mirror of
https://github.com/ClickHouse/ClickHouse.git
synced 2024-09-20 00:30:49 +00:00
Merge branch 'master' into optimize_rewrite_sum_if_to_count_if
This commit is contained in:
Raúl Marín
GitHub
commit
8df67eb7ed
37
.github/workflows/master.yml
vendored
37
.github/workflows/master.yml
vendored
@ -59,16 +59,17 @@ jobs:
|
||||
uses: ./.github/workflows/reusable_docker.yml
|
||||
with:
|
||||
data: ${{ needs.RunConfig.outputs.data }}
|
||||
StyleCheck:
|
||||
needs: [RunConfig, BuildDockers]
|
||||
if: ${{ !failure() && !cancelled() }}
|
||||
uses: ./.github/workflows/reusable_test.yml
|
||||
with:
|
||||
test_name: Style check
|
||||
runner_type: style-checker
|
||||
data: ${{ needs.RunConfig.outputs.data }}
|
||||
run_command: |
|
||||
python3 style_check.py --no-push
|
||||
# Tested in MQ
|
||||
# StyleCheck:
|
||||
# needs: [RunConfig, BuildDockers]
|
||||
# if: ${{ !failure() && !cancelled() }}
|
||||
# uses: ./.github/workflows/reusable_test.yml
|
||||
# with:
|
||||
# test_name: Style check
|
||||
# runner_type: style-checker
|
||||
# data: ${{ needs.RunConfig.outputs.data }}
|
||||
# run_command: |
|
||||
# python3 style_check.py --no-push
|
||||
CompatibilityCheckX86:
|
||||
needs: [RunConfig, BuilderDebRelease]
|
||||
if: ${{ !failure() && !cancelled() }}
|
||||
@ -447,6 +448,14 @@ jobs:
|
||||
test_name: Stateless tests (debug)
|
||||
runner_type: func-tester
|
||||
data: ${{ needs.RunConfig.outputs.data }}
|
||||
FunctionalStatelessTestAsanAzure:
|
||||
needs: [RunConfig, BuilderDebAsan]
|
||||
if: ${{ !failure() && !cancelled() }}
|
||||
uses: ./.github/workflows/reusable_test.yml
|
||||
with:
|
||||
test_name: Stateless tests (azure, asan)
|
||||
runner_type: func-tester
|
||||
data: ${{ needs.RunConfig.outputs.data }}
|
||||
##############################################################################################
|
||||
############################ FUNCTIONAl STATEFUL TESTS #######################################
|
||||
##############################################################################################
|
||||
@ -597,6 +606,14 @@ jobs:
|
||||
test_name: Stress test (tsan)
|
||||
runner_type: stress-tester
|
||||
data: ${{ needs.RunConfig.outputs.data }}
|
||||
StressTestTsanAzure:
|
||||
needs: [RunConfig, BuilderDebTsan]
|
||||
if: ${{ !failure() && !cancelled() }}
|
||||
uses: ./.github/workflows/reusable_test.yml
|
||||
with:
|
||||
test_name: Stress test (azure, tsan)
|
||||
runner_type: stress-tester
|
||||
data: ${{ needs.RunConfig.outputs.data }}
|
||||
StressTestMsan:
|
||||
needs: [RunConfig, BuilderDebMsan]
|
||||
if: ${{ !failure() && !cancelled() }}
|
||||
|
||||
4
.github/workflows/pull_request.yml
vendored
4
.github/workflows/pull_request.yml
vendored
@ -158,7 +158,7 @@ jobs:
|
||||
#
|
||||
FinishCheck:
|
||||
if: ${{ !failure() && !cancelled() }}
|
||||
needs: [Tests_1, Tests_2, Builds_1_Report, Builds_2_Report]
|
||||
needs: [RunConfig, BuildDockers, StyleCheck, FastTest, Builds_1, Builds_2, Builds_1_Report, Builds_2_Report, Tests_1, Tests_2]
|
||||
runs-on: [self-hosted, style-checker]
|
||||
steps:
|
||||
- name: Check out repository code
|
||||
@ -171,7 +171,7 @@ jobs:
|
||||
- name: Finish label
|
||||
run: |
|
||||
cd "$GITHUB_WORKSPACE/tests/ci"
|
||||
python3 finish_check.py
|
||||
python3 finish_check.py ${{ (contains(needs.*.result, 'failure') && github.event_name == 'merge_group') && '--pipeline-failure' || '' }}
|
||||
- name: Auto merge if approved
|
||||
if: ${{ github.event_name != 'merge_group' }}
|
||||
run: |
|
||||
|
||||
@ -16,7 +16,7 @@
|
||||
#ci_set_reduced
|
||||
#ci_set_arm
|
||||
#ci_set_integration
|
||||
#ci_set_analyzer
|
||||
#ci_set_old_analyzer
|
||||
|
||||
## To run specified job in CI:
|
||||
#job_<JOB NAME>
|
||||
|
||||
@ -38,9 +38,9 @@
|
||||
* Optimized function `dotProduct` to omit unnecessary and expensive memory copies. [#60928](https://github.com/ClickHouse/ClickHouse/pull/60928) ([Robert Schulze](https://github.com/rschu1ze)).
|
||||
* 30x faster printing for 256-bit integers. [#61100](https://github.com/ClickHouse/ClickHouse/pull/61100) ([Raúl Marín](https://github.com/Algunenano)).
|
||||
* If the table's primary key contains mostly useless columns, don't keep them in memory. This is controlled by a new setting `primary_key_ratio_of_unique_prefix_values_to_skip_suffix_columns` with the value `0.9` by default, which means: for a composite primary key, if a column changes its value for at least 0.9 of all the times, the next columns after it will be not loaded. [#60255](https://github.com/ClickHouse/ClickHouse/pull/60255) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
|
||||
* Improve the performance of serialized aggregation method when involving multiple `Nullable` columns. [#55809](https://github.com/ClickHouse/ClickHouse/pull/55809) ([Amos Bird](https://github.com/amosbird)).
|
||||
* Lazy build JSON's output to improve performance of ALL JOIN. [#58278](https://github.com/ClickHouse/ClickHouse/pull/58278) ([LiuNeng](https://github.com/liuneng1994)).
|
||||
* Make HTTP/HTTPs connections with external services, such as AWS S3 reusable for all uses cases. Even when response is 3xx or 4xx. [#58845](https://github.com/ClickHouse/ClickHouse/pull/58845) ([Sema Checherinda](https://github.com/CheSema)).
|
||||
* Improve the performance of serialized aggregation methods when involving multiple `Nullable` columns. [#55809](https://github.com/ClickHouse/ClickHouse/pull/55809) ([Amos Bird](https://github.com/amosbird)).
|
||||
* Lazy builds JSON's output to improve performance of ALL JOIN. [#58278](https://github.com/ClickHouse/ClickHouse/pull/58278) ([LiuNeng](https://github.com/liuneng1994)).
|
||||
* Make HTTP/HTTPs connections with external services, such as AWS S3 reusable for all use cases. Even when the response is 3xx or 4xx. [#58845](https://github.com/ClickHouse/ClickHouse/pull/58845) ([Sema Checherinda](https://github.com/CheSema)).
|
||||
* Improvements to aggregate functions `argMin` / `argMax` / `any` / `anyLast` / `anyHeavy`, as well as `ORDER BY {u8/u16/u32/u64/i8/i16/u32/i64) LIMIT 1` queries. [#58640](https://github.com/ClickHouse/ClickHouse/pull/58640) ([Raúl Marín](https://github.com/Algunenano)).
|
||||
* Trivial optimization for column's filter. Peak memory can be reduced to 44% of the original in some cases. [#59698](https://github.com/ClickHouse/ClickHouse/pull/59698) ([李扬](https://github.com/taiyang-li)).
|
||||
* Execute `multiIf` function in a columnar fashion when the result type's underlying type is a number. [#60384](https://github.com/ClickHouse/ClickHouse/pull/60384) ([李扬](https://github.com/taiyang-li)).
|
||||
@ -49,7 +49,7 @@
|
||||
* Optimize data movement between columns of a Nullable number or a Nullable string, which improves some micro-benchmarks. [#60846](https://github.com/ClickHouse/ClickHouse/pull/60846) ([李扬](https://github.com/taiyang-li)).
|
||||
* Operations with the filesystem cache will suffer less from the lock contention. [#61066](https://github.com/ClickHouse/ClickHouse/pull/61066) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
|
||||
* Optimize array join and other JOINs by preventing a wrong compiler's optimization. Close [#61074](https://github.com/ClickHouse/ClickHouse/issues/61074). [#61075](https://github.com/ClickHouse/ClickHouse/pull/61075) ([李扬](https://github.com/taiyang-li)).
|
||||
* If a query with a syntax error contained `COLUMNS` matcher with a regular expression, the regular expression was compiled each time during the parser's backtracking, instead of being compiled once. This was a fundamental error. The compiled regexp was put to AST. But the letter A in AST means "abstract" which means it should not contain heavyweight objects. Parts of AST can be created and discarded during parsing, including a large number of backtracking. This leads to slowness on the parsing side and consequently allows DoS by a readonly user. But the main problem is that it prevents progress in fuzzers. [#61543](https://github.com/ClickHouse/ClickHouse/pull/61543) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
|
||||
* If a query with a syntax error contained the `COLUMNS` matcher with a regular expression, the regular expression was compiled each time during the parser's backtracking, instead of being compiled once. This was a fundamental error. The compiled regexp was put to AST. But the letter A in AST means "abstract" which means it should not contain heavyweight objects. Parts of AST can be created and discarded during parsing, including a large number of backtracking. This leads to slowness on the parsing side and consequently allows DoS by a readonly user. But the main problem is that it prevents progress in fuzzers. [#61543](https://github.com/ClickHouse/ClickHouse/pull/61543) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
|
||||
* Add a new analyzer pass to optimize the IN operator for a single value. [#61564](https://github.com/ClickHouse/ClickHouse/pull/61564) ([LiuNeng](https://github.com/liuneng1994)).
|
||||
* DNSResolver shuffles set of resolved IPs which is needed to uniformly utilize multiple endpoints of AWS S3. [#60965](https://github.com/ClickHouse/ClickHouse/pull/60965) ([Sema Checherinda](https://github.com/CheSema)).
|
||||
|
||||
|
||||
@ -8,9 +8,6 @@ option (SANITIZE "Enable one of the code sanitizers" "")
|
||||
|
||||
set (SAN_FLAGS "${SAN_FLAGS} -g -fno-omit-frame-pointer -DSANITIZER")
|
||||
|
||||
# It's possible to pass an ignore list to sanitizers (-fsanitize-ignorelist). Intentionally not doing this because
|
||||
# 1. out-of-source suppressions are awkward 2. it seems ignore lists don't work after the Clang v16 upgrade (#49829)
|
||||
|
||||
if (SANITIZE)
|
||||
if (SANITIZE STREQUAL "address")
|
||||
set (ASAN_FLAGS "-fsanitize=address -fsanitize-address-use-after-scope")
|
||||
|
||||
2
contrib/curl
vendored
2
contrib/curl
vendored
@ -1 +1 @@
|
||||
Subproject commit 1a05e833f8f7140628b27882b10525fd9ec4b873
|
||||
Subproject commit de7b3e89218467159a7af72d58cea8425946e97d
|
||||
@ -33,14 +33,15 @@ set (SRCS
|
||||
"${LIBRARY_DIR}/lib/curl_memrchr.c"
|
||||
"${LIBRARY_DIR}/lib/curl_multibyte.c"
|
||||
"${LIBRARY_DIR}/lib/curl_ntlm_core.c"
|
||||
"${LIBRARY_DIR}/lib/curl_ntlm_wb.c"
|
||||
"${LIBRARY_DIR}/lib/curl_path.c"
|
||||
"${LIBRARY_DIR}/lib/curl_range.c"
|
||||
"${LIBRARY_DIR}/lib/curl_rtmp.c"
|
||||
"${LIBRARY_DIR}/lib/curl_sasl.c"
|
||||
"${LIBRARY_DIR}/lib/curl_sha512_256.c"
|
||||
"${LIBRARY_DIR}/lib/curl_sspi.c"
|
||||
"${LIBRARY_DIR}/lib/curl_threads.c"
|
||||
"${LIBRARY_DIR}/lib/curl_trc.c"
|
||||
"${LIBRARY_DIR}/lib/cw-out.c"
|
||||
"${LIBRARY_DIR}/lib/dict.c"
|
||||
"${LIBRARY_DIR}/lib/doh.c"
|
||||
"${LIBRARY_DIR}/lib/dynbuf.c"
|
||||
@ -98,6 +99,7 @@ set (SRCS
|
||||
"${LIBRARY_DIR}/lib/psl.c"
|
||||
"${LIBRARY_DIR}/lib/rand.c"
|
||||
"${LIBRARY_DIR}/lib/rename.c"
|
||||
"${LIBRARY_DIR}/lib/request.c"
|
||||
"${LIBRARY_DIR}/lib/rtsp.c"
|
||||
"${LIBRARY_DIR}/lib/select.c"
|
||||
"${LIBRARY_DIR}/lib/sendf.c"
|
||||
|
||||
@ -38,6 +38,7 @@
|
||||
|
||||
#define HAVE_ARPA_INET_H
|
||||
#define HAVE_ERRNO_H
|
||||
#define HAVE_GETSOCKNAME
|
||||
#define HAVE_FCNTL_H
|
||||
#define HAVE_NETDB_H
|
||||
#define HAVE_NETINET_IN_H
|
||||
|
||||
@ -14,11 +14,14 @@ RUN curl -o krb5-libs-1.10.3-65.el6.x86_64.rpm ftp://ftp.pbone.net/mirror/vault.
|
||||
rpm -Uvh libkadm5-1.10.3-65.el6.x86_64.rpm libss-1.41.12-24.el6.x86_64.rpm krb5-libs-1.10.3-65.el6.x86_64.rpm krb5-workstation-1.10.3-65.el6.x86_64.rpm libcom_err-1.41.12-24.el6.x86_64.rpm && \
|
||||
rm -fr *.rpm
|
||||
|
||||
ADD https://archive.apache.org/dist/commons/daemon/source/commons-daemon-1.0.15-src.tar.gz /tmp/commons-daemon-1.0.15-src.tar.gz
|
||||
|
||||
RUN cd /tmp && \
|
||||
curl http://archive.apache.org/dist/commons/daemon/source/commons-daemon-1.0.15-src.tar.gz -o commons-daemon-1.0.15-src.tar.gz && \
|
||||
tar xzf commons-daemon-1.0.15-src.tar.gz && \
|
||||
cd commons-daemon-1.0.15-src/src/native/unix && \
|
||||
./configure && \
|
||||
make && \
|
||||
cp ./jsvc /usr/local/hadoop-2.7.0/sbin && \
|
||||
[ -e /usr/local/hadoop ] || ln -s ./hadoop-2.7.0 /usr/local/hadoop
|
||||
cd /tmp && \
|
||||
rm -rf commons-daemon-1.0.15-src* && \
|
||||
{ [ -e /usr/local/hadoop ] || ln -s ./hadoop-2.7.0 /usr/local/hadoop; }
|
||||
|
||||
@ -19,7 +19,7 @@ ln -s /usr/share/clickhouse-test/clickhouse-test /usr/bin/clickhouse-test
|
||||
# install test configs
|
||||
/usr/share/clickhouse-test/config/install.sh
|
||||
|
||||
azurite-blob --blobHost 0.0.0.0 --blobPort 10000 --debug /azurite_log &
|
||||
azurite-blob --blobHost 0.0.0.0 --blobPort 10000 --silent --inMemoryPersistence &
|
||||
./setup_minio.sh stateful
|
||||
|
||||
config_logs_export_cluster /etc/clickhouse-server/config.d/system_logs_export.yaml
|
||||
@ -87,7 +87,7 @@ function start()
|
||||
tail -n1000 /var/log/clickhouse-server/clickhouse-server.log
|
||||
break
|
||||
fi
|
||||
timeout 120 service clickhouse-server start
|
||||
timeout 120 sudo -E -u clickhouse /usr/bin/clickhouse-server --config /etc/clickhouse-server/config.xml --daemon --pid-file /var/run/clickhouse-server/clickhouse-server.pid
|
||||
sleep 0.5
|
||||
counter=$((counter + 1))
|
||||
done
|
||||
|
||||
@ -42,12 +42,6 @@ source /utils.lib
|
||||
# install test configs
|
||||
/usr/share/clickhouse-test/config/install.sh
|
||||
|
||||
if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then
|
||||
echo "Azure is disabled"
|
||||
else
|
||||
azurite-blob --blobHost 0.0.0.0 --blobPort 10000 --debug /azurite_log &
|
||||
fi
|
||||
|
||||
./setup_minio.sh stateless
|
||||
./setup_hdfs_minicluster.sh
|
||||
|
||||
@ -97,12 +91,11 @@ if [ "$NUM_TRIES" -gt "1" ]; then
|
||||
export THREAD_FUZZER_pthread_mutex_unlock_AFTER_SLEEP_TIME_US_MAX=10000
|
||||
|
||||
mkdir -p /var/run/clickhouse-server
|
||||
# simplest way to forward env variables to server
|
||||
sudo -E -u clickhouse /usr/bin/clickhouse-server --config /etc/clickhouse-server/config.xml --daemon --pid-file /var/run/clickhouse-server/clickhouse-server.pid
|
||||
else
|
||||
sudo clickhouse start
|
||||
fi
|
||||
|
||||
# simplest way to forward env variables to server
|
||||
sudo -E -u clickhouse /usr/bin/clickhouse-server --config /etc/clickhouse-server/config.xml --daemon --pid-file /var/run/clickhouse-server/clickhouse-server.pid
|
||||
|
||||
if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then
|
||||
sudo sed -i "s|<filesystem_caches_path>/var/lib/clickhouse/filesystem_caches/</filesystem_caches_path>|<filesystem_caches_path>/var/lib/clickhouse/filesystem_caches_1/</filesystem_caches_path>|" /etc/clickhouse-server1/config.d/filesystem_caches_path.xml
|
||||
|
||||
@ -212,6 +205,14 @@ function run_tests()
|
||||
ADDITIONAL_OPTIONS+=('--s3-storage')
|
||||
fi
|
||||
|
||||
if [[ -n "$USE_AZURE_STORAGE_FOR_MERGE_TREE" ]] && [[ "$USE_AZURE_STORAGE_FOR_MERGE_TREE" -eq 1 ]]; then
|
||||
# to disable the same tests
|
||||
ADDITIONAL_OPTIONS+=('--s3-storage')
|
||||
# azurite is slow, but with these two settings it can be super slow
|
||||
ADDITIONAL_OPTIONS+=('--no-random-settings')
|
||||
ADDITIONAL_OPTIONS+=('--no-random-merge-tree-settings')
|
||||
fi
|
||||
|
||||
if [[ -n "$USE_SHARED_CATALOG" ]] && [[ "$USE_SHARED_CATALOG" -eq 1 ]]; then
|
||||
ADDITIONAL_OPTIONS+=('--shared-catalog')
|
||||
fi
|
||||
@ -286,7 +287,7 @@ stop_logs_replication
|
||||
failed_to_save_logs=0
|
||||
for table in query_log zookeeper_log trace_log transactions_info_log metric_log
|
||||
do
|
||||
err=$( { clickhouse-client -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.tsv.zst; } 2>&1 )
|
||||
err=$(clickhouse-client -q "select * from system.$table into outfile '/test_output/$table.tsv.gz' format TSVWithNamesAndTypes")
|
||||
echo "$err"
|
||||
[[ "0" != "${#err}" ]] && failed_to_save_logs=1
|
||||
if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then
|
||||
|
||||
@ -279,7 +279,7 @@ function check_logs_for_critical_errors()
|
||||
|
||||
function collect_query_and_trace_logs()
|
||||
{
|
||||
for table in query_log trace_log
|
||||
for table in query_log trace_log metric_log
|
||||
do
|
||||
clickhouse-local --config-file=/etc/clickhouse-server/config.xml --only-system-tables -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.tsv.zst ||:
|
||||
done
|
||||
|
||||
@ -52,7 +52,6 @@ export ZOOKEEPER_FAULT_INJECTION=1
|
||||
# available for dump via clickhouse-local
|
||||
configure
|
||||
|
||||
azurite-blob --blobHost 0.0.0.0 --blobPort 10000 --debug /azurite_log &
|
||||
./setup_minio.sh stateless # to have a proper environment
|
||||
|
||||
config_logs_export_cluster /etc/clickhouse-server/config.d/system_logs_export.yaml
|
||||
|
||||
37
docs/changelogs/v23.8.13.25-lts.md
Normal file
37
docs/changelogs/v23.8.13.25-lts.md
Normal file
@ -0,0 +1,37 @@
|
||||
---
|
||||
sidebar_position: 1
|
||||
sidebar_label: 2024
|
||||
---
|
||||
|
||||
# 2024 Changelog
|
||||
|
||||
### ClickHouse release v23.8.13.25-lts (37e034f903e) FIXME as compared to v23.8.12.13-lts (bdbd0d87e5d)
|
||||
|
||||
#### Improvement
|
||||
* Backported in [#61930](https://github.com/ClickHouse/ClickHouse/issues/61930): Fixed accounting of memory allocated before attaching thread to a query or a user. [#56089](https://github.com/ClickHouse/ClickHouse/pull/56089) ([Nikita Taranov](https://github.com/nickitat)).
|
||||
|
||||
#### Build/Testing/Packaging Improvement
|
||||
* Backported in [#62007](https://github.com/ClickHouse/ClickHouse/issues/62007): Remove from the Keeper Docker image the volumes at /etc/clickhouse-keeper and /var/log/clickhouse-keeper. [#61683](https://github.com/ClickHouse/ClickHouse/pull/61683) ([Tristan](https://github.com/Tristan971)).
|
||||
|
||||
#### Bug Fix (user-visible misbehavior in an official stable release)
|
||||
|
||||
* Fix REPLACE/MOVE PARTITION with zero-copy replication [#54193](https://github.com/ClickHouse/ClickHouse/pull/54193) ([Alexander Tokmakov](https://github.com/tavplubix)).
|
||||
* Fix ATTACH query with external ON CLUSTER [#61365](https://github.com/ClickHouse/ClickHouse/pull/61365) ([Nikolay Degterinsky](https://github.com/evillique)).
|
||||
* Cancel merges before removing moved parts [#61610](https://github.com/ClickHouse/ClickHouse/pull/61610) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)).
|
||||
* Mark CANNOT_PARSE_ESCAPE_SEQUENCE error as parse error to be able to skip it in row input formats [#61883](https://github.com/ClickHouse/ClickHouse/pull/61883) ([Kruglov Pavel](https://github.com/Avogar)).
|
||||
* Try to fix segfault in Hive engine [#62578](https://github.com/ClickHouse/ClickHouse/pull/62578) ([Nikolay Degterinsky](https://github.com/evillique)).
|
||||
|
||||
#### CI Fix or Improvement (changelog entry is not required)
|
||||
|
||||
* Backported in [#62201](https://github.com/ClickHouse/ClickHouse/issues/62201):. [#62190](https://github.com/ClickHouse/ClickHouse/pull/62190) ([Konstantin Bogdanov](https://github.com/thevar1able)).
|
||||
* Backported in [#62796](https://github.com/ClickHouse/ClickHouse/issues/62796): We won't fail the job when GH fails to retrieve the job ID and URLs. [#62651](https://github.com/ClickHouse/ClickHouse/pull/62651) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
|
||||
* Backported in [#62968](https://github.com/ClickHouse/ClickHouse/issues/62968):. [#62932](https://github.com/ClickHouse/ClickHouse/pull/62932) ([Robert Schulze](https://github.com/rschu1ze)).
|
||||
|
||||
#### NO CL CATEGORY
|
||||
|
||||
* Backported in [#62585](https://github.com/ClickHouse/ClickHouse/issues/62585):. [#60078](https://github.com/ClickHouse/ClickHouse/pull/60078) ([Maksim Kita](https://github.com/kitaisreal)).
|
||||
|
||||
#### NOT FOR CHANGELOG / INSIGNIFICANT
|
||||
|
||||
* Throw on query timeout in ZooKeeperRetries [#60922](https://github.com/ClickHouse/ClickHouse/pull/60922) ([Antonio Andelic](https://github.com/antonio2368)).
|
||||
|
||||
@ -4,7 +4,7 @@ sidebar_position: 30
|
||||
sidebar_label: Replicated
|
||||
---
|
||||
|
||||
# [experimental] Replicated
|
||||
# Replicated
|
||||
|
||||
The engine is based on the [Atomic](../../engines/database-engines/atomic.md) engine. It supports replication of metadata via DDL log being written to ZooKeeper and executed on all of the replicas for a given database.
|
||||
|
||||
|
||||
@ -8,6 +8,8 @@ sidebar_label: HDFS
|
||||
|
||||
This engine provides integration with the [Apache Hadoop](https://en.wikipedia.org/wiki/Apache_Hadoop) ecosystem by allowing to manage data on [HDFS](https://hadoop.apache.org/docs/current/hadoop-project-dist/hadoop-hdfs/HdfsDesign.html) via ClickHouse. This engine is similar to the [File](../../../engines/table-engines/special/file.md#table_engines-file) and [URL](../../../engines/table-engines/special/url.md#table_engines-url) engines, but provides Hadoop-specific features.
|
||||
|
||||
This feature is not supported by ClickHouse engineers, and it is known to have a sketchy quality. In case of any problems, fix them yourself and submit a pull request.
|
||||
|
||||
## Usage {#usage}
|
||||
|
||||
``` sql
|
||||
|
||||
@ -10,7 +10,8 @@ The RecipeNLG dataset is available for download [here](https://recipenlg.cs.put.
|
||||
|
||||
1. Go to the download page [https://recipenlg.cs.put.poznan.pl/dataset](https://recipenlg.cs.put.poznan.pl/dataset).
|
||||
1. Accept Terms and Conditions and download zip file.
|
||||
1. Unpack the zip file with `unzip`. You will get the `full_dataset.csv` file.
|
||||
1. Option: Using the `md5sum dataset.zip` to validate the zip file and it should be equal to `3a168dfd0912bb034225619b3586ce76`.
|
||||
1. Unpack the zip file with `unzip dataset.zip`. You will get the `full_dataset.csv` file in the `dataset` directory.
|
||||
|
||||
## Create a Table
|
||||
|
||||
@ -72,7 +73,7 @@ Result:
|
||||
|
||||
``` text
|
||||
┌─count()─┐
|
||||
│ 2231141 │
|
||||
│ 2231142 │
|
||||
└─────────┘
|
||||
```
|
||||
|
||||
@ -115,7 +116,7 @@ Result:
|
||||
│ egg │ 160507 │
|
||||
│ baking powder │ 148277 │
|
||||
│ lemon juice │ 146414 │
|
||||
│ Salt │ 122557 │
|
||||
│ Salt │ 122558 │
|
||||
│ cinnamon │ 117927 │
|
||||
│ sour cream │ 116682 │
|
||||
│ cream cheese │ 114423 │
|
||||
|
||||
@ -327,7 +327,9 @@ Use buffering to avoid situations where a query processing error occurred after
|
||||
|
||||
## Setting a role with query parameters {#setting-role-with-query-parameters}
|
||||
|
||||
In certain scenarios, it might be required to set the granted role first, before executing the statement itself.
|
||||
This is a new feature added in ClickHouse 24.4.
|
||||
|
||||
In specific scenarios, setting the granted role first might be required before executing the statement itself.
|
||||
However, it is not possible to send `SET ROLE` and the statement together, as multi-statements are not allowed:
|
||||
|
||||
```
|
||||
@ -346,7 +348,7 @@ To overcome this limitation, you could use the `role` query parameter instead:
|
||||
curl -sS "http://localhost:8123?role=my_role" --data-binary "SELECT * FROM my_table;"
|
||||
```
|
||||
|
||||
This will be an equivalent of executing `SET ROLE my_role` before the statement.
|
||||
This will be the equivalent of executing `SET ROLE my_role` before the statement.
|
||||
|
||||
Additionally, it is possible to specify multiple `role` query parameters:
|
||||
|
||||
|
||||
@ -7,6 +7,8 @@ toc_max_heading_level: 2
|
||||
|
||||
# Core Settings
|
||||
|
||||
All below settings are also available in table [system.settings](/docs/en/operations/system-tables/settings).
|
||||
|
||||
## additional_table_filters
|
||||
|
||||
An additional filter expression that is applied after reading
|
||||
@ -3931,19 +3933,6 @@ For example, `avg(if(cond, col, null))` can be rewritten to `avgOrNullIf(cond, c
|
||||
Supported only with experimental analyzer (`allow_experimental_analyzer = 1`).
|
||||
:::
|
||||
|
||||
## allow_experimental_database_replicated {#allow_experimental_database_replicated}
|
||||
|
||||
Enables to create databases with [Replicated](../../engines/database-engines/replicated.md) engine.
|
||||
|
||||
Possible values:
|
||||
|
||||
- 0 — Disabled.
|
||||
- 1 — Enabled.
|
||||
|
||||
Default value: `0`.
|
||||
|
||||
Cloud default value: `1`.
|
||||
|
||||
## database_replicated_initial_query_timeout_sec {#database_replicated_initial_query_timeout_sec}
|
||||
|
||||
Sets how long initial DDL query should wait for Replicated database to process previous DDL queue entries in seconds.
|
||||
|
||||
@ -111,7 +111,7 @@ On newer Linux kernels transparent huge pages are alright.
|
||||
$ echo 'madvise' | sudo tee /sys/kernel/mm/transparent_hugepage/enabled
|
||||
```
|
||||
|
||||
If you want to modify the transparent huge pages setting permanently, editing the `/etc/default/grub` to add the `transparent_hugepage=never` to the `GRUB_CMDLINE_LINUX_DEFAULT` option:
|
||||
If you want to modify the transparent huge pages setting permanently, editing the `/etc/default/grub` to add the `transparent_hugepage=madvise` to the `GRUB_CMDLINE_LINUX_DEFAULT` option:
|
||||
|
||||
```bash
|
||||
$ GRUB_CMDLINE_LINUX_DEFAULT="transparent_hugepage=madvise ..."
|
||||
|
||||
@ -1907,6 +1907,12 @@ If the addition results in a value outside the bounds of the data type, the resu
|
||||
date_add(unit, value, date)
|
||||
```
|
||||
|
||||
Alternative syntax:
|
||||
|
||||
``` sql
|
||||
date_add(date, INTERVAL value unit)
|
||||
```
|
||||
|
||||
Aliases: `dateAdd`, `DATE_ADD`.
|
||||
|
||||
**Arguments**
|
||||
@ -1946,6 +1952,20 @@ Result:
|
||||
└───────────────────────────────────────────────┘
|
||||
```
|
||||
|
||||
```sql
|
||||
SELECT date_add(toDate('2018-01-01'), INTERVAL 3 YEAR);
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
```text
|
||||
┌─plus(toDate('2018-01-01'), toIntervalYear(3))─┐
|
||||
│ 2021-01-01 │
|
||||
└───────────────────────────────────────────────┘
|
||||
```
|
||||
|
||||
|
||||
|
||||
**See Also**
|
||||
|
||||
- [addDate](#addDate)
|
||||
@ -1962,6 +1982,13 @@ If the subtraction results in a value outside the bounds of the data type, the r
|
||||
date_sub(unit, value, date)
|
||||
```
|
||||
|
||||
Alternative syntax:
|
||||
|
||||
``` sql
|
||||
date_sub(date, INTERVAL value unit)
|
||||
```
|
||||
|
||||
|
||||
Aliases: `dateSub`, `DATE_SUB`.
|
||||
|
||||
**Arguments**
|
||||
@ -2002,6 +2029,19 @@ Result:
|
||||
└────────────────────────────────────────────────┘
|
||||
```
|
||||
|
||||
``` sql
|
||||
SELECT date_sub(toDate('2018-01-01'), INTERVAL 3 YEAR);
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
``` text
|
||||
┌─minus(toDate('2018-01-01'), toIntervalYear(3))─┐
|
||||
│ 2015-01-01 │
|
||||
└────────────────────────────────────────────────┘
|
||||
```
|
||||
|
||||
|
||||
**See Also**
|
||||
|
||||
- [subDate](#subDate)
|
||||
|
||||
@ -3213,6 +3213,74 @@ Result:
|
||||
└───────────────────────────────────────┘
|
||||
```
|
||||
|
||||
## connectionId
|
||||
|
||||
Retrieves the connection ID of the client that submitted the current query and returns it as a UInt64 integer.
|
||||
|
||||
**Syntax**
|
||||
|
||||
```sql
|
||||
connectionId()
|
||||
```
|
||||
|
||||
**Parameters**
|
||||
|
||||
None.
|
||||
|
||||
**Returned value**
|
||||
|
||||
Returns an integer of type UInt64.
|
||||
|
||||
**Implementation details**
|
||||
|
||||
This function is most useful in debugging scenarios or for internal purposes within the MySQL handler. It was created for compatibility with [MySQL's `CONNECTION_ID` function](https://dev.mysql.com/doc/refman/8.0/en/information-functions.html#function_connection-id) It is not typically used in production queries.
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
```sql
|
||||
SELECT connectionId();
|
||||
```
|
||||
|
||||
```response
|
||||
0
|
||||
```
|
||||
|
||||
## connection_id
|
||||
|
||||
An alias of `connectionId`. Retrieves the connection ID of the client that submitted the current query and returns it as a UInt64 integer.
|
||||
|
||||
**Syntax**
|
||||
|
||||
```sql
|
||||
connection_id()
|
||||
```
|
||||
|
||||
**Parameters**
|
||||
|
||||
None.
|
||||
|
||||
**Returned value**
|
||||
|
||||
Returns an integer of type UInt64.
|
||||
|
||||
**Implementation details**
|
||||
|
||||
This function is most useful in debugging scenarios or for internal purposes within the MySQL handler. It was created for compatibility with [MySQL's `CONNECTION_ID` function](https://dev.mysql.com/doc/refman/8.0/en/information-functions.html#function_connection-id) It is not typically used in production queries.
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
```sql
|
||||
SELECT connection_id();
|
||||
```
|
||||
|
||||
```response
|
||||
0
|
||||
```
|
||||
|
||||
## getClientHTTPHeader
|
||||
|
||||
Get the value of an HTTP header.
|
||||
|
||||
@ -79,9 +79,9 @@ round(expression [, decimal_places])
|
||||
|
||||
The rounded number of the same type as the input number.
|
||||
|
||||
### Examples
|
||||
**Examples**
|
||||
|
||||
**Example of use with Float**
|
||||
Example of usage with Float:
|
||||
|
||||
``` sql
|
||||
SELECT number / 2 AS x, round(x) FROM system.numbers LIMIT 3;
|
||||
@ -95,7 +95,7 @@ SELECT number / 2 AS x, round(x) FROM system.numbers LIMIT 3;
|
||||
└─────┴──────────────────────────┘
|
||||
```
|
||||
|
||||
**Example of use with Decimal**
|
||||
Example of usage with Decimal:
|
||||
|
||||
``` sql
|
||||
SELECT cast(number / 2 AS Decimal(10,4)) AS x, round(x) FROM system.numbers LIMIT 3;
|
||||
@ -124,9 +124,7 @@ SELECT cast(number / 2 AS Decimal(10,4)) AS x, round(x) FROM system.numbers LIM
|
||||
└────────┴──────────────────────────────────────────────────┘
|
||||
```
|
||||
|
||||
**Examples of rounding**
|
||||
|
||||
Rounding to the nearest number.
|
||||
Examples of rounding to the nearest number:
|
||||
|
||||
``` text
|
||||
round(3.2, 0) = 3
|
||||
@ -183,9 +181,7 @@ roundBankers(expression [, decimal_places])
|
||||
|
||||
A value rounded by the banker’s rounding method.
|
||||
|
||||
### Examples
|
||||
|
||||
**Example of use**
|
||||
**Examples**
|
||||
|
||||
Query:
|
||||
|
||||
@ -210,7 +206,7 @@ Result:
|
||||
└─────┴───┘
|
||||
```
|
||||
|
||||
**Examples of Banker’s rounding**
|
||||
Examples of Banker’s rounding:
|
||||
|
||||
``` text
|
||||
roundBankers(0.4) = 0
|
||||
@ -226,25 +222,180 @@ roundBankers(10.755, 2) = 10.76
|
||||
|
||||
- [round](#rounding_functions-round)
|
||||
|
||||
## roundToExp2(num)
|
||||
## roundToExp2
|
||||
|
||||
Accepts a number. If the number is less than one, it returns 0. Otherwise, it rounds the number down to the nearest (whole non-negative) degree of two.
|
||||
Accepts a number. If the number is less than one, it returns `0`. Otherwise, it rounds the number down to the nearest (whole non-negative) degree of two.
|
||||
|
||||
## roundDuration(num)
|
||||
**Syntax**
|
||||
|
||||
Accepts a number. If the number is less than one, it returns 0. Otherwise, it rounds the number down to numbers from the set: 1, 10, 30, 60, 120, 180, 240, 300, 600, 1200, 1800, 3600, 7200, 18000, 36000.
|
||||
```sql
|
||||
roundToExp2(num)
|
||||
```
|
||||
|
||||
## roundAge(num)
|
||||
**Parameters**
|
||||
|
||||
Accepts a number. If the number is
|
||||
- smaller than 1, it returns 0,
|
||||
- between 1 and 17, it returns 17,
|
||||
- between 18 and 24, it returns 18,
|
||||
- between 25 and 34, it returns 25,
|
||||
- between 35 and 44, it returns 35,
|
||||
- between 45 and 54, it returns 45,
|
||||
- larger than 55, it returns 55.
|
||||
- `num`: A number representing an age in years. [UInt](../data-types/int-uint.md)/[Float](../data-types/float.md).
|
||||
|
||||
## roundDown(num, arr)
|
||||
**Returned value**
|
||||
|
||||
- `0`, for `num` $\lt 1$. [UInt8](../data-types/int-uint.md).
|
||||
- `num` rounded down to the nearest (whole non-negative) degree of two. [UInt](../data-types/int-uint.md)/[Float](../data-types/float.md) equivalent to the input type.
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
```sql
|
||||
SELECT *, roundToExp2(*) FROM system.numbers WHERE number IN (0, 2, 5, 10, 19, 50)
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
```response
|
||||
┌─number─┬─roundToExp2(number)─┐
|
||||
│ 0 │ 0 │
|
||||
│ 2 │ 2 │
|
||||
│ 5 │ 4 │
|
||||
│ 10 │ 8 │
|
||||
│ 19 │ 16 │
|
||||
│ 50 │ 32 │
|
||||
└────────┴─────────────────────┘
|
||||
```
|
||||
|
||||
## roundDuration
|
||||
|
||||
Accepts a number. If the number is less than one, it returns `0`. Otherwise, it rounds the number down to numbers from the set of commonly used durations: `1, 10, 30, 60, 120, 180, 240, 300, 600, 1200, 1800, 3600, 7200, 18000, 36000`.
|
||||
|
||||
**Syntax**
|
||||
|
||||
```sql
|
||||
roundDuration(num)
|
||||
```
|
||||
|
||||
**Parameters**
|
||||
|
||||
- `num`: A number to round to one of the numbers in the set of common durations. [UInt](../data-types/int-uint.md)/[Float](../data-types/float.md).
|
||||
|
||||
**Returned value**
|
||||
|
||||
- `0`, for `num` $\lt 1$.
|
||||
- Otherwise, one of: `1, 10, 30, 60, 120, 180, 240, 300, 600, 1200, 1800, 3600, 7200, 18000, 36000`. [UInt16](../data-types/int-uint.md).
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
```sql
|
||||
SELECT *, roundDuration(*) FROM system.numbers WHERE number IN (0, 9, 19, 47, 101, 149, 205, 271, 421, 789, 1423, 2345, 4567, 9876, 24680, 42573)
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
```response
|
||||
┌─number─┬─roundDuration(number)─┐
|
||||
│ 0 │ 0 │
|
||||
│ 9 │ 1 │
|
||||
│ 19 │ 10 │
|
||||
│ 47 │ 30 │
|
||||
│ 101 │ 60 │
|
||||
│ 149 │ 120 │
|
||||
│ 205 │ 180 │
|
||||
│ 271 │ 240 │
|
||||
│ 421 │ 300 │
|
||||
│ 789 │ 600 │
|
||||
│ 1423 │ 1200 │
|
||||
│ 2345 │ 1800 │
|
||||
│ 4567 │ 3600 │
|
||||
│ 9876 │ 7200 │
|
||||
│ 24680 │ 18000 │
|
||||
│ 42573 │ 36000 │
|
||||
└────────┴───────────────────────┘
|
||||
```
|
||||
|
||||
## roundAge
|
||||
|
||||
Accepts a number within various commonly used ranges of human age and returns either a maximum or a minimum within that range.
|
||||
|
||||
**Syntax**
|
||||
|
||||
```sql
|
||||
roundAge(num)
|
||||
```
|
||||
|
||||
**Parameters**
|
||||
|
||||
- `age`: A number representing an age in years. [UInt](../data-types/int-uint.md)/[Float](../data-types/float.md).
|
||||
|
||||
**Returned value**
|
||||
|
||||
- Returns `0`, for $age \lt 1$.
|
||||
- Returns `17`, for $1 \leq age \leq 17$.
|
||||
- Returns `18`, for $18 \leq age \leq 24$.
|
||||
- Returns `25`, for $25 \leq age \leq 34$.
|
||||
- Returns `35`, for $35 \leq age \leq 44$.
|
||||
- Returns `45`, for $45 \leq age \leq 54$.
|
||||
- Returns `55`, for $age \geq 55$.
|
||||
|
||||
Type: [UInt8](../data-types/int-uint.md).
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
```sql
|
||||
SELECT *, roundAge(*) FROM system.numbers WHERE number IN (0, 5, 20, 31, 37, 54, 72);
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
```response
|
||||
┌─number─┬─roundAge(number)─┐
|
||||
│ 0 │ 0 │
|
||||
│ 5 │ 17 │
|
||||
│ 20 │ 18 │
|
||||
│ 31 │ 25 │
|
||||
│ 37 │ 35 │
|
||||
│ 54 │ 45 │
|
||||
│ 72 │ 55 │
|
||||
└────────┴──────────────────┘
|
||||
```
|
||||
|
||||
## roundDown
|
||||
|
||||
Accepts a number and rounds it down to an element in the specified array. If the value is less than the lowest bound, the lowest bound is returned.
|
||||
|
||||
**Syntax**
|
||||
|
||||
```sql
|
||||
roundDown(num, arr)
|
||||
```
|
||||
|
||||
**Parameters**
|
||||
|
||||
- `num`: A number to round down. [Numeric](../data-types/int-uint.md).
|
||||
- `arr`: Array of elements to round `age` down to. [Array](../data-types/array.md) of [UInt](../data-types/int-uint.md)/[Float](../data-types/float.md) type.
|
||||
|
||||
**Returned value**
|
||||
|
||||
- Number rounded down to an element in `arr`. If the value is less than the lowest bound, the lowest bound is returned. [UInt](../data-types/int-uint.md)/[Float](../data-types/float.md) type deduced from the type of `arr`.
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
```sql
|
||||
SELECT *, roundDown(*, [3, 4, 5]) FROM system.numbers WHERE number IN (0, 1, 2, 3, 4, 5)
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
```response
|
||||
┌─number─┬─roundDown(number, [3, 4, 5])─┐
|
||||
│ 0 │ 3 │
|
||||
│ 1 │ 3 │
|
||||
│ 2 │ 3 │
|
||||
│ 3 │ 3 │
|
||||
│ 4 │ 4 │
|
||||
│ 5 │ 5 │
|
||||
└────────┴──────────────────────────────┘
|
||||
```
|
||||
|
||||
@ -817,6 +817,42 @@ Result:
|
||||
└─────────────────────────────────────┘
|
||||
```
|
||||
|
||||
## flattenTuple
|
||||
|
||||
Returns a flattened `output` tuple from a nested named `input` tuple. Elements of the `output` tuple are the paths from the original `input` tuple. For instance: `Tuple(a Int, Tuple(b Int, c Int)) -> Tuple(a Int, b Int, c Int)`. `flattenTuple` can be used to select all paths from type `Object` as separate columns.
|
||||
|
||||
**Syntax**
|
||||
|
||||
```sql
|
||||
flattenTuple(input)
|
||||
```
|
||||
|
||||
**Parameters**
|
||||
|
||||
- `input`: Nested named tuple to flatten. [Tuple](../data-types/tuple).
|
||||
|
||||
**Returned value**
|
||||
|
||||
- `output` tuple whose elements are paths from the original `input`. [Tuple](../data-types/tuple).
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
``` sql
|
||||
CREATE TABLE t_flatten_tuple(t Tuple(t1 Nested(a UInt32, s String), b UInt32, t2 Tuple(k String, v UInt32))) ENGINE = Memory;
|
||||
INSERT INTO t_flatten_tuple VALUES (([(1, 'a'), (2, 'b')], 3, ('c', 4)));
|
||||
SELECT flattenTuple(t) FROM t_flatten_tuple;
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
``` text
|
||||
┌─flattenTuple(t)───────────┐
|
||||
│ ([1,2],['a','b'],3,'c',4) │
|
||||
└───────────────────────────┘
|
||||
```
|
||||
|
||||
## Distance functions
|
||||
|
||||
All supported functions are described in [distance functions documentation](../../sql-reference/functions/distance-functions.md).
|
||||
|
||||
@ -6,24 +6,29 @@ sidebar_label: Embedded Dictionaries
|
||||
|
||||
# Functions for Working with Embedded Dictionaries
|
||||
|
||||
:::note
|
||||
In order for the functions below to work, the server config must specify the paths and addresses for getting all the embedded dictionaries. The dictionaries are loaded at the first call of any of these functions. If the reference lists can’t be loaded, an exception is thrown.
|
||||
|
||||
For information about creating reference lists, see the section “Dictionaries”.
|
||||
As such, the examples shown in this section will throw an exception in [ClickHouse Fiddle](https://fiddle.clickhouse.com/) and in quick release and production deployments by default, unless first configured.
|
||||
:::
|
||||
|
||||
For information about creating reference lists, see the section [“Dictionaries”](../dictionaries#embedded-dictionaries).
|
||||
|
||||
## Multiple Geobases
|
||||
|
||||
ClickHouse supports working with multiple alternative geobases (regional hierarchies) simultaneously, in order to support various perspectives on which countries certain regions belong to.
|
||||
|
||||
The ‘clickhouse-server’ config specifies the file with the regional hierarchy::`<path_to_regions_hierarchy_file>/opt/geo/regions_hierarchy.txt</path_to_regions_hierarchy_file>`
|
||||
The ‘clickhouse-server’ config specifies the file with the regional hierarchy:
|
||||
|
||||
Besides this file, it also searches for files nearby that have the _ symbol and any suffix appended to the name (before the file extension).
|
||||
For example, it will also find the file `/opt/geo/regions_hierarchy_ua.txt`, if present.
|
||||
```<path_to_regions_hierarchy_file>/opt/geo/regions_hierarchy.txt</path_to_regions_hierarchy_file>```
|
||||
|
||||
`ua` is called the dictionary key. For a dictionary without a suffix, the key is an empty string.
|
||||
Besides this file, it also searches for files nearby that have the `_` symbol and any suffix appended to the name (before the file extension).
|
||||
For example, it will also find the file `/opt/geo/regions_hierarchy_ua.txt`, if present. Here `ua` is called the dictionary key. For a dictionary without a suffix, the key is an empty string.
|
||||
|
||||
All the dictionaries are re-loaded in runtime (once every certain number of seconds, as defined in the builtin_dictionaries_reload_interval config parameter, or once an hour by default). However, the list of available dictionaries is defined one time, when the server starts.
|
||||
All the dictionaries are re-loaded during runtime (once every certain number of seconds, as defined in the [`builtin_dictionaries_reload_interval`](../../operations/server-configuration-parameters/settings#builtin-dictionaries-reload-interval) config parameter, or once an hour by default). However, the list of available dictionaries is defined once, when the server starts.
|
||||
|
||||
All functions for working with regions have an optional argument at the end – the dictionary key. It is referred to as the geobase.
|
||||
|
||||
Example:
|
||||
|
||||
``` sql
|
||||
@ -32,13 +37,116 @@ regionToCountry(RegionID, '') – Uses the default dictionary: /opt/geo/regions_
|
||||
regionToCountry(RegionID, 'ua') – Uses the dictionary for the 'ua' key: /opt/geo/regions_hierarchy_ua.txt
|
||||
```
|
||||
|
||||
### regionToCity(id\[, geobase\])
|
||||
### regionToName
|
||||
|
||||
Accepts a UInt32 number – the region ID from the geobase. If this region is a city or part of a city, it returns the region ID for the appropriate city. Otherwise, returns 0.
|
||||
Accepts a region ID and geobase and returns a string of the name of the region in the corresponding language. If the region with the specified ID does not exist, an empty string is returned.
|
||||
|
||||
### regionToArea(id\[, geobase\])
|
||||
**Syntax**
|
||||
|
||||
Converts a region to an area (type 5 in the geobase). In every other way, this function is the same as ‘regionToCity’.
|
||||
``` sql
|
||||
regionToName(id\[, lang\])
|
||||
```
|
||||
**Parameters**
|
||||
|
||||
- `id` — Region ID from the geobase. [UInt32](../data-types/int-uint).
|
||||
- `geobase` — Dictionary key. See [Multiple Geobases](#multiple-geobases). [String](../data-types/string). Optional.
|
||||
|
||||
**Returned value**
|
||||
|
||||
- Name of the region in the corresponding language specified by `geobase`. [String](../data-types/string).
|
||||
- Otherwise, an empty string.
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
``` sql
|
||||
SELECT regionToName(number::UInt32,'en') FROM numbers(0,5);
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
``` text
|
||||
┌─regionToName(CAST(number, 'UInt32'), 'en')─┐
|
||||
│ │
|
||||
│ World │
|
||||
│ USA │
|
||||
│ Colorado │
|
||||
│ Boulder County │
|
||||
└────────────────────────────────────────────┘
|
||||
```
|
||||
|
||||
### regionToCity
|
||||
|
||||
Accepts a region ID from the geobase. If this region is a city or part of a city, it returns the region ID for the appropriate city. Otherwise, returns 0.
|
||||
|
||||
**Syntax**
|
||||
|
||||
```sql
|
||||
regionToCity(id [, geobase])
|
||||
```
|
||||
|
||||
**Parameters**
|
||||
|
||||
- `id` — Region ID from the geobase. [UInt32](../data-types/int-uint).
|
||||
- `geobase` — Dictionary key. See [Multiple Geobases](#multiple-geobases). [String](../data-types/string). Optional.
|
||||
|
||||
**Returned value**
|
||||
|
||||
- Region ID for the appropriate city, if it exists. [UInt32](../data-types/int-uint).
|
||||
- 0, if there is none.
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
```sql
|
||||
SELECT regionToName(number::UInt32, 'en'), regionToCity(number::UInt32) AS id, regionToName(id, 'en') FROM numbers(13);
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
```response
|
||||
┌─regionToName(CAST(number, 'UInt32'), 'en')─┬─id─┬─regionToName(regionToCity(CAST(number, 'UInt32')), 'en')─┐
|
||||
│ │ 0 │ │
|
||||
│ World │ 0 │ │
|
||||
│ USA │ 0 │ │
|
||||
│ Colorado │ 0 │ │
|
||||
│ Boulder County │ 0 │ │
|
||||
│ Boulder │ 5 │ Boulder │
|
||||
│ China │ 0 │ │
|
||||
│ Sichuan │ 0 │ │
|
||||
│ Chengdu │ 8 │ Chengdu │
|
||||
│ America │ 0 │ │
|
||||
│ North America │ 0 │ │
|
||||
│ Eurasia │ 0 │ │
|
||||
│ Asia │ 0 │ │
|
||||
└────────────────────────────────────────────┴────┴──────────────────────────────────────────────────────────┘
|
||||
```
|
||||
|
||||
### regionToArea
|
||||
|
||||
Converts a region to an area (type 5 in the geobase). In every other way, this function is the same as [‘regionToCity’](#regiontocity).
|
||||
|
||||
**Syntax**
|
||||
|
||||
```sql
|
||||
regionToArea(id [, geobase])
|
||||
```
|
||||
|
||||
**Parameters**
|
||||
|
||||
- `id` — Region ID from the geobase. [UInt32](../data-types/int-uint).
|
||||
- `geobase` — Dictionary key. See [Multiple Geobases](#multiple-geobases). [String](../data-types/string). Optional.
|
||||
|
||||
**Returned value**
|
||||
|
||||
- Region ID for the appropriate area, if it exists. [UInt32](../data-types/int-uint).
|
||||
- 0, if there is none.
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
``` sql
|
||||
SELECT DISTINCT regionToName(regionToArea(toUInt32(number), 'ua'))
|
||||
@ -46,6 +154,8 @@ FROM system.numbers
|
||||
LIMIT 15
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
``` text
|
||||
┌─regionToName(regionToArea(toUInt32(number), \'ua\'))─┐
|
||||
│ │
|
||||
@ -66,16 +176,38 @@ LIMIT 15
|
||||
└──────────────────────────────────────────────────────┘
|
||||
```
|
||||
|
||||
### regionToDistrict(id\[, geobase\])
|
||||
### regionToDistrict
|
||||
|
||||
Converts a region to a federal district (type 4 in the geobase). In every other way, this function is the same as ‘regionToCity’.
|
||||
|
||||
**Syntax**
|
||||
|
||||
```sql
|
||||
regionToDistrict(id [, geobase])
|
||||
```
|
||||
|
||||
**Parameters**
|
||||
|
||||
- `id` — Region ID from the geobase. [UInt32](../data-types/int-uint).
|
||||
- `geobase` — Dictionary key. See [Multiple Geobases](#multiple-geobases). [String](../data-types/string). Optional.
|
||||
|
||||
**Returned value**
|
||||
|
||||
- Region ID for the appropriate city, if it exists. [UInt32](../data-types/int-uint).
|
||||
- 0, if there is none.
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
``` sql
|
||||
SELECT DISTINCT regionToName(regionToDistrict(toUInt32(number), 'ua'))
|
||||
FROM system.numbers
|
||||
LIMIT 15
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
``` text
|
||||
┌─regionToName(regionToDistrict(toUInt32(number), \'ua\'))─┐
|
||||
│ │
|
||||
@ -96,17 +228,103 @@ LIMIT 15
|
||||
└──────────────────────────────────────────────────────────┘
|
||||
```
|
||||
|
||||
### regionToCountry(id\[, geobase\])
|
||||
### regionToCountry
|
||||
|
||||
Converts a region to a country. In every other way, this function is the same as ‘regionToCity’.
|
||||
Example: `regionToCountry(toUInt32(213)) = 225` converts Moscow (213) to Russia (225).
|
||||
Converts a region to a country (type 3 in the geobase). In every other way, this function is the same as ‘regionToCity’.
|
||||
|
||||
### regionToContinent(id\[, geobase\])
|
||||
**Syntax**
|
||||
|
||||
Converts a region to a continent. In every other way, this function is the same as ‘regionToCity’.
|
||||
Example: `regionToContinent(toUInt32(213)) = 10001` converts Moscow (213) to Eurasia (10001).
|
||||
```sql
|
||||
regionToCountry(id [, geobase])
|
||||
```
|
||||
|
||||
### regionToTopContinent(id\[, geobase\])
|
||||
**Parameters**
|
||||
|
||||
- `id` — Region ID from the geobase. [UInt32](../data-types/int-uint).
|
||||
- `geobase` — Dictionary key. See [Multiple Geobases](#multiple-geobases). [String](../data-types/string). Optional.
|
||||
|
||||
**Returned value**
|
||||
|
||||
- Region ID for the appropriate country, if it exists. [UInt32](../data-types/int-uint).
|
||||
- 0, if there is none.
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
``` sql
|
||||
SELECT regionToName(number::UInt32, 'en'), regionToCountry(number::UInt32) AS id, regionToName(id, 'en') FROM numbers(13);
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
``` text
|
||||
┌─regionToName(CAST(number, 'UInt32'), 'en')─┬─id─┬─regionToName(regionToCountry(CAST(number, 'UInt32')), 'en')─┐
|
||||
│ │ 0 │ │
|
||||
│ World │ 0 │ │
|
||||
│ USA │ 2 │ USA │
|
||||
│ Colorado │ 2 │ USA │
|
||||
│ Boulder County │ 2 │ USA │
|
||||
│ Boulder │ 2 │ USA │
|
||||
│ China │ 6 │ China │
|
||||
│ Sichuan │ 6 │ China │
|
||||
│ Chengdu │ 6 │ China │
|
||||
│ America │ 0 │ │
|
||||
│ North America │ 0 │ │
|
||||
│ Eurasia │ 0 │ │
|
||||
│ Asia │ 0 │ │
|
||||
└────────────────────────────────────────────┴────┴─────────────────────────────────────────────────────────────┘
|
||||
```
|
||||
|
||||
### regionToContinent
|
||||
|
||||
Converts a region to a continent (type 1 in the geobase). In every other way, this function is the same as ‘regionToCity’.
|
||||
|
||||
**Syntax**
|
||||
|
||||
```sql
|
||||
regionToContinent(id [, geobase])
|
||||
```
|
||||
|
||||
**Parameters**
|
||||
|
||||
- `id` — Region ID from the geobase. [UInt32](../data-types/int-uint).
|
||||
- `geobase` — Dictionary key. See [Multiple Geobases](#multiple-geobases). [String](../data-types/string). Optional.
|
||||
|
||||
**Returned value**
|
||||
|
||||
- Region ID for the appropriate continent, if it exists. [UInt32](../data-types/int-uint).
|
||||
- 0, if there is none.
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
``` sql
|
||||
SELECT regionToName(number::UInt32, 'en'), regionToContinent(number::UInt32) AS id, regionToName(id, 'en') FROM numbers(13);
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
``` text
|
||||
┌─regionToName(CAST(number, 'UInt32'), 'en')─┬─id─┬─regionToName(regionToContinent(CAST(number, 'UInt32')), 'en')─┐
|
||||
│ │ 0 │ │
|
||||
│ World │ 0 │ │
|
||||
│ USA │ 10 │ North America │
|
||||
│ Colorado │ 10 │ North America │
|
||||
│ Boulder County │ 10 │ North America │
|
||||
│ Boulder │ 10 │ North America │
|
||||
│ China │ 12 │ Asia │
|
||||
│ Sichuan │ 12 │ Asia │
|
||||
│ Chengdu │ 12 │ Asia │
|
||||
│ America │ 9 │ America │
|
||||
│ North America │ 10 │ North America │
|
||||
│ Eurasia │ 11 │ Eurasia │
|
||||
│ Asia │ 12 │ Asia │
|
||||
└────────────────────────────────────────────┴────┴───────────────────────────────────────────────────────────────┘
|
||||
```
|
||||
|
||||
### regionToTopContinent
|
||||
|
||||
Finds the highest continent in the hierarchy for the region.
|
||||
|
||||
@ -116,37 +334,175 @@ Finds the highest continent in the hierarchy for the region.
|
||||
regionToTopContinent(id[, geobase])
|
||||
```
|
||||
|
||||
**Arguments**
|
||||
**Parameters**
|
||||
|
||||
- `id` — Region ID from the geobase. [UInt32](../../sql-reference/data-types/int-uint.md).
|
||||
- `geobase` — Dictionary key. See [Multiple Geobases](#multiple-geobases). [String](../../sql-reference/data-types/string.md). Optional.
|
||||
- `id` — Region ID from the geobase. [UInt32](../data-types/int-uint).
|
||||
- `geobase` — Dictionary key. See [Multiple Geobases](#multiple-geobases). [String](../data-types/string). Optional.
|
||||
|
||||
**Returned value**
|
||||
|
||||
- Identifier of the top level continent (the latter when you climb the hierarchy of regions).
|
||||
- Identifier of the top level continent (the latter when you climb the hierarchy of regions).[UInt32](../data-types/int-uint).
|
||||
- 0, if there is none.
|
||||
|
||||
Type: `UInt32`.
|
||||
**Example**
|
||||
|
||||
### regionToPopulation(id\[, geobase\])
|
||||
Query:
|
||||
|
||||
Gets the population for a region.
|
||||
The population can be recorded in files with the geobase. See the section “Dictionaries”.
|
||||
If the population is not recorded for the region, it returns 0.
|
||||
In the geobase, the population might be recorded for child regions, but not for parent regions.
|
||||
``` sql
|
||||
SELECT regionToName(number::UInt32, 'en'), regionToTopContinent(number::UInt32) AS id, regionToName(id, 'en') FROM numbers(13);
|
||||
```
|
||||
|
||||
### regionIn(lhs, rhs\[, geobase\])
|
||||
Result:
|
||||
|
||||
``` text
|
||||
┌─regionToName(CAST(number, 'UInt32'), 'en')─┬─id─┬─regionToName(regionToTopContinent(CAST(number, 'UInt32')), 'en')─┐
|
||||
│ │ 0 │ │
|
||||
│ World │ 0 │ │
|
||||
│ USA │ 9 │ America │
|
||||
│ Colorado │ 9 │ America │
|
||||
│ Boulder County │ 9 │ America │
|
||||
│ Boulder │ 9 │ America │
|
||||
│ China │ 11 │ Eurasia │
|
||||
│ Sichuan │ 11 │ Eurasia │
|
||||
│ Chengdu │ 11 │ Eurasia │
|
||||
│ America │ 9 │ America │
|
||||
│ North America │ 9 │ America │
|
||||
│ Eurasia │ 11 │ Eurasia │
|
||||
│ Asia │ 11 │ Eurasia │
|
||||
└────────────────────────────────────────────┴────┴──────────────────────────────────────────────────────────────────┘
|
||||
```
|
||||
|
||||
### regionToPopulation
|
||||
|
||||
Gets the population for a region. The population can be recorded in files with the geobase. See the section [“Dictionaries”](../dictionaries#embedded-dictionaries). If the population is not recorded for the region, it returns 0. In the geobase, the population might be recorded for child regions, but not for parent regions.
|
||||
|
||||
**Syntax**
|
||||
|
||||
``` sql
|
||||
regionToPopulation(id[, geobase])
|
||||
```
|
||||
|
||||
**Parameters**
|
||||
|
||||
- `id` — Region ID from the geobase. [UInt32](../data-types/int-uint).
|
||||
- `geobase` — Dictionary key. See [Multiple Geobases](#multiple-geobases). [String](../data-types/string). Optional.
|
||||
|
||||
**Returned value**
|
||||
|
||||
- Population for the region. [UInt32](../data-types/int-uint).
|
||||
- 0, if there is none.
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
``` sql
|
||||
SELECT regionToName(number::UInt32, 'en'), regionToPopulation(number::UInt32) AS id, regionToName(id, 'en') FROM numbers(13);
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
``` text
|
||||
┌─regionToName(CAST(number, 'UInt32'), 'en')─┬─population─┐
|
||||
│ │ 0 │
|
||||
│ World │ 4294967295 │
|
||||
│ USA │ 330000000 │
|
||||
│ Colorado │ 5700000 │
|
||||
│ Boulder County │ 330000 │
|
||||
│ Boulder │ 100000 │
|
||||
│ China │ 1500000000 │
|
||||
│ Sichuan │ 83000000 │
|
||||
│ Chengdu │ 20000000 │
|
||||
│ America │ 1000000000 │
|
||||
│ North America │ 600000000 │
|
||||
│ Eurasia │ 4294967295 │
|
||||
│ Asia │ 4294967295 │
|
||||
└────────────────────────────────────────────┴────────────┘
|
||||
```
|
||||
|
||||
### regionIn
|
||||
|
||||
Checks whether a `lhs` region belongs to a `rhs` region. Returns a UInt8 number equal to 1 if it belongs, or 0 if it does not belong.
|
||||
|
||||
**Syntax**
|
||||
|
||||
``` sql
|
||||
regionIn(lhs, rhs\[, geobase\])
|
||||
```
|
||||
|
||||
**Parameters**
|
||||
|
||||
- `lhs` — Lhs region ID from the geobase. [UInt32](../../sql-reference/data-types/int-uint).
|
||||
- `rhs` — Rhs region ID from the geobase. [UInt32](../../sql-reference/data-types/int-uint).
|
||||
- `geobase` — Dictionary key. See [Multiple Geobases](#multiple-geobases). [String](../data-types/string). Optional.
|
||||
|
||||
**Returned value**
|
||||
|
||||
- 1, if it belongs. [UInt8](../../sql-reference/data-types/int-uint).
|
||||
- 0, if it doesn't belong.
|
||||
|
||||
**Implementation details**
|
||||
|
||||
Checks whether a ‘lhs’ region belongs to a ‘rhs’ region. Returns a UInt8 number equal to 1 if it belongs, or 0 if it does not belong.
|
||||
The relationship is reflexive – any region also belongs to itself.
|
||||
|
||||
### regionHierarchy(id\[, geobase\])
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
``` sql
|
||||
SELECT regionToName(n1.number::UInt32, 'en') || (regionIn(n1.number::UInt32, n2.number::UInt32) ? ' is in ' : ' is not in ') || regionToName(n2.number::UInt32, 'en') FROM numbers(1,2) AS n1 CROSS JOIN numbers(1,5) AS n2;
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
``` text
|
||||
World is in World
|
||||
World is not in USA
|
||||
World is not in Colorado
|
||||
World is not in Boulder County
|
||||
World is not in Boulder
|
||||
USA is in World
|
||||
USA is in USA
|
||||
USA is not in Colorado
|
||||
USA is not in Boulder County
|
||||
USA is not in Boulder
|
||||
```
|
||||
|
||||
### regionHierarchy
|
||||
|
||||
Accepts a UInt32 number – the region ID from the geobase. Returns an array of region IDs consisting of the passed region and all parents along the chain.
|
||||
Example: `regionHierarchy(toUInt32(213)) = [213,1,3,225,10001,10000]`.
|
||||
|
||||
### regionToName(id\[, lang\])
|
||||
**Syntax**
|
||||
|
||||
Accepts a UInt32 number – the region ID from the geobase. A string with the name of the language can be passed as a second argument. Supported languages are: ru, en, ua, uk, by, kz, tr. If the second argument is omitted, the language ‘ru’ is used. If the language is not supported, an exception is thrown. Returns a string – the name of the region in the corresponding language. If the region with the specified ID does not exist, an empty string is returned.
|
||||
``` sql
|
||||
regionHierarchy(id\[, geobase\])
|
||||
```
|
||||
|
||||
`ua` and `uk` both mean Ukrainian.
|
||||
**Parameters**
|
||||
|
||||
- `id` — Region ID from the geobase. [UInt32](../data-types/int-uint).
|
||||
- `geobase` — Dictionary key. See [Multiple Geobases](#multiple-geobases). [String](../data-types/string). Optional.
|
||||
|
||||
**Returned value**
|
||||
|
||||
- Array of region IDs consisting of the passed region and all parents along the chain. [Array](../data-types/array)([UInt32](../data-types/int-uint)).
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
``` sql
|
||||
SELECT regionHierarchy(number::UInt32) AS arr, arrayMap(id -> regionToName(id, 'en'), arr) FROM numbers(5);
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
``` text
|
||||
┌─arr────────────┬─arrayMap(lambda(tuple(id), regionToName(id, 'en')), regionHierarchy(CAST(number, 'UInt32')))─┐
|
||||
│ [] │ [] │
|
||||
│ [1] │ ['World'] │
|
||||
│ [2,10,9,1] │ ['USA','North America','America','World'] │
|
||||
│ [3,2,10,9,1] │ ['Colorado','USA','North America','America','World'] │
|
||||
│ [4,3,2,10,9,1] │ ['Boulder County','Colorado','USA','North America','America','World'] │
|
||||
└────────────────┴──────────────────────────────────────────────────────────────────────────────────────────────┘
|
||||
```
|
||||
|
||||
@ -19,25 +19,51 @@ Subquery is another `SELECT` query that may be specified in parenthesis inside `
|
||||
|
||||
## FINAL Modifier
|
||||
|
||||
When `FINAL` is specified, ClickHouse fully merges the data before returning the result and thus performs all data transformations that happen during merges for the given table engine.
|
||||
When `FINAL` is specified, ClickHouse fully merges the data before returning the result. This also performs all data transformations that happen during merges for the given table engine.
|
||||
|
||||
It is applicable when selecting data from ReplacingMergeTree, SummingMergeTree, AggregatingMergeTree, CollapsingMergeTree and VersionedCollapsingMergeTree tables.
|
||||
It is applicable when selecting data from from tables using the following table engines:
|
||||
- `ReplacingMergeTree`
|
||||
- `SummingMergeTree`
|
||||
- `AggregatingMergeTree`
|
||||
- `CollapsingMergeTree`
|
||||
- `VersionedCollapsingMergeTree`
|
||||
|
||||
`SELECT` queries with `FINAL` are executed in parallel. The [max_final_threads](../../../operations/settings/settings.md#max-final-threads) setting limits the number of threads used.
|
||||
|
||||
There are drawbacks to using `FINAL` (see below).
|
||||
|
||||
### Drawbacks
|
||||
|
||||
Queries that use `FINAL` are executed slightly slower than similar queries that do not, because:
|
||||
Queries that use `FINAL` execute slightly slower than similar queries that do not use `FINAL` because:
|
||||
|
||||
- Data is merged during query execution.
|
||||
- Queries with `FINAL` read primary key columns in addition to the columns specified in the query.
|
||||
- Queries with `FINAL` may read primary key columns in addition to the columns specified in the query.
|
||||
|
||||
`FINAL` requires additional compute and memory resources, as the processing that normally would occur at merge time must occur in memory at the time of the query. However, using FINAL is sometimes necessary in order to produce accurate results, and is less expensive than running `OPTIMIZE` to force a merge. It is also sometimes possible to use different queries that assume the background processes of the `MergeTree` engine haven’t happened yet and deal with it by applying aggregation (for example, to discard duplicates). If you need to use FINAL in your queries in order to get the required results, then it is okay to do so but be aware of the additional processing required.
|
||||
`FINAL` requires additional compute and memory resources because the processing that normally would occur at merge time must occur in memory at the time of the query. However, using FINAL is sometimes necessary in order to produce accurate results (as data may not yet be fully merged). It is less expensive than running `OPTIMIZE` to force a merge.
|
||||
|
||||
As an alternative to using `FINAL`, it is sometimes possible to use different queries that assume the background processes of the `MergeTree` engine have not yet occurred and deal with it by applying an aggregation (for example, to discard duplicates). If you need to use `FINAL` in your queries in order to get the required results, it is okay to do so but be aware of the additional processing required.
|
||||
|
||||
`FINAL` can be applied automatically using [FINAL](../../../operations/settings/settings.md#final) setting to all tables in a query using a session or a user profile.
|
||||
|
||||
### Example Usage
|
||||
|
||||
**Using the `FINAL` keyword**
|
||||
|
||||
```sql
|
||||
SELECT x, y FROM mytable FINAL WHERE x > 1;
|
||||
```
|
||||
|
||||
**Using `FINAL` as a query-level setting**
|
||||
|
||||
```sql
|
||||
SELECT x, y FROM mytable WHERE x > 1 SETTINGS final = 1;
|
||||
```
|
||||
|
||||
**Using `FINAL` as a session-level setting**
|
||||
|
||||
```sql
|
||||
SET final = 1;
|
||||
SELECT x, y FROM mytable WHERE x > 1;
|
||||
```
|
||||
|
||||
## Implementation Details
|
||||
|
||||
If the `FROM` clause is omitted, data will be read from the `system.one` table.
|
||||
|
||||
@ -3447,17 +3447,6 @@ SELECT
|
||||
FROM fuse_tbl
|
||||
```
|
||||
|
||||
## allow_experimental_database_replicated {#allow_experimental_database_replicated}
|
||||
|
||||
Позволяет создавать базы данных с движком [Replicated](../../engines/database-engines/replicated.md).
|
||||
|
||||
Возможные значения:
|
||||
|
||||
- 0 — Disabled.
|
||||
- 1 — Enabled.
|
||||
|
||||
Значение по умолчанию: `0`.
|
||||
|
||||
## database_replicated_initial_query_timeout_sec {#database_replicated_initial_query_timeout_sec}
|
||||
|
||||
Устанавливает, как долго начальный DDL-запрос должен ждать, пока реплицированная база данных прецессирует предыдущие записи очереди DDL в секундах.
|
||||
|
||||
@ -5,7 +5,7 @@ sidebar_position: 106
|
||||
|
||||
# argMax {#agg-function-argmax}
|
||||
|
||||
Вычисляет значение `arg` при максимальном значении `val`.
|
||||
Вычисляет значение `arg` при максимальном значении `val`. Если несколько строк имеют одинаковое `val`, в которых равное значение является максимальным, то возвращаемое `arg` не является детерминированным. Обе части, arg и max, ведут себя как агрегатные функции, они обе пропускают Null во время обработки и возвращают не Null значения, если не Null значения доступны.
|
||||
|
||||
**Синтаксис**
|
||||
|
||||
@ -49,3 +49,60 @@ SELECT argMax(user, salary), argMax(tuple(user, salary), salary) FROM salary;
|
||||
│ director │ ('director',5000) │
|
||||
└──────────────────────┴─────────────────────────────────────┘
|
||||
```
|
||||
|
||||
**Дополнительный пример**
|
||||
|
||||
```sql
|
||||
CREATE TABLE test
|
||||
(
|
||||
a Nullable(String),
|
||||
b Nullable(Int64)
|
||||
)
|
||||
ENGINE = Memory AS
|
||||
SELECT *
|
||||
FROM VALUES(('a', 1), ('b', 2), ('c', 2), (NULL, 3), (NULL, NULL), ('d', NULL));
|
||||
|
||||
select * from test;
|
||||
┌─a────┬────b─┐
|
||||
│ a │ 1 │
|
||||
│ b │ 2 │
|
||||
│ c │ 2 │
|
||||
│ ᴺᵁᴸᴸ │ 3 │
|
||||
│ ᴺᵁᴸᴸ │ ᴺᵁᴸᴸ │
|
||||
│ d │ ᴺᵁᴸᴸ │
|
||||
└──────┴──────┘
|
||||
|
||||
SELECT argMax(a, b), max(b) FROM test;
|
||||
┌─argMax(a, b)─┬─max(b)─┐
|
||||
│ b │ 3 │ -- argMax = 'b' потому что это первое not Null значение, max(b) из другой строки!
|
||||
└──────────────┴────────┘
|
||||
|
||||
SELECT argMax(tuple(a), b) FROM test;
|
||||
┌─argMax(tuple(a), b)─┐
|
||||
│ (NULL) │ -- Кортеж `Tuple`, который содержит только `NULL` значения является не `NULL` кортежем, поэтому агрегатыне функции не будут пропускать эту строку с `NULL` значениями.
|
||||
└─────────────────────┘
|
||||
|
||||
SELECT (argMax((a, b), b) as t).1 argMaxA, t.2 argMaxB FROM test;
|
||||
┌─argMaxA─┬─argMaxB─┐
|
||||
│ ᴺᵁᴸᴸ │ 3 │ -- Вы можете использовать кортеж Tuple и получить оба значения для соответсвующего max(b).
|
||||
└─────────┴─────────┘
|
||||
|
||||
SELECT argMax(a, b), max(b) FROM test WHERE a IS NULL AND b IS NULL;
|
||||
┌─argMax(a, b)─┬─max(b)─┐
|
||||
│ ᴺᵁᴸᴸ │ ᴺᵁᴸᴸ │ -- Все агрегированные строки содержат хотя бы одно `NULL` значение, поэтому все строки пропускаются и результатом будет `NULL`.
|
||||
└──────────────┴────────┘
|
||||
|
||||
SELECT argMax(a, (b,a)) FROM test;
|
||||
┌─argMax(a, tuple(b, a))─┐
|
||||
│ c │ -- Есть две строки с b=2, кортеж `Tuple` в функции `Max` позволяет получить не первый `arg`.
|
||||
└────────────────────────┘
|
||||
|
||||
SELECT argMax(a, tuple(b)) FROM test;
|
||||
┌─argMax(a, tuple(b))─┐
|
||||
│ b │ -- Кортеж `Tuple` может использоваться в `Max`, чтобы не пропускать `NULL` значения в `Max`.
|
||||
└─────────────────────┘
|
||||
```
|
||||
|
||||
**Смотрите также**
|
||||
|
||||
- [Tuple](/docs/ru/sql-reference/data-types/tuple.md)
|
||||
|
||||
@ -918,11 +918,13 @@ bool Client::processWithFuzzing(const String & full_query)
|
||||
}
|
||||
|
||||
|
||||
void Client::printHelpMessage(const OptionsDescription & options_description)
|
||||
void Client::printHelpMessage(const OptionsDescription & options_description, bool verbose)
|
||||
{
|
||||
std::cout << options_description.main_description.value() << "\n";
|
||||
std::cout << options_description.external_description.value() << "\n";
|
||||
std::cout << options_description.hosts_and_ports_description.value() << "\n";
|
||||
if (verbose)
|
||||
std::cout << "All settings are documented at https://clickhouse.com/docs/en/operations/settings/settings.\n\n";
|
||||
std::cout << "In addition, --param_name=value can be specified for substitution of parameters for parametrized queries.\n";
|
||||
std::cout << "\nSee also: https://clickhouse.com/docs/en/integrations/sql-clients/cli\n";
|
||||
}
|
||||
|
||||
@ -25,7 +25,7 @@ protected:
|
||||
|
||||
String getName() const override { return "client"; }
|
||||
|
||||
void printHelpMessage(const OptionsDescription & options_description) override;
|
||||
void printHelpMessage(const OptionsDescription & options_description, bool verbose) override;
|
||||
|
||||
void addOptions(OptionsDescription & options_description) override;
|
||||
|
||||
|
||||
@ -774,10 +774,12 @@ void LocalServer::processConfig()
|
||||
}
|
||||
|
||||
|
||||
void LocalServer::printHelpMessage([[maybe_unused]] const OptionsDescription & options_description)
|
||||
void LocalServer::printHelpMessage(const OptionsDescription & options_description, bool verbose)
|
||||
{
|
||||
std::cout << getHelpHeader() << "\n";
|
||||
std::cout << options_description.main_description.value() << "\n";
|
||||
if (verbose)
|
||||
std::cout << "All settings are documented at https://clickhouse.com/docs/en/operations/settings/settings.\n\n";
|
||||
std::cout << getHelpFooter() << "\n";
|
||||
std::cout << "In addition, --param_name=value can be specified for substitution of parameters for parametrized queries.\n";
|
||||
std::cout << "\nSee also: https://clickhouse.com/docs/en/operations/utilities/clickhouse-local/\n";
|
||||
|
||||
@ -36,7 +36,7 @@ protected:
|
||||
|
||||
String getName() const override { return "local"; }
|
||||
|
||||
void printHelpMessage(const OptionsDescription & options_description) override;
|
||||
void printHelpMessage(const OptionsDescription & options_description, bool verbose) override;
|
||||
|
||||
void addOptions(OptionsDescription & options_description) override;
|
||||
|
||||
|
||||
@ -280,11 +280,11 @@ void AccessControl::setUpFromMainConfig(const Poco::Util::AbstractConfiguration
|
||||
|
||||
/// Optional improvements in access control system.
|
||||
/// The default values are false because we need to be compatible with earlier access configurations
|
||||
setEnabledUsersWithoutRowPoliciesCanReadRows(config_.getBool("access_control_improvements.users_without_row_policies_can_read_rows", false));
|
||||
setOnClusterQueriesRequireClusterGrant(config_.getBool("access_control_improvements.on_cluster_queries_require_cluster_grant", false));
|
||||
setSelectFromSystemDatabaseRequiresGrant(config_.getBool("access_control_improvements.select_from_system_db_requires_grant", false));
|
||||
setSelectFromInformationSchemaRequiresGrant(config_.getBool("access_control_improvements.select_from_information_schema_requires_grant", false));
|
||||
setSettingsConstraintsReplacePrevious(config_.getBool("access_control_improvements.settings_constraints_replace_previous", false));
|
||||
setEnabledUsersWithoutRowPoliciesCanReadRows(config_.getBool("access_control_improvements.users_without_row_policies_can_read_rows", true));
|
||||
setOnClusterQueriesRequireClusterGrant(config_.getBool("access_control_improvements.on_cluster_queries_require_cluster_grant", true));
|
||||
setSelectFromSystemDatabaseRequiresGrant(config_.getBool("access_control_improvements.select_from_system_db_requires_grant", true));
|
||||
setSelectFromInformationSchemaRequiresGrant(config_.getBool("access_control_improvements.select_from_information_schema_requires_grant", true));
|
||||
setSettingsConstraintsReplacePrevious(config_.getBool("access_control_improvements.settings_constraints_replace_previous", true));
|
||||
setTableEnginesRequireGrant(config_.getBool("access_control_improvements.table_engines_require_grant", false));
|
||||
|
||||
addStoragesFromMainConfig(config_, config_path_, get_zookeeper_function_);
|
||||
|
||||
@ -1,9 +1,10 @@
|
||||
#include <Analyzer/Passes/QueryAnalysisPass.h>
|
||||
|
||||
#include <boost/algorithm/string.hpp>
|
||||
|
||||
#include <Common/checkStackSize.h>
|
||||
#include <Common/NamePrompter.h>
|
||||
#include <Common/ProfileEvents.h>
|
||||
#include <Analyzer/FunctionSecretArgumentsFinderTreeNode.h>
|
||||
|
||||
#include <IO/WriteBuffer.h>
|
||||
#include <IO/WriteHelpers.h>
|
||||
@ -81,8 +82,8 @@
|
||||
#include <Analyzer/QueryTreeBuilder.h>
|
||||
#include <Analyzer/IQueryTreeNode.h>
|
||||
#include <Analyzer/Identifier.h>
|
||||
|
||||
#include <boost/algorithm/string.hpp>
|
||||
#include <Analyzer/FunctionSecretArgumentsFinderTreeNode.h>
|
||||
#include <Analyzer/RecursiveCTE.h>
|
||||
|
||||
namespace ProfileEvents
|
||||
{
|
||||
@ -740,7 +741,7 @@ struct IdentifierResolveScope
|
||||
/// Identifier lookup to result
|
||||
std::unordered_map<IdentifierLookup, IdentifierResolveState, IdentifierLookupHash> identifier_lookup_to_resolve_state;
|
||||
|
||||
/// Lambda argument can be expression like constant, column, or it can be function
|
||||
/// Argument can be expression like constant, column, function or table expression
|
||||
std::unordered_map<std::string, QueryTreeNodePtr> expression_argument_name_to_node;
|
||||
|
||||
/// Alias name to query expression node
|
||||
@ -1464,7 +1465,8 @@ private:
|
||||
/// Lambdas that are currently in resolve process
|
||||
std::unordered_set<IQueryTreeNode *> lambdas_in_resolve_process;
|
||||
|
||||
std::unordered_set<std::string_view> cte_in_resolve_process;
|
||||
/// CTEs that are currently in resolve process
|
||||
std::unordered_set<std::string_view> ctes_in_resolve_process;
|
||||
|
||||
/// Function name to user defined lambda map
|
||||
std::unordered_map<std::string, QueryTreeNodePtr> function_name_to_user_defined_lambda;
|
||||
@ -2148,9 +2150,9 @@ void QueryAnalyzer::evaluateScalarSubqueryIfNeeded(QueryTreeNodePtr & node, Iden
|
||||
else
|
||||
{
|
||||
/** Make unique column names for tuple.
|
||||
*
|
||||
* Example: SELECT (SELECT 2 AS x, x)
|
||||
*/
|
||||
*
|
||||
* Example: SELECT (SELECT 2 AS x, x)
|
||||
*/
|
||||
makeUniqueColumnNamesInBlock(block);
|
||||
|
||||
scalar_block.insert({
|
||||
@ -3981,6 +3983,9 @@ IdentifierResolveResult QueryAnalyzer::tryResolveIdentifierInParentScopes(const
|
||||
auto * union_node = resolved_identifier->as<UnionNode>();
|
||||
|
||||
bool is_cte = (subquery_node && subquery_node->isCTE()) || (union_node && union_node->isCTE());
|
||||
bool is_table_from_expression_arguments = lookup_result.resolve_place == IdentifierResolvePlace::EXPRESSION_ARGUMENTS &&
|
||||
resolved_identifier->getNodeType() == QueryTreeNodeType::TABLE;
|
||||
bool is_valid_table_expression = is_cte || is_table_from_expression_arguments;
|
||||
|
||||
/** From parent scopes we can resolve table identifiers only as CTE.
|
||||
* Example: SELECT (SELECT 1 FROM a) FROM test_table AS a;
|
||||
@ -3988,14 +3993,10 @@ IdentifierResolveResult QueryAnalyzer::tryResolveIdentifierInParentScopes(const
|
||||
* During child scope table identifier resolve a, table node test_table with alias a from parent scope
|
||||
* is invalid.
|
||||
*/
|
||||
if (identifier_lookup.isTableExpressionLookup() && !is_cte)
|
||||
if (identifier_lookup.isTableExpressionLookup() && !is_valid_table_expression)
|
||||
continue;
|
||||
|
||||
if (is_cte)
|
||||
{
|
||||
return lookup_result;
|
||||
}
|
||||
else if (resolved_identifier->as<ConstantNode>())
|
||||
if (is_valid_table_expression || resolved_identifier->as<ConstantNode>())
|
||||
{
|
||||
return lookup_result;
|
||||
}
|
||||
@ -4071,13 +4072,9 @@ IdentifierResolveResult QueryAnalyzer::tryResolveIdentifier(const IdentifierLook
|
||||
|
||||
if (it->second.resolve_result.isResolved() &&
|
||||
scope.use_identifier_lookup_to_result_cache &&
|
||||
!scope.non_cached_identifier_lookups_during_expression_resolve.contains(identifier_lookup))
|
||||
{
|
||||
if (!it->second.resolve_result.isResolvedFromCTEs() || !cte_in_resolve_process.contains(identifier_lookup.identifier.getFullName()))
|
||||
{
|
||||
return it->second.resolve_result;
|
||||
}
|
||||
}
|
||||
!scope.non_cached_identifier_lookups_during_expression_resolve.contains(identifier_lookup) &&
|
||||
(!it->second.resolve_result.isResolvedFromCTEs() || !ctes_in_resolve_process.contains(identifier_lookup.identifier.getFullName())))
|
||||
return it->second.resolve_result;
|
||||
}
|
||||
else
|
||||
{
|
||||
@ -4150,7 +4147,7 @@ IdentifierResolveResult QueryAnalyzer::tryResolveIdentifier(const IdentifierLook
|
||||
/// To accomplish this behaviour it's not allowed to resolve identifiers to
|
||||
/// CTE that is being resolved.
|
||||
if (cte_query_node_it != scope.cte_name_to_query_node.end()
|
||||
&& !cte_in_resolve_process.contains(full_name))
|
||||
&& !ctes_in_resolve_process.contains(full_name))
|
||||
{
|
||||
resolve_result.resolved_identifier = cte_query_node_it->second;
|
||||
resolve_result.resolve_place = IdentifierResolvePlace::CTE;
|
||||
@ -6296,14 +6293,14 @@ ProjectionNames QueryAnalyzer::resolveExpressionNode(QueryTreeNodePtr & node, Id
|
||||
///
|
||||
/// In this example argument of function `in` is being resolve here. If CTE `test1` is not forbidden,
|
||||
/// `test1` is resolved to CTE (not to the table) in `initializeQueryJoinTreeNode` function.
|
||||
cte_in_resolve_process.insert(cte_name);
|
||||
ctes_in_resolve_process.insert(cte_name);
|
||||
|
||||
if (subquery_node)
|
||||
resolveQuery(resolved_identifier_node, subquery_scope);
|
||||
else
|
||||
resolveUnion(resolved_identifier_node, subquery_scope);
|
||||
|
||||
cte_in_resolve_process.erase(cte_name);
|
||||
ctes_in_resolve_process.erase(cte_name);
|
||||
}
|
||||
}
|
||||
}
|
||||
@ -7874,7 +7871,7 @@ void QueryAnalyzer::resolveQuery(const QueryTreeNodePtr & query_node, Identifier
|
||||
auto & query_node_typed = query_node->as<QueryNode &>();
|
||||
|
||||
if (query_node_typed.isCTE())
|
||||
cte_in_resolve_process.insert(query_node_typed.getCTEName());
|
||||
ctes_in_resolve_process.insert(query_node_typed.getCTEName());
|
||||
|
||||
bool is_rollup_or_cube = query_node_typed.isGroupByWithRollup() || query_node_typed.isGroupByWithCube();
|
||||
|
||||
@ -7956,7 +7953,6 @@ void QueryAnalyzer::resolveQuery(const QueryTreeNodePtr & query_node, Identifier
|
||||
auto * union_node = node->as<UnionNode>();
|
||||
|
||||
bool subquery_is_cte = (subquery_node && subquery_node->isCTE()) || (union_node && union_node->isCTE());
|
||||
|
||||
if (!subquery_is_cte)
|
||||
continue;
|
||||
|
||||
@ -8213,7 +8209,7 @@ void QueryAnalyzer::resolveQuery(const QueryTreeNodePtr & query_node, Identifier
|
||||
query_node_typed.resolveProjectionColumns(std::move(projection_columns));
|
||||
|
||||
if (query_node_typed.isCTE())
|
||||
cte_in_resolve_process.erase(query_node_typed.getCTEName());
|
||||
ctes_in_resolve_process.erase(query_node_typed.getCTEName());
|
||||
}
|
||||
|
||||
void QueryAnalyzer::resolveUnion(const QueryTreeNodePtr & union_node, IdentifierResolveScope & scope)
|
||||
@ -8221,13 +8217,56 @@ void QueryAnalyzer::resolveUnion(const QueryTreeNodePtr & union_node, Identifier
|
||||
auto & union_node_typed = union_node->as<UnionNode &>();
|
||||
|
||||
if (union_node_typed.isCTE())
|
||||
cte_in_resolve_process.insert(union_node_typed.getCTEName());
|
||||
ctes_in_resolve_process.insert(union_node_typed.getCTEName());
|
||||
|
||||
auto & queries_nodes = union_node_typed.getQueries().getNodes();
|
||||
|
||||
for (auto & query_node : queries_nodes)
|
||||
std::optional<RecursiveCTETable> recursive_cte_table;
|
||||
TableNodePtr recursive_cte_table_node;
|
||||
|
||||
if (union_node_typed.isCTE() && union_node_typed.isRecursiveCTE())
|
||||
{
|
||||
auto & non_recursive_query = queries_nodes[0];
|
||||
bool non_recursive_query_is_query_node = non_recursive_query->getNodeType() == QueryTreeNodeType::QUERY;
|
||||
auto & non_recursive_query_mutable_context = non_recursive_query_is_query_node ? non_recursive_query->as<QueryNode &>().getMutableContext()
|
||||
: non_recursive_query->as<UnionNode &>().getMutableContext();
|
||||
|
||||
IdentifierResolveScope non_recursive_subquery_scope(non_recursive_query, &scope /*parent_scope*/);
|
||||
non_recursive_subquery_scope.subquery_depth = scope.subquery_depth + 1;
|
||||
|
||||
if (non_recursive_query_is_query_node)
|
||||
resolveQuery(non_recursive_query, non_recursive_subquery_scope);
|
||||
else
|
||||
resolveUnion(non_recursive_query, non_recursive_subquery_scope);
|
||||
|
||||
auto temporary_table_columns = non_recursive_query_is_query_node
|
||||
? non_recursive_query->as<QueryNode &>().getProjectionColumns()
|
||||
: non_recursive_query->as<UnionNode &>().computeProjectionColumns();
|
||||
|
||||
auto temporary_table_holder = std::make_shared<TemporaryTableHolder>(
|
||||
non_recursive_query_mutable_context,
|
||||
ColumnsDescription{NamesAndTypesList{temporary_table_columns.begin(), temporary_table_columns.end()}},
|
||||
ConstraintsDescription{},
|
||||
nullptr /*query*/,
|
||||
true /*create_for_global_subquery*/);
|
||||
auto temporary_table_storage = temporary_table_holder->getTable();
|
||||
|
||||
recursive_cte_table_node = std::make_shared<TableNode>(temporary_table_storage, non_recursive_query_mutable_context);
|
||||
recursive_cte_table_node->setTemporaryTableName(union_node_typed.getCTEName());
|
||||
|
||||
recursive_cte_table.emplace(std::move(temporary_table_holder), std::move(temporary_table_storage), std::move(temporary_table_columns));
|
||||
}
|
||||
|
||||
size_t queries_nodes_size = queries_nodes.size();
|
||||
for (size_t i = recursive_cte_table.has_value(); i < queries_nodes_size; ++i)
|
||||
{
|
||||
auto & query_node = queries_nodes[i];
|
||||
|
||||
IdentifierResolveScope subquery_scope(query_node, &scope /*parent_scope*/);
|
||||
|
||||
if (recursive_cte_table_node)
|
||||
subquery_scope.expression_argument_name_to_node[union_node_typed.getCTEName()] = recursive_cte_table_node;
|
||||
|
||||
auto query_node_type = query_node->getNodeType();
|
||||
|
||||
if (query_node_type == QueryTreeNodeType::QUERY)
|
||||
@ -8247,8 +8286,19 @@ void QueryAnalyzer::resolveUnion(const QueryTreeNodePtr & union_node, Identifier
|
||||
}
|
||||
}
|
||||
|
||||
if (recursive_cte_table && isStorageUsedInTree(recursive_cte_table->storage, union_node.get()))
|
||||
{
|
||||
if (union_node_typed.getUnionMode() != SelectUnionMode::UNION_ALL)
|
||||
throw Exception(ErrorCodes::UNSUPPORTED_METHOD,
|
||||
"Recursive CTE subquery {} with {} union mode is unsupported, only UNION ALL union mode is supported",
|
||||
union_node_typed.formatASTForErrorMessage(),
|
||||
toString(union_node_typed.getUnionMode()));
|
||||
|
||||
union_node_typed.setRecursiveCTETable(std::move(*recursive_cte_table));
|
||||
}
|
||||
|
||||
if (union_node_typed.isCTE())
|
||||
cte_in_resolve_process.erase(union_node_typed.getCTEName());
|
||||
ctes_in_resolve_process.erase(union_node_typed.getCTEName());
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
@ -14,12 +14,14 @@
|
||||
|
||||
#include <Parsers/ASTExpressionList.h>
|
||||
#include <Parsers/ASTTablesInSelectQuery.h>
|
||||
#include <Parsers/ASTWithElement.h>
|
||||
#include <Parsers/ASTSubquery.h>
|
||||
#include <Parsers/ASTSelectQuery.h>
|
||||
#include <Parsers/ASTSelectWithUnionQuery.h>
|
||||
#include <Parsers/ASTSetQuery.h>
|
||||
|
||||
#include <Analyzer/Utils.h>
|
||||
#include <Analyzer/UnionNode.h>
|
||||
|
||||
namespace DB
|
||||
{
|
||||
@ -107,6 +109,9 @@ void QueryNode::dumpTreeImpl(WriteBuffer & buffer, FormatState & format_state, s
|
||||
if (is_cte)
|
||||
buffer << ", is_cte: " << is_cte;
|
||||
|
||||
if (is_recursive_with)
|
||||
buffer << ", is_recursive_with: " << is_recursive_with;
|
||||
|
||||
if (is_distinct)
|
||||
buffer << ", is_distinct: " << is_distinct;
|
||||
|
||||
@ -259,6 +264,7 @@ bool QueryNode::isEqualImpl(const IQueryTreeNode & rhs, CompareOptions) const
|
||||
|
||||
return is_subquery == rhs_typed.is_subquery &&
|
||||
is_cte == rhs_typed.is_cte &&
|
||||
is_recursive_with == rhs_typed.is_recursive_with &&
|
||||
is_distinct == rhs_typed.is_distinct &&
|
||||
is_limit_with_ties == rhs_typed.is_limit_with_ties &&
|
||||
is_group_by_with_totals == rhs_typed.is_group_by_with_totals &&
|
||||
@ -291,6 +297,7 @@ void QueryNode::updateTreeHashImpl(HashState & state, CompareOptions) const
|
||||
state.update(projection_column_type_name);
|
||||
}
|
||||
|
||||
state.update(is_recursive_with);
|
||||
state.update(is_distinct);
|
||||
state.update(is_limit_with_ties);
|
||||
state.update(is_group_by_with_totals);
|
||||
@ -317,19 +324,20 @@ QueryTreeNodePtr QueryNode::cloneImpl() const
|
||||
{
|
||||
auto result_query_node = std::make_shared<QueryNode>(context);
|
||||
|
||||
result_query_node->is_subquery = is_subquery;
|
||||
result_query_node->is_cte = is_cte;
|
||||
result_query_node->is_distinct = is_distinct;
|
||||
result_query_node->is_limit_with_ties = is_limit_with_ties;
|
||||
result_query_node->is_group_by_with_totals = is_group_by_with_totals;
|
||||
result_query_node->is_group_by_with_rollup = is_group_by_with_rollup;
|
||||
result_query_node->is_group_by_with_cube = is_group_by_with_cube;
|
||||
result_query_node->is_subquery = is_subquery;
|
||||
result_query_node->is_cte = is_cte;
|
||||
result_query_node->is_recursive_with = is_recursive_with;
|
||||
result_query_node->is_distinct = is_distinct;
|
||||
result_query_node->is_limit_with_ties = is_limit_with_ties;
|
||||
result_query_node->is_group_by_with_totals = is_group_by_with_totals;
|
||||
result_query_node->is_group_by_with_rollup = is_group_by_with_rollup;
|
||||
result_query_node->is_group_by_with_cube = is_group_by_with_cube;
|
||||
result_query_node->is_group_by_with_grouping_sets = is_group_by_with_grouping_sets;
|
||||
result_query_node->is_group_by_all = is_group_by_all;
|
||||
result_query_node->is_order_by_all = is_order_by_all;
|
||||
result_query_node->cte_name = cte_name;
|
||||
result_query_node->projection_columns = projection_columns;
|
||||
result_query_node->settings_changes = settings_changes;
|
||||
result_query_node->is_group_by_all = is_group_by_all;
|
||||
result_query_node->is_order_by_all = is_order_by_all;
|
||||
result_query_node->cte_name = cte_name;
|
||||
result_query_node->projection_columns = projection_columns;
|
||||
result_query_node->settings_changes = settings_changes;
|
||||
|
||||
return result_query_node;
|
||||
}
|
||||
@ -337,6 +345,7 @@ QueryTreeNodePtr QueryNode::cloneImpl() const
|
||||
ASTPtr QueryNode::toASTImpl(const ConvertToASTOptions & options) const
|
||||
{
|
||||
auto select_query = std::make_shared<ASTSelectQuery>();
|
||||
select_query->recursive_with = is_recursive_with;
|
||||
select_query->distinct = is_distinct;
|
||||
select_query->limit_with_ties = is_limit_with_ties;
|
||||
select_query->group_by_with_totals = is_group_by_with_totals;
|
||||
@ -347,7 +356,41 @@ ASTPtr QueryNode::toASTImpl(const ConvertToASTOptions & options) const
|
||||
select_query->order_by_all = is_order_by_all;
|
||||
|
||||
if (hasWith())
|
||||
select_query->setExpression(ASTSelectQuery::Expression::WITH, getWith().toAST(options));
|
||||
{
|
||||
const auto & with = getWith();
|
||||
auto expression_list_ast = std::make_shared<ASTExpressionList>();
|
||||
expression_list_ast->children.reserve(with.getNodes().size());
|
||||
|
||||
for (const auto & with_node : with)
|
||||
{
|
||||
auto with_node_ast = with_node->toAST(options);
|
||||
expression_list_ast->children.push_back(with_node_ast);
|
||||
|
||||
const auto * with_query_node = with_node->as<QueryNode>();
|
||||
const auto * with_union_node = with_node->as<UnionNode>();
|
||||
if (!with_query_node && !with_union_node)
|
||||
continue;
|
||||
|
||||
bool is_with_node_cte = with_query_node ? with_query_node->isCTE() : with_union_node->isCTE();
|
||||
if (!is_with_node_cte)
|
||||
continue;
|
||||
|
||||
const auto & with_node_cte_name = with_query_node ? with_query_node->cte_name : with_union_node->getCTEName();
|
||||
|
||||
auto * with_node_ast_subquery = with_node_ast->as<ASTSubquery>();
|
||||
if (with_node_ast_subquery)
|
||||
with_node_ast_subquery->cte_name = "";
|
||||
|
||||
auto with_element_ast = std::make_shared<ASTWithElement>();
|
||||
with_element_ast->name = with_node_cte_name;
|
||||
with_element_ast->subquery = std::move(with_node_ast);
|
||||
with_element_ast->children.push_back(with_element_ast->subquery);
|
||||
|
||||
expression_list_ast->children.back() = std::move(with_element_ast);
|
||||
}
|
||||
|
||||
select_query->setExpression(ASTSelectQuery::Expression::WITH, std::move(expression_list_ast));
|
||||
}
|
||||
|
||||
auto projection_ast = getProjection().toAST(options);
|
||||
auto & projection_expression_list_ast = projection_ast->as<ASTExpressionList &>();
|
||||
|
||||
@ -140,6 +140,18 @@ public:
|
||||
cte_name = std::move(cte_name_value);
|
||||
}
|
||||
|
||||
/// Returns true if query node has RECURSIVE WITH, false otherwise
|
||||
bool isRecursiveWith() const
|
||||
{
|
||||
return is_recursive_with;
|
||||
}
|
||||
|
||||
/// Set query node RECURSIVE WITH value
|
||||
void setIsRecursiveWith(bool is_recursive_with_value)
|
||||
{
|
||||
is_recursive_with = is_recursive_with_value;
|
||||
}
|
||||
|
||||
/// Returns true if query node has DISTINCT, false otherwise
|
||||
bool isDistinct() const
|
||||
{
|
||||
@ -618,6 +630,7 @@ protected:
|
||||
private:
|
||||
bool is_subquery = false;
|
||||
bool is_cte = false;
|
||||
bool is_recursive_with = false;
|
||||
bool is_distinct = false;
|
||||
bool is_limit_with_ties = false;
|
||||
bool is_group_by_with_totals = false;
|
||||
|
||||
@ -271,6 +271,7 @@ QueryTreeNodePtr QueryTreeBuilder::buildSelectExpression(const ASTPtr & select_q
|
||||
current_query_tree->setIsSubquery(is_subquery);
|
||||
current_query_tree->setIsCTE(!cte_name.empty());
|
||||
current_query_tree->setCTEName(cte_name);
|
||||
current_query_tree->setIsRecursiveWith(select_query_typed.recursive_with);
|
||||
current_query_tree->setIsDistinct(select_query_typed.distinct);
|
||||
current_query_tree->setIsLimitWithTies(select_query_typed.limit_with_ties);
|
||||
current_query_tree->setIsGroupByWithTotals(select_query_typed.group_by_with_totals);
|
||||
@ -287,8 +288,22 @@ QueryTreeNodePtr QueryTreeBuilder::buildSelectExpression(const ASTPtr & select_q
|
||||
|
||||
auto select_with_list = select_query_typed.with();
|
||||
if (select_with_list)
|
||||
{
|
||||
current_query_tree->getWithNode() = buildExpressionList(select_with_list, current_context);
|
||||
|
||||
if (select_query_typed.recursive_with)
|
||||
{
|
||||
for (auto & with_node : current_query_tree->getWith().getNodes())
|
||||
{
|
||||
auto * with_union_node = with_node->as<UnionNode>();
|
||||
if (!with_union_node)
|
||||
continue;
|
||||
|
||||
with_union_node->setIsRecursiveCTE(true);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
auto select_expression_list = select_query_typed.select();
|
||||
if (select_expression_list)
|
||||
current_query_tree->getProjectionNode() = buildExpressionList(select_expression_list, current_context);
|
||||
|
||||
@ -165,7 +165,6 @@ private:
|
||||
|
||||
/** ClickHouse query tree pass manager.
|
||||
*
|
||||
* TODO: Support setting optimize_monotonous_functions_in_order_by.
|
||||
* TODO: Add optimizations based on function semantics. Example: SELECT * FROM test_table WHERE id != id. (id is not nullable column).
|
||||
*/
|
||||
|
||||
|
||||
21
src/Analyzer/RecursiveCTE.cpp
Normal file
21
src/Analyzer/RecursiveCTE.cpp
Normal file
@ -0,0 +1,21 @@
|
||||
#include <Analyzer/RecursiveCTE.h>
|
||||
|
||||
#include <Storages/IStorage.h>
|
||||
|
||||
namespace DB
|
||||
{
|
||||
|
||||
RecursiveCTETable::RecursiveCTETable(TemporaryTableHolderPtr holder_,
|
||||
StoragePtr storage_,
|
||||
NamesAndTypes columns_)
|
||||
: holder(std::move(holder_))
|
||||
, storage(std::move(storage_))
|
||||
, columns(std::move(columns_))
|
||||
{}
|
||||
|
||||
StorageID RecursiveCTETable::getStorageID() const
|
||||
{
|
||||
return storage->getStorageID();
|
||||
}
|
||||
|
||||
}
|
||||
51
src/Analyzer/RecursiveCTE.h
Normal file
51
src/Analyzer/RecursiveCTE.h
Normal file
@ -0,0 +1,51 @@
|
||||
#pragma once
|
||||
|
||||
#include <Core/NamesAndTypes.h>
|
||||
|
||||
#include <Interpreters/DatabaseCatalog.h>
|
||||
|
||||
#include <Analyzer/IQueryTreeNode.h>
|
||||
#include <Analyzer/TableNode.h>
|
||||
|
||||
namespace DB
|
||||
{
|
||||
|
||||
/** Recursive CTEs allow to recursively evaluate UNION subqueries.
|
||||
*
|
||||
* Overview:
|
||||
* https://www.postgresql.org/docs/current/queries-with.html#QUERIES-WITH-RECURSIVE
|
||||
*
|
||||
* Current implementation algorithm:
|
||||
*
|
||||
* During query analysis, when we resolve UNION node that is inside WITH RECURSIVE section of parent query we:
|
||||
* 1. First resolve non recursive subquery.
|
||||
* 2. Create temporary table using projection columns of resolved subquery from step 1.
|
||||
* 3. Create temporary table expression node using storage from step 2.
|
||||
* 4. Create resolution scope for recursive subquery. In that scope we add node from step 3 as expression argument with UNION node CTE name.
|
||||
* 5. Resolve recursive subquery.
|
||||
* 6. If in resolved UNION node temporary table expression storage from step 2 is used, we update UNION query with recursive CTE table.
|
||||
*
|
||||
* During query planning if UNION node contains recursive CTE table, we add ReadFromRecursiveCTEStep to query plan. That step is responsible for whole
|
||||
* recursive CTE query execution.
|
||||
*
|
||||
* TODO: Improve locking in ReadFromRecursiveCTEStep.
|
||||
* TODO: Improve query analysis if query contains aggregates, JOINS, GROUP BY, ORDER BY, LIMIT, OFFSET.
|
||||
* TODO: Support SEARCH DEPTH FIRST BY, SEARCH BREADTH FIRST BY syntax.
|
||||
* TODO: Support CYCLE syntax.
|
||||
* TODO: Support UNION DISTINCT recursive CTE mode.
|
||||
*/
|
||||
class RecursiveCTETable
|
||||
{
|
||||
public:
|
||||
RecursiveCTETable(TemporaryTableHolderPtr holder_,
|
||||
StoragePtr storage_,
|
||||
NamesAndTypes columns_);
|
||||
|
||||
StorageID getStorageID() const;
|
||||
|
||||
TemporaryTableHolderPtr holder;
|
||||
StoragePtr storage;
|
||||
NamesAndTypes columns;
|
||||
};
|
||||
|
||||
}
|
||||
@ -33,6 +33,14 @@ TableNode::TableNode(StoragePtr storage_, const ContextPtr & context)
|
||||
{
|
||||
}
|
||||
|
||||
void TableNode::updateStorage(StoragePtr storage_value, const ContextPtr & context)
|
||||
{
|
||||
storage = std::move(storage_value);
|
||||
storage_id = storage->getStorageID();
|
||||
storage_lock = storage->lockForShare(context->getInitialQueryId(), context->getSettingsRef().lock_acquire_timeout);
|
||||
storage_snapshot = storage->getStorageSnapshot(storage->getInMemoryMetadataPtr(), context);
|
||||
}
|
||||
|
||||
void TableNode::dumpTreeImpl(WriteBuffer & buffer, FormatState & format_state, size_t indent) const
|
||||
{
|
||||
buffer << std::string(indent, ' ') << "TABLE id: " << format_state.getNodeId(this);
|
||||
|
||||
@ -32,6 +32,11 @@ public:
|
||||
/// Construct table node with storage, context
|
||||
explicit TableNode(StoragePtr storage_, const ContextPtr & context);
|
||||
|
||||
/** Update table node storage.
|
||||
* After this call storage, storage_id, storage_lock, storage_snapshot will be updated using new storage.
|
||||
*/
|
||||
void updateStorage(StoragePtr storage_value, const ContextPtr & context);
|
||||
|
||||
/// Get storage
|
||||
const StoragePtr & getStorage() const
|
||||
{
|
||||
|
||||
@ -9,6 +9,7 @@
|
||||
|
||||
#include <Parsers/ASTExpressionList.h>
|
||||
#include <Parsers/ASTTablesInSelectQuery.h>
|
||||
#include <Parsers/ASTWithElement.h>
|
||||
#include <Parsers/ASTSubquery.h>
|
||||
#include <Parsers/ASTSelectQuery.h>
|
||||
#include <Parsers/ASTSelectWithUnionQuery.h>
|
||||
@ -20,6 +21,8 @@
|
||||
|
||||
#include <DataTypes/getLeastSupertype.h>
|
||||
|
||||
#include <Storages/IStorage.h>
|
||||
|
||||
#include <Interpreters/Context.h>
|
||||
|
||||
#include <Analyzer/QueryNode.h>
|
||||
@ -49,6 +52,9 @@ UnionNode::UnionNode(ContextMutablePtr context_, SelectUnionMode union_mode_)
|
||||
|
||||
NamesAndTypes UnionNode::computeProjectionColumns() const
|
||||
{
|
||||
if (recursive_cte_table)
|
||||
return recursive_cte_table->columns;
|
||||
|
||||
std::vector<NamesAndTypes> projections;
|
||||
|
||||
NamesAndTypes query_node_projection;
|
||||
@ -90,6 +96,9 @@ NamesAndTypes UnionNode::computeProjectionColumns() const
|
||||
|
||||
void UnionNode::removeUnusedProjectionColumns(const std::unordered_set<std::string> & used_projection_columns)
|
||||
{
|
||||
if (recursive_cte_table)
|
||||
return;
|
||||
|
||||
auto projection_columns = computeProjectionColumns();
|
||||
size_t projection_columns_size = projection_columns.size();
|
||||
std::unordered_set<size_t> used_projection_column_indexes;
|
||||
@ -113,6 +122,9 @@ void UnionNode::removeUnusedProjectionColumns(const std::unordered_set<std::stri
|
||||
|
||||
void UnionNode::removeUnusedProjectionColumns(const std::unordered_set<size_t> & used_projection_columns_indexes)
|
||||
{
|
||||
if (recursive_cte_table)
|
||||
return;
|
||||
|
||||
auto & query_nodes = getQueries().getNodes();
|
||||
for (auto & query_node : query_nodes)
|
||||
{
|
||||
@ -136,6 +148,12 @@ void UnionNode::dumpTreeImpl(WriteBuffer & buffer, FormatState & format_state, s
|
||||
if (is_cte)
|
||||
buffer << ", is_cte: " << is_cte;
|
||||
|
||||
if (is_recursive_cte)
|
||||
buffer << ", is_recursive_cte: " << is_recursive_cte;
|
||||
|
||||
if (recursive_cte_table)
|
||||
buffer << ", recursive_cte_table: " << recursive_cte_table->storage->getStorageID().getNameForLogs();
|
||||
|
||||
if (!cte_name.empty())
|
||||
buffer << ", cte_name: " << cte_name;
|
||||
|
||||
@ -149,14 +167,28 @@ bool UnionNode::isEqualImpl(const IQueryTreeNode & rhs, CompareOptions) const
|
||||
{
|
||||
const auto & rhs_typed = assert_cast<const UnionNode &>(rhs);
|
||||
|
||||
return is_subquery == rhs_typed.is_subquery && is_cte == rhs_typed.is_cte && cte_name == rhs_typed.cte_name &&
|
||||
union_mode == rhs_typed.union_mode;
|
||||
if (recursive_cte_table && rhs_typed.recursive_cte_table &&
|
||||
recursive_cte_table->getStorageID() != rhs_typed.recursive_cte_table->getStorageID())
|
||||
return false;
|
||||
else if ((recursive_cte_table && !rhs_typed.recursive_cte_table) || (!recursive_cte_table && rhs_typed.recursive_cte_table))
|
||||
return false;
|
||||
|
||||
return is_subquery == rhs_typed.is_subquery && is_cte == rhs_typed.is_cte && is_recursive_cte == rhs_typed.is_recursive_cte
|
||||
&& cte_name == rhs_typed.cte_name && union_mode == rhs_typed.union_mode;
|
||||
}
|
||||
|
||||
void UnionNode::updateTreeHashImpl(HashState & state, CompareOptions) const
|
||||
{
|
||||
state.update(is_subquery);
|
||||
state.update(is_cte);
|
||||
state.update(is_recursive_cte);
|
||||
|
||||
if (recursive_cte_table)
|
||||
{
|
||||
auto full_name = recursive_cte_table->getStorageID().getFullNameNotQuoted();
|
||||
state.update(full_name.size());
|
||||
state.update(full_name);
|
||||
}
|
||||
|
||||
state.update(cte_name.size());
|
||||
state.update(cte_name);
|
||||
@ -170,6 +202,8 @@ QueryTreeNodePtr UnionNode::cloneImpl() const
|
||||
|
||||
result_union_node->is_subquery = is_subquery;
|
||||
result_union_node->is_cte = is_cte;
|
||||
result_union_node->is_recursive_cte = is_recursive_cte;
|
||||
result_union_node->recursive_cte_table = recursive_cte_table;
|
||||
result_union_node->cte_name = cte_name;
|
||||
|
||||
return result_union_node;
|
||||
@ -183,14 +217,64 @@ ASTPtr UnionNode::toASTImpl(const ConvertToASTOptions & options) const
|
||||
select_with_union_query->children.push_back(getQueriesNode()->toAST(options));
|
||||
select_with_union_query->list_of_selects = select_with_union_query->children.back();
|
||||
|
||||
if (is_subquery)
|
||||
ASTPtr result_query = std::move(select_with_union_query);
|
||||
bool set_subquery_cte_name = true;
|
||||
|
||||
if (recursive_cte_table)
|
||||
{
|
||||
auto subquery = std::make_shared<ASTSubquery>(std::move(select_with_union_query));
|
||||
subquery->cte_name = cte_name;
|
||||
return subquery;
|
||||
auto recursive_select_query = std::make_shared<ASTSelectQuery>();
|
||||
recursive_select_query->recursive_with = true;
|
||||
|
||||
auto with_element_ast = std::make_shared<ASTWithElement>();
|
||||
with_element_ast->name = cte_name;
|
||||
with_element_ast->subquery = std::make_shared<ASTSubquery>(std::move(result_query));
|
||||
with_element_ast->children.push_back(with_element_ast->subquery);
|
||||
|
||||
auto with_expression_list_ast = std::make_shared<ASTExpressionList>();
|
||||
with_expression_list_ast->children.push_back(std::move(with_element_ast));
|
||||
|
||||
recursive_select_query->setExpression(ASTSelectQuery::Expression::WITH, std::move(with_expression_list_ast));
|
||||
|
||||
auto select_expression_list_ast = std::make_shared<ASTExpressionList>();
|
||||
select_expression_list_ast->children.reserve(recursive_cte_table->columns.size());
|
||||
for (const auto & recursive_cte_table_column : recursive_cte_table->columns)
|
||||
select_expression_list_ast->children.push_back(std::make_shared<ASTIdentifier>(recursive_cte_table_column.name));
|
||||
|
||||
recursive_select_query->setExpression(ASTSelectQuery::Expression::SELECT, std::move(select_expression_list_ast));
|
||||
|
||||
auto table_expression_ast = std::make_shared<ASTTableExpression>();
|
||||
table_expression_ast->children.push_back(std::make_shared<ASTTableIdentifier>(cte_name));
|
||||
table_expression_ast->database_and_table_name = table_expression_ast->children.back();
|
||||
|
||||
auto tables_in_select_query_element_ast = std::make_shared<ASTTablesInSelectQueryElement>();
|
||||
tables_in_select_query_element_ast->children.push_back(std::move(table_expression_ast));
|
||||
tables_in_select_query_element_ast->table_expression = tables_in_select_query_element_ast->children.back();
|
||||
|
||||
ASTPtr tables_in_select_query_ast = std::make_shared<ASTTablesInSelectQuery>();
|
||||
tables_in_select_query_ast->children.push_back(std::move(tables_in_select_query_element_ast));
|
||||
|
||||
recursive_select_query->setExpression(ASTSelectQuery::Expression::TABLES, std::move(tables_in_select_query_ast));
|
||||
|
||||
auto recursive_select_with_union_query = std::make_shared<ASTSelectWithUnionQuery>();
|
||||
auto recursive_select_with_union_query_list_of_selects = std::make_shared<ASTExpressionList>();
|
||||
recursive_select_with_union_query_list_of_selects->children.push_back(std::move(recursive_select_query));
|
||||
recursive_select_with_union_query->children.push_back(std::move(recursive_select_with_union_query_list_of_selects));
|
||||
recursive_select_with_union_query->list_of_selects = recursive_select_with_union_query->children.back();
|
||||
|
||||
result_query = std::move(recursive_select_with_union_query);
|
||||
set_subquery_cte_name = false;
|
||||
}
|
||||
|
||||
return select_with_union_query;
|
||||
if (is_subquery)
|
||||
{
|
||||
auto subquery = std::make_shared<ASTSubquery>(std::move(result_query));
|
||||
if (set_subquery_cte_name)
|
||||
subquery->cte_name = cte_name;
|
||||
|
||||
result_query = std::move(subquery);
|
||||
}
|
||||
|
||||
return result_query;
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
@ -9,6 +9,7 @@
|
||||
#include <Analyzer/IQueryTreeNode.h>
|
||||
#include <Analyzer/ListNode.h>
|
||||
#include <Analyzer/TableExpressionModifiers.h>
|
||||
#include <Analyzer/RecursiveCTE.h>
|
||||
|
||||
#include <Interpreters/Context_fwd.h>
|
||||
|
||||
@ -84,6 +85,42 @@ public:
|
||||
is_cte = is_cte_value;
|
||||
}
|
||||
|
||||
/// Returns true if union node CTE is specified in WITH RECURSIVE, false otherwise
|
||||
bool isRecursiveCTE() const
|
||||
{
|
||||
return is_recursive_cte;
|
||||
}
|
||||
|
||||
/// Set union node is recursive CTE value
|
||||
void setIsRecursiveCTE(bool is_recursive_cte_value)
|
||||
{
|
||||
is_recursive_cte = is_recursive_cte_value;
|
||||
}
|
||||
|
||||
/// Returns true if union node has recursive CTE table, false otherwise
|
||||
bool hasRecursiveCTETable() const
|
||||
{
|
||||
return recursive_cte_table.has_value();
|
||||
}
|
||||
|
||||
/// Returns optional recursive CTE table
|
||||
const std::optional<RecursiveCTETable> & getRecursiveCTETable() const
|
||||
{
|
||||
return recursive_cte_table;
|
||||
}
|
||||
|
||||
/// Returns optional recursive CTE table
|
||||
std::optional<RecursiveCTETable> & getRecursiveCTETable()
|
||||
{
|
||||
return recursive_cte_table;
|
||||
}
|
||||
|
||||
/// Set union node recursive CTE table value
|
||||
void setRecursiveCTETable(RecursiveCTETable recursive_cte_table_value)
|
||||
{
|
||||
recursive_cte_table.emplace(std::move(recursive_cte_table_value));
|
||||
}
|
||||
|
||||
/// Get union node CTE name
|
||||
const std::string & getCTEName() const
|
||||
{
|
||||
@ -154,6 +191,8 @@ protected:
|
||||
private:
|
||||
bool is_subquery = false;
|
||||
bool is_cte = false;
|
||||
bool is_recursive_cte = false;
|
||||
std::optional<RecursiveCTETable> recursive_cte_table;
|
||||
std::string cte_name;
|
||||
ContextMutablePtr context;
|
||||
SelectUnionMode union_mode;
|
||||
|
||||
@ -5,6 +5,7 @@
|
||||
#include <Parsers/ASTSubquery.h>
|
||||
#include <Parsers/ASTFunction.h>
|
||||
|
||||
#include <DataTypes/DataTypesNumber.h>
|
||||
#include <DataTypes/DataTypeString.h>
|
||||
#include <DataTypes/DataTypeTuple.h>
|
||||
#include <DataTypes/DataTypeArray.h>
|
||||
@ -15,6 +16,8 @@
|
||||
#include <Functions/FunctionHelpers.h>
|
||||
#include <Functions/FunctionFactory.h>
|
||||
|
||||
#include <Storages/IStorage.h>
|
||||
|
||||
#include <Interpreters/Context.h>
|
||||
|
||||
#include <Analyzer/InDepthQueryTreeVisitor.h>
|
||||
@ -61,6 +64,36 @@ bool isNodePartOfTree(const IQueryTreeNode * node, const IQueryTreeNode * root)
|
||||
return false;
|
||||
}
|
||||
|
||||
bool isStorageUsedInTree(const StoragePtr & storage, const IQueryTreeNode * root)
|
||||
{
|
||||
std::vector<const IQueryTreeNode *> nodes_to_process;
|
||||
nodes_to_process.push_back(root);
|
||||
|
||||
while (!nodes_to_process.empty())
|
||||
{
|
||||
const auto * subtree_node = nodes_to_process.back();
|
||||
nodes_to_process.pop_back();
|
||||
|
||||
const auto * table_node = subtree_node->as<TableNode>();
|
||||
const auto * table_function_node = subtree_node->as<TableFunctionNode>();
|
||||
|
||||
if (table_node || table_function_node)
|
||||
{
|
||||
const auto & table_storage = table_node ? table_node->getStorage() : table_function_node->getStorage();
|
||||
if (table_storage->getStorageID() == storage->getStorageID())
|
||||
return true;
|
||||
}
|
||||
|
||||
for (const auto & child : subtree_node->getChildren())
|
||||
{
|
||||
if (child)
|
||||
nodes_to_process.push_back(child.get());
|
||||
}
|
||||
}
|
||||
|
||||
return false;
|
||||
}
|
||||
|
||||
bool isNameOfInFunction(const std::string & function_name)
|
||||
{
|
||||
bool is_special_function_in = function_name == "in" ||
|
||||
@ -808,26 +841,87 @@ QueryTreeNodePtr getExpressionSource(const QueryTreeNodePtr & node)
|
||||
return source;
|
||||
}
|
||||
|
||||
QueryTreeNodePtr buildSubqueryToReadColumnsFromTableExpression(QueryTreeNodePtr table_node, const ContextPtr & context)
|
||||
/** There are no limits on the maximum size of the result for the subquery.
|
||||
* Since the result of the query is not the result of the entire query.
|
||||
*/
|
||||
void updateContextForSubqueryExecution(ContextMutablePtr & mutable_context)
|
||||
{
|
||||
/** The subquery in the IN / JOIN section does not have any restrictions on the maximum size of the result.
|
||||
* Because the result of this query is not the result of the entire query.
|
||||
* Constraints work instead
|
||||
* max_rows_in_set, max_bytes_in_set, set_overflow_mode,
|
||||
* max_rows_in_join, max_bytes_in_join, join_overflow_mode,
|
||||
* which are checked separately (in the Set, Join objects).
|
||||
*/
|
||||
Settings subquery_settings = mutable_context->getSettings();
|
||||
subquery_settings.max_result_rows = 0;
|
||||
subquery_settings.max_result_bytes = 0;
|
||||
/// The calculation of extremes does not make sense and is not necessary (if you do it, then the extremes of the subquery can be taken for whole query).
|
||||
subquery_settings.extremes = false;
|
||||
mutable_context->setSettings(subquery_settings);
|
||||
}
|
||||
|
||||
QueryTreeNodePtr buildQueryToReadColumnsFromTableExpression(const NamesAndTypes & columns,
|
||||
const QueryTreeNodePtr & table_expression,
|
||||
ContextMutablePtr & context)
|
||||
{
|
||||
auto projection_columns = columns;
|
||||
|
||||
QueryTreeNodes subquery_projection_nodes;
|
||||
subquery_projection_nodes.reserve(projection_columns.size());
|
||||
|
||||
for (const auto & column : projection_columns)
|
||||
subquery_projection_nodes.push_back(std::make_shared<ColumnNode>(column, table_expression));
|
||||
|
||||
if (subquery_projection_nodes.empty())
|
||||
{
|
||||
auto constant_data_type = std::make_shared<DataTypeUInt64>();
|
||||
subquery_projection_nodes.push_back(std::make_shared<ConstantNode>(1UL, constant_data_type));
|
||||
projection_columns.push_back({"1", std::move(constant_data_type)});
|
||||
}
|
||||
|
||||
updateContextForSubqueryExecution(context);
|
||||
|
||||
auto query_node = std::make_shared<QueryNode>(std::move(context));
|
||||
|
||||
query_node->getProjection().getNodes() = std::move(subquery_projection_nodes);
|
||||
query_node->resolveProjectionColumns(projection_columns);
|
||||
query_node->getJoinTree() = table_expression;
|
||||
|
||||
return query_node;
|
||||
}
|
||||
|
||||
QueryTreeNodePtr buildSubqueryToReadColumnsFromTableExpression(const NamesAndTypes & columns,
|
||||
const QueryTreeNodePtr & table_expression,
|
||||
ContextMutablePtr & context)
|
||||
{
|
||||
auto result = buildQueryToReadColumnsFromTableExpression(columns, table_expression, context);
|
||||
result->as<QueryNode &>().setIsSubquery(true);
|
||||
return result;
|
||||
}
|
||||
|
||||
QueryTreeNodePtr buildQueryToReadColumnsFromTableExpression(const NamesAndTypes & columns,
|
||||
const QueryTreeNodePtr & table_expression,
|
||||
const ContextPtr & context)
|
||||
{
|
||||
auto context_copy = Context::createCopy(context);
|
||||
return buildQueryToReadColumnsFromTableExpression(columns, table_expression, context_copy);
|
||||
}
|
||||
|
||||
QueryTreeNodePtr buildSubqueryToReadColumnsFromTableExpression(const NamesAndTypes & columns,
|
||||
const QueryTreeNodePtr & table_expression,
|
||||
const ContextPtr & context)
|
||||
{
|
||||
auto context_copy = Context::createCopy(context);
|
||||
return buildSubqueryToReadColumnsFromTableExpression(columns, table_expression, context_copy);
|
||||
}
|
||||
|
||||
QueryTreeNodePtr buildSubqueryToReadColumnsFromTableExpression(const QueryTreeNodePtr & table_node, const ContextPtr & context)
|
||||
{
|
||||
const auto & storage_snapshot = table_node->as<TableNode>()->getStorageSnapshot();
|
||||
auto columns_to_select = storage_snapshot->getColumns(GetColumnsOptions(GetColumnsOptions::Ordinary));
|
||||
size_t columns_to_select_size = columns_to_select.size();
|
||||
auto column_nodes_to_select = std::make_shared<ListNode>();
|
||||
column_nodes_to_select->getNodes().reserve(columns_to_select_size);
|
||||
NamesAndTypes projection_columns;
|
||||
projection_columns.reserve(columns_to_select_size);
|
||||
for (auto & column : columns_to_select)
|
||||
{
|
||||
column_nodes_to_select->getNodes().emplace_back(std::make_shared<ColumnNode>(column, table_node));
|
||||
projection_columns.emplace_back(column.name, column.type);
|
||||
}
|
||||
auto subquery_for_table = std::make_shared<QueryNode>(Context::createCopy(context));
|
||||
subquery_for_table->setIsSubquery(true);
|
||||
subquery_for_table->getProjectionNode() = std::move(column_nodes_to_select);
|
||||
subquery_for_table->getJoinTree() = std::move(table_node);
|
||||
subquery_for_table->resolveProjectionColumns(std::move(projection_columns));
|
||||
return subquery_for_table;
|
||||
auto columns_to_select_list = storage_snapshot->getColumns(GetColumnsOptions(GetColumnsOptions::Ordinary));
|
||||
NamesAndTypes columns_to_select(columns_to_select_list.begin(), columns_to_select_list.end());
|
||||
return buildSubqueryToReadColumnsFromTableExpression(columns_to_select, table_node, context);
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
@ -1,9 +1,13 @@
|
||||
#pragma once
|
||||
|
||||
#include <Analyzer/IQueryTreeNode.h>
|
||||
#include <Core/NamesAndTypes.h>
|
||||
|
||||
#include <Storages/IStorage_fwd.h>
|
||||
|
||||
#include <Interpreters/Context_fwd.h>
|
||||
|
||||
#include <Analyzer/IQueryTreeNode.h>
|
||||
|
||||
namespace DB
|
||||
{
|
||||
|
||||
@ -12,6 +16,9 @@ class FunctionNode;
|
||||
/// Returns true if node part of root tree, false otherwise
|
||||
bool isNodePartOfTree(const IQueryTreeNode * node, const IQueryTreeNode * root);
|
||||
|
||||
/// Returns true if storage is used in tree, false otherwise
|
||||
bool isStorageUsedInTree(const StoragePtr & storage, const IQueryTreeNode * root);
|
||||
|
||||
/// Returns true if function name is name of IN function or its variations, false otherwise
|
||||
bool isNameOfInFunction(const std::string & function_name);
|
||||
|
||||
@ -108,7 +115,41 @@ QueryTreeNodePtr createCastFunction(QueryTreeNodePtr node, DataTypePtr result_ty
|
||||
/// Checks that node has only one source and returns it
|
||||
QueryTreeNodePtr getExpressionSource(const QueryTreeNodePtr & node);
|
||||
|
||||
/// Build subquery which we execute for `IN table` function.
|
||||
QueryTreeNodePtr buildSubqueryToReadColumnsFromTableExpression(QueryTreeNodePtr table_node, const ContextPtr & context);
|
||||
/// Update mutable context for subquery execution
|
||||
void updateContextForSubqueryExecution(ContextMutablePtr & mutable_context);
|
||||
|
||||
/** Build query to read specified columns from table expression.
|
||||
* Specified mutable context will be used as query context.
|
||||
*/
|
||||
QueryTreeNodePtr buildQueryToReadColumnsFromTableExpression(const NamesAndTypes & columns,
|
||||
const QueryTreeNodePtr & table_expression,
|
||||
ContextMutablePtr & context);
|
||||
|
||||
/** Build subquery to read specified columns from table expression.
|
||||
* Specified mutable context will be used as query context.
|
||||
*/
|
||||
QueryTreeNodePtr buildSubqueryToReadColumnsFromTableExpression(const NamesAndTypes & columns,
|
||||
const QueryTreeNodePtr & table_expression,
|
||||
ContextMutablePtr & context);
|
||||
|
||||
/** Build query to read specified columns from table expression.
|
||||
* Specified context will be copied and used as query context.
|
||||
*/
|
||||
QueryTreeNodePtr buildQueryToReadColumnsFromTableExpression(const NamesAndTypes & columns,
|
||||
const QueryTreeNodePtr & table_expression,
|
||||
const ContextPtr & context);
|
||||
|
||||
/** Build subquery to read specified columns from table expression.
|
||||
* Specified context will be copied and used as query context.
|
||||
*/
|
||||
QueryTreeNodePtr buildSubqueryToReadColumnsFromTableExpression(const NamesAndTypes & columns,
|
||||
const QueryTreeNodePtr & table_expression,
|
||||
const ContextPtr & context);
|
||||
|
||||
/** Build subquery to read all columns from table expression.
|
||||
* Specified context will be copied and used as query context.
|
||||
*/
|
||||
QueryTreeNodePtr buildSubqueryToReadColumnsFromTableExpression(const QueryTreeNodePtr & table_node, const ContextPtr & context);
|
||||
|
||||
|
||||
}
|
||||
|
||||
@ -221,7 +221,8 @@ std::unique_ptr<WriteBuffer> BackupWriterAzureBlobStorage::writeFile(const Strin
|
||||
key,
|
||||
DBMS_DEFAULT_BUFFER_SIZE,
|
||||
write_settings,
|
||||
settings);
|
||||
settings,
|
||||
threadPoolCallbackRunnerUnsafe<void>(getBackupsIOThreadPool().get(), "BackupWRAzure"));
|
||||
}
|
||||
|
||||
void BackupWriterAzureBlobStorage::removeFile(const String & file_name)
|
||||
|
||||
@ -109,7 +109,7 @@ RestorerFromBackup::~RestorerFromBackup()
|
||||
if (getNumFutures() > 0)
|
||||
{
|
||||
LOG_INFO(log, "Waiting for {} tasks to finish", getNumFutures());
|
||||
waitFutures();
|
||||
waitFutures(/* throw_if_error= */ false);
|
||||
}
|
||||
}
|
||||
|
||||
@ -161,7 +161,7 @@ void RestorerFromBackup::run(Mode mode)
|
||||
setStage(Stage::COMPLETED);
|
||||
}
|
||||
|
||||
void RestorerFromBackup::waitFutures()
|
||||
void RestorerFromBackup::waitFutures(bool throw_if_error)
|
||||
{
|
||||
std::exception_ptr error;
|
||||
|
||||
@ -176,11 +176,7 @@ void RestorerFromBackup::waitFutures()
|
||||
if (futures_to_wait.empty())
|
||||
break;
|
||||
|
||||
/// Wait for all tasks.
|
||||
for (auto & future : futures_to_wait)
|
||||
future.wait();
|
||||
|
||||
/// Check if there is an exception.
|
||||
/// Wait for all tasks to finish.
|
||||
for (auto & future : futures_to_wait)
|
||||
{
|
||||
try
|
||||
@ -197,7 +193,12 @@ void RestorerFromBackup::waitFutures()
|
||||
}
|
||||
|
||||
if (error)
|
||||
std::rethrow_exception(error);
|
||||
{
|
||||
if (throw_if_error)
|
||||
std::rethrow_exception(error);
|
||||
else
|
||||
tryLogException(error, log);
|
||||
}
|
||||
}
|
||||
|
||||
size_t RestorerFromBackup::getNumFutures() const
|
||||
|
||||
@ -130,7 +130,7 @@ private:
|
||||
|
||||
/// Waits until all tasks are processed (including the tasks scheduled while we're waiting).
|
||||
/// Throws an exception if any of the tasks throws an exception.
|
||||
void waitFutures();
|
||||
void waitFutures(bool throw_if_error = true);
|
||||
|
||||
/// Throws an exception if the RESTORE query was cancelled.
|
||||
void checkIsQueryCancelled() const;
|
||||
|
||||
@ -2955,7 +2955,8 @@ void ClientBase::init(int argc, char ** argv)
|
||||
|
||||
/// Common options for clickhouse-client and clickhouse-local.
|
||||
options_description.main_description->add_options()
|
||||
("help", "produce help message")
|
||||
("help", "print usage summary, combine with --verbose to display all options")
|
||||
("verbose", "print query and other debugging info")
|
||||
("version,V", "print version information and exit")
|
||||
("version-clean", "print version in machine-readable format and exit")
|
||||
|
||||
@ -2979,7 +2980,6 @@ void ClientBase::init(int argc, char ** argv)
|
||||
("time,t", "print query execution time to stderr in non-interactive mode (for benchmarks)")
|
||||
|
||||
("echo", "in batch mode, print query before execution")
|
||||
("verbose", "print query and other debugging info")
|
||||
|
||||
("log-level", po::value<std::string>(), "log level")
|
||||
("server_logs_file", po::value<std::string>(), "put server logs into specified file")
|
||||
@ -3008,6 +3008,8 @@ void ClientBase::init(int argc, char ** argv)
|
||||
|
||||
addOptions(options_description);
|
||||
|
||||
OptionsDescription options_description_non_verbose = options_description;
|
||||
|
||||
auto getter = [](const auto & op)
|
||||
{
|
||||
String op_long_name = op->long_name();
|
||||
@ -3042,11 +3044,17 @@ void ClientBase::init(int argc, char ** argv)
|
||||
exit(0); // NOLINT(concurrency-mt-unsafe)
|
||||
}
|
||||
|
||||
if (options.count("verbose"))
|
||||
config().setBool("verbose", true);
|
||||
|
||||
/// Output of help message.
|
||||
if (options.count("help")
|
||||
|| (options.count("host") && options["host"].as<std::string>() == "elp")) /// If user writes -help instead of --help.
|
||||
{
|
||||
printHelpMessage(options_description);
|
||||
if (config().getBool("verbose", false))
|
||||
printHelpMessage(options_description, true);
|
||||
else
|
||||
printHelpMessage(options_description_non_verbose, false);
|
||||
exit(0); // NOLINT(concurrency-mt-unsafe)
|
||||
}
|
||||
|
||||
@ -3113,8 +3121,6 @@ void ClientBase::init(int argc, char ** argv)
|
||||
config().setBool("highlight", options["highlight"].as<bool>());
|
||||
if (options.count("history_file"))
|
||||
config().setString("history_file", options["history_file"].as<std::string>());
|
||||
if (options.count("verbose"))
|
||||
config().setBool("verbose", true);
|
||||
if (options.count("interactive"))
|
||||
config().setBool("interactive", true);
|
||||
if (options.count("pager"))
|
||||
|
||||
@ -121,7 +121,7 @@ protected:
|
||||
};
|
||||
|
||||
virtual void updateLoggerLevel(const String &) {}
|
||||
virtual void printHelpMessage(const OptionsDescription & options_description) = 0;
|
||||
virtual void printHelpMessage(const OptionsDescription & options_description, bool verbose) = 0;
|
||||
virtual void addOptions(OptionsDescription & options_description) = 0;
|
||||
virtual void processOptions(const OptionsDescription & options_description,
|
||||
const CommandLineOptions & options,
|
||||
|
||||
@ -28,28 +28,6 @@ namespace ErrorCodes
|
||||
extern const int USER_SESSION_LIMIT_EXCEEDED;
|
||||
}
|
||||
|
||||
Suggest::Suggest()
|
||||
{
|
||||
/// Keywords may be not up to date with ClickHouse parser.
|
||||
addWords({"CREATE", "DATABASE", "IF", "NOT", "EXISTS", "TEMPORARY", "TABLE", "ON",
|
||||
"CLUSTER", "DEFAULT", "MATERIALIZED", "ALIAS", "ENGINE", "AS", "VIEW", "POPULATE",
|
||||
"SETTINGS", "ATTACH", "DETACH", "DROP", "RENAME", "TO", "ALTER", "ADD",
|
||||
"MODIFY", "CLEAR", "COLUMN", "AFTER", "COPY", "PROJECT", "PRIMARY", "KEY",
|
||||
"CHECK", "PARTITION", "PART", "FREEZE", "FETCH", "FROM", "SHOW", "INTO",
|
||||
"OUTFILE", "FORMAT", "TABLES", "DATABASES", "LIKE", "PROCESSLIST", "CASE", "WHEN",
|
||||
"THEN", "ELSE", "END", "DESCRIBE", "DESC", "USE", "SET", "OPTIMIZE",
|
||||
"FINAL", "DEDUPLICATE", "INSERT", "VALUES", "SELECT", "DISTINCT", "SAMPLE", "ARRAY",
|
||||
"JOIN", "GLOBAL", "LOCAL", "ANY", "ALL", "INNER", "LEFT", "RIGHT",
|
||||
"FULL", "OUTER", "CROSS", "USING", "PREWHERE", "WHERE", "GROUP", "BY",
|
||||
"WITH", "TOTALS", "HAVING", "ORDER", "COLLATE", "LIMIT", "UNION", "AND",
|
||||
"OR", "ASC", "IN", "KILL", "QUERY", "SYNC", "ASYNC", "TEST",
|
||||
"BETWEEN", "TRUNCATE", "USER", "ROLE", "PROFILE", "QUOTA", "POLICY", "ROW",
|
||||
"GRANT", "REVOKE", "OPTION", "ADMIN", "EXCEPT", "REPLACE", "IDENTIFIED", "HOST",
|
||||
"NAME", "READONLY", "WRITABLE", "PERMISSIVE", "FOR", "RESTRICTIVE", "RANDOMIZED", "INTERVAL",
|
||||
"LIMITS", "ONLY", "TRACKING", "IP", "REGEXP", "ILIKE", "CLEANUP", "APPEND",
|
||||
"IGNORE NULLS", "RESPECT NULLS", "OVER", "PASTE", "WINDOW", "QUALIFY"});
|
||||
}
|
||||
|
||||
static String getLoadSuggestionQuery(Int32 suggestion_limit, bool basic_suggestion)
|
||||
{
|
||||
/// NOTE: Once you will update the completion list,
|
||||
@ -82,6 +60,7 @@ static String getLoadSuggestionQuery(Int32 suggestion_limit, bool basic_suggesti
|
||||
add_column("name", "data_type_families", false, {});
|
||||
add_column("name", "merge_tree_settings", false, {});
|
||||
add_column("name", "settings", false, {});
|
||||
add_column("keyword", "keywords", false, {});
|
||||
|
||||
if (!basic_suggestion)
|
||||
{
|
||||
|
||||
@ -17,7 +17,7 @@ namespace DB
|
||||
class Suggest : public LineReader::Suggest, boost::noncopyable
|
||||
{
|
||||
public:
|
||||
Suggest();
|
||||
Suggest() = default;
|
||||
|
||||
~Suggest()
|
||||
{
|
||||
|
||||
@ -39,6 +39,7 @@ static struct InitFiu
|
||||
REGULAR(replicated_merge_tree_commit_zk_fail_when_recovering_from_hw_fault) \
|
||||
REGULAR(use_delayed_remote_source) \
|
||||
REGULAR(cluster_discovery_faults) \
|
||||
REGULAR(replicated_sends_failpoint) \
|
||||
ONCE(smt_commit_merge_mutate_zk_fail_after_op) \
|
||||
ONCE(smt_commit_merge_mutate_zk_fail_before_op) \
|
||||
ONCE(smt_commit_write_zk_fail_after_op) \
|
||||
|
||||
@ -24,6 +24,7 @@ public:
|
||||
void updateHash(SipHash & hash) const override;
|
||||
|
||||
protected:
|
||||
/// 1 byte (`gcd_bytes_size` value) + 1 byte (`bytes_to_skip` value) + `bytes_to_skip` bytes (trash) + `gcd_bytes_size` bytes (gcd value) + (`source_size` - `bytes_to_skip`) bytes (data)
|
||||
UInt32 doCompressData(const char * source, UInt32 source_size, char * dest) const override;
|
||||
void doDecompressData(const char * source, UInt32 source_size, char * dest, UInt32 uncompressed_size) const override;
|
||||
UInt32 getMaxCompressedDataSize(UInt32 uncompressed_size) const override;
|
||||
@ -54,7 +55,7 @@ UInt32 CompressionCodecGCD::getMaxCompressedDataSize(UInt32 uncompressed_size) c
|
||||
{
|
||||
return uncompressed_size
|
||||
+ gcd_bytes_size // To store gcd
|
||||
+ 2; // Local header
|
||||
+ 2; // Values of `gcd_bytes_size` and `bytes_to_skip`
|
||||
}
|
||||
|
||||
uint8_t CompressionCodecGCD::getMethodByte() const
|
||||
@ -147,7 +148,7 @@ void decompressDataForType(const char * source, UInt32 source_size, char * dest,
|
||||
if (source_size - sizeof(T) != output_size)
|
||||
throw Exception(ErrorCodes::CANNOT_DECOMPRESS, "Cannot decompress GCD-encoded data");
|
||||
|
||||
memcpy(dest, source, source_size);
|
||||
memcpy(dest, source, source_size - sizeof(T));
|
||||
return;
|
||||
}
|
||||
|
||||
@ -160,6 +161,7 @@ void decompressDataForType(const char * source, UInt32 source_size, char * dest,
|
||||
source += sizeof(T);
|
||||
dest += sizeof(T);
|
||||
}
|
||||
chassert(source == source_end);
|
||||
}
|
||||
|
||||
}
|
||||
@ -209,6 +211,8 @@ void CompressionCodecGCD::doDecompressData(const char * source, UInt32 source_si
|
||||
throw Exception(ErrorCodes::CANNOT_DECOMPRESS, "Cannot decompress GCD-encoded data. File has wrong header");
|
||||
|
||||
UInt8 bytes_to_skip = uncompressed_size % bytes_size;
|
||||
chassert(bytes_to_skip == static_cast<UInt8>(source[1]));
|
||||
|
||||
UInt32 output_size = uncompressed_size - bytes_to_skip;
|
||||
|
||||
if (static_cast<UInt32>(2 + bytes_to_skip) > source_size)
|
||||
|
||||
@ -67,6 +67,9 @@ static constexpr auto DBMS_DEFAULT_MAX_PARSER_DEPTH = 1000;
|
||||
/// Default limit on the amount of backtracking of recursive descend parser.
|
||||
static constexpr auto DBMS_DEFAULT_MAX_PARSER_BACKTRACKS = 1000000;
|
||||
|
||||
/// Default limit on recursive CTE evaluation depth.
|
||||
static constexpr auto DBMS_RECURSIVE_CTE_MAX_EVALUATION_DEPTH = 1000;
|
||||
|
||||
/// Default limit on query size.
|
||||
static constexpr auto DBMS_DEFAULT_MAX_QUERY_SIZE = 262144;
|
||||
|
||||
|
||||
@ -112,6 +112,7 @@ class IColumn;
|
||||
M(Bool, azure_create_new_file_on_insert, false, "Enables or disables creating a new file on each insert in azure engine tables", 0) \
|
||||
M(Bool, s3_check_objects_after_upload, false, "Check each uploaded object to s3 with head request to be sure that upload was successful", 0) \
|
||||
M(Bool, s3_allow_parallel_part_upload, true, "Use multiple threads for s3 multipart upload. It may lead to slightly higher memory usage", 0) \
|
||||
M(Bool, azure_allow_parallel_part_upload, true, "Use multiple threads for azure multipart upload.", 0) \
|
||||
M(Bool, s3_throw_on_zero_files_match, false, "Throw an error, when ListObjects request cannot match any files", 0) \
|
||||
M(Bool, s3_disable_checksum, false, "Do not calculate a checksum when sending a file to S3. This speeds up writes by avoiding excessive processing passes on a file. It is mostly safe as the data of MergeTree tables is checksummed by ClickHouse anyway, and when S3 is accessed with HTTPS, the TLS layer already provides integrity while transferring through the network. While additional checksums on S3 give defense in depth.", 0) \
|
||||
M(UInt64, s3_retry_attempts, 100, "Setting for Aws::Client::RetryStrategy, Aws::Client does retries itself, 0 means no retries", 0) \
|
||||
@ -603,7 +604,6 @@ class IColumn;
|
||||
M(Bool, optimize_if_chain_to_multiif, false, "Replace if(cond1, then1, if(cond2, ...)) chains to multiIf. Currently it's not beneficial for numeric types.", 0) \
|
||||
M(Bool, optimize_multiif_to_if, true, "Replace 'multiIf' with only one condition to 'if'.", 0) \
|
||||
M(Bool, optimize_if_transform_strings_to_enum, false, "Replaces string-type arguments in If and Transform to enum. Disabled by default cause it could make inconsistent change in distributed query that would lead to its fail.", 0) \
|
||||
M(Bool, optimize_monotonous_functions_in_order_by, false, "Replace monotonous function with its argument in ORDER BY", 0) \
|
||||
M(Bool, optimize_functions_to_subcolumns, false, "Transform functions to subcolumns, if possible, to reduce amount of read data. E.g. 'length(arr)' -> 'arr.size0', 'col IS NULL' -> 'col.null' ", 0) \
|
||||
M(Bool, optimize_using_constraints, false, "Use constraints for query optimization", 0) \
|
||||
M(Bool, optimize_substitute_columns, false, "Use constraints for column substitution", 0) \
|
||||
@ -622,6 +622,7 @@ class IColumn;
|
||||
M(Bool, validate_polygons, true, "Throw exception if polygon is invalid in function pointInPolygon (e.g. self-tangent, self-intersecting). If the setting is false, the function will accept invalid polygons but may silently return wrong result.", 0) \
|
||||
M(UInt64, max_parser_depth, DBMS_DEFAULT_MAX_PARSER_DEPTH, "Maximum parser depth (recursion depth of recursive descend parser).", 0) \
|
||||
M(UInt64, max_parser_backtracks, DBMS_DEFAULT_MAX_PARSER_BACKTRACKS, "Maximum parser backtracking (how many times it tries different alternatives in the recursive descend parsing process).", 0) \
|
||||
M(UInt64, max_recursive_cte_evaluation_depth, DBMS_RECURSIVE_CTE_MAX_EVALUATION_DEPTH, "Maximum limit on recursive CTE evaluation depth", 0) \
|
||||
M(Bool, allow_settings_after_format_in_insert, false, "Allow SETTINGS after FORMAT, but note, that this is not always safe (note: this is a compatibility setting).", 0) \
|
||||
M(Seconds, periodic_live_view_refresh, 60, "Interval after which periodically refreshed live view is forced to refresh.", 0) \
|
||||
M(Bool, transform_null_in, false, "If enabled, NULL values will be matched with 'IN' operator as if they are considered equal.", 0) \
|
||||
@ -706,7 +707,6 @@ class IColumn;
|
||||
M(Bool, engine_file_skip_empty_files, false, "Allows to skip empty files in file table engine", 0) \
|
||||
M(Bool, engine_url_skip_empty_files, false, "Allows to skip empty files in url table engine", 0) \
|
||||
M(Bool, enable_url_encoding, true, " Allows to enable/disable decoding/encoding path in uri in URL table engine", 0) \
|
||||
M(Bool, allow_experimental_database_replicated, false, "Allow to create databases with Replicated engine", 0) \
|
||||
M(UInt64, database_replicated_initial_query_timeout_sec, 300, "How long initial DDL query should wait for Replicated database to precess previous DDL queue entries", 0) \
|
||||
M(Bool, database_replicated_enforce_synchronous_settings, false, "Enforces synchronous waiting for some queries (see also database_atomic_wait_for_drop_and_detach_synchronously, mutation_sync, alter_sync). Not recommended to enable these settings.", 0) \
|
||||
M(UInt64, max_distributed_depth, 5, "Maximum distributed query depth", 0) \
|
||||
@ -739,6 +739,7 @@ class IColumn;
|
||||
M(Bool, query_plan_split_filter, true, "Allow to split filters in the query plan", 0) \
|
||||
M(Bool, query_plan_merge_expressions, true, "Allow to merge expressions in the query plan", 0) \
|
||||
M(Bool, query_plan_filter_push_down, true, "Allow to push down filter by predicate query plan step", 0) \
|
||||
M(Bool, query_plan_convert_outer_join_to_inner_join, true, "Allow to convert OUTER JOIN to INNER JOIN if filter after JOIN always filters default values", 0) \
|
||||
M(Bool, query_plan_optimize_prewhere, true, "Allow to push down filter to PREWHERE expression for supported storages", 0) \
|
||||
M(Bool, query_plan_execute_functions_after_sorting, true, "Allow to re-order functions after sorting", 0) \
|
||||
M(Bool, query_plan_reuse_storage_ordering_for_window_functions, true, "Allow to use the storage sorting for window functions", 0) \
|
||||
@ -933,6 +934,7 @@ class IColumn;
|
||||
MAKE_OBSOLETE(M, Bool, allow_experimental_query_cache, true) \
|
||||
MAKE_OBSOLETE(M, Bool, allow_experimental_alter_materialized_view_structure, true) \
|
||||
MAKE_OBSOLETE(M, Bool, allow_experimental_shared_merge_tree, true) \
|
||||
MAKE_OBSOLETE(M, Bool, allow_experimental_database_replicated, true) \
|
||||
\
|
||||
MAKE_OBSOLETE(M, Milliseconds, async_insert_stale_timeout_ms, 0) \
|
||||
MAKE_OBSOLETE(M, StreamingHandleErrorMode, handle_kafka_error_mode, StreamingHandleErrorMode::DEFAULT) \
|
||||
@ -977,6 +979,7 @@ class IColumn;
|
||||
MAKE_OBSOLETE(M, Bool, allow_experimental_undrop_table_query, true) \
|
||||
MAKE_OBSOLETE(M, Bool, allow_experimental_s3queue, true) \
|
||||
MAKE_OBSOLETE(M, Bool, query_plan_optimize_primary_key, true) \
|
||||
MAKE_OBSOLETE(M, Bool, optimize_monotonous_functions_in_order_by, false) \
|
||||
|
||||
/** The section above is for obsolete settings. Do not add anything there. */
|
||||
|
||||
|
||||
@ -92,9 +92,13 @@ static std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges> sett
|
||||
{"input_format_json_ignore_unnecessary_fields", false, true, "Ignore unnecessary fields and not parse them. Enabling this may not throw exceptions on json strings of invalid format or with duplicated fields"},
|
||||
{"input_format_hive_text_allow_variable_number_of_columns", false, true, "Ignore extra columns in Hive Text input (if file has more columns than expected) and treat missing fields in Hive Text input as default values."},
|
||||
{"first_day_of_week", "Monday", "Monday", "Added a setting for the first day of the week for date/time functions"},
|
||||
{"allow_experimental_database_replicated", false, true, "Database engine Replicated is now in Beta stage"},
|
||||
{"temporary_data_in_cache_reserve_space_wait_lock_timeout_milliseconds", (10 * 60 * 1000), (10 * 60 * 1000), "Wait time to lock cache for sapce reservation in temporary data in filesystem cache"},
|
||||
{"optimize_rewrite_sum_if_to_count_if", false, true, "Only available for the analyzer, where it works correctly"}
|
||||
}},
|
||||
{"azure_allow_parallel_part_upload", "true", "true", "Use multiple threads for azure multipart upload."},
|
||||
{"max_recursive_cte_evaluation_depth", DBMS_RECURSIVE_CTE_MAX_EVALUATION_DEPTH, DBMS_RECURSIVE_CTE_MAX_EVALUATION_DEPTH, "Maximum limit on recursive CTE evaluation depth"},
|
||||
{"query_plan_convert_outer_join_to_inner_join", false, true, "Allow to convert OUTER JOIN to INNER JOIN if filter after JOIN always filters default values"},
|
||||
}},
|
||||
{"24.3", {{"s3_connect_timeout_ms", 1000, 1000, "Introduce new dedicated setting for s3 connection timeout"},
|
||||
{"allow_experimental_shared_merge_tree", false, true, "The setting is obsolete"},
|
||||
{"use_page_cache_for_disks_without_file_cache", false, false, "Added userspace page cache"},
|
||||
|
||||
@ -1139,8 +1139,10 @@ void DatabaseReplicated::recoverLostReplica(const ZooKeeperPtr & current_zookeep
|
||||
}
|
||||
LOG_INFO(log, "All tables are created successfully");
|
||||
|
||||
chassert(max_log_ptr_at_creation || our_log_ptr);
|
||||
UInt32 first_entry_to_mark_finished = new_replica ? max_log_ptr_at_creation : our_log_ptr;
|
||||
/// NOTE first_entry_to_mark_finished can be 0 if our replica has crashed just after creating its nodes in ZK,
|
||||
/// so it's a new replica, but after restarting we don't know max_log_ptr_at_creation anymore...
|
||||
/// It's a very rare case, and it's okay if some queries throw TIMEOUT_EXCEEDED when waiting for all replicas
|
||||
if (first_entry_to_mark_finished)
|
||||
{
|
||||
/// If the replica is new and some of the queries applied during recovery
|
||||
|
||||
@ -154,6 +154,7 @@ void IDisk::copyThroughBuffers(
|
||||
/// Disable parallel write. We already copy in parallel.
|
||||
/// Avoid high memory usage. See test_s3_zero_copy_ttl/test.py::test_move_and_s3_memory_usage
|
||||
write_settings.s3_allow_parallel_part_upload = false;
|
||||
write_settings.azure_allow_parallel_part_upload = false;
|
||||
|
||||
asyncCopy(*this, from_path, *to_disk, to_path, copying_thread_pool, results, copy_root_dir, read_settings, write_settings, cancellation_hook);
|
||||
|
||||
|
||||
@ -282,12 +282,17 @@ std::unique_ptr<WriteBufferFromFileBase> AzureObjectStorage::writeObject( /// NO
|
||||
|
||||
LOG_TEST(log, "Writing file: {}", object.remote_path);
|
||||
|
||||
ThreadPoolCallbackRunnerUnsafe<void> scheduler;
|
||||
if (write_settings.azure_allow_parallel_part_upload)
|
||||
scheduler = threadPoolCallbackRunnerUnsafe<void>(getThreadPoolWriter(), "VFSWrite");
|
||||
|
||||
return std::make_unique<WriteBufferFromAzureBlobStorage>(
|
||||
client.get(),
|
||||
object.remote_path,
|
||||
buf_size,
|
||||
patchSettings(write_settings),
|
||||
settings.get());
|
||||
settings.get(),
|
||||
std::move(scheduler));
|
||||
}
|
||||
|
||||
void AzureObjectStorage::removeObjectImpl(const StoredObject & object, const SharedAzureClientPtr & client_ptr, bool if_exists)
|
||||
|
||||
@ -1169,7 +1169,10 @@ struct ToStartOfHourImpl
|
||||
struct ToYearImpl
|
||||
{
|
||||
static constexpr auto name = "toYear";
|
||||
|
||||
static UInt16 execute(UInt64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toYear(t);
|
||||
}
|
||||
static UInt16 execute(Int64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toYear(t);
|
||||
@ -1217,7 +1220,10 @@ struct ToWeekYearImpl
|
||||
static constexpr auto name = "toWeekYear";
|
||||
|
||||
static constexpr Int8 week_mode = 3;
|
||||
|
||||
static UInt16 execute(UInt64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toYearWeek(t, week_mode).first;
|
||||
}
|
||||
static UInt16 execute(Int64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toYearWeek(t, week_mode).first;
|
||||
@ -1241,7 +1247,10 @@ struct ToWeekYearImpl
|
||||
struct ToWeekOfWeekYearImpl
|
||||
{
|
||||
static constexpr auto name = "toWeekOfWeekYear";
|
||||
|
||||
static UInt16 execute(UInt64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toISOWeek(t);
|
||||
}
|
||||
static UInt16 execute(Int64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toISOWeek(t);
|
||||
@ -1265,7 +1274,10 @@ struct ToWeekOfWeekYearImpl
|
||||
struct ToQuarterImpl
|
||||
{
|
||||
static constexpr auto name = "toQuarter";
|
||||
|
||||
static UInt8 execute(UInt64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toQuarter(t);
|
||||
}
|
||||
static UInt8 execute(Int64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toQuarter(t);
|
||||
@ -1290,7 +1302,10 @@ struct ToQuarterImpl
|
||||
struct ToMonthImpl
|
||||
{
|
||||
static constexpr auto name = "toMonth";
|
||||
|
||||
static UInt8 execute(UInt64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toMonth(t);
|
||||
}
|
||||
static UInt8 execute(Int64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toMonth(t);
|
||||
@ -1315,7 +1330,10 @@ struct ToMonthImpl
|
||||
struct ToDayOfMonthImpl
|
||||
{
|
||||
static constexpr auto name = "toDayOfMonth";
|
||||
|
||||
static UInt8 execute(UInt64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toDayOfMonth(t);
|
||||
}
|
||||
static UInt8 execute(Int64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toDayOfMonth(t);
|
||||
@ -1341,7 +1359,10 @@ struct ToDayOfWeekImpl
|
||||
{
|
||||
static constexpr auto name = "toDayOfWeek";
|
||||
static constexpr bool value_may_be_string = true;
|
||||
|
||||
static UInt8 execute(UInt64 t, UInt8 mode, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toDayOfWeek(t, mode);
|
||||
}
|
||||
static UInt8 execute(Int64 t, UInt8 mode, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toDayOfWeek(t, mode);
|
||||
@ -1365,7 +1386,10 @@ struct ToDayOfWeekImpl
|
||||
struct ToDayOfYearImpl
|
||||
{
|
||||
static constexpr auto name = "toDayOfYear";
|
||||
|
||||
static UInt16 execute(UInt64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toDayOfYear(t);
|
||||
}
|
||||
static UInt16 execute(Int64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toDayOfYear(t);
|
||||
@ -1421,7 +1445,10 @@ public:
|
||||
struct ToHourImpl
|
||||
{
|
||||
static constexpr auto name = "toHour";
|
||||
|
||||
static UInt8 execute(UInt64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toHour(t);
|
||||
}
|
||||
static UInt8 execute(Int64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toHour(t);
|
||||
@ -1446,7 +1473,10 @@ struct ToHourImpl
|
||||
struct TimezoneOffsetImpl
|
||||
{
|
||||
static constexpr auto name = "timezoneOffset";
|
||||
|
||||
static time_t execute(UInt64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.timezoneOffset(t);
|
||||
}
|
||||
static time_t execute(Int64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.timezoneOffset(t);
|
||||
@ -1474,7 +1504,10 @@ struct TimezoneOffsetImpl
|
||||
struct ToMinuteImpl
|
||||
{
|
||||
static constexpr auto name = "toMinute";
|
||||
|
||||
static UInt8 execute(UInt64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toMinute(t);
|
||||
}
|
||||
static UInt8 execute(Int64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toMinute(t);
|
||||
@ -1499,7 +1532,10 @@ struct ToMinuteImpl
|
||||
struct ToSecondImpl
|
||||
{
|
||||
static constexpr auto name = "toSecond";
|
||||
|
||||
static UInt8 execute(UInt64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toSecond(t);
|
||||
}
|
||||
static UInt8 execute(Int64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toSecond(t);
|
||||
@ -1550,7 +1586,10 @@ struct ToMillisecondImpl
|
||||
struct ToISOYearImpl
|
||||
{
|
||||
static constexpr auto name = "toISOYear";
|
||||
|
||||
static UInt16 execute(UInt64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toISOYear(time_zone.toDayNum(t));
|
||||
}
|
||||
static UInt16 execute(Int64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toISOYear(time_zone.toDayNum(t));
|
||||
@ -1607,7 +1646,10 @@ struct ToStartOfISOYearImpl
|
||||
struct ToISOWeekImpl
|
||||
{
|
||||
static constexpr auto name = "toISOWeek";
|
||||
|
||||
static UInt8 execute(UInt64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toISOWeek(time_zone.toDayNum(t));
|
||||
}
|
||||
static UInt8 execute(Int64 t, const DateLUTImpl & time_zone)
|
||||
{
|
||||
return time_zone.toISOWeek(time_zone.toDayNum(t));
|
||||
|
||||
@ -62,8 +62,8 @@ template <> struct InstructionValueTypeMap<DataTypeInt16> { using Instructi
|
||||
template <> struct InstructionValueTypeMap<DataTypeUInt16> { using InstructionValueType = UInt32; };
|
||||
template <> struct InstructionValueTypeMap<DataTypeInt32> { using InstructionValueType = UInt32; };
|
||||
template <> struct InstructionValueTypeMap<DataTypeUInt32> { using InstructionValueType = UInt32; };
|
||||
template <> struct InstructionValueTypeMap<DataTypeInt64> { using InstructionValueType = UInt32; };
|
||||
template <> struct InstructionValueTypeMap<DataTypeUInt64> { using InstructionValueType = UInt32; };
|
||||
template <> struct InstructionValueTypeMap<DataTypeInt64> { using InstructionValueType = Int64; };
|
||||
template <> struct InstructionValueTypeMap<DataTypeUInt64> { using InstructionValueType = UInt64; };
|
||||
template <> struct InstructionValueTypeMap<DataTypeDate> { using InstructionValueType = UInt16; };
|
||||
template <> struct InstructionValueTypeMap<DataTypeDate32> { using InstructionValueType = Int32; };
|
||||
template <> struct InstructionValueTypeMap<DataTypeDateTime> { using InstructionValueType = UInt32; };
|
||||
@ -1017,7 +1017,7 @@ public:
|
||||
else
|
||||
{
|
||||
for (auto & instruction : instructions)
|
||||
instruction.perform(pos, static_cast<UInt32>(vec[i]), 0, 0, *time_zone);
|
||||
instruction.perform(pos, static_cast<T>(vec[i]), 0, 0, *time_zone);
|
||||
}
|
||||
*pos++ = '\0';
|
||||
|
||||
@ -1073,7 +1073,7 @@ public:
|
||||
{
|
||||
/// DateTime/DateTime64 --> insert instruction
|
||||
/// Other types cannot provide the requested data --> write out template
|
||||
if constexpr (is_any_of<T, UInt32, Int64>)
|
||||
if constexpr (is_any_of<T, UInt32, Int64, UInt64>)
|
||||
{
|
||||
Instruction<T> instruction;
|
||||
instruction.setMysqlFunc(std::move(func));
|
||||
@ -1539,7 +1539,7 @@ public:
|
||||
/// If the argument was DateTime, add instruction for printing. If it was date, just append default literal
|
||||
auto add_instruction = [&]([[maybe_unused]] typename Instruction<T>::FuncJoda && func, [[maybe_unused]] const String & default_literal)
|
||||
{
|
||||
if constexpr (is_any_of<T, UInt32, Int64>)
|
||||
if constexpr (is_any_of<T, UInt32, Int64, UInt64>)
|
||||
{
|
||||
Instruction<T> instruction;
|
||||
instruction.setJodaFunc(std::move(func));
|
||||
|
||||
@ -23,6 +23,7 @@ struct WriteSettings
|
||||
size_t filesystem_cache_reserve_space_wait_lock_timeout_milliseconds = 1000;
|
||||
|
||||
bool s3_allow_parallel_part_upload = true;
|
||||
bool azure_allow_parallel_part_upload = true;
|
||||
|
||||
/// Monitoring
|
||||
bool for_object_storage = false; // to choose which profile events should be incremented
|
||||
|
||||
@ -109,4 +109,9 @@ void copyDataWithThrottler(ReadBuffer & from, WriteBuffer & to, size_t bytes, co
|
||||
copyDataImpl(from, to, true, bytes, &is_cancelled, throttler);
|
||||
}
|
||||
|
||||
void copyDataWithThrottler(ReadBuffer & from, WriteBuffer & to, std::function<void()> cancellation_hook, ThrottlerPtr throttler)
|
||||
{
|
||||
copyDataImpl(from, to, false, std::numeric_limits<size_t>::max(), cancellation_hook, throttler);
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
@ -33,5 +33,6 @@ void copyDataMaxBytes(ReadBuffer & from, WriteBuffer & to, size_t max_bytes);
|
||||
/// Same as above but also use throttler to limit maximum speed
|
||||
void copyDataWithThrottler(ReadBuffer & from, WriteBuffer & to, const std::atomic<int> & is_cancelled, ThrottlerPtr throttler);
|
||||
void copyDataWithThrottler(ReadBuffer & from, WriteBuffer & to, size_t bytes, const std::atomic<int> & is_cancelled, ThrottlerPtr throttler);
|
||||
void copyDataWithThrottler(ReadBuffer & from, WriteBuffer & to, std::function<void()> cancellation_hook, ThrottlerPtr throttler);
|
||||
|
||||
}
|
||||
|
||||
@ -2013,6 +2013,63 @@ ActionsDAG::SplitResult ActionsDAG::splitActionsBySortingDescription(const NameS
|
||||
return res;
|
||||
}
|
||||
|
||||
bool ActionsDAG::isFilterAlwaysFalseForDefaultValueInputs(const std::string & filter_name, const Block & input_stream_header)
|
||||
{
|
||||
const auto * filter_node = tryFindInOutputs(filter_name);
|
||||
if (!filter_node)
|
||||
throw Exception(ErrorCodes::LOGICAL_ERROR,
|
||||
"Outputs for ActionsDAG does not contain filter column name {}. DAG:\n{}",
|
||||
filter_name,
|
||||
dumpDAG());
|
||||
|
||||
std::unordered_map<std::string, ColumnWithTypeAndName> input_node_name_to_default_input_column;
|
||||
|
||||
for (const auto * input : inputs)
|
||||
{
|
||||
if (!input_stream_header.has(input->result_name))
|
||||
continue;
|
||||
|
||||
if (input->column)
|
||||
continue;
|
||||
|
||||
auto constant_column = input->result_type->createColumnConst(0, input->result_type->getDefault());
|
||||
auto constant_column_with_type_and_name = ColumnWithTypeAndName{constant_column, input->result_type, input->result_name};
|
||||
input_node_name_to_default_input_column.emplace(input->result_name, std::move(constant_column_with_type_and_name));
|
||||
}
|
||||
|
||||
ActionsDAGPtr filter_with_default_value_inputs;
|
||||
|
||||
try
|
||||
{
|
||||
filter_with_default_value_inputs = buildFilterActionsDAG({filter_node}, input_node_name_to_default_input_column);
|
||||
}
|
||||
catch (const Exception &)
|
||||
{
|
||||
/** It is possible that duing DAG construction, some functions cannot be executed for constant default value inputs
|
||||
* and exception will be thrown.
|
||||
*/
|
||||
return false;
|
||||
}
|
||||
|
||||
const auto * filter_with_default_value_inputs_filter_node = filter_with_default_value_inputs->getOutputs()[0];
|
||||
if (!filter_with_default_value_inputs_filter_node->column || !isColumnConst(*filter_with_default_value_inputs_filter_node->column))
|
||||
return false;
|
||||
|
||||
const auto & constant_type = filter_with_default_value_inputs_filter_node->result_type;
|
||||
auto which_constant_type = WhichDataType(constant_type);
|
||||
if (!which_constant_type.isUInt8() && !which_constant_type.isNothing())
|
||||
return false;
|
||||
|
||||
Field value;
|
||||
filter_with_default_value_inputs_filter_node->column->get(0, value);
|
||||
|
||||
if (value.isNull())
|
||||
return true;
|
||||
|
||||
UInt8 predicate_value = value.safeGet<UInt8>();
|
||||
return predicate_value == 0;
|
||||
}
|
||||
|
||||
ActionsDAG::SplitResult ActionsDAG::splitActionsForFilter(const std::string & column_name) const
|
||||
{
|
||||
const auto * node = tryFindInOutputs(column_name);
|
||||
|
||||
@ -355,6 +355,13 @@ public:
|
||||
/// The second contains the rest.
|
||||
SplitResult splitActionsBySortingDescription(const NameSet & sort_columns) const;
|
||||
|
||||
/** Returns true if filter DAG is always false for inputs with default values.
|
||||
*
|
||||
* @param filter_name - name of filter node in current DAG.
|
||||
* @param input_stream_header - input stream header.
|
||||
*/
|
||||
bool isFilterAlwaysFalseForDefaultValueInputs(const std::string & filter_name, const Block & input_stream_header);
|
||||
|
||||
/// Create actions which may calculate part of filter using only available_inputs.
|
||||
/// If nothing may be calculated, returns nullptr.
|
||||
/// Otherwise, return actions which inputs are from available_inputs.
|
||||
|
||||
@ -82,6 +82,7 @@ void EvictionCandidates::removeQueueEntries(const CachePriorityGuard::Lock & loc
|
||||
auto queue_iterator = candidate->getQueueIterator();
|
||||
queue_iterator->invalidate();
|
||||
|
||||
chassert(candidate->releasable());
|
||||
candidate->file_segment->resetQueueIterator();
|
||||
/// We need to set removed flag in file segment metadata,
|
||||
/// because in dynamic cache resize we first remove queue entries,
|
||||
@ -122,7 +123,13 @@ void EvictionCandidates::evict()
|
||||
while (!key_candidates.candidates.empty())
|
||||
{
|
||||
auto & candidate = key_candidates.candidates.back();
|
||||
chassert(candidate->releasable());
|
||||
if (!candidate->releasable())
|
||||
{
|
||||
throw Exception(ErrorCodes::LOGICAL_ERROR,
|
||||
"Eviction candidate is not releasable: {} (evicting or removed flag: {})",
|
||||
candidate->file_segment->getInfoForLog(), candidate->isEvictingOrRemoved(*locked_key));
|
||||
}
|
||||
|
||||
const auto segment = candidate->file_segment;
|
||||
|
||||
IFileCachePriority::IteratorPtr iterator;
|
||||
|
||||
@ -118,7 +118,7 @@ LockedKeyPtr KeyMetadata::lockNoStateCheck()
|
||||
return std::make_unique<LockedKey>(shared_from_this());
|
||||
}
|
||||
|
||||
bool KeyMetadata::createBaseDirectory()
|
||||
bool KeyMetadata::createBaseDirectory(bool throw_if_failed)
|
||||
{
|
||||
if (!created_base_directory.exchange(true))
|
||||
{
|
||||
@ -131,7 +131,7 @@ bool KeyMetadata::createBaseDirectory()
|
||||
{
|
||||
created_base_directory = false;
|
||||
|
||||
if (e.code() == std::errc::no_space_on_device)
|
||||
if (!throw_if_failed && e.code() == std::errc::no_space_on_device)
|
||||
{
|
||||
LOG_TRACE(cache_metadata->log, "Failed to create base directory for key {}, "
|
||||
"because no space left on device", key);
|
||||
|
||||
@ -113,7 +113,7 @@ struct KeyMetadata : private std::map<size_t, FileSegmentMetadataPtr>,
|
||||
|
||||
LockedKeyPtr tryLock();
|
||||
|
||||
bool createBaseDirectory();
|
||||
bool createBaseDirectory(bool throw_if_failed = false);
|
||||
|
||||
std::string getPath() const;
|
||||
|
||||
|
||||
@ -54,10 +54,18 @@ WriteBufferToFileSegment::WriteBufferToFileSegment(FileSegmentsHolderPtr segment
|
||||
void WriteBufferToFileSegment::nextImpl()
|
||||
{
|
||||
auto downloader [[maybe_unused]] = file_segment->getOrSetDownloader();
|
||||
chassert(downloader == FileSegment::getCallerId());
|
||||
if (downloader != FileSegment::getCallerId())
|
||||
{
|
||||
throw Exception(ErrorCodes::LOGICAL_ERROR,
|
||||
"Failed to set a downloader (current downloader: {}, file segment info: {})",
|
||||
downloader, file_segment->getInfoForLog());
|
||||
}
|
||||
|
||||
SCOPE_EXIT({
|
||||
file_segment->completePartAndResetDownloader();
|
||||
if (file_segment->isDownloader())
|
||||
file_segment->completePartAndResetDownloader();
|
||||
else
|
||||
chassert(false);
|
||||
});
|
||||
|
||||
size_t bytes_to_write = offset();
|
||||
|
||||
@ -245,11 +245,15 @@ void executeQuery(
|
||||
const auto & shard_info = cluster->getShardsInfo()[i];
|
||||
|
||||
auto query_for_shard = query_info.query_tree->clone();
|
||||
if (sharding_key_expr && query_info.optimized_cluster && settings.optimize_skip_unused_shards_rewrite_in && shards > 1)
|
||||
if (sharding_key_expr &&
|
||||
query_info.optimized_cluster &&
|
||||
settings.optimize_skip_unused_shards_rewrite_in &&
|
||||
shards > 1 &&
|
||||
/// TODO: support composite sharding key
|
||||
sharding_key_expr->getRequiredColumns().size() == 1)
|
||||
{
|
||||
OptimizeShardingKeyRewriteInVisitor::Data visitor_data{
|
||||
sharding_key_expr,
|
||||
sharding_key_expr->getSampleBlock().getByPosition(0).type,
|
||||
sharding_key_column_name,
|
||||
shard_info,
|
||||
not_optimized_cluster->getSlotToShard(),
|
||||
@ -282,11 +286,15 @@ void executeQuery(
|
||||
const auto & shard_info = cluster->getShardsInfo()[i];
|
||||
|
||||
ASTPtr query_ast_for_shard = query_info.query->clone();
|
||||
if (sharding_key_expr && query_info.optimized_cluster && settings.optimize_skip_unused_shards_rewrite_in && shards > 1)
|
||||
if (sharding_key_expr &&
|
||||
query_info.optimized_cluster &&
|
||||
settings.optimize_skip_unused_shards_rewrite_in &&
|
||||
shards > 1 &&
|
||||
/// TODO: support composite sharding key
|
||||
sharding_key_expr->getRequiredColumns().size() == 1)
|
||||
{
|
||||
OptimizeShardingKeyRewriteInVisitor::Data visitor_data{
|
||||
sharding_key_expr,
|
||||
sharding_key_expr->getSampleBlock().getByPosition(0).type,
|
||||
sharding_key_column_name,
|
||||
shard_info,
|
||||
not_optimized_cluster->getSlotToShard(),
|
||||
|
||||
@ -1618,6 +1618,33 @@ void Context::addExternalTable(const String & table_name, TemporaryTableHolder &
|
||||
external_tables_mapping.emplace(table_name, std::make_shared<TemporaryTableHolder>(std::move(temporary_table)));
|
||||
}
|
||||
|
||||
void Context::updateExternalTable(const String & table_name, TemporaryTableHolder && temporary_table)
|
||||
{
|
||||
if (isGlobalContext())
|
||||
throw Exception(ErrorCodes::LOGICAL_ERROR, "Global context cannot have external tables");
|
||||
|
||||
auto temporary_table_ptr = std::make_shared<TemporaryTableHolder>(std::move(temporary_table));
|
||||
|
||||
std::lock_guard lock(mutex);
|
||||
auto it = external_tables_mapping.find(table_name);
|
||||
if (it == external_tables_mapping.end())
|
||||
throw Exception(ErrorCodes::TABLE_ALREADY_EXISTS, "Temporary table {} does not exists", backQuoteIfNeed(table_name));
|
||||
it->second = std::move(temporary_table_ptr);
|
||||
}
|
||||
|
||||
void Context::addOrUpdateExternalTable(const String & table_name, TemporaryTableHolder && temporary_table)
|
||||
{
|
||||
if (isGlobalContext())
|
||||
throw Exception(ErrorCodes::LOGICAL_ERROR, "Global context cannot have external tables");
|
||||
|
||||
auto temporary_table_ptr = std::make_shared<TemporaryTableHolder>(std::move(temporary_table));
|
||||
|
||||
std::lock_guard lock(mutex);
|
||||
auto [it, inserted] = external_tables_mapping.emplace(table_name, temporary_table_ptr);
|
||||
if (!inserted)
|
||||
it->second = std::move(temporary_table_ptr);
|
||||
}
|
||||
|
||||
std::shared_ptr<TemporaryTableHolder> Context::findExternalTable(const String & table_name) const
|
||||
{
|
||||
if (isGlobalContext())
|
||||
@ -5256,6 +5283,7 @@ WriteSettings Context::getWriteSettings() const
|
||||
res.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds = settings.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds;
|
||||
|
||||
res.s3_allow_parallel_part_upload = settings.s3_allow_parallel_part_upload;
|
||||
res.azure_allow_parallel_part_upload = settings.azure_allow_parallel_part_upload;
|
||||
|
||||
res.remote_throttler = getRemoteWriteThrottler();
|
||||
res.local_throttler = getLocalWriteThrottler();
|
||||
|
||||
@ -677,6 +677,8 @@ public:
|
||||
|
||||
Tables getExternalTables() const;
|
||||
void addExternalTable(const String & table_name, TemporaryTableHolder && temporary_table);
|
||||
void updateExternalTable(const String & table_name, TemporaryTableHolder && temporary_table);
|
||||
void addOrUpdateExternalTable(const String & table_name, TemporaryTableHolder && temporary_table);
|
||||
std::shared_ptr<TemporaryTableHolder> findExternalTable(const String & table_name) const;
|
||||
std::shared_ptr<TemporaryTableHolder> removeExternalTable(const String & table_name);
|
||||
|
||||
|
||||
@ -113,8 +113,10 @@ struct TemporaryTableHolder : boost::noncopyable, WithContext
|
||||
FutureSetFromSubqueryPtr future_set;
|
||||
};
|
||||
|
||||
using TemporaryTableHolderPtr = std::shared_ptr<TemporaryTableHolder>;
|
||||
|
||||
///TODO maybe remove shared_ptr from here?
|
||||
using TemporaryTablesMapping = std::map<String, std::shared_ptr<TemporaryTableHolder>>;
|
||||
using TemporaryTablesMapping = std::map<String, TemporaryTableHolderPtr>;
|
||||
|
||||
class BackgroundSchedulePoolTaskHolder;
|
||||
|
||||
|
||||
@ -31,8 +31,21 @@ public:
|
||||
}
|
||||
|
||||
std::string getName() const override { return "FullSortingMergeJoin"; }
|
||||
|
||||
const TableJoin & getTableJoin() const override { return *table_join; }
|
||||
|
||||
bool isCloneSupported() const override
|
||||
{
|
||||
return true;
|
||||
}
|
||||
|
||||
std::shared_ptr<IJoin> clone(const std::shared_ptr<TableJoin> & table_join_,
|
||||
const Block &,
|
||||
const Block & right_sample_block_) const override
|
||||
{
|
||||
return std::make_shared<FullSortingMergeJoin>(table_join_, right_sample_block_, null_direction);
|
||||
}
|
||||
|
||||
int getNullDirection() const { return null_direction; }
|
||||
|
||||
bool addBlockToJoin(const Block & /* block */, bool /* check_limits */) override
|
||||
|
||||
@ -236,11 +236,13 @@ static void correctNullabilityInplace(ColumnWithTypeAndName & column, bool nulla
|
||||
}
|
||||
|
||||
HashJoin::HashJoin(std::shared_ptr<TableJoin> table_join_, const Block & right_sample_block_,
|
||||
bool any_take_last_row_, size_t reserve_num, const String & instance_id_)
|
||||
bool any_take_last_row_, size_t reserve_num_, const String & instance_id_)
|
||||
: table_join(table_join_)
|
||||
, kind(table_join->kind())
|
||||
, strictness(table_join->strictness())
|
||||
, any_take_last_row(any_take_last_row_)
|
||||
, reserve_num(reserve_num_)
|
||||
, instance_id(instance_id_)
|
||||
, asof_inequality(table_join->getAsofInequality())
|
||||
, data(std::make_shared<RightTableData>())
|
||||
, right_sample_block(right_sample_block_)
|
||||
@ -325,7 +327,7 @@ HashJoin::HashJoin(std::shared_ptr<TableJoin> table_join_, const Block & right_s
|
||||
}
|
||||
|
||||
for (auto & maps : data->maps)
|
||||
dataMapInit(maps, reserve_num);
|
||||
dataMapInit(maps);
|
||||
}
|
||||
|
||||
HashJoin::Type HashJoin::chooseMethod(JoinKind kind, const ColumnRawPtrs & key_columns, Sizes & key_sizes)
|
||||
@ -487,9 +489,8 @@ struct KeyGetterForType
|
||||
using Type = typename KeyGetterForTypeImpl<type, Value, Mapped>::Type;
|
||||
};
|
||||
|
||||
void HashJoin::dataMapInit(MapsVariant & map, size_t reserve_num)
|
||||
void HashJoin::dataMapInit(MapsVariant & map)
|
||||
{
|
||||
|
||||
if (kind == JoinKind::Cross)
|
||||
return;
|
||||
joinDispatchInit(kind, strictness, map);
|
||||
|
||||
@ -148,13 +148,26 @@ class HashJoin : public IJoin
|
||||
public:
|
||||
HashJoin(
|
||||
std::shared_ptr<TableJoin> table_join_, const Block & right_sample_block,
|
||||
bool any_take_last_row_ = false, size_t reserve_num = 0, const String & instance_id_ = "");
|
||||
bool any_take_last_row_ = false, size_t reserve_num_ = 0, const String & instance_id_ = "");
|
||||
|
||||
~HashJoin() override;
|
||||
|
||||
std::string getName() const override { return "HashJoin"; }
|
||||
|
||||
const TableJoin & getTableJoin() const override { return *table_join; }
|
||||
|
||||
bool isCloneSupported() const override
|
||||
{
|
||||
return true;
|
||||
}
|
||||
|
||||
std::shared_ptr<IJoin> clone(const std::shared_ptr<TableJoin> & table_join_,
|
||||
const Block &,
|
||||
const Block & right_sample_block_) const override
|
||||
{
|
||||
return std::make_shared<HashJoin>(table_join_, right_sample_block_, any_take_last_row, reserve_num, instance_id);
|
||||
}
|
||||
|
||||
/** Add block of data from right hand of JOIN to the map.
|
||||
* Returns false, if some limit was exceeded and you should not insert more data.
|
||||
*/
|
||||
@ -412,7 +425,9 @@ private:
|
||||
/// This join was created from StorageJoin and it is already filled.
|
||||
bool from_storage_join = false;
|
||||
|
||||
bool any_take_last_row; /// Overwrite existing values when encountering the same key again
|
||||
const bool any_take_last_row; /// Overwrite existing values when encountering the same key again
|
||||
const size_t reserve_num;
|
||||
const String instance_id;
|
||||
std::optional<TypeIndex> asof_type;
|
||||
const ASOFJoinInequality asof_inequality;
|
||||
|
||||
@ -454,7 +469,7 @@ private:
|
||||
/// If set HashJoin instance is not available for modification (addBlockToJoin)
|
||||
TableLockHolder storage_join_lock = nullptr;
|
||||
|
||||
void dataMapInit(MapsVariant &, size_t);
|
||||
void dataMapInit(MapsVariant & map);
|
||||
|
||||
void initRightBlockStructure(Block & saved_block_sample);
|
||||
|
||||
|
||||
@ -11,6 +11,11 @@
|
||||
namespace DB
|
||||
{
|
||||
|
||||
namespace ErrorCodes
|
||||
{
|
||||
extern const int UNSUPPORTED_METHOD;
|
||||
}
|
||||
|
||||
struct ExtraBlock;
|
||||
using ExtraBlockPtr = std::shared_ptr<ExtraBlock>;
|
||||
|
||||
@ -52,6 +57,23 @@ public:
|
||||
|
||||
virtual const TableJoin & getTableJoin() const = 0;
|
||||
|
||||
/// Returns true if clone is supported
|
||||
virtual bool isCloneSupported() const
|
||||
{
|
||||
return false;
|
||||
}
|
||||
|
||||
/// Clone underlyhing JOIN algorithm using table join, left sample block, right sample block
|
||||
virtual std::shared_ptr<IJoin> clone(const std::shared_ptr<TableJoin> & table_join_,
|
||||
const Block & left_sample_block_,
|
||||
const Block & right_sample_block_) const
|
||||
{
|
||||
(void)(table_join_);
|
||||
(void)(left_sample_block_);
|
||||
(void)(right_sample_block_);
|
||||
throw Exception(ErrorCodes::UNSUPPORTED_METHOD, "Clone method is not supported for {}", getName());
|
||||
}
|
||||
|
||||
/// Add block of data from right hand of JOIN.
|
||||
/// @returns false, if some limit was exceeded and you should not insert more data.
|
||||
virtual bool addBlockToJoin(const Block & block, bool check_limits = true) = 0; /// NOLINT
|
||||
|
||||
@ -258,15 +258,6 @@ BlockIO InterpreterCreateQuery::createDatabase(ASTCreateQuery & create)
|
||||
"Enable allow_experimental_database_materialized_mysql to use it");
|
||||
}
|
||||
|
||||
if (create.storage->engine->name == "Replicated"
|
||||
&& !getContext()->getSettingsRef().allow_experimental_database_replicated
|
||||
&& !internal && !create.attach)
|
||||
{
|
||||
throw Exception(ErrorCodes::UNKNOWN_DATABASE_ENGINE,
|
||||
"Replicated is an experimental database engine. "
|
||||
"Enable allow_experimental_database_replicated to use it");
|
||||
}
|
||||
|
||||
if (create.storage->engine->name == "MaterializedPostgreSQL"
|
||||
&& !getContext()->getSettingsRef().allow_experimental_database_materialized_postgresql
|
||||
&& !internal && !create.attach)
|
||||
|
||||
@ -27,7 +27,9 @@ public:
|
||||
const StoragePtr & storage_,
|
||||
const SelectQueryOptions & select_query_options_);
|
||||
|
||||
/// Initialize interpreter with query tree
|
||||
/** Initialize interpreter with query tree.
|
||||
* No query tree passes are applied.
|
||||
*/
|
||||
InterpreterSelectQueryAnalyzer(const QueryTreeNodePtr & query_tree_,
|
||||
const ContextPtr & context_,
|
||||
const SelectQueryOptions & select_query_options_);
|
||||
|
||||
@ -508,6 +508,7 @@ static void validateUpdateColumns(
|
||||
/// because their sizes couldn't change, since sizes of all nested subcolumns must be consistent.
|
||||
static std::optional<std::vector<ASTPtr>> getExpressionsOfUpdatedNestedSubcolumns(
|
||||
const String & column_name,
|
||||
NameSet affected_materialized,
|
||||
const NamesAndTypesList & all_columns,
|
||||
const std::unordered_map<String, ASTPtr> & column_to_update_expression)
|
||||
{
|
||||
@ -520,6 +521,10 @@ static std::optional<std::vector<ASTPtr>> getExpressionsOfUpdatedNestedSubcolumn
|
||||
auto split = Nested::splitName(column.name);
|
||||
if (isArray(column.type) && split.first == source_name && !split.second.empty())
|
||||
{
|
||||
// Materialized nested columns shall never be part of the update expression
|
||||
if (affected_materialized.contains(column.name))
|
||||
continue;
|
||||
|
||||
auto it = column_to_update_expression.find(column.name);
|
||||
if (it == column_to_update_expression.end())
|
||||
return {};
|
||||
@ -655,7 +660,10 @@ void MutationsInterpreter::prepare(bool dry_run)
|
||||
if (materialized_it != column_to_affected_materialized.end())
|
||||
for (const auto & mat_column : materialized_it->second)
|
||||
affected_materialized.emplace(mat_column);
|
||||
}
|
||||
|
||||
for (const auto & [column_name, update_expr] : command.column_to_update_expression)
|
||||
{
|
||||
/// When doing UPDATE column = expression WHERE condition
|
||||
/// we will replace column to the result of the following expression:
|
||||
///
|
||||
@ -689,7 +697,7 @@ void MutationsInterpreter::prepare(bool dry_run)
|
||||
{
|
||||
std::shared_ptr<ASTFunction> function = nullptr;
|
||||
|
||||
auto nested_update_exprs = getExpressionsOfUpdatedNestedSubcolumns(column_name, all_columns, command.column_to_update_expression);
|
||||
auto nested_update_exprs = getExpressionsOfUpdatedNestedSubcolumns(column_name, affected_materialized, all_columns, command.column_to_update_expression);
|
||||
if (!nested_update_exprs)
|
||||
{
|
||||
function = makeASTFunction("validateNestedArraySizes",
|
||||
|
||||
@ -38,25 +38,27 @@ Field executeFunctionOnField(
|
||||
return (*ret.column)[0];
|
||||
}
|
||||
|
||||
/// @param sharding_column_value - one of values from IN
|
||||
/// @param column_value - one of values from IN
|
||||
/// @param sharding_column_name - name of that column
|
||||
/// @return true if shard may contain such value (or it is unknown), otherwise false.
|
||||
bool shardContains(
|
||||
Field sharding_column_value,
|
||||
Field column_value,
|
||||
const std::string & sharding_column_name,
|
||||
const OptimizeShardingKeyRewriteInMatcher::Data & data)
|
||||
{
|
||||
/// Type of column in storage (used for implicit conversion from i.e. String to Int)
|
||||
const DataTypePtr & column_type = data.sharding_key_expr->getSampleBlock().getByName(sharding_column_name).type;
|
||||
/// Implicit conversion.
|
||||
sharding_column_value = convertFieldToType(sharding_column_value, *data.sharding_key_type);
|
||||
column_value = convertFieldToType(column_value, *column_type);
|
||||
|
||||
/// NULL is not allowed in sharding key,
|
||||
/// so it should be safe to assume that shard cannot contain it.
|
||||
if (sharding_column_value.isNull())
|
||||
if (column_value.isNull())
|
||||
return false;
|
||||
|
||||
Field sharding_value = executeFunctionOnField(
|
||||
sharding_column_value, sharding_column_name,
|
||||
data.sharding_key_expr, data.sharding_key_type,
|
||||
column_value, sharding_column_name,
|
||||
data.sharding_key_expr, column_type,
|
||||
data.sharding_key_column_name);
|
||||
/// The value from IN can be non-numeric,
|
||||
/// but in this case it should be convertible to numeric type, let's try.
|
||||
|
||||
@ -28,8 +28,6 @@ struct OptimizeShardingKeyRewriteInMatcher
|
||||
{
|
||||
/// Expression of sharding_key for the Distributed() table
|
||||
const ExpressionActionsPtr & sharding_key_expr;
|
||||
/// Type of sharding_key column.
|
||||
const DataTypePtr & sharding_key_type;
|
||||
/// Name of the column for sharding_expr
|
||||
const std::string & sharding_key_column_name;
|
||||
/// Info for the current shard (to compare shard_num with calculated)
|
||||
|
||||
@ -56,6 +56,14 @@ struct SelectQueryOptions
|
||||
std::optional<UInt32> shard_num;
|
||||
std::optional<UInt32> shard_count;
|
||||
|
||||
/** During read from MergeTree parts will be removed from snapshot after they are not needed.
|
||||
* This optimization will break subsequent execution of the same query tree, because table node
|
||||
* will no more have valid snapshot.
|
||||
*
|
||||
* TODO: Implement this functionality in safer way
|
||||
*/
|
||||
bool merge_tree_enable_remove_parts_from_snapshot_optimization = true;
|
||||
|
||||
SelectQueryOptions( /// NOLINT(google-explicit-constructor)
|
||||
QueryProcessingStage::Enum stage = QueryProcessingStage::Complete,
|
||||
size_t depth = 0,
|
||||
|
||||
@ -243,7 +243,10 @@ public:
|
||||
table_join.strictness = strictness;
|
||||
}
|
||||
|
||||
TableJoin(const TableJoin & rhs) = default;
|
||||
|
||||
JoinKind kind() const { return table_join.kind; }
|
||||
void setKind(JoinKind kind) { table_join.kind = kind; }
|
||||
JoinStrictness strictness() const { return table_join.strictness; }
|
||||
bool sameStrictnessAndKind(JoinStrictness, JoinKind) const;
|
||||
const SizeLimits & sizeLimits() const { return size_limits; }
|
||||
|
||||
@ -111,7 +111,7 @@ FileSegmentsHolderPtr TemporaryDataOnDisk::createCacheFile(size_t max_file_size)
|
||||
CreateFileSegmentSettings(FileSegmentKind::Temporary, /* unbounded */ true), FileCache::getCommonUser());
|
||||
|
||||
chassert(holder->size() == 1);
|
||||
holder->back().getKeyMetadata()->createBaseDirectory();
|
||||
holder->back().getKeyMetadata()->createBaseDirectory(/* throw_if_failed */true);
|
||||
|
||||
return holder;
|
||||
}
|
||||
|
||||
@ -367,92 +367,6 @@ std::unordered_set<String> getDistinctNames(const ASTSelectQuery & select)
|
||||
return names;
|
||||
}
|
||||
|
||||
/// Replace monotonous functions in ORDER BY if they don't participate in GROUP BY expression,
|
||||
/// has a single argument and not an aggregate functions.
|
||||
void optimizeMonotonousFunctionsInOrderBy(ASTSelectQuery * select_query, ContextPtr context,
|
||||
const TablesWithColumns & tables_with_columns,
|
||||
const TreeRewriterResult & result)
|
||||
{
|
||||
auto order_by = select_query->orderBy();
|
||||
if (!order_by)
|
||||
return;
|
||||
|
||||
/// Do not apply optimization for Distributed and Merge storages,
|
||||
/// because we can't get the sorting key of their underlying tables
|
||||
/// and we can break the matching of the sorting key for `read_in_order`
|
||||
/// optimization by removing monotonous functions from the prefix of key.
|
||||
if (result.is_remote_storage || (result.storage && result.storage->getName() == "Merge"))
|
||||
return;
|
||||
|
||||
for (const auto & child : order_by->children)
|
||||
{
|
||||
auto * order_by_element = child->as<ASTOrderByElement>();
|
||||
|
||||
if (!order_by_element || order_by_element->children.empty())
|
||||
throw Exception(ErrorCodes::UNKNOWN_TYPE_OF_AST_NODE, "Bad ORDER BY expression AST");
|
||||
|
||||
if (order_by_element->with_fill)
|
||||
return;
|
||||
}
|
||||
|
||||
std::unordered_set<String> group_by_hashes;
|
||||
if (auto group_by = select_query->groupBy())
|
||||
{
|
||||
if (select_query->group_by_with_grouping_sets)
|
||||
{
|
||||
for (auto & set : group_by->children)
|
||||
{
|
||||
for (auto & elem : set->children)
|
||||
{
|
||||
const auto hash = elem->getTreeHash(/*ignore_aliases=*/ true);
|
||||
const auto key = toString(hash);
|
||||
group_by_hashes.insert(key);
|
||||
}
|
||||
}
|
||||
}
|
||||
else
|
||||
{
|
||||
for (auto & elem : group_by->children)
|
||||
{
|
||||
const auto hash = elem->getTreeHash(/*ignore_aliases=*/ true);
|
||||
const auto key = toString(hash);
|
||||
group_by_hashes.insert(key);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
auto sorting_key_columns = result.storage_snapshot ? result.storage_snapshot->metadata->getSortingKeyColumns() : Names{};
|
||||
|
||||
bool is_sorting_key_prefix = true;
|
||||
for (size_t i = 0; i < order_by->children.size(); ++i)
|
||||
{
|
||||
auto * order_by_element = order_by->children[i]->as<ASTOrderByElement>();
|
||||
|
||||
auto & ast_func = order_by_element->children[0];
|
||||
if (!ast_func->as<ASTFunction>())
|
||||
continue;
|
||||
|
||||
if (i >= sorting_key_columns.size() || ast_func->getColumnName() != sorting_key_columns[i])
|
||||
is_sorting_key_prefix = false;
|
||||
|
||||
/// If order by expression matches the sorting key, do not remove
|
||||
/// functions to allow execute reading in order of key.
|
||||
if (is_sorting_key_prefix)
|
||||
continue;
|
||||
|
||||
MonotonicityCheckVisitor::Data data{tables_with_columns, context, group_by_hashes};
|
||||
MonotonicityCheckVisitor(data).visit(ast_func);
|
||||
|
||||
if (!data.isRejected())
|
||||
{
|
||||
ast_func = data.identifier->clone();
|
||||
ast_func->setAlias("");
|
||||
if (!data.monotonicity.is_positive)
|
||||
order_by_element->direction *= -1;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/// If ORDER BY has argument x followed by f(x) transforms it to ORDER BY x.
|
||||
/// Optimize ORDER BY x, y, f(x), g(x, y), f(h(x)), t(f(x), g(x)) into ORDER BY x, y
|
||||
/// in case if f(), g(), h(), t() are deterministic (in scope of query).
|
||||
@ -779,10 +693,6 @@ void TreeOptimizer::apply(ASTPtr & query, TreeRewriterResult & result,
|
||||
if (settings.optimize_redundant_functions_in_order_by)
|
||||
optimizeRedundantFunctionsInOrderBy(select_query, context);
|
||||
|
||||
/// Replace monotonous functions with its argument
|
||||
if (settings.optimize_monotonous_functions_in_order_by)
|
||||
optimizeMonotonousFunctionsInOrderBy(select_query, context, tables_with_columns, result);
|
||||
|
||||
/// Remove duplicate items from ORDER BY.
|
||||
/// Execute it after all order by optimizations,
|
||||
/// because they can produce duplicated columns.
|
||||
|
||||
@ -44,6 +44,7 @@ ASTPtr ASTSelectQuery::clone() const
|
||||
|
||||
void ASTSelectQuery::updateTreeHashImpl(SipHash & hash_state, bool ignore_aliases) const
|
||||
{
|
||||
hash_state.update(recursive_with);
|
||||
hash_state.update(distinct);
|
||||
hash_state.update(group_by_with_totals);
|
||||
hash_state.update(group_by_with_rollup);
|
||||
@ -64,6 +65,10 @@ void ASTSelectQuery::formatImpl(const FormatSettings & s, FormatState & state, F
|
||||
if (with())
|
||||
{
|
||||
s.ostr << (s.hilite ? hilite_keyword : "") << indent_str << "WITH" << (s.hilite ? hilite_none : "");
|
||||
|
||||
if (recursive_with)
|
||||
s.ostr << (s.hilite ? hilite_keyword : "") << " RECURSIVE" << (s.hilite ? hilite_none : "");
|
||||
|
||||
s.one_line
|
||||
? with()->formatImpl(s, state, frame)
|
||||
: with()->as<ASTExpressionList &>().formatImplMultiline(s, state, frame);
|
||||
|
||||
@ -83,6 +83,7 @@ public:
|
||||
|
||||
ASTPtr clone() const override;
|
||||
|
||||
bool recursive_with = false;
|
||||
bool distinct = false;
|
||||
bool group_by_all = false;
|
||||
bool group_by_with_totals = false;
|
||||
|
||||
@ -13,7 +13,7 @@ class ASTSubquery : public ASTWithAlias
|
||||
{
|
||||
public:
|
||||
// Stored the name when the subquery is defined in WITH clause. For example:
|
||||
// WITH (SELECT 1) AS a SELECT * FROM a AS b; cte_name will be `a`.
|
||||
// WITH a AS (SELECT 1) SELECT * FROM a AS b; cte_name will be `a`.
|
||||
String cte_name;
|
||||
|
||||
/** Get the text that identifies this element. */
|
||||
|
||||
@ -516,6 +516,7 @@ namespace DB
|
||||
MR_MACROS(WITH_REPLACE_OPTION, "WITH REPLACE OPTION") \
|
||||
MR_MACROS(WITH_TIES, "WITH TIES") \
|
||||
MR_MACROS(WITH, "WITH") \
|
||||
MR_MACROS(RECURSIVE, "RECURSIVE") \
|
||||
MR_MACROS(WK, "WK") \
|
||||
MR_MACROS(WRITABLE, "WRITABLE") \
|
||||
MR_MACROS(WW, "WW") \
|
||||
|
||||
@ -357,6 +357,16 @@ bool ParserFilterClause::parseImpl(Pos & pos, ASTPtr & node, Expected & expected
|
||||
return false;
|
||||
}
|
||||
|
||||
if (function.name == "count")
|
||||
{
|
||||
/// Remove child from function.arguments if it's '*' because countIf(*) is not supported.
|
||||
/// See https://github.com/ClickHouse/ClickHouse/issues/61004
|
||||
std::erase_if(function.arguments->children, [] (const ASTPtr & child)
|
||||
{
|
||||
return typeid_cast<const ASTAsterisk *>(child.get()) || typeid_cast<const ASTQualifiedAsterisk *>(child.get());
|
||||
});
|
||||
}
|
||||
|
||||
function.name += "If";
|
||||
function.arguments->children.push_back(condition->children[0]);
|
||||
return true;
|
||||
|
||||
@ -46,6 +46,7 @@ bool ParserSelectQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
|
||||
ParserKeyword s_where(Keyword::WHERE);
|
||||
ParserKeyword s_group_by(Keyword::GROUP_BY);
|
||||
ParserKeyword s_with(Keyword::WITH);
|
||||
ParserKeyword s_recursive(Keyword::RECURSIVE);
|
||||
ParserKeyword s_totals(Keyword::TOTALS);
|
||||
ParserKeyword s_having(Keyword::HAVING);
|
||||
ParserKeyword s_window(Keyword::WINDOW);
|
||||
@ -103,6 +104,8 @@ bool ParserSelectQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
|
||||
{
|
||||
if (s_with.ignore(pos, expected))
|
||||
{
|
||||
select_query->recursive_with = s_recursive.ignore(pos, expected);
|
||||
|
||||
if (!ParserList(std::make_unique<ParserWithElement>(), std::make_unique<ParserToken>(TokenType::Comma))
|
||||
.parse(pos, with_expression_list, expected))
|
||||
return false;
|
||||
|
||||
@ -1,4 +1,4 @@
|
||||
#include <cassert>
|
||||
#include <Parsers/CommonParsers.h>
|
||||
|
||||
#include <Parsers/obfuscateQueries.h>
|
||||
#include <Parsers/Lexer.h>
|
||||
@ -11,6 +11,12 @@
|
||||
#include <IO/ReadHelpers.h>
|
||||
#include <IO/ReadBufferFromMemory.h>
|
||||
|
||||
#include <algorithm>
|
||||
#include <cassert>
|
||||
#include <iterator>
|
||||
|
||||
#include <boost/algorithm/string.hpp>
|
||||
|
||||
|
||||
namespace DB
|
||||
{
|
||||
@ -24,413 +30,54 @@ namespace ErrorCodes
|
||||
namespace
|
||||
{
|
||||
|
||||
const std::unordered_set<std::string_view> keywords
|
||||
const std::unordered_set<std::string> & getObfuscateKeywords()
|
||||
{
|
||||
"!=",
|
||||
"",
|
||||
"%",
|
||||
"*",
|
||||
"+",
|
||||
"-",
|
||||
"->",
|
||||
".",
|
||||
"/",
|
||||
":",
|
||||
"::",
|
||||
"<",
|
||||
"<=",
|
||||
"<>",
|
||||
"=",
|
||||
"==",
|
||||
"<=>",
|
||||
">",
|
||||
">=",
|
||||
"?",
|
||||
"[",
|
||||
"]+",
|
||||
"]+|[",
|
||||
"^[",
|
||||
"||",
|
||||
"]+$",
|
||||
"ACCESS",
|
||||
"ACTION",
|
||||
"ADD",
|
||||
"ADMIN",
|
||||
"AFTER",
|
||||
"ALGORITHM",
|
||||
"ALIAS",
|
||||
"ALL",
|
||||
"ALLOWED_LATENESS",
|
||||
"ALTER",
|
||||
"AND",
|
||||
"ANTI",
|
||||
"ANY",
|
||||
"APPLY",
|
||||
"ARRAY",
|
||||
"AS",
|
||||
"ASC",
|
||||
"ASCENDING",
|
||||
"ASOF",
|
||||
"ASSUME",
|
||||
"AST",
|
||||
"ASYNC",
|
||||
"ATTACH",
|
||||
"AUTO_INCREMENT",
|
||||
"BACKUP",
|
||||
"BASE_BACKUP",
|
||||
"BEGIN",
|
||||
"BETWEEN",
|
||||
"BIDIRECTIONAL",
|
||||
"BOTH",
|
||||
"BY",
|
||||
"CACHE",
|
||||
"CACHES",
|
||||
"CASCADE",
|
||||
"CASE",
|
||||
"CASEWITHEXPRESSION",
|
||||
"CAST",
|
||||
"CHANGE",
|
||||
"CHANGEABLE_IN_READONLY",
|
||||
"CHANGED",
|
||||
"CHAR",
|
||||
"CHARACTER",
|
||||
"CHECK",
|
||||
"CLEANUP",
|
||||
"CLEAR",
|
||||
"CLUSTER",
|
||||
"CLUSTER_HOST_IDS",
|
||||
"CLUSTERS",
|
||||
"CN",
|
||||
"CODEC",
|
||||
"COLLATE",
|
||||
"COLLECTION",
|
||||
"COLUMN",
|
||||
"COLUMNS",
|
||||
"COMMENT",
|
||||
"COMMIT",
|
||||
"COMPRESSION",
|
||||
"CONCAT",
|
||||
"CONSTRAINT",
|
||||
"CREATE",
|
||||
"CROSS",
|
||||
"CUBE",
|
||||
"CURRENT",
|
||||
"CURRENT_USER",
|
||||
"DATABASE",
|
||||
"DATABASES",
|
||||
"DATE",
|
||||
"DATE_ADD",
|
||||
"DATEADD",
|
||||
"DATE_DIFF",
|
||||
"DATEDIFF",
|
||||
"DATE_SUB",
|
||||
"DATESUB",
|
||||
"DAY",
|
||||
"DD",
|
||||
"DDL",
|
||||
"DEDUPLICATE",
|
||||
"DEFAULT",
|
||||
"DELAY",
|
||||
"DELETE",
|
||||
"DESC",
|
||||
"DESCENDING",
|
||||
"DESCRIBE",
|
||||
"DETACH",
|
||||
"DETACHED",
|
||||
"DICTIONARIES",
|
||||
"DICTIONARY",
|
||||
"DISK",
|
||||
"DISTINCT",
|
||||
"DIV",
|
||||
"DOUBLE_SHA1_HASH",
|
||||
"DROP",
|
||||
"ELSE",
|
||||
"EMPTY",
|
||||
"ENABLED",
|
||||
"END",
|
||||
"ENFORCED",
|
||||
"ENGINE",
|
||||
"EPHEMERAL",
|
||||
"EQUALS",
|
||||
"ESTIMATE",
|
||||
"EVENT",
|
||||
"EVENTS",
|
||||
"EXCEPT",
|
||||
"EXCHANGE",
|
||||
"EXISTS",
|
||||
"EXPLAIN",
|
||||
"EXPRESSION",
|
||||
"EXTERNAL",
|
||||
"EXTRACT",
|
||||
"FALSE",
|
||||
"FETCH",
|
||||
"FILE",
|
||||
"FILESYSTEM",
|
||||
"FILL",
|
||||
"FILTER",
|
||||
"FINAL",
|
||||
"FIRST",
|
||||
"FOLLOWING",
|
||||
"FOR",
|
||||
"FOREIGN",
|
||||
"FORMAT",
|
||||
"FREEZE",
|
||||
"FROM",
|
||||
"FULL",
|
||||
"FULLTEXT",
|
||||
"FUNCTION",
|
||||
"GLOBAL",
|
||||
"GRANT",
|
||||
"GRANTEES",
|
||||
"GRANTS",
|
||||
"GRANULARITY",
|
||||
"GREATER",
|
||||
"GREATEROREQUALS",
|
||||
"GROUP",
|
||||
"GROUPING",
|
||||
"GROUPS",
|
||||
"HASH",
|
||||
"HAVING",
|
||||
"HDFS",
|
||||
"HH",
|
||||
"HIERARCHICAL",
|
||||
"HOST",
|
||||
"HOUR",
|
||||
"ID",
|
||||
"IDENTIFIED",
|
||||
"IF",
|
||||
"ILIKE",
|
||||
"IN",
|
||||
"INDEX",
|
||||
"INFILE",
|
||||
"INHERIT",
|
||||
"INJECTIVE",
|
||||
"INNER",
|
||||
"INSERT",
|
||||
"INTERPOLATE",
|
||||
"INTERSECT",
|
||||
"INTERVAL",
|
||||
"INTO",
|
||||
"INVISIBLE",
|
||||
"IP",
|
||||
"IS",
|
||||
"IS_OBJECT_ID",
|
||||
"JOIN",
|
||||
"KEY",
|
||||
"KEYED",
|
||||
"KILL",
|
||||
"LAMBDA",
|
||||
"LARGE",
|
||||
"LAST",
|
||||
"LAYOUT",
|
||||
"LEADING",
|
||||
"LEFT",
|
||||
"LESS",
|
||||
"LESSOREQUALS",
|
||||
"LEVEL",
|
||||
"LIFETIME",
|
||||
"LIKE",
|
||||
"LIMIT",
|
||||
"LIMITS",
|
||||
"LINEAR",
|
||||
"LIST",
|
||||
"LITERAL",
|
||||
"LIVE",
|
||||
"LOCAL",
|
||||
"LTRIM",
|
||||
"MATCH",
|
||||
"MATERIALIZE",
|
||||
"MATERIALIZED",
|
||||
"MAX",
|
||||
"MCS",
|
||||
"MEMORY",
|
||||
"MI",
|
||||
"MICROSECOND",
|
||||
"MILLISECOND",
|
||||
"MIN",
|
||||
"MINUS",
|
||||
"MINUTE",
|
||||
"MM",
|
||||
"MOD",
|
||||
"MODIFY",
|
||||
"MONTH",
|
||||
"MOVE",
|
||||
"MS",
|
||||
"MULTIIF",
|
||||
"MUTATION",
|
||||
"NAME",
|
||||
"NAMED",
|
||||
"NANOSECOND",
|
||||
"NEXT",
|
||||
"NO",
|
||||
"NONE",
|
||||
"NOT",
|
||||
"NOTEQUALS",
|
||||
"NOTIN",
|
||||
"NS",
|
||||
"NULL",
|
||||
"NULLS",
|
||||
"OBJECT",
|
||||
"OFFSET",
|
||||
"ON",
|
||||
"ONLY",
|
||||
"OPTIMIZE",
|
||||
"OPTION",
|
||||
"OR",
|
||||
"ORDER",
|
||||
"OUTER",
|
||||
"OUTFILE",
|
||||
"OVER",
|
||||
"OVERRIDE",
|
||||
"PART",
|
||||
"PARTIAL",
|
||||
"PARTITION",
|
||||
"PARTITIONS",
|
||||
"PART_MOVE_TO_SHARD",
|
||||
"PERMANENTLY",
|
||||
"PERMISSIVE",
|
||||
"PIPELINE",
|
||||
"PLAN",
|
||||
"PLUS",
|
||||
"POLICY",
|
||||
"POPULATE",
|
||||
"POSITION",
|
||||
"PRECEDING",
|
||||
"PRECISION",
|
||||
"PREWHERE",
|
||||
"PRIMARY",
|
||||
"PRIVILEGES",
|
||||
"PROCESSLIST",
|
||||
"PROFILE",
|
||||
"PROJECTION",
|
||||
"QQ",
|
||||
"QUARTER",
|
||||
"QUERY",
|
||||
"QUOTA",
|
||||
"RANDOMIZED",
|
||||
"RANGE",
|
||||
"READONLY",
|
||||
"REALM",
|
||||
"RECOMPRESS",
|
||||
"REFERENCES",
|
||||
"REFRESH",
|
||||
"REGEXP",
|
||||
"REGEXPQUOTEMETA",
|
||||
"REMOVE",
|
||||
"RENAME",
|
||||
"REPLACE",
|
||||
"REPLACEREGEXPALL",
|
||||
"REPLACEREGEXPONE",
|
||||
"RESET",
|
||||
"RESTORE",
|
||||
"RESTRICT",
|
||||
"RESTRICTIVE",
|
||||
"RESUME",
|
||||
"REVOKE",
|
||||
"RIGHT",
|
||||
"ROLE",
|
||||
"ROLES",
|
||||
"ROLLBACK",
|
||||
"ROLLUP",
|
||||
"ROW",
|
||||
"ROWS",
|
||||
"RTRIM",
|
||||
"S3",
|
||||
"SALT",
|
||||
"SAMPLE",
|
||||
"SECOND",
|
||||
"SELECT",
|
||||
"SEMI",
|
||||
"SERVER",
|
||||
"SET",
|
||||
"SETS",
|
||||
"SETTING",
|
||||
"SETTINGS",
|
||||
"SHA256_HASH",
|
||||
"SHARD",
|
||||
"SHOW",
|
||||
"SIGNED",
|
||||
"SIMPLE",
|
||||
"SINGLEVALUEORNULL",
|
||||
"SNAPSHOT",
|
||||
"SOURCE",
|
||||
"SPATIAL",
|
||||
"SS",
|
||||
"STDOUT",
|
||||
"STEP",
|
||||
"STORAGE",
|
||||
"STRICT",
|
||||
"STRICTLY_ASCENDING",
|
||||
"SUBPARTITION",
|
||||
"SUBPARTITIONS",
|
||||
"SUBSTRING",
|
||||
"SUSPEND",
|
||||
"SYNC",
|
||||
"SYNTAX",
|
||||
"SYSTEM",
|
||||
"TABLE",
|
||||
"TABLES",
|
||||
"TEMPORARY",
|
||||
"TEST",
|
||||
"THAN",
|
||||
"THEN",
|
||||
"TIES",
|
||||
"TIMESTAMP",
|
||||
"TIMESTAMP_ADD",
|
||||
"TIMESTAMPADD",
|
||||
"TIMESTAMP_DIFF",
|
||||
"TIMESTAMPDIFF",
|
||||
"TIMESTAMP_SUB",
|
||||
"TIMESTAMPSUB",
|
||||
"TO",
|
||||
"TODATE",
|
||||
"TODATETIME",
|
||||
"TOP",
|
||||
"TOTALS",
|
||||
"TRACKING",
|
||||
"TRAILING",
|
||||
"TRANSACTION",
|
||||
"TREE",
|
||||
"TRIGGER",
|
||||
"TRIM",
|
||||
"TRIMBOTH",
|
||||
"TRIMLEFT",
|
||||
"TRIMRIGHT",
|
||||
"TRUE",
|
||||
"TRUNCATE",
|
||||
"TTL",
|
||||
"TUPLE",
|
||||
"TYPE",
|
||||
"UNBOUNDED",
|
||||
"UNFREEZE",
|
||||
"UNION",
|
||||
"UNIQUE",
|
||||
"UNSIGNED",
|
||||
"UNTUPLE",
|
||||
"UPDATE",
|
||||
"URL",
|
||||
"USE",
|
||||
"USER",
|
||||
"USING",
|
||||
"UUID",
|
||||
"VALUES",
|
||||
"VARYING",
|
||||
"VIEW",
|
||||
"VIEWIFPERMITTED",
|
||||
"VISIBLE",
|
||||
"VOLUME",
|
||||
"WATCH",
|
||||
"WATERMARK",
|
||||
"WEEK",
|
||||
"WHEN",
|
||||
"WHERE",
|
||||
"WINDOW",
|
||||
"WITH",
|
||||
"WK",
|
||||
"WRITABLE",
|
||||
"YEAR",
|
||||
"YYYY",
|
||||
"ZKPATH"
|
||||
auto initialize = []()
|
||||
{
|
||||
std::unordered_set<std::string> instance = {
|
||||
"!=",
|
||||
"",
|
||||
"%",
|
||||
"*",
|
||||
"+",
|
||||
"-",
|
||||
"->",
|
||||
".",
|
||||
"/",
|
||||
":",
|
||||
"::",
|
||||
"<",
|
||||
"<=",
|
||||
"<>",
|
||||
"=",
|
||||
"==",
|
||||
"<=>",
|
||||
">",
|
||||
">=",
|
||||
"?",
|
||||
"[",
|
||||
"]+",
|
||||
"]+|[",
|
||||
"^[",
|
||||
"||",
|
||||
"]+$"
|
||||
};
|
||||
|
||||
for (const auto & keyword : getAllKeyWords())
|
||||
{
|
||||
/// The keyword may consist of several tokens (ORDER BY or GROUP BY)
|
||||
/// We will split them and add separately.
|
||||
std::vector<std::string> tokens;
|
||||
boost::split(tokens, keyword, [](char c) { return c == ' '; });
|
||||
for (const auto & token : tokens)
|
||||
instance.insert(token);
|
||||
}
|
||||
|
||||
return instance;
|
||||
};
|
||||
|
||||
static std::unordered_set<std::string> instance = initialize();
|
||||
return instance;
|
||||
};
|
||||
|
||||
/// We want to keep some words inside quotes. For example we want to keep HOUR inside:
|
||||
@ -1309,11 +956,9 @@ void obfuscateQueries(
|
||||
|
||||
if (token.type == TokenType::BareWord)
|
||||
{
|
||||
std::string whole_token_uppercase(whole_token);
|
||||
Poco::toUpperInPlace(whole_token_uppercase);
|
||||
auto whole_token_uppercase = Poco::toUpper(toString(whole_token));
|
||||
|
||||
if (keywords.contains(whole_token_uppercase)
|
||||
|| known_identifier_func(whole_token))
|
||||
if (getObfuscateKeywords().contains(whole_token_uppercase) || known_identifier_func(whole_token))
|
||||
{
|
||||
/// Keep keywords as is.
|
||||
result.write(token.begin, token.size());
|
||||
|
||||
@ -90,7 +90,7 @@ public:
|
||||
|
||||
auto subquery_to_execute = in_second_argument;
|
||||
if (in_second_argument->as<TableNode>())
|
||||
subquery_to_execute = buildSubqueryToReadColumnsFromTableExpression(std::move(subquery_to_execute), planner_context.getQueryContext());
|
||||
subquery_to_execute = buildSubqueryToReadColumnsFromTableExpression(subquery_to_execute, planner_context.getQueryContext());
|
||||
|
||||
sets.addFromSubquery(set_key, std::move(subquery_to_execute), settings);
|
||||
}
|
||||
|
||||
@ -35,6 +35,7 @@
|
||||
#include <Processors/QueryPlan/LimitByStep.h>
|
||||
#include <Processors/QueryPlan/WindowStep.h>
|
||||
#include <Processors/QueryPlan/ReadNothingStep.h>
|
||||
#include <Processors/QueryPlan/ReadFromRecursiveCTEStep.h>
|
||||
#include <QueryPipeline/QueryPipelineBuilder.h>
|
||||
|
||||
#include <Interpreters/Context.h>
|
||||
@ -1249,6 +1250,21 @@ void Planner::buildPlanForUnionNode()
|
||||
|| union_mode == SelectUnionMode::INTERSECT_DEFAULT)
|
||||
throw Exception(ErrorCodes::BAD_ARGUMENTS, "UNION mode must be initialized");
|
||||
|
||||
if (union_node.hasRecursiveCTETable())
|
||||
{
|
||||
const auto & recursive_cte_table = *union_node.getRecursiveCTETable();
|
||||
|
||||
ColumnsWithTypeAndName recursive_cte_columns;
|
||||
recursive_cte_columns.reserve(recursive_cte_table.columns.size());
|
||||
for (const auto & recursive_cte_table_column : recursive_cte_table.columns)
|
||||
recursive_cte_columns.emplace_back(recursive_cte_table_column.type, recursive_cte_table_column.name);
|
||||
|
||||
auto read_from_recursive_cte_step = std::make_unique<ReadFromRecursiveCTEStep>(Block(std::move(recursive_cte_columns)), query_tree);
|
||||
read_from_recursive_cte_step->setStepDescription(query_tree->toAST()->formatForErrorMessage());
|
||||
query_plan.addStep(std::move(read_from_recursive_cte_step));
|
||||
return;
|
||||
}
|
||||
|
||||
const auto & union_queries_nodes = union_node.getQueries().getNodes();
|
||||
size_t queries_size = union_queries_nodes.size();
|
||||
|
||||
@ -1366,6 +1382,7 @@ void Planner::buildPlanForQueryNode()
|
||||
select_query_info.has_window = hasWindowFunctionNodes(query_tree);
|
||||
select_query_info.has_aggregates = hasAggregateFunctionNodes(query_tree);
|
||||
select_query_info.need_aggregate = query_node.hasGroupBy() || select_query_info.has_aggregates;
|
||||
select_query_info.merge_tree_enable_remove_parts_from_snapshot_optimization = select_query_options.merge_tree_enable_remove_parts_from_snapshot_optimization;
|
||||
|
||||
if (!select_query_info.has_window && query_node.hasQualify())
|
||||
{
|
||||
|
||||
@ -1089,18 +1089,6 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_
|
||||
auto right_plan = std::move(right_join_tree_query_plan.query_plan);
|
||||
auto right_plan_output_columns = right_plan.getCurrentDataStream().header.getColumnsWithTypeAndName();
|
||||
|
||||
// {
|
||||
// WriteBufferFromOwnString buf;
|
||||
// left_plan.explainPlan(buf, {.header = true, .actions = true});
|
||||
// std::cerr << "left plan \n "<< buf.str() << std::endl;
|
||||
// }
|
||||
|
||||
// {
|
||||
// WriteBufferFromOwnString buf;
|
||||
// right_plan.explainPlan(buf, {.header = true, .actions = true});
|
||||
// std::cerr << "right plan \n "<< buf.str() << std::endl;
|
||||
// }
|
||||
|
||||
JoinClausesAndActions join_clauses_and_actions;
|
||||
JoinKind join_kind = join_node.getKind();
|
||||
JoinStrictness join_strictness = join_node.getStrictness();
|
||||
|
||||
@ -160,26 +160,6 @@ ASTPtr queryNodeToDistributedSelectQuery(const QueryTreeNodePtr & query_node)
|
||||
return ast;
|
||||
}
|
||||
|
||||
/** There are no limits on the maximum size of the result for the subquery.
|
||||
* Since the result of the query is not the result of the entire query.
|
||||
*/
|
||||
void updateContextForSubqueryExecution(ContextMutablePtr & mutable_context)
|
||||
{
|
||||
/** The subquery in the IN / JOIN section does not have any restrictions on the maximum size of the result.
|
||||
* Because the result of this query is not the result of the entire query.
|
||||
* Constraints work instead
|
||||
* max_rows_in_set, max_bytes_in_set, set_overflow_mode,
|
||||
* max_rows_in_join, max_bytes_in_join, join_overflow_mode,
|
||||
* which are checked separately (in the Set, Join objects).
|
||||
*/
|
||||
Settings subquery_settings = mutable_context->getSettings();
|
||||
subquery_settings.max_result_rows = 0;
|
||||
subquery_settings.max_result_bytes = 0;
|
||||
/// The calculation of extremes does not make sense and is not necessary (if you do it, then the extremes of the subquery can be taken for whole query).
|
||||
subquery_settings.extremes = false;
|
||||
mutable_context->setSettings(subquery_settings);
|
||||
}
|
||||
|
||||
namespace
|
||||
{
|
||||
|
||||
@ -422,38 +402,6 @@ QueryTreeNodePtr replaceTableExpressionsWithDummyTables(
|
||||
return query_node->cloneAndReplace(replacement_map);
|
||||
}
|
||||
|
||||
QueryTreeNodePtr buildSubqueryToReadColumnsFromTableExpression(const NamesAndTypes & columns,
|
||||
const QueryTreeNodePtr & table_expression,
|
||||
const ContextPtr & context)
|
||||
{
|
||||
auto projection_columns = columns;
|
||||
|
||||
QueryTreeNodes subquery_projection_nodes;
|
||||
subquery_projection_nodes.reserve(projection_columns.size());
|
||||
|
||||
for (const auto & column : projection_columns)
|
||||
subquery_projection_nodes.push_back(std::make_shared<ColumnNode>(column, table_expression));
|
||||
|
||||
if (subquery_projection_nodes.empty())
|
||||
{
|
||||
auto constant_data_type = std::make_shared<DataTypeUInt64>();
|
||||
subquery_projection_nodes.push_back(std::make_shared<ConstantNode>(1UL, constant_data_type));
|
||||
projection_columns.push_back({"1", std::move(constant_data_type)});
|
||||
}
|
||||
|
||||
auto context_copy = Context::createCopy(context);
|
||||
updateContextForSubqueryExecution(context_copy);
|
||||
|
||||
auto query_node = std::make_shared<QueryNode>(std::move(context_copy));
|
||||
|
||||
query_node->getProjection().getNodes() = std::move(subquery_projection_nodes);
|
||||
query_node->resolveProjectionColumns(projection_columns);
|
||||
query_node->getJoinTree() = table_expression;
|
||||
query_node->setIsSubquery(true);
|
||||
|
||||
return query_node;
|
||||
}
|
||||
|
||||
SelectQueryInfo buildSelectQueryInfo(const QueryTreeNodePtr & query_tree, const PlannerContextPtr & planner_context)
|
||||
{
|
||||
SelectQueryInfo select_query_info;
|
||||
|
||||
Some files were not shown because too many files have changed in this diff Show More
Loading…
Reference in New Issue
Block a user