include diff --git a/src/Compression/fuzzers/lz4_decompress_fuzzer.cpp b/src/Compression/fuzzers/lz4_decompress_fuzzer.cpp index 85c4c9bd329..f03fc716c2c 100644 --- a/src/Compression/fuzzers/lz4_decompress_fuzzer.cpp +++ b/src/Compression/fuzzers/lz4_decompress_fuzzer.cpp @@ -1,4 +1,3 @@ -#include #include #include diff --git a/src/Coordination/FourLetterCommand.cpp b/src/Coordination/FourLetterCommand.cpp index d1ff03482b6..379a24e5aab 100644 --- a/src/Coordination/FourLetterCommand.cpp +++ b/src/Coordination/FourLetterCommand.cpp @@ -577,7 +577,6 @@ String FeatureFlagsCommand::run() } return ret.str(); - } } diff --git a/src/Coordination/SnapshotableHashTable.h b/src/Coordination/SnapshotableHashTable.h index 7db546bd4c8..093126237ef 100644 --- a/src/Coordination/SnapshotableHashTable.h +++ b/src/Coordination/SnapshotableHashTable.h @@ -6,7 +6,7 @@ #include #include #include -#include + namespace DB { diff --git a/src/Coordination/SummingStateMachine.cpp b/src/Coordination/SummingStateMachine.cpp index 4c21f1d8658..b70509fe6f0 100644 --- a/src/Coordination/SummingStateMachine.cpp +++ b/src/Coordination/SummingStateMachine.cpp @@ -1,5 +1,4 @@ #include -#include #include namespace DB diff --git a/src/Coordination/pathUtils.cpp b/src/Coordination/pathUtils.cpp index afa42b4a639..25f8e25cf06 100644 --- a/src/Coordination/pathUtils.cpp +++ b/src/Coordination/pathUtils.cpp @@ -1,5 +1,4 @@ #include -#include namespace DB { diff --git a/src/Core/MySQL/MySQLCharset.cpp b/src/Core/MySQL/MySQLCharset.cpp index 869941ebd84..787e4edcf88 100644 --- a/src/Core/MySQL/MySQLCharset.cpp +++ b/src/Core/MySQL/MySQLCharset.cpp @@ -1,6 +1,5 @@ #include "MySQLCharset.h" #include "config.h" -#include #include #if USE_ICU diff --git a/src/Core/fuzzers/names_and_types_fuzzer.cpp b/src/Core/fuzzers/names_and_types_fuzzer.cpp index cc4a2920c66..94f0872fff6 100644 --- a/src/Core/fuzzers/names_and_types_fuzzer.cpp +++ b/src/Core/fuzzers/names_and_types_fuzzer.cpp @@ -1,4 +1,3 @@ -#include #include #include diff --git a/src/Daemon/BaseDaemon.h b/src/Daemon/BaseDaemon.h index 7aa1e8ad1a0..952cf61d8e0 100644 --- a/src/Daemon/BaseDaemon.h +++ b/src/Daemon/BaseDaemon.h @@ -2,7 +2,6 @@ #include #include -#include #include #include #include diff --git a/src/Daemon/SentryWriter.cpp b/src/Daemon/SentryWriter.cpp index e38d339d088..942d1e306ae 100644 --- a/src/Daemon/SentryWriter.cpp +++ b/src/Daemon/SentryWriter.cpp @@ -3,7 +3,6 @@ #include #include -#include #include #include #include @@ -13,7 +12,6 @@ #include #include #include -#include #include "config.h" #include "config_version.h" diff --git a/src/DataTypes/tests/gtest_DataType_deserializeAsText.cpp b/src/DataTypes/tests/gtest_DataType_deserializeAsText.cpp index b755bd109d0..0373e55a62d 100644 --- a/src/DataTypes/tests/gtest_DataType_deserializeAsText.cpp +++ b/src/DataTypes/tests/gtest_DataType_deserializeAsText.cpp @@ -1,8 +1,6 @@ #include #include #include -#include -#include #include #include diff --git a/src/Dictionaries/HashedDictionary.cpp b/src/Dictionaries/HashedDictionary.cpp index 5f25600db8f..e336ca80856 100644 --- a/src/Dictionaries/HashedDictionary.cpp +++ b/src/Dictionaries/HashedDictionary.cpp @@ -25,6 +25,7 @@ #include #include + namespace CurrentMetrics { extern const Metric HashedDictionaryThreads; diff --git a/src/Dictionaries/PolygonDictionaryUtils.h b/src/Dictionaries/PolygonDictionaryUtils.h index 94b8b961577..0238ef0b2b9 100644 --- a/src/Dictionaries/PolygonDictionaryUtils.h +++ b/src/Dictionaries/PolygonDictionaryUtils.h @@ -13,6 +13,7 @@ #include + namespace DB { diff --git a/src/Disks/IO/ReadBufferFromAzureBlobStorage.cpp b/src/Disks/IO/ReadBufferFromAzureBlobStorage.cpp index 129bb97be09..cd8502c913f 100644 --- a/src/Disks/IO/ReadBufferFromAzureBlobStorage.cpp +++ b/src/Disks/IO/ReadBufferFromAzureBlobStorage.cpp @@ -102,6 +102,19 @@ bool ReadBufferFromAzureBlobStorage::nextImpl() size_t bytes_read = 0; size_t sleep_time_with_backoff_milliseconds = 100; + + auto handle_exception = [&, this](const auto & e, size_t i) + { + LOG_INFO(log, "Exception caught during Azure Read for file {} at attempt {}/{}: {}", path, i + 1, max_single_read_retries, e.Message); + if (i + 1 == max_single_read_retries) + throw; + + sleepForMilliseconds(sleep_time_with_backoff_milliseconds); + sleep_time_with_backoff_milliseconds *= 2; + initialized = false; + initialize(); + }; + for (size_t i = 0; i < max_single_read_retries; ++i) { try @@ -111,16 +124,13 @@ bool ReadBufferFromAzureBlobStorage::nextImpl() read_settings.remote_throttler->add(bytes_read, ProfileEvents::RemoteReadThrottlerBytes, ProfileEvents::RemoteReadThrottlerSleepMicroseconds); break; } + catch (const Azure::Core::Http::TransportException & e) + { + handle_exception(e, i); + } catch (const Azure::Storage::StorageException & e) { - LOG_INFO(log, "Exception caught during Azure Read for file {} at attempt {}: {}", path, i, e.Message); - if (i + 1 == max_single_read_retries) - throw; - - sleepForMilliseconds(sleep_time_with_backoff_milliseconds); - sleep_time_with_backoff_milliseconds *= 2; - initialized = false; - initialize(); + handle_exception(e, i); } } @@ -211,6 +221,17 @@ void ReadBufferFromAzureBlobStorage::initialize() blob_client = std::make_unique(blob_container_client->GetBlobClient(path)); size_t sleep_time_with_backoff_milliseconds = 100; + + auto handle_exception = [&, this](const auto & e, size_t i) + { + LOG_INFO(log, "Exception caught during Azure Download for file {} at offset {} at attempt {}/{}: {}", path, offset, i + 1, max_single_download_retries, e.Message); + if (i + 1 == max_single_download_retries) + throw; + + sleepForMilliseconds(sleep_time_with_backoff_milliseconds); + sleep_time_with_backoff_milliseconds *= 2; + }; + for (size_t i = 0; i < max_single_download_retries; ++i) { try @@ -219,14 +240,13 @@ void ReadBufferFromAzureBlobStorage::initialize() data_stream = std::move(download_response.Value.BodyStream); break; } + catch (const Azure::Core::Http::TransportException & e) + { + handle_exception(e, i); + } catch (const Azure::Core::RequestFailedException & e) { - LOG_INFO(log, "Exception caught during Azure Download for file {} at offset {} at attempt {} : {}", path, offset, i + 1, e.Message); - if (i + 1 == max_single_download_retries) - throw; - - sleepForMilliseconds(sleep_time_with_backoff_milliseconds); - sleep_time_with_backoff_milliseconds *= 2; + handle_exception(e,i); } } diff --git a/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp b/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp index cc75f18bc0d..dd33395fbfa 100644 --- a/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp +++ b/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp @@ -2,7 +2,6 @@ #include -#include #include #include #include diff --git a/src/Formats/ReadSchemaUtils.cpp b/src/Formats/ReadSchemaUtils.cpp index cbfa05bfba6..b40fb20dadf 100644 --- a/src/Formats/ReadSchemaUtils.cpp +++ b/src/Formats/ReadSchemaUtils.cpp @@ -1,13 +1,11 @@ -#include -#include #include -#include -#include #include #include #include #include #include +#include + namespace DB { @@ -53,6 +51,7 @@ ColumnsDescription readSchemaFromFormat( bool retry, ContextPtr & context, std::unique_ptr & buf) +try { NamesAndTypesList names_and_types; if (FormatFactory::instance().checkIfFormatHasExternalSchemaReader(format_name)) @@ -209,12 +208,23 @@ ColumnsDescription readSchemaFromFormat( ErrorCodes::BAD_ARGUMENTS, "{} file format doesn't support schema inference. You must specify the structure manually", format_name); + /// Some formats like CSVWithNames can contain empty column names. We don't support empty column names and further processing can fail with an exception. Let's just remove columns with empty names from the structure. names_and_types.erase( std::remove_if(names_and_types.begin(), names_and_types.end(), [](const NameAndTypePair & pair) { return pair.name.empty(); }), names_and_types.end()); return ColumnsDescription(names_and_types); } +catch (Exception & e) +{ + if (!buf) + throw; + auto file_name = getFileNameFromReadBuffer(*buf); + if (!file_name.empty()) + e.addMessage(fmt::format("(in file/uri {})", file_name)); + throw; +} + ColumnsDescription readSchemaFromFormat( const String & format_name, diff --git a/src/Functions/FunctionBinaryArithmetic.h b/src/Functions/FunctionBinaryArithmetic.h index 723abc7c308..86a5ce29d9c 100644 --- a/src/Functions/FunctionBinaryArithmetic.h +++ b/src/Functions/FunctionBinaryArithmetic.h @@ -1173,21 +1173,8 @@ class FunctionBinaryArithmetic : public IFunction const auto * left_array_col = typeid_cast(arguments[0].column.get()); const auto * right_array_col = typeid_cast(arguments[1].column.get()); - const auto & left_offsets = left_array_col->getOffsets(); - const auto & right_offsets = right_array_col->getOffsets(); - - chassert(left_offsets.size() == right_offsets.size() && "Unexpected difference in number of offsets"); - /// Unpacking non-const arrays and checking sizes of them. - for (auto offset_index = 0U; offset_index < left_offsets.size(); ++offset_index) - { - if (left_offsets[offset_index] != right_offsets[offset_index]) - { - throw Exception(ErrorCodes::SIZES_OF_ARRAYS_DONT_MATCH, - "Cannot apply operation for arrays of different sizes. Size of the first argument: {}, size of the second argument: {}", - *left_array_col->getOffsets().data(), - *right_array_col ->getOffsets().data()); - } - } + if (!left_array_col->hasEqualOffsets(*right_array_col)) + throw Exception(ErrorCodes::SIZES_OF_ARRAYS_DONT_MATCH, "Two arguments for function {} must have equal sizes", getName()); const auto & left_array_type = typeid_cast(arguments[0].type.get())->getNestedType(); new_arguments[0] = {left_array_col->getDataPtr(), left_array_type, arguments[0].name}; @@ -1198,6 +1185,7 @@ class FunctionBinaryArithmetic : public IFunction result_array_type = typeid_cast(result_type.get())->getNestedType(); size_t rows_count = 0; + const auto & left_offsets = left_array_col->getOffsets(); if (!left_offsets.empty()) rows_count = left_offsets.back(); auto res = executeImpl(new_arguments, result_array_type, rows_count); diff --git a/src/Functions/FunctionStartsEndsWith.h b/src/Functions/FunctionStartsEndsWith.h index b15bab47ae0..7671129fcfc 100644 --- a/src/Functions/FunctionStartsEndsWith.h +++ b/src/Functions/FunctionStartsEndsWith.h @@ -28,10 +28,24 @@ namespace ErrorCodes struct NameStartsWith { static constexpr auto name = "startsWith"; + static constexpr auto is_utf8 = false; }; struct NameEndsWith { static constexpr auto name = "endsWith"; + static constexpr auto is_utf8 = false; +}; + +struct NameStartsWithUTF8 +{ + static constexpr auto name = "startsWithUTF8"; + static constexpr auto is_utf8 = true; +}; + +struct NameEndsWithUTF8 +{ + static constexpr auto name = "endsWithUTF8"; + static constexpr auto is_utf8 = true; }; DECLARE_MULTITARGET_CODE( @@ -41,6 +55,7 @@ class FunctionStartsEndsWith : public IFunction { public: static constexpr auto name = Name::name; + static constexpr auto is_utf8 = Name::is_utf8; String getName() const override { @@ -64,7 +79,8 @@ public: DataTypePtr getReturnTypeImpl(const DataTypes & arguments) const override { - if (isStringOrFixedString(arguments[0]) && isStringOrFixedString(arguments[1])) + if (!is_utf8 && isStringOrFixedString(arguments[0]) && isStringOrFixedString(arguments[1]) + || isString(arguments[0]) && isString(arguments[1])) return std::make_shared(); if (isArray(arguments[0]) && isArray(arguments[1])) @@ -78,8 +94,11 @@ public: ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { auto data_type = arguments[0].type; - if (isStringOrFixedString(*data_type)) + + if (!is_utf8 && isStringOrFixedString(*data_type)) return executeImplString(arguments, {}, input_rows_count); + if (is_utf8 && isString(*data_type)) + return executeImplStringUTF8(arguments, {}, input_rows_count); if (isArray(data_type)) return executeImplArray(arguments, {}, input_rows_count); return {}; @@ -131,7 +150,6 @@ private: typename ColumnVector::Container & vec_res = col_res->getData(); vec_res.resize(input_rows_count); - if (const ColumnString * haystack = checkAndGetColumn(haystack_column)) dispatch(StringSource(*haystack), needle_column, vec_res); else if (const ColumnFixedString * haystack_fixed = checkAndGetColumn(haystack_column)) @@ -146,6 +164,26 @@ private: return col_res; } + ColumnPtr executeImplStringUTF8(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const + { + const IColumn * haystack_column = arguments[0].column.get(); + const IColumn * needle_column = arguments[1].column.get(); + + auto col_res = ColumnVector::create(); + typename ColumnVector::Container & vec_res = col_res->getData(); + + vec_res.resize(input_rows_count); + if (const ColumnString * haystack = checkAndGetColumn(haystack_column)) + dispatchUTF8(UTF8StringSource(*haystack), needle_column, vec_res); + else if (const ColumnConst * haystack_const = checkAndGetColumnConst(haystack_column)) + dispatchUTF8>(ConstSource(*haystack_const), needle_column, vec_res); + else + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal combination of columns as arguments of function {}", getName()); + + return col_res; + } + + template void dispatch(HaystackSource haystack_source, const IColumn * needle_column, PaddedPODArray & res_data) const { @@ -161,6 +199,17 @@ private: throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal combination of columns as arguments of function {}", getName()); } + template + void dispatchUTF8(HaystackSource haystack_source, const IColumn * needle_column, PaddedPODArray & res_data) const + { + if (const ColumnString * needle = checkAndGetColumn(needle_column)) + execute(haystack_source, UTF8StringSource(*needle), res_data); + else if (const ColumnConst * needle_const = checkAndGetColumnConst(needle_column)) + execute>(haystack_source, ConstSource(*needle_const), res_data); + else + throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Illegal combination of columns as arguments of function {}", getName()); + } + template static void execute(HaystackSource haystack_source, NeedleSource needle_source, PaddedPODArray & res_data) { @@ -172,18 +221,27 @@ private: auto needle = needle_source.getWhole(); if (needle.size > haystack.size) - { res_data[row_num] = false; - } else { - if constexpr (std::is_same_v) - { + if constexpr (std::is_same_v) /// startsWith res_data[row_num] = StringRef(haystack.data, needle.size) == StringRef(needle.data, needle.size); - } - else /// endsWith - { + else if constexpr (std::is_same_v) /// endsWith res_data[row_num] = StringRef(haystack.data + haystack.size - needle.size, needle.size) == StringRef(needle.data, needle.size); + else /// startsWithUTF8 or endsWithUTF8 + { + auto length = UTF8::countCodePoints(needle.data, needle.size); + + if constexpr (std::is_same_v) + { + auto slice = haystack_source.getSliceFromLeft(0, length); + res_data[row_num] = StringRef(slice.data, slice.size) == StringRef(needle.data, needle.size); + } + else + { + auto slice = haystack_source.getSliceFromRight(length); + res_data[row_num] = StringRef(slice.data, slice.size) == StringRef(needle.data, needle.size); + } } } diff --git a/src/Functions/array/arrayEnumerateRanked.h b/src/Functions/array/arrayEnumerateRanked.h index 0733f1e2d43..2e28a0f455d 100644 --- a/src/Functions/array/arrayEnumerateRanked.h +++ b/src/Functions/array/arrayEnumerateRanked.h @@ -13,7 +13,6 @@ #include #include -// for better debug: #include /** The function will enumerate distinct values of the passed multidimensional arrays looking inside at the specified depths. * This is very unusual function made as a special order for our dear customer - Metrica web analytics system. diff --git a/src/Functions/array/arrayShuffle.cpp b/src/Functions/array/arrayShuffle.cpp index 7c20e195098..faa5ae47b29 100644 --- a/src/Functions/array/arrayShuffle.cpp +++ b/src/Functions/array/arrayShuffle.cpp @@ -16,6 +16,7 @@ #include #include + namespace DB { diff --git a/src/Functions/endsWithUTF8.cpp b/src/Functions/endsWithUTF8.cpp new file mode 100644 index 00000000000..1b042452298 --- /dev/null +++ b/src/Functions/endsWithUTF8.cpp @@ -0,0 +1,21 @@ +#include +#include +#include + + +namespace DB +{ + +using FunctionEndsWithUTF8 = FunctionStartsEndsWith; + +REGISTER_FUNCTION(EndsWithUTF8) +{ + factory.registerFunction(FunctionDocumentation{ + .description = R"( +Returns whether string `str` ends with `suffix`, the difference between `endsWithUTF8` and `endsWith` is that `endsWithUTF8` match `str` and `suffix` by UTF-8 characters. + )", + .examples{{"endsWithUTF8", "select endsWithUTF8('富强民主文明和谐', '富强');", ""}}, + .categories{"String"}}); +} + +} diff --git a/src/Functions/evalMLMethod.cpp b/src/Functions/evalMLMethod.cpp index 346c8249905..4d5657f0aab 100644 --- a/src/Functions/evalMLMethod.cpp +++ b/src/Functions/evalMLMethod.cpp @@ -5,7 +5,6 @@ #include #include -#include #include diff --git a/src/Functions/extractAllGroups.h b/src/Functions/extractAllGroups.h index 3a7987be93e..f9637e59461 100644 --- a/src/Functions/extractAllGroups.h +++ b/src/Functions/extractAllGroups.h @@ -14,8 +14,6 @@ #include #include -#include - namespace DB { diff --git a/src/Functions/keyvaluepair/impl/StateHandler.h b/src/Functions/keyvaluepair/impl/StateHandler.h index 27c1a0b44be..178974e9d36 100644 --- a/src/Functions/keyvaluepair/impl/StateHandler.h +++ b/src/Functions/keyvaluepair/impl/StateHandler.h @@ -2,7 +2,6 @@ #include -#include namespace DB { diff --git a/src/Functions/keyvaluepair/tests/gtest_extractKeyValuePairs.cpp b/src/Functions/keyvaluepair/tests/gtest_extractKeyValuePairs.cpp index 507d9c0e5c7..55a08023cbd 100644 --- a/src/Functions/keyvaluepair/tests/gtest_extractKeyValuePairs.cpp +++ b/src/Functions/keyvaluepair/tests/gtest_extractKeyValuePairs.cpp @@ -9,11 +9,11 @@ #include #include -#include #include #include #include + namespace { using namespace DB; @@ -174,5 +174,3 @@ INSTANTIATE_TEST_SUITE_P(InvalidEscapeSeqInValue, extractKVPairKeyValuePairExtra } ) ); - - diff --git a/src/Functions/startsWithUTF8.cpp b/src/Functions/startsWithUTF8.cpp new file mode 100644 index 00000000000..3f46916d760 --- /dev/null +++ b/src/Functions/startsWithUTF8.cpp @@ -0,0 +1,21 @@ +#include +#include +#include + + +namespace DB +{ + +using FunctionStartsWithUTF8 = FunctionStartsEndsWith; + +REGISTER_FUNCTION(StartsWithUTF8) +{ + factory.registerFunction(FunctionDocumentation{ + .description = R"( +Returns whether string `str` starts with `prefix`, the difference between `startsWithUTF8` and `startsWith` is that `startsWithUTF8` match `str` and `suffix` by UTF-8 characters. + )", + .examples{{"startsWithUTF8", "select startsWithUTF8('富强民主文明和谐', '富强');", ""}}, + .categories{"String"}}); +} + +} diff --git a/src/Functions/translate.cpp b/src/Functions/translate.cpp index 83779eee23c..836cb4de2f3 100644 --- a/src/Functions/translate.cpp +++ b/src/Functions/translate.cpp @@ -6,6 +6,8 @@ #include #include #include +#include + namespace DB { diff --git a/src/IO/Archives/LibArchiveReader.h b/src/IO/Archives/LibArchiveReader.h index 700e8f70d04..ef50d9de56e 100644 --- a/src/IO/Archives/LibArchiveReader.h +++ b/src/IO/Archives/LibArchiveReader.h @@ -4,7 +4,6 @@ #include -#include namespace DB { diff --git a/src/IO/HTTPCommon.h b/src/IO/HTTPCommon.h index 082491b2851..f10fd748200 100644 --- a/src/IO/HTTPCommon.h +++ b/src/IO/HTTPCommon.h @@ -1,6 +1,5 @@ #pragma once -#include #include #include diff --git a/src/IO/ReadBufferFromIStream.h b/src/IO/ReadBufferFromIStream.h index 67cc60c053f..8c3f62728b5 100644 --- a/src/IO/ReadBufferFromIStream.h +++ b/src/IO/ReadBufferFromIStream.h @@ -1,7 +1,5 @@ #pragma once -#include - #include #include diff --git a/src/IO/S3/SessionAwareIOStream.h b/src/IO/S3/SessionAwareIOStream.h index f7e42f99f51..babe52545d1 100644 --- a/src/IO/S3/SessionAwareIOStream.h +++ b/src/IO/S3/SessionAwareIOStream.h @@ -1,6 +1,6 @@ #pragma once -#include +#include namespace DB::S3 diff --git a/src/IO/StdStreamFromReadBuffer.h b/src/IO/StdStreamFromReadBuffer.h index eae939a28b5..ff327dc342e 100644 --- a/src/IO/StdStreamFromReadBuffer.h +++ b/src/IO/StdStreamFromReadBuffer.h @@ -1,7 +1,6 @@ #pragma once #include -#include #include diff --git a/src/IO/VarInt.h b/src/IO/VarInt.h index 2a2743e3407..8d10055a3df 100644 --- a/src/IO/VarInt.h +++ b/src/IO/VarInt.h @@ -1,6 +1,5 @@ #pragma once -#include #include #include #include diff --git a/src/IO/WriteBuffer.h b/src/IO/WriteBuffer.h index ea032302235..d29ca6d5c6c 100644 --- a/src/IO/WriteBuffer.h +++ b/src/IO/WriteBuffer.h @@ -2,7 +2,6 @@ #include #include -#include #include #include diff --git a/src/IO/WriteBufferFromOStream.h b/src/IO/WriteBufferFromOStream.h index 5a933739cb1..3f9d3ee3d92 100644 --- a/src/IO/WriteBufferFromOStream.h +++ b/src/IO/WriteBufferFromOStream.h @@ -1,6 +1,6 @@ #pragma once -#include +#include #include #include diff --git a/src/IO/examples/read_buffer_from_hdfs.cpp b/src/IO/examples/read_buffer_from_hdfs.cpp index da4e5298681..977dd2ae227 100644 --- a/src/IO/examples/read_buffer_from_hdfs.cpp +++ b/src/IO/examples/read_buffer_from_hdfs.cpp @@ -1,4 +1,3 @@ -#include #include #include #include diff --git a/src/Interpreters/Access/InterpreterCreateQuotaQuery.cpp b/src/Interpreters/Access/InterpreterCreateQuotaQuery.cpp index 83a620d11c6..e271497ff5c 100644 --- a/src/Interpreters/Access/InterpreterCreateQuotaQuery.cpp +++ b/src/Interpreters/Access/InterpreterCreateQuotaQuery.cpp @@ -14,6 +14,12 @@ namespace DB { + +namespace ErrorCodes +{ + extern const int ACCESS_ENTITY_ALREADY_EXISTS; +} + namespace { void updateQuotaFromQueryImpl( @@ -90,6 +96,15 @@ BlockIO InterpreterCreateQuotaQuery::execute() if (query.roles) roles_from_query = RolesOrUsersSet{*query.roles, access_control, getContext()->getUserID()}; + IAccessStorage * storage = &access_control; + MultipleAccessStorage::StoragePtr storage_ptr; + + if (!query.storage_name.empty()) + { + storage_ptr = access_control.getStorageByName(query.storage_name); + storage = storage_ptr.get(); + } + if (query.alter) { auto update_func = [&](const AccessEntityPtr & entity) -> AccessEntityPtr @@ -100,11 +115,11 @@ BlockIO InterpreterCreateQuotaQuery::execute() }; if (query.if_exists) { - auto ids = access_control.find(query.names); - access_control.tryUpdate(ids, update_func); + auto ids = storage->find(query.names); + storage->tryUpdate(ids, update_func); } else - access_control.update(access_control.getIDs(query.names), update_func); + storage->update(storage->getIDs(query.names), update_func); } else { @@ -116,12 +131,21 @@ BlockIO InterpreterCreateQuotaQuery::execute() new_quotas.emplace_back(std::move(new_quota)); } + if (!query.storage_name.empty()) + { + for (const auto & name : query.names) + { + if (auto another_storage_ptr = access_control.findExcludingStorage(AccessEntityType::QUOTA, name, storage_ptr)) + throw Exception(ErrorCodes::ACCESS_ENTITY_ALREADY_EXISTS, "Quota {} already exists in storage {}", name, another_storage_ptr->getStorageName()); + } + } + if (query.if_not_exists) - access_control.tryInsert(new_quotas); + storage->tryInsert(new_quotas); else if (query.or_replace) - access_control.insertOrReplace(new_quotas); + storage->insertOrReplace(new_quotas); else - access_control.insert(new_quotas); + storage->insert(new_quotas); } return {}; diff --git a/src/Interpreters/Access/InterpreterCreateRoleQuery.cpp b/src/Interpreters/Access/InterpreterCreateRoleQuery.cpp index 2b22857d5de..d0c41c9e8f0 100644 --- a/src/Interpreters/Access/InterpreterCreateRoleQuery.cpp +++ b/src/Interpreters/Access/InterpreterCreateRoleQuery.cpp @@ -8,6 +8,12 @@ namespace DB { + +namespace ErrorCodes +{ + extern const int ACCESS_ENTITY_ALREADY_EXISTS; +} + namespace { void updateRoleFromQueryImpl( @@ -52,6 +58,15 @@ BlockIO InterpreterCreateRoleQuery::execute() if (!query.cluster.empty()) return executeDDLQueryOnCluster(query_ptr, getContext()); + IAccessStorage * storage = &access_control; + MultipleAccessStorage::StoragePtr storage_ptr; + + if (!query.storage_name.empty()) + { + storage_ptr = access_control.getStorageByName(query.storage_name); + storage = storage_ptr.get(); + } + if (query.alter) { auto update_func = [&](const AccessEntityPtr & entity) -> AccessEntityPtr @@ -62,11 +77,11 @@ BlockIO InterpreterCreateRoleQuery::execute() }; if (query.if_exists) { - auto ids = access_control.find(query.names); - access_control.tryUpdate(ids, update_func); + auto ids = storage->find(query.names); + storage->tryUpdate(ids, update_func); } else - access_control.update(access_control.getIDs(query.names), update_func); + storage->update(storage->getIDs(query.names), update_func); } else { @@ -78,12 +93,21 @@ BlockIO InterpreterCreateRoleQuery::execute() new_roles.emplace_back(std::move(new_role)); } + if (!query.storage_name.empty()) + { + for (const auto & name : query.names) + { + if (auto another_storage_ptr = access_control.findExcludingStorage(AccessEntityType::ROLE, name, storage_ptr)) + throw Exception(ErrorCodes::ACCESS_ENTITY_ALREADY_EXISTS, "Role {} already exists in storage {}", name, another_storage_ptr->getStorageName()); + } + } + if (query.if_not_exists) - access_control.tryInsert(new_roles); + storage->tryInsert(new_roles); else if (query.or_replace) - access_control.insertOrReplace(new_roles); + storage->insertOrReplace(new_roles); else - access_control.insert(new_roles); + storage->insert(new_roles); } return {}; diff --git a/src/Interpreters/Access/InterpreterCreateRowPolicyQuery.cpp b/src/Interpreters/Access/InterpreterCreateRowPolicyQuery.cpp index 87dc9039c17..a938d7afc16 100644 --- a/src/Interpreters/Access/InterpreterCreateRowPolicyQuery.cpp +++ b/src/Interpreters/Access/InterpreterCreateRowPolicyQuery.cpp @@ -14,6 +14,12 @@ namespace DB { + +namespace ErrorCodes +{ + extern const int ACCESS_ENTITY_ALREADY_EXISTS; +} + namespace { void updateRowPolicyFromQueryImpl( @@ -66,6 +72,16 @@ BlockIO InterpreterCreateRowPolicyQuery::execute() if (query.roles) roles_from_query = RolesOrUsersSet{*query.roles, access_control, getContext()->getUserID()}; + IAccessStorage * storage = &access_control; + MultipleAccessStorage::StoragePtr storage_ptr; + + if (!query.storage_name.empty()) + { + storage_ptr = access_control.getStorageByName(query.storage_name); + storage = storage_ptr.get(); + } + + Strings names = query.names->toStrings(); if (query.alter) { auto update_func = [&](const AccessEntityPtr & entity) -> AccessEntityPtr @@ -74,14 +90,13 @@ BlockIO InterpreterCreateRowPolicyQuery::execute() updateRowPolicyFromQueryImpl(*updated_policy, query, {}, roles_from_query); return updated_policy; }; - Strings names = query.names->toStrings(); if (query.if_exists) { - auto ids = access_control.find(names); - access_control.tryUpdate(ids, update_func); + auto ids = storage->find(names); + storage->tryUpdate(ids, update_func); } else - access_control.update(access_control.getIDs(names), update_func); + storage->update(storage->getIDs(names), update_func); } else { @@ -93,12 +108,21 @@ BlockIO InterpreterCreateRowPolicyQuery::execute() new_policies.emplace_back(std::move(new_policy)); } + if (!query.storage_name.empty()) + { + for (const auto & name : names) + { + if (auto another_storage_ptr = access_control.findExcludingStorage(AccessEntityType::ROW_POLICY, name, storage_ptr)) + throw Exception(ErrorCodes::ACCESS_ENTITY_ALREADY_EXISTS, "Row policy {} already exists in storage {}", name, another_storage_ptr->getStorageName()); + } + } + if (query.if_not_exists) - access_control.tryInsert(new_policies); + storage->tryInsert(new_policies); else if (query.or_replace) - access_control.insertOrReplace(new_policies); + storage->insertOrReplace(new_policies); else - access_control.insert(new_policies); + storage->insert(new_policies); } return {}; diff --git a/src/Interpreters/Access/InterpreterCreateSettingsProfileQuery.cpp b/src/Interpreters/Access/InterpreterCreateSettingsProfileQuery.cpp index 8a79bab0b0d..77401b94e1c 100644 --- a/src/Interpreters/Access/InterpreterCreateSettingsProfileQuery.cpp +++ b/src/Interpreters/Access/InterpreterCreateSettingsProfileQuery.cpp @@ -10,6 +10,12 @@ namespace DB { + +namespace ErrorCodes +{ + extern const int ACCESS_ENTITY_ALREADY_EXISTS; +} + namespace { void updateSettingsProfileFromQueryImpl( @@ -67,6 +73,16 @@ BlockIO InterpreterCreateSettingsProfileQuery::execute() if (query.to_roles) roles_from_query = RolesOrUsersSet{*query.to_roles, access_control, getContext()->getUserID()}; + + IAccessStorage * storage = &access_control; + MultipleAccessStorage::StoragePtr storage_ptr; + + if (!query.storage_name.empty()) + { + storage_ptr = access_control.getStorageByName(query.storage_name); + storage = storage_ptr.get(); + } + if (query.alter) { auto update_func = [&](const AccessEntityPtr & entity) -> AccessEntityPtr @@ -77,11 +93,11 @@ BlockIO InterpreterCreateSettingsProfileQuery::execute() }; if (query.if_exists) { - auto ids = access_control.find(query.names); - access_control.tryUpdate(ids, update_func); + auto ids = storage->find(query.names); + storage->tryUpdate(ids, update_func); } else - access_control.update(access_control.getIDs(query.names), update_func); + storage->update(storage->getIDs(query.names), update_func); } else { @@ -93,12 +109,21 @@ BlockIO InterpreterCreateSettingsProfileQuery::execute() new_profiles.emplace_back(std::move(new_profile)); } + if (!query.storage_name.empty()) + { + for (const auto & name : query.names) + { + if (auto another_storage_ptr = access_control.findExcludingStorage(AccessEntityType::SETTINGS_PROFILE, name, storage_ptr)) + throw Exception(ErrorCodes::ACCESS_ENTITY_ALREADY_EXISTS, "Settings profile {} already exists in storage {}", name, another_storage_ptr->getStorageName()); + } + } + if (query.if_not_exists) - access_control.tryInsert(new_profiles); + storage->tryInsert(new_profiles); else if (query.or_replace) - access_control.insertOrReplace(new_profiles); + storage->insertOrReplace(new_profiles); else - access_control.insert(new_profiles); + storage->insert(new_profiles); } return {}; diff --git a/src/Interpreters/Access/InterpreterCreateUserQuery.cpp b/src/Interpreters/Access/InterpreterCreateUserQuery.cpp index 475ee270506..07fae5ba914 100644 --- a/src/Interpreters/Access/InterpreterCreateUserQuery.cpp +++ b/src/Interpreters/Access/InterpreterCreateUserQuery.cpp @@ -17,6 +17,7 @@ namespace DB namespace ErrorCodes { extern const int BAD_ARGUMENTS; + extern const int ACCESS_ENTITY_ALREADY_EXISTS; } namespace { @@ -139,6 +140,16 @@ BlockIO InterpreterCreateUserQuery::execute() if (!query.cluster.empty()) return executeDDLQueryOnCluster(query_ptr, getContext()); + IAccessStorage * storage = &access_control; + MultipleAccessStorage::StoragePtr storage_ptr; + + if (!query.storage_name.empty()) + { + storage_ptr = access_control.getStorageByName(query.storage_name); + storage = storage_ptr.get(); + } + + Strings names = query.names->toStrings(); if (query.alter) { std::optional grantees_from_query; @@ -152,14 +163,13 @@ BlockIO InterpreterCreateUserQuery::execute() return updated_user; }; - Strings names = query.names->toStrings(); if (query.if_exists) { - auto ids = access_control.find(names); - access_control.tryUpdate(ids, update_func); + auto ids = storage->find(names); + storage->tryUpdate(ids, update_func); } else - access_control.update(access_control.getIDs(names), update_func); + storage->update(storage->getIDs(names), update_func); } else { @@ -171,13 +181,22 @@ BlockIO InterpreterCreateUserQuery::execute() new_users.emplace_back(std::move(new_user)); } + if (!query.storage_name.empty()) + { + for (const auto & name : names) + { + if (auto another_storage_ptr = access_control.findExcludingStorage(AccessEntityType::USER, name, storage_ptr)) + throw Exception(ErrorCodes::ACCESS_ENTITY_ALREADY_EXISTS, "User {} already exists in storage {}", name, another_storage_ptr->getStorageName()); + } + } + std::vector ids; if (query.if_not_exists) - ids = access_control.tryInsert(new_users); + ids = storage->tryInsert(new_users); else if (query.or_replace) - ids = access_control.insertOrReplace(new_users); + ids = storage->insertOrReplace(new_users); else - ids = access_control.insert(new_users); + ids = storage->insert(new_users); if (query.grantees) { diff --git a/src/Interpreters/Access/InterpreterDropAccessEntityQuery.cpp b/src/Interpreters/Access/InterpreterDropAccessEntityQuery.cpp index d4c37064065..54e3b95226c 100644 --- a/src/Interpreters/Access/InterpreterDropAccessEntityQuery.cpp +++ b/src/Interpreters/Access/InterpreterDropAccessEntityQuery.cpp @@ -26,18 +26,26 @@ BlockIO InterpreterDropAccessEntityQuery::execute() query.replaceEmptyDatabase(getContext()->getCurrentDatabase()); - auto do_drop = [&](const Strings & names) + auto do_drop = [&](const Strings & names, const String & storage_name) { + IAccessStorage * storage = &access_control; + MultipleAccessStorage::StoragePtr storage_ptr; + if (!storage_name.empty()) + { + storage_ptr = access_control.getStorageByName(storage_name); + storage = storage_ptr.get(); + } + if (query.if_exists) - access_control.tryRemove(access_control.find(query.type, names)); + storage->tryRemove(storage->find(query.type, names)); else - access_control.remove(access_control.getIDs(query.type, names)); + storage->remove(storage->getIDs(query.type, names)); }; if (query.type == AccessEntityType::ROW_POLICY) - do_drop(query.row_policy_names->toStrings()); + do_drop(query.row_policy_names->toStrings(), query.storage_name); else - do_drop(query.names); + do_drop(query.names, query.storage_name); return {}; } diff --git a/src/Interpreters/Access/InterpreterMoveAccessEntityQuery.cpp b/src/Interpreters/Access/InterpreterMoveAccessEntityQuery.cpp new file mode 100644 index 00000000000..49e90783a59 --- /dev/null +++ b/src/Interpreters/Access/InterpreterMoveAccessEntityQuery.cpp @@ -0,0 +1,93 @@ +#include +#include +#include +#include +#include +#include + + +namespace DB +{ +namespace ErrorCodes +{ + extern const int NOT_IMPLEMENTED; + extern const int ACCESS_ENTITY_NOT_FOUND; +} + + +BlockIO InterpreterMoveAccessEntityQuery::execute() +{ + auto & query = query_ptr->as(); + auto & access_control = getContext()->getAccessControl(); + getContext()->checkAccess(getRequiredAccess()); + + if (!query.cluster.empty()) + return executeDDLQueryOnCluster(query_ptr, getContext()); + + query.replaceEmptyDatabase(getContext()->getCurrentDatabase()); + + std::vector ids; + if (query.type == AccessEntityType::ROW_POLICY) + ids = access_control.getIDs(query.type, query.row_policy_names->toStrings()); + else + ids = access_control.getIDs(query.type, query.names); + + /// Validate that all entities are from the same storage. + const auto source_storage = access_control.findStorage(ids.front()); + if (!source_storage->exists(ids)) + throw Exception(ErrorCodes::ACCESS_ENTITY_NOT_FOUND, "All access entities must be from the same storage in order to be moved"); + + access_control.moveAccessEntities(ids, source_storage->getStorageName(), query.storage_name); + return {}; +} + + +AccessRightsElements InterpreterMoveAccessEntityQuery::getRequiredAccess() const +{ + const auto & query = query_ptr->as(); + AccessRightsElements res; + switch (query.type) + { + case AccessEntityType::USER: + { + res.emplace_back(AccessType::DROP_USER); + res.emplace_back(AccessType::CREATE_USER); + return res; + } + case AccessEntityType::ROLE: + { + res.emplace_back(AccessType::DROP_ROLE); + res.emplace_back(AccessType::CREATE_ROLE); + return res; + } + case AccessEntityType::SETTINGS_PROFILE: + { + res.emplace_back(AccessType::DROP_SETTINGS_PROFILE); + res.emplace_back(AccessType::CREATE_SETTINGS_PROFILE); + return res; + } + case AccessEntityType::ROW_POLICY: + { + if (query.row_policy_names) + { + for (const auto & row_policy_name : query.row_policy_names->full_names) + { + res.emplace_back(AccessType::DROP_ROW_POLICY, row_policy_name.database, row_policy_name.table_name); + res.emplace_back(AccessType::CREATE_ROW_POLICY, row_policy_name.database, row_policy_name.table_name); + } + } + return res; + } + case AccessEntityType::QUOTA: + { + res.emplace_back(AccessType::DROP_QUOTA); + res.emplace_back(AccessType::CREATE_QUOTA); + return res; + } + case AccessEntityType::MAX: + break; + } + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "{}: type is not supported by DROP query", toString(query.type)); +} + +} diff --git a/src/Interpreters/Access/InterpreterMoveAccessEntityQuery.h b/src/Interpreters/Access/InterpreterMoveAccessEntityQuery.h new file mode 100644 index 00000000000..1a70bf7c065 --- /dev/null +++ b/src/Interpreters/Access/InterpreterMoveAccessEntityQuery.h @@ -0,0 +1,24 @@ +#pragma once + +#include +#include + + +namespace DB +{ +class AccessRightsElements; + +class InterpreterMoveAccessEntityQuery : public IInterpreter, WithMutableContext +{ +public: + InterpreterMoveAccessEntityQuery(const ASTPtr & query_ptr_, ContextMutablePtr context_) : WithMutableContext(context_), query_ptr(query_ptr_) {} + + BlockIO execute() override; + +private: + AccessRightsElements getRequiredAccess() const; + + ASTPtr query_ptr; +}; + +} diff --git a/src/Interpreters/AsynchronousInsertQueue.cpp b/src/Interpreters/AsynchronousInsertQueue.cpp index 4f15192f46c..c8a3f77556a 100644 --- a/src/Interpreters/AsynchronousInsertQueue.cpp +++ b/src/Interpreters/AsynchronousInsertQueue.cpp @@ -460,7 +460,6 @@ try const auto * log = &Poco::Logger::get("AsynchronousInsertQueue"); const auto & insert_query = assert_cast(*key.query); auto insert_context = Context::createCopy(global_context); - DB::CurrentThread::QueryScope query_scope_holder(insert_context); bool internal = false; // To enable logging this query bool async_insert = true; @@ -482,6 +481,9 @@ try insert_context->setInitialQueryStartTime(query_start_time); insert_context->setCurrentQueryId(insert_query_id); insert_context->setInitialQueryId(insert_query_id); + + DB::CurrentThread::QueryScope query_scope_holder(insert_context); + size_t log_queries_cut_to_length = insert_context->getSettingsRef().log_queries_cut_to_length; String query_for_logging = insert_query.hasSecretParts() ? insert_query.formatForLogging(log_queries_cut_to_length) diff --git a/src/Interpreters/Context.cpp b/src/Interpreters/Context.cpp index a210a9efbc7..8f98759f592 100644 --- a/src/Interpreters/Context.cpp +++ b/src/Interpreters/Context.cpp @@ -2,14 +2,10 @@ #include #include #include -#include #include -#include #include #include -#include #include -#include #include #include #include @@ -17,12 +13,10 @@ #include #include #include -#include #include #include #include #include -#include #include #include #include @@ -34,8 +28,6 @@ #include #include #include -#include -#include #include #include #include @@ -44,7 +36,6 @@ #include #include #include -#include #include #include #include @@ -56,7 +47,6 @@ #include #include #include -#include #include #include #include @@ -70,7 +60,6 @@ #include #include #include -#include #include #include #include @@ -87,8 +76,6 @@ #include #include #include -#include -#include #include #include #include @@ -98,14 +85,12 @@ #include #include #include -#include #include #include #include #include #include #include -#include #include #include #include @@ -118,12 +103,8 @@ #include #include #include -#include #include -#if USE_ROCKSDB -#include -#endif namespace fs = std::filesystem; diff --git a/src/Interpreters/GraceHashJoin.cpp b/src/Interpreters/GraceHashJoin.cpp index 5d72cf20740..89ea3a326cc 100644 --- a/src/Interpreters/GraceHashJoin.cpp +++ b/src/Interpreters/GraceHashJoin.cpp @@ -1,15 +1,11 @@ #include #include #include -#include #include #include #include -#include -#include -#include #include #include @@ -18,6 +14,9 @@ #include +#include + + namespace CurrentMetrics { extern const Metric TemporaryFilesForJoin; diff --git a/src/Interpreters/InterpreterFactory.cpp b/src/Interpreters/InterpreterFactory.cpp index 9cd1f2a251c..27c94119750 100644 --- a/src/Interpreters/InterpreterFactory.cpp +++ b/src/Interpreters/InterpreterFactory.cpp @@ -39,6 +39,7 @@ #include #include #include +#include #include #include #include @@ -96,6 +97,7 @@ #include #include #include +#include #include #include #include @@ -314,6 +316,10 @@ std::unique_ptr InterpreterFactory::get(ASTPtr & query, ContextMut { return std::make_unique(query, context); } + else if (query->as()) + { + return std::make_unique(query, context); + } else if (query->as()) { return std::make_unique(query, context); diff --git a/src/Interpreters/InterpreterKillQueryQuery.cpp b/src/Interpreters/InterpreterKillQueryQuery.cpp index 3330159aff5..82fe2f57a0e 100644 --- a/src/Interpreters/InterpreterKillQueryQuery.cpp +++ b/src/Interpreters/InterpreterKillQueryQuery.cpp @@ -24,7 +24,6 @@ #include #include #include -#include #include diff --git a/src/Interpreters/Session.cpp b/src/Interpreters/Session.cpp index de2a779b740..f8bd70afdb6 100644 --- a/src/Interpreters/Session.cpp +++ b/src/Interpreters/Session.cpp @@ -520,6 +520,8 @@ ContextMutablePtr Session::makeSessionContext() {}, session_context->getSettingsRef().max_sessions_for_user); + recordLoginSucess(session_context); + return session_context; } @@ -582,6 +584,8 @@ ContextMutablePtr Session::makeSessionContext(const String & session_name_, std: { session_name_ }, max_sessions_for_user); + recordLoginSucess(session_context); + return session_context; } @@ -655,24 +659,38 @@ ContextMutablePtr Session::makeQueryContextImpl(const ClientInfo * client_info_t if (user_id) user = query_context->getUser(); - if (!notified_session_log_about_login) - { - if (auto session_log = getSessionLog()) - { - session_log->addLoginSuccess( - auth_id, - named_session ? std::optional(named_session->key.second) : std::nullopt, - *query_context, - user); - - notified_session_log_about_login = true; - } - } + /// Interserver does not create session context + recordLoginSucess(query_context); return query_context; } +void Session::recordLoginSucess(ContextPtr login_context) const +{ + if (notified_session_log_about_login) + return; + + if (!login_context) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Session or query context must be created"); + + if (auto session_log = getSessionLog()) + { + const auto & settings = login_context->getSettingsRef(); + const auto access = login_context->getAccess(); + + session_log->addLoginSuccess(auth_id, + named_session ? named_session->key.second : "", + settings, + access, + getClientInfo(), + user); + } + + notified_session_log_about_login = true; +} + + void Session::releaseSessionID() { if (!named_session) diff --git a/src/Interpreters/Session.h b/src/Interpreters/Session.h index 51c0e3c71fa..81ef987b428 100644 --- a/src/Interpreters/Session.h +++ b/src/Interpreters/Session.h @@ -97,6 +97,8 @@ public: private: std::shared_ptr getSessionLog() const; ContextMutablePtr makeQueryContextImpl(const ClientInfo * client_info_to_copy, ClientInfo * client_info_to_move) const; + void recordLoginSucess(ContextPtr login_context) const; + mutable bool notified_session_log_about_login = false; const UUID auth_id; diff --git a/src/Interpreters/SessionLog.cpp b/src/Interpreters/SessionLog.cpp index 0a8a7fc18c5..b847eaf9824 100644 --- a/src/Interpreters/SessionLog.cpp +++ b/src/Interpreters/SessionLog.cpp @@ -199,12 +199,13 @@ void SessionLogElement::appendToBlock(MutableColumns & columns) const columns[i++]->insertData(auth_failure_reason.data(), auth_failure_reason.length()); } -void SessionLog::addLoginSuccess(const UUID & auth_id, std::optional session_id, const Context & login_context, const UserPtr & login_user) +void SessionLog::addLoginSuccess(const UUID & auth_id, + const String & session_id, + const Settings & settings, + const ContextAccessPtr & access, + const ClientInfo & client_info, + const UserPtr & login_user) { - const auto access = login_context.getAccess(); - const auto & settings = login_context.getSettingsRef(); - const auto & client_info = login_context.getClientInfo(); - DB::SessionLogElement log_entry(auth_id, SESSION_LOGIN_SUCCESS); log_entry.client_info = client_info; @@ -215,8 +216,7 @@ void SessionLog::addLoginSuccess(const UUID & auth_id, std::optional ses } log_entry.external_auth_server = login_user ? login_user->auth_data.getLDAPServerName() : ""; - if (session_id) - log_entry.session_id = *session_id; + log_entry.session_id = session_id; if (const auto roles_info = access->getRolesInfo()) log_entry.roles = roles_info->getCurrentRolesNames(); diff --git a/src/Interpreters/SessionLog.h b/src/Interpreters/SessionLog.h index 1282ac09c4d..8757bc12270 100644 --- a/src/Interpreters/SessionLog.h +++ b/src/Interpreters/SessionLog.h @@ -20,6 +20,7 @@ enum SessionLogElementType : int8_t class ContextAccess; struct User; using UserPtr = std::shared_ptr; +using ContextAccessPtr = std::shared_ptr; /** A struct which will be inserted as row into session_log table. * @@ -72,7 +73,13 @@ class SessionLog : public SystemLog using SystemLog::SystemLog; public: - void addLoginSuccess(const UUID & auth_id, std::optional session_id, const Context & login_context, const UserPtr & login_user); + void addLoginSuccess(const UUID & auth_id, + const String & session_id, + const Settings & settings, + const ContextAccessPtr & access, + const ClientInfo & client_info, + const UserPtr & login_user); + void addLoginFailure(const UUID & auth_id, const ClientInfo & info, const std::optional & user, const Exception & reason); void addLogOut(const UUID & auth_id, const UserPtr & login_user, const ClientInfo & client_info); }; diff --git a/src/Interpreters/SystemLog.h b/src/Interpreters/SystemLog.h index 437b1b2a6bb..cf287ad7775 100644 --- a/src/Interpreters/SystemLog.h +++ b/src/Interpreters/SystemLog.h @@ -111,7 +111,7 @@ public: * and new table get created - as if previous table was not exist. */ SystemLog(ContextPtr context_, - const SystemLogSettings& settings_, + const SystemLogSettings & settings_, std::shared_ptr> queue_ = nullptr); /** Append a record into log. @@ -131,8 +131,6 @@ protected: using Base::queue; private: - - /* Saving thread data */ const StorageID table_id; const String storage_def; diff --git a/src/Loggers/OwnSplitChannel.cpp b/src/Loggers/OwnSplitChannel.cpp index cdf8402745f..618d453c238 100644 --- a/src/Loggers/OwnSplitChannel.cpp +++ b/src/Loggers/OwnSplitChannel.cpp @@ -1,7 +1,6 @@ #include "OwnSplitChannel.h" #include "OwnFormattingChannel.h" -#include #include #include #include diff --git a/src/Parsers/ASTSelectWithUnionQuery.cpp b/src/Parsers/ASTSelectWithUnionQuery.cpp index 9550752b1f3..48b4ae3c38d 100644 --- a/src/Parsers/ASTSelectWithUnionQuery.cpp +++ b/src/Parsers/ASTSelectWithUnionQuery.cpp @@ -5,7 +5,6 @@ #include #include -#include namespace DB { diff --git a/src/Parsers/Access/ASTCreateQuotaQuery.cpp b/src/Parsers/Access/ASTCreateQuotaQuery.cpp index 56abedf5235..091b62b0a9f 100644 --- a/src/Parsers/Access/ASTCreateQuotaQuery.cpp +++ b/src/Parsers/Access/ASTCreateQuotaQuery.cpp @@ -170,6 +170,12 @@ void ASTCreateQuotaQuery::formatImpl(const FormatSettings & settings, FormatStat settings.ostr << (settings.hilite ? hilite_keyword : "") << " OR REPLACE" << (settings.hilite ? hilite_none : ""); formatNames(names, settings); + + if (!storage_name.empty()) + settings.ostr << (settings.hilite ? IAST::hilite_keyword : "") + << " IN " << (settings.hilite ? IAST::hilite_none : "") + << backQuoteIfNeed(storage_name); + formatOnCluster(settings); if (!new_name.empty()) diff --git a/src/Parsers/Access/ASTCreateQuotaQuery.h b/src/Parsers/Access/ASTCreateQuotaQuery.h index f5eb59800ec..aecbbb00f9a 100644 --- a/src/Parsers/Access/ASTCreateQuotaQuery.h +++ b/src/Parsers/Access/ASTCreateQuotaQuery.h @@ -38,6 +38,7 @@ public: Strings names; String new_name; std::optional key_type; + String storage_name; struct Limits { diff --git a/src/Parsers/Access/ASTCreateRoleQuery.cpp b/src/Parsers/Access/ASTCreateRoleQuery.cpp index d624b9a9157..eeeb34c97e4 100644 --- a/src/Parsers/Access/ASTCreateRoleQuery.cpp +++ b/src/Parsers/Access/ASTCreateRoleQuery.cpp @@ -71,6 +71,12 @@ void ASTCreateRoleQuery::formatImpl(const FormatSettings & format, FormatState & format.ostr << (format.hilite ? hilite_keyword : "") << " OR REPLACE" << (format.hilite ? hilite_none : ""); formatNames(names, format); + + if (!storage_name.empty()) + format.ostr << (format.hilite ? IAST::hilite_keyword : "") + << " IN " << (format.hilite ? IAST::hilite_none : "") + << backQuoteIfNeed(storage_name); + formatOnCluster(format); if (!new_name.empty()) diff --git a/src/Parsers/Access/ASTCreateRoleQuery.h b/src/Parsers/Access/ASTCreateRoleQuery.h index 42d1a4031b6..4e465553164 100644 --- a/src/Parsers/Access/ASTCreateRoleQuery.h +++ b/src/Parsers/Access/ASTCreateRoleQuery.h @@ -28,6 +28,7 @@ public: Strings names; String new_name; + String storage_name; std::shared_ptr settings; diff --git a/src/Parsers/Access/ASTCreateRowPolicyQuery.cpp b/src/Parsers/Access/ASTCreateRowPolicyQuery.cpp index ca888be2cfe..e95959703ee 100644 --- a/src/Parsers/Access/ASTCreateRowPolicyQuery.cpp +++ b/src/Parsers/Access/ASTCreateRowPolicyQuery.cpp @@ -168,6 +168,11 @@ void ASTCreateRowPolicyQuery::formatImpl(const FormatSettings & settings, Format settings.ostr << " "; names->format(settings); + if (!storage_name.empty()) + settings.ostr << (settings.hilite ? IAST::hilite_keyword : "") + << " IN " << (settings.hilite ? IAST::hilite_none : "") + << backQuoteIfNeed(storage_name); + formatOnCluster(settings); assert(names->cluster.empty()); diff --git a/src/Parsers/Access/ASTCreateRowPolicyQuery.h b/src/Parsers/Access/ASTCreateRowPolicyQuery.h index 32d98fab822..3f2418e7287 100644 --- a/src/Parsers/Access/ASTCreateRowPolicyQuery.h +++ b/src/Parsers/Access/ASTCreateRowPolicyQuery.h @@ -35,6 +35,7 @@ public: bool if_exists = false; bool if_not_exists = false; bool or_replace = false; + String storage_name; std::shared_ptr names; String new_short_name; diff --git a/src/Parsers/Access/ASTCreateSettingsProfileQuery.cpp b/src/Parsers/Access/ASTCreateSettingsProfileQuery.cpp index 56ddef433ef..3379486d789 100644 --- a/src/Parsers/Access/ASTCreateSettingsProfileQuery.cpp +++ b/src/Parsers/Access/ASTCreateSettingsProfileQuery.cpp @@ -81,6 +81,12 @@ void ASTCreateSettingsProfileQuery::formatImpl(const FormatSettings & format, Fo format.ostr << (format.hilite ? hilite_keyword : "") << " OR REPLACE" << (format.hilite ? hilite_none : ""); formatNames(names, format); + + if (!storage_name.empty()) + format.ostr << (format.hilite ? IAST::hilite_keyword : "") + << " IN " << (format.hilite ? IAST::hilite_none : "") + << backQuoteIfNeed(storage_name); + formatOnCluster(format); if (!new_name.empty()) diff --git a/src/Parsers/Access/ASTCreateSettingsProfileQuery.h b/src/Parsers/Access/ASTCreateSettingsProfileQuery.h index c1a64998f29..be01aae1e26 100644 --- a/src/Parsers/Access/ASTCreateSettingsProfileQuery.h +++ b/src/Parsers/Access/ASTCreateSettingsProfileQuery.h @@ -28,6 +28,7 @@ public: bool if_exists = false; bool if_not_exists = false; bool or_replace = false; + String storage_name; Strings names; String new_name; diff --git a/src/Parsers/Access/ASTCreateUserQuery.cpp b/src/Parsers/Access/ASTCreateUserQuery.cpp index 0611545adf0..b7290c16013 100644 --- a/src/Parsers/Access/ASTCreateUserQuery.cpp +++ b/src/Parsers/Access/ASTCreateUserQuery.cpp @@ -208,6 +208,11 @@ void ASTCreateUserQuery::formatImpl(const FormatSettings & format, FormatState & format.ostr << " "; names->format(format); + if (!storage_name.empty()) + format.ostr << (format.hilite ? IAST::hilite_keyword : "") + << " IN " << (format.hilite ? IAST::hilite_none : "") + << backQuoteIfNeed(storage_name); + formatOnCluster(format); if (new_name) diff --git a/src/Parsers/Access/ASTCreateUserQuery.h b/src/Parsers/Access/ASTCreateUserQuery.h index 62ddbfd0040..06940e67d91 100644 --- a/src/Parsers/Access/ASTCreateUserQuery.h +++ b/src/Parsers/Access/ASTCreateUserQuery.h @@ -45,6 +45,7 @@ public: std::shared_ptr names; std::optional new_name; + String storage_name; std::shared_ptr auth_data; diff --git a/src/Parsers/Access/ASTDropAccessEntityQuery.cpp b/src/Parsers/Access/ASTDropAccessEntityQuery.cpp index 88f2d7bce63..bcd7105d0e9 100644 --- a/src/Parsers/Access/ASTDropAccessEntityQuery.cpp +++ b/src/Parsers/Access/ASTDropAccessEntityQuery.cpp @@ -53,6 +53,11 @@ void ASTDropAccessEntityQuery::formatImpl(const FormatSettings & settings, Forma else formatNames(names, settings); + if (!storage_name.empty()) + settings.ostr << (settings.hilite ? hilite_keyword : "") + << " FROM " << (settings.hilite ? hilite_none : "") + << backQuoteIfNeed(storage_name); + formatOnCluster(settings); } diff --git a/src/Parsers/Access/ASTDropAccessEntityQuery.h b/src/Parsers/Access/ASTDropAccessEntityQuery.h index f3a065c50df..32f4a8f8047 100644 --- a/src/Parsers/Access/ASTDropAccessEntityQuery.h +++ b/src/Parsers/Access/ASTDropAccessEntityQuery.h @@ -21,6 +21,7 @@ public: AccessEntityType type; bool if_exists = false; Strings names; + String storage_name; std::shared_ptr row_policy_names; String getID(char) const override; diff --git a/src/Parsers/Access/ASTMoveAccessEntityQuery.cpp b/src/Parsers/Access/ASTMoveAccessEntityQuery.cpp new file mode 100644 index 00000000000..285f07854c0 --- /dev/null +++ b/src/Parsers/Access/ASTMoveAccessEntityQuery.cpp @@ -0,0 +1,64 @@ +#include +#include +#include +#include + + +namespace DB +{ +namespace +{ + void formatNames(const Strings & names, const IAST::FormatSettings & settings) + { + bool need_comma = false; + for (const auto & name : names) + { + if (std::exchange(need_comma, true)) + settings.ostr << ','; + settings.ostr << ' ' << backQuoteIfNeed(name); + } + } +} + +String ASTMoveAccessEntityQuery::getID(char) const +{ + return String("MOVE ") + toString(type) + " query"; +} + +ASTPtr ASTMoveAccessEntityQuery::clone() const +{ + auto res = std::make_shared(*this); + + if (row_policy_names) + res->row_policy_names = std::static_pointer_cast(row_policy_names->clone()); + + return res; +} + +void ASTMoveAccessEntityQuery::formatImpl(const FormatSettings & settings, FormatState &, FormatStateStacked) const +{ + settings.ostr << (settings.hilite ? hilite_keyword : "") + << "MOVE " << AccessEntityTypeInfo::get(type).name + << (settings.hilite ? hilite_none : ""); + + if (type == AccessEntityType::ROW_POLICY) + { + settings.ostr << " "; + row_policy_names->format(settings); + } + else + formatNames(names, settings); + + settings.ostr << (settings.hilite ? hilite_keyword : "") + << " TO " << (settings.hilite ? hilite_none : "") + << backQuoteIfNeed(storage_name); + + formatOnCluster(settings); +} + +void ASTMoveAccessEntityQuery::replaceEmptyDatabase(const String & current_database) const +{ + if (row_policy_names) + row_policy_names->replaceEmptyDatabase(current_database); +} +} diff --git a/src/Parsers/Access/ASTMoveAccessEntityQuery.h b/src/Parsers/Access/ASTMoveAccessEntityQuery.h new file mode 100644 index 00000000000..aa2b3b0f98c --- /dev/null +++ b/src/Parsers/Access/ASTMoveAccessEntityQuery.h @@ -0,0 +1,32 @@ +#pragma once + +#include +#include +#include + + +namespace DB +{ +class ASTRowPolicyNames; + +/** MOVE {USER | ROLE | QUOTA | [ROW] POLICY | [SETTINGS] PROFILE} [IF EXISTS] name [,...] [ON [database.]table [,...]] TO storage_name + */ +class ASTMoveAccessEntityQuery : public IAST, public ASTQueryWithOnCluster +{ +public: + AccessEntityType type; + Strings names; + std::shared_ptr row_policy_names; + + String storage_name; + + String getID(char) const override; + ASTPtr clone() const override; + void formatImpl(const FormatSettings & settings, FormatState &, FormatStateStacked) const override; + ASTPtr getRewrittenASTWithoutOnCluster(const WithoutOnClusterASTRewriteParams &) const override { return removeOnCluster(clone()); } + + void replaceEmptyDatabase(const String & current_database) const; + + QueryKind getQueryKind() const override { return QueryKind::Move; } +}; +} diff --git a/src/Parsers/Access/ParserCreateQuotaQuery.cpp b/src/Parsers/Access/ParserCreateQuotaQuery.cpp index cd94fbb410f..bc3512e3fc0 100644 --- a/src/Parsers/Access/ParserCreateQuotaQuery.cpp +++ b/src/Parsers/Access/ParserCreateQuotaQuery.cpp @@ -1,4 +1,5 @@ #include +#include #include #include #include @@ -289,6 +290,7 @@ bool ParserCreateQuotaQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe std::optional key_type; std::vector all_limits; String cluster; + String storage_name; while (true) { @@ -311,6 +313,9 @@ bool ParserCreateQuotaQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe if (cluster.empty() && parseOnCluster(pos, expected, cluster)) continue; + if (storage_name.empty() && ParserKeyword{"IN"}.ignore(pos, expected) && parseAccessStorageName(pos, expected, storage_name)) + continue; + break; } @@ -333,6 +338,7 @@ bool ParserCreateQuotaQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe query->key_type = key_type; query->all_limits = std::move(all_limits); query->roles = std::move(roles); + query->storage_name = std::move(storage_name); return true; } diff --git a/src/Parsers/Access/ParserCreateRoleQuery.cpp b/src/Parsers/Access/ParserCreateRoleQuery.cpp index da9749958ee..99a97f6901d 100644 --- a/src/Parsers/Access/ParserCreateRoleQuery.cpp +++ b/src/Parsers/Access/ParserCreateRoleQuery.cpp @@ -1,3 +1,4 @@ +#include #include #include #include @@ -91,6 +92,7 @@ bool ParserCreateRoleQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec String new_name; std::shared_ptr settings; String cluster; + String storage_name; while (true) { @@ -110,6 +112,9 @@ bool ParserCreateRoleQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec if (cluster.empty() && parseOnCluster(pos, expected, cluster)) continue; + if (storage_name.empty() && ParserKeyword{"IN"}.ignore(pos, expected) && parseAccessStorageName(pos, expected, storage_name)) + continue; + break; } @@ -125,6 +130,7 @@ bool ParserCreateRoleQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec query->names = std::move(names); query->new_name = std::move(new_name); query->settings = std::move(settings); + query->storage_name = std::move(storage_name); return true; } diff --git a/src/Parsers/Access/ParserCreateRowPolicyQuery.cpp b/src/Parsers/Access/ParserCreateRowPolicyQuery.cpp index 2c25fc14e7d..f9e9466e35d 100644 --- a/src/Parsers/Access/ParserCreateRowPolicyQuery.cpp +++ b/src/Parsers/Access/ParserCreateRowPolicyQuery.cpp @@ -1,3 +1,4 @@ +#include #include #include #include @@ -245,6 +246,7 @@ bool ParserCreateRowPolicyQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & String new_short_name; std::optional is_restrictive; std::vector> filters; + String storage_name; while (true) { @@ -271,6 +273,9 @@ bool ParserCreateRowPolicyQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & if (cluster.empty() && parseOnCluster(pos, expected, cluster)) continue; + if (storage_name.empty() && ParserKeyword{"IN"}.ignore(pos, expected) && parseAccessStorageName(pos, expected, storage_name)) + continue; + break; } @@ -294,6 +299,7 @@ bool ParserCreateRowPolicyQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & query->is_restrictive = is_restrictive; query->filters = std::move(filters); query->roles = std::move(roles); + query->storage_name = std::move(storage_name); return true; } diff --git a/src/Parsers/Access/ParserCreateSettingsProfileQuery.cpp b/src/Parsers/Access/ParserCreateSettingsProfileQuery.cpp index c58a3035dc6..fcb6ebd8806 100644 --- a/src/Parsers/Access/ParserCreateSettingsProfileQuery.cpp +++ b/src/Parsers/Access/ParserCreateSettingsProfileQuery.cpp @@ -1,3 +1,4 @@ +#include #include #include #include @@ -111,6 +112,7 @@ bool ParserCreateSettingsProfileQuery::parseImpl(Pos & pos, ASTPtr & node, Expec String new_name; std::shared_ptr settings; String cluster; + String storage_name; while (true) { @@ -130,6 +132,9 @@ bool ParserCreateSettingsProfileQuery::parseImpl(Pos & pos, ASTPtr & node, Expec if (cluster.empty() && parseOnCluster(pos, expected, cluster)) continue; + if (storage_name.empty() && ParserKeyword{"IN"}.ignore(pos, expected) && parseAccessStorageName(pos, expected, storage_name)) + continue; + break; } @@ -152,6 +157,7 @@ bool ParserCreateSettingsProfileQuery::parseImpl(Pos & pos, ASTPtr & node, Expec query->new_name = std::move(new_name); query->settings = std::move(settings); query->to_roles = std::move(to_roles); + query->storage_name = std::move(storage_name); return true; } diff --git a/src/Parsers/Access/ParserCreateUserQuery.cpp b/src/Parsers/Access/ParserCreateUserQuery.cpp index 0344fb99c04..577b6772ab2 100644 --- a/src/Parsers/Access/ParserCreateUserQuery.cpp +++ b/src/Parsers/Access/ParserCreateUserQuery.cpp @@ -1,3 +1,4 @@ +#include #include #include #include @@ -414,6 +415,7 @@ bool ParserCreateUserQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec std::shared_ptr grantees; std::shared_ptr default_database; String cluster; + String storage_name; while (true) { @@ -480,6 +482,9 @@ bool ParserCreateUserQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec } } + if (storage_name.empty() && ParserKeyword{"IN"}.ignore(pos, expected) && parseAccessStorageName(pos, expected, storage_name)) + continue; + break; } @@ -514,6 +519,7 @@ bool ParserCreateUserQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expec query->settings = std::move(settings); query->grantees = std::move(grantees); query->default_database = std::move(default_database); + query->storage_name = std::move(storage_name); if (query->auth_data) query->children.push_back(query->auth_data); diff --git a/src/Parsers/Access/ParserDropAccessEntityQuery.cpp b/src/Parsers/Access/ParserDropAccessEntityQuery.cpp index d21164bc1a2..7fd34a6f8e4 100644 --- a/src/Parsers/Access/ParserDropAccessEntityQuery.cpp +++ b/src/Parsers/Access/ParserDropAccessEntityQuery.cpp @@ -1,3 +1,4 @@ +#include #include #include #include @@ -53,6 +54,7 @@ bool ParserDropAccessEntityQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & Strings names; std::shared_ptr row_policy_names; + String storage_name; String cluster; if ((type == AccessEntityType::USER) || (type == AccessEntityType::ROLE)) @@ -76,6 +78,9 @@ bool ParserDropAccessEntityQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & return false; } + if (ParserKeyword{"FROM"}.ignore(pos, expected)) + parseAccessStorageName(pos, expected, storage_name); + if (cluster.empty()) parseOnCluster(pos, expected, cluster); @@ -87,6 +92,7 @@ bool ParserDropAccessEntityQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & query->cluster = std::move(cluster); query->names = std::move(names); query->row_policy_names = std::move(row_policy_names); + query->storage_name = std::move(storage_name); return true; } diff --git a/src/Parsers/Access/ParserMoveAccessEntityQuery.cpp b/src/Parsers/Access/ParserMoveAccessEntityQuery.cpp new file mode 100644 index 00000000000..798e200bfb3 --- /dev/null +++ b/src/Parsers/Access/ParserMoveAccessEntityQuery.cpp @@ -0,0 +1,94 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace DB +{ +namespace +{ + bool parseEntityType(IParserBase::Pos & pos, Expected & expected, AccessEntityType & type) + { + for (auto i : collections::range(AccessEntityType::MAX)) + { + const auto & type_info = AccessEntityTypeInfo::get(i); + if (ParserKeyword{type_info.name}.ignore(pos, expected) + || (!type_info.alias.empty() && ParserKeyword{type_info.alias}.ignore(pos, expected))) + { + type = i; + return true; + } + } + return false; + } + + + bool parseOnCluster(IParserBase::Pos & pos, Expected & expected, String & cluster) + { + return IParserBase::wrapParseImpl(pos, [&] + { + return ParserKeyword{"ON"}.ignore(pos, expected) && ASTQueryWithOnCluster::parse(pos, cluster, expected); + }); + } +} + + +bool ParserMoveAccessEntityQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected) +{ + if (!ParserKeyword{"MOVE"}.ignore(pos, expected)) + return false; + + AccessEntityType type; + if (!parseEntityType(pos, expected, type)) + return false; + + Strings names; + std::shared_ptr row_policy_names; + String storage_name; + String cluster; + + if ((type == AccessEntityType::USER) || (type == AccessEntityType::ROLE)) + { + if (!parseUserNames(pos, expected, names)) + return false; + } + else if (type == AccessEntityType::ROW_POLICY) + { + ParserRowPolicyNames parser; + ASTPtr ast; + parser.allowOnCluster(); + if (!parser.parse(pos, ast, expected)) + return false; + row_policy_names = typeid_cast>(ast); + cluster = std::exchange(row_policy_names->cluster, ""); + } + else + { + if (!parseIdentifiersOrStringLiterals(pos, expected, names)) + return false; + } + + if (!ParserKeyword{"TO"}.ignore(pos, expected) || !parseAccessStorageName(pos, expected, storage_name)) + return false; + + if (cluster.empty()) + parseOnCluster(pos, expected, cluster); + + auto query = std::make_shared(); + node = query; + + query->type = type; + query->cluster = std::move(cluster); + query->names = std::move(names); + query->row_policy_names = std::move(row_policy_names); + query->storage_name = std::move(storage_name); + + return true; +} +} diff --git a/src/Parsers/Access/ParserMoveAccessEntityQuery.h b/src/Parsers/Access/ParserMoveAccessEntityQuery.h new file mode 100644 index 00000000000..3accec883dd --- /dev/null +++ b/src/Parsers/Access/ParserMoveAccessEntityQuery.h @@ -0,0 +1,17 @@ +#pragma once + +#include + + +namespace DB +{ +/** Parses queries like + * MOVE {USER | ROLE | QUOTA | [ROW] POLICY | [SETTINGS] PROFILE} [IF EXISTS] name [,...] [ON [database.]table [,...]] TO storage_name + */ +class ParserMoveAccessEntityQuery : public IParserBase +{ +protected: + const char * getName() const override { return "MOVE access entity query"; } + bool parseImpl(Pos & pos, ASTPtr & node, Expected & expected) override; +}; +} diff --git a/src/Parsers/Access/parseUserName.h b/src/Parsers/Access/parseUserName.h index fbad78e7bce..10c548d8826 100644 --- a/src/Parsers/Access/parseUserName.h +++ b/src/Parsers/Access/parseUserName.h @@ -2,6 +2,7 @@ #include #include +#include namespace DB diff --git a/src/Parsers/IAST.h b/src/Parsers/IAST.h index d217876459f..2b097025b07 100644 --- a/src/Parsers/IAST.h +++ b/src/Parsers/IAST.h @@ -302,6 +302,7 @@ public: Alter, Grant, Revoke, + Move, System, Set, Use, diff --git a/src/Parsers/ParserQuery.cpp b/src/Parsers/ParserQuery.cpp index 77e7b58e6b1..7ed69940bed 100644 --- a/src/Parsers/ParserQuery.cpp +++ b/src/Parsers/ParserQuery.cpp @@ -27,6 +27,7 @@ #include #include #include +#include #include @@ -54,6 +55,7 @@ bool ParserQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected) ParserCreateIndexQuery create_index_p; ParserDropIndexQuery drop_index_p; ParserDropAccessEntityQuery drop_access_entity_p; + ParserMoveAccessEntityQuery move_access_entity_p; ParserGrantQuery grant_p; ParserSetRoleQuery set_role_p; ParserExternalDDLQuery external_ddl_p; @@ -80,6 +82,7 @@ bool ParserQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected) || create_index_p.parse(pos, node, expected) || drop_index_p.parse(pos, node, expected) || drop_access_entity_p.parse(pos, node, expected) + || move_access_entity_p.parse(pos, node, expected) || grant_p.parse(pos, node, expected) || external_ddl_p.parse(pos, node, expected) || transaction_control_p.parse(pos, node, expected) diff --git a/src/Processors/Executors/StreamingFormatExecutor.cpp b/src/Processors/Executors/StreamingFormatExecutor.cpp index 2223721439e..46818989032 100644 --- a/src/Processors/Executors/StreamingFormatExecutor.cpp +++ b/src/Processors/Executors/StreamingFormatExecutor.cpp @@ -1,6 +1,5 @@ #include #include -#include namespace DB { diff --git a/src/Processors/Formats/IRowInputFormat.cpp b/src/Processors/Formats/IRowInputFormat.cpp index 8bcf293033b..7395d9b916a 100644 --- a/src/Processors/Formats/IRowInputFormat.cpp +++ b/src/Processors/Formats/IRowInputFormat.cpp @@ -86,7 +86,21 @@ void IRowInputFormat::logError() Chunk IRowInputFormat::generate() { if (total_rows == 0) - readPrefix(); + { + try + { + readPrefix(); + } + catch (Exception & e) + { + auto file_name = getFileNameFromReadBuffer(getReadBuffer()); + if (!file_name.empty()) + e.addMessage(fmt::format("(in file/uri {})", file_name)); + + e.addMessage("(while reading header)"); + throw; + } + } const Block & header = getPort().getHeader(); diff --git a/src/Processors/Formats/Impl/ODBCDriver2BlockOutputFormat.cpp b/src/Processors/Formats/Impl/ODBCDriver2BlockOutputFormat.cpp index 43294355f2f..c8c00086e8c 100644 --- a/src/Processors/Formats/Impl/ODBCDriver2BlockOutputFormat.cpp +++ b/src/Processors/Formats/Impl/ODBCDriver2BlockOutputFormat.cpp @@ -3,9 +3,6 @@ #include #include #include - - -#include #include diff --git a/src/Processors/IAccumulatingTransform.cpp b/src/Processors/IAccumulatingTransform.cpp index 00d5b2ee089..ea3c3c2c1b0 100644 --- a/src/Processors/IAccumulatingTransform.cpp +++ b/src/Processors/IAccumulatingTransform.cpp @@ -1,5 +1,4 @@ #include -#include namespace DB { diff --git a/src/Processors/ResizeProcessor.cpp b/src/Processors/ResizeProcessor.cpp index 8167fae9baf..3a8c6fb2bff 100644 --- a/src/Processors/ResizeProcessor.cpp +++ b/src/Processors/ResizeProcessor.cpp @@ -1,5 +1,4 @@ #include -#include namespace DB { diff --git a/src/Processors/Transforms/CountingTransform.cpp b/src/Processors/Transforms/CountingTransform.cpp index 646256d60c0..3dfb9fe178f 100644 --- a/src/Processors/Transforms/CountingTransform.cpp +++ b/src/Processors/Transforms/CountingTransform.cpp @@ -1,4 +1,3 @@ -#include #include #include diff --git a/src/Processors/Transforms/ExceptionKeepingTransform.cpp b/src/Processors/Transforms/ExceptionKeepingTransform.cpp index 3c40c078225..b50f66b0240 100644 --- a/src/Processors/Transforms/ExceptionKeepingTransform.cpp +++ b/src/Processors/Transforms/ExceptionKeepingTransform.cpp @@ -2,7 +2,6 @@ #include #include #include -#include namespace DB { diff --git a/src/Processors/Transforms/SquashingChunksTransform.cpp b/src/Processors/Transforms/SquashingChunksTransform.cpp index e89aec31655..7de9538e435 100644 --- a/src/Processors/Transforms/SquashingChunksTransform.cpp +++ b/src/Processors/Transforms/SquashingChunksTransform.cpp @@ -1,5 +1,4 @@ #include -#include namespace DB { diff --git a/src/Server/HTTP/HTTPServerResponse.h b/src/Server/HTTP/HTTPServerResponse.h index f5b7a70dc79..236a56e2323 100644 --- a/src/Server/HTTP/HTTPServerResponse.h +++ b/src/Server/HTTP/HTTPServerResponse.h @@ -5,9 +5,9 @@ #include #include -#include #include + namespace DB { diff --git a/src/Server/HTTPHandler.cpp b/src/Server/HTTPHandler.cpp index ff5690a3b07..10d5e7a0242 100644 --- a/src/Server/HTTPHandler.cpp +++ b/src/Server/HTTPHandler.cpp @@ -561,8 +561,7 @@ void HTTPHandler::processQuery( session->makeSessionContext(); } - auto client_info = session->getClientInfo(); - auto context = session->makeQueryContext(std::move(client_info)); + auto context = session->makeQueryContext(); /// This parameter is used to tune the behavior of output formats (such as Native) for compatibility. if (params.has("client_protocol_version")) diff --git a/src/Storages/MergeTree/GinIndexStore.cpp b/src/Storages/MergeTree/GinIndexStore.cpp index 91e831270d4..5b798ecc8a5 100644 --- a/src/Storages/MergeTree/GinIndexStore.cpp +++ b/src/Storages/MergeTree/GinIndexStore.cpp @@ -11,7 +11,6 @@ #include #include #include -#include #include #include diff --git a/src/Storages/MergeTree/MergeTreeIndexSet.cpp b/src/Storages/MergeTree/MergeTreeIndexSet.cpp index 05ea7d15603..eaa695c7f7a 100644 --- a/src/Storages/MergeTree/MergeTreeIndexSet.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexSet.cpp @@ -256,10 +256,6 @@ MergeTreeIndexConditionSet::MergeTreeIndexConditionSet( if (!key_columns.contains(name)) key_columns.insert(name); - ASTPtr ast_filter_node = buildFilterNode(query_info.query); - if (!ast_filter_node) - return; - if (context->getSettingsRef().allow_experimental_analyzer) { if (!query_info.filter_actions_dag) @@ -280,6 +276,10 @@ MergeTreeIndexConditionSet::MergeTreeIndexConditionSet( } else { + ASTPtr ast_filter_node = buildFilterNode(query_info.query); + if (!ast_filter_node) + return; + if (checkASTUseless(ast_filter_node)) return; @@ -457,11 +457,10 @@ const ActionsDAG::Node * MergeTreeIndexConditionSet::operatorFromDAG(const Actio if (arguments_size != 1) return nullptr; - auto bit_wrapper_function = FunctionFactory::instance().get("__bitWrapperFunc", context); - const auto & bit_wrapper_func_node = result_dag->addFunction(bit_wrapper_function, {arguments[0]}, {}); + const ActionsDAG::Node * argument = &traverseDAG(*arguments[0], result_dag, context, node_to_result_node); auto bit_swap_last_two_function = FunctionFactory::instance().get("__bitSwapLastTwo", context); - return &result_dag->addFunction(bit_swap_last_two_function, {&bit_wrapper_func_node}, {}); + return &result_dag->addFunction(bit_swap_last_two_function, {argument}, {}); } else if (function_name == "and" || function_name == "indexHint" || function_name == "or") { diff --git a/src/Storages/StorageFile.cpp b/src/Storages/StorageFile.cpp index e3908c75a58..5c03540de9a 100644 --- a/src/Storages/StorageFile.cpp +++ b/src/Storages/StorageFile.cpp @@ -396,9 +396,9 @@ std::unique_ptr createReadBuffer( throw Exception(ErrorCodes::CANNOT_COMPILE_REGEXP, "Cannot compile regex from glob ({}): {}", current_path, matcher->error()); - return reader->readFile([matcher = std::move(matcher)](const std::string & path) + return reader->readFile([my_matcher = std::move(matcher)](const std::string & path) { - return re2::RE2::FullMatch(path, *matcher); + return re2::RE2::FullMatch(path, *my_matcher); }); } else diff --git a/src/Storages/System/StorageSystemRoleGrants.cpp b/src/Storages/System/StorageSystemRoleGrants.cpp index cf5a24f88cd..8fcd0fd7cf5 100644 --- a/src/Storages/System/StorageSystemRoleGrants.cpp +++ b/src/Storages/System/StorageSystemRoleGrants.cpp @@ -3,6 +3,7 @@ #include #include #include +#include #include #include #include @@ -22,6 +23,7 @@ NamesAndTypesList StorageSystemRoleGrants::getNamesAndTypes() {"user_name", std::make_shared(std::make_shared())}, {"role_name", std::make_shared(std::make_shared())}, {"granted_role_name", std::make_shared()}, + {"granted_role_id", std::make_shared()}, {"granted_role_is_default", std::make_shared()}, {"with_admin_option", std::make_shared()}, }; @@ -45,12 +47,14 @@ void StorageSystemRoleGrants::fillData(MutableColumns & res_columns, ContextPtr auto & column_role_name = assert_cast(assert_cast(*res_columns[column_index]).getNestedColumn()); auto & column_role_name_null_map = assert_cast(*res_columns[column_index++]).getNullMapData(); auto & column_granted_role_name = assert_cast(*res_columns[column_index++]); + auto & column_granted_role_id = assert_cast(*res_columns[column_index++]).getData(); auto & column_is_default = assert_cast(*res_columns[column_index++]).getData(); auto & column_admin_option = assert_cast(*res_columns[column_index++]).getData(); auto add_row = [&](const String & grantee_name, AccessEntityType grantee_type, const String & granted_role_name, + const UUID & granted_role_id, bool is_default, bool with_admin_option) { @@ -72,6 +76,7 @@ void StorageSystemRoleGrants::fillData(MutableColumns & res_columns, ContextPtr assert(false); column_granted_role_name.insertData(granted_role_name.data(), granted_role_name.length()); + column_granted_role_id.push_back(granted_role_id.toUnderType()); column_is_default.push_back(is_default); column_admin_option.push_back(with_admin_option); }; @@ -90,7 +95,7 @@ void StorageSystemRoleGrants::fillData(MutableColumns & res_columns, ContextPtr continue; bool is_default = !default_roles || default_roles->match(role_id); - add_row(grantee_name, grantee_type, *role_name, is_default, element.admin_option); + add_row(grantee_name, grantee_type, *role_name, role_id, is_default, element.admin_option); } } }; diff --git a/src/Storages/fuzzers/mergetree_checksum_fuzzer.cpp b/src/Storages/fuzzers/mergetree_checksum_fuzzer.cpp index 9a5a68f09cc..0f2ce8a2e44 100644 --- a/src/Storages/fuzzers/mergetree_checksum_fuzzer.cpp +++ b/src/Storages/fuzzers/mergetree_checksum_fuzzer.cpp @@ -1,4 +1,3 @@ -#include #include #include diff --git a/tests/analyzer_tech_debt.txt b/tests/analyzer_tech_debt.txt index db21b86753a..15d46403da9 100644 --- a/tests/analyzer_tech_debt.txt +++ b/tests/analyzer_tech_debt.txt @@ -6,7 +6,6 @@ 00725_memory_tracking 00754_distributed_optimize_skip_select_on_unused_shards 00754_distributed_optimize_skip_select_on_unused_shards_with_prewhere -00838_unique_index 00927_asof_joins 00940_order_by_read_in_order_query_plan 00945_bloom_filter_index diff --git a/tests/ci/build_check.py b/tests/ci/build_check.py index 592c27c2c68..9e6370c2e6e 100644 --- a/tests/ci/build_check.py +++ b/tests/ci/build_check.py @@ -1,12 +1,13 @@ #!/usr/bin/env python3 +from pathlib import Path +from typing import List, Tuple import subprocess import logging import json import os import sys import time -from typing import List, Tuple from ci_config import CI_CONFIG, BuildConfig from docker_pull_helper import get_image_with_version @@ -18,18 +19,19 @@ from env_helper import ( S3_DOWNLOAD, TEMP_PATH, ) +from git_helper import Git, git_runner from pr_info import PRInfo from s3_helper import S3Helper from tee_popen import TeePopen from version_helper import ( ClickHouseVersion, - Git, get_version_from_repo, update_version_local, ) from clickhouse_helper import ( ClickHouseHelper, prepare_tests_results_for_clickhouse, + get_instance_type, ) from stopwatch import Stopwatch @@ -50,7 +52,7 @@ def _can_export_binaries(build_config: BuildConfig) -> bool: def get_packager_cmd( build_config: BuildConfig, packager_path: str, - output_path: str, + output_path: Path, build_version: str, image_version: str, official: bool, @@ -59,8 +61,8 @@ def get_packager_cmd( comp = build_config.compiler cmake_flags = "-DENABLE_CLICKHOUSE_SELF_EXTRACTING=1" cmd = ( - f"cd {packager_path} && CMAKE_FLAGS='{cmake_flags}' ./packager --output-dir={output_path} " - f"--package-type={package_type} --compiler={comp}" + f"cd {packager_path} && CMAKE_FLAGS='{cmake_flags}' ./packager " + f"--output-dir={output_path} --package-type={package_type} --compiler={comp}" ) if build_config.debug_build: @@ -78,6 +80,7 @@ def get_packager_cmd( cmd += " --additional-pkgs" cmd += f" --docker-image-version={image_version}" + cmd += " --with-profiler" cmd += f" --version={build_version}" if _can_export_binaries(build_config): @@ -90,13 +93,13 @@ def get_packager_cmd( def build_clickhouse( - packager_cmd: str, logs_path: str, build_output_path: str -) -> Tuple[str, bool]: - build_log_path = os.path.join(logs_path, BUILD_LOG_NAME) + packager_cmd: str, logs_path: Path, build_output_path: Path +) -> Tuple[Path, bool]: + build_log_path = logs_path / BUILD_LOG_NAME success = False with TeePopen(packager_cmd, build_log_path) as process: retcode = process.wait() - if os.path.exists(build_output_path): + if build_output_path.exists(): build_results = os.listdir(build_output_path) else: build_results = [] @@ -217,7 +220,7 @@ def upload_master_static_binaries( pr_info: PRInfo, build_config: BuildConfig, s3_helper: S3Helper, - build_output_path: str, + build_output_path: Path, ) -> None: """Upload binary artifacts to a static S3 links""" static_binary_name = build_config.static_binary_name @@ -229,7 +232,7 @@ def upload_master_static_binaries( return s3_path = "/".join((pr_info.base_ref, static_binary_name, "clickhouse")) - binary = os.path.join(build_output_path, "clickhouse") + binary = build_output_path / "clickhouse" url = s3_helper.upload_build_file_to_s3(binary, s3_path) print(f"::notice ::Binary static URL: {url}") @@ -242,8 +245,8 @@ def main(): build_config = CI_CONFIG.build_config[build_name] - if not os.path.exists(TEMP_PATH): - os.makedirs(TEMP_PATH) + temp_path = Path(TEMP_PATH) + os.makedirs(temp_path, exist_ok=True) pr_info = PRInfo() @@ -282,9 +285,8 @@ def main(): logging.info("Build short name %s", build_name) - build_output_path = os.path.join(TEMP_PATH, build_name) - if not os.path.exists(build_output_path): - os.makedirs(build_output_path) + build_output_path = temp_path / build_name + os.makedirs(build_output_path, exist_ok=True) packager_cmd = get_packager_cmd( build_config, @@ -297,9 +299,8 @@ def main(): logging.info("Going to run packager with %s", packager_cmd) - logs_path = os.path.join(TEMP_PATH, "build_log") - if not os.path.exists(logs_path): - os.makedirs(logs_path) + logs_path = temp_path / "build_log" + os.makedirs(logs_path, exist_ok=True) start = time.time() log_path, success = build_clickhouse(packager_cmd, logs_path, build_output_path) @@ -320,8 +321,8 @@ def main(): # FIXME performance performance_urls = [] - performance_path = os.path.join(build_output_path, "performance.tar.zst") - if os.path.exists(performance_path): + performance_path = build_output_path / "performance.tar.zst" + if performance_path.exists(): performance_urls.append( s3_helper.upload_build_file_to_s3(performance_path, s3_performance_path) ) @@ -344,9 +345,9 @@ def main(): print("::notice ::Build URLs: {}".format("\n".join(build_urls))) - if os.path.exists(log_path): + if log_path.exists(): log_url = s3_helper.upload_build_file_to_s3( - log_path, s3_path_prefix + "/" + os.path.basename(log_path) + log_path, s3_path_prefix + "/" + log_path.name ) logging.info("Log url %s", log_url) else: @@ -360,7 +361,79 @@ def main(): upload_master_static_binaries(pr_info, build_config, s3_helper, build_output_path) + # Upload profile data ch_helper = ClickHouseHelper() + + clickhouse_ci_logs_host = os.getenv("CLICKHOUSE_CI_LOGS_HOST", "") + if clickhouse_ci_logs_host: + instance_type = get_instance_type() + query = f"""INSERT INTO build_time_trace +( + pull_request_number, + commit_sha, + check_start_time, + check_name, + instance_type, + file, + library, + time, + pid, + tid, + ph, + ts, + dur, + cat, + name, + detail, + count, + avgMs, + args_name +) +SELECT {pr_info.number}, '{pr_info.sha}', '{stopwatch.start_time_str}', '{build_name}', '{instance_type}', * +FROM input(' + file String, + library String, + time DateTime64(6), + pid UInt32, + tid UInt32, + ph String, + ts UInt64, + dur UInt64, + cat String, + name String, + detail String, + count UInt64, + avgMs UInt64, + args_name String') +FORMAT JSONCompactEachRow""" + + auth = { + "X-ClickHouse-User": "ci", + "X-ClickHouse-Key": os.getenv("CLICKHOUSE_CI_LOGS_PASSWORD", ""), + } + url = f"https://{clickhouse_ci_logs_host}/" + profiles_dir = temp_path / "profiles_source" + os.makedirs(profiles_dir, exist_ok=True) + logging.info("Processing profile JSON files from {GIT_REPO_ROOT}/build_docker") + git_runner( + "./utils/prepare-time-trace/prepare-time-trace.sh " + f"build_docker {profiles_dir.absolute()}" + ) + profile_data_file = temp_path / "profile.json" + with open(profile_data_file, "wb") as profile_fd: + for profile_sourse in os.listdir(profiles_dir): + with open(profiles_dir / profile_sourse, "rb") as ps_fd: + profile_fd.write(ps_fd.read()) + + logging.info( + "::notice ::Log Uploading profile data, path: %s, size: %s, query: %s", + profile_data_file, + profile_data_file.stat().st_size, + query, + ) + ch_helper.insert_file(url, auth, query, profile_data_file) + + # Upload statistics to CI database prepared_events = prepare_tests_results_for_clickhouse( pr_info, [], diff --git a/tests/ci/clickhouse_helper.py b/tests/ci/clickhouse_helper.py index 7f98f6e52d2..b165cda37e8 100644 --- a/tests/ci/clickhouse_helper.py +++ b/tests/ci/clickhouse_helper.py @@ -1,5 +1,6 @@ #!/usr/bin/env python3 -from typing import List +from pathlib import Path +from typing import Dict, List, Optional import json import logging import time @@ -16,30 +17,60 @@ class InsertException(Exception): class ClickHouseHelper: - def __init__(self, url=None): + def __init__( + self, url: Optional[str] = None, auth: Optional[Dict[str, str]] = None + ): if url is None: url = get_parameter_from_ssm("clickhouse-test-stat-url") self.url = url - self.auth = { + self.auth = auth or { "X-ClickHouse-User": get_parameter_from_ssm("clickhouse-test-stat-login"), "X-ClickHouse-Key": get_parameter_from_ssm("clickhouse-test-stat-password"), } @staticmethod - def _insert_json_str_info_impl(url, auth, db, table, json_str): + def insert_file( + url: str, + auth: Optional[Dict[str, str]], + query: str, + file: Path, + additional_options: Optional[Dict[str, str]] = None, + ) -> None: + params = { + "query": query, + "date_time_input_format": "best_effort", + "send_logs_level": "warning", + } + if additional_options: + for k, v in additional_options.items(): + params[k] = v + + with open(file, "rb") as data_fd: + ClickHouseHelper._insert_post( + url, params=params, data=data_fd, headers=auth + ) + + @staticmethod + def insert_json_str(url, auth, db, table, json_str): params = { "database": db, "query": f"INSERT INTO {table} FORMAT JSONEachRow", "date_time_input_format": "best_effort", "send_logs_level": "warning", } + ClickHouseHelper._insert_post(url, params=params, data=json_str, headers=auth) + + @staticmethod + def _insert_post(*args, **kwargs): + url = "" + if args: + url = args[0] + url = kwargs.get("url", url) for i in range(5): try: - response = requests.post( - url, params=params, data=json_str, headers=auth - ) + response = requests.post(*args, **kwargs) except Exception as e: error = f"Received exception while sending data to {url} on {i} attempt: {e}" logging.warning(error) @@ -51,13 +82,8 @@ class ClickHouseHelper: break error = ( - "Cannot insert data into clickhouse at try " - + str(i) - + ": HTTP code " - + str(response.status_code) - + ": '" - + str(response.text) - + "'" + f"Cannot insert data into clickhouse at try {i}: HTTP code " + f"{response.status_code}: '{response.text}'" ) if response.status_code >= 500: @@ -76,7 +102,7 @@ class ClickHouseHelper: raise InsertException(error) def _insert_json_str_info(self, db, table, json_str): - self._insert_json_str_info_impl(self.url, self.auth, db, table, json_str) + self.insert_json_str(self.url, self.auth, db, table, json_str) def insert_event_into(self, db, table, event, safe=True): event_str = json.dumps(event) diff --git a/tests/ci/stress.py b/tests/ci/stress.py index 60ddeb85d35..eb829cf519c 100755 --- a/tests/ci/stress.py +++ b/tests/ci/stress.py @@ -202,13 +202,14 @@ def prepare_for_hung_check(drop_databases): call( make_query_command( """ - select sleepEachRow(( - select maxOrDefault(300 - elapsed) + 1 - from system.processes - where query not like '%from system.processes%' and elapsed < 300 + SELECT sleepEachRow(( + SELECT maxOrDefault(300 - elapsed) + 1 + FROM system.processes + WHERE query NOT LIKE '%FROM system.processes%' AND elapsed < 300 ) / 300) - from numbers(300) - format Null + FROM numbers(300) + FORMAT Null + SETTINGS function_sleep_max_microseconds_per_block = 0 """ ), shell=True, diff --git a/tests/integration/test_cluster_copier/test_two_nodes.py b/tests/integration/test_cluster_copier/test_two_nodes.py index 2b6fcf6cac2..1bd3561f24f 100644 --- a/tests/integration/test_cluster_copier/test_two_nodes.py +++ b/tests/integration/test_cluster_copier/test_two_nodes.py @@ -595,24 +595,3 @@ def execute_task(started_cluster, task, cmd_options): task.check() finally: zk.delete(task.zk_task_path, recursive=True) - - -# Tests -@pytest.mark.skip(reason="Too flaky :(") -def test_different_schema(started_cluster): - execute_task(started_cluster, TaskWithDifferentSchema(started_cluster), []) - - -@pytest.mark.skip(reason="Too flaky :(") -def test_ttl_columns(started_cluster): - execute_task(started_cluster, TaskTTL(started_cluster), []) - - -@pytest.mark.skip(reason="Too flaky :(") -def test_skip_index(started_cluster): - execute_task(started_cluster, TaskSkipIndex(started_cluster), []) - - -@pytest.mark.skip(reason="Too flaky :(") -def test_ttl_move_to_volume(started_cluster): - execute_task(started_cluster, TaskTTLMoveToVolume(started_cluster), []) diff --git a/tests/integration/test_dictionaries_all_layouts_separate_sources/test_cassandra.py b/tests/integration/test_dictionaries_all_layouts_separate_sources/test_cassandra.py deleted file mode 100644 index 90287e19bd0..00000000000 --- a/tests/integration/test_dictionaries_all_layouts_separate_sources/test_cassandra.py +++ /dev/null @@ -1,99 +0,0 @@ -import os -import math -import pytest - -# FIXME This test is too flaky -# https://github.com/ClickHouse/ClickHouse/issues/33006 - -pytestmark = pytest.mark.skip - -from .common import * - -from helpers.cluster import ClickHouseCluster -from helpers.dictionary import Field, Row, Dictionary, DictionaryStructure, Layout -from helpers.external_sources import SourceCassandra - -SOURCE = None -cluster = None -node = None -simple_tester = None -complex_tester = None -ranged_tester = None -test_name = "cassandra" - - -def setup_module(module): - global cluster - global node - global simple_tester - global complex_tester - global ranged_tester - - cluster = ClickHouseCluster(__file__) - - SOURCE = SourceCassandra( - "Cassandra", - None, - cluster.cassandra_port, - cluster.cassandra_host, - cluster.cassandra_port, - "", - "", - ) - - simple_tester = SimpleLayoutTester(test_name) - simple_tester.cleanup() - simple_tester.create_dictionaries(SOURCE) - - complex_tester = ComplexLayoutTester(test_name) - complex_tester.create_dictionaries(SOURCE) - - ranged_tester = RangedLayoutTester(test_name) - ranged_tester.create_dictionaries(SOURCE) - # Since that all .xml configs were created - - main_configs = [] - main_configs.append(os.path.join("configs", "disable_ssl_verification.xml")) - - dictionaries = simple_tester.list_dictionaries() - - node = cluster.add_instance( - "cass_node", - main_configs=main_configs, - dictionaries=dictionaries, - with_cassandra=True, - ) - - -def teardown_module(module): - simple_tester.cleanup() - - -@pytest.fixture(scope="module") -def started_cluster(): - try: - cluster.start() - - simple_tester.prepare(cluster) - complex_tester.prepare(cluster) - ranged_tester.prepare(cluster) - - yield cluster - - finally: - cluster.shutdown() - - -@pytest.mark.parametrize("layout_name", sorted(LAYOUTS_SIMPLE)) -def test_simple(started_cluster, layout_name): - simple_tester.execute(layout_name, node) - - -@pytest.mark.parametrize("layout_name", sorted(LAYOUTS_COMPLEX)) -def test_complex(started_cluster, layout_name): - complex_tester.execute(layout_name, node) - - -@pytest.mark.parametrize("layout_name", sorted(LAYOUTS_RANGED)) -def test_ranged(started_cluster, layout_name): - ranged_tester.execute(layout_name, node) diff --git a/tests/integration/test_dictionaries_complex_key_cache_string/test.py b/tests/integration/test_dictionaries_complex_key_cache_string/test.py index ae9cd4e7c91..bca7122a2a8 100644 --- a/tests/integration/test_dictionaries_complex_key_cache_string/test.py +++ b/tests/integration/test_dictionaries_complex_key_cache_string/test.py @@ -27,71 +27,3 @@ def started_cluster(): yield cluster finally: cluster.shutdown() - - -@pytest.mark.skip(reason="SSD cache test can run on disk only") -@pytest.mark.parametrize("type", ["memory", "ssd"]) -def test_memory_consumption(started_cluster, type): - node = started_cluster.instances[f"node_{type}"] - node.query( - "insert into radars_table select toString(rand() % 5000), '{0}', '{0}' from numbers(1000)".format( - "w" * 8 - ) - ) - node.query( - "insert into radars_table select toString(rand() % 5000), '{0}', '{0}' from numbers(1000)".format( - "x" * 16 - ) - ) - node.query( - "insert into radars_table select toString(rand() % 5000), '{0}', '{0}' from numbers(1000)".format( - "y" * 32 - ) - ) - node.query( - "insert into radars_table select toString(rand() % 5000), '{0}', '{0}' from numbers(1000)".format( - "z" * 64 - ) - ) - - # Fill dictionary - node.query( - "select dictGetString('radars', 'client_id', tuple(toString(number))) from numbers(0, 5000)" - ) - - allocated_first = int( - node.query( - "select bytes_allocated from system.dictionaries where name = 'radars'" - ).strip() - ) - - alloc_array = [] - for i in range(5): - node.query( - "select dictGetString('radars', 'client_id', tuple(toString(number))) from numbers(0, 5000)" - ) - - allocated = int( - node.query( - "select bytes_allocated from system.dictionaries where name = 'radars'" - ).strip() - ) - alloc_array.append(allocated) - - # size doesn't grow - assert all(allocated_first >= a for a in alloc_array) - - for i in range(5): - node.query( - "select dictGetString('radars', 'client_id', tuple(toString(number))) from numbers(0, 5000)" - ) - - allocated = int( - node.query( - "select bytes_allocated from system.dictionaries where name = 'radars'" - ).strip() - ) - alloc_array.append(allocated) - - # size doesn't grow - assert all(allocated_first >= a for a in alloc_array) diff --git a/tests/integration/test_dictionary_allow_read_expired_keys/test_default_reading.py b/tests/integration/test_dictionary_allow_read_expired_keys/test_default_reading.py index 85c45d5df3c..921fb4e4154 100644 --- a/tests/integration/test_dictionary_allow_read_expired_keys/test_default_reading.py +++ b/tests/integration/test_dictionary_allow_read_expired_keys/test_default_reading.py @@ -38,7 +38,6 @@ def started_cluster(): cluster.shutdown() -# @pytest.mark.skip(reason="debugging") def test_default_reading(started_cluster): assert None != dictionary_node.get_process_pid( "clickhouse" diff --git a/tests/integration/test_dictionary_allow_read_expired_keys/test_default_string.py b/tests/integration/test_dictionary_allow_read_expired_keys/test_default_string.py index 92d681698bc..6eedf63f95c 100644 --- a/tests/integration/test_dictionary_allow_read_expired_keys/test_default_string.py +++ b/tests/integration/test_dictionary_allow_read_expired_keys/test_default_string.py @@ -53,7 +53,6 @@ def started_cluster(): cluster.shutdown() -# @pytest.mark.skip(reason="debugging") def test_return_real_values(started_cluster): assert None != dictionary_node.get_process_pid( "clickhouse" diff --git a/tests/integration/test_dictionary_allow_read_expired_keys/test_dict_get.py b/tests/integration/test_dictionary_allow_read_expired_keys/test_dict_get.py index 05f638ec337..8e45af44640 100644 --- a/tests/integration/test_dictionary_allow_read_expired_keys/test_dict_get.py +++ b/tests/integration/test_dictionary_allow_read_expired_keys/test_dict_get.py @@ -38,7 +38,6 @@ def started_cluster(): cluster.shutdown() -# @pytest.mark.skip(reason="debugging") def test_simple_dict_get(started_cluster): assert None != dictionary_node.get_process_pid( "clickhouse" diff --git a/tests/integration/test_dictionary_allow_read_expired_keys/test_dict_get_or_default.py b/tests/integration/test_dictionary_allow_read_expired_keys/test_dict_get_or_default.py index 1da8fd3325a..8d6242f4711 100644 --- a/tests/integration/test_dictionary_allow_read_expired_keys/test_dict_get_or_default.py +++ b/tests/integration/test_dictionary_allow_read_expired_keys/test_dict_get_or_default.py @@ -38,7 +38,6 @@ def started_cluster(): cluster.shutdown() -# @pytest.mark.skip(reason="debugging") def test_simple_dict_get_or_default(started_cluster): assert None != dictionary_node.get_process_pid( "clickhouse" diff --git a/tests/integration/test_disabled_access_control_improvements/test_row_policy.py b/tests/integration/test_disabled_access_control_improvements/test_row_policy.py index 509b4de1a37..b620e88e7eb 100644 --- a/tests/integration/test_disabled_access_control_improvements/test_row_policy.py +++ b/tests/integration/test_disabled_access_control_improvements/test_row_policy.py @@ -76,7 +76,7 @@ def test_introspection(): "mydb", "filtered_table1", "6068883a-0e9d-f802-7e22-0144f8e66d3c", - "users.xml", + "users_xml", "1", 0, 0, @@ -89,7 +89,7 @@ def test_introspection(): "mydb", "filtered_table2", "c019e957-c60b-d54e-cc52-7c90dac5fb01", - "users.xml", + "users_xml", "1", 0, 0, @@ -102,7 +102,7 @@ def test_introspection(): "mydb", "filtered_table3", "4cb080d0-44e8-dbef-6026-346655143628", - "users.xml", + "users_xml", "1", 0, 0, @@ -115,7 +115,7 @@ def test_introspection(): "mydb", "local", "5b23c389-7e18-06bf-a6bc-dd1afbbc0a97", - "users.xml", + "users_xml", "a = 1", 0, 0, @@ -128,7 +128,7 @@ def test_introspection(): "mydb", "filtered_table1", "9e8a8f62-4965-2b5e-8599-57c7b99b3549", - "users.xml", + "users_xml", "a = 1", 0, 0, @@ -141,7 +141,7 @@ def test_introspection(): "mydb", "filtered_table2", "cffae79d-b9bf-a2ef-b798-019c18470b25", - "users.xml", + "users_xml", "a + b < 1 or c - d > 5", 0, 0, @@ -154,7 +154,7 @@ def test_introspection(): "mydb", "filtered_table3", "12fc5cef-e3da-3940-ec79-d8be3911f42b", - "users.xml", + "users_xml", "c = 1", 0, 0, @@ -167,7 +167,7 @@ def test_introspection(): "mydb", "local", "cdacaeb5-1d97-f99d-2bb0-4574f290629c", - "users.xml", + "users_xml", "1", 0, 0, diff --git a/tests/integration/test_extreme_deduplication/test.py b/tests/integration/test_extreme_deduplication/test.py index 03451933171..3632369154a 100644 --- a/tests/integration/test_extreme_deduplication/test.py +++ b/tests/integration/test_extreme_deduplication/test.py @@ -71,73 +71,3 @@ def test_deduplication_window_in_seconds(started_cluster): assert TSV.toMat(node.query("SELECT count() FROM simple"))[0][0] == "3" node1.query("""DROP TABLE simple ON CLUSTER test_cluster""") - - -# Currently this test just reproduce incorrect behavior that sould be fixed -@pytest.mark.skip(reason="Flapping test") -def test_deduplication_works_in_case_of_intensive_inserts(started_cluster): - inserters = [] - fetchers = [] - - node1.query( - """ - CREATE TABLE simple ON CLUSTER test_cluster (date Date, id UInt32) - ENGINE = ReplicatedMergeTree('/clickhouse/tables/{shard}/simple', '{replica}') PARTITION BY toYYYYMM(date) ORDER BY id""" - ) - - node1.query("INSERT INTO simple VALUES (0, 0)") - - for node in nodes: - host = node.ip_address - - inserters.append( - CommandRequest( - ["/bin/bash"], - timeout=10, - stdin=""" -set -e -for i in `seq 1000`; do - {} --host {} -q "INSERT INTO simple VALUES (0, 0)" -done -""".format( - cluster.get_client_cmd(), host - ), - ) - ) - - fetchers.append( - CommandRequest( - ["/bin/bash"], - timeout=10, - stdin=""" -set -e -for i in `seq 1000`; do - res=`{} --host {} -q "SELECT count() FROM simple"` - if [[ $? -ne 0 || $res -ne 1 ]]; then - echo "Selected $res elements! Host: {}" 1>&2 - exit -1 - fi; -done -""".format( - cluster.get_client_cmd(), host, node.name - ), - ) - ) - - # There were not errors during INSERTs - for inserter in inserters: - try: - inserter.get_answer() - except QueryTimeoutExceedException: - # Only timeout is accepted - pass - - # There were not errors during SELECTs - for fetcher in fetchers: - try: - fetcher.get_answer() - except QueryTimeoutExceedException: - # Only timeout is accepted - pass - - node1.query("""DROP TABLE simple ON CLUSTER test_cluster""") diff --git a/tests/integration/test_grant_and_revoke/test.py b/tests/integration/test_grant_and_revoke/test.py index 8a8eb11d91f..0bbaa69691e 100644 --- a/tests/integration/test_grant_and_revoke/test.py +++ b/tests/integration/test_grant_and_revoke/test.py @@ -471,7 +471,7 @@ def test_introspection(): [ [ "A", - "local directory", + "local_directory", "no_password", "{}", "['::/0']", @@ -484,7 +484,7 @@ def test_introspection(): ], [ "B", - "local directory", + "local_directory", "no_password", "{}", "['::/0']", diff --git a/tests/integration/test_hive_query/configs/config.xml b/tests/integration/test_hive_query/configs/config.xml deleted file mode 100644 index 7de1391e56c..00000000000 --- a/tests/integration/test_hive_query/configs/config.xml +++ /dev/null @@ -1,32 +0,0 @@ - - - - - - - localhost - 9000 - - - - - - - true - /tmp/clickhouse_local_cache,/tmp/clickhouse_local_cache1 - 207374182400 - 1048576 - - - - /etc/clickhouse-server/extra_conf.d/hdfs-site.xml - - - - system - metric_log
- 1000 - 1000 - - - diff --git a/tests/integration/test_hive_query/configs/hdfs-site.xml b/tests/integration/test_hive_query/configs/hdfs-site.xml deleted file mode 100644 index 82c525ea414..00000000000 --- a/tests/integration/test_hive_query/configs/hdfs-site.xml +++ /dev/null @@ -1,6 +0,0 @@ - - - dfs.replication - 1 - - diff --git a/tests/integration/test_hive_query/data/prepare_hive_data.sh b/tests/integration/test_hive_query/data/prepare_hive_data.sh deleted file mode 100755 index 495ea201870..00000000000 --- a/tests/integration/test_hive_query/data/prepare_hive_data.sh +++ /dev/null @@ -1,15 +0,0 @@ -#!/bin/bash -hive -e "create database test" - -hive -e "drop table if exists test.demo; create table test.demo(id string, score int) PARTITIONED BY(day string) ROW FORMAT SERDE 'org.apache.hadoop.hive.ql.io.parquet.serde.ParquetHiveSerDe' STORED AS INPUTFORMAT 'org.apache.hadoop.hive.ql.io.parquet.MapredParquetInputFormat' OUTPUTFORMAT 'org.apache.hadoop.hive.ql.io.parquet.MapredParquetOutputFormat'; create table test.demo_orc(id string, score int) PARTITIONED BY(day string) ROW FORMAT SERDE 'org.apache.hadoop.hive.ql.io.orc.OrcSerde' STORED AS INPUTFORMAT 'org.apache.hadoop.hive.ql.io.orc.OrcInputFormat' OUTPUTFORMAT 'org.apache.hadoop.hive.ql.io.orc.OrcOutputFormat'; " -hive -e "drop table if exists test.parquet_demo; create table test.parquet_demo(id string, score int) PARTITIONED BY(day string, hour string) ROW FORMAT SERDE 'org.apache.hadoop.hive.ql.io.parquet.serde.ParquetHiveSerDe' STORED AS INPUTFORMAT 'org.apache.hadoop.hive.ql.io.parquet.MapredParquetInputFormat' OUTPUTFORMAT 'org.apache.hadoop.hive.ql.io.parquet.MapredParquetOutputFormat'" -hive -e "drop table if exists test.demo_text; create table test.demo_text(id string, score int, day string)row format delimited fields terminated by ','; load data local inpath '/demo_data.txt' into table test.demo_text " -hive -e "set hive.exec.dynamic.partition.mode=nonstrict;insert into test.demo partition(day) select * from test.demo_text; insert into test.demo_orc partition(day) select * from test.demo_text" - -hive -e "set hive.exec.dynamic.partition.mode=nonstrict;insert into test.parquet_demo partition(day, hour) select id, score, day, '00' as hour from test.demo;" -hive -e "set hive.exec.dynamic.partition.mode=nonstrict;insert into test.parquet_demo partition(day, hour) select id, score, day, '01' as hour from test.demo;" - -hive -e "drop table if exists test.test_hive_types; CREATE TABLE test.test_hive_types( f_tinyint tinyint, f_smallint smallint, f_int int, f_integer int, f_bigint bigint, f_float float, f_double double, f_decimal decimal(10,0), f_timestamp timestamp, f_date date, f_string string, f_varchar varchar(100), f_char char(100), f_bool boolean, f_array_int array, f_array_string array, f_array_float array, f_map_int map, f_map_string map, f_map_float map, f_struct struct>) PARTITIONED BY( day string) ROW FORMAT SERDE 'org.apache.hadoop.hive.ql.io.parquet.serde.ParquetHiveSerDe' STORED AS INPUTFORMAT 'org.apache.hadoop.hive.ql.io.parquet.MapredParquetInputFormat' OUTPUTFORMAT 'org.apache.hadoop.hive.ql.io.parquet.MapredParquetOutputFormat';" - -hive -e "insert into test.test_hive_types partition(day='2022-02-20') select 1, 2, 3, 4, 5, 6.11, 7.22, 8.333, '2022-02-20 14:47:04', '2022-02-20', 'hello world', 'hello world', 'hello world', true, array(1,2,3), array('hello world', 'hello world'), array(float(1.1),float(1.2)), map('a', 100, 'b', 200, 'c', 300), map('a', 'aa', 'b', 'bb', 'c', 'cc'), map('a', float(111.1), 'b', float(222.2), 'c', float(333.3)), named_struct('a', 'aaa', 'b', 200, 'c', float(333.3), 'd', named_struct('x', 10, 'y', 'xyz')); insert into test.test_hive_types partition(day='2022-02-19') select 1, 2, 3, 4, 5, 6.11, 7.22, 8.333, '2022-02-19 14:47:04', '2022-02-19', 'hello world', 'hello world', 'hello world', true, array(1,2,3), array('hello world', 'hello world'), array(float(1.1),float(1.2)), map('a', 100, 'b', 200, 'c', 300), map('a', 'aa', 'b', 'bb', 'c', 'cc'), map('a', float(111.1), 'b', float(222.2), 'c', float(333.3)), named_struct('a', 'aaa', 'b', 200, 'c', float(333.3), 'd', named_struct('x', 11, 'y', 'abc'));" - diff --git a/tests/integration/test_hive_query/test.py b/tests/integration/test_hive_query/test.py deleted file mode 100644 index 791ae03f9f6..00000000000 --- a/tests/integration/test_hive_query/test.py +++ /dev/null @@ -1,498 +0,0 @@ -import pytest - -# FIXME This test is too flaky -# https://github.com/ClickHouse/ClickHouse/issues/43541 - -pytestmark = pytest.mark.skip - -import logging -import os - -import time -from helpers.cluster import ClickHouseCluster -from helpers.test_tools import TSV - -logging.getLogger().setLevel(logging.INFO) -logging.getLogger().addHandler(logging.StreamHandler()) - -SCRIPT_DIR = os.path.dirname(os.path.realpath(__file__)) - - -@pytest.fixture(scope="module") -def started_cluster(): - try: - cluster = ClickHouseCluster(__file__) - cluster.add_instance( - "h0_0_0", - main_configs=["configs/config.xml"], - extra_configs=["configs/hdfs-site.xml", "data/prepare_hive_data.sh"], - with_hive=True, - ) - - logging.info("Starting cluster ...") - cluster.start() - cluster.copy_file_to_container( - "roottesthivequery_hdfs1_1", - "/ClickHouse/tests/integration/test_hive_query/data/prepare_hive_data.sh", - "/prepare_hive_data.sh", - ) - cluster.exec_in_container( - "roottesthivequery_hdfs1_1", ["bash", "-c", "bash /prepare_hive_data.sh"] - ) - yield cluster - finally: - cluster.shutdown() - - -def test_create_parquet_table(started_cluster): - logging.info("Start testing creating hive table ...") - node = started_cluster.instances["h0_0_0"] - test_passed = False - for i in range(10): - node.query("set input_format_parquet_allow_missing_columns = true") - result = node.query( - """ -DROP TABLE IF EXISTS default.demo_parquet; -CREATE TABLE default.demo_parquet (`id` Nullable(String), `score` Nullable(Int32), `day` Nullable(String)) ENGINE = Hive('thrift://hivetest:9083', 'test', 'demo') PARTITION BY(day) - """ - ) - logging.info("create result {}".format(result)) - if result.strip() == "": - test_passed = True - break - time.sleep(60) - assert test_passed - - -def test_create_parquet_table_1(started_cluster): - logging.info("Start testing creating hive table ...") - node = started_cluster.instances["h0_0_0"] - for i in range(10): - node.query("set input_format_parquet_allow_missing_columns = true") - result = node.query( - """ -DROP TABLE IF EXISTS default.demo_parquet_parts; -CREATE TABLE default.demo_parquet_parts (`id` Nullable(String), `score` Nullable(Int32), `day` Nullable(String), `hour` String) ENGINE = Hive('thrift://hivetest:9083', 'test', 'parquet_demo') PARTITION BY(day, hour); - """ - ) - logging.info("create result {}".format(result)) - if result.strip() == "": - test_passed = True - break - time.sleep(60) - assert test_passed - - -def test_create_orc_table(started_cluster): - logging.info("Start testing creating hive table ...") - node = started_cluster.instances["h0_0_0"] - test_passed = False - for i in range(10): - result = node.query( - """ - DROP TABLE IF EXISTS default.demo_orc; - CREATE TABLE default.demo_orc (`id` Nullable(String), `score` Nullable(Int32), `day` Nullable(String)) ENGINE = Hive('thrift://hivetest:9083', 'test', 'demo_orc') PARTITION BY(day) - """ - ) - logging.info("create result {}".format(result)) - if result.strip() == "": - test_passed = True - break - time.sleep(60) - - assert test_passed - - -def test_create_text_table(started_cluster): - logging.info("Start testing creating hive table ...") - node = started_cluster.instances["h0_0_0"] - result = node.query( - """ - DROP TABLE IF EXISTS default.demo_text; - CREATE TABLE default.demo_text (`id` Nullable(String), `score` Nullable(Int32), `day` Nullable(String)) ENGINE = Hive('thrift://hivetest:9083', 'test', 'demo_text') PARTITION BY (tuple()) - """ - ) - logging.info("create result {}".format(result)) - - assert result.strip() == "" - - -def test_parquet_groupby(started_cluster): - logging.info("Start testing groupby ...") - node = started_cluster.instances["h0_0_0"] - result = node.query( - """ - SELECT day, count(*) FROM default.demo_parquet group by day order by day - """ - ) - expected_result = """2021-11-01 1 -2021-11-05 2 -2021-11-11 1 -2021-11-16 2 -""" - assert result == expected_result - - -def test_parquet_in_filter(started_cluster): - logging.info("Start testing groupby ...") - node = started_cluster.instances["h0_0_0"] - result = node.query( - """ - SELECT count(*) FROM default.demo_parquet_parts where day = '2021-11-05' and hour in ('00') - """ - ) - expected_result = """2 -""" - logging.info("query result:{}".format(result)) - assert result == expected_result - - -def test_orc_groupby(started_cluster): - logging.info("Start testing groupby ...") - node = started_cluster.instances["h0_0_0"] - result = node.query( - """ - DROP TABLE IF EXISTS default.demo_orc; - CREATE TABLE default.demo_orc (`id` Nullable(String), `score` Nullable(Int32), `day` Nullable(String)) ENGINE = Hive('thrift://hivetest:9083', 'test', 'demo_orc') PARTITION BY(day); - SELECT day, count(*) FROM default.demo_orc group by day order by day - """ - ) - expected_result = """2021-11-01 1 -2021-11-05 2 -2021-11-11 1 -2021-11-16 2 -""" - assert result == expected_result - - -@pytest.mark.parametrize( - "table,use_local_cache_for_remote_storage,enable_orc_file_minmax_index,enable_orc_stripe_minmax_index", - [ - pytest.param( - "demo_orc_no_cache_no_index", - "false", - "false", - "false", - id="demo_orc_no_cache_no_index", - ), - pytest.param( - "demo_orc_with_cache_no_index", - "true", - "false", - "false", - id="demo_orc_with_cache_no_index", - ), - pytest.param( - "demo_orc_no_cache_file_index", - "false", - "true", - "false", - id="demo_orc_no_cache_file_index", - ), - pytest.param( - "demo_orc_with_cache_file_index", - "true", - "true", - "false", - id="demo_orc_with_cache_file_index", - ), - pytest.param( - "demo_orc_no_cache_stripe_index", - "false", - "true", - "true", - id="demo_orc_no_cache_stripe_index", - ), - pytest.param( - "demo_orc_with_cache_stripe_index", - "true", - "true", - "true", - id="demo_orc_with_cache_stripe_index", - ), - ], -) -def test_orc_minmax_index( - started_cluster, - table, - use_local_cache_for_remote_storage, - enable_orc_file_minmax_index, - enable_orc_stripe_minmax_index, -): - node = started_cluster.instances["h0_0_0"] - result = node.query( - """ - DROP TABLE IF EXISTS default.{table}; - CREATE TABLE default.{table} (`id` Nullable(String), `score` Nullable(Int32), `day` Nullable(String)) ENGINE = Hive('thrift://hivetest:9083', 'test', 'demo_orc') PARTITION BY(day) - SETTINGS enable_orc_file_minmax_index = {enable_orc_file_minmax_index}, enable_orc_stripe_minmax_index = {enable_orc_stripe_minmax_index}; - """.format( - table=table, - enable_orc_file_minmax_index=enable_orc_file_minmax_index, - enable_orc_stripe_minmax_index=enable_orc_stripe_minmax_index, - ) - ) - assert result.strip() == "" - - for i in range(2): - result = node.query( - """ - SELECT day, id, score FROM default.{table} where day >= '2021-11-05' and day <= '2021-11-16' and score >= 15 and score <= 30 order by day, id - SETTINGS use_local_cache_for_remote_storage = {use_local_cache_for_remote_storage} - """.format( - table=table, - use_local_cache_for_remote_storage=use_local_cache_for_remote_storage, - ) - ) - - assert ( - result - == """2021-11-05 abd 15 -2021-11-16 aaa 22 -""" - ) - - -@pytest.mark.parametrize( - "table,use_local_cache_for_remote_storage,enable_parquet_rowgroup_minmax_index", - [ - pytest.param( - "demo_parquet_no_cache_no_index", - "false", - "false", - id="demo_parquet_no_cache_no_index", - ), - pytest.param( - "demo_parquet_with_cache_no_index", - "true", - "false", - id="demo_parquet_with_cache_no_index", - ), - pytest.param( - "demo_parquet_no_cache_rowgroup_index", - "false", - "true", - id="demo_parquet_no_cache_rowgroup_index", - ), - pytest.param( - "demo_parquet_with_cache_rowgroup_index", - "true", - "true", - id="demo_parquet_with_cache_rowgroup_index", - ), - ], -) -def test_parquet_minmax_index( - started_cluster, - table, - use_local_cache_for_remote_storage, - enable_parquet_rowgroup_minmax_index, -): - node = started_cluster.instances["h0_0_0"] - result = node.query( - """ - DROP TABLE IF EXISTS default.{table}; - CREATE TABLE default.{table} (`id` Nullable(String), `score` Nullable(Int32), `day` Nullable(String)) ENGINE = Hive('thrift://hivetest:9083', 'test', 'demo') PARTITION BY(day) - SETTINGS enable_parquet_rowgroup_minmax_index = {enable_parquet_rowgroup_minmax_index} - """.format( - table=table, - enable_parquet_rowgroup_minmax_index=enable_parquet_rowgroup_minmax_index, - ) - ) - assert result.strip() == "" - - for i in range(2): - result = node.query( - """ - SELECT day, id, score FROM default.{table} where day >= '2021-11-05' and day <= '2021-11-16' and score >= 15 and score <= 30 order by day, id - SETTINGS use_local_cache_for_remote_storage = {use_local_cache_for_remote_storage} - """.format( - table=table, - use_local_cache_for_remote_storage=use_local_cache_for_remote_storage, - ) - ) - - assert ( - result - == """2021-11-05 abd 15 -2021-11-16 aaa 22 -""" - ) - - -def test_hive_columns_prunning(started_cluster): - logging.info("Start testing groupby ...") - node = started_cluster.instances["h0_0_0"] - result = node.query( - """ - SELECT count(*) FROM default.demo_parquet_parts where day = '2021-11-05' - """ - ) - expected_result = """4 -""" - logging.info("query result:{}".format(result)) - assert result == expected_result - - -def test_text_count(started_cluster): - node = started_cluster.instances["h0_0_0"] - result = node.query( - """ - DROP TABLE IF EXISTS default.demo_orc; - CREATE TABLE default.demo_orc (`id` Nullable(String), `score` Nullable(Int32), `day` Nullable(String)) ENGINE = Hive('thrift://hivetest:9083', 'test', 'demo_orc') PARTITION BY(day); - SELECT day, count(*) FROM default.demo_orc group by day order by day SETTINGS format_csv_delimiter = '\x01' - """ - ) - expected_result = """2021-11-01 1 -2021-11-05 2 -2021-11-11 1 -2021-11-16 2 -""" - assert result == expected_result - - -def test_parquet_groupby_with_cache(started_cluster): - logging.info("Start testing groupby ...") - node = started_cluster.instances["h0_0_0"] - result = node.query( - """ - SELECT day, count(*) FROM default.demo_parquet group by day order by day - """ - ) - expected_result = """2021-11-01 1 -2021-11-05 2 -2021-11-11 1 -2021-11-16 2 -""" - assert result == expected_result - - -def test_parquet_groupby_by_hive_function(started_cluster): - logging.info("Start testing groupby ...") - node = started_cluster.instances["h0_0_0"] - result = node.query( - """ - SELECT day, count(*) FROM hive('thrift://hivetest:9083', 'test', 'demo', '`id` Nullable(String), `score` Nullable(Int32), `day` Nullable(String)', 'day') group by day order by day - """ - ) - expected_result = """2021-11-01 1 -2021-11-05 2 -2021-11-11 1 -2021-11-16 2 -""" - assert result == expected_result - - -def test_cache_read_bytes(started_cluster): - node = started_cluster.instances["h0_0_0"] - result = node.query( - """ - CREATE TABLE IF NOT EXISTS default.demo_parquet_1 (`id` Nullable(String), `score` Nullable(Int32), `day` Nullable(String)) ENGINE = Hive('thrift://hivetest:9083', 'test', 'demo') PARTITION BY(day) - """ - ) - test_passed = False - for i in range(10): - result = node.query( - """ - SELECT * FROM default.demo_parquet_1 settings input_format_parquet_allow_missing_columns = true - """ - ) - node.query("system flush logs") - result = node.query( - "select sum(ProfileEvent_ExternalDataSourceLocalCacheReadBytes) from system.metric_log where ProfileEvent_ExternalDataSourceLocalCacheReadBytes > 0" - ) - if result.strip() == "0": - logging.info("ProfileEvent_ExternalDataSourceLocalCacheReadBytes == 0") - time.sleep(10) - continue - test_passed = True - break - assert test_passed - - -def test_cache_dir_use(started_cluster): - node = started_cluster.instances["h0_0_0"] - result0 = node.exec_in_container( - ["bash", "-c", "ls /tmp/clickhouse_local_cache | wc -l"] - ) - result1 = node.exec_in_container( - ["bash", "-c", "ls /tmp/clickhouse_local_cache1 | wc -l"] - ) - assert result0 != "0" and result1 != "0" - - -def test_hive_struct_type(started_cluster): - node = started_cluster.instances["h0_0_0"] - result = node.query( - """ - CREATE TABLE IF NOT EXISTS default.test_hive_types (`f_tinyint` Int8, `f_smallint` Int16, `f_int` Int32, `f_integer` Int32, `f_bigint` Int64, `f_float` Float32, `f_double` Float64, `f_decimal` Float64, `f_timestamp` DateTime, `f_date` Date, `f_string` String, `f_varchar` String, `f_char` String, `f_bool` Boolean, `f_array_int` Array(Int32), `f_array_string` Array(String), `f_array_float` Array(Float32), `f_map_int` Map(String, Int32), `f_map_string` Map(String, String), `f_map_float` Map(String, Float32), `f_struct` Tuple(a String, b Int32, c Float32, d Tuple(x Int32, y String)), `day` String) ENGINE = Hive('thrift://hivetest:9083', 'test', 'test_hive_types') PARTITION BY (day) - """ - ) - result = node.query( - """ - SELECT * FROM default.test_hive_types WHERE day = '2022-02-20' SETTINGS input_format_parquet_import_nested=1 - """ - ) - expected_result = """1 2 3 4 5 6.11 7.22 8 2022-02-20 14:47:04 2022-02-20 hello world hello world hello world true [1,2,3] ['hello world','hello world'] [1.1,1.2] {'a':100,'b':200,'c':300} {'a':'aa','b':'bb','c':'cc'} {'a':111.1,'b':222.2,'c':333.3} ('aaa',200,333.3,(10,'xyz')) 2022-02-20""" - assert result.strip() == expected_result - - result = node.query( - """ - SELECT day, f_struct.a, f_struct.d.x FROM default.test_hive_types WHERE day = '2022-02-20' SETTINGS input_format_parquet_import_nested=1 - """ - ) - expected_result = """2022-02-20 aaa 10""" - - -def test_table_alter_add(started_cluster): - node = started_cluster.instances["h0_0_0"] - result = node.query("DROP TABLE IF EXISTS default.demo_parquet_1") - result = node.query( - """ -CREATE TABLE IF NOT EXISTS default.demo_parquet_1 (`score` Nullable(Int32), `day` Nullable(String)) ENGINE = Hive('thrift://hivetest:9083', 'test', 'demo') PARTITION BY(day) - """ - ) - result = node.query( - """ -ALTER TABLE default.demo_parquet_1 ADD COLUMN id Nullable(String) FIRST - """ - ) - result = node.query("""DESC default.demo_parquet_1 FORMAT TSV""") - - expected_result = "id\tNullable(String)\t\t\t\t\t\nscore\tNullable(Int32)\t\t\t\t\t\nday\tNullable(String)" - assert result.strip() == expected_result - - -def test_table_alter_drop(started_cluster): - node = started_cluster.instances["h0_0_0"] - result = node.query("DROP TABLE IF EXISTS default.demo_parquet_1") - result = node.query( - """ -CREATE TABLE IF NOT EXISTS default.demo_parquet_1 (`id` Nullable(String), `score` Nullable(Int32), `day` Nullable(String)) ENGINE = Hive('thrift://hivetest:9083', 'test', 'demo') PARTITION BY(day) - """ - ) - result = node.query( - """ -ALTER TABLE default.demo_parquet_1 DROP COLUMN id - """ - ) - - result = node.query("""DESC default.demo_parquet_1 FORMAT TSV""") - expected_result = """score\tNullable(Int32)\t\t\t\t\t\nday\tNullable(String)""" - assert result.strip() == expected_result - - -def test_table_alter_comment(started_cluster): - node = started_cluster.instances["h0_0_0"] - result = node.query("DROP TABLE IF EXISTS default.demo_parquet_1") - result = node.query( - """ -CREATE TABLE IF NOT EXISTS default.demo_parquet_1 (`id` Nullable(String), `score` Nullable(Int32), `day` Nullable(String)) ENGINE = Hive('thrift://hivetest:9083', 'test', 'demo') PARTITION BY(day) - """ - ) - - result = node.query( - """ALTER TABLE default.demo_parquet_1 COMMENT COLUMN id 'Text comment'""" - ) - result = node.query("""DESC default.demo_parquet_1 FORMAT TSV""") - expected_result = """id\tNullable(String)\t\t\tText comment\t\t\nscore\tNullable(Int32)\t\t\t\t\t\nday\tNullable(String)""" - assert result.strip() == expected_result diff --git a/tests/integration/test_host_regexp_multiple_ptr_records/configs/host_regexp.xml b/tests/integration/test_host_regexp_multiple_ptr_records/configs/host_regexp.xml deleted file mode 100644 index 7a2141e6c7e..00000000000 --- a/tests/integration/test_host_regexp_multiple_ptr_records/configs/host_regexp.xml +++ /dev/null @@ -1,11 +0,0 @@ - - - - - - test1\.example\.com$ - - default - - - \ No newline at end of file diff --git a/tests/integration/test_host_regexp_multiple_ptr_records/configs/listen_host.xml b/tests/integration/test_host_regexp_multiple_ptr_records/configs/listen_host.xml deleted file mode 100644 index 58ef55cd3f3..00000000000 --- a/tests/integration/test_host_regexp_multiple_ptr_records/configs/listen_host.xml +++ /dev/null @@ -1,5 +0,0 @@ - - :: - 0.0.0.0 - 1 - diff --git a/tests/integration/test_host_regexp_multiple_ptr_records/coredns_config/Corefile b/tests/integration/test_host_regexp_multiple_ptr_records/coredns_config/Corefile deleted file mode 100644 index 3edf37dafa5..00000000000 --- a/tests/integration/test_host_regexp_multiple_ptr_records/coredns_config/Corefile +++ /dev/null @@ -1,8 +0,0 @@ -. { - hosts /example.com { - reload "20ms" - fallthrough - } - forward . 127.0.0.11 - log -} diff --git a/tests/integration/test_host_regexp_multiple_ptr_records/coredns_config/example.com b/tests/integration/test_host_regexp_multiple_ptr_records/coredns_config/example.com deleted file mode 100644 index 9beb415c290..00000000000 --- a/tests/integration/test_host_regexp_multiple_ptr_records/coredns_config/example.com +++ /dev/null @@ -1 +0,0 @@ -filled in runtime, but needs to exist in order to be volume mapped in docker \ No newline at end of file diff --git a/tests/integration/test_host_regexp_multiple_ptr_records/test.py b/tests/integration/test_host_regexp_multiple_ptr_records/test.py deleted file mode 100644 index 36f76140a2c..00000000000 --- a/tests/integration/test_host_regexp_multiple_ptr_records/test.py +++ /dev/null @@ -1,115 +0,0 @@ -import pytest - -# FIXME This test is too flaky -# https://github.com/ClickHouse/ClickHouse/issues/51471 - -pytestmark = pytest.mark.skip - -import socket -from helpers.cluster import ClickHouseCluster, get_docker_compose_path, run_and_check -from time import sleep -import os - -DOCKER_COMPOSE_PATH = get_docker_compose_path() -SCRIPT_DIR = os.path.dirname(os.path.realpath(__file__)) - -cluster = ClickHouseCluster(__file__) - -ch_server = cluster.add_instance( - "clickhouse-server", - with_coredns=True, - main_configs=["configs/listen_host.xml"], - user_configs=["configs/host_regexp.xml"], - ipv6_address="2001:3984:3989::1:1111", -) - -client = cluster.add_instance( - "clickhouse-client", - ipv6_address="2001:3984:3989::1:1112", -) - - -@pytest.fixture(scope="module") -def started_cluster(): - global cluster - try: - cluster.start() - yield cluster - - finally: - cluster.shutdown() - - -def check_ptr_record(ip, hostname): - try: - host, aliaslist, ipaddrlist = socket.gethostbyaddr(ip) - if hostname.lower() == host.lower(): - return True - except socket.herror: - pass - return False - - -def setup_dns_server(ip): - domains_string = "test3.example.com test2.example.com test1.example.com" - example_file_path = f'{ch_server.env_variables["COREDNS_CONFIG_DIR"]}/example.com' - run_and_check(f"echo '{ip} {domains_string}' > {example_file_path}", shell=True) - - # DNS server takes time to reload the configuration. - for try_num in range(10): - if all(check_ptr_record(ip, host) for host in domains_string.split()): - break - sleep(1) - - -def setup_ch_server(dns_server_ip): - ch_server.exec_in_container( - (["bash", "-c", f"echo 'nameserver {dns_server_ip}' > /etc/resolv.conf"]) - ) - ch_server.exec_in_container( - (["bash", "-c", "echo 'options ndots:0' >> /etc/resolv.conf"]) - ) - ch_server.query("SYSTEM DROP DNS CACHE") - - -def build_endpoint_v4(ip): - return f"'http://{ip}:8123/?query=SELECT+1&user=test_dns'" - - -def build_endpoint_v6(ip): - return build_endpoint_v4(f"[{ip}]") - - -def test_host_regexp_multiple_ptr_v4_fails_with_wrong_resolution(started_cluster): - server_ip = cluster.get_instance_ip("clickhouse-server") - random_ip = "9.9.9.9" - dns_server_ip = cluster.get_instance_ip(cluster.coredns_host) - - setup_dns_server(random_ip) - setup_ch_server(dns_server_ip) - - endpoint = build_endpoint_v4(server_ip) - - assert "1\n" != client.exec_in_container(["bash", "-c", f"curl {endpoint}"]) - - -def test_host_regexp_multiple_ptr_v4(started_cluster): - server_ip = cluster.get_instance_ip("clickhouse-server") - client_ip = cluster.get_instance_ip("clickhouse-client") - dns_server_ip = cluster.get_instance_ip(cluster.coredns_host) - - setup_dns_server(client_ip) - setup_ch_server(dns_server_ip) - - endpoint = build_endpoint_v4(server_ip) - - assert "1\n" == client.exec_in_container(["bash", "-c", f"curl {endpoint}"]) - - -def test_host_regexp_multiple_ptr_v6(started_cluster): - setup_dns_server(client.ipv6_address) - setup_ch_server(cluster.get_instance_global_ipv6(cluster.coredns_host)) - - endpoint = build_endpoint_v6(ch_server.ipv6_address) - - assert "1\n" == client.exec_in_container(["bash", "-c", f"curl -6 {endpoint}"]) diff --git a/tests/integration/test_insert_into_distributed_through_materialized_view/test.py b/tests/integration/test_insert_into_distributed_through_materialized_view/test.py index a5f92002450..b1eb0df2d43 100644 --- a/tests/integration/test_insert_into_distributed_through_materialized_view/test.py +++ b/tests/integration/test_insert_into_distributed_through_materialized_view/test.py @@ -108,74 +108,6 @@ def test_reconnect(started_cluster): assert remote.query("SELECT count(*) FROM local1").strip() == "3" -@pytest.mark.skip(reason="Flapping test") -def test_inserts_batching(started_cluster): - instance = instance_test_inserts_batching - - with PartitionManager() as pm: - pm.partition_instances(instance, remote) - - instance.query("INSERT INTO local2_source(d, x) VALUES ('2000-01-01', 1)") - # Sleep a bit so that this INSERT forms a batch of its own. - time.sleep(0.2) - - instance.query("INSERT INTO local2_source(x, d) VALUES (2, '2000-01-01')") - - for i in range(3, 7): - instance.query( - "INSERT INTO local2_source(d, x) VALUES ('2000-01-01', {})".format(i) - ) - - for i in range(7, 9): - instance.query( - "INSERT INTO local2_source(x, d) VALUES ({}, '2000-01-01')".format(i) - ) - - instance.query("INSERT INTO local2_source(d, x) VALUES ('2000-01-01', 9)") - - # After ALTER the structure of the saved blocks will be different - instance.query("DROP TABLE local2_view") - instance.query("ALTER TABLE distributed ADD COLUMN s String") - - # Memory Engine doesn't support ALTER so we just DROP/CREATE everything - instance.query("DROP TABLE local2_source") - instance.query( - "CREATE TABLE local2_source (d Date, x UInt32, s String) ENGINE = Memory" - ) - instance.query( - "CREATE MATERIALIZED VIEW local2_view to distributed AS SELECT d,x,s FROM local2_source" - ) - - for i in range(10, 13): - instance.query( - "INSERT INTO local2_source(d, x) VALUES ('2000-01-01', {})".format(i) - ) - - time.sleep(1.0) - - result = remote.query( - "SELECT _part, groupArray(x) FROM local2 GROUP BY _part ORDER BY _part" - ) - - # Explanation: as merges are turned off on remote instance, active parts in local2 table correspond 1-to-1 - # to inserted blocks. - # Batches of max 3 rows are formed as min_insert_block_size_rows = 3. - # Blocks: - # 1. Failed batch that is retried with the same contents. - # 2. Full batch of inserts regardless of the order of columns thanks to the view. - # 3. Full batch of inserts regardless order of columns thanks to the view. - # 4. Full batch of inserts after ALTER (that have different block structure). - # 5. What was left to insert before ALTER. - expected = """\ -20000101_20000101_1_1_0 [1] -20000101_20000101_2_2_0 [2,3,4] -20000101_20000101_3_3_0 [5,6,7] -20000101_20000101_4_4_0 [10,11,12] -20000101_20000101_5_5_0 [8,9] -""" - assert TSV(result) == TSV(expected) - - def test_inserts_local(started_cluster): instance = instance_test_inserts_local_cluster instance.query("INSERT INTO local_source VALUES ('2000-01-01', 1)") diff --git a/tests/integration/test_keeper_client/test.py b/tests/integration/test_keeper_client/test.py index 00c7908eeed..8f7056a5afd 100644 --- a/tests/integration/test_keeper_client/test.py +++ b/tests/integration/test_keeper_client/test.py @@ -1,6 +1,7 @@ import pytest from helpers.client import CommandRequest from helpers.cluster import ClickHouseCluster +from helpers.test_tools import TSV cluster = ClickHouseCluster(__file__) @@ -13,7 +14,7 @@ node = cluster.add_instance( ) -@pytest.fixture(scope="module") +@pytest.fixture(scope="module", autouse=True) def started_cluster(): try: cluster.start() @@ -23,41 +24,123 @@ def started_cluster(): cluster.shutdown() -def test_base_commands(started_cluster): - _ = started_cluster - - command = CommandRequest( +def keeper_query(query: str): + return CommandRequest( [ - started_cluster.server_bin_path, + cluster.server_bin_path, "keeper-client", "--host", str(cluster.get_instance_ip("zoo1")), "--port", str(cluster.zookeeper_port), "-q", - "create test_create_zk_node1 testvalue1;create test_create_zk_node_2 testvalue2;get test_create_zk_node1;", + query, ], stdin="", ) + +def test_big_family(): + command = keeper_query( + "touch test_big_family;" + "touch test_big_family/1;" + "touch test_big_family/1/1;" + "touch test_big_family/1/2;" + "touch test_big_family/1/3;" + "touch test_big_family/1/4;" + "touch test_big_family/1/5;" + "touch test_big_family/2;" + "touch test_big_family/2/1;" + "touch test_big_family/2/2;" + "touch test_big_family/2/3;" + "find_big_family test_big_family;" + ) + + assert command.get_answer() == TSV( + [ + ["/test_big_family/1", "5"], + ["/test_big_family/2", "3"], + ["/test_big_family/2/3", "0"], + ["/test_big_family/2/2", "0"], + ["/test_big_family/2/1", "0"], + ["/test_big_family/1/5", "0"], + ["/test_big_family/1/4", "0"], + ["/test_big_family/1/3", "0"], + ["/test_big_family/1/2", "0"], + ["/test_big_family/1/1", "0"], + ] + ) + + command = keeper_query("find_big_family test_big_family 1;") + + assert command.get_answer() == TSV( + [ + ["/test_big_family/1", "5"], + ] + ) + + +def test_find_super_nodes(): + command = keeper_query( + "touch test_find_super_nodes;" + "touch test_find_super_nodes/1;" + "touch test_find_super_nodes/1/1;" + "touch test_find_super_nodes/1/2;" + "touch test_find_super_nodes/1/3;" + "touch test_find_super_nodes/1/4;" + "touch test_find_super_nodes/1/5;" + "touch test_find_super_nodes/2;" + "touch test_find_super_nodes/2/1;" + "touch test_find_super_nodes/2/2;" + "touch test_find_super_nodes/2/3;" + "touch test_find_super_nodes/2/4;" + "cd test_find_super_nodes;" + "find_super_nodes 4;" + ) + + assert command.get_answer() == TSV( + [ + ["/test_find_super_nodes/1", "5"], + ["/test_find_super_nodes/2", "4"], + ] + ) + + +def test_delete_stale_backups(): + command = keeper_query( + "touch /clickhouse;" + "touch /clickhouse/backups;" + "touch /clickhouse/backups/1;" + "touch /clickhouse/backups/1/stage;" + "touch /clickhouse/backups/1/stage/alive123;" + "touch /clickhouse/backups/2;" + "touch /clickhouse/backups/2/stage;" + "touch /clickhouse/backups/2/stage/dead123;" + "delete_stale_backups;" + "y;" + "ls clickhouse/backups;" + ) + + assert command.get_answer() == ( + "You are going to delete all inactive backups in /clickhouse/backups. Continue?\n" + 'Found backup "/clickhouse/backups/1", checking if it\'s active\n' + 'Backup "/clickhouse/backups/1" is active, not going to delete\n' + 'Found backup "/clickhouse/backups/2", checking if it\'s active\n' + 'Backup "/clickhouse/backups/2" is not active, deleting it\n' + "1\n" + ) + + +def test_base_commands(): + command = keeper_query( + "create test_create_zk_node1 testvalue1;" + "create test_create_zk_node_2 testvalue2;" + "get test_create_zk_node1;" + ) + assert command.get_answer() == "testvalue1\n" -def test_four_letter_word_commands(started_cluster): - _ = started_cluster - - command = CommandRequest( - [ - started_cluster.server_bin_path, - "keeper-client", - "--host", - str(cluster.get_instance_ip("zoo1")), - "--port", - str(cluster.zookeeper_port), - "-q", - "ruok", - ], - stdin="", - ) - +def test_four_letter_word_commands(): + command = keeper_query("ruok") assert command.get_answer() == "imok\n" diff --git a/tests/integration/test_keeper_reconfig_replace_leader_in_one_command/test.py b/tests/integration/test_keeper_reconfig_replace_leader_in_one_command/test.py index 1ec44d8a002..b099d0513e1 100644 --- a/tests/integration/test_keeper_reconfig_replace_leader_in_one_command/test.py +++ b/tests/integration/test_keeper_reconfig_replace_leader_in_one_command/test.py @@ -41,81 +41,3 @@ def started_cluster(): def get_fake_zk(node): return ku.get_fake_zk(cluster, node) - - -@pytest.mark.skip(reason="test is flaky because changes are not properly waited for") -def test_reconfig_replace_leader_in_one_command(started_cluster): - """ - Remove leader from a cluster of 3 and add a new node to this cluster in a single command - """ - - zk1 = get_fake_zk(node1) - config = ku.get_config_str(zk1) - - assert len(config.split("\n")) == 3 - assert "node1" in config - assert "node2" in config - assert "node3" in config - assert "node4" not in config - - for i in range(100): - zk1.create(f"/test_four_{i}", b"somedata") - - zk2 = get_fake_zk(node2) - zk2.sync("/test_four_0") - ku.wait_configs_equal(config, zk2) - - zk3 = get_fake_zk(node3) - zk3.sync("/test_four_0") - ku.wait_configs_equal(config, zk3) - - for i in range(100): - assert zk2.exists(f"/test_four_{i}") is not None - assert zk3.exists(f"/test_four_{i}") is not None - - assert ku.is_leader(cluster, node1) - node4.start_clickhouse() - config, _ = zk2.reconfig( - joining="server.4=node4:9234", leaving="1", new_members=None - ) - config = config.decode("utf-8") - - print("After removing 1 and adding 4", config) - assert len(config.split("\n")) == 3 - assert "node1" not in config - assert "node2" in config - assert "node3" in config - assert "node4" in config - - ku.wait_until_connected(cluster, node4) - time.sleep(1) - - zk4 = get_fake_zk(node4) - zk4.sync("/test_four_0") - ku.wait_configs_equal(config, zk4) - - for i in range(100): - assert zk4.exists(f"test_four_{i}") is not None - zk4.create(f"/test_four_{100 + i}", b"somedata") - - with pytest.raises(Exception): - zk1.stop() - zk1.close() - zk1 = get_fake_zk(node1) - zk1.sync("/test_four_0") - - zk2.stop() - zk2.close() - zk2 = get_fake_zk(node2) - zk2.sync("/test_four_0") - ku.wait_configs_equal(config, zk2) - - zk3.stop() - zk3.close() - zk3 = get_fake_zk(node3) - zk3.sync("/test_four_0") - ku.wait_configs_equal(config, zk3) - - for i in range(200): - assert zk2.exists(f"test_four_{i}") is not None - assert zk3.exists(f"test_four_{i}") is not None diff --git a/tests/integration/test_hive_query/__init__.py b/tests/integration/test_multi_access_storage_role_management/__init__.py similarity index 100% rename from tests/integration/test_hive_query/__init__.py rename to tests/integration/test_multi_access_storage_role_management/__init__.py diff --git a/tests/integration/test_multi_access_storage_role_management/configs/memory.xml b/tests/integration/test_multi_access_storage_role_management/configs/memory.xml new file mode 100644 index 00000000000..c2f60c8bbc1 --- /dev/null +++ b/tests/integration/test_multi_access_storage_role_management/configs/memory.xml @@ -0,0 +1,5 @@ + + + + + diff --git a/tests/integration/test_multi_access_storage_role_management/configs/roles.xml b/tests/integration/test_multi_access_storage_role_management/configs/roles.xml new file mode 100644 index 00000000000..faaf1ee27ae --- /dev/null +++ b/tests/integration/test_multi_access_storage_role_management/configs/roles.xml @@ -0,0 +1,9 @@ + + + + + GRANT ALL ON *.* WITH GRANT OPTION + + + + diff --git a/tests/integration/test_multi_access_storage_role_management/test.py b/tests/integration/test_multi_access_storage_role_management/test.py new file mode 100644 index 00000000000..77aea411394 --- /dev/null +++ b/tests/integration/test_multi_access_storage_role_management/test.py @@ -0,0 +1,153 @@ +import pytest +import os +from helpers.cluster import ClickHouseCluster +from helpers.test_tools import TSV +from helpers.client import QueryRuntimeException + +cluster = ClickHouseCluster(__file__) +node = cluster.add_instance( + "node", stay_alive=True, main_configs=["configs/memory.xml"] +) + +SCRIPT_DIR = os.path.dirname(os.path.realpath(__file__)) + + +@pytest.fixture(scope="module", autouse=True) +def started_cluster(): + try: + cluster.start() + node.query("CREATE DATABASE mydb") + node.query("CREATE TABLE mydb.table1(x UInt32) ENGINE=Log") + + node.query("CREATE USER test_user") + node.query("CREATE USER test_user2") + yield cluster + + finally: + cluster.shutdown() + + +def execute_test_for_access_type(access_type: str, system_table_name: str): + node.query(f"CREATE {access_type} test1 IN local_directory") + node.query(f"CREATE {access_type} test2 IN local_directory") + node.query(f"CREATE {access_type} test3 IN local_directory") + + node.query(f"CREATE {access_type} test4 IN memory") + node.query(f"CREATE {access_type} test5 IN memory") + node.query(f"CREATE {access_type} test6 IN memory") + + # Already exists + with pytest.raises(QueryRuntimeException): + node.query(f"CREATE {access_type} test1 IN memory") + + node.query(f"MOVE {access_type} test1 TO memory") + assert node.query( + f"SELECT storage FROM system.{system_table_name} WHERE name = 'test1'" + ) == TSV(["memory"]) + + node.query(f"MOVE {access_type} test2 TO local_directory") + assert node.query( + f"SELECT storage FROM system.{system_table_name} WHERE name = 'test2'" + ) == TSV(["local_directory"]) + + node.query(f"MOVE {access_type} test2,test3 TO memory") + assert node.query( + f"SELECT storage FROM system.{system_table_name} WHERE name = 'test2'" + ) == TSV(["memory"]) + assert node.query( + f"SELECT storage FROM system.{system_table_name} WHERE name = 'test3'" + ) == TSV(["memory"]) + + node.query(f"MOVE {access_type} test4,test5 TO local_directory") + + # Different storages + with pytest.raises(QueryRuntimeException): + node.query(f"MOVE {access_type} test4,test1 TO memory") + + # Doesn't exist + with pytest.raises(QueryRuntimeException): + node.query(f"MOVE {access_type} test7 TO local_directory") + + # Storage doesn't exist + with pytest.raises(QueryRuntimeException): + node.query(f"MOVE {access_type} test6 TO non_existing_storage") + + # Unwriteable storage + with pytest.raises(QueryRuntimeException): + node.query(f"MOVE {access_type} test6 TO users_xml") + + node.query(f"DROP {access_type} test1") + node.query(f"DROP {access_type} test2") + node.query(f"DROP {access_type} test3") + node.query(f"DROP {access_type} test4") + node.query(f"DROP {access_type} test5") + node.query(f"DROP {access_type} test6") + + +def test_roles(): + execute_test_for_access_type("ROLE", "roles") + + +def test_users(): + execute_test_for_access_type("USER", "users") + + +def test_settings_profiles(): + execute_test_for_access_type("SETTINGS PROFILE", "settings_profiles") + + +def test_quotas(): + execute_test_for_access_type("QUOTA", "quotas") + + +def test_role_from_different_storages(): + node.query("CREATE ROLE default_role") + node.query("GRANT SELECT ON system.* TO default_role") + + assert node.query("SHOW GRANTS FOR default_role") == TSV( + ["GRANT SELECT ON system.* TO default_role"] + ) + assert node.query("SHOW ROLES") == TSV(["default_role"]) + + node.query("GRANT default_role TO test_user") + + node.copy_file_to_container( + os.path.join(SCRIPT_DIR, "configs/roles.xml"), + "/etc/clickhouse-server/users.d/roles.xml", + ) + + node.restart_clickhouse() + + assert node.query("SELECT name, storage FROM system.roles") == TSV( + [["default_role", "users_xml"], ["default_role", "local_directory"]] + ) + + # Role from users.xml will have priority + assert node.query("SHOW GRANTS FOR default_role") == TSV( + ["GRANT ALL ON *.* TO default_role WITH GRANT OPTION"] + ) + + node.query("GRANT default_role TO test_user") + node.query("GRANT default_role TO test_user2") + assert node.query( + "SELECT granted_role_id FROM system.role_grants WHERE user_name = 'test_user2'" + ) == TSV( + [ + "62bedbf3-7fb1-94cb-3a35-e479693223b3" + ] # roles from users.xml have deterministic ids + ) + + node.query("DROP ROLE default_role FROM local_directory") + assert node.query( + "SELECT granted_role_id FROM system.role_grants WHERE user_name = 'test_user'" + ) == TSV(["62bedbf3-7fb1-94cb-3a35-e479693223b3"]) + + # Already exists + with pytest.raises(QueryRuntimeException): + node.query("CREATE ROLE default_role IN memory") + + node.query("CREATE ROLE other_role IN memory") + + assert node.query( + "SELECT storage FROM system.roles WHERE name = 'other_role'" + ) == TSV(["memory"]) diff --git a/tests/integration/test_mutations_with_merge_tree/test.py b/tests/integration/test_mutations_with_merge_tree/test.py index e7d7abc8c7c..0b4a750b38e 100644 --- a/tests/integration/test_mutations_with_merge_tree/test.py +++ b/tests/integration/test_mutations_with_merge_tree/test.py @@ -27,111 +27,6 @@ def started_cluster(): cluster.shutdown() -@pytest.mark.skip(reason="Skipping mutations in partition does not work") -def test_mutations_in_partition_background(started_cluster): - try: - numbers = 100 - - name = "test_mutations_in_partition" - instance_test_mutations.query( - f"""CREATE TABLE {name} (date Date, a UInt64, b String) ENGINE = MergeTree() ORDER BY tuple() PARTITION BY a""" - ) - instance_test_mutations.query( - f"""INSERT INTO {name} SELECT '2019-07-29' AS date, number, toString(number) FROM numbers({numbers})""" - ) - - for i in range(0, numbers, 3): - instance_test_mutations.query( - f"""ALTER TABLE {name} DELETE IN PARTITION {i} WHERE a = {i}""" - ) - - for i in range(1, numbers, 3): - instance_test_mutations.query( - f"""ALTER TABLE {name} UPDATE b = 'changed' IN PARTITION {i} WHERE a = {i} """ - ) - - def count_and_changed(): - return instance_test_mutations.query( - f"SELECT count(), countIf(b == 'changed') FROM {name} SETTINGS force_index_by_date = 0, force_primary_key = 0 FORMAT CSV" - ).splitlines() - - all_done = False - for wait_times_for_mutation in range( - 100 - ): # wait for replication 80 seconds max - time.sleep(0.8) - - if count_and_changed() == ["66,33"]: - all_done = True - break - - print( - instance_test_mutations.query( - f"SELECT mutation_id, command, parts_to_do, is_done, latest_failed_part, latest_fail_reason, parts_to_do_names FROM system.mutations WHERE table = '{name}' SETTINGS force_index_by_date = 0, force_primary_key = 0 FORMAT TSVWithNames" - ) - ) - - assert (count_and_changed(), all_done) == (["66,33"], True) - assert instance_test_mutations.query( - f"SELECT count(), sum(is_done) FROM system.mutations WHERE table = '{name}' SETTINGS force_index_by_date = 0, force_primary_key = 0 FORMAT CSV" - ).splitlines() == ["67,67"] - - finally: - instance_test_mutations.query(f"""DROP TABLE {name}""") - - -@pytest.mark.skip(reason="Skipping mutations in partition does not work") -@pytest.mark.parametrize("sync", [("last",), ("all",)]) -def test_mutations_in_partition_sync(started_cluster, sync): - try: - numbers = 10 - - name = "test_mutations_in_partition_sync" - instance_test_mutations.query( - f"""CREATE TABLE {name} (date Date, a UInt64, b String) ENGINE = MergeTree() ORDER BY tuple() PARTITION BY a""" - ) - instance_test_mutations.query( - f"""INSERT INTO {name} SELECT '2019-07-29' AS date, number, toString(number) FROM numbers({numbers})""" - ) - - for i in range(0, numbers, 3): - instance_test_mutations.query( - f"""ALTER TABLE {name} DELETE IN PARTITION {i} WHERE a = {i}""" - + (" SETTINGS mutations_sync = 1" if sync == "all" else "") - ) - - for reverse_index, i in reversed( - list(enumerate(reversed(range(1, numbers, 3)))) - ): - instance_test_mutations.query( - f"""ALTER TABLE {name} UPDATE b = 'changed' IN PARTITION {i} WHERE a = {i}""" - + ( - " SETTINGS mutations_sync = 1" - if not reverse_index or sync == "all" - else "" - ) - ) - - def count_and_changed(): - return instance_test_mutations.query( - f"SELECT count(), countIf(b == 'changed') FROM {name} SETTINGS force_index_by_date = 0, force_primary_key = 0 FORMAT CSV" - ).splitlines() - - print( - instance_test_mutations.query( - f"SELECT mutation_id, command, parts_to_do, is_done, latest_failed_part, latest_fail_reason FROM system.mutations WHERE table = '{name}' SETTINGS force_index_by_date = 0, force_primary_key = 0 FORMAT TSVWithNames" - ) - ) - - assert count_and_changed() == ["6,3"] - assert instance_test_mutations.query( - f"SELECT count(), sum(is_done) FROM system.mutations WHERE table = '{name}' SETTINGS force_index_by_date = 0, force_primary_key = 0 FORMAT CSV" - ).splitlines() == ["7,7"] - - finally: - instance_test_mutations.query(f"""DROP TABLE {name}""") - - def test_mutations_with_merge_background_task(started_cluster): instance_test_mutations.query( """SYSTEM STOP MERGES test_mutations_with_ast_elements""" @@ -190,92 +85,3 @@ def test_mutations_with_truncate_table(started_cluster): ).rstrip() == "0" ) - - -@pytest.mark.skip(reason="Skipping mutations in partition does not work") -def test_mutations_will_not_hang_for_non_existing_parts_sync(started_cluster): - try: - numbers = 100 - - name = "test_mutations_will_not_hang_for_non_existing_parts_sync" - instance_test_mutations.query( - f"""CREATE TABLE {name} (date Date, a UInt64, b String) ENGINE = MergeTree() ORDER BY tuple() PARTITION BY a""" - ) - instance_test_mutations.query( - f"""INSERT INTO {name} SELECT '2019-07-29' AS date, number, toString(number) FROM numbers({numbers})""" - ) - - for i in range(0, numbers, 3): - instance_test_mutations.query( - f"""ALTER TABLE {name} DELETE IN PARTITION {i+1000} WHERE a = {i} SETTINGS mutations_sync = 1""" - ) - - def count(): - return instance_test_mutations.query( - f"SELECT count() FROM {name} SETTINGS force_index_by_date = 0, force_primary_key = 0 FORMAT CSV" - ).splitlines() - - print( - instance_test_mutations.query( - f"SELECT mutation_id, command, parts_to_do, is_done, latest_failed_part, latest_fail_reason, parts_to_do_names FROM system.mutations WHERE table = '{name}' SETTINGS force_index_by_date = 0, force_primary_key = 0 FORMAT TSVWithNames" - ) - ) - - assert count() == [f"{numbers}"] - assert instance_test_mutations.query( - f"SELECT count(), sum(is_done) FROM system.mutations WHERE table = '{name}' SETTINGS force_index_by_date = 0, force_primary_key = 0 FORMAT CSV" - ).splitlines() == [f"34,34"] - - finally: - instance_test_mutations.query(f"""DROP TABLE {name}""") - - -@pytest.mark.skip(reason="Skipping mutations in partition does not work") -def test_mutations_will_not_hang_for_non_existing_parts_async(started_cluster): - try: - numbers = 100 - - name = "test_mutations_will_not_hang_for_non_existing_parts_async" - instance_test_mutations.query( - f"""CREATE TABLE {name} (date Date, a UInt64, b String) ENGINE = MergeTree() ORDER BY tuple() PARTITION BY a""" - ) - instance_test_mutations.query( - f"""INSERT INTO {name} SELECT '2019-07-29' AS date, number, toString(number) FROM numbers({numbers})""" - ) - - for i in range(0, numbers, 3): - instance_test_mutations.query( - f"""ALTER TABLE {name} DELETE IN PARTITION {i+1000} WHERE a = {i}""" - ) - - def count(): - return instance_test_mutations.query( - f"SELECT count() FROM {name} SETTINGS force_index_by_date = 0, force_primary_key = 0 FORMAT CSV" - ).splitlines() - - def count_and_sum_is_done(): - return instance_test_mutations.query( - f"SELECT count(), sum(is_done) FROM system.mutations WHERE table = '{name}' SETTINGS force_index_by_date = 0, force_primary_key = 0 FORMAT CSV" - ).splitlines() - - all_done = False - for wait_times_for_mutation in range( - 100 - ): # wait for replication 80 seconds max - time.sleep(0.8) - - if count_and_sum_is_done() == ["34,34"]: - all_done = True - break - - print( - instance_test_mutations.query( - f"SELECT mutation_id, command, parts_to_do, is_done, latest_failed_part, latest_fail_reason, parts_to_do_names FROM system.mutations WHERE table = '{name}' SETTINGS force_index_by_date = 0, force_primary_key = 0 FORMAT TSVWithNames" - ) - ) - - assert count() == [f"{numbers}"] - assert count_and_sum_is_done() == ["34,34"] - - finally: - instance_test_mutations.query(f"""DROP TABLE {name}""") diff --git a/tests/integration/test_polymorphic_parts/test.py b/tests/integration/test_polymorphic_parts/test.py index 48bcca7d505..debb509de90 100644 --- a/tests/integration/test_polymorphic_parts/test.py +++ b/tests/integration/test_polymorphic_parts/test.py @@ -414,54 +414,6 @@ def start_cluster_diff_versions(): cluster.shutdown() -@pytest.mark.skip(reason="compatability is temporary broken") -def test_polymorphic_parts_diff_versions(start_cluster_diff_versions): - # Check that replication with Wide parts works between different versions. - - node_old = node7 - node_new = node8 - - insert_random_data("polymorphic_table", node7, 100) - node8.query("SYSTEM SYNC REPLICA polymorphic_table", timeout=20) - - assert node8.query("SELECT count() FROM polymorphic_table") == "100\n" - assert ( - node8.query( - "SELECT DISTINCT part_type FROM system.parts WHERE table = 'polymorphic_table' and active" - ) - == "Wide\n" - ) - - -@pytest.mark.skip(reason="compatability is temporary broken") -def test_polymorphic_parts_diff_versions_2(start_cluster_diff_versions): - # Replication doesn't work on old version if part is created in compact format, because - # this version doesn't know anything about it. It's considered to be ok. - - node_old = node7 - node_new = node8 - - insert_random_data("polymorphic_table_2", node_new, 100) - - assert node_new.query("SELECT count() FROM polymorphic_table_2") == "100\n" - assert node_old.query("SELECT count() FROM polymorphic_table_2") == "0\n" - with pytest.raises(Exception): - node_old.query("SYSTEM SYNC REPLICA polymorphic_table_2", timeout=3) - - node_old.restart_with_latest_version(fix_metadata=True) - - node_old.query("SYSTEM SYNC REPLICA polymorphic_table_2", timeout=20) - - # Works after update - assert node_old.query("SELECT count() FROM polymorphic_table_2") == "100\n" - assert ( - node_old.query( - "SELECT DISTINCT part_type FROM system.parts WHERE table = 'polymorphic_table_2' and active" - ) - == "Compact\n" - ) - - def test_polymorphic_parts_non_adaptive(start_cluster): node1.query("SYSTEM STOP MERGES") node2.query("SYSTEM STOP MERGES") diff --git a/tests/integration/test_profile_max_sessions_for_user/test.py b/tests/integration/test_profile_max_sessions_for_user/test.py index 2930262f63e..65587933fed 100755 --- a/tests/integration/test_profile_max_sessions_for_user/test.py +++ b/tests/integration/test_profile_max_sessions_for_user/test.py @@ -27,10 +27,7 @@ proto_dir = os.path.join(SCRIPT_DIR, "./protos") gen_dir = os.path.join(SCRIPT_DIR, "./_gen") os.makedirs(gen_dir, exist_ok=True) run_and_check( - "python3 -m grpc_tools.protoc -I{proto_dir} --python_out={gen_dir} --grpc_python_out={gen_dir} \ - {proto_dir}/clickhouse_grpc.proto".format( - proto_dir=proto_dir, gen_dir=gen_dir - ), + f"python3 -m grpc_tools.protoc -I{proto_dir} --python_out={gen_dir} --grpc_python_out={gen_dir} {proto_dir}/clickhouse_grpc.proto", shell=True, ) diff --git a/tests/integration/test_quota/test.py b/tests/integration/test_quota/test.py index 4f98b9a0d0d..cec14b0af73 100644 --- a/tests/integration/test_quota/test.py +++ b/tests/integration/test_quota/test.py @@ -105,7 +105,7 @@ def test_quota_from_users_xml(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", [31556952], 0, @@ -252,7 +252,7 @@ def test_simpliest_quota(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", "[]", 0, @@ -326,7 +326,7 @@ def test_tracking_quota(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", "[31556952]", 0, @@ -446,7 +446,7 @@ def test_exceed_quota(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", "[31556952]", 0, @@ -523,7 +523,7 @@ def test_exceed_quota(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", "[31556952]", 0, @@ -613,7 +613,7 @@ def test_add_remove_interval(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", [31556952], 0, @@ -675,7 +675,7 @@ def test_add_remove_interval(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", "[31556952,63113904]", 0, @@ -824,7 +824,7 @@ def test_add_remove_interval(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", [31556952], 0, @@ -914,7 +914,7 @@ def test_add_remove_interval(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", "[]", 0, @@ -986,7 +986,7 @@ def test_add_remove_interval(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", [31556952], 0, @@ -1048,7 +1048,7 @@ def test_add_remove_quota(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", [31556952], 0, @@ -1111,7 +1111,7 @@ def test_add_remove_quota(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", "[31556952]", 0, @@ -1121,7 +1121,7 @@ def test_add_remove_quota(): [ "myQuota2", "4590510c-4d13-bf21-ec8a-c2187b092e73", - "users.xml", + "users_xml", "['client_key','user_name']", "[3600,2629746]", 0, @@ -1214,7 +1214,7 @@ def test_add_remove_quota(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", "[31556952]", 0, @@ -1283,7 +1283,7 @@ def test_add_remove_quota(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", "[31556952]", 0, @@ -1346,7 +1346,7 @@ def test_reload_users_xml_by_timer(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", "[31556952]", 0, @@ -1385,7 +1385,7 @@ def test_reload_users_xml_by_timer(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", ["user_name"], "[31556952]", 0, @@ -1554,7 +1554,7 @@ def test_query_inserts(): [ "myQuota", "e651da9c-a748-8703-061a-7e5e5096dae7", - "users.xml", + "users_xml", "['user_name']", [31556952], 0, diff --git a/tests/integration/test_replicated_merge_tree_hdfs_zero_copy/__init__.py b/tests/integration/test_replicated_merge_tree_hdfs_zero_copy/__init__.py deleted file mode 100644 index e69de29bb2d..00000000000 diff --git a/tests/integration/test_replicated_merge_tree_hdfs_zero_copy/configs/config.d/storage_conf.xml b/tests/integration/test_replicated_merge_tree_hdfs_zero_copy/configs/config.d/storage_conf.xml deleted file mode 100644 index bb4aba94e0b..00000000000 --- a/tests/integration/test_replicated_merge_tree_hdfs_zero_copy/configs/config.d/storage_conf.xml +++ /dev/null @@ -1,93 +0,0 @@ - - - - - hdfs - hdfs://hdfs1:9000/clickhouse1/ - - true - - - hdfs - hdfs://hdfs1:9000/clickhouse1/ - - true - - - hdfs - hdfs://hdfs1:9000/clickhouse2/ - - true - - - - - -
- hdfs1 -
- - - - -
- default -
- - hdfs1 - - - 0.0 - - - -
- hdfs2 -
- - hdfs1 - - - - - -
- hdfs1_again -
- - hdfs1 - - - - - - - - 1024000 - 1 - true - - - - - - - node1 - 9000 - - - - - node2 - 9000 - - - - - - - test_cluster - 1 - - true - diff --git a/tests/integration/test_replicated_merge_tree_hdfs_zero_copy/test.py b/tests/integration/test_replicated_merge_tree_hdfs_zero_copy/test.py deleted file mode 100644 index eb3d62eb718..00000000000 --- a/tests/integration/test_replicated_merge_tree_hdfs_zero_copy/test.py +++ /dev/null @@ -1,322 +0,0 @@ -import pytest - -# FIXME This test is too flaky -# https://github.com/ClickHouse/ClickHouse/issues/42561 - -pytestmark = pytest.mark.skip - -import logging -from string import Template -import time - -from helpers.cluster import ClickHouseCluster -from helpers.test_tools import assert_eq_with_retry - -from pyhdfs import HdfsClient - -SHARDS = 2 -FILES_OVERHEAD_PER_TABLE = 1 # format_version.txt -FILES_OVERHEAD_PER_PART_COMPACT = 7 - - -def wait_for_hdfs_objects(cluster, fp, expected, num_tries=30): - fs = HdfsClient(hosts=cluster.hdfs_ip) - while num_tries > 0: - num_hdfs_objects = len(fs.listdir(fp)) - if num_hdfs_objects == expected: - break - num_tries -= 1 - time.sleep(1) - assert len(fs.listdir(fp)) == expected - - -@pytest.fixture(scope="module") -def cluster(): - try: - cluster = ClickHouseCluster(__file__) - cluster.add_instance( - "node1", - main_configs=["configs/config.d/storage_conf.xml"], - macros={"replica": "node1"}, - with_zookeeper=True, - with_hdfs=True, - ) - cluster.add_instance( - "node2", - main_configs=["configs/config.d/storage_conf.xml"], - macros={"replica": "node2"}, - with_zookeeper=True, - with_hdfs=True, - ) - logging.info("Starting cluster...") - cluster.start() - if cluster.instances["node1"].is_debug_build(): - # https://github.com/ClickHouse/ClickHouse/issues/27814 - pytest.skip( - "libhdfs3 calls rand function which does not pass harmful check in debug build" - ) - logging.info("Cluster started") - - fs = HdfsClient(hosts=cluster.hdfs_ip) - fs.mkdirs("/clickhouse1") - fs.mkdirs("/clickhouse2") - logging.info("Created HDFS directory") - - yield cluster - finally: - cluster.shutdown() - - -def test_hdfs_zero_copy_replication_insert(cluster): - node1 = cluster.instances["node1"] - node2 = cluster.instances["node2"] - try: - node1.query( - """ - CREATE TABLE hdfs_test ON CLUSTER test_cluster (dt DateTime, id Int64) - ENGINE=ReplicatedMergeTree('/clickhouse/tables/{cluster}/{shard}/hdfs_test', '{replica}') - ORDER BY (dt, id) - SETTINGS storage_policy='hdfs_only' - """ - ) - wait_for_hdfs_objects( - cluster, "/clickhouse1", SHARDS * FILES_OVERHEAD_PER_TABLE - ) - - node1.query("INSERT INTO hdfs_test VALUES (now() - INTERVAL 3 DAY, 10)") - node2.query("SYSTEM SYNC REPLICA hdfs_test", timeout=30) - assert node1.query("SELECT count() FROM hdfs_test FORMAT Values") == "(1)" - assert node2.query("SELECT count() FROM hdfs_test FORMAT Values") == "(1)" - assert ( - node1.query("SELECT id FROM hdfs_test ORDER BY dt FORMAT Values") == "(10)" - ) - assert ( - node2.query("SELECT id FROM hdfs_test ORDER BY dt FORMAT Values") == "(10)" - ) - assert ( - node1.query( - "SELECT partition_id,disk_name FROM system.parts WHERE table='hdfs_test' FORMAT Values" - ) - == "('all','hdfs1')" - ) - assert ( - node2.query( - "SELECT partition_id,disk_name FROM system.parts WHERE table='hdfs_test' FORMAT Values" - ) - == "('all','hdfs1')" - ) - wait_for_hdfs_objects( - cluster, - "/clickhouse1", - SHARDS * FILES_OVERHEAD_PER_TABLE + FILES_OVERHEAD_PER_PART_COMPACT, - ) - finally: - node1.query("DROP TABLE IF EXISTS hdfs_test SYNC") - node2.query("DROP TABLE IF EXISTS hdfs_test SYNC") - - -@pytest.mark.parametrize( - ("storage_policy", "init_objects"), - [("hybrid", 0), ("tiered", 0), ("tiered_copy", FILES_OVERHEAD_PER_TABLE)], -) -def test_hdfs_zero_copy_replication_single_move(cluster, storage_policy, init_objects): - node1 = cluster.instances["node1"] - try: - node1.query( - Template( - """ - CREATE TABLE single_node_move_test (dt DateTime, id Int64) - ENGINE=ReplicatedMergeTree('/clickhouse/tables/{cluster}/{shard}/single_node_move_test', '{replica}') - ORDER BY (dt, id) - SETTINGS storage_policy='$policy',temporary_directories_lifetime=1 - """ - ).substitute(policy=storage_policy) - ) - wait_for_hdfs_objects(cluster, "/clickhouse1", init_objects) - - node1.query( - "INSERT INTO single_node_move_test VALUES (now() - INTERVAL 3 DAY, 10), (now() - INTERVAL 1 DAY, 11)" - ) - assert ( - node1.query( - "SELECT id FROM single_node_move_test ORDER BY dt FORMAT Values" - ) - == "(10),(11)" - ) - - node1.query( - "ALTER TABLE single_node_move_test MOVE PARTITION ID 'all' TO VOLUME 'external'" - ) - assert ( - node1.query( - "SELECT partition_id,disk_name FROM system.parts WHERE table='single_node_move_test' FORMAT Values" - ) - == "('all','hdfs1')" - ) - assert ( - node1.query( - "SELECT id FROM single_node_move_test ORDER BY dt FORMAT Values" - ) - == "(10),(11)" - ) - wait_for_hdfs_objects( - cluster, "/clickhouse1", init_objects + FILES_OVERHEAD_PER_PART_COMPACT - ) - - node1.query( - "ALTER TABLE single_node_move_test MOVE PARTITION ID 'all' TO VOLUME 'main'" - ) - assert ( - node1.query( - "SELECT id FROM single_node_move_test ORDER BY dt FORMAT Values" - ) - == "(10),(11)" - ) - finally: - node1.query("DROP TABLE IF EXISTS single_node_move_test SYNC") - - -@pytest.mark.parametrize( - ("storage_policy", "init_objects"), - [("hybrid", 0), ("tiered", 0), ("tiered_copy", SHARDS * FILES_OVERHEAD_PER_TABLE)], -) -def test_hdfs_zero_copy_replication_move(cluster, storage_policy, init_objects): - node1 = cluster.instances["node1"] - node2 = cluster.instances["node2"] - try: - node1.query( - Template( - """ - CREATE TABLE move_test ON CLUSTER test_cluster (dt DateTime, id Int64) - ENGINE=ReplicatedMergeTree('/clickhouse/tables/{cluster}/{shard}/move_test', '{replica}') - ORDER BY (dt, id) - SETTINGS storage_policy='$policy' - """ - ).substitute(policy=storage_policy) - ) - wait_for_hdfs_objects(cluster, "/clickhouse1", init_objects) - - node1.query( - "INSERT INTO move_test VALUES (now() - INTERVAL 3 DAY, 10), (now() - INTERVAL 1 DAY, 11)" - ) - node2.query("SYSTEM SYNC REPLICA move_test", timeout=30) - - assert ( - node1.query("SELECT id FROM move_test ORDER BY dt FORMAT Values") - == "(10),(11)" - ) - assert ( - node2.query("SELECT id FROM move_test ORDER BY dt FORMAT Values") - == "(10),(11)" - ) - - node1.query( - "ALTER TABLE move_test MOVE PARTITION ID 'all' TO VOLUME 'external'" - ) - wait_for_hdfs_objects( - cluster, "/clickhouse1", init_objects + FILES_OVERHEAD_PER_PART_COMPACT - ) - - node2.query( - "ALTER TABLE move_test MOVE PARTITION ID 'all' TO VOLUME 'external'" - ) - assert ( - node1.query( - "SELECT partition_id,disk_name FROM system.parts WHERE table='move_test' FORMAT Values" - ) - == "('all','hdfs1')" - ) - assert ( - node2.query( - "SELECT partition_id,disk_name FROM system.parts WHERE table='move_test' FORMAT Values" - ) - == "('all','hdfs1')" - ) - assert ( - node1.query("SELECT id FROM move_test ORDER BY dt FORMAT Values") - == "(10),(11)" - ) - assert ( - node2.query("SELECT id FROM move_test ORDER BY dt FORMAT Values") - == "(10),(11)" - ) - wait_for_hdfs_objects( - cluster, "/clickhouse1", init_objects + FILES_OVERHEAD_PER_PART_COMPACT - ) - finally: - node1.query("DROP TABLE IF EXISTS move_test SYNC") - node2.query("DROP TABLE IF EXISTS move_test SYNC") - - -@pytest.mark.parametrize(("storage_policy"), ["hybrid", "tiered", "tiered_copy"]) -def test_hdfs_zero_copy_with_ttl_move(cluster, storage_policy): - node1 = cluster.instances["node1"] - node2 = cluster.instances["node2"] - try: - node1.query( - Template( - """ - CREATE TABLE ttl_move_test ON CLUSTER test_cluster (dt DateTime, id Int64) - ENGINE=ReplicatedMergeTree('/clickhouse/tables/{cluster}/{shard}/ttl_move_test', '{replica}') - ORDER BY (dt, id) - TTL dt + INTERVAL 2 DAY TO VOLUME 'external' - SETTINGS storage_policy='$policy' - """ - ).substitute(policy=storage_policy) - ) - - node1.query("INSERT INTO ttl_move_test VALUES (now() - INTERVAL 3 DAY, 10)") - node1.query("INSERT INTO ttl_move_test VALUES (now() - INTERVAL 1 DAY, 11)") - - node1.query("OPTIMIZE TABLE ttl_move_test FINAL") - node2.query("SYSTEM SYNC REPLICA ttl_move_test", timeout=30) - - assert_eq_with_retry(node1, "SELECT count() FROM ttl_move_test", "2") - assert_eq_with_retry(node2, "SELECT count() FROM ttl_move_test", "2") - assert ( - node1.query("SELECT id FROM ttl_move_test ORDER BY id FORMAT Values") - == "(10),(11)" - ) - assert ( - node2.query("SELECT id FROM ttl_move_test ORDER BY id FORMAT Values") - == "(10),(11)" - ) - finally: - node1.query("DROP TABLE IF EXISTS ttl_move_test SYNC") - node2.query("DROP TABLE IF EXISTS ttl_move_test SYNC") - - -def test_hdfs_zero_copy_with_ttl_delete(cluster): - node1 = cluster.instances["node1"] - node2 = cluster.instances["node2"] - try: - node1.query( - """ - CREATE TABLE ttl_delete_test ON CLUSTER test_cluster (dt DateTime, id Int64) - ENGINE=ReplicatedMergeTree('/clickhouse/tables/{cluster}/{shard}/ttl_delete_test', '{replica}') - ORDER BY (dt, id) - TTL dt + INTERVAL 2 DAY - SETTINGS storage_policy='tiered' - """ - ) - - node1.query("INSERT INTO ttl_delete_test VALUES (now() - INTERVAL 3 DAY, 10)") - node1.query("INSERT INTO ttl_delete_test VALUES (now() - INTERVAL 1 DAY, 11)") - - node1.query("OPTIMIZE TABLE ttl_delete_test FINAL") - node2.query("SYSTEM SYNC REPLICA ttl_delete_test", timeout=30) - - assert_eq_with_retry(node1, "SELECT count() FROM ttl_delete_test", "1") - assert_eq_with_retry(node2, "SELECT count() FROM ttl_delete_test", "1") - - assert ( - node1.query("SELECT id FROM ttl_delete_test ORDER BY id FORMAT Values") - == "(11)" - ) - assert ( - node2.query("SELECT id FROM ttl_delete_test ORDER BY id FORMAT Values") - == "(11)" - ) - finally: - node1.query("DROP TABLE IF EXISTS ttl_delete_test SYNC") - node2.query("DROP TABLE IF EXISTS ttl_delete_test SYNC") diff --git a/tests/integration/test_reverse_dns_query/__init__.py b/tests/integration/test_reverse_dns_query/__init__.py deleted file mode 100644 index e69de29bb2d..00000000000 diff --git a/tests/integration/test_reverse_dns_query/configs/config.xml b/tests/integration/test_reverse_dns_query/configs/config.xml deleted file mode 100644 index 5ce55afa2a7..00000000000 --- a/tests/integration/test_reverse_dns_query/configs/config.xml +++ /dev/null @@ -1,3 +0,0 @@ - - 1 - diff --git a/tests/integration/test_reverse_dns_query/configs/listen_host.xml b/tests/integration/test_reverse_dns_query/configs/listen_host.xml deleted file mode 100644 index 58ef55cd3f3..00000000000 --- a/tests/integration/test_reverse_dns_query/configs/listen_host.xml +++ /dev/null @@ -1,5 +0,0 @@ - - :: - 0.0.0.0 - 1 - diff --git a/tests/integration/test_reverse_dns_query/configs/reverse_dns_function.xml b/tests/integration/test_reverse_dns_query/configs/reverse_dns_function.xml deleted file mode 100644 index 35d0a07c6a6..00000000000 --- a/tests/integration/test_reverse_dns_query/configs/reverse_dns_function.xml +++ /dev/null @@ -1,3 +0,0 @@ - - 1 - diff --git a/tests/integration/test_reverse_dns_query/coredns_config/Corefile b/tests/integration/test_reverse_dns_query/coredns_config/Corefile deleted file mode 100644 index 84d297f7cdf..00000000000 --- a/tests/integration/test_reverse_dns_query/coredns_config/Corefile +++ /dev/null @@ -1,4 +0,0 @@ -. { - forward . 127.0.0.11 - log -} diff --git a/tests/integration/test_reverse_dns_query/test.py b/tests/integration/test_reverse_dns_query/test.py deleted file mode 100644 index 6a39bd7b586..00000000000 --- a/tests/integration/test_reverse_dns_query/test.py +++ /dev/null @@ -1,56 +0,0 @@ -import pytest - -# FIXME This test is broken -# https://github.com/ClickHouse/ClickHouse/issues/53194 - -pytestmark = pytest.mark.skip - -from helpers.cluster import ClickHouseCluster, get_docker_compose_path, run_and_check -from time import sleep -import os - -DOCKER_COMPOSE_PATH = get_docker_compose_path() -SCRIPT_DIR = os.path.dirname(os.path.realpath(__file__)) - -cluster = ClickHouseCluster(__file__) - -ch_server = cluster.add_instance( - "clickhouse-server", - with_coredns=True, - main_configs=[ - "configs/config.xml", - "configs/reverse_dns_function.xml", - "configs/listen_host.xml", - ], -) - - -@pytest.fixture(scope="module") -def started_cluster(): - global cluster - try: - cluster.start() - yield cluster - - finally: - cluster.shutdown() - - -def setup_ch_server(dns_server_ip): - ch_server.exec_in_container( - (["bash", "-c", f"echo 'nameserver {dns_server_ip}' > /etc/resolv.conf"]) - ) - ch_server.exec_in_container( - (["bash", "-c", "echo 'options ndots:0' >> /etc/resolv.conf"]) - ) - ch_server.query("SYSTEM DROP DNS CACHE") - - -def test_reverse_dns_query(started_cluster): - dns_server_ip = cluster.get_instance_ip(cluster.coredns_host) - - setup_ch_server(dns_server_ip) - - for _ in range(0, 200): - response = ch_server.query("select reverseDNSQuery('2001:4860:4860::8888')") - assert response == "['dns.google']\n" diff --git a/tests/integration/test_role/test.py b/tests/integration/test_role/test.py index 9a352f43eab..0e1bbea9cea 100644 --- a/tests/integration/test_role/test.py +++ b/tests/integration/test_role/test.py @@ -287,7 +287,7 @@ def test_introspection(): assert instance.query( "SELECT name, storage from system.roles WHERE name IN ('R1', 'R2') ORDER BY name" - ) == TSV([["R1", "local directory"], ["R2", "local directory"]]) + ) == TSV([["R1", "local_directory"], ["R2", "local_directory"]]) assert instance.query( "SELECT * from system.grants WHERE user_name IN ('A', 'B') OR role_name IN ('R1', 'R2') ORDER BY user_name, role_name, access_type, database, table, column, is_partial_revoke, grant_option" @@ -301,7 +301,7 @@ def test_introspection(): ) assert instance.query( - "SELECT * from system.role_grants WHERE user_name IN ('A', 'B') OR role_name IN ('R1', 'R2') ORDER BY user_name, role_name, granted_role_name" + "SELECT user_name, role_name, granted_role_name, granted_role_is_default, with_admin_option from system.role_grants WHERE user_name IN ('A', 'B') OR role_name IN ('R1', 'R2') ORDER BY user_name, role_name, granted_role_name" ) == TSV([["A", "\\N", "R1", 1, 0], ["B", "\\N", "R2", 1, 1]]) assert instance.query( diff --git a/tests/integration/test_row_policy/test.py b/tests/integration/test_row_policy/test.py index 1933823f5d2..acf544c98aa 100644 --- a/tests/integration/test_row_policy/test.py +++ b/tests/integration/test_row_policy/test.py @@ -420,7 +420,7 @@ def test_introspection(): "mydb", "local", "5b23c389-7e18-06bf-a6bc-dd1afbbc0a97", - "users.xml", + "users_xml", "a = 1", 0, 0, @@ -433,7 +433,7 @@ def test_introspection(): "mydb", "filtered_table1", "9e8a8f62-4965-2b5e-8599-57c7b99b3549", - "users.xml", + "users_xml", "a = 1", 0, 0, @@ -446,7 +446,7 @@ def test_introspection(): "mydb", "filtered_table2", "cffae79d-b9bf-a2ef-b798-019c18470b25", - "users.xml", + "users_xml", "a + b < 1 or c - d > 5", 0, 0, @@ -459,7 +459,7 @@ def test_introspection(): "mydb", "filtered_table3", "12fc5cef-e3da-3940-ec79-d8be3911f42b", - "users.xml", + "users_xml", "c = 1", 0, 0, diff --git a/tests/integration/test_s3_zero_copy_replication/test.py b/tests/integration/test_s3_zero_copy_replication/test.py index 2a4e0eece08..0ca81a27802 100644 --- a/tests/integration/test_s3_zero_copy_replication/test.py +++ b/tests/integration/test_s3_zero_copy_replication/test.py @@ -153,96 +153,6 @@ def test_s3_zero_copy_replication(started_cluster, policy): node2.query("DROP TABLE IF EXISTS s3_test SYNC") -@pytest.mark.skip(reason="Test is flaky (and never was stable)") -def test_s3_zero_copy_on_hybrid_storage(started_cluster): - node1 = cluster.instances["node1"] - node2 = cluster.instances["node2"] - - node1.query( - """ - CREATE TABLE hybrid_test ON CLUSTER test_cluster (id UInt32, value String) - ENGINE=ReplicatedMergeTree('/clickhouse/tables/hybrid_test', '{}') - ORDER BY id - SETTINGS storage_policy='hybrid',temporary_directories_lifetime=1 - """.format( - "{replica}" - ) - ) - - node1.query("INSERT INTO hybrid_test VALUES (0,'data'),(1,'data')") - node2.query("SYSTEM SYNC REPLICA hybrid_test", timeout=30) - - assert ( - node1.query("SELECT * FROM hybrid_test ORDER BY id FORMAT Values") - == "(0,'data'),(1,'data')" - ) - assert ( - node2.query("SELECT * FROM hybrid_test ORDER BY id FORMAT Values") - == "(0,'data'),(1,'data')" - ) - - assert ( - node1.query( - "SELECT partition_id,disk_name FROM system.parts WHERE table='hybrid_test' FORMAT Values" - ) - == "('all','default')" - ) - assert ( - node2.query( - "SELECT partition_id,disk_name FROM system.parts WHERE table='hybrid_test' FORMAT Values" - ) - == "('all','default')" - ) - - node1.query("ALTER TABLE hybrid_test MOVE PARTITION ID 'all' TO DISK 's31'") - - assert ( - node1.query( - "SELECT partition_id,disk_name FROM system.parts WHERE table='hybrid_test' FORMAT Values" - ) - == "('all','s31')" - ) - assert ( - node2.query( - "SELECT partition_id,disk_name FROM system.parts WHERE table='hybrid_test' FORMAT Values" - ) - == "('all','default')" - ) - - # Total objects in S3 - s3_objects = get_large_objects_count(cluster, size=0) - - node2.query("ALTER TABLE hybrid_test MOVE PARTITION ID 'all' TO DISK 's31'") - - assert ( - node1.query( - "SELECT partition_id,disk_name FROM system.parts WHERE table='hybrid_test' FORMAT Values" - ) - == "('all','s31')" - ) - assert ( - node2.query( - "SELECT partition_id,disk_name FROM system.parts WHERE table='hybrid_test' FORMAT Values" - ) - == "('all','s31')" - ) - - # Check that after moving partition on node2 no new obects on s3 - wait_for_large_objects_count(cluster, s3_objects, size=0) - - assert ( - node1.query("SELECT * FROM hybrid_test ORDER BY id FORMAT Values") - == "(0,'data'),(1,'data')" - ) - assert ( - node2.query("SELECT * FROM hybrid_test ORDER BY id FORMAT Values") - == "(0,'data'),(1,'data')" - ) - - node1.query("DROP TABLE IF EXISTS hybrid_test SYNC") - node2.query("DROP TABLE IF EXISTS hybrid_test SYNC") - - def insert_data_time(node, table, number_of_mb, time, start=0): values = ",".join( f"({x},{time})" diff --git a/tests/integration/test_s3_zero_copy_ttl/__init__.py b/tests/integration/test_s3_zero_copy_ttl/__init__.py deleted file mode 100644 index e69de29bb2d..00000000000 diff --git a/tests/integration/test_s3_zero_copy_ttl/configs/max_delayed_streams.xml b/tests/integration/test_s3_zero_copy_ttl/configs/max_delayed_streams.xml deleted file mode 100644 index 54f7152690b..00000000000 --- a/tests/integration/test_s3_zero_copy_ttl/configs/max_delayed_streams.xml +++ /dev/null @@ -1,9 +0,0 @@ - - - - - - 10 - - - diff --git a/tests/integration/test_s3_zero_copy_ttl/configs/s3.xml b/tests/integration/test_s3_zero_copy_ttl/configs/s3.xml deleted file mode 100644 index 7bb7fa875e4..00000000000 --- a/tests/integration/test_s3_zero_copy_ttl/configs/s3.xml +++ /dev/null @@ -1,39 +0,0 @@ - - - - - s3 - http://minio1:9001/root/data/ - minio - minio123 - - - - - - -
- default -
- - s3_disk - - - - - -
- s3_disk -
- - - - - - - true - 1.0 - - - true - diff --git a/tests/integration/test_s3_zero_copy_ttl/test.py b/tests/integration/test_s3_zero_copy_ttl/test.py deleted file mode 100644 index 04bff4a44fb..00000000000 --- a/tests/integration/test_s3_zero_copy_ttl/test.py +++ /dev/null @@ -1,94 +0,0 @@ -#!/usr/bin/env python3 -import time - -import pytest -from helpers.cluster import ClickHouseCluster - -cluster = ClickHouseCluster(__file__) -node1 = cluster.add_instance( - "node1", main_configs=["configs/s3.xml"], with_minio=True, with_zookeeper=True -) -node2 = cluster.add_instance( - "node2", main_configs=["configs/s3.xml"], with_minio=True, with_zookeeper=True -) -node3 = cluster.add_instance( - "node3", main_configs=["configs/s3.xml"], with_minio=True, with_zookeeper=True -) - - -@pytest.fixture(scope="module") -def started_cluster(): - try: - cluster.start() - - yield cluster - finally: - cluster.shutdown() - - -def test_ttl_move_and_s3(started_cluster): - for i, node in enumerate([node1, node2, node3]): - node.query( - """ - CREATE TABLE s3_test_with_ttl (date DateTime, id UInt32, value String) - ENGINE=ReplicatedMergeTree('/clickhouse/tables/s3_test', '{}') - ORDER BY id - PARTITION BY id - TTL date TO DISK 's3_disk' - SETTINGS storage_policy='s3_and_default', temporary_directories_lifetime=1 - """.format( - i - ) - ) - - node1.query("SYSTEM STOP MOVES s3_test_with_ttl") - - node2.query("SYSTEM STOP MOVES s3_test_with_ttl") - - for i in range(30): - if i % 2 == 0: - node = node1 - else: - node = node2 - - node.query( - f"INSERT INTO s3_test_with_ttl SELECT now() + 5, {i}, randomPrintableASCII(1048570)" - ) - - node1.query("SYSTEM SYNC REPLICA s3_test_with_ttl") - node2.query("SYSTEM SYNC REPLICA s3_test_with_ttl") - node3.query("SYSTEM SYNC REPLICA s3_test_with_ttl") - - assert node1.query("SELECT COUNT() FROM s3_test_with_ttl") == "30\n" - assert node2.query("SELECT COUNT() FROM s3_test_with_ttl") == "30\n" - - node1.query("SYSTEM START MOVES s3_test_with_ttl") - node2.query("SYSTEM START MOVES s3_test_with_ttl") - - assert node1.query("SELECT COUNT() FROM s3_test_with_ttl") == "30\n" - assert node2.query("SELECT COUNT() FROM s3_test_with_ttl") == "30\n" - - for attempt in reversed(range(5)): - time.sleep(5) - - print( - node1.query( - "SELECT * FROM system.parts WHERE table = 's3_test_with_ttl' FORMAT Vertical" - ) - ) - - minio = cluster.minio_client - objects = minio.list_objects(cluster.minio_bucket, "data/", recursive=True) - counter = 0 - for obj in objects: - print(f"Objectname: {obj.object_name}, metadata: {obj.metadata}") - counter += 1 - - print(f"Total objects: {counter}") - - if counter == 330: - break - - print(f"Attempts remaining: {attempt}") - - assert counter == 330 diff --git a/tests/integration/test_s3_zero_copy_ttl/test_ttl_move_memory_usage.py b/tests/integration/test_s3_zero_copy_ttl/test_ttl_move_memory_usage.py deleted file mode 100644 index 5fbe426074f..00000000000 --- a/tests/integration/test_s3_zero_copy_ttl/test_ttl_move_memory_usage.py +++ /dev/null @@ -1,81 +0,0 @@ -#!/usr/bin/env python3 -import time - -import pytest - -# FIXME This test is too flaky -# https://github.com/ClickHouse/ClickHouse/issues/45887 - -pytestmark = pytest.mark.skip - -from helpers.cluster import ClickHouseCluster - - -single_node_cluster = ClickHouseCluster(__file__) -small_node = single_node_cluster.add_instance( - "small_node", main_configs=["configs/s3.xml"], with_minio=True -) - - -@pytest.fixture(scope="module") -def started_single_node_cluster(): - try: - single_node_cluster.start() - - yield single_node_cluster - finally: - single_node_cluster.shutdown() - - -def test_move_and_s3_memory_usage(started_single_node_cluster): - if small_node.is_built_with_sanitizer() or small_node.is_debug_build(): - pytest.skip("Disabled for debug and sanitizers. Too slow.") - - small_node.query( - "CREATE TABLE s3_test_with_ttl (x UInt32, a String codec(NONE), b String codec(NONE), c String codec(NONE), d String codec(NONE), e String codec(NONE)) engine = MergeTree order by x partition by x SETTINGS storage_policy='s3_and_default'" - ) - - for _ in range(10): - small_node.query( - "insert into s3_test_with_ttl select 0, repeat('a', 100), repeat('b', 100), repeat('c', 100), repeat('d', 100), repeat('e', 100) from zeros(400000) settings max_block_size = 8192, max_insert_block_size=10000000, min_insert_block_size_rows=10000000" - ) - - # After this, we should have 5 columns per 10 * 100 * 400000 ~ 400 MB; total ~2G data in partition - small_node.query( - "optimize table s3_test_with_ttl final", - settings={ - "send_logs_level": "error", - "allow_prefetched_read_pool_for_remote_filesystem": 0, - }, - ) - - small_node.query("system flush logs") - # Will take memory usage from metric_log. - # It is easier then specifying total memory limit (insert queries can hit this limit). - small_node.query("truncate table system.metric_log") - - small_node.query( - "alter table s3_test_with_ttl move partition 0 to volume 'external'", - settings={ - "send_logs_level": "error", - "allow_prefetched_read_pool_for_remote_filesystem": 0, - }, - ) - small_node.query("system flush logs") - max_usage = small_node.query( - """ - select max(m.val - am.val * 4096) from - (select toStartOfMinute(event_time) as time, max(CurrentMetric_MemoryTracking) as val from system.metric_log group by time) as m join - (select toStartOfMinute(event_time) as time, min(value) as val from system.asynchronous_metric_log where metric='jemalloc.arenas.all.pdirty' group by time) as am using time;""" - ) - # 3G limit is a big one. However, we can hit it anyway with parallel s3 writes enabled. - # Also actual value can be bigger because of memory drift. - # Increase it a little bit if test fails. - assert int(max_usage) < 3e9 - res = small_node.query( - "select * from system.errors where last_error_message like '%Memory limit%' limit 1", - settings={ - "allow_prefetched_read_pool_for_remote_filesystem": 0, - }, - ) - assert res == "" diff --git a/tests/integration/test_s3_zero_copy_ttl/test_vertical_merge_memory_usage.py b/tests/integration/test_s3_zero_copy_ttl/test_vertical_merge_memory_usage.py deleted file mode 100644 index fb9f3eb67b9..00000000000 --- a/tests/integration/test_s3_zero_copy_ttl/test_vertical_merge_memory_usage.py +++ /dev/null @@ -1,46 +0,0 @@ -#!/usr/bin/env python3 -import time - -import pytest -from helpers.cluster import ClickHouseCluster - - -single_node_cluster = ClickHouseCluster(__file__) -small_node = single_node_cluster.add_instance( - "small_node", - main_configs=["configs/s3.xml"], - user_configs=["configs/max_delayed_streams.xml"], - with_minio=True, -) - - -@pytest.fixture(scope="module") -def started_single_node_cluster(): - try: - single_node_cluster.start() - - yield single_node_cluster - finally: - single_node_cluster.shutdown() - - -def test_vertical_merge_memory_usage(started_single_node_cluster): - if small_node.is_built_with_sanitizer() or small_node.is_debug_build(): - pytest.skip("Disabled for debug and sanitizers. Too slow.") - - small_node.query( - "create table tvm2 (c0 UInt64, c1 UInt64, c2 UInt64, c3 UInt64, c4 UInt64, c5 UInt64, c6 UInt64, c7 UInt64, c8 UInt64, c9 UInt64, c10 UInt64, c11 UInt64, c12 UInt64, c13 UInt64, c14 UInt64, c15 UInt64, c16 UInt64, c17 UInt64, c18 UInt64, c19 UInt64, c20 UInt64, c21 UInt64, c22 UInt64, c23 UInt64, c24 UInt64, c25 UInt64, c26 UInt64, c27 UInt64, c28 UInt64, c29 UInt64, c30 UInt64, c31 UInt64, c32 UInt64, c33 UInt64, c34 UInt64, c35 UInt64, c36 UInt64, c37 UInt64, c38 UInt64, c39 UInt64, c40 UInt64, c41 UInt64, c42 UInt64, c43 UInt64, c44 UInt64, c45 UInt64, c46 UInt64, c47 UInt64, c48 UInt64, c49 UInt64, c50 UInt64, c51 UInt64, c52 UInt64, c53 UInt64, c54 UInt64, c55 UInt64, c56 UInt64, c57 UInt64, c58 UInt64, c59 UInt64, c60 UInt64, c61 UInt64, c62 UInt64, c63 UInt64, c64 UInt64, c65 UInt64, c66 UInt64, c67 UInt64, c68 UInt64, c69 UInt64, c70 UInt64, c71 UInt64, c72 UInt64, c73 UInt64, c74 UInt64, c75 UInt64, c76 UInt64, c77 UInt64, c78 UInt64, c79 UInt64, c80 UInt64, c81 UInt64, c82 UInt64, c83 UInt64, c84 UInt64, c85 UInt64, c86 UInt64, c87 UInt64, c88 UInt64, c89 UInt64, c90 UInt64, c91 UInt64, c92 UInt64, c93 UInt64, c94 UInt64, c95 UInt64, c96 UInt64, c97 UInt64, c98 UInt64, c99 UInt64, c100 UInt64, c101 UInt64, c102 UInt64, c103 UInt64, c104 UInt64, c105 UInt64, c106 UInt64, c107 UInt64, c108 UInt64, c109 UInt64, c110 UInt64, c111 UInt64, c112 UInt64, c113 UInt64, c114 UInt64, c115 UInt64, c116 UInt64, c117 UInt64, c118 UInt64, c119 UInt64, c120 UInt64, c121 UInt64, c122 UInt64, c123 UInt64, c124 UInt64, c125 UInt64, c126 UInt64, c127 UInt64, c128 UInt64, c129 UInt64, c130 UInt64, c131 UInt64, c132 UInt64, c133 UInt64, c134 UInt64, c135 UInt64, c136 UInt64, c137 UInt64, c138 UInt64, c139 UInt64, c140 UInt64, c141 UInt64, c142 UInt64, c143 UInt64, c144 UInt64, c145 UInt64, c146 UInt64, c147 UInt64, c148 UInt64, c149 UInt64, c150 UInt64, c151 UInt64, c152 UInt64, c153 UInt64, c154 UInt64, c155 UInt64, c156 UInt64, c157 UInt64, c158 UInt64, c159 UInt64, c160 UInt64, c161 UInt64, c162 UInt64, c163 UInt64, c164 UInt64, c165 UInt64, c166 UInt64, c167 UInt64, c168 UInt64, c169 UInt64, c170 UInt64, c171 UInt64, c172 UInt64, c173 UInt64, c174 UInt64, c175 UInt64, c176 UInt64, c177 UInt64, c178 UInt64, c179 UInt64, c180 UInt64, c181 UInt64, c182 UInt64, c183 UInt64, c184 UInt64, c185 UInt64, c186 UInt64, c187 UInt64, c188 UInt64, c189 UInt64, c190 UInt64, c191 UInt64, c192 UInt64, c193 UInt64, c194 UInt64, c195 UInt64, c196 UInt64, c197 UInt64, c198 UInt64, c199 UInt64, c200 UInt64, c201 UInt64, c202 UInt64, c203 UInt64, c204 UInt64, c205 UInt64, c206 UInt64, c207 UInt64, c208 UInt64, c209 UInt64, c210 UInt64, c211 UInt64, c212 UInt64, c213 UInt64, c214 UInt64, c215 UInt64, c216 UInt64, c217 UInt64, c218 UInt64, c219 UInt64, c220 UInt64, c221 UInt64, c222 UInt64, c223 UInt64, c224 UInt64, c225 UInt64, c226 UInt64, c227 UInt64, c228 UInt64, c229 UInt64, c230 UInt64, c231 UInt64, c232 UInt64, c233 UInt64, c234 UInt64, c235 UInt64, c236 UInt64, c237 UInt64, c238 UInt64, c239 UInt64, c240 UInt64, c241 UInt64, c242 UInt64, c243 UInt64, c244 UInt64, c245 UInt64, c246 UInt64, c247 UInt64, c248 UInt64, c249 UInt64, c250 UInt64, c251 UInt64, c252 UInt64, c253 UInt64, c254 UInt64, c255 UInt64, c256 UInt64, c257 UInt64, c258 UInt64, c259 UInt64, c260 UInt64, c261 UInt64, c262 UInt64, c263 UInt64, c264 UInt64, c265 UInt64, c266 UInt64, c267 UInt64, c268 UInt64, c269 UInt64, c270 UInt64, c271 UInt64, c272 UInt64, c273 UInt64, c274 UInt64, c275 UInt64, c276 UInt64, c277 UInt64, c278 UInt64, c279 UInt64, c280 UInt64, c281 UInt64, c282 UInt64, c283 UInt64, c284 UInt64, c285 UInt64, c286 UInt64, c287 UInt64, c288 UInt64, c289 UInt64, c290 UInt64, c291 UInt64, c292 UInt64, c293 UInt64, c294 UInt64, c295 UInt64, c296 UInt64, c297 UInt64, c298 UInt64, c299 UInt64) engine = MergeTree order by tuple() settings min_rows_for_wide_part = 10, min_bytes_for_wide_part=0, storage_policy = 's3', vertical_merge_algorithm_min_rows_to_activate=1" - ) - - small_node.query( - "insert into tvm2 select number + 0, number + 1, number + 2, number + 3, number + 4, number + 5, number + 6, number + 7, number + 8, number + 9, number + 10, number + 11, number + 12, number + 13, number + 14, number + 15, number + 16, number + 17, number + 18, number + 19, number + 20, number + 21, number + 22, number + 23, number + 24, number + 25, number + 26, number + 27, number + 28, number + 29, number + 30, number + 31, number + 32, number + 33, number + 34, number + 35, number + 36, number + 37, number + 38, number + 39, number + 40, number + 41, number + 42, number + 43, number + 44, number + 45, number + 46, number + 47, number + 48, number + 49, number + 50, number + 51, number + 52, number + 53, number + 54, number + 55, number + 56, number + 57, number + 58, number + 59, number + 60, number + 61, number + 62, number + 63, number + 64, number + 65, number + 66, number + 67, number + 68, number + 69, number + 70, number + 71, number + 72, number + 73, number + 74, number + 75, number + 76, number + 77, number + 78, number + 79, number + 80, number + 81, number + 82, number + 83, number + 84, number + 85, number + 86, number + 87, number + 88, number + 89, number + 90, number + 91, number + 92, number + 93, number + 94, number + 95, number + 96, number + 97, number + 98, number + 99, number + 100, number + 101, number + 102, number + 103, number + 104, number + 105, number + 106, number + 107, number + 108, number + 109, number + 110, number + 111, number + 112, number + 113, number + 114, number + 115, number + 116, number + 117, number + 118, number + 119, number + 120, number + 121, number + 122, number + 123, number + 124, number + 125, number + 126, number + 127, number + 128, number + 129, number + 130, number + 131, number + 132, number + 133, number + 134, number + 135, number + 136, number + 137, number + 138, number + 139, number + 140, number + 141, number + 142, number + 143, number + 144, number + 145, number + 146, number + 147, number + 148, number + 149, number + 150, number + 151, number + 152, number + 153, number + 154, number + 155, number + 156, number + 157, number + 158, number + 159, number + 160, number + 161, number + 162, number + 163, number + 164, number + 165, number + 166, number + 167, number + 168, number + 169, number + 170, number + 171, number + 172, number + 173, number + 174, number + 175, number + 176, number + 177, number + 178, number + 179, number + 180, number + 181, number + 182, number + 183, number + 184, number + 185, number + 186, number + 187, number + 188, number + 189, number + 190, number + 191, number + 192, number + 193, number + 194, number + 195, number + 196, number + 197, number + 198, number + 199, number + 200, number + 201, number + 202, number + 203, number + 204, number + 205, number + 206, number + 207, number + 208, number + 209, number + 210, number + 211, number + 212, number + 213, number + 214, number + 215, number + 216, number + 217, number + 218, number + 219, number + 220, number + 221, number + 222, number + 223, number + 224, number + 225, number + 226, number + 227, number + 228, number + 229, number + 230, number + 231, number + 232, number + 233, number + 234, number + 235, number + 236, number + 237, number + 238, number + 239, number + 240, number + 241, number + 242, number + 243, number + 244, number + 245, number + 246, number + 247, number + 248, number + 249, number + 250, number + 251, number + 252, number + 253, number + 254, number + 255, number + 256, number + 257, number + 258, number + 259, number + 260, number + 261, number + 262, number + 263, number + 264, number + 265, number + 266, number + 267, number + 268, number + 269, number + 270, number + 271, number + 272, number + 273, number + 274, number + 275, number + 276, number + 277, number + 278, number + 279, number + 280, number + 281, number + 282, number + 283, number + 284, number + 285, number + 286, number + 287, number + 288, number + 289, number + 290, number + 291, number + 292, number + 293, number + 294, number + 295, number + 296, number + 297, number + 298, number + 299 from numbers(20)" - ) - small_node.query("optimize table tvm2 final") - small_node.query("system flush logs") - - # Should be about 25M - res = small_node.query( - "select formatReadableSize(peak_memory_usage), * from system.part_log where table = 'tvm2' and database = currentDatabase() and event_date >= today() - 1 and event_type = 'MergeParts' and peak_memory_usage > 100000000 format Vertical" - ) - - assert res == "" diff --git a/tests/integration/test_session_log/.gitignore b/tests/integration/test_session_log/.gitignore new file mode 100644 index 00000000000..edf565ec632 --- /dev/null +++ b/tests/integration/test_session_log/.gitignore @@ -0,0 +1 @@ +_gen diff --git a/tests/integration/test_host_regexp_multiple_ptr_records/__init__.py b/tests/integration/test_session_log/__init__.py similarity index 100% rename from tests/integration/test_host_regexp_multiple_ptr_records/__init__.py rename to tests/integration/test_session_log/__init__.py diff --git a/tests/integration/test_session_log/configs/log.xml b/tests/integration/test_session_log/configs/log.xml new file mode 100644 index 00000000000..7a079b81e69 --- /dev/null +++ b/tests/integration/test_session_log/configs/log.xml @@ -0,0 +1,9 @@ + + + trace + /var/log/clickhouse-server/clickhouse-server.log + /var/log/clickhouse-server/clickhouse-server.err.log + 1000M + 10 + + \ No newline at end of file diff --git a/tests/integration/test_session_log/configs/ports.xml b/tests/integration/test_session_log/configs/ports.xml new file mode 100644 index 00000000000..fbaefc16b3a --- /dev/null +++ b/tests/integration/test_session_log/configs/ports.xml @@ -0,0 +1,9 @@ + + 5433 + 9001 + 9100 + + + false + + \ No newline at end of file diff --git a/tests/integration/test_session_log/configs/session_log.xml b/tests/integration/test_session_log/configs/session_log.xml new file mode 100644 index 00000000000..a0e4e3e2216 --- /dev/null +++ b/tests/integration/test_session_log/configs/session_log.xml @@ -0,0 +1,9 @@ + + + system + session_log
+ + toYYYYMM(event_date) + 7500 + + diff --git a/tests/integration/test_session_log/configs/users.xml b/tests/integration/test_session_log/configs/users.xml new file mode 100644 index 00000000000..0416dfadc8a --- /dev/null +++ b/tests/integration/test_session_log/configs/users.xml @@ -0,0 +1,23 @@ + + + + 0 + + + + + + + pass + + + pass + + + pass + + + pass + + + \ No newline at end of file diff --git a/tests/integration/test_session_log/protos/clickhouse_grpc.proto b/tests/integration/test_session_log/protos/clickhouse_grpc.proto new file mode 120000 index 00000000000..25d15f11e3b --- /dev/null +++ b/tests/integration/test_session_log/protos/clickhouse_grpc.proto @@ -0,0 +1 @@ +../../../../src/Server/grpc_protos/clickhouse_grpc.proto \ No newline at end of file diff --git a/tests/integration/test_session_log/test.py b/tests/integration/test_session_log/test.py new file mode 100644 index 00000000000..dbb39993ce3 --- /dev/null +++ b/tests/integration/test_session_log/test.py @@ -0,0 +1,289 @@ +import os + +import grpc +import pymysql.connections +import psycopg2 as py_psql +import pytest +import random +import sys +import threading + +from helpers.cluster import ClickHouseCluster, run_and_check + +POSTGRES_SERVER_PORT = 5433 +MYSQL_SERVER_PORT = 9001 +GRPC_PORT = 9100 +SESSION_LOG_MATCHING_FIELDS = "auth_id, auth_type, client_version_major, client_version_minor, client_version_patch, interface" + +SCRIPT_DIR = os.path.dirname(os.path.realpath(__file__)) +DEFAULT_ENCODING = "utf-8" + +# Use grpcio-tools to generate *pb2.py files from *.proto. +proto_dir = os.path.join(SCRIPT_DIR, "./protos") +gen_dir = os.path.join(SCRIPT_DIR, "./_gen") +os.makedirs(gen_dir, exist_ok=True) +run_and_check( + f"python3 -m grpc_tools.protoc -I{proto_dir} --python_out={gen_dir} --grpc_python_out={gen_dir} {proto_dir}/clickhouse_grpc.proto", + shell=True, +) + +sys.path.append(gen_dir) + +import clickhouse_grpc_pb2 +import clickhouse_grpc_pb2_grpc + +cluster = ClickHouseCluster(__file__) +instance = cluster.add_instance( + "node", + main_configs=[ + "configs/ports.xml", + "configs/log.xml", + "configs/session_log.xml", + ], + user_configs=["configs/users.xml"], + # Bug in TSAN reproduces in this test https://github.com/grpc/grpc/issues/29550#issuecomment-1188085387 + env_variables={ + "TSAN_OPTIONS": "report_atomic_races=0 " + os.getenv("TSAN_OPTIONS", default="") + }, +) + + +def grpc_get_url(): + return f"{instance.ip_address}:{GRPC_PORT}" + + +def grpc_create_insecure_channel(): + channel = grpc.insecure_channel(grpc_get_url()) + grpc.channel_ready_future(channel).result(timeout=2) + return channel + + +session_id_counter = 0 + + +def next_session_id(): + global session_id_counter + session_id = session_id_counter + session_id_counter += 1 + return str(session_id) + + +def grpc_query(query, user_, pass_, raise_exception): + try: + query_info = clickhouse_grpc_pb2.QueryInfo( + query=query, + session_id=next_session_id(), + user_name=user_, + password=pass_, + ) + channel = grpc_create_insecure_channel() + stub = clickhouse_grpc_pb2_grpc.ClickHouseStub(channel) + result = stub.ExecuteQuery(query_info) + if result and result.HasField("exception"): + raise Exception(result.exception.display_text) + + return result.output.decode(DEFAULT_ENCODING) + except Exception: + assert raise_exception + + +def postgres_query(query, user_, pass_, raise_exception): + try: + client = py_psql.connect( + host=instance.ip_address, + port=POSTGRES_SERVER_PORT, + user=user_, + password=pass_, + database="default", + ) + cursor = client.cursor() + cursor.execute(query) + cursor.fetchall() + except Exception: + assert raise_exception + + +def mysql_query(query, user_, pass_, raise_exception): + try: + client = pymysql.connections.Connection( + host=instance.ip_address, + user=user_, + password=pass_, + database="default", + port=MYSQL_SERVER_PORT, + ) + cursor = client.cursor(pymysql.cursors.DictCursor) + if raise_exception: + with pytest.raises(Exception): + cursor.execute(query) + else: + cursor.execute(query) + cursor.fetchall() + except Exception: + assert raise_exception + + +@pytest.fixture(scope="module") +def started_cluster(): + try: + cluster.start() + yield cluster + finally: + cluster.shutdown() + + +def test_grpc_session(started_cluster): + grpc_query("SELECT 1", "grpc_user", "pass", False) + grpc_query("SELECT 2", "grpc_user", "wrong_pass", True) + grpc_query("SELECT 3", "wrong_grpc_user", "pass", True) + + instance.query("SYSTEM FLUSH LOGS") + login_success_records = instance.query( + "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='grpc_user' AND type = 'LoginSuccess'" + ) + assert login_success_records == "grpc_user\t1\t1\n" + logout_records = instance.query( + "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='grpc_user' AND type = 'Logout'" + ) + assert logout_records == "grpc_user\t1\t1\n" + login_failure_records = instance.query( + "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='grpc_user' AND type = 'LoginFailure'" + ) + assert login_failure_records == "grpc_user\t1\t1\n" + logins_and_logouts = instance.query( + f"SELECT COUNT(*) FROM (SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'grpc_user' AND type = 'LoginSuccess' INTERSECT SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'grpc_user' AND type = 'Logout')" + ) + assert logins_and_logouts == "1\n" + + +def test_mysql_session(started_cluster): + mysql_query("SELECT 1", "mysql_user", "pass", False) + mysql_query("SELECT 2", "mysql_user", "wrong_pass", True) + mysql_query("SELECT 3", "wrong_mysql_user", "pass", True) + + instance.query("SYSTEM FLUSH LOGS") + login_success_records = instance.query( + "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='mysql_user' AND type = 'LoginSuccess'" + ) + assert login_success_records == "mysql_user\t1\t1\n" + logout_records = instance.query( + "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='mysql_user' AND type = 'Logout'" + ) + assert logout_records == "mysql_user\t1\t1\n" + login_failure_records = instance.query( + "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='mysql_user' AND type = 'LoginFailure'" + ) + assert login_failure_records == "mysql_user\t1\t1\n" + logins_and_logouts = instance.query( + f"SELECT COUNT(*) FROM (SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'mysql_user' AND type = 'LoginSuccess' INTERSECT SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'mysql_user' AND type = 'Logout')" + ) + assert logins_and_logouts == "1\n" + + +def test_postgres_session(started_cluster): + postgres_query("SELECT 1", "postgres_user", "pass", False) + postgres_query("SELECT 2", "postgres_user", "wrong_pass", True) + postgres_query("SELECT 3", "wrong_postgres_user", "pass", True) + + instance.query("SYSTEM FLUSH LOGS") + login_success_records = instance.query( + "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='postgres_user' AND type = 'LoginSuccess'" + ) + assert login_success_records == "postgres_user\t1\t1\n" + logout_records = instance.query( + "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='postgres_user' AND type = 'Logout'" + ) + assert logout_records == "postgres_user\t1\t1\n" + login_failure_records = instance.query( + "SELECT user, client_port <> 0, client_address <> toIPv6('::') FROM system.session_log WHERE user='postgres_user' AND type = 'LoginFailure'" + ) + assert login_failure_records == "postgres_user\t1\t1\n" + logins_and_logouts = instance.query( + f"SELECT COUNT(*) FROM (SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'postgres_user' AND type = 'LoginSuccess' INTERSECT SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'postgres_user' AND type = 'Logout')" + ) + assert logins_and_logouts == "1\n" + + +def test_parallel_sessions(started_cluster): + thread_list = [] + for _ in range(10): + # Sleep time does not significantly matter here, + # test should pass even without sleeping. + for function in [postgres_query, grpc_query, mysql_query]: + thread = threading.Thread( + target=function, + args=( + f"SELECT sleep({random.uniform(0.03, 0.04)})", + "parallel_user", + "pass", + False, + ), + ) + thread.start() + thread_list.append(thread) + thread = threading.Thread( + target=function, + args=( + f"SELECT sleep({random.uniform(0.03, 0.04)})", + "parallel_user", + "wrong_pass", + True, + ), + ) + thread.start() + thread_list.append(thread) + thread = threading.Thread( + target=function, + args=( + f"SELECT sleep({random.uniform(0.03, 0.04)})", + "wrong_parallel_user", + "pass", + True, + ), + ) + thread.start() + thread_list.append(thread) + + for thread in thread_list: + thread.join() + + instance.query("SYSTEM FLUSH LOGS") + port_0_sessions = instance.query( + f"SELECT COUNT(*) FROM system.session_log WHERE user = 'parallel_user'" + ) + assert port_0_sessions == "90\n" + + port_0_sessions = instance.query( + f"SELECT COUNT(*) FROM system.session_log WHERE user = 'parallel_user' AND client_port = 0" + ) + assert port_0_sessions == "0\n" + + address_0_sessions = instance.query( + f"SELECT COUNT(*) FROM system.session_log WHERE user = 'parallel_user' AND client_address = toIPv6('::')" + ) + assert address_0_sessions == "0\n" + + grpc_sessions = instance.query( + f"SELECT COUNT(*) FROM system.session_log WHERE user = 'parallel_user' AND interface = 'gRPC'" + ) + assert grpc_sessions == "30\n" + + mysql_sessions = instance.query( + f"SELECT COUNT(*) FROM system.session_log WHERE user = 'parallel_user' AND interface = 'MySQL'" + ) + assert mysql_sessions == "30\n" + + postgres_sessions = instance.query( + f"SELECT COUNT(*) FROM system.session_log WHERE user = 'parallel_user' AND interface = 'PostgreSQL'" + ) + assert postgres_sessions == "30\n" + + logins_and_logouts = instance.query( + f"SELECT COUNT(*) FROM (SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'parallel_user' AND type = 'LoginSuccess' INTERSECT SELECT {SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = 'parallel_user' AND type = 'Logout')" + ) + assert logins_and_logouts == "30\n" + + logout_failure_sessions = instance.query( + f"SELECT COUNT(*) FROM system.session_log WHERE user = 'parallel_user' AND type = 'LoginFailure'" + ) + assert logout_failure_sessions == "30\n" diff --git a/tests/integration/test_settings_profile/test.py b/tests/integration/test_settings_profile/test.py index 13acc3f1042..8770cc82064 100644 --- a/tests/integration/test_settings_profile/test.py +++ b/tests/integration/test_settings_profile/test.py @@ -88,7 +88,7 @@ def test_smoke(): ) ) assert system_settings_profile("xyz") == [ - ["xyz", "local directory", 1, 0, "['robin']", "[]"] + ["xyz", "local_directory", 1, 0, "['robin']", "[]"] ] assert system_settings_profile_elements(profile_name="xyz") == [ [ @@ -120,7 +120,7 @@ def test_smoke(): instance.query("SET max_memory_usage = 80000000", user="robin") instance.query("SET max_memory_usage = 120000000", user="robin") assert system_settings_profile("xyz") == [ - ["xyz", "local directory", 1, 0, "[]", "[]"] + ["xyz", "local_directory", 1, 0, "[]", "[]"] ] assert system_settings_profile_elements(user_name="robin") == [] @@ -201,7 +201,7 @@ def test_settings_from_granted_role(): ) ) assert system_settings_profile("xyz") == [ - ["xyz", "local directory", 2, 0, "[]", "[]"] + ["xyz", "local_directory", 2, 0, "[]", "[]"] ] assert system_settings_profile_elements(profile_name="xyz") == [ [ @@ -276,7 +276,7 @@ def test_settings_from_granted_role(): ) ) assert system_settings_profile("xyz") == [ - ["xyz", "local directory", 2, 0, "['worker']", "[]"] + ["xyz", "local_directory", 2, 0, "['worker']", "[]"] ] instance.query("ALTER SETTINGS PROFILE xyz TO NONE") @@ -293,7 +293,7 @@ def test_settings_from_granted_role(): ) instance.query("SET max_memory_usage = 120000000", user="robin") assert system_settings_profile("xyz") == [ - ["xyz", "local directory", 2, 0, "[]", "[]"] + ["xyz", "local_directory", 2, 0, "[]", "[]"] ] @@ -323,7 +323,7 @@ def test_inheritance(): ) assert system_settings_profile("xyz") == [ - ["xyz", "local directory", 1, 0, "[]", "[]"] + ["xyz", "local_directory", 1, 0, "[]", "[]"] ] assert system_settings_profile_elements(profile_name="xyz") == [ [ @@ -340,7 +340,7 @@ def test_inheritance(): ] ] assert system_settings_profile("alpha") == [ - ["alpha", "local directory", 1, 0, "['robin']", "[]"] + ["alpha", "local_directory", 1, 0, "['robin']", "[]"] ] assert system_settings_profile_elements(profile_name="alpha") == [ ["alpha", "\\N", "\\N", 0, "\\N", "\\N", "\\N", "\\N", "\\N", "xyz"] diff --git a/tests/integration/test_storage_nats/__init__.py b/tests/integration/test_storage_nats/__init__.py deleted file mode 100644 index e69de29bb2d..00000000000 diff --git a/tests/integration/test_storage_nats/clickhouse_path/format_schemas/nats.proto b/tests/integration/test_storage_nats/clickhouse_path/format_schemas/nats.proto deleted file mode 100755 index 090ed917cdd..00000000000 --- a/tests/integration/test_storage_nats/clickhouse_path/format_schemas/nats.proto +++ /dev/null @@ -1,6 +0,0 @@ -syntax = "proto3"; - - message ProtoKeyValue { - uint64 key = 1; - string value = 2; - } diff --git a/tests/integration/test_storage_nats/configs/macros.xml b/tests/integration/test_storage_nats/configs/macros.xml deleted file mode 100644 index 4aa547e049e..00000000000 --- a/tests/integration/test_storage_nats/configs/macros.xml +++ /dev/null @@ -1,7 +0,0 @@ - - - nats1:4444 - macro - JSONEachRow - - diff --git a/tests/integration/test_storage_nats/configs/named_collection.xml b/tests/integration/test_storage_nats/configs/named_collection.xml deleted file mode 100644 index 15817f321f0..00000000000 --- a/tests/integration/test_storage_nats/configs/named_collection.xml +++ /dev/null @@ -1,13 +0,0 @@ - - - - nats1:4444 - named - JSONEachRow - 111 - 12 - click - house - - - diff --git a/tests/integration/test_storage_nats/configs/nats.xml b/tests/integration/test_storage_nats/configs/nats.xml deleted file mode 100644 index 0a8be9122ad..00000000000 --- a/tests/integration/test_storage_nats/configs/nats.xml +++ /dev/null @@ -1,6 +0,0 @@ - - - click - house - - diff --git a/tests/integration/test_storage_nats/configs/users.xml b/tests/integration/test_storage_nats/configs/users.xml deleted file mode 100644 index 2cef0a6de3c..00000000000 --- a/tests/integration/test_storage_nats/configs/users.xml +++ /dev/null @@ -1,7 +0,0 @@ - - - - 1 - - - diff --git a/tests/integration/test_storage_nats/nats_certs.sh b/tests/integration/test_storage_nats/nats_certs.sh deleted file mode 100755 index 689221c39e4..00000000000 --- a/tests/integration/test_storage_nats/nats_certs.sh +++ /dev/null @@ -1,13 +0,0 @@ -#!/bin/bash -set -euxo pipefail - -mkdir -p "${NATS_CERT_DIR}/ca" -mkdir -p "${NATS_CERT_DIR}/nats" -openssl req -newkey rsa:4096 -x509 -days 365 -nodes -batch -keyout "${NATS_CERT_DIR}/ca/ca-key.pem" -out "${NATS_CERT_DIR}/ca/ca-cert.pem" -subj "/C=RU/ST=Some-State/O=Internet Widgits Pty Ltd/CN=ca" -openssl req -newkey rsa:4096 -nodes -batch -keyout "${NATS_CERT_DIR}/nats/server-key.pem" -out "${NATS_CERT_DIR}/nats/server-req.pem" -subj "/C=RU/ST=Some-State/O=Internet Widgits Pty Ltd/CN=server" -openssl x509 -req -days 365 -in "${NATS_CERT_DIR}/nats/server-req.pem" -CA "${NATS_CERT_DIR}/ca/ca-cert.pem" -CAkey "${NATS_CERT_DIR}/ca/ca-key.pem" -CAcreateserial -out "${NATS_CERT_DIR}/nats/server-cert.pem" -extfile <( -cat <<-EOF -subjectAltName = DNS:localhost, DNS:nats1 -EOF -) -rm -f "${NATS_CERT_DIR}/nats/server-req.pem" diff --git a/tests/integration/test_storage_nats/nats_pb2.py b/tests/integration/test_storage_nats/nats_pb2.py deleted file mode 100644 index e9e5cb72363..00000000000 --- a/tests/integration/test_storage_nats/nats_pb2.py +++ /dev/null @@ -1,37 +0,0 @@ -# -*- coding: utf-8 -*- -# Generated by the protocol buffer compiler. DO NOT EDIT! -# source: clickhouse_path/format_schemas/nats.proto -"""Generated protocol buffer code.""" -from google.protobuf import descriptor as _descriptor -from google.protobuf import descriptor_pool as _descriptor_pool -from google.protobuf import message as _message -from google.protobuf import reflection as _reflection -from google.protobuf import symbol_database as _symbol_database - -# @@protoc_insertion_point(imports) - -_sym_db = _symbol_database.Default() - - -DESCRIPTOR = _descriptor_pool.Default().AddSerializedFile( - b'\n)clickhouse_path/format_schemas/nats.proto"+\n\rProtoKeyValue\x12\x0b\n\x03key\x18\x01 \x01(\x04\x12\r\n\x05value\x18\x02 \x01(\tb\x06proto3' -) - - -_PROTOKEYVALUE = DESCRIPTOR.message_types_by_name["ProtoKeyValue"] -ProtoKeyValue = _reflection.GeneratedProtocolMessageType( - "ProtoKeyValue", - (_message.Message,), - { - "DESCRIPTOR": _PROTOKEYVALUE, - "__module__": "clickhouse_path.format_schemas.nats_pb2" - # @@protoc_insertion_point(class_scope:ProtoKeyValue) - }, -) -_sym_db.RegisterMessage(ProtoKeyValue) - -if _descriptor._USE_C_DESCRIPTORS == False: - DESCRIPTOR._options = None - _PROTOKEYVALUE._serialized_start = 45 - _PROTOKEYVALUE._serialized_end = 88 -# @@protoc_insertion_point(module_scope) diff --git a/tests/integration/test_storage_nats/test.py b/tests/integration/test_storage_nats/test.py deleted file mode 100644 index 4d7e4cf813d..00000000000 --- a/tests/integration/test_storage_nats/test.py +++ /dev/null @@ -1,1875 +0,0 @@ -import pytest - -# FIXME This test is too flaky -# https://github.com/ClickHouse/ClickHouse/issues/39185 - -pytestmark = pytest.mark.skip - -import json -import os.path as p -import random -import subprocess -import threading -import logging -import time -from random import randrange -import math - -import asyncio -from google.protobuf.internal.encoder import _VarintBytes -from helpers.client import QueryRuntimeException -from helpers.cluster import ClickHouseCluster, check_nats_is_available, nats_connect_ssl -from helpers.test_tools import TSV - -from . import nats_pb2 - -cluster = ClickHouseCluster(__file__) -instance = cluster.add_instance( - "instance", - main_configs=[ - "configs/nats.xml", - "configs/macros.xml", - "configs/named_collection.xml", - ], - user_configs=["configs/users.xml"], - with_nats=True, - clickhouse_path_dir="clickhouse_path", -) - - -# Helpers - - -def wait_nats_to_start(nats_port, ssl_ctx=None, timeout=180): - start = time.time() - while time.time() - start < timeout: - try: - if asyncio.run(check_nats_is_available(nats_port, ssl_ctx=ssl_ctx)): - logging.debug("NATS is available") - return - time.sleep(0.5) - except Exception as ex: - logging.debug("Can't connect to NATS " + str(ex)) - time.sleep(0.5) - - -def nats_check_result(result, check=False, ref_file="test_nats_json.reference"): - fpath = p.join(p.dirname(__file__), ref_file) - with open(fpath) as reference: - if check: - assert TSV(result) == TSV(reference) - else: - return TSV(result) == TSV(reference) - - -def kill_nats(nats_id): - p = subprocess.Popen(("docker", "stop", nats_id), stdout=subprocess.PIPE) - p.communicate() - return p.returncode == 0 - - -def revive_nats(nats_id, nats_port): - p = subprocess.Popen(("docker", "start", nats_id), stdout=subprocess.PIPE) - p.communicate() - wait_nats_to_start(nats_port) - - -# Fixtures - - -@pytest.fixture(scope="module") -def nats_cluster(): - try: - cluster.start() - logging.debug("nats_id is {}".format(instance.cluster.nats_docker_id)) - instance.query("CREATE DATABASE test") - - yield cluster - - finally: - cluster.shutdown() - - -@pytest.fixture(autouse=True) -def nats_setup_teardown(): - print("NATS is available - running test") - yield # run test - instance.query("DROP DATABASE test SYNC") - instance.query("CREATE DATABASE test") - - -# Tests - - -async def nats_produce_messages(cluster_inst, subject, messages=(), bytes=None): - nc = await nats_connect_ssl( - cluster_inst.nats_port, - user="click", - password="house", - ssl_ctx=cluster_inst.nats_ssl_context, - ) - logging.debug("NATS connection status: " + str(nc.is_connected)) - - for message in messages: - await nc.publish(subject, message.encode()) - if bytes is not None: - await nc.publish(subject, bytes) - logging.debug("Finished publising to " + subject) - - await nc.close() - return messages - - -def check_table_is_ready(instance, table_name): - try: - instance.query("SELECT * FROM {}".format(table_name)) - return True - except Exception: - return False - - -def test_nats_select(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'select', - nats_format = 'JSONEachRow', - nats_row_delimiter = '\\n'; - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - messages = [] - for i in range(50): - messages.append(json.dumps({"key": i, "value": i})) - asyncio.run(nats_produce_messages(nats_cluster, "select", messages)) - - # The order of messages in select * from test.nats is not guaranteed, so sleep to collect everything in one select - time.sleep(1) - - result = "" - while True: - result += instance.query( - "SELECT * FROM test.nats ORDER BY key", ignore_error=True - ) - if nats_check_result(result): - break - - nats_check_result(result, True) - - -def test_nats_select_empty(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'empty', - nats_format = 'TSV', - nats_row_delimiter = '\\n'; - """ - ) - - assert int(instance.query("SELECT count() FROM test.nats")) == 0 - - -def test_nats_json_without_delimiter(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'json', - nats_format = 'JSONEachRow'; - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - messages = "" - for i in range(25): - messages += json.dumps({"key": i, "value": i}) + "\n" - - all_messages = [messages] - asyncio.run(nats_produce_messages(nats_cluster, "json", all_messages)) - - messages = "" - for i in range(25, 50): - messages += json.dumps({"key": i, "value": i}) + "\n" - all_messages = [messages] - asyncio.run(nats_produce_messages(nats_cluster, "json", all_messages)) - - time.sleep(1) - - result = "" - time_limit_sec = 60 - deadline = time.monotonic() + time_limit_sec - - while time.monotonic() < deadline: - result += instance.query( - "SELECT * FROM test.nats ORDER BY key", ignore_error=True - ) - if nats_check_result(result): - break - - nats_check_result(result, True) - - -def test_nats_csv_with_delimiter(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'csv', - nats_format = 'CSV', - nats_row_delimiter = '\\n'; - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - messages = [] - for i in range(50): - messages.append("{i}, {i}".format(i=i)) - - asyncio.run(nats_produce_messages(nats_cluster, "csv", messages)) - - time.sleep(1) - - result = "" - for _ in range(60): - result += instance.query( - "SELECT * FROM test.nats ORDER BY key", ignore_error=True - ) - if nats_check_result(result): - break - - nats_check_result(result, True) - - -def test_nats_tsv_with_delimiter(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'tsv', - nats_format = 'TSV', - nats_row_delimiter = '\\n'; - CREATE TABLE test.view (key UInt64, value UInt64) - ENGINE = MergeTree() - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.nats; - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - messages = [] - for i in range(50): - messages.append("{i}\t{i}".format(i=i)) - - asyncio.run(nats_produce_messages(nats_cluster, "tsv", messages)) - - result = "" - for _ in range(60): - result = instance.query("SELECT * FROM test.view ORDER BY key") - if nats_check_result(result): - break - - nats_check_result(result, True) - - -# - - -def test_nats_macros(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = '{nats_url}', - nats_subjects = '{nats_subjects}', - nats_format = '{nats_format}' - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - message = "" - for i in range(50): - message += json.dumps({"key": i, "value": i}) + "\n" - asyncio.run(nats_produce_messages(nats_cluster, "macro", [message])) - - time.sleep(1) - - result = "" - for _ in range(60): - result += instance.query( - "SELECT * FROM test.nats ORDER BY key", ignore_error=True - ) - if nats_check_result(result): - break - - nats_check_result(result, True) - - -def test_nats_materialized_view(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'mv', - nats_format = 'JSONEachRow', - nats_row_delimiter = '\\n'; - CREATE TABLE test.view (key UInt64, value UInt64) - ENGINE = MergeTree() - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.nats; - - CREATE TABLE test.view2 (key UInt64, value UInt64) - ENGINE = MergeTree() - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer2 TO test.view2 AS - SELECT * FROM test.nats group by (key, value); - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - messages = [] - for i in range(50): - messages.append(json.dumps({"key": i, "value": i})) - - asyncio.run(nats_produce_messages(nats_cluster, "mv", messages)) - - time_limit_sec = 60 - deadline = time.monotonic() + time_limit_sec - - while time.monotonic() < deadline: - result = instance.query("SELECT * FROM test.view ORDER BY key") - if nats_check_result(result): - break - - nats_check_result(result, True) - - deadline = time.monotonic() + time_limit_sec - - while time.monotonic() < deadline: - result = instance.query("SELECT * FROM test.view2 ORDER BY key") - if nats_check_result(result): - break - - nats_check_result(result, True) - - -def test_nats_materialized_view_with_subquery(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'mvsq', - nats_format = 'JSONEachRow', - nats_row_delimiter = '\\n'; - CREATE TABLE test.view (key UInt64, value UInt64) - ENGINE = MergeTree() - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM (SELECT * FROM test.nats); - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - messages = [] - for i in range(50): - messages.append(json.dumps({"key": i, "value": i})) - asyncio.run(nats_produce_messages(nats_cluster, "mvsq", messages)) - - time_limit_sec = 60 - deadline = time.monotonic() + time_limit_sec - - while time.monotonic() < deadline: - result = instance.query("SELECT * FROM test.view ORDER BY key") - if nats_check_result(result): - break - - nats_check_result(result, True) - - -def test_nats_many_materialized_views(nats_cluster): - instance.query( - """ - DROP TABLE IF EXISTS test.view1; - DROP TABLE IF EXISTS test.view2; - DROP TABLE IF EXISTS test.consumer1; - DROP TABLE IF EXISTS test.consumer2; - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'mmv', - nats_format = 'JSONEachRow', - nats_row_delimiter = '\\n'; - CREATE TABLE test.view1 (key UInt64, value UInt64) - ENGINE = MergeTree() - ORDER BY key; - CREATE TABLE test.view2 (key UInt64, value UInt64) - ENGINE = MergeTree() - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer1 TO test.view1 AS - SELECT * FROM test.nats; - CREATE MATERIALIZED VIEW test.consumer2 TO test.view2 AS - SELECT * FROM test.nats; - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - messages = [] - for i in range(50): - messages.append(json.dumps({"key": i, "value": i})) - asyncio.run(nats_produce_messages(nats_cluster, "mmv", messages)) - - time_limit_sec = 60 - deadline = time.monotonic() + time_limit_sec - - while time.monotonic() < deadline: - result1 = instance.query("SELECT * FROM test.view1 ORDER BY key") - result2 = instance.query("SELECT * FROM test.view2 ORDER BY key") - if nats_check_result(result1) and nats_check_result(result2): - break - - instance.query( - """ - DROP TABLE test.consumer1; - DROP TABLE test.consumer2; - DROP TABLE test.view1; - DROP TABLE test.view2; - """ - ) - - nats_check_result(result1, True) - nats_check_result(result2, True) - - -def test_nats_protobuf(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value String) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'pb', - nats_format = 'Protobuf', - nats_schema = 'nats.proto:ProtoKeyValue'; - CREATE TABLE test.view (key UInt64, value UInt64) - ENGINE = MergeTree() - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.nats; - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - data = b"" - for i in range(0, 20): - msg = nats_pb2.ProtoKeyValue() - msg.key = i - msg.value = str(i) - serialized_msg = msg.SerializeToString() - data = data + _VarintBytes(len(serialized_msg)) + serialized_msg - asyncio.run(nats_produce_messages(nats_cluster, "pb", bytes=data)) - data = b"" - for i in range(20, 21): - msg = nats_pb2.ProtoKeyValue() - msg.key = i - msg.value = str(i) - serialized_msg = msg.SerializeToString() - data = data + _VarintBytes(len(serialized_msg)) + serialized_msg - asyncio.run(nats_produce_messages(nats_cluster, "pb", bytes=data)) - data = b"" - for i in range(21, 50): - msg = nats_pb2.ProtoKeyValue() - msg.key = i - msg.value = str(i) - serialized_msg = msg.SerializeToString() - data = data + _VarintBytes(len(serialized_msg)) + serialized_msg - asyncio.run(nats_produce_messages(nats_cluster, "pb", bytes=data)) - - result = "" - time_limit_sec = 60 - deadline = time.monotonic() + time_limit_sec - - while time.monotonic() < deadline: - result = instance.query("SELECT * FROM test.view ORDER BY key") - if nats_check_result(result): - break - - nats_check_result(result, True) - - -def test_nats_big_message(nats_cluster): - # Create batchs of messages of size ~100Kb - nats_messages = 1000 - batch_messages = 1000 - messages = [ - json.dumps({"key": i, "value": "x" * 100}) * batch_messages - for i in range(nats_messages) - ] - - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value String) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'big', - nats_format = 'JSONEachRow'; - CREATE TABLE test.view (key UInt64, value String) - ENGINE = MergeTree - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.nats; - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - asyncio.run(nats_produce_messages(nats_cluster, "big", messages)) - - while True: - result = instance.query("SELECT count() FROM test.view") - if int(result) == batch_messages * nats_messages: - break - - assert ( - int(result) == nats_messages * batch_messages - ), "ClickHouse lost some messages: {}".format(result) - - -def test_nats_mv_combo(nats_cluster): - NUM_MV = 5 - NUM_CONSUMERS = 4 - - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'combo', - nats_num_consumers = {}, - nats_format = 'JSONEachRow', - nats_row_delimiter = '\\n'; - """.format( - NUM_CONSUMERS - ) - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - for mv_id in range(NUM_MV): - instance.query( - """ - DROP TABLE IF EXISTS test.combo_{0}; - DROP TABLE IF EXISTS test.combo_{0}_mv; - CREATE TABLE test.combo_{0} (key UInt64, value UInt64) - ENGINE = MergeTree() - ORDER BY key; - CREATE MATERIALIZED VIEW test.combo_{0}_mv TO test.combo_{0} AS - SELECT * FROM test.nats; - """.format( - mv_id - ) - ) - - time.sleep(2) - - i = [0] - messages_num = 10000 - - def produce(): - messages = [] - for _ in range(messages_num): - messages.append(json.dumps({"key": i[0], "value": i[0]})) - i[0] += 1 - asyncio.run(nats_produce_messages(nats_cluster, "combo", messages)) - - threads = [] - threads_num = 20 - - for _ in range(threads_num): - threads.append(threading.Thread(target=produce)) - for thread in threads: - time.sleep(random.uniform(0, 1)) - thread.start() - - while True: - result = 0 - for mv_id in range(NUM_MV): - result += int( - instance.query("SELECT count() FROM test.combo_{0}".format(mv_id)) - ) - if int(result) == messages_num * threads_num * NUM_MV: - break - time.sleep(1) - - for thread in threads: - thread.join() - - for mv_id in range(NUM_MV): - instance.query( - """ - DROP TABLE test.combo_{0}_mv; - DROP TABLE test.combo_{0}; - """.format( - mv_id - ) - ) - - assert ( - int(result) == messages_num * threads_num * NUM_MV - ), "ClickHouse lost some messages: {}".format(result) - - -def test_nats_insert(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'insert', - nats_format = 'TSV', - nats_row_delimiter = '\\n'; - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - values = [] - for i in range(50): - values.append("({i}, {i})".format(i=i)) - values = ",".join(values) - - insert_messages = [] - - async def sub_to_nats(): - nc = await nats_connect_ssl( - nats_cluster.nats_port, - user="click", - password="house", - ssl_ctx=nats_cluster.nats_ssl_context, - ) - sub = await nc.subscribe("insert") - await sub.unsubscribe(50) - async for msg in sub.messages: - insert_messages.append(msg.data.decode()) - - await sub.drain() - await nc.drain() - - def run_sub(): - asyncio.run(sub_to_nats()) - - thread = threading.Thread(target=run_sub) - thread.start() - time.sleep(1) - - while True: - try: - instance.query("INSERT INTO test.nats VALUES {}".format(values)) - break - except QueryRuntimeException as e: - if "Local: Timed out." in str(e): - continue - else: - raise - thread.join() - - result = "\n".join(insert_messages) - nats_check_result(result, True) - - -def test_nats_many_subjects_insert_wrong(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'insert1,insert2.>,insert3.*.foo', - nats_format = 'TSV', - nats_row_delimiter = '\\n'; - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - values = [] - for i in range(50): - values.append("({i}, {i})".format(i=i)) - values = ",".join(values) - - # no subject specified - instance.query_and_get_error("INSERT INTO test.nats VALUES {}".format(values)) - - # can't insert into wildcard subjects - instance.query_and_get_error( - "INSERT INTO test.nats SETTINGS stream_like_engine_insert_queue='insert2.>' VALUES {}".format( - values - ) - ) - instance.query_and_get_error( - "INSERT INTO test.nats SETTINGS stream_like_engine_insert_queue='insert3.*.foo' VALUES {}".format( - values - ) - ) - - # specified subject is not among engine's subjects - instance.query_and_get_error( - "INSERT INTO test.nats SETTINGS stream_like_engine_insert_queue='insert4' VALUES {}".format( - values - ) - ) - instance.query_and_get_error( - "INSERT INTO test.nats SETTINGS stream_like_engine_insert_queue='insert3.foo.baz' VALUES {}".format( - values - ) - ) - instance.query_and_get_error( - "INSERT INTO test.nats SETTINGS stream_like_engine_insert_queue='foo.insert2' VALUES {}".format( - values - ) - ) - - -def test_nats_many_subjects_insert_right(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'right_insert1,right_insert2', - nats_format = 'TSV', - nats_row_delimiter = '\\n'; - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - values = [] - for i in range(50): - values.append("({i}, {i})".format(i=i)) - values = ",".join(values) - - insert_messages = [] - - async def sub_to_nats(): - nc = await nats_connect_ssl( - nats_cluster.nats_port, - user="click", - password="house", - ssl_ctx=nats_cluster.nats_ssl_context, - ) - sub = await nc.subscribe("right_insert1") - await sub.unsubscribe(50) - async for msg in sub.messages: - insert_messages.append(msg.data.decode()) - - await sub.drain() - await nc.drain() - - def run_sub(): - asyncio.run(sub_to_nats()) - - thread = threading.Thread(target=run_sub) - thread.start() - time.sleep(1) - - while True: - try: - instance.query( - "INSERT INTO test.nats SETTINGS stream_like_engine_insert_queue='right_insert1' VALUES {}".format( - values - ) - ) - break - except QueryRuntimeException as e: - if "Local: Timed out." in str(e): - continue - else: - raise - thread.join() - - result = "\n".join(insert_messages) - nats_check_result(result, True) - - -def test_nats_many_inserts(nats_cluster): - instance.query( - """ - DROP TABLE IF EXISTS test.nats_many; - DROP TABLE IF EXISTS test.nats_consume; - DROP TABLE IF EXISTS test.view_many; - DROP TABLE IF EXISTS test.consumer_many; - CREATE TABLE test.nats_many (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'many_inserts', - nats_format = 'TSV', - nats_row_delimiter = '\\n'; - CREATE TABLE test.nats_consume (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'many_inserts', - nats_format = 'TSV', - nats_row_delimiter = '\\n'; - CREATE TABLE test.view_many (key UInt64, value UInt64) - ENGINE = MergeTree - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer_many TO test.view_many AS - SELECT * FROM test.nats_consume; - """ - ) - while not check_table_is_ready(instance, "test.nats_consume"): - logging.debug("Table test.nats_consume is not yet ready") - time.sleep(0.5) - - messages_num = 10000 - values = [] - for i in range(messages_num): - values.append("({i}, {i})".format(i=i)) - values = ",".join(values) - - def insert(): - while True: - try: - instance.query("INSERT INTO test.nats_many VALUES {}".format(values)) - break - except QueryRuntimeException as e: - if "Local: Timed out." in str(e): - continue - else: - raise - - threads = [] - threads_num = 10 - for _ in range(threads_num): - threads.append(threading.Thread(target=insert)) - for thread in threads: - time.sleep(random.uniform(0, 1)) - thread.start() - - for thread in threads: - thread.join() - - time_limit_sec = 300 - deadline = time.monotonic() + time_limit_sec - - while time.monotonic() < deadline: - result = instance.query("SELECT count() FROM test.view_many") - print(result, messages_num * threads_num) - if int(result) >= messages_num * threads_num: - break - time.sleep(1) - - instance.query( - """ - DROP TABLE test.nats_consume; - DROP TABLE test.nats_many; - DROP TABLE test.consumer_many; - DROP TABLE test.view_many; - """ - ) - - assert ( - int(result) == messages_num * threads_num - ), "ClickHouse lost some messages or got duplicated ones. Total count: {}".format( - result - ) - - -def test_nats_overloaded_insert(nats_cluster): - instance.query( - """ - DROP TABLE IF EXISTS test.view_overload; - DROP TABLE IF EXISTS test.consumer_overload; - DROP TABLE IF EXISTS test.nats_consume; - CREATE TABLE test.nats_consume (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'over', - nats_num_consumers = 5, - nats_max_block_size = 10000, - nats_format = 'TSV', - nats_row_delimiter = '\\n'; - CREATE TABLE test.nats_overload (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'over', - nats_format = 'TSV', - nats_row_delimiter = '\\n'; - CREATE TABLE test.view_overload (key UInt64, value UInt64) - ENGINE = MergeTree - ORDER BY key - SETTINGS old_parts_lifetime=5, cleanup_delay_period=2, cleanup_delay_period_random_add=3, - cleanup_thread_preferred_points_per_iteration=0; - CREATE MATERIALIZED VIEW test.consumer_overload TO test.view_overload AS - SELECT * FROM test.nats_consume; - """ - ) - while not check_table_is_ready(instance, "test.nats_consume"): - logging.debug("Table test.nats_consume is not yet ready") - time.sleep(0.5) - - messages_num = 100000 - - def insert(): - values = [] - for i in range(messages_num): - values.append("({i}, {i})".format(i=i)) - values = ",".join(values) - - while True: - try: - instance.query( - "INSERT INTO test.nats_overload VALUES {}".format(values) - ) - break - except QueryRuntimeException as e: - if "Local: Timed out." in str(e): - continue - else: - raise - - threads = [] - threads_num = 5 - for _ in range(threads_num): - threads.append(threading.Thread(target=insert)) - for thread in threads: - time.sleep(random.uniform(0, 1)) - thread.start() - - time_limit_sec = 300 - deadline = time.monotonic() + time_limit_sec - - while time.monotonic() < deadline: - result = instance.query("SELECT count() FROM test.view_overload") - time.sleep(1) - if int(result) >= messages_num * threads_num: - break - - instance.query( - """ - DROP TABLE test.consumer_overload; - DROP TABLE test.view_overload; - DROP TABLE test.nats_consume; - DROP TABLE test.nats_overload; - """ - ) - - for thread in threads: - thread.join() - - assert ( - int(result) == messages_num * threads_num - ), "ClickHouse lost some messages or got duplicated ones. Total count: {}".format( - result - ) - - -def test_nats_virtual_column(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats_virtuals (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'virtuals', - nats_format = 'JSONEachRow'; - CREATE MATERIALIZED VIEW test.view Engine=Log AS - SELECT value, key, _subject FROM test.nats_virtuals; - """ - ) - while not check_table_is_ready(instance, "test.nats_virtuals"): - logging.debug("Table test.nats_virtuals is not yet ready") - time.sleep(0.5) - - message_num = 10 - i = 0 - messages = [] - for _ in range(message_num): - messages.append(json.dumps({"key": i, "value": i})) - i += 1 - - asyncio.run(nats_produce_messages(nats_cluster, "virtuals", messages)) - - while True: - result = instance.query("SELECT count() FROM test.view") - time.sleep(1) - if int(result) == message_num: - break - - result = instance.query( - """ - SELECT key, value, _subject - FROM test.view ORDER BY key - """ - ) - - expected = """\ -0 0 virtuals -1 1 virtuals -2 2 virtuals -3 3 virtuals -4 4 virtuals -5 5 virtuals -6 6 virtuals -7 7 virtuals -8 8 virtuals -9 9 virtuals -""" - - instance.query( - """ - DROP TABLE test.nats_virtuals; - DROP TABLE test.view; - """ - ) - - assert TSV(result) == TSV(expected) - - -def test_nats_virtual_column_with_materialized_view(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats_virtuals_mv (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'virtuals_mv', - nats_format = 'JSONEachRow'; - CREATE TABLE test.view (key UInt64, value UInt64, subject String) ENGINE = MergeTree() - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT *, _subject as subject - FROM test.nats_virtuals_mv; - """ - ) - while not check_table_is_ready(instance, "test.nats_virtuals_mv"): - logging.debug("Table test.nats_virtuals_mv is not yet ready") - time.sleep(0.5) - - message_num = 10 - i = 0 - messages = [] - for _ in range(message_num): - messages.append(json.dumps({"key": i, "value": i})) - i += 1 - - asyncio.run(nats_produce_messages(nats_cluster, "virtuals_mv", messages)) - - while True: - result = instance.query("SELECT count() FROM test.view") - time.sleep(1) - if int(result) == message_num: - break - - result = instance.query("SELECT key, value, subject FROM test.view ORDER BY key") - expected = """\ -0 0 virtuals_mv -1 1 virtuals_mv -2 2 virtuals_mv -3 3 virtuals_mv -4 4 virtuals_mv -5 5 virtuals_mv -6 6 virtuals_mv -7 7 virtuals_mv -8 8 virtuals_mv -9 9 virtuals_mv -""" - - instance.query( - """ - DROP TABLE test.consumer; - DROP TABLE test.view; - DROP TABLE test.nats_virtuals_mv - """ - ) - - assert TSV(result) == TSV(expected) - - -def test_nats_many_consumers_to_each_queue(nats_cluster): - instance.query( - """ - DROP TABLE IF EXISTS test.destination; - CREATE TABLE test.destination(key UInt64, value UInt64) - ENGINE = MergeTree() - ORDER BY key; - """ - ) - - num_tables = 4 - for table_id in range(num_tables): - print(("Setting up table {}".format(table_id))) - instance.query( - """ - DROP TABLE IF EXISTS test.many_consumers_{0}; - DROP TABLE IF EXISTS test.many_consumers_{0}_mv; - CREATE TABLE test.many_consumers_{0} (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'many_consumers', - nats_num_consumers = 2, - nats_queue_group = 'many_consumers', - nats_format = 'JSONEachRow', - nats_row_delimiter = '\\n'; - CREATE MATERIALIZED VIEW test.many_consumers_{0}_mv TO test.destination AS - SELECT key, value FROM test.many_consumers_{0}; - """.format( - table_id - ) - ) - while not check_table_is_ready( - instance, "test.many_consumers_{}".format(table_id) - ): - logging.debug( - "Table test.many_consumers_{} is not yet ready".format(table_id) - ) - time.sleep(0.5) - - i = [0] - messages_num = 1000 - - def produce(): - messages = [] - for _ in range(messages_num): - messages.append(json.dumps({"key": i[0], "value": i[0]})) - i[0] += 1 - asyncio.run(nats_produce_messages(nats_cluster, "many_consumers", messages)) - - threads = [] - threads_num = 20 - - for _ in range(threads_num): - threads.append(threading.Thread(target=produce)) - for thread in threads: - time.sleep(random.uniform(0, 1)) - thread.start() - - result1 = "" - while True: - result1 = instance.query("SELECT count() FROM test.destination") - time.sleep(1) - if int(result1) == messages_num * threads_num: - break - - for thread in threads: - thread.join() - - for consumer_id in range(num_tables): - instance.query( - """ - DROP TABLE test.many_consumers_{0}; - DROP TABLE test.many_consumers_{0}_mv; - """.format( - consumer_id - ) - ) - - instance.query( - """ - DROP TABLE test.destination; - """ - ) - - assert ( - int(result1) == messages_num * threads_num - ), "ClickHouse lost some messages: {}".format(result1) - - -def test_nats_restore_failed_connection_without_losses_on_write(nats_cluster): - instance.query( - """ - DROP TABLE IF EXISTS test.consume; - CREATE TABLE test.view (key UInt64, value UInt64) - ENGINE = MergeTree - ORDER BY key; - CREATE TABLE test.consume (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'producer_reconnect', - nats_format = 'JSONEachRow', - nats_num_consumers = 2, - nats_row_delimiter = '\\n'; - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.consume; - DROP TABLE IF EXISTS test.producer_reconnect; - CREATE TABLE test.producer_reconnect (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'producer_reconnect', - nats_format = 'JSONEachRow', - nats_row_delimiter = '\\n'; - """ - ) - while not check_table_is_ready(instance, "test.consume"): - logging.debug("Table test.consume is not yet ready") - time.sleep(0.5) - - messages_num = 100000 - values = [] - for i in range(messages_num): - values.append("({i}, {i})".format(i=i)) - values = ",".join(values) - - while True: - try: - instance.query( - "INSERT INTO test.producer_reconnect VALUES {}".format(values) - ) - break - except QueryRuntimeException as e: - if "Local: Timed out." in str(e): - continue - else: - raise - - while int(instance.query("SELECT count() FROM test.view")) == 0: - time.sleep(0.1) - - kill_nats(nats_cluster.nats_docker_id) - time.sleep(4) - revive_nats(nats_cluster.nats_docker_id, nats_cluster.nats_port) - - while True: - result = instance.query("SELECT count(DISTINCT key) FROM test.view") - time.sleep(1) - if int(result) == messages_num: - break - - instance.query( - """ - DROP TABLE test.consume; - DROP TABLE test.producer_reconnect; - """ - ) - - assert int(result) == messages_num, "ClickHouse lost some messages: {}".format( - result - ) - - -def test_nats_no_connection_at_startup_1(nats_cluster): - # no connection when table is initialized - nats_cluster.pause_container("nats1") - instance.query_and_get_error( - """ - CREATE TABLE test.cs (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'cs', - nats_format = 'JSONEachRow', - nats_num_consumers = '5', - nats_row_delimiter = '\\n'; - """ - ) - nats_cluster.unpause_container("nats1") - - -def test_nats_no_connection_at_startup_2(nats_cluster): - instance.query( - """ - CREATE TABLE test.cs (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'cs', - nats_format = 'JSONEachRow', - nats_num_consumers = '5', - nats_row_delimiter = '\\n'; - CREATE TABLE test.view (key UInt64, value UInt64) - ENGINE = MergeTree - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.cs; - """ - ) - - instance.query("DETACH TABLE test.cs") - nats_cluster.pause_container("nats1") - instance.query("ATTACH TABLE test.cs") - nats_cluster.unpause_container("nats1") - while not check_table_is_ready(instance, "test.cs"): - logging.debug("Table test.cs is not yet ready") - time.sleep(0.5) - - messages_num = 1000 - messages = [] - for i in range(messages_num): - messages.append(json.dumps({"key": i, "value": i})) - asyncio.run(nats_produce_messages(nats_cluster, "cs", messages)) - - for _ in range(20): - result = instance.query("SELECT count() FROM test.view") - time.sleep(1) - if int(result) == messages_num: - break - - instance.query( - """ - DROP TABLE test.consumer; - DROP TABLE test.cs; - """ - ) - - assert int(result) == messages_num, "ClickHouse lost some messages: {}".format( - result - ) - - -def test_nats_format_factory_settings(nats_cluster): - instance.query( - """ - CREATE TABLE test.format_settings ( - id String, date DateTime - ) ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'format_settings', - nats_format = 'JSONEachRow', - date_time_input_format = 'best_effort'; - """ - ) - while not check_table_is_ready(instance, "test.format_settings"): - logging.debug("Table test.format_settings is not yet ready") - time.sleep(0.5) - - message = json.dumps( - {"id": "format_settings_test", "date": "2021-01-19T14:42:33.1829214Z"} - ) - expected = instance.query( - """SELECT parseDateTimeBestEffort(CAST('2021-01-19T14:42:33.1829214Z', 'String'))""" - ) - - asyncio.run(nats_produce_messages(nats_cluster, "format_settings", [message])) - - while True: - result = instance.query("SELECT date FROM test.format_settings") - if result == expected: - break - - instance.query( - """ - CREATE TABLE test.view ( - id String, date DateTime - ) ENGINE = MergeTree ORDER BY id; - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.format_settings; - """ - ) - - asyncio.run(nats_produce_messages(nats_cluster, "format_settings", [message])) - while True: - result = instance.query("SELECT date FROM test.view") - if result == expected: - break - - instance.query( - """ - DROP TABLE test.consumer; - DROP TABLE test.format_settings; - """ - ) - - assert result == expected - - -def test_nats_bad_args(nats_cluster): - instance.query_and_get_error( - """ - CREATE TABLE test.drop (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_secure = true, - nats_format = 'JSONEachRow'; - """ - ) - - -def test_nats_drop_mv(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'mv', - nats_format = 'JSONEachRow'; - CREATE TABLE test.view (key UInt64, value UInt64) - ENGINE = MergeTree() - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.nats; - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - messages = [] - for i in range(20): - messages.append(json.dumps({"key": i, "value": i})) - asyncio.run(nats_produce_messages(nats_cluster, "mv", messages)) - - instance.query("DROP VIEW test.consumer") - messages = [] - for i in range(20, 40): - messages.append(json.dumps({"key": i, "value": i})) - asyncio.run(nats_produce_messages(nats_cluster, "mv", messages)) - - instance.query( - """ - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.nats; - """ - ) - messages = [] - for i in range(40, 50): - messages.append(json.dumps({"key": i, "value": i})) - asyncio.run(nats_produce_messages(nats_cluster, "mv", messages)) - - while True: - result = instance.query("SELECT * FROM test.view ORDER BY key") - if nats_check_result(result): - break - - nats_check_result(result, True) - - instance.query("DROP VIEW test.consumer") - messages = [] - for i in range(50, 60): - messages.append(json.dumps({"key": i, "value": i})) - asyncio.run(nats_produce_messages(nats_cluster, "mv", messages)) - - count = 0 - while True: - count = int(instance.query("SELECT count() FROM test.nats")) - if count: - break - - assert count > 0 - - -def test_nats_predefined_configuration(nats_cluster): - instance.query( - """ - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS(nats1) """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - asyncio.run( - nats_produce_messages( - nats_cluster, "named", [json.dumps({"key": 1, "value": 2})] - ) - ) - while True: - result = instance.query( - "SELECT * FROM test.nats ORDER BY key", ignore_error=True - ) - if result == "1\t2\n": - break - - -def test_format_with_prefix_and_suffix(nats_cluster): - instance.query( - """ - DROP TABLE IF EXISTS test.nats; - - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'custom', - nats_format = 'CustomSeparated'; - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - insert_messages = [] - - async def sub_to_nats(): - nc = await nats_connect_ssl( - nats_cluster.nats_port, - user="click", - password="house", - ssl_ctx=nats_cluster.nats_ssl_context, - ) - sub = await nc.subscribe("custom") - await sub.unsubscribe(2) - async for msg in sub.messages: - insert_messages.append(msg.data.decode()) - - await sub.drain() - await nc.drain() - - def run_sub(): - asyncio.run(sub_to_nats()) - - thread = threading.Thread(target=run_sub) - thread.start() - time.sleep(1) - - instance.query( - "INSERT INTO test.nats select number*10 as key, number*100 as value from numbers(2) settings format_custom_result_before_delimiter='\n', format_custom_result_after_delimiter='\n'" - ) - - thread.join() - - assert ( - "".join(insert_messages) - == "\n0\t0\n\n\n10\t100\n\n" - ) - - -def test_max_rows_per_message(nats_cluster): - instance.query( - """ - DROP TABLE IF EXISTS test.view; - DROP TABLE IF EXISTS test.nats; - - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'custom1', - nats_format = 'CustomSeparated', - nats_max_rows_per_message = 3, - format_custom_result_before_delimiter = '\n', - format_custom_result_after_delimiter = '\n'; - - CREATE MATERIALIZED VIEW test.view Engine=Log AS - SELECT key, value FROM test.nats; - """ - ) - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - num_rows = 5 - - insert_messages = [] - - async def sub_to_nats(): - nc = await nats_connect_ssl( - nats_cluster.nats_port, - user="click", - password="house", - ssl_ctx=nats_cluster.nats_ssl_context, - ) - sub = await nc.subscribe("custom1") - await sub.unsubscribe(2) - async for msg in sub.messages: - insert_messages.append(msg.data.decode()) - - await sub.drain() - await nc.drain() - - def run_sub(): - asyncio.run(sub_to_nats()) - - thread = threading.Thread(target=run_sub) - thread.start() - time.sleep(1) - - instance.query( - f"INSERT INTO test.nats select number*10 as key, number*100 as value from numbers({num_rows}) settings format_custom_result_before_delimiter='\n', format_custom_result_after_delimiter='\n'" - ) - - thread.join() - - assert ( - "".join(insert_messages) - == "\n0\t0\n10\t100\n20\t200\n\n\n30\t300\n40\t400\n\n" - ) - - attempt = 0 - rows = 0 - while attempt < 100: - rows = int(instance.query("SELECT count() FROM test.view")) - if rows == num_rows: - break - attempt += 1 - - assert rows == num_rows - - result = instance.query("SELECT * FROM test.view") - assert result == "0\t0\n10\t100\n20\t200\n30\t300\n40\t400\n" - - -def test_row_based_formats(nats_cluster): - num_rows = 10 - - for format_name in [ - "TSV", - "TSVWithNamesAndTypes", - "TSKV", - "CSV", - "CSVWithNamesAndTypes", - "CustomSeparatedWithNamesAndTypes", - "Values", - "JSON", - "JSONEachRow", - "JSONCompactEachRow", - "JSONCompactEachRowWithNamesAndTypes", - "JSONObjectEachRow", - "Avro", - "RowBinary", - "RowBinaryWithNamesAndTypes", - "MsgPack", - ]: - print(format_name) - - instance.query( - f""" - DROP TABLE IF EXISTS test.view; - DROP TABLE IF EXISTS test.nats; - - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = '{format_name}', - nats_format = '{format_name}'; - - CREATE MATERIALIZED VIEW test.view Engine=Log AS - SELECT key, value FROM test.nats; - """ - ) - - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - insert_messages = 0 - - async def sub_to_nats(): - nc = await nats_connect_ssl( - nats_cluster.nats_port, - user="click", - password="house", - ssl_ctx=nats_cluster.nats_ssl_context, - ) - sub = await nc.subscribe(format_name) - await sub.unsubscribe(2) - async for msg in sub.messages: - nonlocal insert_messages - insert_messages += 1 - - await sub.drain() - await nc.drain() - - def run_sub(): - asyncio.run(sub_to_nats()) - - thread = threading.Thread(target=run_sub) - thread.start() - time.sleep(1) - - instance.query( - f"INSERT INTO test.nats select number*10 as key, number*100 as value from numbers({num_rows})" - ) - - thread.join() - - assert insert_messages == 2 - - attempt = 0 - rows = 0 - while attempt < 100: - rows = int(instance.query("SELECT count() FROM test.view")) - if rows == num_rows: - break - attempt += 1 - - assert rows == num_rows - - expected = "" - for i in range(num_rows): - expected += str(i * 10) + "\t" + str(i * 100) + "\n" - - result = instance.query("SELECT * FROM test.view") - assert result == expected - - -def test_block_based_formats_1(nats_cluster): - instance.query( - """ - DROP TABLE IF EXISTS test.nats; - - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = 'PrettySpace', - nats_format = 'PrettySpace'; - """ - ) - - insert_messages = [] - - async def sub_to_nats(): - nc = await nats_connect_ssl( - nats_cluster.nats_port, - user="click", - password="house", - ssl_ctx=nats_cluster.nats_ssl_context, - ) - sub = await nc.subscribe("PrettySpace") - await sub.unsubscribe(3) - async for msg in sub.messages: - insert_messages.append(msg.data.decode()) - - await sub.drain() - await nc.drain() - - def run_sub(): - asyncio.run(sub_to_nats()) - - thread = threading.Thread(target=run_sub) - thread.start() - time.sleep(1) - - attempt = 0 - while attempt < 100: - try: - instance.query( - "INSERT INTO test.nats SELECT number * 10 as key, number * 100 as value FROM numbers(5) settings max_block_size=2, optimize_trivial_insert_select=0;" - ) - break - except Exception: - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - attempt += 1 - thread.join() - - data = [] - for message in insert_messages: - splitted = message.split("\n") - assert splitted[0] == " \x1b[1mkey\x1b[0m \x1b[1mvalue\x1b[0m" - assert splitted[1] == "" - assert splitted[-1] == "" - data += [line.split() for line in splitted[2:-1]] - - assert data == [ - ["0", "0"], - ["10", "100"], - ["20", "200"], - ["30", "300"], - ["40", "400"], - ] - - -def test_block_based_formats_2(nats_cluster): - num_rows = 100 - - for format_name in [ - "JSONColumns", - "Native", - "Arrow", - "Parquet", - "ORC", - "JSONCompactColumns", - ]: - print(format_name) - - instance.query( - f""" - DROP TABLE IF EXISTS test.view; - DROP TABLE IF EXISTS test.nats; - - CREATE TABLE test.nats (key UInt64, value UInt64) - ENGINE = NATS - SETTINGS nats_url = 'nats1:4444', - nats_subjects = '{format_name}', - nats_format = '{format_name}'; - - CREATE MATERIALIZED VIEW test.view Engine=Log AS - SELECT key, value FROM test.nats; - """ - ) - - while not check_table_is_ready(instance, "test.nats"): - logging.debug("Table test.nats is not yet ready") - time.sleep(0.5) - - insert_messages = 0 - - async def sub_to_nats(): - nc = await nats_connect_ssl( - nats_cluster.nats_port, - user="click", - password="house", - ssl_ctx=nats_cluster.nats_ssl_context, - ) - sub = await nc.subscribe(format_name) - await sub.unsubscribe(9) - async for msg in sub.messages: - nonlocal insert_messages - insert_messages += 1 - - await sub.drain() - await nc.drain() - - def run_sub(): - asyncio.run(sub_to_nats()) - - thread = threading.Thread(target=run_sub) - thread.start() - time.sleep(1) - - instance.query( - f"INSERT INTO test.nats SELECT number * 10 as key, number * 100 as value FROM numbers({num_rows}) settings max_block_size=12, optimize_trivial_insert_select=0;" - ) - - thread.join() - - assert insert_messages == 9 - - attempt = 0 - rows = 0 - while attempt < 100: - rows = int(instance.query("SELECT count() FROM test.view")) - if rows == num_rows: - break - attempt += 1 - - assert rows == num_rows - - result = instance.query("SELECT * FROM test.view ORDER by key") - expected = "" - for i in range(num_rows): - expected += str(i * 10) + "\t" + str(i * 100) + "\n" - assert result == expected - - -if __name__ == "__main__": - cluster.start() - input("Cluster created, press any key to destroy...") - cluster.shutdown() diff --git a/tests/integration/test_storage_nats/test_nats_json.reference b/tests/integration/test_storage_nats/test_nats_json.reference deleted file mode 100644 index 959bb2aad74..00000000000 --- a/tests/integration/test_storage_nats/test_nats_json.reference +++ /dev/null @@ -1,50 +0,0 @@ -0 0 -1 1 -2 2 -3 3 -4 4 -5 5 -6 6 -7 7 -8 8 -9 9 -10 10 -11 11 -12 12 -13 13 -14 14 -15 15 -16 16 -17 17 -18 18 -19 19 -20 20 -21 21 -22 22 -23 23 -24 24 -25 25 -26 26 -27 27 -28 28 -29 29 -30 30 -31 31 -32 32 -33 33 -34 34 -35 35 -36 36 -37 37 -38 38 -39 39 -40 40 -41 41 -42 42 -43 43 -44 44 -45 45 -46 46 -47 47 -48 48 -49 49 diff --git a/tests/integration/test_storage_postgresql_replica/__init__.py b/tests/integration/test_storage_postgresql_replica/__init__.py deleted file mode 100644 index e69de29bb2d..00000000000 diff --git a/tests/integration/test_storage_postgresql_replica/configs/log_conf.xml b/tests/integration/test_storage_postgresql_replica/configs/log_conf.xml deleted file mode 100644 index 27c7107ce5e..00000000000 --- a/tests/integration/test_storage_postgresql_replica/configs/log_conf.xml +++ /dev/null @@ -1,11 +0,0 @@ - - - trace - /var/log/clickhouse-server/log.log - /var/log/clickhouse-server/log.err.log - 1000M - 10 - /var/log/clickhouse-server/stderr.log - /var/log/clickhouse-server/stdout.log - - diff --git a/tests/integration/test_storage_postgresql_replica/test.py b/tests/integration/test_storage_postgresql_replica/test.py deleted file mode 100644 index 66495700102..00000000000 --- a/tests/integration/test_storage_postgresql_replica/test.py +++ /dev/null @@ -1,780 +0,0 @@ -import pytest - -# FIXME Tests with MaterializedPostgresSQL are temporarily disabled -# https://github.com/ClickHouse/ClickHouse/issues/36898 -# https://github.com/ClickHouse/ClickHouse/issues/38677 -# https://github.com/ClickHouse/ClickHouse/pull/39272#issuecomment-1190087190 - -pytestmark = pytest.mark.skip - -import time -import psycopg2 -import os.path as p - -from helpers.cluster import ClickHouseCluster -from helpers.test_tools import assert_eq_with_retry -from psycopg2.extensions import ISOLATION_LEVEL_AUTOCOMMIT -from helpers.test_tools import TSV - -import threading - -cluster = ClickHouseCluster(__file__) -instance = cluster.add_instance( - "instance", - main_configs=["configs/log_conf.xml"], - with_postgres=True, - stay_alive=True, -) - -postgres_table_template = """ - CREATE TABLE IF NOT EXISTS {} ( - key Integer NOT NULL, value Integer, PRIMARY KEY(key)) - """ - -queries = [ - "INSERT INTO {} select i, i from generate_series(0, 10000) as t(i);", - "DELETE FROM {} WHERE (value*value) % 3 = 0;", - "UPDATE {} SET value = value + 125 WHERE key % 2 = 0;", - "UPDATE {} SET key=key+20000 WHERE key%2=0", - "INSERT INTO {} select i, i from generate_series(40000, 50000) as t(i);", - "DELETE FROM {} WHERE key % 10 = 0;", - "UPDATE {} SET value = value + 101 WHERE key % 2 = 1;", - "UPDATE {} SET key=key+80000 WHERE key%2=1", - "DELETE FROM {} WHERE value % 2 = 0;", - "UPDATE {} SET value = value + 2000 WHERE key % 5 = 0;", - "INSERT INTO {} select i, i from generate_series(200000, 250000) as t(i);", - "DELETE FROM {} WHERE value % 3 = 0;", - "UPDATE {} SET value = value * 2 WHERE key % 3 = 0;", - "UPDATE {} SET key=key+500000 WHERE key%2=1", - "INSERT INTO {} select i, i from generate_series(1000000, 1050000) as t(i);", - "DELETE FROM {} WHERE value % 9 = 2;", - "UPDATE {} SET key=key+10000000", - "UPDATE {} SET value = value + 2 WHERE key % 3 = 1;", - "DELETE FROM {} WHERE value%5 = 0;", -] - - -@pytest.mark.timeout(30) -def check_tables_are_synchronized( - table_name, order_by="key", postgres_database="postgres_database" -): - while True: - expected = instance.query( - "select * from {}.{} order by {};".format( - postgres_database, table_name, order_by - ) - ) - result = instance.query( - "select * from test.{} order by {};".format(table_name, order_by) - ) - if result == expected: - break - - assert result == expected - - -def get_postgres_conn( - ip, port, database=False, auto_commit=True, database_name="postgres_database" -): - if database == True: - conn_string = "host={} port={} dbname='{}' user='postgres' password='mysecretpassword'".format( - ip, port, database_name - ) - else: - conn_string = ( - "host={} port={} user='postgres' password='mysecretpassword'".format( - ip, port - ) - ) - - conn = psycopg2.connect(conn_string) - if auto_commit: - conn.set_isolation_level(ISOLATION_LEVEL_AUTOCOMMIT) - conn.autocommit = True - return conn - - -def create_postgres_db(cursor, name): - cursor.execute("CREATE DATABASE {}".format(name)) - - -def create_clickhouse_postgres_db(ip, port, name="postgres_database"): - instance.query( - """ - CREATE DATABASE {} - ENGINE = PostgreSQL('{}:{}', '{}', 'postgres', 'mysecretpassword')""".format( - name, ip, port, name - ) - ) - - -def create_materialized_table(ip, port, table_name="postgresql_replica"): - instance.query( - f""" - CREATE TABLE test.{table_name} (key Int64, value Int64) - ENGINE = MaterializedPostgreSQL( - '{ip}:{port}', 'postgres_database', '{table_name}', 'postgres', 'mysecretpassword') - PRIMARY KEY key; """ - ) - - -def create_postgres_table(cursor, table_name, replica_identity_full=False): - cursor.execute("DROP TABLE IF EXISTS {}".format(table_name)) - cursor.execute(postgres_table_template.format(table_name)) - if replica_identity_full: - cursor.execute("ALTER TABLE {} REPLICA IDENTITY FULL;".format(table_name)) - - -def postgresql_replica_check_result( - result, check=False, ref_file="test_postgresql_replica.reference" -): - fpath = p.join(p.dirname(__file__), ref_file) - with open(fpath) as reference: - if check: - assert TSV(result) == TSV(reference) - else: - return TSV(result) == TSV(reference) - - -@pytest.fixture(scope="module") -def started_cluster(): - try: - cluster.start() - conn = get_postgres_conn(ip=cluster.postgres_ip, port=cluster.postgres_port) - cursor = conn.cursor() - create_postgres_db(cursor, "postgres_database") - create_clickhouse_postgres_db( - ip=cluster.postgres_ip, port=cluster.postgres_port - ) - - instance.query("CREATE DATABASE test") - yield cluster - - finally: - cluster.shutdown() - - -@pytest.mark.timeout(320) -def test_initial_load_from_snapshot(started_cluster): - conn = get_postgres_conn( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - database=True, - ) - cursor = conn.cursor() - create_postgres_table(cursor, "postgresql_replica") - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT number, number from numbers(50)" - ) - - instance.query("DROP TABLE IF EXISTS test.postgresql_replica") - create_materialized_table( - ip=started_cluster.postgres_ip, port=started_cluster.postgres_port - ) - - result = instance.query("SELECT * FROM test.postgresql_replica ORDER BY key;") - while postgresql_replica_check_result(result) == False: - time.sleep(0.2) - result = instance.query("SELECT * FROM test.postgresql_replica ORDER BY key;") - - cursor.execute("DROP TABLE postgresql_replica;") - postgresql_replica_check_result(result, True) - instance.query(f"DROP TABLE test.postgresql_replica SYNC") - - -@pytest.mark.timeout(320) -def test_no_connection_at_startup(started_cluster): - conn = get_postgres_conn( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - database=True, - ) - cursor = conn.cursor() - create_postgres_table(cursor, "postgresql_replica") - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT number, number from numbers(50)" - ) - - instance.query("DROP TABLE IF EXISTS test.postgresql_replica") - create_materialized_table( - ip=started_cluster.postgres_ip, port=started_cluster.postgres_port - ) - time.sleep(3) - - instance.query("DETACH TABLE test.postgresql_replica") - started_cluster.pause_container("postgres1") - - instance.query("ATTACH TABLE test.postgresql_replica") - time.sleep(3) - started_cluster.unpause_container("postgres1") - - result = instance.query("SELECT count() FROM test.postgresql_replica;") - while int(result) == 0: - time.sleep(0.5) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - - result = instance.query("SELECT * FROM test.postgresql_replica ORDER BY key;") - cursor.execute("DROP TABLE postgresql_replica;") - postgresql_replica_check_result(result, True) - instance.query(f"DROP TABLE test.postgresql_replica SYNC") - - -@pytest.mark.timeout(320) -def test_detach_attach_is_ok(started_cluster): - conn = get_postgres_conn( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - database=True, - ) - cursor = conn.cursor() - create_postgres_table(cursor, "postgresql_replica") - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT number, number from numbers(50)" - ) - - instance.query("DROP TABLE IF EXISTS test.postgresql_replica") - create_materialized_table( - ip=started_cluster.postgres_ip, port=started_cluster.postgres_port - ) - - result = instance.query("SELECT count() FROM test.postgresql_replica;") - while int(result) == 0: - time.sleep(0.2) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - - result = instance.query("SELECT * FROM test.postgresql_replica ORDER BY key;") - postgresql_replica_check_result(result, True) - - instance.query("DETACH TABLE test.postgresql_replica") - instance.query("ATTACH TABLE test.postgresql_replica") - - result = instance.query("SELECT * FROM test.postgresql_replica ORDER BY key;") - while postgresql_replica_check_result(result) == False: - time.sleep(0.5) - result = instance.query("SELECT * FROM test.postgresql_replica ORDER BY key;") - - cursor.execute("DROP TABLE postgresql_replica;") - postgresql_replica_check_result(result, True) - instance.query(f"DROP TABLE test.postgresql_replica SYNC") - - -@pytest.mark.timeout(320) -def test_replicating_insert_queries(started_cluster): - conn = get_postgres_conn( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - database=True, - ) - cursor = conn.cursor() - create_postgres_table(cursor, "postgresql_replica") - - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT number, number from numbers(10)" - ) - - instance.query("DROP TABLE IF EXISTS test.postgresql_replica") - create_materialized_table( - ip=started_cluster.postgres_ip, port=started_cluster.postgres_port - ) - - result = instance.query("SELECT count() FROM test.postgresql_replica;") - while int(result) != 10: - time.sleep(0.2) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT 10 + number, 10 + number from numbers(10)" - ) - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT 20 + number, 20 + number from numbers(10)" - ) - - result = instance.query("SELECT count() FROM test.postgresql_replica;") - while int(result) != 30: - time.sleep(0.2) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT 30 + number, 30 + number from numbers(10)" - ) - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT 40 + number, 40 + number from numbers(10)" - ) - - result = instance.query("SELECT count() FROM test.postgresql_replica;") - while int(result) != 50: - time.sleep(0.2) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - - result = instance.query("SELECT * FROM test.postgresql_replica ORDER BY key;") - cursor.execute("DROP TABLE postgresql_replica;") - postgresql_replica_check_result(result, True) - instance.query(f"DROP TABLE test.postgresql_replica SYNC") - - -@pytest.mark.timeout(320) -def test_replicating_delete_queries(started_cluster): - conn = get_postgres_conn( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - database=True, - ) - cursor = conn.cursor() - create_postgres_table(cursor, "postgresql_replica") - - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT number, number from numbers(50)" - ) - - instance.query("DROP TABLE IF EXISTS test.postgresql_replica") - create_materialized_table( - ip=started_cluster.postgres_ip, port=started_cluster.postgres_port - ) - - result = instance.query("SELECT * FROM test.postgresql_replica ORDER BY key;") - while postgresql_replica_check_result(result) == False: - time.sleep(0.2) - result = instance.query("SELECT * FROM test.postgresql_replica ORDER BY key;") - - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT 50 + number, 50 + number from numbers(50)" - ) - - result = instance.query("SELECT count() FROM test.postgresql_replica;") - while int(result) != 100: - time.sleep(0.5) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - - cursor.execute("DELETE FROM postgresql_replica WHERE key > 49;") - - result = instance.query("SELECT * FROM test.postgresql_replica ORDER BY key;") - while postgresql_replica_check_result(result) == False: - time.sleep(0.5) - result = instance.query("SELECT * FROM test.postgresql_replica ORDER BY key;") - - cursor.execute("DROP TABLE postgresql_replica;") - postgresql_replica_check_result(result, True) - - -@pytest.mark.timeout(320) -def test_replicating_update_queries(started_cluster): - conn = get_postgres_conn( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - database=True, - ) - cursor = conn.cursor() - create_postgres_table(cursor, "postgresql_replica") - - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT number, number + 10 from numbers(50)" - ) - - instance.query("DROP TABLE IF EXISTS test.postgresql_replica") - create_materialized_table( - ip=started_cluster.postgres_ip, port=started_cluster.postgres_port - ) - - result = instance.query("SELECT count() FROM test.postgresql_replica;") - while int(result) != 50: - time.sleep(0.2) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - - cursor.execute("UPDATE postgresql_replica SET value = value - 10;") - - result = instance.query("SELECT * FROM test.postgresql_replica ORDER BY key;") - while postgresql_replica_check_result(result) == False: - time.sleep(0.5) - result = instance.query("SELECT * FROM test.postgresql_replica ORDER BY key;") - - cursor.execute("DROP TABLE postgresql_replica;") - postgresql_replica_check_result(result, True) - - -@pytest.mark.timeout(320) -def test_resume_from_written_version(started_cluster): - conn = get_postgres_conn( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - database=True, - ) - cursor = conn.cursor() - create_postgres_table(cursor, "postgresql_replica") - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT number, number + 10 from numbers(50)" - ) - - instance.query("DROP TABLE IF EXISTS test.postgresql_replica") - create_materialized_table( - ip=started_cluster.postgres_ip, port=started_cluster.postgres_port - ) - - result = instance.query("SELECT count() FROM test.postgresql_replica;") - while int(result) != 50: - time.sleep(0.2) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT 50 + number, 50 + number from numbers(50)" - ) - - result = instance.query("SELECT count() FROM test.postgresql_replica;") - while int(result) != 100: - time.sleep(0.2) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - - instance.query("DETACH TABLE test.postgresql_replica") - - cursor.execute("DELETE FROM postgresql_replica WHERE key > 49;") - cursor.execute("UPDATE postgresql_replica SET value = value - 10;") - - instance.query("ATTACH TABLE test.postgresql_replica") - - result = instance.query("SELECT * FROM test.postgresql_replica ORDER BY key;") - while postgresql_replica_check_result(result) == False: - time.sleep(0.5) - result = instance.query("SELECT * FROM test.postgresql_replica ORDER BY key;") - - cursor.execute("DROP TABLE postgresql_replica;") - postgresql_replica_check_result(result, True) - - -@pytest.mark.timeout(320) -def test_many_replication_messages(started_cluster): - conn = get_postgres_conn( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - database=True, - ) - cursor = conn.cursor() - create_postgres_table(cursor, "postgresql_replica") - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT number, number from numbers(100000)" - ) - - instance.query("DROP TABLE IF EXISTS test.postgresql_replica") - create_materialized_table( - ip=started_cluster.postgres_ip, port=started_cluster.postgres_port - ) - - result = instance.query("SELECT count() FROM test.postgresql_replica;") - while int(result) != 100000: - time.sleep(0.2) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - print("SYNC OK") - - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT number, number from numbers(100000, 100000)" - ) - - result = instance.query("SELECT count() FROM test.postgresql_replica;") - while int(result) != 200000: - time.sleep(1) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - print("INSERT OK") - - result = instance.query("SELECT key FROM test.postgresql_replica ORDER BY key;") - expected = instance.query("SELECT number from numbers(200000)") - assert result == expected - - cursor.execute("UPDATE postgresql_replica SET value = key + 1 WHERE key < 100000;") - - result = instance.query( - "SELECT key FROM test.postgresql_replica WHERE value = key + 1 ORDER BY key;" - ) - expected = instance.query("SELECT number from numbers(100000)") - - while result != expected: - time.sleep(1) - result = instance.query( - "SELECT key FROM test.postgresql_replica WHERE value = key + 1 ORDER BY key;" - ) - print("UPDATE OK") - - cursor.execute("DELETE FROM postgresql_replica WHERE key % 2 = 1;") - cursor.execute("DELETE FROM postgresql_replica WHERE key != value;") - - result = instance.query( - "SELECT count() FROM (SELECT * FROM test.postgresql_replica);" - ) - while int(result) != 50000: - time.sleep(1) - result = instance.query( - "SELECT count() FROM (SELECT * FROM test.postgresql_replica);" - ) - print("DELETE OK") - - cursor.execute("DROP TABLE postgresql_replica;") - - -@pytest.mark.timeout(320) -def test_connection_loss(started_cluster): - conn = get_postgres_conn( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - database=True, - ) - cursor = conn.cursor() - create_postgres_table(cursor, "postgresql_replica") - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT number, number from numbers(50)" - ) - - instance.query("DROP TABLE IF EXISTS test.postgresql_replica") - create_materialized_table( - ip=started_cluster.postgres_ip, port=started_cluster.postgres_port - ) - - i = 50 - while i < 100000: - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT {} + number, number from numbers(10000)".format( - i - ) - ) - i += 10000 - - started_cluster.pause_container("postgres1") - - result = instance.query("SELECT count() FROM test.postgresql_replica;") - print(int(result)) - time.sleep(6) - - started_cluster.unpause_container("postgres1") - - result = instance.query("SELECT count() FROM test.postgresql_replica;") - while int(result) < 100050: - time.sleep(1) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - - cursor.execute("DROP TABLE postgresql_replica;") - assert int(result) == 100050 - - -@pytest.mark.timeout(320) -def test_clickhouse_restart(started_cluster): - pytest.skip("Temporary disabled (FIXME)") - conn = get_postgres_conn( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - database=True, - ) - cursor = conn.cursor() - create_postgres_table(cursor, "postgresql_replica") - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT number, number from numbers(50)" - ) - - instance.query("DROP TABLE IF EXISTS test.postgresql_replica") - create_materialized_table( - ip=started_cluster.postgres_ip, port=started_cluster.postgres_port - ) - - i = 50 - while i < 100000: - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT {} + number, number from numbers(10000)".format( - i - ) - ) - i += 10000 - - instance.restart_clickhouse() - - result = instance.query("SELECT count() FROM test.postgresql_replica;") - while int(result) < 100050: - time.sleep(1) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - - cursor.execute("DROP TABLE postgresql_replica;") - print(result) - assert int(result) == 100050 - - -def test_rename_table(started_cluster): - conn = get_postgres_conn( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - database=True, - ) - cursor = conn.cursor() - create_postgres_table(cursor, "postgresql_replica") - - instance.query("DROP TABLE IF EXISTS test.postgresql_replica") - create_materialized_table( - ip=started_cluster.postgres_ip, port=started_cluster.postgres_port - ) - - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT number, number from numbers(25)" - ) - - result = instance.query("SELECT count() FROM test.postgresql_replica;") - while int(result) != 25: - time.sleep(0.5) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - - instance.query( - "RENAME TABLE test.postgresql_replica TO test.postgresql_replica_renamed" - ) - assert ( - int(instance.query("SELECT count() FROM test.postgresql_replica_renamed;")) - == 25 - ) - - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT number, number from numbers(25, 25)" - ) - - result = instance.query("SELECT count() FROM test.postgresql_replica_renamed;") - while int(result) != 50: - time.sleep(0.5) - result = instance.query("SELECT count() FROM test.postgresql_replica_renamed;") - - result = instance.query( - "SELECT * FROM test.postgresql_replica_renamed ORDER BY key;" - ) - postgresql_replica_check_result(result, True) - cursor.execute("DROP TABLE postgresql_replica;") - instance.query("DROP TABLE IF EXISTS test.postgresql_replica_renamed") - - -def test_virtual_columns(started_cluster): - conn = get_postgres_conn( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - database=True, - ) - cursor = conn.cursor() - create_postgres_table(cursor, "postgresql_replica") - - instance.query("DROP TABLE IF EXISTS test.postgresql_replica") - create_materialized_table( - ip=started_cluster.postgres_ip, port=started_cluster.postgres_port - ) - - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT number, number from numbers(10)" - ) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - while int(result) != 10: - time.sleep(0.5) - result = instance.query("SELECT count() FROM test.postgresql_replica;") - - # just check that it works, no check with `expected` because _version is taken as LSN, which will be different each time. - result = instance.query( - "SELECT key, value, _sign, _version FROM test.postgresql_replica;" - ) - print(result) - cursor.execute("DROP TABLE postgresql_replica;") - instance.query(f"DROP TABLE test.postgresql_replica SYNC") - - -def test_abrupt_connection_loss_while_heavy_replication(started_cluster): - instance.query("DROP DATABASE IF EXISTS test_database") - conn = get_postgres_conn( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - database=True, - ) - cursor = conn.cursor() - table_name = "postgresql_replica" - create_postgres_table(cursor, table_name) - - instance.query(f"DROP TABLE IF EXISTS test.{table_name}") - create_materialized_table( - ip=started_cluster.postgres_ip, port=started_cluster.postgres_port - ) - - for i in range(len(queries)): - query = queries[i].format(table_name) - cursor.execute(query) - print("query {}".format(query.format(table_name))) - - started_cluster.pause_container("postgres1") - - result = instance.query("SELECT count() FROM test.postgresql_replica") - print(result) # Just debug - - started_cluster.unpause_container("postgres1") - - check_tables_are_synchronized("postgresql_replica") - - result = instance.query("SELECT count() FROM test.postgresql_replica") - print(result) # Just debug - instance.query(f"DROP TABLE test.postgresql_replica SYNC") - - -def test_abrupt_server_restart_while_heavy_replication(started_cluster): - # FIXME (kssenii) temporary disabled - if instance.is_built_with_sanitizer(): - pytest.skip("Temporary disabled (FIXME)") - - conn = get_postgres_conn( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - database=True, - ) - cursor = conn.cursor() - table_name = "postgresql_replica_697" - create_postgres_table(cursor, table_name) - - instance.query(f"INSERT INTO postgres_database.{table_name} SELECT -1, 1") - instance.query(f"DROP TABLE IF EXISTS test.{table_name} SYNC") - create_materialized_table( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - table_name=table_name, - ) - - n = 1 - while int(instance.query(f"select count() from test.{table_name}")) != 1: - sleep(1) - n += 1 - if n > 10: - break - - for query in queries: - cursor.execute(query.format(table_name)) - print("query {}".format(query.format(table_name))) - - instance.restart_clickhouse() - - result = instance.query(f"SELECT count() FROM test.{table_name}") - print(result) # Just debug - - check_tables_are_synchronized(table_name) - - result = instance.query(f"SELECT count() FROM test.{table_name}") - print(result) # Just debug - instance.query(f"DROP TABLE test.{table_name} SYNC") - - -def test_drop_table_immediately(started_cluster): - conn = get_postgres_conn( - ip=started_cluster.postgres_ip, - port=started_cluster.postgres_port, - database=True, - ) - cursor = conn.cursor() - create_postgres_table(cursor, "postgresql_replica") - instance.query( - "INSERT INTO postgres_database.postgresql_replica SELECT number, number from numbers(100000)" - ) - - instance.query("DROP TABLE IF EXISTS test.postgresql_replica") - create_materialized_table( - ip=started_cluster.postgres_ip, port=started_cluster.postgres_port - ) - instance.query("DROP TABLE test.postgresql_replica") - create_materialized_table( - ip=started_cluster.postgres_ip, port=started_cluster.postgres_port - ) - check_tables_are_synchronized("postgresql_replica") - instance.query(f"DROP TABLE test.postgresql_replica SYNC") - - -if __name__ == "__main__": - cluster.start() - input("Cluster created, press any key to destroy...") - cluster.shutdown() diff --git a/tests/integration/test_storage_postgresql_replica/test_postgresql_replica.reference b/tests/integration/test_storage_postgresql_replica/test_postgresql_replica.reference deleted file mode 100644 index 959bb2aad74..00000000000 --- a/tests/integration/test_storage_postgresql_replica/test_postgresql_replica.reference +++ /dev/null @@ -1,50 +0,0 @@ -0 0 -1 1 -2 2 -3 3 -4 4 -5 5 -6 6 -7 7 -8 8 -9 9 -10 10 -11 11 -12 12 -13 13 -14 14 -15 15 -16 16 -17 17 -18 18 -19 19 -20 20 -21 21 -22 22 -23 23 -24 24 -25 25 -26 26 -27 27 -28 28 -29 29 -30 30 -31 31 -32 32 -33 33 -34 34 -35 35 -36 36 -37 37 -38 38 -39 39 -40 40 -41 41 -42 42 -43 43 -44 44 -45 45 -46 46 -47 47 -48 48 -49 49 diff --git a/tests/integration/test_storage_rabbitmq/test.py b/tests/integration/test_storage_rabbitmq/test.py index 751279f5e5a..80d2050b394 100644 --- a/tests/integration/test_storage_rabbitmq/test.py +++ b/tests/integration/test_storage_rabbitmq/test.py @@ -511,69 +511,6 @@ def test_rabbitmq_many_materialized_views(rabbitmq_cluster): rabbitmq_check_result(result2, True) -@pytest.mark.skip(reason="clichouse_path with rabbitmq.proto fails to be exported") -def test_rabbitmq_protobuf(rabbitmq_cluster): - instance.query( - """ - CREATE TABLE test.rabbitmq (key UInt64, value String) - ENGINE = RabbitMQ - SETTINGS rabbitmq_host_port = 'rabbitmq1:5672', - rabbitmq_exchange_name = 'pb', - rabbitmq_format = 'Protobuf', - rabbitmq_flush_interval_ms=1000, - rabbitmq_max_block_size=100, - rabbitmq_schema = 'rabbitmq.proto:KeyValueProto'; - CREATE TABLE test.view (key UInt64, value UInt64) - ENGINE = MergeTree() - ORDER BY key; - CREATE MATERIALIZED VIEW test.consumer TO test.view AS - SELECT * FROM test.rabbitmq; - """ - ) - - credentials = pika.PlainCredentials("root", "clickhouse") - parameters = pika.ConnectionParameters( - rabbitmq_cluster.rabbitmq_ip, rabbitmq_cluster.rabbitmq_port, "/", credentials - ) - connection = pika.BlockingConnection(parameters) - channel = connection.channel() - - data = "" - for i in range(0, 20): - msg = rabbitmq_pb2.KeyValueProto() - msg.key = i - msg.value = str(i) - serialized_msg = msg.SerializeToString() - data = data + _VarintBytes(len(serialized_msg)) + serialized_msg - channel.basic_publish(exchange="pb", routing_key="", body=data) - data = "" - for i in range(20, 21): - msg = rabbitmq_pb2.KeyValueProto() - msg.key = i - msg.value = str(i) - serialized_msg = msg.SerializeToString() - data = data + _VarintBytes(len(serialized_msg)) + serialized_msg - channel.basic_publish(exchange="pb", routing_key="", body=data) - data = "" - for i in range(21, 50): - msg = rabbitmq_pb2.KeyValueProto() - msg.key = i - msg.value = str(i) - serialized_msg = msg.SerializeToString() - data = data + _VarintBytes(len(serialized_msg)) + serialized_msg - channel.basic_publish(exchange="pb", routing_key="", body=data) - - connection.close() - - result = "" - while True: - result = instance.query("SELECT * FROM test.view ORDER BY key") - if rabbitmq_check_result(result): - break - - rabbitmq_check_result(result, True) - - def test_rabbitmq_big_message(rabbitmq_cluster): # Create batchs of messages of size ~100Kb rabbitmq_messages = 1000 diff --git a/tests/integration/test_ttl_move/test.py b/tests/integration/test_ttl_move/test.py index a2f28e21666..c1c076277bb 100644 --- a/tests/integration/test_ttl_move/test.py +++ b/tests/integration/test_ttl_move/test.py @@ -1529,106 +1529,6 @@ def test_concurrent_alter_with_ttl_move(started_cluster, name, engine): node1.query("DROP TABLE IF EXISTS {name} SYNC".format(name=name)) -@pytest.mark.skip(reason="Flacky test") -@pytest.mark.parametrize( - "name,positive", - [ - pytest.param("test_double_move_while_select_negative", 0, id="negative"), - pytest.param("test_double_move_while_select_positive", 1, id="positive"), - ], -) -def test_double_move_while_select(started_cluster, name, positive): - name = unique_table_name(name) - - try: - node1.query( - """ - CREATE TABLE {name} ( - n Int64, - s String - ) ENGINE = MergeTree - ORDER BY tuple() - PARTITION BY n - SETTINGS storage_policy='small_jbod_with_external',temporary_directories_lifetime=1 - """.format( - name=name - ) - ) - - node1.query( - "INSERT INTO {name} VALUES (1, randomPrintableASCII(10*1024*1024))".format( - name=name - ) - ) - - parts = node1.query( - "SELECT name FROM system.parts WHERE table = '{name}' AND active = 1".format( - name=name - ) - ).splitlines() - assert len(parts) == 1 - - node1.query( - "ALTER TABLE {name} MOVE PART '{part}' TO DISK 'external'".format( - name=name, part=parts[0] - ) - ) - - def long_select(): - if positive: - node1.query( - "SELECT sleep(3), sleep(2), sleep(1), n FROM {name}".format( - name=name - ) - ) - - thread = threading.Thread(target=long_select) - thread.start() - - time.sleep(1) - - node1.query( - "ALTER TABLE {name} MOVE PART '{part}' TO DISK 'jbod1'".format( - name=name, part=parts[0] - ) - ) - - # Fill jbod1 to force ClickHouse to make move of partition 1 to external. - node1.query( - "INSERT INTO {name} VALUES (2, randomPrintableASCII(9*1024*1024))".format( - name=name - ) - ) - node1.query( - "INSERT INTO {name} VALUES (3, randomPrintableASCII(9*1024*1024))".format( - name=name - ) - ) - node1.query( - "INSERT INTO {name} VALUES (4, randomPrintableASCII(9*1024*1024))".format( - name=name - ) - ) - - wait_parts_mover(node1, name, retry_count=40) - - # If SELECT locked old part on external, move shall fail. - assert node1.query( - "SELECT disk_name FROM system.parts WHERE table = '{name}' AND active = 1 AND name = '{part}'".format( - name=name, part=parts[0] - ) - ).splitlines() == ["jbod1" if positive else "external"] - - thread.join() - - assert node1.query( - "SELECT n FROM {name} ORDER BY n".format(name=name) - ).splitlines() == ["1", "2", "3", "4"] - - finally: - node1.query("DROP TABLE IF EXISTS {name} SYNC".format(name=name)) - - @pytest.mark.parametrize( "name,engine,positive", [ diff --git a/tests/integration/test_user_directories/test.py b/tests/integration/test_user_directories/test.py index 45afb86f464..704fb30b2fd 100644 --- a/tests/integration/test_user_directories/test.py +++ b/tests/integration/test_user_directories/test.py @@ -38,14 +38,14 @@ def test_old_style(): assert node.query("SELECT * FROM system.user_directories") == TSV( [ [ - "users.xml", - "users.xml", + "users_xml", + "users_xml", '{"path":"\\\\/etc\\\\/clickhouse-server\\\\/users2.xml"}', 1, ], [ - "local directory", - "local directory", + "local_directory", + "local_directory", '{"path":"\\\\/var\\\\/lib\\\\/clickhouse\\\\/access2\\\\/"}', 2, ], @@ -62,20 +62,20 @@ def test_local_directories(): assert node.query("SELECT * FROM system.user_directories") == TSV( [ [ - "users.xml", - "users.xml", + "users_xml", + "users_xml", '{"path":"\\\\/etc\\\\/clickhouse-server\\\\/users3.xml"}', 1, ], [ - "local directory", - "local directory", + "local_directory", + "local_directory", '{"path":"\\\\/var\\\\/lib\\\\/clickhouse\\\\/access3\\\\/"}', 2, ], [ "local directory (ro)", - "local directory", + "local_directory", '{"path":"\\\\/var\\\\/lib\\\\/clickhouse\\\\/access3-ro\\\\/","readonly":true}', 3, ], @@ -92,8 +92,8 @@ def test_relative_path(): assert node.query("SELECT * FROM system.user_directories") == TSV( [ [ - "users.xml", - "users.xml", + "users_xml", + "users_xml", '{"path":"\\\\/etc\\\\/clickhouse-server\\\\/users4.xml"}', 1, ] @@ -110,8 +110,8 @@ def test_memory(): assert node.query("SELECT * FROM system.user_directories") == TSV( [ [ - "users.xml", - "users.xml", + "users_xml", + "users_xml", '{"path":"\\\\/etc\\\\/clickhouse-server\\\\/users5.xml"}', 1, ], @@ -129,20 +129,20 @@ def test_mixed_style(): assert node.query("SELECT * FROM system.user_directories") == TSV( [ [ - "users.xml", - "users.xml", + "users_xml", + "users_xml", '{"path":"\\\\/etc\\\\/clickhouse-server\\\\/users6.xml"}', 1, ], [ - "local directory", - "local directory", + "local_directory", + "local_directory", '{"path":"\\\\/var\\\\/lib\\\\/clickhouse\\\\/access6\\\\/"}', 2, ], [ - "local directory", - "local directory", + "local_directory", + "local_directory", '{"path":"\\\\/var\\\\/lib\\\\/clickhouse\\\\/access6a\\\\/"}', 3, ], @@ -160,14 +160,14 @@ def test_duplicates(): assert node.query("SELECT * FROM system.user_directories") == TSV( [ [ - "users.xml", - "users.xml", + "users_xml", + "users_xml", '{"path":"\\\\/etc\\\\/clickhouse-server\\\\/users7.xml"}', 1, ], [ - "local directory", - "local directory", + "local_directory", + "local_directory", '{"path":"\\\\/var\\\\/lib\\\\/clickhouse\\\\/access7\\\\/"}', 2, ], diff --git a/tests/integration/test_zookeeper_config/test.py b/tests/integration/test_zookeeper_config/test.py index 5e36a97b3ca..0c0f77ec597 100644 --- a/tests/integration/test_zookeeper_config/test.py +++ b/tests/integration/test_zookeeper_config/test.py @@ -57,11 +57,10 @@ def test_chroot_with_same_root(started_cluster): for j in range(2): # Second insert to test deduplication node.query("INSERT INTO simple VALUES ({0}, {0})".format(i)) - # Replication might take time - - assert_eq_with_retry(node1, "select count() from simple", "2\n") - - assert_eq_with_retry(node2, "select count() from simple", "2\n") + node1.query("SYSTEM SYNC REPLICA simple") + assert_eq_with_retry(node1, "select count() from simple", "2") + node2.query("SYSTEM SYNC REPLICA simple") + assert_eq_with_retry(node2, "select count() from simple", "2") def test_chroot_with_different_root(started_cluster): @@ -78,5 +77,7 @@ def test_chroot_with_different_root(started_cluster): for j in range(2): # Second insert to test deduplication node.query("INSERT INTO simple_different VALUES ({0}, {0})".format(i)) - assert node1.query("select count() from simple_different").strip() == "1" - assert node3.query("select count() from simple_different").strip() == "1" + node1.query("SYSTEM SYNC REPLICA simple_different") + assert_eq_with_retry(node1, "select count() from simple_different", "1") + node3.query("SYSTEM SYNC REPLICA simple_different") + assert_eq_with_retry(node3, "select count() from simple_different", "1") diff --git a/tests/queries/0_stateless/00417_kill_query.reference b/tests/queries/0_stateless/00417_kill_query.reference index 1a3b47964c0..0dd1d78ac82 100644 --- a/tests/queries/0_stateless/00417_kill_query.reference +++ b/tests/queries/0_stateless/00417_kill_query.reference @@ -1,2 +1,2 @@ -SELECT sleep(1) FROM system.numbers LIMIT 30 -SELECT sleep(1) FROM system.numbers LIMIT 31 +SELECT sleep(1) FROM system.numbers LIMIT 300 +SELECT sleep(1) FROM system.numbers LIMIT 301 diff --git a/tests/queries/0_stateless/00417_kill_query.sh b/tests/queries/0_stateless/00417_kill_query.sh index cd5b788a147..e64e93fd4a0 100755 --- a/tests/queries/0_stateless/00417_kill_query.sh +++ b/tests/queries/0_stateless/00417_kill_query.sh @@ -7,19 +7,26 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -QUERY_FIELND_NUM=4 +QUERY_FIELD_NUM=4 -$CLICKHOUSE_CLIENT --max_block_size=1 -q "SELECT sleep(1) FROM system.numbers LIMIT 30" &>/dev/null & -sleep 1 -$CLICKHOUSE_CLIENT -q "KILL QUERY WHERE current_database = '${CLICKHOUSE_DATABASE}' and query LIKE 'SELECT sleep(%' AND (elapsed >= 0.) SYNC" | cut -f $QUERY_FIELND_NUM +$CLICKHOUSE_CLIENT --max_block_size=1 -q "SELECT sleep(1) FROM system.numbers LIMIT 300" &>/dev/null & -# 31 is for the query to be different from the previous one -$CLICKHOUSE_CLIENT --max_block_size=1 -q "SELECT sleep(1) FROM system.numbers LIMIT 31" &>/dev/null & -sleep 1 -$CLICKHOUSE_CLIENT -q "KILL QUERY WHERE current_database = '${CLICKHOUSE_DATABASE}' and query = 'SELECT sleep(1) FROM system.numbers LIMIT 31' ASYNC" | cut -f $QUERY_FIELND_NUM +while true +do + $CLICKHOUSE_CLIENT --query "KILL QUERY WHERE current_database = '${CLICKHOUSE_DATABASE}' AND query LIKE 'SELECT sleep(%' AND (elapsed >= 0.) SYNC" | cut -f $QUERY_FIELD_NUM | grep '.' && break + sleep 0.1 +done + +# 31 is for the query to be different from the previous one +$CLICKHOUSE_CLIENT --max_block_size=1 -q "SELECT sleep(1) FROM system.numbers LIMIT 301" &>/dev/null & + +while true +do + $CLICKHOUSE_CLIENT --query "KILL QUERY WHERE current_database = '${CLICKHOUSE_DATABASE}' AND query = 'SELECT sleep(1) FROM system.numbers LIMIT 301' ASYNC" | cut -f $QUERY_FIELD_NUM | grep '.' && break + sleep 0.1 +done $CLICKHOUSE_CLIENT -q "KILL QUERY WHERE 0 ASYNC" $CLICKHOUSE_CLIENT -q "KILL QUERY WHERE 0 FORMAT TabSeparated" $CLICKHOUSE_CLIENT -q "KILL QUERY WHERE 0 SYNC FORMAT TabSeparated" $CLICKHOUSE_CLIENT -q "KILL QUERY WHERE 1 TEST" &>/dev/null - diff --git a/tests/queries/0_stateless/01292_create_user.reference b/tests/queries/0_stateless/01292_create_user.reference index eb89a5ed38c..b7c30b304bf 100644 --- a/tests/queries/0_stateless/01292_create_user.reference +++ b/tests/queries/0_stateless/01292_create_user.reference @@ -106,10 +106,10 @@ CREATE USER u2_01292 DEFAULT ROLE r1_01292, r2_01292 SETTINGS readonly = 1 CREATE USER u3_01292 HOST LIKE \'%.%.myhost.com\' DEFAULT ROLE r1_01292, r2_01292 CREATE USER u4_01292 HOST LIKE \'%.%.myhost.com\' DEFAULT ROLE r1_01292, r2_01292 -- system.users -u1_01292 local directory plaintext_password {} [] ['localhost'] [] [] 1 [] [] -u2_01292 local directory no_password {} [] [] [] ['%.%.myhost.com'] 0 [] [] -u3_01292 local directory sha256_password {} ['192.169.1.1','192.168.0.0/16'] ['localhost'] [] [] 0 ['r1_01292'] [] -u4_01292 local directory double_sha1_password {} ['::/0'] [] [] [] 1 [] ['r1_01292'] +u1_01292 local_directory plaintext_password {} [] ['localhost'] [] [] 1 [] [] +u2_01292 local_directory no_password {} [] [] [] ['%.%.myhost.com'] 0 [] [] +u3_01292 local_directory sha256_password {} ['192.169.1.1','192.168.0.0/16'] ['localhost'] [] [] 0 ['r1_01292'] [] +u4_01292 local_directory double_sha1_password {} ['::/0'] [] [] [] 1 [] ['r1_01292'] -- system.settings_profile_elements \N u1_01292 \N 0 readonly 1 \N \N \N \N \N u2_01292 \N 0 \N \N \N \N \N default diff --git a/tests/queries/0_stateless/01293_create_role.reference b/tests/queries/0_stateless/01293_create_role.reference index 9b3c4eabd47..42f091bddd5 100644 --- a/tests/queries/0_stateless/01293_create_role.reference +++ b/tests/queries/0_stateless/01293_create_role.reference @@ -28,7 +28,7 @@ CREATE ROLE r2_01293 CREATE ROLE r1_01293 SETTINGS readonly = 1 CREATE ROLE r2_01293 SETTINGS readonly = 1 -- system.roles -r1_01293 local directory +r1_01293 local_directory -- system.settings_profile_elements \N \N r1_01293 0 readonly 1 \N \N \N \N \N \N r2_01293 0 \N \N \N \N \N default diff --git a/tests/queries/0_stateless/01293_create_role.sql b/tests/queries/0_stateless/01293_create_role.sql index f22edfeec3e..fd75d62964d 100644 --- a/tests/queries/0_stateless/01293_create_role.sql +++ b/tests/queries/0_stateless/01293_create_role.sql @@ -1,3 +1,5 @@ +-- Tags: no-parallel + DROP ROLE IF EXISTS r1_01293, r2_01293, r3_01293, r4_01293, r5_01293, r6_01293, r7_01293, r8_01293, r9_01293; DROP ROLE IF EXISTS r2_01293_renamed; DROP ROLE IF EXISTS r1_01293@'%', 'r2_01293@%.myhost.com'; diff --git a/tests/queries/0_stateless/01294_create_settings_profile.reference b/tests/queries/0_stateless/01294_create_settings_profile.reference index dbb73bca851..a10d5758752 100644 --- a/tests/queries/0_stateless/01294_create_settings_profile.reference +++ b/tests/queries/0_stateless/01294_create_settings_profile.reference @@ -53,12 +53,12 @@ CREATE SETTINGS PROFILE s4_01294 SETTINGS INHERIT readonly, INHERIT readonly CREATE SETTINGS PROFILE s5_01294 SETTINGS INHERIT readonly, readonly = 1 CREATE SETTINGS PROFILE s6_01294 SETTINGS INHERIT readonly, readonly CONST -- system.settings_profiles -s1_01294 local directory 0 0 [] [] -s2_01294 local directory 1 0 ['r1_01294'] [] -s3_01294 local directory 1 0 ['r1_01294'] [] -s4_01294 local directory 1 0 ['r1_01294'] [] -s5_01294 local directory 3 0 ['u1_01294'] [] -s6_01294 local directory 0 1 [] ['r1_01294','u1_01294'] +s1_01294 local_directory 0 0 [] [] +s2_01294 local_directory 1 0 ['r1_01294'] [] +s3_01294 local_directory 1 0 ['r1_01294'] [] +s4_01294 local_directory 1 0 ['r1_01294'] [] +s5_01294 local_directory 3 0 ['u1_01294'] [] +s6_01294 local_directory 0 1 [] ['r1_01294','u1_01294'] -- system.settings_profile_elements s2_01294 \N \N 0 readonly 0 \N \N \N \N s3_01294 \N \N 0 max_memory_usage 5000000 4000000 6000000 CONST \N diff --git a/tests/queries/0_stateless/01294_create_settings_profile.sql b/tests/queries/0_stateless/01294_create_settings_profile.sql index 565b4e70367..f71eefa6975 100644 --- a/tests/queries/0_stateless/01294_create_settings_profile.sql +++ b/tests/queries/0_stateless/01294_create_settings_profile.sql @@ -1,3 +1,5 @@ +-- Tags: no-parallel + DROP SETTINGS PROFILE IF EXISTS s1_01294, s2_01294, s3_01294, s4_01294, s5_01294, s6_01294, s7_01294, s8_01294, s9_01294, s10_01294; DROP SETTINGS PROFILE IF EXISTS s2_01294_renamed; DROP USER IF EXISTS u1_01294; diff --git a/tests/queries/0_stateless/01295_create_row_policy.reference b/tests/queries/0_stateless/01295_create_row_policy.reference index 6e3169b7fec..d73d9752bc1 100644 --- a/tests/queries/0_stateless/01295_create_row_policy.reference +++ b/tests/queries/0_stateless/01295_create_row_policy.reference @@ -30,6 +30,6 @@ CREATE ROW POLICY p5_01295 ON db2.table2 FOR SELECT USING a = b CREATE ROW POLICY p1_01295 ON db.table FOR SELECT USING 1 TO ALL CREATE ROW POLICY p2_01295 ON db.table FOR SELECT USING 1 TO ALL -- system.row_policies -p1_01295 ON db.table p1_01295 db table local directory (a < b) AND (c > d) 0 0 [] [] -p2_01295 ON db.table p2_01295 db table local directory id = currentUser() 1 0 ['u1_01295'] [] -p3_01295 ON db.table p3_01295 db table local directory 1 0 1 [] ['r1_01295'] +p1_01295 ON db.table p1_01295 db table local_directory (a < b) AND (c > d) 0 0 [] [] +p2_01295 ON db.table p2_01295 db table local_directory id = currentUser() 1 0 ['u1_01295'] [] +p3_01295 ON db.table p3_01295 db table local_directory 1 0 1 [] ['r1_01295'] diff --git a/tests/queries/0_stateless/01295_create_row_policy.sql b/tests/queries/0_stateless/01295_create_row_policy.sql index b484d0ea0f3..5ccd815c89a 100644 --- a/tests/queries/0_stateless/01295_create_row_policy.sql +++ b/tests/queries/0_stateless/01295_create_row_policy.sql @@ -1,3 +1,5 @@ +-- Tags: no-parallel + DROP ROW POLICY IF EXISTS p1_01295, p2_01295, p3_01295, p4_01295, p5_01295, p6_01295, p7_01295, p8_01295, p9_01295, p10_01295 ON db.table; DROP ROW POLICY IF EXISTS p2_01295_renamed ON db.table; DROP ROW POLICY IF EXISTS p3_01295 ON db.table, db2.table2; diff --git a/tests/queries/0_stateless/01297_create_quota.reference b/tests/queries/0_stateless/01297_create_quota.reference index 6b844e0d2f6..308bbf79024 100644 --- a/tests/queries/0_stateless/01297_create_quota.reference +++ b/tests/queries/0_stateless/01297_create_quota.reference @@ -52,10 +52,10 @@ CREATE QUOTA q2_01297 FOR INTERVAL 1 day MAX errors = 5 CREATE QUOTA q1_01297 FOR INTERVAL 1 day TRACKING ONLY TO r1_01297 CREATE QUOTA q2_01297 FOR INTERVAL 1 day TRACKING ONLY TO r1_01297 -- system.quotas -q1_01297 local directory ['user_name'] [] 0 ['r1_01297'] [] -q2_01297 local directory [] [5259492] 0 ['r1_01297','u1_01297'] [] -q3_01297 local directory ['client_key','user_name'] [5259492,15778476] 0 [] [] -q4_01297 local directory [] [604800] 1 [] ['u1_01297'] +q1_01297 local_directory ['user_name'] [] 0 ['r1_01297'] [] +q2_01297 local_directory [] [5259492] 0 ['r1_01297','u1_01297'] [] +q3_01297 local_directory ['client_key','user_name'] [5259492,15778476] 0 [] [] +q4_01297 local_directory [] [604800] 1 [] ['u1_01297'] -- system.quota_limits q2_01297 5259492 0 100 \N \N 11 1000 10000 1001 10001 2.5 \N q3_01297 5259492 0 \N \N \N \N 1002 \N \N \N \N \N diff --git a/tests/queries/0_stateless/02117_show_create_table_system.reference b/tests/queries/0_stateless/02117_show_create_table_system.reference index 5b678537248..0616756328b 100644 --- a/tests/queries/0_stateless/02117_show_create_table_system.reference +++ b/tests/queries/0_stateless/02117_show_create_table_system.reference @@ -960,6 +960,7 @@ CREATE TABLE system.role_grants `user_name` Nullable(String), `role_name` Nullable(String), `granted_role_name` String, + `granted_role_id` UUID, `granted_role_is_default` UInt8, `with_admin_option` UInt8 ) diff --git a/tests/queries/0_stateless/02263_format_insert_settings.sh b/tests/queries/0_stateless/02263_format_insert_settings.sh index 3d5f780a38c..efb3d39ab6c 100755 --- a/tests/queries/0_stateless/02263_format_insert_settings.sh +++ b/tests/queries/0_stateless/02263_format_insert_settings.sh @@ -23,10 +23,12 @@ function run_format_both() # NOTE: that those queries may work slow, due to stack trace obtaining run_format 'insert into foo settings max_threads=1' 2> >(grep -m1 -o "Syntax error (query): failed at position .* (end of query):") + # compatibility run_format 'insert into foo format tsv settings max_threads=1' 2> >(grep -m1 -F -o "Can't format ASTInsertQuery with data, since data will be lost.") run_format_both 'insert into foo format tsv settings max_threads=1' --allow_settings_after_format_in_insert run_format 'insert into foo settings max_threads=1 format tsv settings max_threads=1' --allow_settings_after_format_in_insert 2> >(grep -m1 -F -o "You have SETTINGS before and after FORMAT") + # and via server (since this is a separate code path) $CLICKHOUSE_CLIENT -q 'drop table if exists data_02263' $CLICKHOUSE_CLIENT -q 'create table data_02263 (key Int) engine=Memory()' diff --git a/tests/queries/0_stateless/02273_full_sort_join.sql.j2 b/tests/queries/0_stateless/02273_full_sort_join.sql.j2 index 6500306356c..2a13e11e116 100644 --- a/tests/queries/0_stateless/02273_full_sort_join.sql.j2 +++ b/tests/queries/0_stateless/02273_full_sort_join.sql.j2 @@ -1,4 +1,4 @@ --- Tags: long +-- Tags: long, no-random-settings DROP TABLE IF EXISTS t1; DROP TABLE IF EXISTS t2; diff --git a/tests/queries/0_stateless/02790_async_queries_in_query_log.reference b/tests/queries/0_stateless/02790_async_queries_in_query_log.reference index e202a38f068..aa18817f4e6 100644 --- a/tests/queries/0_stateless/02790_async_queries_in_query_log.reference +++ b/tests/queries/0_stateless/02790_async_queries_in_query_log.reference @@ -36,6 +36,14 @@ exception_code: 0 system.query_views_log +system.part_log +Row 1: +────── +database: default +table: async_insert_landing +partition_id: all +rows: 4 + system.query_log Row 1: ────── @@ -85,6 +93,14 @@ written_bytes: 0 status: QueryFinish exception_code: 0 +system.part_log +Row 1: +────── +database: default +table: async_insert_landing +partition_id: all +rows: 3 + system.query_log Row 1: ────── @@ -133,3 +149,11 @@ written_rows: 0 written_bytes: 0 status: Exc*****onWhileProcessing exception_code: 395 + +system.part_log +Row 1: +────── +database: default +table: async_insert_landing +partition_id: all +rows: 3 diff --git a/tests/queries/0_stateless/02790_async_queries_in_query_log.sh b/tests/queries/0_stateless/02790_async_queries_in_query_log.sh index d1c8fe122cb..1ff97031acb 100755 --- a/tests/queries/0_stateless/02790_async_queries_in_query_log.sh +++ b/tests/queries/0_stateless/02790_async_queries_in_query_log.sh @@ -28,7 +28,7 @@ function print_flush_query_logs() FROM system.query_log WHERE event_date >= yesterday() - AND initial_query_id = (SELECT flush_query_id FROM system.asynchronous_insert_log WHERE query_id = '$1') + AND initial_query_id = (SELECT flush_query_id FROM system.asynchronous_insert_log WHERE event_date >= yesterday() AND query_id = '$1') -- AND current_database = currentDatabase() -- Just to silence style check: this is not ok for this test since the query uses default values ORDER BY type DESC FORMAT Vertical" @@ -50,7 +50,21 @@ function print_flush_query_logs() FROM system.query_views_log WHERE event_date >= yesterday() - AND initial_query_id = (SELECT flush_query_id FROM system.asynchronous_insert_log WHERE query_id = '$1') + AND initial_query_id = (SELECT flush_query_id FROM system.asynchronous_insert_log WHERE event_date >= yesterday() AND query_id = '$1') + FORMAT Vertical" + + echo "" + echo "system.part_log" + ${CLICKHOUSE_CLIENT} -q " + SELECT + database, + table, + partition_id, + rows + FROM system.part_log + WHERE + event_date >= yesterday() + AND query_id = (SELECT flush_query_id FROM system.asynchronous_insert_log WHERE event_date >= yesterday() AND query_id = '$1') FORMAT Vertical" } diff --git a/tests/queries/0_stateless/02833_concurrrent_sessions.reference b/tests/queries/0_stateless/02833_concurrrent_sessions.reference new file mode 100644 index 00000000000..bfe507e8eac --- /dev/null +++ b/tests/queries/0_stateless/02833_concurrrent_sessions.reference @@ -0,0 +1,34 @@ +sessions: +150 +port_0_sessions: +0 +address_0_sessions: +0 +tcp_sessions +60 +http_sessions +30 +http_with_session_id_sessions +30 +my_sql_sessions +30 +Corresponding LoginSuccess/Logout +10 +LoginFailure +10 +Corresponding LoginSuccess/Logout +10 +LoginFailure +10 +Corresponding LoginSuccess/Logout +10 +LoginFailure +10 +Corresponding LoginSuccess/Logout +10 +LoginFailure +10 +Corresponding LoginSuccess/Logout +10 +LoginFailure +10 diff --git a/tests/queries/0_stateless/02833_concurrrent_sessions.sh b/tests/queries/0_stateless/02833_concurrrent_sessions.sh new file mode 100755 index 00000000000..26b48462a76 --- /dev/null +++ b/tests/queries/0_stateless/02833_concurrrent_sessions.sh @@ -0,0 +1,138 @@ +#!/usr/bin/env bash +# Tags: no-fasttest, no-debug + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +readonly PID=$$ + +# Each user uses a separate thread. +readonly TCP_USERS=( "02833_TCP_USER_${PID}"_{1,2} ) # 2 concurrent TCP users +readonly HTTP_USERS=( "02833_HTTP_USER_${PID}" ) +readonly HTTP_WITH_SESSION_ID_SESSION_USERS=( "02833_HTTP_WITH_SESSION_ID_USER_${PID}" ) +readonly MYSQL_USERS=( "02833_MYSQL_USER_${PID}") +readonly ALL_USERS=( "${TCP_USERS[@]}" "${HTTP_USERS[@]}" "${HTTP_WITH_SESSION_ID_SESSION_USERS[@]}" "${MYSQL_USERS[@]}" ) + +readonly TCP_USERS_SQL_COLLECTION_STRING="$( echo "${TCP_USERS[*]}" | sed "s/[^[:space:]]\+/'&'/g" | sed 's/[[:space:]]/,/g' )" +readonly HTTP_USERS_SQL_COLLECTION_STRING="$( echo "${HTTP_USERS[*]}" | sed "s/[^[:space:]]\+/'&'/g" | sed 's/[[:space:]]/,/g' )" +readonly HTTP_WITH_SESSION_ID_USERS_SQL_COLLECTION_STRING="$( echo "${HTTP_WITH_SESSION_ID_SESSION_USERS[*]}" | sed "s/[^[:space:]]\+/'&'/g" | sed 's/[[:space:]]/,/g' )" +readonly MYSQL_USERS_SQL_COLLECTION_STRING="$( echo "${MYSQL_USERS[*]}" | sed "s/[^[:space:]]\+/'&'/g" | sed 's/[[:space:]]/,/g' )" +readonly ALL_USERS_SQL_COLLECTION_STRING="$( echo "${ALL_USERS[*]}" | sed "s/[^[:space:]]\+/'&'/g" | sed 's/[[:space:]]/,/g' )" + +readonly SESSION_LOG_MATCHING_FIELDS="auth_id, auth_type, client_version_major, client_version_minor, client_version_patch, interface" + +for user in "${ALL_USERS[@]}"; do + ${CLICKHOUSE_CLIENT} -q "CREATE USER IF NOT EXISTS ${user} IDENTIFIED WITH plaintext_password BY 'pass'" + ${CLICKHOUSE_CLIENT} -q "GRANT SELECT ON system.* TO ${user}" + ${CLICKHOUSE_CLIENT} -q "GRANT SELECT ON INFORMATION_SCHEMA.* TO ${user}"; +done + +# All _session functions execute in separate threads. +# These functions try to create a session with successful login and logout. +# Sleep a small, random amount of time to make concurrency more intense. +# and try to login with an invalid password. +function tcp_session() +{ + local user=$1 + local i=0 + while (( (i++) < 10 )); do + # login logout + ${CLICKHOUSE_CLIENT} -q "SELECT 1, sleep(0.01${RANDOM})" --user="${user}" --password="pass" + # login failure + ${CLICKHOUSE_CLIENT} -q "SELECT 2" --user="${user}" --password 'invalid' + done +} + +function http_session() +{ + local user=$1 + local i=0 + while (( (i++) < 10 )); do + # login logout + ${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&user=${user}&password=pass" -d "SELECT 3, sleep(0.01${RANDOM})" + + # login failure + ${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&user=${user}&password=wrong" -d "SELECT 4" + done +} + +function http_with_session_id_session() +{ + local user=$1 + local i=0 + while (( (i++) < 10 )); do + # login logout + ${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&session_id=${user}&user=${user}&password=pass" -d "SELECT 5, sleep 0.01${RANDOM}" + + # login failure + ${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&session_id=${user}&user=${user}&password=wrong" -d "SELECT 6" + done +} + +function mysql_session() +{ + local user=$1 + local i=0 + while (( (i++) < 10 )); do + # login logout + ${CLICKHOUSE_CLIENT} -q "SELECT 1, sleep(0.01${RANDOM}) FROM mysql('127.0.0.1:9004', 'system', 'one', '${user}', 'pass')" + + # login failure + ${CLICKHOUSE_CLIENT} -q "SELECT 1 FROM mysql('127.0.0.1:9004', 'system', 'one', '${user}', 'wrong', SETTINGS connection_max_tries=1)" + done +} + +${CLICKHOUSE_CLIENT} -q "SYSTEM FLUSH LOGS" +${CLICKHOUSE_CLIENT} -q "DELETE FROM system.session_log WHERE user IN (${ALL_USERS_SQL_COLLECTION_STRING})" + +export -f tcp_session; +export -f http_session; +export -f http_with_session_id_session; +export -f mysql_session; + +for user in "${TCP_USERS[@]}"; do + timeout 60s bash -c "tcp_session ${user}" >/dev/null 2>&1 & +done + +for user in "${HTTP_USERS[@]}"; do + timeout 60s bash -c "http_session ${user}" >/dev/null 2>&1 & +done + +for user in "${HTTP_WITH_SESSION_ID_SESSION_USERS[@]}"; do + timeout 60s bash -c "http_with_session_id_session ${user}" >/dev/null 2>&1 & +done + +for user in "${MYSQL_USERS[@]}"; do + timeout 60s bash -c "mysql_session ${user}" >/dev/null 2>&1 & +done + +wait + +${CLICKHOUSE_CLIENT} -q "SYSTEM FLUSH LOGS" + +echo "sessions:" +${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user IN (${ALL_USERS_SQL_COLLECTION_STRING})" + +echo "port_0_sessions:" +${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user IN (${ALL_USERS_SQL_COLLECTION_STRING}) AND client_port = 0" + +echo "address_0_sessions:" +${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user IN (${ALL_USERS_SQL_COLLECTION_STRING}) AND client_address = toIPv6('::')" + +echo "tcp_sessions" +${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user IN (${TCP_USERS_SQL_COLLECTION_STRING}) AND interface = 'TCP'" +echo "http_sessions" +${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user IN (${HTTP_USERS_SQL_COLLECTION_STRING}) AND interface = 'HTTP'" +echo "http_with_session_id_sessions" +${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user IN (${HTTP_WITH_SESSION_ID_USERS_SQL_COLLECTION_STRING}) AND interface = 'HTTP'" +echo "my_sql_sessions" +${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user IN (${MYSQL_USERS_SQL_COLLECTION_STRING}) AND interface = 'MySQL'" + +for user in "${ALL_USERS[@]}"; do + ${CLICKHOUSE_CLIENT} -q "DROP USER ${user}" + echo "Corresponding LoginSuccess/Logout" + ${CLICKHOUSE_CLIENT} -q "SELECT COUNT(*) FROM (SELECT ${SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '${user}' AND type = 'LoginSuccess' INTERSECT SELECT ${SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '${user}' AND type = 'Logout')" + echo "LoginFailure" + ${CLICKHOUSE_CLIENT} -q "SELECT COUNT(*) FROM system.session_log WHERE user = '${user}' AND type = 'LoginFailure'" + done diff --git a/tests/queries/0_stateless/02833_starts_ends_with_utf8.reference b/tests/queries/0_stateless/02833_starts_ends_with_utf8.reference new file mode 100644 index 00000000000..ca2a5bc50f8 --- /dev/null +++ b/tests/queries/0_stateless/02833_starts_ends_with_utf8.reference @@ -0,0 +1,29 @@ +-- { echoOn } +select startsWithUTF8('富强民主文明和谐', '富强'); +1 +select startsWithUTF8('富强民主文明和谐', '\xe5'); +0 +select startsWithUTF8('富强民主文明和谐', ''); +1 +SELECT startsWithUTF8('123', '123'); +1 +SELECT startsWithUTF8('123', '12'); +1 +SELECT startsWithUTF8('123', '1234'); +0 +SELECT startsWithUTF8('123', ''); +1 +select endsWithUTF8('富强民主文明和谐', '和谐'); +1 +select endsWithUTF8('富强民主文明和谐', '\x90'); +0 +select endsWithUTF8('富强民主文明和谐', ''); +1 +SELECT endsWithUTF8('123', '3'); +1 +SELECT endsWithUTF8('123', '23'); +1 +SELECT endsWithUTF8('123', '32'); +0 +SELECT endsWithUTF8('123', ''); +1 diff --git a/tests/queries/0_stateless/02833_starts_ends_with_utf8.sql b/tests/queries/0_stateless/02833_starts_ends_with_utf8.sql new file mode 100644 index 00000000000..3a783dc280e --- /dev/null +++ b/tests/queries/0_stateless/02833_starts_ends_with_utf8.sql @@ -0,0 +1,19 @@ +-- { echoOn } +select startsWithUTF8('富强民主文明和谐', '富强'); +select startsWithUTF8('富强民主文明和谐', '\xe5'); +select startsWithUTF8('富强民主文明和谐', ''); + +SELECT startsWithUTF8('123', '123'); +SELECT startsWithUTF8('123', '12'); +SELECT startsWithUTF8('123', '1234'); +SELECT startsWithUTF8('123', ''); + +select endsWithUTF8('富强民主文明和谐', '和谐'); +select endsWithUTF8('富强民主文明和谐', '\x90'); +select endsWithUTF8('富强民主文明和谐', ''); + +SELECT endsWithUTF8('123', '3'); +SELECT endsWithUTF8('123', '23'); +SELECT endsWithUTF8('123', '32'); +SELECT endsWithUTF8('123', ''); +-- { echoOff } diff --git a/tests/queries/0_stateless/02834_remote_session_log.reference b/tests/queries/0_stateless/02834_remote_session_log.reference new file mode 100644 index 00000000000..e2680982ab0 --- /dev/null +++ b/tests/queries/0_stateless/02834_remote_session_log.reference @@ -0,0 +1,13 @@ +0 +0 +0 +0 +client_port 0 connections: +0 +client_address '::' connections: +0 +login failures: +0 +TCP Login and logout count is equal +HTTP Login and logout count is equal +MySQL Login and logout count is equal diff --git a/tests/queries/0_stateless/02834_remote_session_log.sh b/tests/queries/0_stateless/02834_remote_session_log.sh new file mode 100755 index 00000000000..3bedfb6c9ee --- /dev/null +++ b/tests/queries/0_stateless/02834_remote_session_log.sh @@ -0,0 +1,56 @@ +#!/usr/bin/env bash +# Tags: no-fasttest + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +readonly PID=$$ +readonly TEST_USER=$"02834_USER_${PID}" +readonly SESSION_LOG_MATCHING_FIELDS="auth_id, auth_type, client_version_major, client_version_minor, client_version_patch, interface" + +${CLICKHOUSE_CLIENT} -q "CREATE USER IF NOT EXISTS ${TEST_USER} IDENTIFIED WITH plaintext_password BY 'pass'" +${CLICKHOUSE_CLIENT} -q "GRANT SELECT ON INFORMATION_SCHEMA.* TO ${TEST_USER}" +${CLICKHOUSE_CLIENT} -q "GRANT SELECT ON system.* TO ${TEST_USER}" +${CLICKHOUSE_CLIENT} -q "GRANT CREATE TEMPORARY TABLE, MYSQL, REMOTE ON *.* TO ${TEST_USER}" + +${CLICKHOUSE_CLIENT} -q "SYSTEM FLUSH LOGS" +${CLICKHOUSE_CLIENT} -q "DELETE FROM system.session_log WHERE user = '${TEST_USER}'" + +${CLICKHOUSE_CURL} -sS -X POST "${CLICKHOUSE_URL}&user=${TEST_USER}&password=pass" \ + -d "SELECT * FROM remote('127.0.0.1:${CLICKHOUSE_PORT_TCP}', 'system', 'one', '${TEST_USER}', 'pass')" + +${CLICKHOUSE_CURL} -sS -X POST "${CLICKHOUSE_URL}&user=${TEST_USER}&password=pass" \ + -d "SELECT * FROM mysql('127.0.0.1:9004', 'system', 'one', '${TEST_USER}', 'pass')" + +${CLICKHOUSE_CLIENT} -q "SELECT * FROM remote('127.0.0.1:${CLICKHOUSE_PORT_TCP}', 'system', 'one', '${TEST_USER}', 'pass')" -u "${TEST_USER}" --password "pass" +${CLICKHOUSE_CLIENT} -q "SELECT * FROM mysql('127.0.0.1:9004', 'system', 'one', '${TEST_USER}', 'pass')" -u "${TEST_USER}" --password "pass" + +${CLICKHOUSE_CLIENT} -q "SYSTEM FLUSH LOGS" + +echo "client_port 0 connections:" +${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user = '${TEST_USER}' and client_port = 0" + +echo "client_address '::' connections:" +${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user = '${TEST_USER}' and client_address = toIPv6('::')" + +echo "login failures:" +${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user = '${TEST_USER}' and type = 'LoginFailure'" + +# remote(...) function sometimes reuses old cached sessions for query execution. +# This makes LoginSuccess/Logout entries count unstable, but success and logouts must always match. + +for interface in 'TCP' 'HTTP' 'MySQL' +do + LOGIN_COUNT=`${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'LoginSuccess' AND interface = '${interface}'"` + CORRESPONDING_LOGOUT_RECORDS_COUNT=`${CLICKHOUSE_CLIENT} -q "SELECT COUNT(*) FROM (SELECT ${SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'LoginSuccess' AND interface = '${interface}' INTERSECT SELECT ${SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'Logout' AND interface = '${interface}')"` + + if [ "$LOGIN_COUNT" == "$CORRESPONDING_LOGOUT_RECORDS_COUNT" ]; then + echo "${interface} Login and logout count is equal" + else + TOTAL_LOGOUT_COUNT=`${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'Logout' AND interface = '${interface}'"` + echo "${interface} Login count ${LOGIN_COUNT} != corresponding logout count ${CORRESPONDING_LOGOUT_RECORDS_COUNT}. TOTAL_LOGOUT_COUNT ${TOTAL_LOGOUT_COUNT}" + fi +done + +${CLICKHOUSE_CLIENT} -q "DROP USER ${TEST_USER}" diff --git a/tests/queries/0_stateless/02835_drop_user_during_session.reference b/tests/queries/0_stateless/02835_drop_user_during_session.reference new file mode 100644 index 00000000000..7252faab8c6 --- /dev/null +++ b/tests/queries/0_stateless/02835_drop_user_during_session.reference @@ -0,0 +1,8 @@ +port_0_sessions: +0 +address_0_sessions: +0 +Corresponding LoginSuccess/Logout +9 +LoginFailure +0 diff --git a/tests/queries/0_stateless/02835_drop_user_during_session.sh b/tests/queries/0_stateless/02835_drop_user_during_session.sh new file mode 100755 index 00000000000..347ebd22f96 --- /dev/null +++ b/tests/queries/0_stateless/02835_drop_user_during_session.sh @@ -0,0 +1,114 @@ +#!/usr/bin/env bash +# Tags: no-debug + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +readonly PID=$$ + +readonly TEST_USER="02835_USER_${PID}" +readonly TEST_ROLE="02835_ROLE_${PID}" +readonly TEST_PROFILE="02835_PROFILE_${PID}" +readonly SESSION_LOG_MATCHING_FIELDS="auth_id, auth_type, client_version_major, client_version_minor, client_version_patch, interface" + +function tcp_session() +{ + local user=$1 + ${CLICKHOUSE_CLIENT} -q "SELECT COUNT(*) FROM system.numbers" --user="${user}" +} + +function http_session() +{ + local user=$1 + ${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&user=${user}&password=pass" -d "SELECT COUNT(*) FROM system.numbers" +} + +function http_with_session_id_session() +{ + local user=$1 + ${CLICKHOUSE_CURL} -sS "${CLICKHOUSE_URL}&user=${user}&password=pass" -d "SELECT COUNT(*) FROM system.numbers" +} + +# Busy-waits until user $1, specified amount of queries ($2) will run simultaneously. +function wait_for_queries_start() +{ + local user=$1 + local queries_count=$2 + # 10 seconds waiting + counter=0 retries=100 + while [[ $counter -lt $retries ]]; do + result=$($CLICKHOUSE_CLIENT --query "SELECT COUNT(*) FROM system.processes WHERE user = '${user}'") + if [[ $result == "${queries_count}" ]]; then + break; + fi + sleep 0.1 + ((++counter)) + done +} + +${CLICKHOUSE_CLIENT} -q "SYSTEM FLUSH LOGS" +${CLICKHOUSE_CLIENT} -q "DELETE FROM system.session_log WHERE user = '${TEST_USER}'" + +# DROP USE CASE +${CLICKHOUSE_CLIENT} -q "CREATE USER IF NOT EXISTS ${TEST_USER}" +${CLICKHOUSE_CLIENT} -q "GRANT SELECT ON system.numbers TO ${TEST_USER}" + +export -f tcp_session; +export -f http_session; +export -f http_with_session_id_session; + +timeout 10s bash -c "tcp_session ${TEST_USER}" >/dev/null 2>&1 & +timeout 10s bash -c "http_session ${TEST_USER}" >/dev/null 2>&1 & +timeout 10s bash -c "http_with_session_id_session ${TEST_USER}" >/dev/null 2>&1 & + +wait_for_queries_start $TEST_USER 3 +${CLICKHOUSE_CLIENT} -q "DROP USER ${TEST_USER}" +${CLICKHOUSE_CLIENT} -q "KILL QUERY WHERE user = '${TEST_USER}' SYNC" >/dev/null & + +wait + +# DROP ROLE CASE +${CLICKHOUSE_CLIENT} -q "CREATE ROLE IF NOT EXISTS ${TEST_ROLE}" +${CLICKHOUSE_CLIENT} -q "CREATE USER ${TEST_USER} DEFAULT ROLE ${TEST_ROLE}" +${CLICKHOUSE_CLIENT} -q "GRANT SELECT ON system.numbers TO ${TEST_USER}" + +timeout 10s bash -c "tcp_session ${TEST_USER}" >/dev/null 2>&1 & +timeout 10s bash -c "http_session ${TEST_USER}" >/dev/null 2>&1 & +timeout 10s bash -c "http_with_session_id_session ${TEST_USER}" >/dev/null 2>&1 & + +wait_for_queries_start $TEST_USER 3 +${CLICKHOUSE_CLIENT} -q "DROP ROLE ${TEST_ROLE}" +${CLICKHOUSE_CLIENT} -q "DROP USER ${TEST_USER}" + +${CLICKHOUSE_CLIENT} -q "KILL QUERY WHERE user = '${TEST_USER}' SYNC" >/dev/null & + +wait + +# DROP PROFILE CASE +${CLICKHOUSE_CLIENT} -q "CREATE SETTINGS PROFILE IF NOT EXISTS '${TEST_PROFILE}'" +${CLICKHOUSE_CLIENT} -q "CREATE USER ${TEST_USER} SETTINGS PROFILE '${TEST_PROFILE}'" +${CLICKHOUSE_CLIENT} -q "GRANT SELECT ON system.numbers TO ${TEST_USER}" + +timeout 10s bash -c "tcp_session ${TEST_USER}" >/dev/null 2>&1 & +timeout 10s bash -c "http_session ${TEST_USER}" >/dev/null 2>&1 & +timeout 10s bash -c "http_with_session_id_session ${TEST_USER}" >/dev/null 2>&1 & + +wait_for_queries_start $TEST_USER 3 +${CLICKHOUSE_CLIENT} -q "DROP SETTINGS PROFILE '${TEST_PROFILE}'" +${CLICKHOUSE_CLIENT} -q "DROP USER ${TEST_USER}" + +${CLICKHOUSE_CLIENT} -q "KILL QUERY WHERE user = '${TEST_USER}' SYNC" >/dev/null & + +wait + +${CLICKHOUSE_CLIENT} -q "SYSTEM FLUSH LOGS" + +echo "port_0_sessions:" +${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user = '${TEST_USER}' AND client_port = 0" +echo "address_0_sessions:" +${CLICKHOUSE_CLIENT} -q "SELECT count(*) FROM system.session_log WHERE user = '${TEST_USER}' AND client_address = toIPv6('::')" +echo "Corresponding LoginSuccess/Logout" +${CLICKHOUSE_CLIENT} -q "SELECT COUNT(*) FROM (SELECT ${SESSION_LOG_MATCHING_FIELDS} FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'LoginSuccess' INTERSECT SELECT ${SESSION_LOG_MATCHING_FIELDS}, FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'Logout')" +echo "LoginFailure" +${CLICKHOUSE_CLIENT} -q "SELECT COUNT(*) FROM system.session_log WHERE user = '${TEST_USER}' AND type = 'LoginFailure'" diff --git a/tests/queries/0_stateless/02836_file_diagnostics_while_reading_header.reference b/tests/queries/0_stateless/02836_file_diagnostics_while_reading_header.reference new file mode 100644 index 00000000000..6829f972684 --- /dev/null +++ b/tests/queries/0_stateless/02836_file_diagnostics_while_reading_header.reference @@ -0,0 +1,2 @@ +in file/uri +test.csv diff --git a/tests/queries/0_stateless/02836_file_diagnostics_while_reading_header.sh b/tests/queries/0_stateless/02836_file_diagnostics_while_reading_header.sh new file mode 100755 index 00000000000..d1b5ffa2af8 --- /dev/null +++ b/tests/queries/0_stateless/02836_file_diagnostics_while_reading_header.sh @@ -0,0 +1,11 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +FILENAME="${CLICKHOUSE_TMP}/test.csv" + +printf 'Bad\rHeader\n123\n' > "${FILENAME}" +${CLICKHOUSE_LOCAL} --query "SELECT * FROM file('${CLICKHOUSE_TMP}/t*e*s*t.csv')" 2>&1 | grep -o -P 'in file/uri|test\.csv' +rm "${FILENAME}" diff --git a/utils/CMakeLists.txt b/utils/CMakeLists.txt index 47dd2fc9f2d..4e1184cc9a5 100644 --- a/utils/CMakeLists.txt +++ b/utils/CMakeLists.txt @@ -20,7 +20,6 @@ if (ENABLE_UTILS) add_subdirectory (zookeeper-cli) add_subdirectory (zookeeper-dump-tree) add_subdirectory (zookeeper-remove-by-list) - add_subdirectory (wikistat-loader) add_subdirectory (check-marks) add_subdirectory (checksum-for-compressed-block) add_subdirectory (check-mysql-binlog) diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt index aefb14d098f..778e554c40d 100644 --- a/utils/check-style/aspell-ignore/en/aspell-dict.txt +++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt @@ -1404,6 +1404,7 @@ encodings encryptions endian endsWith +endsWithUTF enum enum's enums @@ -2210,6 +2211,7 @@ src stacktrace stacktraces startsWith +startsWithUTF statbox stateful stddev diff --git a/utils/prepare-time-trace/prepare-time-trace.sh b/utils/prepare-time-trace/prepare-time-trace.sh index 300a32b0fd1..7cacdec8c94 100755 --- a/utils/prepare-time-trace/prepare-time-trace.sh +++ b/utils/prepare-time-trace/prepare-time-trace.sh @@ -35,7 +35,6 @@ ENGINE = MergeTree ORDER BY (date, file, name, args_name); INPUT_DIR=$1 OUTPUT_DIR=$2 -EXTRA_COLUMN_VALUES=$3 find "$INPUT_DIR" -name '*.json' | grep -P '\.(c|cpp|cc|cxx)\.json$' | xargs -P $(nproc) -I{} bash -c " @@ -43,7 +42,7 @@ find "$INPUT_DIR" -name '*.json' | grep -P '\.(c|cpp|cc|cxx)\.json$' | xargs -P LIBRARY_NAME=\$(echo '{}' | sed -r -e 's!^.*/CMakeFiles/([^/]+)\.dir/.*\$!\1!') START_TIME=\$(jq '.beginningOfTime' '{}') - jq -c '.traceEvents[] | [${EXTRA_COLUMN_VALUES} \"'\"\$ORIGINAL_FILENAME\"'\", \"'\"\$LIBRARY_NAME\"'\", '\$START_TIME', .pid, .tid, .ph, .ts, .dur, .cat, .name, .args.detail, .args.count, .args[\"avg ms\"], .args.name]' '{}' > \"${OUTPUT_DIR}/\$\$\" + jq -c '.traceEvents[] | [\"'\"\$ORIGINAL_FILENAME\"'\", \"'\"\$LIBRARY_NAME\"'\", '\$START_TIME', .pid, .tid, .ph, .ts, .dur, .cat, .name, .args.detail, .args.count, .args[\"avg ms\"], .args.name]' '{}' > \"${OUTPUT_DIR}/\$\$\" " # Now you can upload it as follows: diff --git a/utils/wikistat-loader/CMakeLists.txt b/utils/wikistat-loader/CMakeLists.txt deleted file mode 100644 index fc5416dea2e..00000000000 --- a/utils/wikistat-loader/CMakeLists.txt +++ /dev/null @@ -1,2 +0,0 @@ -clickhouse_add_executable (wikistat-loader main.cpp ${SRCS}) -target_link_libraries (wikistat-loader PRIVATE clickhouse_common_io boost::program_options) diff --git a/utils/wikistat-loader/main.cpp b/utils/wikistat-loader/main.cpp deleted file mode 100644 index 493f1df05da..00000000000 --- a/utils/wikistat-loader/main.cpp +++ /dev/null @@ -1,225 +0,0 @@ -#include - -#include -#include -#include -#include -#include -#include -#include - - -/** Reads uncompressed wikistat data from stdin, - * and writes transformed data in tsv format, - * ready to be loaded into ClickHouse. - * - * Input data has format: - * - * aa Wikipedia 1 17224 - * aa.b Main_Page 2 21163 - * - * project, optional subproject, path, hits, total size in bytes. - */ - - -template -static void readString(std::string & s, DB::ReadBuffer & buf) -{ - s.clear(); - - while (!buf.eof()) - { - const char * next_pos; - - if (break_at_dot) - next_pos = find_first_symbols<' ', '\n', '.'>(buf.position(), buf.buffer().end()); - else - next_pos = find_first_symbols<' ', '\n'>(buf.position(), buf.buffer().end()); - - s.append(buf.position(), next_pos - buf.position()); - buf.position() += next_pos - buf.position(); - - if (!buf.hasPendingData()) - continue; - - if (*buf.position() == ' ' || *buf.position() == '\n' || (break_at_dot && *buf.position() == '.')) - return; - } -} - - -/** Reads path before whitespace and decodes %xx sequences (to more compact and handy representation), - * except %2F '/', %26 '&', %3D '=', %3F '?', %23 '#' (to not break structure of URL). - */ -static void readPath(std::string & s, DB::ReadBuffer & buf) -{ - s.clear(); - - while (!buf.eof()) - { - const char * next_pos = find_first_symbols<' ', '\n', '%'>(buf.position(), buf.buffer().end()); - - s.append(buf.position(), next_pos - buf.position()); - buf.position() += next_pos - buf.position(); - - if (!buf.hasPendingData()) - continue; - - if (*buf.position() == ' ' || *buf.position() == '\n') - return; - - if (*buf.position() == '%') - { - ++buf.position(); - - char c1; - char c2; - - if (buf.eof() || *buf.position() == ' ') - break; - - DB::readChar(c1, buf); - - if (buf.eof() || *buf.position() == ' ') - break; - - DB::readChar(c2, buf); - - if ((c1 == '2' && (c2 == 'f' || c2 == '6' || c2 == '3' || c2 == 'F')) - || (c1 == '3' && (c2 == 'd' || c2 == 'f' || c2 == 'D' || c2 == 'F'))) - { - s += '%'; - s += c1; - s += c2; - } - else - s += static_cast(static_cast(unhex(c1)) * 16 + static_cast(unhex(c2))); - } - } -} - - -static void skipUntilNewline(DB::ReadBuffer & buf) -{ - while (!buf.eof()) - { - const char * next_pos = find_first_symbols<'\n'>(buf.position(), buf.buffer().end()); - - buf.position() += next_pos - buf.position(); - - if (!buf.hasPendingData()) - continue; - - if (*buf.position() == '\n') - { - ++buf.position(); - return; - } - } -} - - -namespace DB -{ - namespace ErrorCodes - { - extern const int CANNOT_PARSE_INPUT_ASSERTION_FAILED; - } -} - - -int main(int argc, char ** argv) -try -{ - boost::program_options::options_description desc("Allowed options"); - desc.add_options() - ("help,h", "produce help message") - ("time", boost::program_options::value()->required(), - "time of data in YYYY-MM-DD hh:mm:ss form") - ; - - boost::program_options::variables_map options; - boost::program_options::store(boost::program_options::parse_command_line(argc, argv, desc), options); - - if (options.count("help")) - { - std::cout << "Reads uncompressed wikistat data from stdin and writes transformed data in tsv format." << std::endl; - std::cout << "Usage: " << argv[0] << " --time='YYYY-MM-DD hh:00:00' < in > out" << std::endl; - std::cout << desc << std::endl; - return 1; - } - - std::string time_str = options.at("time").as(); - LocalDateTime time(time_str); - LocalDate date(time_str); - - DB::ReadBufferFromFileDescriptor in(STDIN_FILENO); - DB::WriteBufferFromFileDescriptor out(STDOUT_FILENO); - - std::string project; - std::string subproject; - std::string path; - UInt64 hits = 0; - UInt64 size = 0; - - size_t row_num = 0; - while (!in.eof()) - { - try - { - ++row_num; - readString(project, in); - - if (in.eof()) - break; - - if (*in.position() == '.') - readString(subproject, in); - else - subproject.clear(); - - DB::assertChar(' ', in); - readPath(path, in); - DB::assertChar(' ', in); - DB::readIntText(hits, in); - DB::assertChar(' ', in); - DB::readIntText(size, in); - DB::assertChar('\n', in); - } - catch (const DB::Exception & e) - { - /// Sometimes, input data has errors. For example, look at first lines in pagecounts-20130210-130000.gz - /// To save rest of data, just skip lines with errors. - if (e.code() == DB::ErrorCodes::CANNOT_PARSE_INPUT_ASSERTION_FAILED) - { - std::cerr << "At row " << row_num << ": " << DB::getCurrentExceptionMessage(false) << '\n'; - skipUntilNewline(in); - continue; - } - else - throw; - } - - DB::writeText(date, out); - DB::writeChar('\t', out); - DB::writeText(time, out); - DB::writeChar('\t', out); - DB::writeText(project, out); - DB::writeChar('\t', out); - DB::writeText(subproject, out); - DB::writeChar('\t', out); - DB::writeText(path, out); - DB::writeChar('\t', out); - DB::writeText(hits, out); - DB::writeChar('\t', out); - DB::writeText(size, out); - DB::writeChar('\n', out); - } - - return 0; -} -catch (...) -{ - std::cerr << DB::getCurrentExceptionMessage(true) << '\n'; - throw; -}