thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-11-10 01:25:21 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'master' into disable-logical-expression-optimizer-with-aliases

Browse Source
This commit is contained in:
Nikolai Kochetov 2023-03-13 15:12:02 +01:00 committed by GitHub
commit ceeb36b596
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
737 changed files with 13495 additions and 4625 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

43
.github/workflows/backport_branches.yml vendored
View File

@ -79,7 +79,7 @@ jobs:
with:
name: changed_images
path: ${{ runner.temp }}/changed_images.json
CompatibilityCheck:
CompatibilityCheckX86:
needs: [BuilderDebRelease]
runs-on: [self-hosted, style-checker]
steps:
@ -98,12 +98,43 @@ jobs:
uses: actions/download-artifact@v3
with:
path: ${{ env.REPORTS_PATH }}
- name: CompatibilityCheck
- name: CompatibilityCheckX86
run: |
sudo rm -fr "$TEMP_PATH"
mkdir -p "$TEMP_PATH"
cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
cd "$REPO_COPY/tests/ci" && python3 compatibility_check.py
cd "$REPO_COPY/tests/ci" && python3 compatibility_check.py --check-name "Compatibility check (amd64)" --check-glibc --check-distributions
- name: Cleanup
if: always()
run: |
docker ps --quiet | xargs --no-run-if-empty docker kill ||:
docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||:
sudo rm -fr "$TEMP_PATH"
CompatibilityCheckAarch64:
needs: [BuilderDebAarch64]
runs-on: [self-hosted, style-checker]
steps:
- name: Set envs
run: |
cat >> "$GITHUB_ENV" << 'EOF'
TEMP_PATH=${{runner.temp}}/compatibility_check
REPO_COPY=${{runner.temp}}/compatibility_check/ClickHouse
REPORTS_PATH=${{runner.temp}}/reports_dir
EOF
- name: Check out repository code
uses: ClickHouse/checkout@v1
with:
clear-repository: true
- name: Download json reports
uses: actions/download-artifact@v3
with:
path: ${{ env.REPORTS_PATH }}
- name: CompatibilityCheckAarch64
run: |
sudo rm -fr "$TEMP_PATH"
mkdir -p "$TEMP_PATH"
cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
cd "$REPO_COPY/tests/ci" && python3 compatibility_check.py --check-name "Compatibility check (aarch64)" --check-glibc
- name: Cleanup
if: always()
run: |
@ -421,7 +452,8 @@ jobs:
- name: Check docker clickhouse/clickhouse-server building
run: |
cd "$GITHUB_WORKSPACE/tests/ci"
python3 docker_server.py --release-type head --no-push
python3 docker_server.py --release-type head --no-push \
--image-repo clickhouse/clickhouse-server --image-path docker/server
python3 docker_server.py --release-type head --no-push --no-ubuntu \
--image-repo clickhouse/clickhouse-keeper --image-path docker/keeper
- name: Cleanup
@ -741,7 +773,8 @@ jobs:
- FunctionalStatefulTestDebug
- StressTestTsan
- IntegrationTestsRelease
- CompatibilityCheck
- CompatibilityCheckX86
- CompatibilityCheckAarch64
runs-on: [self-hosted, style-checker]
steps:
- name: Check out repository code

1
.github/workflows/cherry_pick.yml vendored
View File

@ -35,7 +35,6 @@ jobs:
fetch-depth: 0
- name: Cherry pick
run: |
sudo pip install GitPython
cd "$GITHUB_WORKSPACE/tests/ci"
python3 cherry_pick.py
- name: Cleanup

43
.github/workflows/master.yml vendored
View File

@ -110,7 +110,7 @@ jobs:
docker ps --quiet | xargs --no-run-if-empty docker kill ||:
docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||:
sudo rm -fr "$TEMP_PATH"
CompatibilityCheck:
CompatibilityCheckX86:
needs: [BuilderDebRelease]
runs-on: [self-hosted, style-checker]
steps:
@ -129,12 +129,43 @@ jobs:
uses: actions/download-artifact@v3
with:
path: ${{ env.REPORTS_PATH }}
- name: CompatibilityCheck
- name: CompatibilityCheckX86
run: |
sudo rm -fr "$TEMP_PATH"
mkdir -p "$TEMP_PATH"
cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
cd "$REPO_COPY/tests/ci" && python3 compatibility_check.py
cd "$REPO_COPY/tests/ci" && python3 compatibility_check.py --check-name "Compatibility check (amd64)" --check-glibc --check-distributions
- name: Cleanup
if: always()
run: |
docker ps --quiet | xargs --no-run-if-empty docker kill ||:
docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||:
sudo rm -fr "$TEMP_PATH"
CompatibilityCheckAarch64:
needs: [BuilderDebAarch64]
runs-on: [self-hosted, style-checker]
steps:
- name: Set envs
run: |
cat >> "$GITHUB_ENV" << 'EOF'
TEMP_PATH=${{runner.temp}}/compatibility_check
REPO_COPY=${{runner.temp}}/compatibility_check/ClickHouse
REPORTS_PATH=${{runner.temp}}/reports_dir
EOF
- name: Check out repository code
uses: ClickHouse/checkout@v1
with:
clear-repository: true
- name: Download json reports
uses: actions/download-artifact@v3
with:
path: ${{ env.REPORTS_PATH }}
- name: CompatibilityCheckAarch64
run: |
sudo rm -fr "$TEMP_PATH"
mkdir -p "$TEMP_PATH"
cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
cd "$REPO_COPY/tests/ci" && python3 compatibility_check.py --check-name "Compatibility check (aarch64)" --check-glibc
- name: Cleanup
if: always()
run: |
@ -829,7 +860,8 @@ jobs:
- name: Check docker clickhouse/clickhouse-server building
run: |
cd "$GITHUB_WORKSPACE/tests/ci"
python3 docker_server.py --release-type head
python3 docker_server.py --release-type head \
--image-repo clickhouse/clickhouse-server --image-path docker/server
python3 docker_server.py --release-type head --no-ubuntu \
--image-repo clickhouse/clickhouse-keeper --image-path docker/keeper
- name: Cleanup
@ -3124,7 +3156,8 @@ jobs:
- PerformanceComparisonX86-1
- PerformanceComparisonX86-2
- PerformanceComparisonX86-3
- CompatibilityCheck
- CompatibilityCheckX86
- CompatibilityCheckAarch64
- ASTFuzzerTestDebug
- ASTFuzzerTestAsan
- ASTFuzzerTestTsan

44
.github/workflows/pull_request.yml vendored
View File

@ -37,7 +37,6 @@ jobs:
cd "$GITHUB_WORKSPACE/tests/ci"
python3 run_check.py
PythonUnitTests:
needs: CheckLabels
runs-on: [self-hosted, style-checker]
steps:
- name: Check out repository code
@ -174,7 +173,7 @@ jobs:
docker ps --quiet | xargs --no-run-if-empty docker kill ||:
docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||:
sudo rm -fr "$TEMP_PATH" "$CACHES_PATH"
CompatibilityCheck:
CompatibilityCheckX86:
needs: [BuilderDebRelease]
runs-on: [self-hosted, style-checker]
steps:
@ -193,12 +192,43 @@ jobs:
uses: actions/download-artifact@v3
with:
path: ${{ env.REPORTS_PATH }}
- name: CompatibilityCheck
- name: CompatibilityCheckX86
run: |
sudo rm -fr "$TEMP_PATH"
mkdir -p "$TEMP_PATH"
cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
cd "$REPO_COPY/tests/ci" && python3 compatibility_check.py
cd "$REPO_COPY/tests/ci" && python3 compatibility_check.py --check-name "Compatibility check (amd64)" --check-glibc --check-distributions
- name: Cleanup
if: always()
run: |
docker ps --quiet | xargs --no-run-if-empty docker kill ||:
docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||:
sudo rm -fr "$TEMP_PATH"
CompatibilityCheckAarch64:
needs: [BuilderDebAarch64]
runs-on: [self-hosted, style-checker]
steps:
- name: Set envs
run: |
cat >> "$GITHUB_ENV" << 'EOF'
TEMP_PATH=${{runner.temp}}/compatibility_check
REPO_COPY=${{runner.temp}}/compatibility_check/ClickHouse
REPORTS_PATH=${{runner.temp}}/reports_dir
EOF
- name: Check out repository code
uses: ClickHouse/checkout@v1
with:
clear-repository: true
- name: Download json reports
uses: actions/download-artifact@v3
with:
path: ${{ env.REPORTS_PATH }}
- name: CompatibilityCheckAarch64
run: |
sudo rm -fr "$TEMP_PATH"
mkdir -p "$TEMP_PATH"
cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
cd "$REPO_COPY/tests/ci" && python3 compatibility_check.py --check-name "Compatibility check (aarch64)" --check-glibc
- name: Cleanup
if: always()
run: |
@ -886,7 +916,8 @@ jobs:
- name: Check docker clickhouse/clickhouse-server building
run: |
cd "$GITHUB_WORKSPACE/tests/ci"
python3 docker_server.py --release-type head --no-push
python3 docker_server.py --release-type head --no-push \
--image-repo clickhouse/clickhouse-server --image-path docker/server
python3 docker_server.py --release-type head --no-push --no-ubuntu \
--image-repo clickhouse/clickhouse-keeper --image-path docker/keeper
- name: Cleanup
@ -4792,7 +4823,8 @@ jobs:
- UnitTestsMsan
- UnitTestsUBsan
- UnitTestsReleaseClang
- CompatibilityCheck
- CompatibilityCheckX86
- CompatibilityCheckAarch64
- IntegrationTestsFlakyCheck
- SQLancerTestRelease
- SQLancerTestDebug

31
.github/workflows/release.yml vendored
View File

@ -7,15 +7,28 @@ on: # yamllint disable-line rule:truthy
release:
types:
- published
workflow_dispatch:
inputs:
tag:
description: 'Release tag'
required: true
type: string
jobs:
ReleasePublish:
runs-on: [self-hosted, style-checker]
steps:
- name: Set tag from input
if: github.event_name == 'workflow_dispatch'
run: |
echo "GITHUB_TAG=${{ github.event.inputs.tag }}" >> "$GITHUB_ENV"
- name: Set tag from REF
if: github.event_name == 'release'
run: |
echo "GITHUB_TAG=${GITHUB_REF#refs/tags/}" >> "$GITHUB_ENV"
- name: Deploy packages and assets
run: |
GITHUB_TAG="${GITHUB_REF#refs/tags/}"
curl --silent --data '' \
curl --silent --data '' --no-buffer \
'${{ secrets.PACKAGES_RELEASE_URL }}/release/'"${GITHUB_TAG}"'?binary=binary_darwin&binary=binary_darwin_aarch64&sync=true'
############################################################################################
##################################### Docker images #######################################
@ -23,16 +36,26 @@ jobs:
DockerServerImages:
runs-on: [self-hosted, style-checker]
steps:
- name: Set tag from input
if: github.event_name == 'workflow_dispatch'
run: |
echo "GITHUB_TAG=${{ github.event.inputs.tag }}" >> "$GITHUB_ENV"
- name: Set tag from REF
if: github.event_name == 'release'
run: |
echo "GITHUB_TAG=${GITHUB_REF#refs/tags/}" >> "$GITHUB_ENV"
- name: Check out repository code
uses: ClickHouse/checkout@v1
with:
clear-repository: true
fetch-depth: 0 # otherwise we will have no version info
ref: ${{ env.GITHUB_TAG }}
- name: Check docker clickhouse/clickhouse-server building
run: |
cd "$GITHUB_WORKSPACE/tests/ci"
python3 docker_server.py --release-type auto --version "${{ github.ref }}"
python3 docker_server.py --release-type auto --version "${{ github.ref }}" --no-ubuntu \
python3 docker_server.py --release-type auto --version "$GITHUB_TAG" \
--image-repo clickhouse/clickhouse-server --image-path docker/server
python3 docker_server.py --release-type auto --version "$GITHUB_TAG" --no-ubuntu \
--image-repo clickhouse/clickhouse-keeper --image-path docker/keeper
- name: Cleanup
if: always()

43
.github/workflows/release_branches.yml vendored
View File

@ -71,7 +71,7 @@ jobs:
with:
name: changed_images
path: ${{ runner.temp }}/changed_images.json
CompatibilityCheck:
CompatibilityCheckX86:
needs: [BuilderDebRelease]
runs-on: [self-hosted, style-checker]
steps:
@ -90,12 +90,43 @@ jobs:
uses: actions/download-artifact@v3
with:
path: ${{ env.REPORTS_PATH }}
- name: CompatibilityCheck
- name: CompatibilityCheckX86
run: |
sudo rm -fr "$TEMP_PATH"
mkdir -p "$TEMP_PATH"
cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
cd "$REPO_COPY/tests/ci" && python3 compatibility_check.py
cd "$REPO_COPY/tests/ci" && python3 compatibility_check.py --check-name "Compatibility check (amd64)" --check-glibc --check-distributions
- name: Cleanup
if: always()
run: |
docker ps --quiet | xargs --no-run-if-empty docker kill ||:
docker ps --all --quiet | xargs --no-run-if-empty docker rm -f ||:
sudo rm -fr "$TEMP_PATH"
CompatibilityCheckAarch64:
needs: [BuilderDebAarch64]
runs-on: [self-hosted, style-checker]
steps:
- name: Set envs
run: |
cat >> "$GITHUB_ENV" << 'EOF'
TEMP_PATH=${{runner.temp}}/compatibility_check
REPO_COPY=${{runner.temp}}/compatibility_check/ClickHouse
REPORTS_PATH=${{runner.temp}}/reports_dir
EOF
- name: Check out repository code
uses: ClickHouse/checkout@v1
with:
clear-repository: true
- name: Download json reports
uses: actions/download-artifact@v3
with:
path: ${{ env.REPORTS_PATH }}
- name: CompatibilityCheckAarch64
run: |
sudo rm -fr "$TEMP_PATH"
mkdir -p "$TEMP_PATH"
cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
cd "$REPO_COPY/tests/ci" && python3 compatibility_check.py --check-name "Compatibility check (aarch64)" --check-glibc
- name: Cleanup
if: always()
run: |
@ -494,7 +525,8 @@ jobs:
- name: Check docker clickhouse/clickhouse-server building
run: |
cd "$GITHUB_WORKSPACE/tests/ci"
python3 docker_server.py --release-type head --no-push
python3 docker_server.py --release-type head --no-push \
--image-repo clickhouse/clickhouse-server --image-path docker/server
python3 docker_server.py --release-type head --no-push --no-ubuntu \
--image-repo clickhouse/clickhouse-keeper --image-path docker/keeper
- name: Cleanup
@ -1947,7 +1979,8 @@ jobs:
- IntegrationTestsTsan1
- IntegrationTestsTsan2
- IntegrationTestsTsan3
- CompatibilityCheck
- CompatibilityCheckX86
- CompatibilityCheckAarch64
runs-on: [self-hosted, style-checker]
steps:
- name: Check out repository code

2
README.md
View File

@ -14,7 +14,7 @@ curl https://clickhouse.com/ | sh
* [Tutorial](https://clickhouse.com/docs/en/getting_started/tutorial/) shows how to set up and query a small ClickHouse cluster.
* [Documentation](https://clickhouse.com/docs/en/) provides more in-depth information.
* [YouTube channel](https://www.youtube.com/c/ClickHouseDB) has a lot of content about ClickHouse in video format.
* [Slack](https://join.slack.com/t/clickhousedb/shared_invite/zt-1gh9ds7f4-PgDhJAaF8ad5RbWBAAjzFg) and [Telegram](https://telegram.me/clickhouse_en) allow chatting with ClickHouse users in real-time.
* [Slack](https://clickhouse.com/slack) and [Telegram](https://telegram.me/clickhouse_en) allow chatting with ClickHouse users in real-time.
* [Blog](https://clickhouse.com/blog/) contains various ClickHouse-related articles, as well as announcements and reports about events.
* [Code Browser (Woboq)](https://clickhouse.com/codebrowser/ClickHouse/index.html) with syntax highlight and navigation.
* [Code Browser (github.dev)](https://github.dev/ClickHouse/ClickHouse) with syntax highlight, powered by github.dev.

3
base/base/TypeList.h
View File

@ -4,7 +4,6 @@
#include <type_traits>
#include <utility>
#include "defines.h"
#include "TypePair.h"
/// General-purpose typelist. Easy on compilation times as it does not use recursion.
template <typename ...Args>
@ -28,7 +27,7 @@ namespace TypeListUtils /// In some contexts it's more handy to use functions in
constexpr Root<Args...> changeRoot(TypeList<Args...>) { return {}; }
template <typename F, typename ...Args>
constexpr void forEach(TypeList<Args...>, F && f) { (std::forward<F>(f)(Id<Args>{}), ...); }
constexpr void forEach(TypeList<Args...>, F && f) { (std::forward<F>(f)(TypeList<Args>{}), ...); }
}
template <typename TypeListLeft, typename TypeListRight>

4
base/base/TypePair.h
View File

@ -1,4 +0,0 @@
#pragma once
template <typename T, typename V> struct TypePair {};
template <typename T> struct Id {};

214
base/base/hex.h Normal file
View File

@ -0,0 +1,214 @@
#pragma once
#include <cstring>
#include "types.h"
/// Maps 0..15 to 0..9A..F or 0..9a..f correspondingly.
constexpr inline std::string_view hex_digit_to_char_uppercase_table = "0123456789ABCDEF";
constexpr inline std::string_view hex_digit_to_char_lowercase_table = "0123456789abcdef";
constexpr char hexDigitUppercase(unsigned char c)
{
return hex_digit_to_char_uppercase_table[c];
}
constexpr char hexDigitLowercase(unsigned char c)
{
return hex_digit_to_char_lowercase_table[c];
}
/// Maps 0..255 to 00..FF or 00..ff correspondingly
constexpr inline std::string_view hex_byte_to_char_uppercase_table = //
"000102030405060708090A0B0C0D0E0F"
"101112131415161718191A1B1C1D1E1F"
"202122232425262728292A2B2C2D2E2F"
"303132333435363738393A3B3C3D3E3F"
"404142434445464748494A4B4C4D4E4F"
"505152535455565758595A5B5C5D5E5F"
"606162636465666768696A6B6C6D6E6F"
"707172737475767778797A7B7C7D7E7F"
"808182838485868788898A8B8C8D8E8F"
"909192939495969798999A9B9C9D9E9F"
"A0A1A2A3A4A5A6A7A8A9AAABACADAEAF"
"B0B1B2B3B4B5B6B7B8B9BABBBCBDBEBF"
"C0C1C2C3C4C5C6C7C8C9CACBCCCDCECF"
"D0D1D2D3D4D5D6D7D8D9DADBDCDDDEDF"
"E0E1E2E3E4E5E6E7E8E9EAEBECEDEEEF"
"F0F1F2F3F4F5F6F7F8F9FAFBFCFDFEFF";
constexpr inline std::string_view hex_byte_to_char_lowercase_table = //
"000102030405060708090a0b0c0d0e0f"
"101112131415161718191a1b1c1d1e1f"
"202122232425262728292a2b2c2d2e2f"
"303132333435363738393a3b3c3d3e3f"
"404142434445464748494a4b4c4d4e4f"
"505152535455565758595a5b5c5d5e5f"
"606162636465666768696a6b6c6d6e6f"
"707172737475767778797a7b7c7d7e7f"
"808182838485868788898a8b8c8d8e8f"
"909192939495969798999a9b9c9d9e9f"
"a0a1a2a3a4a5a6a7a8a9aaabacadaeaf"
"b0b1b2b3b4b5b6b7b8b9babbbcbdbebf"
"c0c1c2c3c4c5c6c7c8c9cacbcccdcecf"
"d0d1d2d3d4d5d6d7d8d9dadbdcdddedf"
"e0e1e2e3e4e5e6e7e8e9eaebecedeeef"
"f0f1f2f3f4f5f6f7f8f9fafbfcfdfeff";
inline void writeHexByteUppercase(UInt8 byte, void * out)
{
memcpy(out, &hex_byte_to_char_uppercase_table[static_cast<size_t>(byte) * 2], 2);
}
inline void writeHexByteLowercase(UInt8 byte, void * out)
{
memcpy(out, &hex_byte_to_char_lowercase_table[static_cast<size_t>(byte) * 2], 2);
}
constexpr inline std::string_view bin_byte_to_char_table = //
"0000000000000001000000100000001100000100000001010000011000000111"
"0000100000001001000010100000101100001100000011010000111000001111"
"0001000000010001000100100001001100010100000101010001011000010111"
"0001100000011001000110100001101100011100000111010001111000011111"
"0010000000100001001000100010001100100100001001010010011000100111"
"0010100000101001001010100010101100101100001011010010111000101111"
"0011000000110001001100100011001100110100001101010011011000110111"
"0011100000111001001110100011101100111100001111010011111000111111"
"0100000001000001010000100100001101000100010001010100011001000111"
"0100100001001001010010100100101101001100010011010100111001001111"
"0101000001010001010100100101001101010100010101010101011001010111"
"0101100001011001010110100101101101011100010111010101111001011111"
"0110000001100001011000100110001101100100011001010110011001100111"
"0110100001101001011010100110101101101100011011010110111001101111"
"0111000001110001011100100111001101110100011101010111011001110111"
"0111100001111001011110100111101101111100011111010111111001111111"
"1000000010000001100000101000001110000100100001011000011010000111"
"1000100010001001100010101000101110001100100011011000111010001111"
"1001000010010001100100101001001110010100100101011001011010010111"
"1001100010011001100110101001101110011100100111011001111010011111"
"1010000010100001101000101010001110100100101001011010011010100111"
"1010100010101001101010101010101110101100101011011010111010101111"
"1011000010110001101100101011001110110100101101011011011010110111"
"1011100010111001101110101011101110111100101111011011111010111111"
"1100000011000001110000101100001111000100110001011100011011000111"
"1100100011001001110010101100101111001100110011011100111011001111"
"1101000011010001110100101101001111010100110101011101011011010111"
"1101100011011001110110101101101111011100110111011101111011011111"
"1110000011100001111000101110001111100100111001011110011011100111"
"1110100011101001111010101110101111101100111011011110111011101111"
"1111000011110001111100101111001111110100111101011111011011110111"
"1111100011111001111110101111101111111100111111011111111011111111";
inline void writeBinByte(UInt8 byte, void * out)
{
memcpy(out, &bin_byte_to_char_table[static_cast<size_t>(byte) * 8], 8);
}
/// Produces hex representation of an unsigned int with leading zeros (for checksums)
template <typename TUInt>
inline void writeHexUIntImpl(TUInt uint_, char * out, std::string_view table)
{
union
{
TUInt value;
UInt8 uint8[sizeof(TUInt)];
};
value = uint_;
for (size_t i = 0; i < sizeof(TUInt); ++i)
{
if constexpr (std::endian::native == std::endian::little)
memcpy(out + i * 2, &table[static_cast<size_t>(uint8[sizeof(TUInt) - 1 - i]) * 2], 2);
else
memcpy(out + i * 2, &table[static_cast<size_t>(uint8[i]) * 2], 2);
}
}
template <typename TUInt>
inline void writeHexUIntUppercase(TUInt uint_, char * out)
{
writeHexUIntImpl(uint_, out, hex_byte_to_char_uppercase_table);
}
template <typename TUInt>
inline void writeHexUIntLowercase(TUInt uint_, char * out)
{
writeHexUIntImpl(uint_, out, hex_byte_to_char_lowercase_table);
}
template <typename TUInt>
std::string getHexUIntUppercase(TUInt uint_)
{
std::string res(sizeof(TUInt) * 2, '\0');
writeHexUIntUppercase(uint_, res.data());
return res;
}
template <typename TUInt>
std::string getHexUIntLowercase(TUInt uint_)
{
std::string res(sizeof(TUInt) * 2, '\0');
writeHexUIntLowercase(uint_, res.data());
return res;
}
/// Maps 0..9, A..F, a..f to 0..15. Other chars are mapped to implementation specific value.
constexpr inline std::string_view hex_char_to_digit_table
= {"\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff"
"\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff"
"\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff"
"\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\xff\xff\xff\xff\xff\xff" //0-9
"\xff\x0a\x0b\x0c\x0d\x0e\x0f\xff\xff\xff\xff\xff\xff\xff\xff\xff" //A-Z
"\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff"
"\xff\x0a\x0b\x0c\x0d\x0e\x0f\xff\xff\xff\xff\xff\xff\xff\xff\xff" //a-z
"\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff"
"\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff"
"\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff"
"\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff"
"\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff"
"\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff"
"\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff"
"\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff"
"\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff\xff",
256};
constexpr UInt8 unhex(char c)
{
return hex_char_to_digit_table[static_cast<UInt8>(c)];
}
constexpr UInt8 unhex2(const char * data)
{
return static_cast<UInt8>(unhex(data[0])) * 0x10 + static_cast<UInt8>(unhex(data[1]));
}
constexpr UInt16 unhex4(const char * data)
{
return static_cast<UInt16>(unhex(data[0])) * 0x1000 + static_cast<UInt16>(unhex(data[1])) * 0x100
+ static_cast<UInt16>(unhex(data[2])) * 0x10 + static_cast<UInt16>(unhex(data[3]));
}
template <typename TUInt>
constexpr TUInt unhexUInt(const char * data)
{
TUInt res = 0;
if constexpr ((sizeof(TUInt) <= 8) || ((sizeof(TUInt) % 8) != 0))
{
for (size_t i = 0; i < sizeof(TUInt) * 2; ++i, ++data)
{
res <<= 4;
res += unhex(*data);
}
}
else
{
for (size_t i = 0; i < sizeof(TUInt) / 8; ++i, data += 16)
{
res <<= 64;
res += unhexUInt<UInt64>(data);
}
}
return res;
}

13
base/base/interpolate.h Normal file
View File

@ -0,0 +1,13 @@
#pragma once
#include <cassert>
#include <cmath>
/** Linear interpolation in logarithmic coordinates.
* Exponential interpolation is related to linear interpolation
* exactly in same way as geometric mean is related to arithmetic mean.
*/
constexpr double interpolateExponential(double min, double max, double ratio)
{
assert(min > 0 && ratio >= 0 && ratio <= 1);
return min * std::pow(max / min, ratio);
}

8
base/poco/MongoDB/include/Poco/MongoDB/Connection.h
View File

@ -90,6 +90,9 @@ namespace MongoDB
Poco::Net::SocketAddress address() const;
/// Returns the address of the MongoDB server.
const std::string & uri() const;
/// Returns the uri on which the connection was made.
void connect(const std::string & hostAndPort);
/// Connects to the given MongoDB server.
@ -148,6 +151,7 @@ namespace MongoDB
private:
Poco::Net::SocketAddress _address;
Poco::Net::StreamSocket _socket;
std::string _uri;
};
@ -158,6 +162,10 @@ namespace MongoDB
{
return _address;
}
inline const std::string & Connection::uri() const
{
return _uri;
}
}

195
base/poco/MongoDB/src/Connection.cpp
View File

@ -145,68 +145,155 @@ void Connection::connect(const Poco::Net::StreamSocket& socket)
void Connection::connect(const std::string& uri, SocketFactory& socketFactory)
{
Poco::URI theURI(uri);
if (theURI.getScheme() != "mongodb") throw Poco::UnknownURISchemeException(uri);
std::vector<std::string> strAddresses;
std::string newURI;
std::string userInfo = theURI.getUserInfo();
std::string host = theURI.getHost();
Poco::UInt16 port = theURI.getPort();
if (port == 0) port = 27017;
if (uri.find(',') != std::string::npos)
{
size_t pos;
size_t head = 0;
if ((pos = uri.find("@")) != std::string::npos)
{
head = pos + 1;
}
else if ((pos = uri.find("://")) != std::string::npos)
{
head = pos + 3;
}
std::string databaseName = theURI.getPath();
if (!databaseName.empty() && databaseName[0] == '/') databaseName.erase(0, 1);
if (databaseName.empty()) databaseName = "admin";
std::string tempstr;
std::string::const_iterator it = uri.begin();
it += head;
size_t tail = head;
for (;it != uri.end() && *it != '?' && *it != '/'; ++it)
{
tempstr += *it;
tail++;
}
bool ssl = false;
Poco::Timespan connectTimeout;
Poco::Timespan socketTimeout;
std::string authMechanism = Database::AUTH_SCRAM_SHA1;
it = tempstr.begin();
std::string token;
for (;it != tempstr.end(); ++it)
{
if (*it == ',')
{
newURI = uri.substr(0, head) + token + uri.substr(tail, uri.length());
strAddresses.push_back(newURI);
token = "";
}
else
{
token += *it;
}
}
newURI = uri.substr(0, head) + token + uri.substr(tail, uri.length());
strAddresses.push_back(newURI);
}
else
{
strAddresses.push_back(uri);
}
Poco::URI::QueryParameters params = theURI.getQueryParameters();
for (Poco::URI::QueryParameters::const_iterator it = params.begin(); it != params.end(); ++it)
{
if (it->first == "ssl")
{
ssl = (it->second == "true");
}
else if (it->first == "connectTimeoutMS")
{
connectTimeout = static_cast<Poco::Timespan::TimeDiff>(1000)*Poco::NumberParser::parse(it->second);
}
else if (it->first == "socketTimeoutMS")
{
socketTimeout = static_cast<Poco::Timespan::TimeDiff>(1000)*Poco::NumberParser::parse(it->second);
}
else if (it->first == "authMechanism")
{
authMechanism = it->second;
}
}
newURI = strAddresses.front();
Poco::URI theURI(newURI);
if (theURI.getScheme() != "mongodb") throw Poco::UnknownURISchemeException(uri);
connect(socketFactory.createSocket(host, port, connectTimeout, ssl));
std::string userInfo = theURI.getUserInfo();
std::string databaseName = theURI.getPath();
if (!databaseName.empty() && databaseName[0] == '/') databaseName.erase(0, 1);
if (databaseName.empty()) databaseName = "admin";
if (socketTimeout > 0)
{
_socket.setSendTimeout(socketTimeout);
_socket.setReceiveTimeout(socketTimeout);
}
bool ssl = false;
Poco::Timespan connectTimeout;
Poco::Timespan socketTimeout;
std::string authMechanism = Database::AUTH_SCRAM_SHA1;
std::string readPreference="primary";
if (!userInfo.empty())
{
std::string username;
std::string password;
std::string::size_type pos = userInfo.find(':');
if (pos != std::string::npos)
{
username.assign(userInfo, 0, pos++);
password.assign(userInfo, pos, userInfo.size() - pos);
}
else username = userInfo;
Poco::URI::QueryParameters params = theURI.getQueryParameters();
for (Poco::URI::QueryParameters::const_iterator it = params.begin(); it != params.end(); ++it)
{
if (it->first == "ssl")
{
ssl = (it->second == "true");
}
else if (it->first == "connectTimeoutMS")
{
connectTimeout = static_cast<Poco::Timespan::TimeDiff>(1000)*Poco::NumberParser::parse(it->second);
}
else if (it->first == "socketTimeoutMS")
{
socketTimeout = static_cast<Poco::Timespan::TimeDiff>(1000)*Poco::NumberParser::parse(it->second);
}
else if (it->first == "authMechanism")
{
authMechanism = it->second;
}
else if (it->first == "readPreference")
{
readPreference= it->second;
}
}
Database database(databaseName);
if (!database.authenticate(*this, username, password, authMechanism))
throw Poco::NoPermissionException(Poco::format("Access to MongoDB database %s denied for user %s", databaseName, username));
}
for (std::vector<std::string>::const_iterator it = strAddresses.cbegin();it != strAddresses.cend(); ++it)
{
newURI = *it;
theURI = Poco::URI(newURI);
std::string host = theURI.getHost();
Poco::UInt16 port = theURI.getPort();
if (port == 0) port = 27017;
connect(socketFactory.createSocket(host, port, connectTimeout, ssl));
_uri = newURI;
if (socketTimeout > 0)
{
_socket.setSendTimeout(socketTimeout);
_socket.setReceiveTimeout(socketTimeout);
}
if (strAddresses.size() > 1)
{
Poco::MongoDB::QueryRequest request("admin.$cmd");
request.setNumberToReturn(1);
request.selector().add("isMaster", 1);
Poco::MongoDB::ResponseMessage response;
sendRequest(request, response);
_uri = newURI;
if (!response.documents().empty())
{
Poco::MongoDB::Document::Ptr doc = response.documents()[0];
if (doc->get<bool>("ismaster") && readPreference == "primary")
{
break;
}
else if (!doc->get<bool>("ismaster") && readPreference == "secondary")
{
break;
}
else if (it + 1 == strAddresses.cend())
{
throw Poco::URISyntaxException(uri);
}
}
}
}
if (!userInfo.empty())
{
std::string username;
std::string password;
std::string::size_type pos = userInfo.find(':');
if (pos != std::string::npos)
{
username.assign(userInfo, 0, pos++);
password.assign(userInfo, pos, userInfo.size() - pos);
}
else username = userInfo;
Database database(databaseName);
if (!database.authenticate(*this, username, password, authMechanism))
throw Poco::NoPermissionException(Poco::format("Access to MongoDB database %s denied for user %s", databaseName, username));
}
}

20
contrib/arrow-cmake/CMakeLists.txt
View File

@ -115,6 +115,13 @@ configure_file("${ORC_SOURCE_SRC_DIR}/Adaptor.hh.in" "${ORC_BUILD_INCLUDE_DIR}/A
# ARROW_ORC + adapters/orc/CMakefiles
set(ORC_SRCS
"${CMAKE_CURRENT_BINARY_DIR}/orc_proto.pb.h"
"${ORC_SOURCE_SRC_DIR}/sargs/ExpressionTree.cc"
"${ORC_SOURCE_SRC_DIR}/sargs/Literal.cc"
"${ORC_SOURCE_SRC_DIR}/sargs/PredicateLeaf.cc"
"${ORC_SOURCE_SRC_DIR}/sargs/SargsApplier.cc"
"${ORC_SOURCE_SRC_DIR}/sargs/SearchArgument.cc"
"${ORC_SOURCE_SRC_DIR}/sargs/TruthValue.cc"
"${ORC_SOURCE_SRC_DIR}/Exceptions.cc"
"${ORC_SOURCE_SRC_DIR}/OrcFile.cc"
"${ORC_SOURCE_SRC_DIR}/Reader.cc"
@ -129,13 +136,20 @@ set(ORC_SRCS
"${ORC_SOURCE_SRC_DIR}/MemoryPool.cc"
"${ORC_SOURCE_SRC_DIR}/RLE.cc"
"${ORC_SOURCE_SRC_DIR}/RLEv1.cc"
"${ORC_SOURCE_SRC_DIR}/RLEv2.cc"
"${ORC_SOURCE_SRC_DIR}/RleDecoderV2.cc"
"${ORC_SOURCE_SRC_DIR}/RleEncoderV2.cc"
"${ORC_SOURCE_SRC_DIR}/RLEV2Util.cc"
"${ORC_SOURCE_SRC_DIR}/Statistics.cc"
"${ORC_SOURCE_SRC_DIR}/StripeStream.cc"
"${ORC_SOURCE_SRC_DIR}/Timezone.cc"
"${ORC_SOURCE_SRC_DIR}/TypeImpl.cc"
"${ORC_SOURCE_SRC_DIR}/Vector.cc"
"${ORC_SOURCE_SRC_DIR}/Writer.cc"
"${ORC_SOURCE_SRC_DIR}/Adaptor.cc"
"${ORC_SOURCE_SRC_DIR}/BloomFilter.cc"
"${ORC_SOURCE_SRC_DIR}/Murmur3.cc"
"${ORC_SOURCE_SRC_DIR}/BlockBuffer.cc"
"${ORC_SOURCE_SRC_DIR}/wrap/orc-proto-wrapper.cc"
"${ORC_SOURCE_SRC_DIR}/io/InputStream.cc"
"${ORC_SOURCE_SRC_DIR}/io/OutputStream.cc"
"${ORC_ADDITION_SOURCE_DIR}/orc_proto.pb.cc"
@ -358,6 +372,9 @@ SET(ARROW_SRCS "${LIBRARY_DIR}/util/compression_zlib.cc" ${ARROW_SRCS})
add_definitions(-DARROW_WITH_ZSTD)
SET(ARROW_SRCS "${LIBRARY_DIR}/util/compression_zstd.cc" ${ARROW_SRCS})
add_definitions(-DARROW_WITH_BROTLI)
SET(ARROW_SRCS "${LIBRARY_DIR}/util/compression_brotli.cc" ${ARROW_SRCS})
add_library(_arrow ${ARROW_SRCS})
@ -372,6 +389,7 @@ target_link_libraries(_arrow PRIVATE
ch_contrib::snappy
ch_contrib::zlib
ch_contrib::zstd
ch_contrib::brotli
)
target_link_libraries(_arrow PUBLIC _orc)

4
contrib/libfarmhash/CMakeLists.txt
View File

@ -6,6 +6,10 @@ if (MSVC)
target_compile_definitions (_farmhash PRIVATE FARMHASH_NO_BUILTIN_EXPECT=1)
endif ()
if (ARCH_S390X)
add_compile_definitions(WORDS_BIGENDIAN)
endif ()
target_include_directories (_farmhash BEFORE PUBLIC ${CMAKE_CURRENT_SOURCE_DIR})
add_library(ch_contrib::farmhash ALIAS _farmhash)

2
contrib/orc vendored

@ -1 +1 @@
Subproject commit f9a393ed2433a60034795284f82d093b348f2102
Subproject commit c5d7755ba0b9a95631c8daea4d094101f26ec761

2
docker/keeper/Dockerfile
View File

@ -29,7 +29,7 @@ RUN arch=${TARGETARCH:-amd64} \
esac
ARG REPOSITORY="https://s3.amazonaws.com/clickhouse-builds/22.4/31c367d3cd3aefd316778601ff6565119fe36682/package_release"
ARG VERSION="23.2.1.2537"
ARG VERSION="23.2.4.12"
ARG PACKAGES="clickhouse-keeper"
# user/group precreated explicitly with fixed uid/gid on purpose.

2
docker/server/Dockerfile.alpine
View File

@ -33,7 +33,7 @@ RUN arch=${TARGETARCH:-amd64} \
# lts / testing / prestable / etc
ARG REPO_CHANNEL="stable"
ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}"
ARG VERSION="23.2.1.2537"
ARG VERSION="23.2.4.12"
ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static"
# user/group precreated explicitly with fixed uid/gid on purpose.

2
docker/server/Dockerfile.ubuntu
View File

@ -22,7 +22,7 @@ RUN sed -i "s|http://archive.ubuntu.com|${apt_archive}|g" /etc/apt/sources.list
ARG REPO_CHANNEL="stable"
ARG REPOSITORY="deb https://packages.clickhouse.com/deb ${REPO_CHANNEL} main"
ARG VERSION="23.2.1.2537"
ARG VERSION="23.2.4.12"
ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static"
# set non-empty deb_location_url url to create a docker image

13
docker/test/upgrade/run.sh
View File

@ -60,6 +60,13 @@ install_packages previous_release_package_folder
export USE_S3_STORAGE_FOR_MERGE_TREE=1
# Previous version may not be ready for fault injections
export ZOOKEEPER_FAULT_INJECTION=0
# force_sync=false doesn't work correctly on some older versions
sudo cat /etc/clickhouse-server/config.d/keeper_port.xml \
| sed "s|<force_sync>false</force_sync>|<force_sync>true</force_sync>|" \
> /etc/clickhouse-server/config.d/keeper_port.xml.tmp
sudo mv /etc/clickhouse-server/config.d/keeper_port.xml.tmp /etc/clickhouse-server/config.d/keeper_port.xml
configure
# But we still need default disk because some tables loaded only into it
@ -161,8 +168,10 @@ rg -Fav -e "Code: 236. DB::Exception: Cancelled merging parts" \
-e "Authentication failed" \
-e "Cannot flush" \
-e "Container already exists" \
/var/log/clickhouse-server/clickhouse-server.upgrade.log | zgrep -Fa "<Error>" > /test_output/upgrade_error_messages.txt \
&& echo -e "Error message in clickhouse-server.log (see upgrade_error_messages.txt)$FAIL$(head_escaped /test_output/bc_check_error_messages.txt)" \
clickhouse-server.upgrade.log \
| grep -av -e "_repl_01111_.*Mapping for table with UUID" \
| zgrep -Fa "<Error>" > /test_output/upgrade_error_messages.txt \
&& echo -e "Error message in clickhouse-server.log (see upgrade_error_messages.txt)$FAIL$(head_escaped /test_output/upgrade_error_messages.txt)" \
>> /test_output/test_results.tsv \
|| echo -e "No Error messages after server upgrade$OK" >> /test_output/test_results.tsv

55
docs/changelogs/v22.12.4.76-stable.md Normal file
View File

@ -0,0 +1,55 @@
---
sidebar_position: 1
sidebar_label: 2023
---
# 2023 Changelog
### ClickHouse release v22.12.4.76-stable (cb5772db805) FIXME as compared to v22.12.3.5-stable (893de538f02)
#### Performance Improvement
* Backported in [#45704](https://github.com/ClickHouse/ClickHouse/issues/45704): Fixed performance of short `SELECT` queries that read from tables with large number of`Array`/`Map`/`Nested` columns. [#45630](https://github.com/ClickHouse/ClickHouse/pull/45630) ([Anton Popov](https://github.com/CurtizJ)).
* Backported in [#46378](https://github.com/ClickHouse/ClickHouse/issues/46378): Fix too big memory usage for vertical merges on non-remote disk. Respect `max_insert_delayed_streams_for_parallel_write` for the remote disk. [#46275](https://github.com/ClickHouse/ClickHouse/pull/46275) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
#### Bug Fix
* Backported in [#45672](https://github.com/ClickHouse/ClickHouse/issues/45672): Fix wiping sensitive info in logs. [#45603](https://github.com/ClickHouse/ClickHouse/pull/45603) ([Vitaly Baranov](https://github.com/vitlibar)).
#### Build/Testing/Packaging Improvement
* Backported in [#45200](https://github.com/ClickHouse/ClickHouse/issues/45200): Fix zookeeper downloading, update the version, and optimize the image size. [#44853](https://github.com/ClickHouse/ClickHouse/pull/44853) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Backported in [#46116](https://github.com/ClickHouse/ClickHouse/issues/46116): Remove the dependency on the `adduser` tool from the packages, because we don't use it. This fixes [#44934](https://github.com/ClickHouse/ClickHouse/issues/44934). [#45011](https://github.com/ClickHouse/ClickHouse/pull/45011) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
* Backported in [#46035](https://github.com/ClickHouse/ClickHouse/issues/46035): Add systemd.service file for clickhouse-keeper. Fixes [#44293](https://github.com/ClickHouse/ClickHouse/issues/44293). [#45568](https://github.com/ClickHouse/ClickHouse/pull/45568) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Backported in [#46484](https://github.com/ClickHouse/ClickHouse/issues/46484): Get rid of unnecessary build for standalone clickhouse-keeper. [#46367](https://github.com/ClickHouse/ClickHouse/pull/46367) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Backported in [#46509](https://github.com/ClickHouse/ClickHouse/issues/46509): Some time ago the ccache compression was changed to `zst`, but `gz` archives are downloaded by default. It fixes it by prioritizing zst archive. [#46490](https://github.com/ClickHouse/ClickHouse/pull/46490) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Backported in [#47058](https://github.com/ClickHouse/ClickHouse/issues/47058): Fix error during server startup on old distros (e.g. Amazon Linux 2) and on ARM that glibc 2.28 symbols are not found. [#47008](https://github.com/ClickHouse/ClickHouse/pull/47008) ([Robert Schulze](https://github.com/rschu1ze)).
#### Bug Fix (user-visible misbehavior in official stable or prestable release)
* Backported in [#45904](https://github.com/ClickHouse/ClickHouse/issues/45904): Fixed bug with non-parsable default value for EPHEMERAL column in table metadata. [#44026](https://github.com/ClickHouse/ClickHouse/pull/44026) ([Yakov Olkhovskiy](https://github.com/yakov-olkhovskiy)).
* Backported in [#45321](https://github.com/ClickHouse/ClickHouse/issues/45321): Fixed a bug in normalization of a `DEFAULT` expression in `CREATE TABLE` statement. The second argument of function `in` (or the right argument of operator `IN`) might be replaced with the result of its evaluation during CREATE query execution. Fixes [#44496](https://github.com/ClickHouse/ClickHouse/issues/44496). [#44547](https://github.com/ClickHouse/ClickHouse/pull/44547) ([Alexander Tokmakov](https://github.com/tavplubix)).
* Backported in [#45000](https://github.com/ClickHouse/ClickHouse/issues/45000): Another fix for `Cannot read all data` error which could happen while reading `LowCardinality` dictionary from remote fs. Fixes [#44709](https://github.com/ClickHouse/ClickHouse/issues/44709). [#44875](https://github.com/ClickHouse/ClickHouse/pull/44875) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
* Backported in [#45553](https://github.com/ClickHouse/ClickHouse/issues/45553): Fix `SELECT ... FROM system.dictionaries` exception when there is a dictionary with a bad structure (e.g. incorrect type in xml config). [#45399](https://github.com/ClickHouse/ClickHouse/pull/45399) ([Aleksei Filatov](https://github.com/aalexfvk)).
* Backported in [#46226](https://github.com/ClickHouse/ClickHouse/issues/46226): A couple of seg faults have been reported around `c-ares`. All of the recent stack traces observed fail on inserting into `std::unodered_set<>`. I believe I have found the root cause of this, it seems to be unprocessed queries. Prior to this PR, CH calls `poll` to wait on the file descriptors in the `c-ares` channel. According to the [poll docs](https://man7.org/linux/man-pages/man2/poll.2.html), a negative return value means an error has ocurred. Because of this, we would abort the execution and return failure. The problem is that `poll` will also return a negative value if a system interrupt occurs. A system interrupt does not mean the processing has failed or ended, but we would abort it anyways because we were checking for negative values. Once the execution is aborted, the whole stack is destroyed, which includes the `std::unordered_set<std::string>` passed to the `void *` parameter of the c-ares callback. Once c-ares completed the request, the callback would be invoked and would access an invalid memory address causing a segfault. [#45629](https://github.com/ClickHouse/ClickHouse/pull/45629) ([Arthur Passos](https://github.com/arthurpassos)).
* Backported in [#46218](https://github.com/ClickHouse/ClickHouse/issues/46218): Fix reading of non existing nested columns with multiple level in compact parts. [#46045](https://github.com/ClickHouse/ClickHouse/pull/46045) ([Azat Khuzhin](https://github.com/azat)).
* Backported in [#46446](https://github.com/ClickHouse/ClickHouse/issues/46446): Fix possible `LOGICAL_ERROR` in asynchronous inserts with invalid data sent in format `VALUES`. [#46350](https://github.com/ClickHouse/ClickHouse/pull/46350) ([Anton Popov](https://github.com/CurtizJ)).
* Backported in [#46678](https://github.com/ClickHouse/ClickHouse/issues/46678): Fix an invalid processing of constant `LowCardinality` argument in function `arrayMap`. This bug could lead to a segfault in release, and logical error `Bad cast` in debug build. [#46569](https://github.com/ClickHouse/ClickHouse/pull/46569) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
* Backported in [#46872](https://github.com/ClickHouse/ClickHouse/issues/46872): Fix a bug in the `Map` data type. This closes [#46855](https://github.com/ClickHouse/ClickHouse/issues/46855). [#46856](https://github.com/ClickHouse/ClickHouse/pull/46856) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
* Backported in [#46954](https://github.com/ClickHouse/ClickHouse/issues/46954): Fix result of LIKE predicates which translate to substring searches and contain quoted non-LIKE metacharacters. [#46875](https://github.com/ClickHouse/ClickHouse/pull/46875) ([Robert Schulze](https://github.com/rschu1ze)).
#### NOT FOR CHANGELOG / INSIGNIFICANT
* Automatically merge green backport PRs and green approved PRs [#41110](https://github.com/ClickHouse/ClickHouse/pull/41110) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Improve release scripts [#45074](https://github.com/ClickHouse/ClickHouse/pull/45074) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Fix wrong approved_at, simplify conditions [#45302](https://github.com/ClickHouse/ClickHouse/pull/45302) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Get rid of artifactory in favor of r2 + ch-repos-manager [#45421](https://github.com/ClickHouse/ClickHouse/pull/45421) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Another attempt to fix automerge, or at least to have debug footprint [#45476](https://github.com/ClickHouse/ClickHouse/pull/45476) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Trim refs/tags/ from GITHUB_TAG in release workflow [#45636](https://github.com/ClickHouse/ClickHouse/pull/45636) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Add check for running workflows to merge_pr.py [#45803](https://github.com/ClickHouse/ClickHouse/pull/45803) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Get rid of progress timestamps in release publishing [#45818](https://github.com/ClickHouse/ClickHouse/pull/45818) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Add necessary dependency for sanitizers [#45959](https://github.com/ClickHouse/ClickHouse/pull/45959) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Add helping logging to auto-merge script [#46080](https://github.com/ClickHouse/ClickHouse/pull/46080) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Fix write buffer destruction order for vertical merge. [#46205](https://github.com/ClickHouse/ClickHouse/pull/46205) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
* Improve install_check.py [#46458](https://github.com/ClickHouse/ClickHouse/pull/46458) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Fix dependencies for InstallPackagesTestAarch64 [#46597](https://github.com/ClickHouse/ClickHouse/pull/46597) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Get rid of legacy DocsReleaseChecks [#46665](https://github.com/ClickHouse/ClickHouse/pull/46665) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Reduce updates of Mergeable Check [#46781](https://github.com/ClickHouse/ClickHouse/pull/46781) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).

29
docs/changelogs/v22.12.5.34-stable.md Normal file
View File

@ -0,0 +1,29 @@
---
sidebar_position: 1
sidebar_label: 2023
---
# 2023 Changelog
### ClickHouse release v22.12.5.34-stable (b82d6401ca1) FIXME as compared to v22.12.4.76-stable (cb5772db805)
#### Improvement
* Backported in [#46983](https://github.com/ClickHouse/ClickHouse/issues/46983): - Apply `ALTER TABLE table_name ON CLUSTER cluster MOVE PARTITION|PART partition_expr TO DISK|VOLUME 'disk_name'` to all replicas. Because `ALTER TABLE t MOVE` is not replicated. [#46402](https://github.com/ClickHouse/ClickHouse/pull/46402) ([lizhuoyu5](https://github.com/lzydmxy)).
#### Bug Fix (user-visible misbehavior in official stable or prestable release)
* Backported in [#45729](https://github.com/ClickHouse/ClickHouse/issues/45729): Fix key description when encountering duplicate primary keys. This can happen in projections. See [#45590](https://github.com/ClickHouse/ClickHouse/issues/45590) for details. [#45686](https://github.com/ClickHouse/ClickHouse/pull/45686) ([Amos Bird](https://github.com/amosbird)).
* Backported in [#46398](https://github.com/ClickHouse/ClickHouse/issues/46398): Fix `SYSTEM UNFREEZE` queries failing with the exception `CANNOT_PARSE_INPUT_ASSERTION_FAILED`. [#46325](https://github.com/ClickHouse/ClickHouse/pull/46325) ([Aleksei Filatov](https://github.com/aalexfvk)).
* Backported in [#46903](https://github.com/ClickHouse/ClickHouse/issues/46903): - Fix incorrect alias recursion in QueryNormalizer. [#46609](https://github.com/ClickHouse/ClickHouse/pull/46609) ([Raúl Marín](https://github.com/Algunenano)).
* Backported in [#47210](https://github.com/ClickHouse/ClickHouse/issues/47210): `INSERT` queries through native TCP protocol and HTTP protocol were not canceled correctly in some cases. It could lead to a partially applied query if a client canceled the query, or if a client died or, in rare cases, on network errors. As a result, it could lead to not working deduplication. Fixes [#27667](https://github.com/ClickHouse/ClickHouse/issues/27667) and [#45377](https://github.com/ClickHouse/ClickHouse/issues/45377). [#46681](https://github.com/ClickHouse/ClickHouse/pull/46681) ([Alexander Tokmakov](https://github.com/tavplubix)).
* Backported in [#47157](https://github.com/ClickHouse/ClickHouse/issues/47157): - Fix arithmetic operations in aggregate optimization with `min` and `max`. [#46705](https://github.com/ClickHouse/ClickHouse/pull/46705) ([Duc Canh Le](https://github.com/canhld94)).
* Backported in [#46881](https://github.com/ClickHouse/ClickHouse/issues/46881): Fix MSan report in the `maxIntersections` function. This closes [#43126](https://github.com/ClickHouse/ClickHouse/issues/43126). [#46847](https://github.com/ClickHouse/ClickHouse/pull/46847) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
* Backported in [#47359](https://github.com/ClickHouse/ClickHouse/issues/47359): Fix possible deadlock on distributed query cancellation. [#47161](https://github.com/ClickHouse/ClickHouse/pull/47161) ([Kruglov Pavel](https://github.com/Avogar)).
#### NOT FOR CHANGELOG / INSIGNIFICANT
* Use /etc/default/clickhouse in systemd too [#47003](https://github.com/ClickHouse/ClickHouse/pull/47003) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Update typing for a new PyGithub version [#47123](https://github.com/ClickHouse/ClickHouse/pull/47123) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Follow-up to [#46681](https://github.com/ClickHouse/ClickHouse/issues/46681) [#47284](https://github.com/ClickHouse/ClickHouse/pull/47284) ([Alexander Tokmakov](https://github.com/tavplubix)).
* Add a manual trigger for release workflow [#47302](https://github.com/ClickHouse/ClickHouse/pull/47302) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).

40
docs/changelogs/v22.8.14.53-lts.md Normal file
View File

@ -0,0 +1,40 @@
---
sidebar_position: 1
sidebar_label: 2023
---
# 2023 Changelog
### ClickHouse release v22.8.14.53-lts (4ea67c40077) FIXME as compared to v22.8.13.20-lts (e4817946d18)
#### Performance Improvement
* Backported in [#45845](https://github.com/ClickHouse/ClickHouse/issues/45845): Fixed performance of short `SELECT` queries that read from tables with large number of`Array`/`Map`/`Nested` columns. [#45630](https://github.com/ClickHouse/ClickHouse/pull/45630) ([Anton Popov](https://github.com/CurtizJ)).
* Backported in [#46374](https://github.com/ClickHouse/ClickHouse/issues/46374): Fix too big memory usage for vertical merges on non-remote disk. Respect `max_insert_delayed_streams_for_parallel_write` for the remote disk. [#46275](https://github.com/ClickHouse/ClickHouse/pull/46275) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
* Backported in [#46358](https://github.com/ClickHouse/ClickHouse/issues/46358): Allow using Vertical merge algorithm with parts in Compact format. This will allow ClickHouse server to use much less memory for background operations. This closes [#46084](https://github.com/ClickHouse/ClickHouse/issues/46084). [#46282](https://github.com/ClickHouse/ClickHouse/pull/46282) ([Anton Popov](https://github.com/CurtizJ)).
#### Build/Testing/Packaging Improvement
* Backported in [#46112](https://github.com/ClickHouse/ClickHouse/issues/46112): Remove the dependency on the `adduser` tool from the packages, because we don't use it. This fixes [#44934](https://github.com/ClickHouse/ClickHouse/issues/44934). [#45011](https://github.com/ClickHouse/ClickHouse/pull/45011) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
* Backported in [#46482](https://github.com/ClickHouse/ClickHouse/issues/46482): Get rid of unnecessary build for standalone clickhouse-keeper. [#46367](https://github.com/ClickHouse/ClickHouse/pull/46367) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Backported in [#46505](https://github.com/ClickHouse/ClickHouse/issues/46505): Some time ago the ccache compression was changed to `zst`, but `gz` archives are downloaded by default. It fixes it by prioritizing zst archive. [#46490](https://github.com/ClickHouse/ClickHouse/pull/46490) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
#### Bug Fix (user-visible misbehavior in official stable or prestable release)
* Backported in [#45908](https://github.com/ClickHouse/ClickHouse/issues/45908): Fixed bug with non-parsable default value for EPHEMERAL column in table metadata. [#44026](https://github.com/ClickHouse/ClickHouse/pull/44026) ([Yakov Olkhovskiy](https://github.com/yakov-olkhovskiy)).
* Backported in [#46238](https://github.com/ClickHouse/ClickHouse/issues/46238): A couple of seg faults have been reported around `c-ares`. All of the recent stack traces observed fail on inserting into `std::unodered_set<>`. I believe I have found the root cause of this, it seems to be unprocessed queries. Prior to this PR, CH calls `poll` to wait on the file descriptors in the `c-ares` channel. According to the [poll docs](https://man7.org/linux/man-pages/man2/poll.2.html), a negative return value means an error has ocurred. Because of this, we would abort the execution and return failure. The problem is that `poll` will also return a negative value if a system interrupt occurs. A system interrupt does not mean the processing has failed or ended, but we would abort it anyways because we were checking for negative values. Once the execution is aborted, the whole stack is destroyed, which includes the `std::unordered_set<std::string>` passed to the `void *` parameter of the c-ares callback. Once c-ares completed the request, the callback would be invoked and would access an invalid memory address causing a segfault. [#45629](https://github.com/ClickHouse/ClickHouse/pull/45629) ([Arthur Passos](https://github.com/arthurpassos)).
* Backported in [#45727](https://github.com/ClickHouse/ClickHouse/issues/45727): Fix key description when encountering duplicate primary keys. This can happen in projections. See [#45590](https://github.com/ClickHouse/ClickHouse/issues/45590) for details. [#45686](https://github.com/ClickHouse/ClickHouse/pull/45686) ([Amos Bird](https://github.com/amosbird)).
* Backported in [#46394](https://github.com/ClickHouse/ClickHouse/issues/46394): Fix `SYSTEM UNFREEZE` queries failing with the exception `CANNOT_PARSE_INPUT_ASSERTION_FAILED`. [#46325](https://github.com/ClickHouse/ClickHouse/pull/46325) ([Aleksei Filatov](https://github.com/aalexfvk)).
* Backported in [#46442](https://github.com/ClickHouse/ClickHouse/issues/46442): Fix possible `LOGICAL_ERROR` in asynchronous inserts with invalid data sent in format `VALUES`. [#46350](https://github.com/ClickHouse/ClickHouse/pull/46350) ([Anton Popov](https://github.com/CurtizJ)).
* Backported in [#46674](https://github.com/ClickHouse/ClickHouse/issues/46674): Fix an invalid processing of constant `LowCardinality` argument in function `arrayMap`. This bug could lead to a segfault in release, and logical error `Bad cast` in debug build. [#46569](https://github.com/ClickHouse/ClickHouse/pull/46569) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
* Backported in [#46879](https://github.com/ClickHouse/ClickHouse/issues/46879): Fix MSan report in the `maxIntersections` function. This closes [#43126](https://github.com/ClickHouse/ClickHouse/issues/43126). [#46847](https://github.com/ClickHouse/ClickHouse/pull/46847) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
* Backported in [#46871](https://github.com/ClickHouse/ClickHouse/issues/46871): Fix a bug in the `Map` data type. This closes [#46855](https://github.com/ClickHouse/ClickHouse/issues/46855). [#46856](https://github.com/ClickHouse/ClickHouse/pull/46856) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
#### NOT FOR CHANGELOG / INSIGNIFICANT
* Another attempt to fix automerge, or at least to have debug footprint [#45476](https://github.com/ClickHouse/ClickHouse/pull/45476) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Add check for running workflows to merge_pr.py [#45803](https://github.com/ClickHouse/ClickHouse/pull/45803) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Get rid of progress timestamps in release publishing [#45818](https://github.com/ClickHouse/ClickHouse/pull/45818) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Add necessary dependency for sanitizers [#45959](https://github.com/ClickHouse/ClickHouse/pull/45959) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Add helping logging to auto-merge script [#46080](https://github.com/ClickHouse/ClickHouse/pull/46080) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Fix write buffer destruction order for vertical merge. [#46205](https://github.com/ClickHouse/ClickHouse/pull/46205) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
* Get rid of legacy DocsReleaseChecks [#46665](https://github.com/ClickHouse/ClickHouse/pull/46665) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).

28
docs/changelogs/v22.8.15.23-lts.md Normal file
View File

@ -0,0 +1,28 @@
---
sidebar_position: 1
sidebar_label: 2023
---
# 2023 Changelog
### ClickHouse release v22.8.15.23-lts (d36fa168bbf) FIXME as compared to v22.8.14.53-lts (4ea67c40077)
#### Improvement
* Backported in [#46981](https://github.com/ClickHouse/ClickHouse/issues/46981): - Apply `ALTER TABLE table_name ON CLUSTER cluster MOVE PARTITION|PART partition_expr TO DISK|VOLUME 'disk_name'` to all replicas. Because `ALTER TABLE t MOVE` is not replicated. [#46402](https://github.com/ClickHouse/ClickHouse/pull/46402) ([lizhuoyu5](https://github.com/lzydmxy)).
#### Bug Fix
* Backported in [#47336](https://github.com/ClickHouse/ClickHouse/issues/47336): Sometimes after changing a role that could be not reflected on the access rights of a user who uses that role. This PR fixes that. [#46772](https://github.com/ClickHouse/ClickHouse/pull/46772) ([Vitaly Baranov](https://github.com/vitlibar)).
#### Bug Fix (user-visible misbehavior in official stable or prestable release)
* Backported in [#46901](https://github.com/ClickHouse/ClickHouse/issues/46901): - Fix incorrect alias recursion in QueryNormalizer. [#46609](https://github.com/ClickHouse/ClickHouse/pull/46609) ([Raúl Marín](https://github.com/Algunenano)).
* Backported in [#47156](https://github.com/ClickHouse/ClickHouse/issues/47156): - Fix arithmetic operations in aggregate optimization with `min` and `max`. [#46705](https://github.com/ClickHouse/ClickHouse/pull/46705) ([Duc Canh Le](https://github.com/canhld94)).
* Backported in [#46987](https://github.com/ClickHouse/ClickHouse/issues/46987): Fix result of LIKE predicates which translate to substring searches and contain quoted non-LIKE metacharacters. [#46875](https://github.com/ClickHouse/ClickHouse/pull/46875) ([Robert Schulze](https://github.com/rschu1ze)).
* Backported in [#47357](https://github.com/ClickHouse/ClickHouse/issues/47357): Fix possible deadlock on distributed query cancellation. [#47161](https://github.com/ClickHouse/ClickHouse/pull/47161) ([Kruglov Pavel](https://github.com/Avogar)).
#### NOT FOR CHANGELOG / INSIGNIFICANT
* Reduce updates of Mergeable Check [#46781](https://github.com/ClickHouse/ClickHouse/pull/46781) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Update typing for a new PyGithub version [#47123](https://github.com/ClickHouse/ClickHouse/pull/47123) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Add a manual trigger for release workflow [#47302](https://github.com/ClickHouse/ClickHouse/pull/47302) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).

47
docs/changelogs/v23.1.4.58-stable.md Normal file
View File

@ -0,0 +1,47 @@
---
sidebar_position: 1
sidebar_label: 2023
---
# 2023 Changelog
### ClickHouse release v23.1.4.58-stable (9ed562163a5) FIXME as compared to v23.1.3.5-stable (548b494bcce)
#### Performance Improvement
* Backported in [#46380](https://github.com/ClickHouse/ClickHouse/issues/46380): Fix too big memory usage for vertical merges on non-remote disk. Respect `max_insert_delayed_streams_for_parallel_write` for the remote disk. [#46275](https://github.com/ClickHouse/ClickHouse/pull/46275) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
#### Improvement
* Backported in [#46985](https://github.com/ClickHouse/ClickHouse/issues/46985): - Apply `ALTER TABLE table_name ON CLUSTER cluster MOVE PARTITION|PART partition_expr TO DISK|VOLUME 'disk_name'` to all replicas. Because `ALTER TABLE t MOVE` is not replicated. [#46402](https://github.com/ClickHouse/ClickHouse/pull/46402) ([lizhuoyu5](https://github.com/lzydmxy)).
* Backported in [#46778](https://github.com/ClickHouse/ClickHouse/issues/46778): Backward compatibility for T64 codec support for IPv4. [#46747](https://github.com/ClickHouse/ClickHouse/pull/46747) ([Yakov Olkhovskiy](https://github.com/yakov-olkhovskiy)).
* Backported in [#47020](https://github.com/ClickHouse/ClickHouse/issues/47020): Allow IPv4 in range(). [#46995](https://github.com/ClickHouse/ClickHouse/pull/46995) ([Yakov Olkhovskiy](https://github.com/yakov-olkhovskiy)).
#### Build/Testing/Packaging Improvement
* Backported in [#46031](https://github.com/ClickHouse/ClickHouse/issues/46031): Add systemd.service file for clickhouse-keeper. Fixes [#44293](https://github.com/ClickHouse/ClickHouse/issues/44293). [#45568](https://github.com/ClickHouse/ClickHouse/pull/45568) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Backported in [#46477](https://github.com/ClickHouse/ClickHouse/issues/46477): Get rid of unnecessary build for standalone clickhouse-keeper. [#46367](https://github.com/ClickHouse/ClickHouse/pull/46367) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Backported in [#46511](https://github.com/ClickHouse/ClickHouse/issues/46511): Some time ago the ccache compression was changed to `zst`, but `gz` archives are downloaded by default. It fixes it by prioritizing zst archive. [#46490](https://github.com/ClickHouse/ClickHouse/pull/46490) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
#### Bug Fix (user-visible misbehavior in official stable or prestable release)
* Backported in [#46228](https://github.com/ClickHouse/ClickHouse/issues/46228): A couple of seg faults have been reported around `c-ares`. All of the recent stack traces observed fail on inserting into `std::unodered_set<>`. I believe I have found the root cause of this, it seems to be unprocessed queries. Prior to this PR, CH calls `poll` to wait on the file descriptors in the `c-ares` channel. According to the [poll docs](https://man7.org/linux/man-pages/man2/poll.2.html), a negative return value means an error has ocurred. Because of this, we would abort the execution and return failure. The problem is that `poll` will also return a negative value if a system interrupt occurs. A system interrupt does not mean the processing has failed or ended, but we would abort it anyways because we were checking for negative values. Once the execution is aborted, the whole stack is destroyed, which includes the `std::unordered_set<std::string>` passed to the `void *` parameter of the c-ares callback. Once c-ares completed the request, the callback would be invoked and would access an invalid memory address causing a segfault. [#45629](https://github.com/ClickHouse/ClickHouse/pull/45629) ([Arthur Passos](https://github.com/arthurpassos)).
* Backported in [#46967](https://github.com/ClickHouse/ClickHouse/issues/46967): Backward compatibility - allow implicit narrowing conversion from UInt64 to IPv4 - required for "INSERT ... VALUES ..." expression. [#45865](https://github.com/ClickHouse/ClickHouse/pull/45865) ([Yakov Olkhovskiy](https://github.com/yakov-olkhovskiy)).
* Backported in [#46220](https://github.com/ClickHouse/ClickHouse/issues/46220): Fix reading of non existing nested columns with multiple level in compact parts. [#46045](https://github.com/ClickHouse/ClickHouse/pull/46045) ([Azat Khuzhin](https://github.com/azat)).
* Backported in [#46751](https://github.com/ClickHouse/ClickHouse/issues/46751): Follow-up fix for Replace domain IP types (IPv4, IPv6) with native https://github.com/ClickHouse/ClickHouse/pull/43221. [#46087](https://github.com/ClickHouse/ClickHouse/pull/46087) ([Yakov Olkhovskiy](https://github.com/yakov-olkhovskiy)).
* Backported in [#46448](https://github.com/ClickHouse/ClickHouse/issues/46448): Fix possible `LOGICAL_ERROR` in asynchronous inserts with invalid data sent in format `VALUES`. [#46350](https://github.com/ClickHouse/ClickHouse/pull/46350) ([Anton Popov](https://github.com/CurtizJ)).
* Backported in [#46680](https://github.com/ClickHouse/ClickHouse/issues/46680): Fix an invalid processing of constant `LowCardinality` argument in function `arrayMap`. This bug could lead to a segfault in release, and logical error `Bad cast` in debug build. [#46569](https://github.com/ClickHouse/ClickHouse/pull/46569) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
* Backported in [#46873](https://github.com/ClickHouse/ClickHouse/issues/46873): Fix a bug in the `Map` data type. This closes [#46855](https://github.com/ClickHouse/ClickHouse/issues/46855). [#46856](https://github.com/ClickHouse/ClickHouse/pull/46856) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
* Backported in [#46956](https://github.com/ClickHouse/ClickHouse/issues/46956): Fix result of LIKE predicates which translate to substring searches and contain quoted non-LIKE metacharacters. [#46875](https://github.com/ClickHouse/ClickHouse/pull/46875) ([Robert Schulze](https://github.com/rschu1ze)).
#### NOT FOR CHANGELOG / INSIGNIFICANT
* Another attempt to fix automerge, or at least to have debug footprint [#45476](https://github.com/ClickHouse/ClickHouse/pull/45476) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Support DELETE ON CLUSTER [#45786](https://github.com/ClickHouse/ClickHouse/pull/45786) ([Alexander Gololobov](https://github.com/davenger)).
* Add check for running workflows to merge_pr.py [#45803](https://github.com/ClickHouse/ClickHouse/pull/45803) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Add necessary dependency for sanitizers [#45959](https://github.com/ClickHouse/ClickHouse/pull/45959) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Add helping logging to auto-merge script [#46080](https://github.com/ClickHouse/ClickHouse/pull/46080) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Fix write buffer destruction order for vertical merge. [#46205](https://github.com/ClickHouse/ClickHouse/pull/46205) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
* Wait for background tasks in ~UploadHelper [#46334](https://github.com/ClickHouse/ClickHouse/pull/46334) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
* Improve install_check.py [#46458](https://github.com/ClickHouse/ClickHouse/pull/46458) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Fix dependencies for InstallPackagesTestAarch64 [#46597](https://github.com/ClickHouse/ClickHouse/pull/46597) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Get rid of legacy DocsReleaseChecks [#46665](https://github.com/ClickHouse/ClickHouse/pull/46665) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Reduce updates of Mergeable Check [#46781](https://github.com/ClickHouse/ClickHouse/pull/46781) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).

28
docs/changelogs/v23.1.5.24-stable.md Normal file
View File

@ -0,0 +1,28 @@
---
sidebar_position: 1
sidebar_label: 2023
---
# 2023 Changelog
### ClickHouse release v23.1.5.24-stable (0e51b53ba99) FIXME as compared to v23.1.4.58-stable (9ed562163a5)
#### Build/Testing/Packaging Improvement
* Backported in [#47060](https://github.com/ClickHouse/ClickHouse/issues/47060): Fix error during server startup on old distros (e.g. Amazon Linux 2) and on ARM that glibc 2.28 symbols are not found. [#47008](https://github.com/ClickHouse/ClickHouse/pull/47008) ([Robert Schulze](https://github.com/rschu1ze)).
#### Bug Fix (user-visible misbehavior in official stable or prestable release)
* Backported in [#46401](https://github.com/ClickHouse/ClickHouse/issues/46401): Fix `SYSTEM UNFREEZE` queries failing with the exception `CANNOT_PARSE_INPUT_ASSERTION_FAILED`. [#46325](https://github.com/ClickHouse/ClickHouse/pull/46325) ([Aleksei Filatov](https://github.com/aalexfvk)).
* Backported in [#46905](https://github.com/ClickHouse/ClickHouse/issues/46905): - Fix incorrect alias recursion in QueryNormalizer. [#46609](https://github.com/ClickHouse/ClickHouse/pull/46609) ([Raúl Marín](https://github.com/Algunenano)).
* Backported in [#47211](https://github.com/ClickHouse/ClickHouse/issues/47211): `INSERT` queries through native TCP protocol and HTTP protocol were not canceled correctly in some cases. It could lead to a partially applied query if a client canceled the query, or if a client died or, in rare cases, on network errors. As a result, it could lead to not working deduplication. Fixes [#27667](https://github.com/ClickHouse/ClickHouse/issues/27667) and [#45377](https://github.com/ClickHouse/ClickHouse/issues/45377). [#46681](https://github.com/ClickHouse/ClickHouse/pull/46681) ([Alexander Tokmakov](https://github.com/tavplubix)).
* Backported in [#47118](https://github.com/ClickHouse/ClickHouse/issues/47118): - Fix arithmetic operations in aggregate optimization with `min` and `max`. [#46705](https://github.com/ClickHouse/ClickHouse/pull/46705) ([Duc Canh Le](https://github.com/canhld94)).
* Backported in [#46883](https://github.com/ClickHouse/ClickHouse/issues/46883): Fix MSan report in the `maxIntersections` function. This closes [#43126](https://github.com/ClickHouse/ClickHouse/issues/43126). [#46847](https://github.com/ClickHouse/ClickHouse/pull/46847) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
* Backported in [#47361](https://github.com/ClickHouse/ClickHouse/issues/47361): Fix possible deadlock on distributed query cancellation. [#47161](https://github.com/ClickHouse/ClickHouse/pull/47161) ([Kruglov Pavel](https://github.com/Avogar)).
#### NOT FOR CHANGELOG / INSIGNIFICANT
* Use /etc/default/clickhouse in systemd too [#47003](https://github.com/ClickHouse/ClickHouse/pull/47003) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Update typing for a new PyGithub version [#47123](https://github.com/ClickHouse/ClickHouse/pull/47123) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* Follow-up to [#46681](https://github.com/ClickHouse/ClickHouse/issues/46681) [#47284](https://github.com/ClickHouse/ClickHouse/pull/47284) ([Alexander Tokmakov](https://github.com/tavplubix)).
* Add a manual trigger for release workflow [#47302](https://github.com/ClickHouse/ClickHouse/pull/47302) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).

30
docs/changelogs/v23.2.2.20-stable.md Normal file
View File

@ -0,0 +1,30 @@
---
sidebar_position: 1
sidebar_label: 2023
---
# 2023 Changelog
### ClickHouse release v23.2.2.20-stable (f6c269c8df2) FIXME as compared to v23.2.1.2537-stable (52bf836e03a)
#### Improvement
* Backported in [#46914](https://github.com/ClickHouse/ClickHouse/issues/46914): Allow PREWHERE for Merge with different DEFAULT expression for column. [#46831](https://github.com/ClickHouse/ClickHouse/pull/46831) ([Azat Khuzhin](https://github.com/azat)).
* Backported in [#47022](https://github.com/ClickHouse/ClickHouse/issues/47022): Allow IPv4 in range(). [#46995](https://github.com/ClickHouse/ClickHouse/pull/46995) ([Yakov Olkhovskiy](https://github.com/yakov-olkhovskiy)).
#### Bug Fix
* Backported in [#46828](https://github.com/ClickHouse/ClickHouse/issues/46828): Combined PREWHERE column accumulated from multiple PREWHERE in some cases didn't contain 0's from previous steps. The fix is to apply final filter if we know that it wasn't applied from more than 1 last step. [#46785](https://github.com/ClickHouse/ClickHouse/pull/46785) ([Alexander Gololobov](https://github.com/davenger)).
#### Build/Testing/Packaging Improvement
* Backported in [#47062](https://github.com/ClickHouse/ClickHouse/issues/47062): Fix error during server startup on old distros (e.g. Amazon Linux 2) and on ARM that glibc 2.28 symbols are not found. [#47008](https://github.com/ClickHouse/ClickHouse/pull/47008) ([Robert Schulze](https://github.com/rschu1ze)).
#### Bug Fix (user-visible misbehavior in official stable or prestable release)
* Backported in [#46895](https://github.com/ClickHouse/ClickHouse/issues/46895): Fixed a bug in automatic retries of `DROP TABLE` query with `ReplicatedMergeTree` tables and `Atomic` databases. In rare cases it could lead to `Can't get data for node /zk_path/log_pointer` and `The specified key does not exist` errors if ZooKeeper session expired during DROP and a new replicated table with the same path in ZooKeeper was created in parallel. [#46384](https://github.com/ClickHouse/ClickHouse/pull/46384) ([Alexander Tokmakov](https://github.com/tavplubix)).
* Backported in [#46865](https://github.com/ClickHouse/ClickHouse/issues/46865): Fix a bug in the `Map` data type. This closes [#46855](https://github.com/ClickHouse/ClickHouse/issues/46855). [#46856](https://github.com/ClickHouse/ClickHouse/pull/46856) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
* Backported in [#46958](https://github.com/ClickHouse/ClickHouse/issues/46958): Fix result of LIKE predicates which translate to substring searches and contain quoted non-LIKE metacharacters. [#46875](https://github.com/ClickHouse/ClickHouse/pull/46875) ([Robert Schulze](https://github.com/rschu1ze)).
#### NOT FOR CHANGELOG / INSIGNIFICANT
* More concise logging at trace level for PREWHERE steps [#46771](https://github.com/ClickHouse/ClickHouse/pull/46771) ([Alexander Gololobov](https://github.com/davenger)).
* Reduce updates of Mergeable Check [#46781](https://github.com/ClickHouse/ClickHouse/pull/46781) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).

23
docs/changelogs/v23.2.3.17-stable.md Normal file
View File

@ -0,0 +1,23 @@
---
sidebar_position: 1
sidebar_label: 2023
---
# 2023 Changelog
### ClickHouse release v23.2.3.17-stable (dec18bf7281) FIXME as compared to v23.2.2.20-stable (f6c269c8df2)
#### Bug Fix (user-visible misbehavior in official stable or prestable release)
* Backported in [#46907](https://github.com/ClickHouse/ClickHouse/issues/46907): - Fix incorrect alias recursion in QueryNormalizer. [#46609](https://github.com/ClickHouse/ClickHouse/pull/46609) ([Raúl Marín](https://github.com/Algunenano)).
* Backported in [#47091](https://github.com/ClickHouse/ClickHouse/issues/47091): - Fix arithmetic operations in aggregate optimization with `min` and `max`. [#46705](https://github.com/ClickHouse/ClickHouse/pull/46705) ([Duc Canh Le](https://github.com/canhld94)).
* Backported in [#46885](https://github.com/ClickHouse/ClickHouse/issues/46885): Fix MSan report in the `maxIntersections` function. This closes [#43126](https://github.com/ClickHouse/ClickHouse/issues/43126). [#46847](https://github.com/ClickHouse/ClickHouse/pull/46847) ([Alexey Milovidov](https://github.com/alexey-milovidov)).
* Backported in [#47067](https://github.com/ClickHouse/ClickHouse/issues/47067): Fix typo in systemd service, which causes the systemd service start to fail. [#47051](https://github.com/ClickHouse/ClickHouse/pull/47051) ([Palash Goel](https://github.com/palash-goel)).
* Backported in [#47259](https://github.com/ClickHouse/ClickHouse/issues/47259): Fix concrete columns PREWHERE support. [#47154](https://github.com/ClickHouse/ClickHouse/pull/47154) ([Azat Khuzhin](https://github.com/azat)).
#### NOT FOR CHANGELOG / INSIGNIFICANT
* Use /etc/default/clickhouse in systemd too [#47003](https://github.com/ClickHouse/ClickHouse/pull/47003) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).
* do flushUntrackedMemory when context switches [#47102](https://github.com/ClickHouse/ClickHouse/pull/47102) ([Sema Checherinda](https://github.com/CheSema)).
* Update typing for a new PyGithub version [#47123](https://github.com/ClickHouse/ClickHouse/pull/47123) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).

20
docs/changelogs/v23.2.4.12-stable.md Normal file
View File

@ -0,0 +1,20 @@
---
sidebar_position: 1
sidebar_label: 2023
---
# 2023 Changelog
### ClickHouse release v23.2.4.12-stable (8fe866cb035) FIXME as compared to v23.2.3.17-stable (dec18bf7281)
#### Bug Fix (user-visible misbehavior in official stable or prestable release)
* Backported in [#47277](https://github.com/ClickHouse/ClickHouse/issues/47277): Fix IPv4/IPv6 serialization/deserialization in binary formats that was broken in https://github.com/ClickHouse/ClickHouse/pull/43221. Closes [#46522](https://github.com/ClickHouse/ClickHouse/issues/46522). [#46616](https://github.com/ClickHouse/ClickHouse/pull/46616) ([Kruglov Pavel](https://github.com/Avogar)).
* Backported in [#47212](https://github.com/ClickHouse/ClickHouse/issues/47212): `INSERT` queries through native TCP protocol and HTTP protocol were not canceled correctly in some cases. It could lead to a partially applied query if a client canceled the query, or if a client died or, in rare cases, on network errors. As a result, it could lead to not working deduplication. Fixes [#27667](https://github.com/ClickHouse/ClickHouse/issues/27667) and [#45377](https://github.com/ClickHouse/ClickHouse/issues/45377). [#46681](https://github.com/ClickHouse/ClickHouse/pull/46681) ([Alexander Tokmakov](https://github.com/tavplubix)).
* Backported in [#47363](https://github.com/ClickHouse/ClickHouse/issues/47363): Fix possible deadlock on distributed query cancellation. [#47161](https://github.com/ClickHouse/ClickHouse/pull/47161) ([Kruglov Pavel](https://github.com/Avogar)).
#### NOT FOR CHANGELOG / INSIGNIFICANT
* Follow-up to [#46681](https://github.com/ClickHouse/ClickHouse/issues/46681) [#47284](https://github.com/ClickHouse/ClickHouse/pull/47284) ([Alexander Tokmakov](https://github.com/tavplubix)).
* Add a manual trigger for release workflow [#47302](https://github.com/ClickHouse/ClickHouse/pull/47302) ([Mikhail f. Shiryaev](https://github.com/Felixoid)).

2
docs/en/development/architecture.md
View File

@ -172,7 +172,7 @@ Global thread pool is `GlobalThreadPool` singleton class. To allocate thread fro
Global pool is universal and all pools described below are implemented on top of it. This can be thought of as a hierarchy of pools. Any specialized pool takes its threads from the global pool using `ThreadPool` class. So the main purpose of any specialized pool is to apply limit on the number of simultaneous jobs and do job scheduling. If there are more jobs scheduled than threads in a pool, `ThreadPool` accumulates jobs in a queue with priorities. Each job has an integer priority. Default priority is zero. All jobs with higher priority values are started before any job with lower priority value. But there is no difference between already executing jobs, thus priority matters only when the pool in overloaded.
IO thread pool is implemented as a plain `ThreadPool` accessible via `IOThreadPool::get()` method. It is configured in the same way as global pool with `max_io_thread_pool_size`, `max_io_thread_pool_free_size` and `io_thread_pool_queue_size` settings. The main purpose of IO thread pool is to avoid exhaustion of the global pool with IO jobs, which could prevent queries from fully utilizing CPU.
IO thread pool is implemented as a plain `ThreadPool` accessible via `IOThreadPool::get()` method. It is configured in the same way as global pool with `max_io_thread_pool_size`, `max_io_thread_pool_free_size` and `io_thread_pool_queue_size` settings. The main purpose of IO thread pool is to avoid exhaustion of the global pool with IO jobs, which could prevent queries from fully utilizing CPU. Backup to S3 does significant amount of IO operations and to avoid impact on interactive queries there is a separate `BackupsIOThreadPool` configured with `max_backups_io_thread_pool_size`, `max_backups_io_thread_pool_free_size` and `backups_io_thread_pool_queue_size` settings.
For periodic task execution there is `BackgroundSchedulePool` class. You can register tasks using `BackgroundSchedulePool::TaskHolder` objects and the pool ensures that no task runs two jobs at the same time. It also allows you to postpone task execution to a specific instant in the future or temporarily deactivate task. Global `Context` provides a few instances of this class for different purposes. For general purpose tasks `Context::getSchedulePool()` is used.

123
docs/en/development/build-cross-s390x.md Normal file
View File

@ -0,0 +1,123 @@
---
slug: /en/development/build-cross-s390x
sidebar_position: 69
title: How to Build, Run and Debug ClickHouse on Linux for s390x (zLinux)
sidebar_label: Build on Linux for s390x (zLinux)
---
As of writing (2023/3/10) building for s390x considered to be experimental. Not all features can be enabled, has broken features and is currently under active development.
## Building
As s390x does not support boringssl, it uses OpenSSL and has two related build options.
- By default, the s390x build will dynamically link to OpenSSL libraries. It will build OpenSSL shared objects, so it's not necessary to install OpenSSL beforehand. (This option is recommended in all cases.)
- Another option is to build OpenSSL in-tree. In this case two build flags need to be supplied to cmake
```bash
-DENABLE_OPENSSL_DYNAMIC=0 -DENABLE_OPENSSL=1
```
These instructions assume that the host machine is x86_64 and has all the tooling required to build natively based on the [build instructions](../development/build.md). It also assumes that the host is Ubuntu 22.04 but the following instructions should also work on Ubuntu 20.04.
In addition to installing the tooling used to build natively, the following additional packages need to be installed:
```bash
apt-get install binutils-s390x-linux-gnu libc6-dev-s390x-cross gcc-s390x-linux-gnu binfmt-support qemu-user-static
```
If you wish to cross compile rust code install the rust cross compile target for s390x:
```bash
rustup target add s390x-unknown-linux-gnu
```
To build for s390x:
```bash
cmake -DCMAKE_TOOLCHAIN_FILE=cmake/linux/toolchain-s390x.cmake ..
ninja
```
## Running
Once built, the binary can be run with, eg.:
```bash
qemu-s390x-static -L /usr/s390x-linux-gnu ./clickhouse
```
## Debugging
Install LLDB:
```bash
apt-get install lldb-15
```
To Debug a s390x executable, run clickhouse using QEMU in debug mode:
```bash
qemu-s390x-static -g 31338 -L /usr/s390x-linux-gnu ./clickhouse
```
In another shell run LLDB and attach, replace `<Clickhouse Parent Directory>` and `<build directory>` with the values corresponding to your environment.
```bash
lldb-15
(lldb) target create ./clickhouse
Current executable set to '/<Clickhouse Parent Directory>/ClickHouse/<build directory>/programs/clickhouse' (s390x).
(lldb) settings set target.source-map <build directory> /<Clickhouse Parent Directory>/ClickHouse
(lldb) gdb-remote 31338
Process 1 stopped
* thread #1, stop reason = signal SIGTRAP
frame #0: 0x0000004020e74cd0
-> 0x4020e74cd0: lgr %r2, %r15
0x4020e74cd4: aghi %r15, -160
0x4020e74cd8: xc 0(8,%r15), 0(%r15)
0x4020e74cde: brasl %r14, 275429939040
(lldb) b main
Breakpoint 1: 9 locations.
(lldb) c
Process 1 resuming
Process 1 stopped
* thread #1, stop reason = breakpoint 1.1
frame #0: 0x0000004005cd9fc0 clickhouse`main(argc_=1, argv_=0x0000004020e594a8) at main.cpp:450:17
447 #if !defined(FUZZING_MODE)
448 int main(int argc_, char ** argv_)
449 {
-> 450 inside_main = true;
451 SCOPE_EXIT({ inside_main = false; });
452
453 /// PHDR cache is required for query profiler to work reliably
```
## Visual Studio Code integration
- (CodeLLDB extension)[https://github.com/vadimcn/vscode-lldb] is required for visual debugging, the (Command Variable)[https://github.com/rioj7/command-variable] extension can help dynamic launches if using (cmake variants)[https://github.com/microsoft/vscode-cmake-tools/blob/main/docs/variants.md].
- Make sure to set the backend to your llvm installation eg. `"lldb.library": "/usr/lib/x86_64-linux-gnu/liblldb-15.so"`
- Launcher:
```json
{
"version": "0.2.0",
"configurations": [
{
"name": "Debug",
"type": "lldb",
"request": "custom",
"targetCreateCommands": ["target create ${command:cmake.launchTargetDirectory}/clickhouse"],
"processCreateCommands": ["settings set target.source-map ${input:targetdir} ${workspaceFolder}", "gdb-remote 31338"],
"sourceMap": { "${input:targetdir}": "${workspaceFolder}" },
}
],
"inputs": [
{
"id": "targetdir",
"type": "command",
"command": "extension.commandvariable.transform",
"args": {
"text": "${command:cmake.launchTargetDirectory}",
"find": ".*/([^/]+)/[^/]+$",
"replace": "$1"
}
}
]
}
```
- Make sure to run the clickhouse executable in debug mode prior to launch. (It is also possible to create a `preLaunchTask` that automates this)

2
docs/en/development/contrib.md
View File

@ -1,6 +1,6 @@
---
slug: /en/development/contrib
sidebar_position: 71
sidebar_position: 72
sidebar_label: Third-Party Libraries
description: A list of third-party libraries used
---

2
docs/en/development/developer-instruction.md
View File

@ -67,7 +67,7 @@ It generally means that the SSH keys for connecting to GitHub are missing. These
You can also clone the repository via https protocol:
git clone --recursive--shallow-submodules https://github.com/ClickHouse/ClickHouse.git
git clone --recursive --shallow-submodules https://github.com/ClickHouse/ClickHouse.git
This, however, will not let you send your changes to the server. You can still use it temporarily and add the SSH keys later replacing the remote address of the repository with `git remote` command.

2
docs/en/development/style.md
View File

@ -1,6 +1,6 @@
---
slug: /en/development/style
sidebar_position: 69
sidebar_position: 70
sidebar_label: C++ Guide
description: A list of recommendations regarding coding style, naming convention, formatting and more
---

5
docs/en/development/tests.md
View File

@ -1,6 +1,6 @@
---
slug: /en/development/tests
sidebar_position: 70
sidebar_position: 71
sidebar_label: Testing
title: ClickHouse Testing
description: Most of ClickHouse features can be tested with functional tests and they are mandatory to use for every change in ClickHouse code that can be tested that way.
@ -31,6 +31,9 @@ folder and run the following command:
PATH=$PATH:<path to clickhouse-client> tests/clickhouse-test 01428_hash_set_nan_key
```
Test results (`stderr` and `stdout`) are written to files `01428_hash_set_nan_key.[stderr|stdout]` which
are located near the test file itself (so for `queries/0_stateless/foo.sql` output will be in `queries/0_stateless/foo.stdout`).
For more options, see `tests/clickhouse-test --help`. You can simply run all tests or run subset of tests filtered by substring in test name: `./clickhouse-test substring`. There are also options to run tests in parallel or in randomized order.
### Adding a New Test

8
docs/en/engines/table-engines/integrations/kafka.md
View File

@ -19,8 +19,8 @@ Kafka lets you:
``` sql
CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster]
(
name1 [type1] [DEFAULT|MATERIALIZED|ALIAS expr1],
name2 [type2] [DEFAULT|MATERIALIZED|ALIAS expr2],
name1 [type1],
name2 [type2],
...
) ENGINE = Kafka()
SETTINGS
@ -113,6 +113,10 @@ Kafka(kafka_broker_list, kafka_topic_list, kafka_group_name, kafka_format
</details>
:::info
The Kafka table engine doesn't support columns with [default value](../../../sql-reference/statements/create/table.md#default_value). If you need columns with default value, you can add them at materialized view level (see below).
:::
## Description {#description}
The delivered messages are tracked automatically, so each message in a group is only counted once. If you want to get the data twice, then create a copy of the table with another group name.

45
docs/en/engines/table-engines/mergetree-family/mergetree.md
View File

@ -450,29 +450,32 @@ INDEX sample_index3 (lower(str), str) TYPE ngrambf_v1(3, 256, 2, 0) GRANULARITY
Conditions in the `WHERE` clause contains calls of the functions that operate with columns. If the column is a part of an index, ClickHouse tries to use this index when performing the functions. ClickHouse supports different subsets of functions for using indexes.
The `set` index can be used with all functions. Function subsets for other indexes are shown in the table below.
Indexes of type `set` can be utilized by all functions. The other index types are supported as follows:
| Function (operator) / Index | primary key | minmax | ngrambf_v1 | tokenbf_v1 | bloom_filter |
|------------------------------------------------------------------------------------------------------------|-------------|--------|-------------|-------------|---------------|
| [equals (=, ==)](/docs/en/sql-reference/functions/comparison-functions.md/#function-equals) | ✔ | ✔ | ✔ | ✔ | ✔ |
| [notEquals(!=, &lt;&gt;)](/docs/en/sql-reference/functions/comparison-functions.md/#function-notequals) | ✔ | ✔ | ✔ | ✔ | ✔ |
| [like](/docs/en/sql-reference/functions/string-search-functions.md/#function-like) | ✔ | ✔ | ✔ | ✔ | ✗ |
| [notLike](/docs/en/sql-reference/functions/string-search-functions.md/#function-notlike) | ✔ | ✔ | ✔ | ✔ | ✗ |
| [startsWith](/docs/en/sql-reference/functions/string-functions.md/#startswith) | ✔ | ✔ | ✔ | ✔ | ✗ |
| [endsWith](/docs/en/sql-reference/functions/string-functions.md/#endswith) | ✗ | ✗ | ✔ | ✔ | ✗ |
| [multiSearchAny](/docs/en/sql-reference/functions/string-search-functions.md/#function-multisearchany) | ✗ | ✗ | ✔ | ✗ | ✗ |
| [in](/docs/en/sql-reference/functions/in-functions#in-functions) | ✔ | ✔ | ✔ | ✔ | ✔ |
| [notIn](/docs/en/sql-reference/functions/in-functions#in-functions) | ✔ | ✔ | ✔ | ✔ | ✔ |
| [less (<)](/docs/en/sql-reference/functions/comparison-functions.md/#function-less) | ✔ | ✔ | ✗ | ✗ | ✗ |
| [greater (>)](/docs/en/sql-reference/functions/comparison-functions.md/#function-greater) | ✔ | ✔ | ✗ | ✗ | ✗ |
| [lessOrEquals (<=)](/docs/en/sql-reference/functions/comparison-functions.md/#function-lessorequals) | ✔ | ✔ | ✗ | ✗ | ✗ |
| [greaterOrEquals (>=)](/docs/en/sql-reference/functions/comparison-functions.md/#function-greaterorequals) | ✔ | ✔ | ✗ | ✗ | ✗ |
| [empty](/docs/en/sql-reference/functions/array-functions#function-empty) | ✔ | ✔ | ✗ | ✗ | ✗ |
| [notEmpty](/docs/en/sql-reference/functions/array-functions#function-notempty) | ✔ | ✔ | ✗ | ✗ | ✗ |
| hasToken | ✗ | ✗ | ✗ | ✔ | ✗ |
| hasTokenOrNull | ✗ | ✗ | ✗ | ✔ | ✗ |
| hasTokenCaseInsensitive | ✗ | ✗ | ✗ | ✔ | ✗ |
| hasTokenCaseInsensitiveOrNull | ✗ | ✗ | ✗ | ✔ | ✗ |
|------------------------------------------------------------------------------------------------------------|-------------|--------|------------|------------|--------------|
| [equals (=, ==)](/docs/en/sql-reference/functions/comparison-functions.md/#function-equals) | ✔ | ✔ | ✔ | ✔ | ✔ |
| [notEquals(!=, &lt;&gt;)](/docs/en/sql-reference/functions/comparison-functions.md/#function-notequals) | ✔ | ✔ | ✔ | ✔ | ✔ |
| [like](/docs/en/sql-reference/functions/string-search-functions.md/#function-like) | ✔ | ✔ | ✔ | ✔ | ✗ |
| [notLike](/docs/en/sql-reference/functions/string-search-functions.md/#function-notlike) | ✔ | ✔ | ✔ | ✔ | ✗ |
| [startsWith](/docs/en/sql-reference/functions/string-functions.md/#startswith) | ✔ | ✔ | ✔ | ✔ | ✗ |
| [endsWith](/docs/en/sql-reference/functions/string-functions.md/#endswith) | ✗ | ✗ | ✔ | ✔ | ✗ |
| [multiSearchAny](/docs/en/sql-reference/functions/string-search-functions.md/#function-multisearchany) | ✗ | ✗ | ✔ | ✗ | ✗ |
| [in](/docs/en/sql-reference/functions/in-functions#in-functions) | ✔ | ✔ | ✔ | ✔ | ✔ |
| [notIn](/docs/en/sql-reference/functions/in-functions#in-functions) | ✔ | ✔ | ✔ | ✔ | ✔ |
| [less (<)](/docs/en/sql-reference/functions/comparison-functions.md/#function-less) | ✔ | ✔ | ✗ | ✗ | ✗ |
| [greater (>)](/docs/en/sql-reference/functions/comparison-functions.md/#function-greater) | ✔ | ✔ | ✗ | ✗ | ✗ |
| [lessOrEquals (<=)](/docs/en/sql-reference/functions/comparison-functions.md/#function-lessorequals) | ✔ | ✔ | ✗ | ✗ | ✗ |
| [greaterOrEquals (>=)](/docs/en/sql-reference/functions/comparison-functions.md/#function-greaterorequals) | ✔ | ✔ | ✗ | ✗ | ✗ |
| [empty](/docs/en/sql-reference/functions/array-functions#function-empty) | ✔ | ✔ | ✗ | ✗ | ✗ |
| [notEmpty](/docs/en/sql-reference/functions/array-functions#function-notempty) | ✔ | ✔ | ✗ | ✗ | ✗ |
| [has](/docs/en/sql-reference/functions/array-functions#function-has) | ✗ | ✗ | ✔ | ✔ | ✔ |
| [hasAny](/docs/en/sql-reference/functions/array-functions#function-hasAny) | ✗ | ✗ | ✗ | ✗ | ✔ |
| [hasAll](/docs/en/sql-reference/functions/array-functions#function-hasAll) | ✗ | ✗ | ✗ | ✗ | ✔ |
| hasToken | ✗ | ✗ | ✗ | ✔ | ✗ |
| hasTokenOrNull | ✗ | ✗ | ✗ | ✔ | ✗ |
| hasTokenCaseInsensitive | ✗ | ✗ | ✗ | ✔ | ✗ |
| hasTokenCaseInsensitiveOrNull | ✗ | ✗ | ✗ | ✔ | ✗ |
Functions with a constant argument that is less than ngram size can’t be used by `ngrambf_v1` for query optimization.

88
docs/en/getting-started/example-datasets/nyc-taxi.md
View File

File diff suppressed because one or more lines are too long

212
docs/en/interfaces/formats.md
View File

@ -1232,50 +1232,52 @@ Each row is formatted as a single document and each column is formatted as a sin
For output it uses the following correspondence between ClickHouse types and BSON types:
| ClickHouse type | BSON Type |
|-----------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------|
| [Bool](/docs/en/sql-reference/data-types/boolean.md) | `\x08` boolean |
| [Int8/UInt8](/docs/en/sql-reference/data-types/int-uint.md) | `\x10` int32 |
| [Int16UInt16](/docs/en/sql-reference/data-types/int-uint.md) | `\x10` int32 |
| [Int32](/docs/en/sql-reference/data-types/int-uint.md) | `\x10` int32 |
| [UInt32](/docs/en/sql-reference/data-types/int-uint.md) | `\x12` int64 |
| [Int64/UInt64](/docs/en/sql-reference/data-types/int-uint.md) | `\x12` int64 |
| [Float32/Float64](/docs/en/sql-reference/data-types/float.md) | `\x01` double |
| [Date](/docs/en/sql-reference/data-types/date.md)/[Date32](/docs/en/sql-reference/data-types/date32.md) | `\x10` int32 |
| [DateTime](/docs/en/sql-reference/data-types/datetime.md) | `\x12` int64 |
| [DateTime64](/docs/en/sql-reference/data-types/datetime64.md) | `\x09` datetime |
| [Decimal32](/docs/en/sql-reference/data-types/decimal.md) | `\x10` int32 |
| [Decimal64](/docs/en/sql-reference/data-types/decimal.md) | `\x12` int64 |
| [Decimal128](/docs/en/sql-reference/data-types/decimal.md) | `\x05` binary, `\x00` binary subtype, size = 16 |
| [Decimal256](/docs/en/sql-reference/data-types/decimal.md) | `\x05` binary, `\x00` binary subtype, size = 32 |
| [Int128/UInt128](/docs/en/sql-reference/data-types/int-uint.md) | `\x05` binary, `\x00` binary subtype, size = 16 |
| [Int256/UInt256](/docs/en/sql-reference/data-types/int-uint.md) | `\x05` binary, `\x00` binary subtype, size = 32 |
| ClickHouse type | BSON Type |
|-----------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------|
| [Bool](/docs/en/sql-reference/data-types/boolean.md) | `\x08` boolean |
| [Int8/UInt8](/docs/en/sql-reference/data-types/int-uint.md) | `\x10` int32 |
| [Int16UInt16](/docs/en/sql-reference/data-types/int-uint.md) | `\x10` int32 |
| [Int32](/docs/en/sql-reference/data-types/int-uint.md) | `\x10` int32 |
| [UInt32](/docs/en/sql-reference/data-types/int-uint.md) | `\x12` int64 |
| [Int64/UInt64](/docs/en/sql-reference/data-types/int-uint.md) | `\x12` int64 |
| [Float32/Float64](/docs/en/sql-reference/data-types/float.md) | `\x01` double |
| [Date](/docs/en/sql-reference/data-types/date.md)/[Date32](/docs/en/sql-reference/data-types/date32.md) | `\x10` int32 |
| [DateTime](/docs/en/sql-reference/data-types/datetime.md) | `\x12` int64 |
| [DateTime64](/docs/en/sql-reference/data-types/datetime64.md) | `\x09` datetime |
| [Decimal32](/docs/en/sql-reference/data-types/decimal.md) | `\x10` int32 |
| [Decimal64](/docs/en/sql-reference/data-types/decimal.md) | `\x12` int64 |
| [Decimal128](/docs/en/sql-reference/data-types/decimal.md) | `\x05` binary, `\x00` binary subtype, size = 16 |
| [Decimal256](/docs/en/sql-reference/data-types/decimal.md) | `\x05` binary, `\x00` binary subtype, size = 32 |
| [Int128/UInt128](/docs/en/sql-reference/data-types/int-uint.md) | `\x05` binary, `\x00` binary subtype, size = 16 |
| [Int256/UInt256](/docs/en/sql-reference/data-types/int-uint.md) | `\x05` binary, `\x00` binary subtype, size = 32 |
| [String](/docs/en/sql-reference/data-types/string.md)/[FixedString](/docs/en/sql-reference/data-types/fixedstring.md) | `\x05` binary, `\x00` binary subtype or \x02 string if setting output_format_bson_string_as_string is enabled |
| [UUID](/docs/en/sql-reference/data-types/uuid.md) | `\x05` binary, `\x04` uuid subtype, size = 16 |
| [Array](/docs/en/sql-reference/data-types/array.md) | `\x04` array |
| [Tuple](/docs/en/sql-reference/data-types/tuple.md) | `\x04` array |
| [Named Tuple](/docs/en/sql-reference/data-types/tuple.md) | `\x03` document |
| [Map](/docs/en/sql-reference/data-types/map.md) (with String keys) | `\x03` document |
| [UUID](/docs/en/sql-reference/data-types/uuid.md) | `\x05` binary, `\x04` uuid subtype, size = 16 |
| [Array](/docs/en/sql-reference/data-types/array.md) | `\x04` array |
| [Tuple](/docs/en/sql-reference/data-types/tuple.md) | `\x04` array |
| [Named Tuple](/docs/en/sql-reference/data-types/tuple.md) | `\x03` document |
| [Map](/docs/en/sql-reference/data-types/map.md) (with String keys) | `\x03` document |
| [IPv4](/docs/en/sql-reference/data-types/domains/ipv4.md) | `\x10` int32 |
| [IPv6](/docs/en/sql-reference/data-types/domains/ipv6.md) | `\x05` binary, `\x00` binary subtype |
For input it uses the following correspondence between BSON types and ClickHouse types:
| BSON Type | ClickHouse Type |
|------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------|
| `\x01` double | [Float32/Float64](/docs/en/sql-reference/data-types/float.md) |
| `\x02` string | [String](/docs/en/sql-reference/data-types/string.md)/[FixedString](/docs/en/sql-reference/data-types/fixedstring.md) |
| `\x03` document | [Map](/docs/en/sql-reference/data-types/map.md)/[Named Tuple](/docs/en/sql-reference/data-types/tuple.md) |
| `\x04` array | [Array](/docs/en/sql-reference/data-types/array.md)/[Tuple](/docs/en/sql-reference/data-types/tuple.md) |
| `\x05` binary, `\x00` binary subtype | [String](/docs/en/sql-reference/data-types/string.md)/[FixedString](/docs/en/sql-reference/data-types/fixedstring.md) |
| `\x05` binary, `\x02` old binary subtype | [String](/docs/en/sql-reference/data-types/string.md)/[FixedString](/docs/en/sql-reference/data-types/fixedstring.md) |
| `\x05` binary, `\x03` old uuid subtype | [UUID](/docs/en/sql-reference/data-types/uuid.md) |
| `\x05` binary, `\x04` uuid subtype | [UUID](/docs/en/sql-reference/data-types/uuid.md) |
| `\x07` ObjectId | [String](/docs/en/sql-reference/data-types/string.md)/[FixedString](/docs/en/sql-reference/data-types/fixedstring.md) |
| `\x08` boolean | [Bool](/docs/en/sql-reference/data-types/boolean.md) |
| `\x09` datetime | [DateTime64](/docs/en/sql-reference/data-types/datetime64.md) |
| `\x0A` null value | [NULL](/docs/en/sql-reference/data-types/nullable.md) |
| `\x0D` JavaScript code | [String](/docs/en/sql-reference/data-types/string.md)/[FixedString](/docs/en/sql-reference/data-types/fixedstring.md) |
| `\x0E` symbol | [String](/docs/en/sql-reference/data-types/string.md)/[FixedString](/docs/en/sql-reference/data-types/fixedstring.md) |
| `\x10` int32 | [Int32/UInt32](/docs/en/sql-reference/data-types/int-uint.md)/[Decimal32](/docs/en/sql-reference/data-types/decimal.md) |
| BSON Type | ClickHouse Type |
|------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| `\x01` double | [Float32/Float64](/docs/en/sql-reference/data-types/float.md) |
| `\x02` string | [String](/docs/en/sql-reference/data-types/string.md)/[FixedString](/docs/en/sql-reference/data-types/fixedstring.md) |
| `\x03` document | [Map](/docs/en/sql-reference/data-types/map.md)/[Named Tuple](/docs/en/sql-reference/data-types/tuple.md) |
| `\x04` array | [Array](/docs/en/sql-reference/data-types/array.md)/[Tuple](/docs/en/sql-reference/data-types/tuple.md) |
| `\x05` binary, `\x00` binary subtype | [String](/docs/en/sql-reference/data-types/string.md)/[FixedString](/docs/en/sql-reference/data-types/fixedstring.md)/[IPv6](/docs/en/sql-reference/data-types/domains/ipv6.md) |
| `\x05` binary, `\x02` old binary subtype | [String](/docs/en/sql-reference/data-types/string.md)/[FixedString](/docs/en/sql-reference/data-types/fixedstring.md) |
| `\x05` binary, `\x03` old uuid subtype | [UUID](/docs/en/sql-reference/data-types/uuid.md) |
| `\x05` binary, `\x04` uuid subtype | [UUID](/docs/en/sql-reference/data-types/uuid.md) |
| `\x07` ObjectId | [String](/docs/en/sql-reference/data-types/string.md)/[FixedString](/docs/en/sql-reference/data-types/fixedstring.md) |
| `\x08` boolean | [Bool](/docs/en/sql-reference/data-types/boolean.md) |
| `\x09` datetime | [DateTime64](/docs/en/sql-reference/data-types/datetime64.md) |
| `\x0A` null value | [NULL](/docs/en/sql-reference/data-types/nullable.md) |
| `\x0D` JavaScript code | [String](/docs/en/sql-reference/data-types/string.md)/[FixedString](/docs/en/sql-reference/data-types/fixedstring.md) |
| `\x0E` symbol | [String](/docs/en/sql-reference/data-types/string.md)/[FixedString](/docs/en/sql-reference/data-types/fixedstring.md) |
| `\x10` int32 | [Int32/UInt32](/docs/en/sql-reference/data-types/int-uint.md)/[Decimal32](/docs/en/sql-reference/data-types/decimal.md)/[IPv4](/docs/en/sql-reference/data-types/domains/ipv4.md) |
| `\x12` int64 | [Int64/UInt64](/docs/en/sql-reference/data-types/int-uint.md)/[Decimal64](/docs/en/sql-reference/data-types/decimal.md)/[DateTime64](/docs/en/sql-reference/data-types/datetime64.md) |
Other BSON types are not supported. Also, it performs conversion between different integer types (for example, you can insert BSON int32 value into ClickHouse UInt8).
@ -1608,23 +1610,25 @@ See also [Format Schema](#formatschema).
The table below shows supported data types and how they match ClickHouse [data types](/docs/en/sql-reference/data-types/index.md) in `INSERT` and `SELECT` queries.
| CapnProto data type (`INSERT`) | ClickHouse data type | CapnProto data type (`SELECT`) |
|--------------------------------|-----------------------------------------------------------|--------------------------------|
| `UINT8`, `BOOL` | [UInt8](/docs/en/sql-reference/data-types/int-uint.md) | `UINT8` |
| `INT8` | [Int8](/docs/en/sql-reference/data-types/int-uint.md) | `INT8` |
| `UINT16` | [UInt16](/docs/en/sql-reference/data-types/int-uint.md), [Date](/docs/en/sql-reference/data-types/date.md) | `UINT16` |
| `INT16` | [Int16](/docs/en/sql-reference/data-types/int-uint.md) | `INT16` |
| `UINT32` | [UInt32](/docs/en/sql-reference/data-types/int-uint.md), [DateTime](/docs/en/sql-reference/data-types/datetime.md) | `UINT32` |
| `INT32` | [Int32](/docs/en/sql-reference/data-types/int-uint.md) | `INT32` |
| `UINT64` | [UInt64](/docs/en/sql-reference/data-types/int-uint.md) | `UINT64` |
| `INT64` | [Int64](/docs/en/sql-reference/data-types/int-uint.md), [DateTime64](/docs/en/sql-reference/data-types/datetime.md) | `INT64` |
| `FLOAT32` | [Float32](/docs/en/sql-reference/data-types/float.md) | `FLOAT32` |
| `FLOAT64` | [Float64](/docs/en/sql-reference/data-types/float.md) | `FLOAT64` |
| `TEXT, DATA` | [String](/docs/en/sql-reference/data-types/string.md), [FixedString](/docs/en/sql-reference/data-types/fixedstring.md) | `TEXT, DATA` |
| `union(T, Void), union(Void, T)` | [Nullable(T)](/docs/en/sql-reference/data-types/date.md) | `union(T, Void), union(Void, T)` |
| `ENUM` | [Enum(8\|16)](/docs/en/sql-reference/data-types/enum.md) | `ENUM` |
| `LIST` | [Array](/docs/en/sql-reference/data-types/array.md) | `LIST` |
| `STRUCT` | [Tuple](/docs/en/sql-reference/data-types/tuple.md) | `STRUCT` |
| CapnProto data type (`INSERT`) | ClickHouse data type | CapnProto data type (`SELECT`) |
|----------------------------------|------------------------------------------------------------------------------------------------------------------------|------------------------------|
| `UINT8`, `BOOL` | [UInt8](/docs/en/sql-reference/data-types/int-uint.md) | `UINT8` |
| `INT8` | [Int8](/docs/en/sql-reference/data-types/int-uint.md) | `INT8` |
| `UINT16` | [UInt16](/docs/en/sql-reference/data-types/int-uint.md), [Date](/docs/en/sql-reference/data-types/date.md) | `UINT16` |
| `INT16` | [Int16](/docs/en/sql-reference/data-types/int-uint.md) | `INT16` |
| `UINT32` | [UInt32](/docs/en/sql-reference/data-types/int-uint.md), [DateTime](/docs/en/sql-reference/data-types/datetime.md) | `UINT32` |
| `INT32` | [Int32](/docs/en/sql-reference/data-types/int-uint.md) | `INT32` |
| `UINT64` | [UInt64](/docs/en/sql-reference/data-types/int-uint.md) | `UINT64` |
| `INT64` | [Int64](/docs/en/sql-reference/data-types/int-uint.md), [DateTime64](/docs/en/sql-reference/data-types/datetime.md) | `INT64` |
| `FLOAT32` | [Float32](/docs/en/sql-reference/data-types/float.md) | `FLOAT32` |
| `FLOAT64` | [Float64](/docs/en/sql-reference/data-types/float.md) | `FLOAT64` |
| `TEXT, DATA` | [String](/docs/en/sql-reference/data-types/string.md), [FixedString](/docs/en/sql-reference/data-types/fixedstring.md) | `TEXT, DATA` |
| `union(T, Void), union(Void, T)` | [Nullable(T)](/docs/en/sql-reference/data-types/date.md) | `union(T, Void), union(Void, T)` |
| `ENUM` | [Enum(8\ |16)](/docs/en/sql-reference/data-types/enum.md) | `ENUM` |
| `LIST` | [Array](/docs/en/sql-reference/data-types/array.md) | `LIST` |
| `STRUCT` | [Tuple](/docs/en/sql-reference/data-types/tuple.md) | `STRUCT` |
| `UINT32` | [IPv4](/docs/en/sql-reference/data-types/domains/ipv4.md) | `UINT32` |
| `DATA` | [IPv6](/docs/en/sql-reference/data-types/domains/ipv6.md) | `DATA` |
For working with `Enum` in CapnProto format use the [format_capn_proto_enum_comparising_mode](/docs/en/operations/settings/settings-formats.md/#format_capn_proto_enum_comparising_mode) setting.
@ -1804,21 +1808,23 @@ ClickHouse Avro format supports reading and writing [Avro data files](https://av
The table below shows supported data types and how they match ClickHouse [data types](/docs/en/sql-reference/data-types/index.md) in `INSERT` and `SELECT` queries.
| Avro data type `INSERT` | ClickHouse data type | Avro data type `SELECT` |
|---------------------------------------------|----------------------------------------------------------------------------------------------------|------------------------------|
| `boolean`, `int`, `long`, `float`, `double` | [Int(8\|16\|32)](/docs/en/sql-reference/data-types/int-uint.md), [UInt(8\|16\|32)](/docs/en/sql-reference/data-types/int-uint.md) | `int` |
| `boolean`, `int`, `long`, `float`, `double` | [Int64](/docs/en/sql-reference/data-types/int-uint.md), [UInt64](/docs/en/sql-reference/data-types/int-uint.md) | `long` |
| `boolean`, `int`, `long`, `float`, `double` | [Float32](/docs/en/sql-reference/data-types/float.md) | `float` |
| `boolean`, `int`, `long`, `float`, `double` | [Float64](/docs/en/sql-reference/data-types/float.md) | `double` |
| `bytes`, `string`, `fixed`, `enum` | [String](/docs/en/sql-reference/data-types/string.md) | `bytes` or `string` \* |
| `bytes`, `string`, `fixed` | [FixedString(N)](/docs/en/sql-reference/data-types/fixedstring.md) | `fixed(N)` |
| `enum` | [Enum(8\|16)](/docs/en/sql-reference/data-types/enum.md) | `enum` |
| `array(T)` | [Array(T)](/docs/en/sql-reference/data-types/array.md) | `array(T)` |
| `union(null, T)`, `union(T, null)` | [Nullable(T)](/docs/en/sql-reference/data-types/date.md) | `union(null, T)` |
| `null` | [Nullable(Nothing)](/docs/en/sql-reference/data-types/special-data-types/nothing.md) | `null` |
| `int (date)` \** | [Date](/docs/en/sql-reference/data-types/date.md) | `int (date)` \** |
| `long (timestamp-millis)` \** | [DateTime64(3)](/docs/en/sql-reference/data-types/datetime.md) | `long (timestamp-millis)` \* |
| `long (timestamp-micros)` \** | [DateTime64(6)](/docs/en/sql-reference/data-types/datetime.md) | `long (timestamp-micros)` \* |
| Avro data type `INSERT` | ClickHouse data type | Avro data type `SELECT` |
|---------------------------------------------|-----------------------------------------------------------------------------------------------------------------|-------------------------------------------------|
| `boolean`, `int`, `long`, `float`, `double` | [Int(8\ | 16\ |32)](/docs/en/sql-reference/data-types/int-uint.md), [UInt(8\|16\|32)](/docs/en/sql-reference/data-types/int-uint.md) | `int` |
| `boolean`, `int`, `long`, `float`, `double` | [Int64](/docs/en/sql-reference/data-types/int-uint.md), [UInt64](/docs/en/sql-reference/data-types/int-uint.md) | `long` |
| `boolean`, `int`, `long`, `float`, `double` | [Float32](/docs/en/sql-reference/data-types/float.md) | `float` |
| `boolean`, `int`, `long`, `float`, `double` | [Float64](/docs/en/sql-reference/data-types/float.md) | `double` |
| `bytes`, `string`, `fixed`, `enum` | [String](/docs/en/sql-reference/data-types/string.md) | `bytes` or `string` \* |
| `bytes`, `string`, `fixed` | [FixedString(N)](/docs/en/sql-reference/data-types/fixedstring.md) | `fixed(N)` |
| `enum` | [Enum(8\ | 16)](/docs/en/sql-reference/data-types/enum.md) | `enum` |
| `array(T)` | [Array(T)](/docs/en/sql-reference/data-types/array.md) | `array(T)` |
| `union(null, T)`, `union(T, null)` | [Nullable(T)](/docs/en/sql-reference/data-types/date.md) | `union(null, T)` |
| `null` | [Nullable(Nothing)](/docs/en/sql-reference/data-types/special-data-types/nothing.md) | `null` |
| `int (date)` \** | [Date](/docs/en/sql-reference/data-types/date.md) | `int (date)` \** |
| `long (timestamp-millis)` \** | [DateTime64(3)](/docs/en/sql-reference/data-types/datetime.md) | `long (timestamp-millis)` \* |
| `long (timestamp-micros)` \** | [DateTime64(6)](/docs/en/sql-reference/data-types/datetime.md) | `long (timestamp-micros)` \* |
| `int` | [IPv4](/docs/en/sql-reference/data-types/domains/ipv4.md) | `int` |
| `fixed(16)` | [IPv6](/docs/en/sql-reference/data-types/domains/ipv6.md) | `fixed(16)` |
\* `bytes` is default, controlled by [output_format_avro_string_column_pattern](/docs/en/operations/settings/settings-formats.md/#output_format_avro_string_column_pattern)
\** [Avro logical types](https://avro.apache.org/docs/current/spec.html#Logical+Types)
@ -1918,28 +1924,30 @@ Setting `format_avro_schema_registry_url` needs to be configured in `users.xml`
The table below shows supported data types and how they match ClickHouse [data types](/docs/en/sql-reference/data-types/index.md) in `INSERT` and `SELECT` queries.
| Parquet data type (`INSERT`) | ClickHouse data type | Parquet data type (`SELECT`) |
|-----------------------------------------------|-----------------------------------------------------------------|------------------------------|
| `BOOL` | [Bool](/docs/en/sql-reference/data-types/boolean.md) | `BOOL` |
| `UINT8`, `BOOL` | [UInt8](/docs/en/sql-reference/data-types/int-uint.md) | `UINT8` |
| `INT8` | [Int8](/docs/en/sql-reference/data-types/int-uint.md) | `INT8` |
| `UINT16` | [UInt16](/docs/en/sql-reference/data-types/int-uint.md) | `UINT16` |
| `INT16` | [Int16](/docs/en/sql-reference/data-types/int-uint.md) | `INT16` |
| `UINT32` | [UInt32](/docs/en/sql-reference/data-types/int-uint.md) | `UINT32` |
| `INT32` | [Int32](/docs/en/sql-reference/data-types/int-uint.md) | `INT32` |
| `UINT64` | [UInt64](/docs/en/sql-reference/data-types/int-uint.md) | `UINT64` |
| `INT64` | [Int64](/docs/en/sql-reference/data-types/int-uint.md) | `INT64` |
| `FLOAT` | [Float32](/docs/en/sql-reference/data-types/float.md) | `FLOAT` |
| `DOUBLE` | [Float64](/docs/en/sql-reference/data-types/float.md) | `DOUBLE` |
| `DATE` | [Date32](/docs/en/sql-reference/data-types/date.md) | `DATE` |
| `TIME (ms)` | [DateTime](/docs/en/sql-reference/data-types/datetime.md) | `UINT32` |
| `TIMESTAMP`, `TIME (us, ns)` | [DateTime64](/docs/en/sql-reference/data-types/datetime64.md) | `TIMESTAMP` |
| `STRING`, `BINARY` | [String](/docs/en/sql-reference/data-types/string.md) | `BINARY` |
| `STRING`, `BINARY`, `FIXED_LENGTH_BYTE_ARRAY` | [FixedString](/docs/en/sql-reference/data-types/fixedstring.md) | `FIXED_LENGTH_BYTE_ARRAY` |
| `DECIMAL` | [Decimal](/docs/en/sql-reference/data-types/decimal.md) | `DECIMAL` |
| `LIST` | [Array](/docs/en/sql-reference/data-types/array.md) | `LIST` |
| `STRUCT` | [Tuple](/docs/en/sql-reference/data-types/tuple.md) | `STRUCT` |
| `MAP` | [Map](/docs/en/sql-reference/data-types/map.md) | `MAP` |
| Parquet data type (`INSERT`) | ClickHouse data type | Parquet data type (`SELECT`) |
|----------------------------------------------------|-----------------------------------------------------------------|------------------------------|
| `BOOL` | [Bool](/docs/en/sql-reference/data-types/boolean.md) | `BOOL` |
| `UINT8`, `BOOL` | [UInt8](/docs/en/sql-reference/data-types/int-uint.md) | `UINT8` |
| `INT8` | [Int8](/docs/en/sql-reference/data-types/int-uint.md) | `INT8` |
| `UINT16` | [UInt16](/docs/en/sql-reference/data-types/int-uint.md) | `UINT16` |
| `INT16` | [Int16](/docs/en/sql-reference/data-types/int-uint.md) | `INT16` |
| `UINT32` | [UInt32](/docs/en/sql-reference/data-types/int-uint.md) | `UINT32` |
| `INT32` | [Int32](/docs/en/sql-reference/data-types/int-uint.md) | `INT32` |
| `UINT64` | [UInt64](/docs/en/sql-reference/data-types/int-uint.md) | `UINT64` |
| `INT64` | [Int64](/docs/en/sql-reference/data-types/int-uint.md) | `INT64` |
| `FLOAT` | [Float32](/docs/en/sql-reference/data-types/float.md) | `FLOAT` |
| `DOUBLE` | [Float64](/docs/en/sql-reference/data-types/float.md) | `DOUBLE` |
| `DATE` | [Date32](/docs/en/sql-reference/data-types/date.md) | `DATE` |
| `TIME (ms)` | [DateTime](/docs/en/sql-reference/data-types/datetime.md) | `UINT32` |
| `TIMESTAMP`, `TIME (us, ns)` | [DateTime64](/docs/en/sql-reference/data-types/datetime64.md) | `TIMESTAMP` |
| `STRING`, `BINARY` | [String](/docs/en/sql-reference/data-types/string.md) | `BINARY` |
| `STRING`, `BINARY`, `FIXED_LENGTH_BYTE_ARRAY` | [FixedString](/docs/en/sql-reference/data-types/fixedstring.md) | `FIXED_LENGTH_BYTE_ARRAY` |
| `DECIMAL` | [Decimal](/docs/en/sql-reference/data-types/decimal.md) | `DECIMAL` |
| `LIST` | [Array](/docs/en/sql-reference/data-types/array.md) | `LIST` |
| `STRUCT` | [Tuple](/docs/en/sql-reference/data-types/tuple.md) | `STRUCT` |
| `MAP` | [Map](/docs/en/sql-reference/data-types/map.md) | `MAP` |
| `UINT32` | [IPv4](/docs/en/sql-reference/data-types/domains/ipv4.md) | `UINT32` |
| `FIXED_LENGTH_BYTE_ARRAY` | [IPv6](/docs/en/sql-reference/data-types/domains/ipv6.md) | `FIXED_LENGTH_BYTE_ARRAY` |
Arrays can be nested and can have a value of the `Nullable` type as an argument. `Tuple` and `Map` types also can be nested.
@ -1973,6 +1981,7 @@ To exchange data with Hadoop, you can use [HDFS table engine](/docs/en/engines/t
- [input_format_parquet_skip_columns_with_unsupported_types_in_schema_inference](/docs/en/operations/settings/settings-formats.md/#input_format_parquet_skip_columns_with_unsupported_types_in_schema_inference) - allow skipping columns with unsupported types while schema inference for Parquet format. Default value - `false`.
- [output_format_parquet_fixed_string_as_fixed_byte_array](/docs/en/operations/settings/settings-formats.md/#output_format_parquet_fixed_string_as_fixed_byte_array) - use Parquet FIXED_LENGTH_BYTE_ARRAY type instead of Binary/String for FixedString columns. Default value - `true`.
- [output_format_parquet_version](/docs/en/operations/settings/settings-formats.md/#output_format_parquet_version) - The version of Parquet format used in output format. Default value - `2.latest`.
- [output_format_parquet_compression_method](/docs/en/operations/settings/settings-formats.md/#output_format_parquet_compression_method) - compression method used in output Parquet format. Default value - `snappy`.
## Arrow {#data-format-arrow}
@ -2007,6 +2016,8 @@ The table below shows supported data types and how they match ClickHouse [data t
| `LIST` | [Array](/docs/en/sql-reference/data-types/array.md) | `LIST` |
| `STRUCT` | [Tuple](/docs/en/sql-reference/data-types/tuple.md) | `STRUCT` |
| `MAP` | [Map](/docs/en/sql-reference/data-types/map.md) | `MAP` |
| `UINT32` | [IPv4](/docs/en/sql-reference/data-types/domains/ipv4.md) | `UINT32` |
| `FIXED_SIZE_BINARY`, `BINARY` | [IPv6](/docs/en/sql-reference/data-types/domains/ipv6.md) | `FIXED_SIZE_BINARY` |
Arrays can be nested and can have a value of the `Nullable` type as an argument. `Tuple` and `Map` types also can be nested.
@ -2041,6 +2052,7 @@ $ clickhouse-client --query="SELECT * FROM {some_table} FORMAT Arrow" > {filenam
- [input_format_arrow_allow_missing_columns](/docs/en/operations/settings/settings-formats.md/#input_format_arrow_allow_missing_columns) - allow missing columns while reading Arrow data. Default value - `false`.
- [input_format_arrow_skip_columns_with_unsupported_types_in_schema_inference](/docs/en/operations/settings/settings-formats.md/#input_format_arrow_skip_columns_with_unsupported_types_in_schema_inference) - allow skipping columns with unsupported types while schema inference for Arrow format. Default value - `false`.
- [output_format_arrow_fixed_string_as_fixed_byte_array](/docs/en/operations/settings/settings-formats.md/#output_format_arrow_fixed_string_as_fixed_byte_array) - use Arrow FIXED_SIZE_BINARY type instead of Binary/String for FixedString columns. Default value - `true`.
- [output_format_arrow_compression_method](/docs/en/operations/settings/settings-formats.md/#output_format_arrow_compression_method) - compression method used in output Arrow format. Default value - `none`.
## ArrowStream {#data-format-arrow-stream}
@ -2054,8 +2066,8 @@ $ clickhouse-client --query="SELECT * FROM {some_table} FORMAT Arrow" > {filenam
The table below shows supported data types and how they match ClickHouse [data types](/docs/en/sql-reference/data-types/index.md) in `INSERT` and `SELECT` queries.
| ORC data type (`INSERT`) | ClickHouse data type | ORC data type (`SELECT`) |
|---------------------------------------|---------------------------------------------------------|--------------------------|
| ORC data type (`INSERT`) | ClickHouse data type | ORC data type (`SELECT`) |
|---------------------------------------|---------------------------------------------------------------|--------------------------|
| `Boolean` | [UInt8](/docs/en/sql-reference/data-types/int-uint.md) | `Boolean` |
| `Tinyint` | [Int8](/docs/en/sql-reference/data-types/int-uint.md) | `Tinyint` |
| `Smallint` | [Int16](/docs/en/sql-reference/data-types/int-uint.md) | `Smallint` |
@ -2070,6 +2082,7 @@ The table below shows supported data types and how they match ClickHouse [data t
| `List` | [Array](/docs/en/sql-reference/data-types/array.md) | `List` |
| `Struct` | [Tuple](/docs/en/sql-reference/data-types/tuple.md) | `Struct` |
| `Map` | [Map](/docs/en/sql-reference/data-types/map.md) | `Map` |
| `-` | [IPv4](/docs/en/sql-reference/data-types/int-uint.md) | `Int` |
Other types are not supported.
@ -2096,6 +2109,7 @@ $ clickhouse-client --query="SELECT * FROM {some_table} FORMAT ORC" > {filename.
### Arrow format settings {#parquet-format-settings}
- [output_format_arrow_string_as_string](/docs/en/operations/settings/settings-formats.md/#output_format_arrow_string_as_string) - use Arrow String type instead of Binary for String columns. Default value - `false`.
- [output_format_orc_compression_method](/docs/en/operations/settings/settings-formats.md/#output_format_orc_compression_method) - compression method used in output ORC format. Default value - `none`.
- [input_format_arrow_import_nested](/docs/en/operations/settings/settings-formats.md/#input_format_arrow_import_nested) - allow inserting array of structs into Nested table in Arrow input format. Default value - `false`.
- [input_format_arrow_case_insensitive_column_matching](/docs/en/operations/settings/settings-formats.md/#input_format_arrow_case_insensitive_column_matching) - ignore case when matching Arrow columns with ClickHouse columns. Default value - `false`.
- [input_format_arrow_allow_missing_columns](/docs/en/operations/settings/settings-formats.md/#input_format_arrow_allow_missing_columns) - allow missing columns while reading Arrow data. Default value - `false`.
@ -2264,8 +2278,8 @@ ClickHouse supports reading and writing [MessagePack](https://msgpack.org/) data
### Data Types Matching {#data-types-matching-msgpack}
| MessagePack data type (`INSERT`) | ClickHouse data type | MessagePack data type (`SELECT`) |
|--------------------------------------------------------------------|-----------------------------------------------------------|------------------------------------|
| MessagePack data type (`INSERT`) | ClickHouse data type | MessagePack data type (`SELECT`) |
|--------------------------------------------------------------------|-----------------------------------------------------------------|------------------------------------|
| `uint N`, `positive fixint` | [UIntN](/docs/en/sql-reference/data-types/int-uint.md) | `uint N` |
| `int N`, `negative fixint` | [IntN](/docs/en/sql-reference/data-types/int-uint.md) | `int N` |
| `bool` | [UInt8](/docs/en/sql-reference/data-types/int-uint.md) | `uint 8` |
@ -2278,6 +2292,8 @@ ClickHouse supports reading and writing [MessagePack](https://msgpack.org/) data
| `uint 64` | [DateTime64](/docs/en/sql-reference/data-types/datetime.md) | `uint 64` |
| `fixarray`, `array 16`, `array 32` | [Array](/docs/en/sql-reference/data-types/array.md) | `fixarray`, `array 16`, `array 32` |
| `fixmap`, `map 16`, `map 32` | [Map](/docs/en/sql-reference/data-types/map.md) | `fixmap`, `map 16`, `map 32` |
| `uint 32` | [IPv4](/docs/en/sql-reference/data-types/domains/ipv4.md) | `uint 32` |
| `bin 8` | [String](/docs/en/sql-reference/data-types/string.md) | `bin 8` |
Example:

80
docs/en/interfaces/schema-inference.md
View File

@ -117,7 +117,7 @@ clickhouse-local --file='hobbies.jsonl' --table='hobbies' --query='SELECT * FROM
4 47 Brayan ['movies','skydiving']
```
# Using structure from insertion table {#using-structure-from-insertion-table}
## Using structure from insertion table {#using-structure-from-insertion-table}
When table functions `file/s3/url/hdfs` are used to insert data into a table,
there is an option to use the structure from the insertion table instead of extracting it from the data.
@ -222,7 +222,7 @@ INSERT INTO hobbies4 SELECT id, empty(hobbies) ? NULL : hobbies[1] FROM file(hob
In this case, there are some operations performed on the column `hobbies` in the `SELECT` query to insert it into the table, so ClickHouse cannot use the structure from the insertion table, and schema inference will be used.
# Schema inference cache {#schema-inference-cache}
## Schema inference cache {#schema-inference-cache}
For most input formats schema inference reads some data to determine its structure and this process can take some time.
To prevent inferring the same schema every time ClickHouse read the data from the same file, the inferred schema is cached and when accessing the same file again, ClickHouse will use the schema from the cache.
@ -326,14 +326,14 @@ SELECT count() FROM system.schema_inference_cache WHERE storage='S3'
└─────────┘
```
# Text formats {#text-formats}
## Text formats {#text-formats}
For text formats, ClickHouse reads the data row by row, extracts column values according to the format,
and then uses some recursive parsers and heuristics to determine the type for each value. The maximum number of rows read from the data in schema inference
is controlled by the setting `input_format_max_rows_to_read_for_schema_inference` with default value 25000.
By default, all inferred types are [Nullable](../sql-reference/data-types/nullable.md), but you can change this by setting `schema_inference_make_columns_nullable` (see examples in the [settings](#settings-for-text-formats) section).
## JSON formats {#json-formats}
### JSON formats {#json-formats}
In JSON formats ClickHouse parses values according to the JSON specification and then tries to find the most appropriate data type for them.
@ -464,9 +464,9 @@ most likely this column contains only Nulls or empty Arrays/Maps.
...
```
### JSON settings {#json-settings}
#### JSON settings {#json-settings}
#### input_format_json_read_objects_as_strings
##### input_format_json_read_objects_as_strings
Enabling this setting allows reading nested JSON objects as strings.
This setting can be used to read nested JSON objects without using JSON object type.
@ -486,7 +486,7 @@ DESC format(JSONEachRow, $$
└──────┴──────────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
#### input_format_json_try_infer_numbers_from_strings
##### input_format_json_try_infer_numbers_from_strings
Enabling this setting allows inferring numbers from string values.
@ -507,7 +507,7 @@ DESC format(JSONEachRow, $$
└───────┴─────────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
#### input_format_json_read_numbers_as_strings
##### input_format_json_read_numbers_as_strings
Enabling this setting allows reading numeric values as strings.
@ -528,7 +528,7 @@ DESC format(JSONEachRow, $$
└───────┴──────────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
#### input_format_json_read_bools_as_numbers
##### input_format_json_read_bools_as_numbers
Enabling this setting allows reading Bool values as numbers.
@ -549,7 +549,7 @@ DESC format(JSONEachRow, $$
└───────┴─────────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
## CSV {#csv}
### CSV {#csv}
In CSV format ClickHouse extracts column values from the row according to delimiters. ClickHouse expects all types except numbers and strings to be enclosed in double quotes. If the value is in double quotes, ClickHouse tries to parse
the data inside quotes using the recursive parser and then tries to find the most appropriate data type for it. If the value is not in double quotes, ClickHouse tries to parse it as a number,
@ -726,7 +726,7 @@ $$)
└──────────────┴───────────────┘
```
## TSV/TSKV {#tsv-tskv}
### TSV/TSKV {#tsv-tskv}
In TSV/TSKV formats ClickHouse extracts column value from the row according to tabular delimiters and then parses extracted value using
the recursive parser to determine the most appropriate type. If the type cannot be determined, ClickHouse treats this value as String.
@ -1019,7 +1019,7 @@ DESC format(TSV, '[1,2,3] 42.42 Hello World!')
└──────┴──────────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
## CustomSeparated {#custom-separated}
### CustomSeparated {#custom-separated}
In CustomSeparated format ClickHouse first extracts all column values from the row according to specified delimiters and then tries to infer
the data type for each value according to escaping rule.
@ -1080,7 +1080,7 @@ $$)
└────────┴───────────────┴────────────┘
```
## Template {#template}
### Template {#template}
In Template format ClickHouse first extracts all column values from the row according to the specified template and then tries to infer the
data type for each value according to its escaping rule.
@ -1120,7 +1120,7 @@ $$)
└──────────┴────────────────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
## Regexp {#regexp}
### Regexp {#regexp}
Similar to Template, in Regexp format ClickHouse first extracts all column values from the row according to specified regular expression and then tries to infer
data type for each value according to the specified escaping rule.
@ -1142,9 +1142,9 @@ Line: value_1=2, value_2="Some string 2", value_3="[4, 5, NULL]"$$)
└──────┴────────────────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
## Settings for text formats {settings-for-text-formats}
### Settings for text formats {#settings-for-text-formats}
### input_format_max_rows_to_read_for_schema_inference
#### input_format_max_rows_to_read_for_schema_inference
This setting controls the maximum number of rows to be read while schema inference.
The more rows are read, the more time is spent on schema inference, but the greater the chance to
@ -1152,7 +1152,7 @@ correctly determine the types (especially when the data contains a lot of nulls)
Default value: `25000`.
### column_names_for_schema_inference
#### column_names_for_schema_inference
The list of column names to use in schema inference for formats without explicit column names. Specified names will be used instead of default `c1,c2,c3,...`. The format: `column1,column2,column3,...`.
@ -1169,7 +1169,7 @@ DESC format(TSV, 'Hello, World! 42 [1, 2, 3]') settings column_names_for_schema_
└──────┴────────────────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
### schema_inference_hints
#### schema_inference_hints
The list of column names and types to use in schema inference instead of automatically determined types. The format: 'column_name1 column_type1, column_name2 column_type2, ...'.
This setting can be used to specify the types of columns that could not be determined automatically or for optimizing the schema.
@ -1189,7 +1189,7 @@ DESC format(JSONEachRow, '{"id" : 1, "age" : 25, "name" : "Josh", "status" : nul
└─────────┴─────────────────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
### schema_inference_make_columns_nullable
#### schema_inference_make_columns_nullable
Controls making inferred types `Nullable` in schema inference for formats without information about nullability.
If the setting is enabled, all inferred type will be `Nullable`, if disabled, the inferred type will be `Nullable` only if the column contains `NULL` in a sample that is parsed during schema inference.
@ -1232,7 +1232,7 @@ DESC format(JSONEachRow, $$
└─────────┴──────────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
### input_format_try_infer_integers
#### input_format_try_infer_integers
If enabled, ClickHouse will try to infer integers instead of floats in schema inference for text formats.
If all numbers in the column from sample data are integers, the result type will be `Int64`, if at least one number is float, the result type will be `Float64`.
@ -1289,7 +1289,7 @@ DESC format(JSONEachRow, $$
└────────┴───────────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
### input_format_try_infer_datetimes
#### input_format_try_infer_datetimes
If enabled, ClickHouse will try to infer type `DateTime64` from string fields in schema inference for text formats.
If all fields from a column in sample data were successfully parsed as datetimes, the result type will be `DateTime64(9)`,
@ -1337,7 +1337,7 @@ DESC format(JSONEachRow, $$
Note: Parsing datetimes during schema inference respect setting [date_time_input_format](/docs/en/operations/settings/settings-formats.md#date_time_input_format)
### input_format_try_infer_dates
#### input_format_try_infer_dates
If enabled, ClickHouse will try to infer type `Date` from string fields in schema inference for text formats.
If all fields from a column in sample data were successfully parsed as dates, the result type will be `Date`,
@ -1383,14 +1383,14 @@ DESC format(JSONEachRow, $$
└──────┴──────────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
# Self describing formats {#self-describing-formats}
## Self describing formats {#self-describing-formats}
Self-describing formats contain information about the structure of the data in the data itself,
it can be some header with a description, a binary type tree, or some kind of table.
To automatically infer a schema from files in such formats, ClickHouse reads a part of the data containing
information about the types and converts it into a schema of the ClickHouse table.
## Formats with -WithNamesAndTypes suffix {#formats-with-names-and-types}
### Formats with -WithNamesAndTypes suffix {#formats-with-names-and-types}
ClickHouse supports some text formats with the suffix -WithNamesAndTypes. This suffix means that the data contains two additional rows with column names and types before the actual data.
While schema inference for such formats, ClickHouse reads the first two rows and extracts column names and types.
@ -1412,7 +1412,7 @@ $$)
└──────┴──────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
## JSON formats with metadata {#json-with-metadata}
### JSON formats with metadata {#json-with-metadata}
Some JSON input formats ([JSON](formats.md#json), [JSONCompact](formats.md#json-compact), [JSONColumnsWithMetadata](formats.md#jsoncolumnswithmetadata)) contain metadata with column names and types.
In schema inference for such formats, ClickHouse reads this metadata.
@ -1465,7 +1465,7 @@ $$)
└──────┴──────────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
## Avro {#avro}
### Avro {#avro}
In Avro format ClickHouse reads its schema from the data and converts it to ClickHouse schema using the following type matches:
@ -1485,7 +1485,7 @@ In Avro format ClickHouse reads its schema from the data and converts it to Clic
Other Avro types are not supported.
## Parquet {#parquet}
### Parquet {#parquet}
In Parquet format ClickHouse reads its schema from the data and converts it to ClickHouse schema using the following type matches:
@ -1513,7 +1513,7 @@ In Parquet format ClickHouse reads its schema from the data and converts it to C
Other Parquet types are not supported. By default, all inferred types are inside `Nullable`, but it can be changed using the setting `schema_inference_make_columns_nullable`.
## Arrow {#arrow}
### Arrow {#arrow}
In Arrow format ClickHouse reads its schema from the data and converts it to ClickHouse schema using the following type matches:
@ -1541,7 +1541,7 @@ In Arrow format ClickHouse reads its schema from the data and converts it to Cli
Other Arrow types are not supported. By default, all inferred types are inside `Nullable`, but it can be changed using the setting `schema_inference_make_columns_nullable`.
## ORC {#orc}
### ORC {#orc}
In ORC format ClickHouse reads its schema from the data and converts it to ClickHouse schema using the following type matches:
@ -1564,17 +1564,17 @@ In ORC format ClickHouse reads its schema from the data and converts it to Click
Other ORC types are not supported. By default, all inferred types are inside `Nullable`, but it can be changed using the setting `schema_inference_make_columns_nullable`.
## Native {#native}
### Native {#native}
Native format is used inside ClickHouse and contains the schema in the data.
In schema inference, ClickHouse reads the schema from the data without any transformations.
# Formats with external schema {#formats-with-external-schema}
## Formats with external schema {#formats-with-external-schema}
Such formats require a schema describing the data in a separate file in a specific schema language.
To automatically infer a schema from files in such formats, ClickHouse reads external schema from a separate file and transforms it to a ClickHouse table schema.
# Protobuf {#protobuf}
### Protobuf {#protobuf}
In schema inference for Protobuf format ClickHouse uses the following type matches:
@ -1592,7 +1592,7 @@ In schema inference for Protobuf format ClickHouse uses the following type match
| `repeated T` | [Array(T)](../sql-reference/data-types/array.md) |
| `message`, `group` | [Tuple](../sql-reference/data-types/tuple.md) |
# CapnProto {#capnproto}
### CapnProto {#capnproto}
In schema inference for CapnProto format ClickHouse uses the following type matches:
@ -1615,13 +1615,13 @@ In schema inference for CapnProto format ClickHouse uses the following type matc
| `struct` | [Tuple](../sql-reference/data-types/tuple.md) |
| `union(T, Void)`, `union(Void, T)` | [Nullable(T)](../sql-reference/data-types/nullable.md) |
# Strong-typed binary formats {#strong-typed-binary-formats}
## Strong-typed binary formats {#strong-typed-binary-formats}
In such formats, each serialized value contains information about its type (and possibly about its name), but there is no information about the whole table.
In schema inference for such formats, ClickHouse reads data row by row (up to `input_format_max_rows_to_read_for_schema_inference` rows) and extracts
the type (and possibly name) for each value from the data and then converts these types to ClickHouse types.
## MsgPack {msgpack}
### MsgPack {#msgpack}
In MsgPack format there is no delimiter between rows, to use schema inference for this format you should specify the number of columns in the table
using the setting `input_format_msgpack_number_of_columns`. ClickHouse uses the following type matches:
@ -1641,7 +1641,7 @@ using the setting `input_format_msgpack_number_of_columns`. ClickHouse uses the
By default, all inferred types are inside `Nullable`, but it can be changed using the setting `schema_inference_make_columns_nullable`.
## BSONEachRow {#bsoneachrow}
### BSONEachRow {#bsoneachrow}
In BSONEachRow each row of data is presented as a BSON document. In schema inference ClickHouse reads BSON documents one by one and extracts
values, names, and types from the data and then transforms these types to ClickHouse types using the following type matches:
@ -1661,11 +1661,11 @@ values, names, and types from the data and then transforms these types to ClickH
By default, all inferred types are inside `Nullable`, but it can be changed using the setting `schema_inference_make_columns_nullable`.
# Formats with constant schema {#formats-with-constant-schema}
## Formats with constant schema {#formats-with-constant-schema}
Data in such formats always have the same schema.
## LineAsString {#line-as-string}
### LineAsString {#line-as-string}
In this format, ClickHouse reads the whole line from the data into a single column with `String` data type. The inferred type for this format is always `String` and the column name is `line`.
@ -1680,7 +1680,7 @@ DESC format(LineAsString, 'Hello\nworld!')
└──────┴────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
## JSONAsString {#json-as-string}
### JSONAsString {#json-as-string}
In this format, ClickHouse reads the whole JSON object from the data into a single column with `String` data type. The inferred type for this format is always `String` and the column name is `json`.
@ -1695,7 +1695,7 @@ DESC format(JSONAsString, '{"x" : 42, "y" : "Hello, World!"}')
└──────┴────────┴──────────────┴────────────────────┴─────────┴──────────────────┴────────────────┘
```
## JSONAsObject {#json-as-object}
### JSONAsObject {#json-as-object}
In this format, ClickHouse reads the whole JSON object from the data into a single column with `Object('json')` data type. Inferred type for this format is always `String` and the column name is `json`.

76
docs/en/operations/server-configuration-parameters/settings.md
View File

@ -967,6 +967,7 @@ The maximum number of jobs that can be scheduled on the Global Thread pool. Incr
Possible values:
- Positive integer.
- 0 — No limit.
Default value: `10000`.
@ -976,6 +977,69 @@ Default value: `10000`.
<thread_pool_queue_size>12000</thread_pool_queue_size>
```
## max_io_thread_pool_size {#max-io-thread-pool-size}
ClickHouse uses threads from the IO Thread pool to do some IO operations (e.g. to interact with S3). `max_io_thread_pool_size` limits the maximum number of threads in the pool.
Possible values:
- Positive integer.
Default value: `100`.
## max_io_thread_pool_free_size {#max-io-thread-pool-free-size}
If the number of **idle** threads in the IO Thread pool exceeds `max_io_thread_pool_free_size`, ClickHouse will release resources occupied by idling threads and decrease the pool size. Threads can be created again if necessary.
Possible values:
- Positive integer.
Default value: `0`.
## io_thread_pool_queue_size {#io-thread-pool-queue-size}
The maximum number of jobs that can be scheduled on the IO Thread pool.
Possible values:
- Positive integer.
- 0 — No limit.
Default value: `10000`.
## max_backups_io_thread_pool_size {#max-backups-io-thread-pool-size}
ClickHouse uses threads from the Backups IO Thread pool to do S3 backup IO operations. `max_backups_io_thread_pool_size` limits the maximum number of threads in the pool.
Possible values:
- Positive integer.
Default value: `1000`.
## max_backups_io_thread_pool_free_size {#max-backups-io-thread-pool-free-size}
If the number of **idle** threads in the Backups IO Thread pool exceeds `max_backup_io_thread_pool_free_size`, ClickHouse will release resources occupied by idling threads and decrease the pool size. Threads can be created again if necessary.
Possible values:
- Positive integer.
- Zero.
Default value: `0`.
## backups_io_thread_pool_queue_size {#backups-io-thread-pool-queue-size}
The maximum number of jobs that can be scheduled on the Backups IO Thread pool. It is recommended to keep this queue unlimited due to the current S3 backup logic.
Possible values:
- Positive integer.
- 0 — No limit.
Default value: `0`.
## background_pool_size {#background_pool_size}
Sets the number of threads performing background merges and mutations for tables with MergeTree engines. This setting is also could be applied at server startup from the `default` profile configuration for backward compatibility at the ClickHouse server start. You can only increase the number of threads at runtime. To lower the number of threads you have to restart the server. By adjusting this setting, you manage CPU and disk load. Smaller pool size utilizes less CPU and disk resources, but background processes advance slower which might eventually impact query performance.
@ -1254,12 +1318,12 @@ Settings:
``` xml
<prometheus>
<endpoint>/metrics</endpoint>
<port>8001</port>
<metrics>true</metrics>
<events>true</events>
<asynchronous_metrics>true</asynchronous_metrics>
</prometheus>
<endpoint>/metrics</endpoint>
<port>9363</port>
<metrics>true</metrics>
<events>true</events>
<asynchronous_metrics>true</asynchronous_metrics>
</prometheus>
```
## query_log {#server_configuration_parameters-query-log}

30
docs/en/operations/settings/settings-formats.md
View File

@ -1014,6 +1014,12 @@ Use Arrow FIXED_SIZE_BINARY type instead of Binary/String for FixedString column
Enabled by default.
### output_format_arrow_compression_method {#output_format_arrow_compression_method}
Compression method used in output Arrow format. Supported codecs: `lz4_frame`, `zstd`, `none` (uncompressed)
Default value: `none`.
## ORC format settings {#orc-format-settings}
### input_format_orc_import_nested {#input_format_orc_import_nested}
@ -1057,6 +1063,12 @@ Use ORC String type instead of Binary for String columns.
Disabled by default.
### output_format_orc_compression_method {#output_format_orc_compression_method}
Compression method used in output ORC format. Supported codecs: `lz4`, `snappy`, `zlib`, `zstd`, `none` (uncompressed)
Default value: `none`.
## Parquet format settings {#parquet-format-settings}
### input_format_parquet_import_nested {#input_format_parquet_import_nested}
@ -1112,6 +1124,12 @@ The version of Parquet format used in output format. Supported versions: `1.0`,
Default value: `2.latest`.
### output_format_parquet_compression_method {#output_format_parquet_compression_method}
Compression method used in output Parquet format. Supported codecs: `snappy`, `lz4`, `brotli`, `zstd`, `gzip`, `none` (uncompressed)
Default value: `snappy`.
## Hive format settings {#hive-format-settings}
### input_format_hive_text_fields_delimiter {#input_format_hive_text_fields_delimiter}
@ -1474,7 +1492,7 @@ Default value: `65505`.
The name of table that will be used in the output INSERT statement.
Default value: `'table''`.
Default value: `table`.
### output_format_sql_insert_include_column_names {#output_format_sql_insert_include_column_names}
@ -1514,4 +1532,12 @@ Disabled by default.
The maximum allowed size for String in RowBinary format. It prevents allocating large amount of memory in case of corrupted data. 0 means there is no limit.
Default value: `1GiB`
Default value: `1GiB`.
## Native format settings {#native-format-settings}
### input_format_native_allow_types_conversion {#input_format_native_allow_types_conversion}
Allow types conversion in Native input format between columns from input data and requested columns.
Enabled by default.

45
docs/en/operations/settings/settings.md
View File

@ -1248,7 +1248,9 @@ Possible values:
Default value: 1.
:::warning
Disable this setting if you use [max_parallel_replicas](#settings-max_parallel_replicas).
Disable this setting if you use [max_parallel_replicas](#settings-max_parallel_replicas) without [parallel_replicas_custom_key](#settings-parallel_replicas_custom_key).
If [parallel_replicas_custom_key](#settings-parallel_replicas_custom_key) is set, disable this setting only if it's used on a cluster with multiple shards containing multiple replicas.
If it's used on a cluster with a single shard and multiple replicas, disabling this setting will have negative effects.
:::
## totals_mode {#totals-mode}
@ -1273,16 +1275,47 @@ Default value: `1`.
**Additional Info**
This setting is useful for replicated tables with a sampling key. A query may be processed faster if it is executed on several servers in parallel. But the query performance may degrade in the following cases:
This options will produce different results depending on the settings used.
:::warning
This setting will produce incorrect results when joins or subqueries are involved, and all tables don't meet certain requirements. See [Distributed Subqueries and max_parallel_replicas](../../sql-reference/operators/in.md/#max_parallel_replica-subqueries) for more details.
:::
### Parallel processing using `SAMPLE` key
A query may be processed faster if it is executed on several servers in parallel. But the query performance may degrade in the following cases:
- The position of the sampling key in the partitioning key does not allow efficient range scans.
- Adding a sampling key to the table makes filtering by other columns less efficient.
- The sampling key is an expression that is expensive to calculate.
- The cluster latency distribution has a long tail, so that querying more servers increases the query overall latency.
:::warning
This setting will produce incorrect results when joins or subqueries are involved, and all tables don't meet certain requirements. See [Distributed Subqueries and max_parallel_replicas](../../sql-reference/operators/in.md/#max_parallel_replica-subqueries) for more details.
:::
### Parallel processing using [parallel_replicas_custom_key](#settings-parallel_replicas_custom_key)
This setting is useful for any replicated table.
## parallel_replicas_custom_key {#settings-parallel_replicas_custom_key}
An arbitrary integer expression that can be used to split work between replicas for a specific table.
The value can be any integer expression.
A query may be processed faster if it is executed on several servers in parallel but it depends on the used [parallel_replicas_custom_key](#settings-parallel_replicas_custom_key)
and [parallel_replicas_custom_key_filter_type](#settings-parallel_replicas_custom_key_filter_type).
Simple expressions using primary keys are preferred.
If the setting is used on a cluster that consists of a single shard with multiple replicas, those replicas will be converted into virtual shards.
Otherwise, it will behave same as for `SAMPLE` key, it will use multiple replicas of each shard.
## parallel_replicas_custom_key_filter_type {#settings-parallel_replicas_custom_key_filter_type}
How to use `parallel_replicas_custom_key` expression for splitting work between replicas.
Possible values:
- `default` — Use the default implementation using modulo operation on the `parallel_replicas_custom_key`.
- `range` — Split the entire value space of the expression in the ranges. This type of filtering is useful if values of `parallel_replicas_custom_key` are uniformly spread across the entire integer space, e.g. hash values.
Default value: `default`.
## compile_expressions {#compile-expressions}
@ -1515,7 +1548,7 @@ Enables or disables asynchronous inserts. This makes sense only for insertion ov
If enabled, the data is combined into batches before the insertion into tables, so it is possible to do small and frequent insertions into ClickHouse (up to 15000 queries per second) without buffer tables.
The data is inserted either after the [async_insert_max_data_size](#async-insert-max-data-size) is exceeded or after [async_insert_busy_timeout_ms](#async-insert-busy-timeout-ms) milliseconds since the first `INSERT` query. If the [async_insert_stale_timeout_ms](#async-insert-stale-timeout-ms) is set to a non-zero value, the data is inserted after `async_insert_stale_timeout_ms` milliseconds since the last query.
The data is inserted either after the [async_insert_max_data_size](#async-insert-max-data-size) is exceeded or after [async_insert_busy_timeout_ms](#async-insert-busy-timeout-ms) milliseconds since the first `INSERT` query. If the [async_insert_stale_timeout_ms](#async-insert-stale-timeout-ms) is set to a non-zero value, the data is inserted after `async_insert_stale_timeout_ms` milliseconds since the last query. Also the buffer will be flushed to disk if at least [async_insert_max_query_number](#async-insert-max-query-number) async insert queries per block were received. This last setting takes effect only if [async_insert_deduplicate](#async-insert-deduplicate) is enabled.
If [wait_for_async_insert](#wait-for-async-insert) is enabled, every client will wait for the data to be processed and flushed to the table. Otherwise, the query would be processed almost instantly, even if the data is not inserted.

24
docs/en/operations/storing-data.md
View File

@ -135,11 +135,13 @@ Example of configuration for versions later or equal to 22.8:
</cache>
</disks>
<policies>
<volumes>
<main>
<disk>cache</disk>
</main>
</volumes>
<s3-cache>
<volumes>
<main>
<disk>cache</disk>
</main>
</volumes>
</s3-cache>
<policies>
</storage_configuration>
```
@ -159,11 +161,13 @@ Example of configuration for versions earlier than 22.8:
</s3>
</disks>
<policies>
<volumes>
<main>
<disk>s3</disk>
</main>
</volumes>
<s3-cache>
<volumes>
<main>
<disk>s3</disk>
</main>
</volumes>
</s3-cache>
<policies>
</storage_configuration>
```

8
docs/en/operations/system-tables/opentelemetry_span_log.md
View File

@ -15,6 +15,13 @@ Columns:
- `operation_name` ([String](../../sql-reference/data-types/string.md)) — The name of the operation.
- `kind` ([Enum8](../../sql-reference/data-types/enum.md)) — The [SpanKind](https://opentelemetry.io/docs/reference/specification/trace/api/#spankind) of the span.
- `INTERNAL` — Indicates that the span represents an internal operation within an application.
- `SERVER` — Indicates that the span covers server-side handling of a synchronous RPC or other remote request.
- `CLIENT` — Indicates that the span describes a request to some remote service.
- `PRODUCER` — Indicates that the span describes the initiators of an asynchronous request. This parent span will often end before the corresponding child CONSUMER span, possibly even before the child span starts.
- `CONSUMER` - Indicates that the span describes a child of an asynchronous PRODUCER request.
- `start_time_us` ([UInt64](../../sql-reference/data-types/int-uint.md)) — The start time of the `trace span` (in microseconds).
- `finish_time_us` ([UInt64](../../sql-reference/data-types/int-uint.md)) — The finish time of the `trace span` (in microseconds).
@ -42,6 +49,7 @@ trace_id: cdab0847-0d62-61d5-4d38-dd65b19a1914
span_id: 701487461015578150
parent_span_id: 2991972114672045096
operation_name: DB::Block DB::InterpreterSelectQuery::getSampleBlockImpl()
kind: INTERNAL
start_time_us: 1612374594529090
finish_time_us: 1612374594529108
finish_date: 2021-02-03

4
docs/en/operations/utilities/clickhouse-local.md
View File

@ -14,10 +14,6 @@ Accepts data that represent tables and queries them using [ClickHouse SQL dialec
By default `clickhouse-local` does not have access to data on the same host, but it supports loading server configuration using `--config-file` argument.
:::warning
It is not recommended to load production server configuration into `clickhouse-local` because data can be damaged in case of human error.
:::
For temporary data, a unique temporary data directory is created by default.
## Usage {#usage}

61
docs/en/sql-reference/aggregate-functions/reference/exponentialmovingaverage.md
View File

@ -11,15 +11,15 @@ sidebar_title: exponentialMovingAverage
**Syntax**
```sql
exponentialMovingAverage(x)(value, timestamp)
exponentialMovingAverage(x)(value, timeunit)
```
Each `value` corresponds to the determinate `timestamp`. The half-life `x` is the time lag at which the exponential weights decay by one-half. The function returns a weighted average: the older the time point, the less weight the corresponding value is considered to be.
Each `value` corresponds to the determinate `timeunit`. The half-life `x` is the time lag at which the exponential weights decay by one-half. The function returns a weighted average: the older the time point, the less weight the corresponding value is considered to be.
**Arguments**
- `value` — Value. [Integer](../../../sql-reference/data-types/int-uint.md), [Float](../../../sql-reference/data-types/float.md) or [Decimal](../../../sql-reference/data-types/decimal.md).
- `timestamp` — Timestamp. [Integer](../../../sql-reference/data-types/int-uint.md), [Float](../../../sql-reference/data-types/float.md) or [Decimal](../../../sql-reference/data-types/decimal.md).
- `timeunit` — Timeunit. [Integer](../../../sql-reference/data-types/int-uint.md), [Float](../../../sql-reference/data-types/float.md) or [Decimal](../../../sql-reference/data-types/decimal.md). Timeunit is not timestamp (seconds), it's -- an index of the time interval. Can be calculated using [intDiv](../../functions/arithmetic-functions/#intdiva-b).
**Parameters**
@ -148,3 +148,58 @@ Result:
│ 1 │ 49 │ 0.825 │ █████████████████████████████████████████▎│
└───────┴──────┴──────────────────────┴────────────────────────────────────────────┘
```
```sql
CREATE TABLE data
ENGINE = Memory AS
SELECT
10 AS value,
toDateTime('2020-01-01') + (3600 * number) AS time
FROM numbers_mt(10);
-- Calculate timeunit using intDiv
SELECT
value,
time,
exponentialMovingAverage(1)(value, intDiv(toUInt32(time), 3600)) OVER (ORDER BY time ASC) AS res,
intDiv(toUInt32(time), 3600) AS timeunit
FROM data
ORDER BY time ASC;
┌─value─┬────────────────time─┬─────────res─┬─timeunit─┐
│ 10 │ 2020-01-01 00:00:00 │ 5 │ 438288 │
│ 10 │ 2020-01-01 01:00:00 │ 7.5 │ 438289 │
│ 10 │ 2020-01-01 02:00:00 │ 8.75 │ 438290 │
│ 10 │ 2020-01-01 03:00:00 │ 9.375 │ 438291 │
│ 10 │ 2020-01-01 04:00:00 │ 9.6875 │ 438292 │
│ 10 │ 2020-01-01 05:00:00 │ 9.84375 │ 438293 │
│ 10 │ 2020-01-01 06:00:00 │ 9.921875 │ 438294 │
│ 10 │ 2020-01-01 07:00:00 │ 9.9609375 │ 438295 │
│ 10 │ 2020-01-01 08:00:00 │ 9.98046875 │ 438296 │
│ 10 │ 2020-01-01 09:00:00 │ 9.990234375 │ 438297 │
└───────┴─────────────────────┴─────────────┴──────────┘
-- Calculate timeunit using toRelativeHourNum
SELECT
value,
time,
exponentialMovingAverage(1)(value, toRelativeHourNum(time)) OVER (ORDER BY time ASC) AS res,
toRelativeHourNum(time) AS timeunit
FROM data
ORDER BY time ASC;
┌─value─┬────────────────time─┬─────────res─┬─timeunit─┐
│ 10 │ 2020-01-01 00:00:00 │ 5 │ 438288 │
│ 10 │ 2020-01-01 01:00:00 │ 7.5 │ 438289 │
│ 10 │ 2020-01-01 02:00:00 │ 8.75 │ 438290 │
│ 10 │ 2020-01-01 03:00:00 │ 9.375 │ 438291 │
│ 10 │ 2020-01-01 04:00:00 │ 9.6875 │ 438292 │
│ 10 │ 2020-01-01 05:00:00 │ 9.84375 │ 438293 │
│ 10 │ 2020-01-01 06:00:00 │ 9.921875 │ 438294 │
│ 10 │ 2020-01-01 07:00:00 │ 9.9609375 │ 438295 │
│ 10 │ 2020-01-01 08:00:00 │ 9.98046875 │ 438296 │
│ 10 │ 2020-01-01 09:00:00 │ 9.990234375 │ 438297 │
└───────┴─────────────────────┴─────────────┴──────────┘
```

39
docs/en/sql-reference/functions/date-time-functions.md
View File

@ -1126,15 +1126,48 @@ Rounds the time to the half hour.
## toYYYYMM
Converts a date or date with time to a UInt32 number containing the year and month number (YYYY \* 100 + MM).
Converts a date or date with time to a UInt32 number containing the year and month number (YYYY \* 100 + MM). Accepts a second optional timezone argument. If provided, the timezone must be a string constant.
### example
```sql
SELECT
toYYYYMM(now(), 'US/Eastern')
```
```response
┌─toYYYYMM(now(), 'US/Eastern')─┐
│ 202303 │
└───────────────────────────────┘
```
## toYYYYMMDD
Converts a date or date with time to a UInt32 number containing the year and month number (YYYY \* 10000 + MM \* 100 + DD).
Converts a date or date with time to a UInt32 number containing the year and month number (YYYY \* 10000 + MM \* 100 + DD). Accepts a second optional timezone argument. If provided, the timezone must be a string constant.
### example
```sql
SELECT
toYYYYMMDD(now(), 'US/Eastern')
```
```response
┌─toYYYYMMDD(now(), 'US/Eastern')─┐
│ 20230302 │
└─────────────────────────────────┘
```
## toYYYYMMDDhhmmss
Converts a date or date with time to a UInt64 number containing the year and month number (YYYY \* 10000000000 + MM \* 100000000 + DD \* 1000000 + hh \* 10000 + mm \* 100 + ss).
Converts a date or date with time to a UInt64 number containing the year and month number (YYYY \* 10000000000 + MM \* 100000000 + DD \* 1000000 + hh \* 10000 + mm \* 100 + ss). Accepts a second optional timezone argument. If provided, the timezone must be a string constant.
### example
```sql
SELECT
toYYYYMMDDhhmmss(now(), 'US/Eastern')
```
```response
┌─toYYYYMMDDhhmmss(now(), 'US/Eastern')─┐
│ 20230302112209 │
└───────────────────────────────────────┘
```
## addYears, addMonths, addWeeks, addDays, addHours, addMinutes, addSeconds, addQuarters

10
docs/en/sql-reference/functions/ip-address-functions.md
View File

@ -280,12 +280,20 @@ SELECT
## toIPv4OrDefault(string)
Same as `toIPv4`, but if the IPv4 address has an invalid format, it returns 0.
Same as `toIPv4`, but if the IPv4 address has an invalid format, it returns `0.0.0.0` (0 IPv4).
## toIPv4OrNull(string)
Same as `toIPv4`, but if the IPv4 address has an invalid format, it returns null.
## toIPv6OrDefault(string)
Same as `toIPv6`, but if the IPv6 address has an invalid format, it returns `::` (0 IPv6).
## toIPv6OrNull(string)
Same as `toIPv6`, but if the IPv6 address has an invalid format, it returns null.
## toIPv6
Converts a string form of IPv6 address to [IPv6](../../sql-reference/data-types/domains/ipv6.md) type. If the IPv6 address has an invalid format, returns an empty value.

2
docs/en/sql-reference/functions/string-functions.md
View File

@ -330,7 +330,7 @@ repeat(s, n)
**Arguments**
- `s` — The string to repeat. [String](../../sql-reference/data-types/string.md).
- `n` — The number of times to repeat the string. [UInt](../../sql-reference/data-types/int-uint.md).
- `n` — The number of times to repeat the string. [UInt or Int](../../sql-reference/data-types/int-uint.md).
**Returned value**

11
docs/en/sql-reference/operators/in.md
View File

@ -233,8 +233,9 @@ If `some_predicate` is not selective enough, it will return large amount of data
### Distributed Subqueries and max_parallel_replicas
When max_parallel_replicas is greater than 1, distributed queries are further transformed. For example, the following:
When [max_parallel_replicas](#settings-max_parallel_replicas) is greater than 1, distributed queries are further transformed.
For example, the following:
```sql
SELECT CounterID, count() FROM distributed_table_1 WHERE UserID IN (SELECT UserID FROM local_table_2 WHERE CounterID < 100)
SETTINGS max_parallel_replicas=3
@ -247,8 +248,12 @@ SELECT CounterID, count() FROM local_table_1 WHERE UserID IN (SELECT UserID FROM
SETTINGS parallel_replicas_count=3, parallel_replicas_offset=M
```
where M is between 1 and 3 depending on which replica the local query is executing on. These settings affect every MergeTree-family table in the query and have the same effect as applying `SAMPLE 1/3 OFFSET (M-1)/3` on each table.
where M is between 1 and 3 depending on which replica the local query is executing on.
Therefore adding the max_parallel_replicas setting will only produce correct results if both tables have the same replication scheme and are sampled by UserID or a subkey of it. In particular, if local_table_2 does not have a sampling key, incorrect results will be produced. The same rule applies to JOIN.
These settings affect every MergeTree-family table in the query and have the same effect as applying `SAMPLE 1/3 OFFSET (M-1)/3` on each table.
Therefore adding the [max_parallel_replicas](#settings-max_parallel_replicas) setting will only produce correct results if both tables have the same replication scheme and are sampled by UserID or a subkey of it. In particular, if local_table_2 does not have a sampling key, incorrect results will be produced. The same rule applies to JOIN.
One workaround if local_table_2 does not meet the requirements, is to use `GLOBAL IN` or `GLOBAL JOIN`.
If a table doesn't have a sampling key, more flexible options for [parallel_replicas_custom_key](#settings-parallel_replicas_custom_key) can be used that can produce different and more optimal behaviour.

37
docs/en/sql-reference/statements/create/table.md
View File

@ -110,25 +110,23 @@ If the type is not `Nullable` and if `NULL` is specified, it will be treated as
See also [data_type_default_nullable](../../../operations/settings/settings.md#data_type_default_nullable) setting.
## Default Values
## Default Values {#default_values}
The column description can specify an expression for a default value, in one of the following ways: `DEFAULT expr`, `MATERIALIZED expr`, `ALIAS expr`.
The column description can specify a default value expression in the form of `DEFAULT expr`, `MATERIALIZED expr`, or `ALIAS expr`. Example: `URLDomain String DEFAULT domain(URL)`.
Example: `URLDomain String DEFAULT domain(URL)`.
The expression `expr` is optional. If it is omitted, the column type must be specified explicitly and the default value will be `0` for numeric columns, `''` (the empty string) for string columns, `[]` (the empty array) for array columns, `1970-01-01` for date columns, or `NULL` for nullable columns.
If an expression for the default value is not defined, the default values will be set to zeros for numbers, empty strings for strings, empty arrays for arrays, and `1970-01-01` for dates or zero unix timestamp for DateTime, NULL for Nullable.
The column type of a default value column can be omitted in which case it is infered from `expr`'s type. For example the type of column `EventDate DEFAULT toDate(EventTime)` will be date.
If the default expression is defined, the column type is optional. If there isn’t an explicitly defined type, the default expression type is used. Example: `EventDate DEFAULT toDate(EventTime)` – the ‘Date’ type will be used for the ‘EventDate’ column.
If both a data type and a default value expression are specified, an implicit type casting function inserted which converts the expression to the specified type. Example: `Hits UInt32 DEFAULT 0` is internally represented as `Hits UInt32 DEFAULT toUInt32(0)`.
If the data type and default expression are defined explicitly, this expression will be cast to the specified type using type casting functions. Example: `Hits UInt32 DEFAULT 0` means the same thing as `Hits UInt32 DEFAULT toUInt32(0)`.
Default expressions may be defined as an arbitrary expression from table constants and columns. When creating and changing the table structure, it checks that expressions do not contain loops. For INSERT, it checks that expressions are resolvable – that all columns they can be calculated from have been passed.
A default value expression `expr` may reference arbitrary table columns and constants. ClickHouse checks that changes of the table structure do not introduce loops in the expression calculation. For INSERT, it checks that expressions are resolvable – that all columns they can be calculated from have been passed.
### DEFAULT
`DEFAULT expr`
Normal default value. If the INSERT query does not specify the corresponding column, it will be filled in by computing the corresponding expression.
Normal default value. If the value of such a column is not specified in an INSERT query, it is computed from `expr`.
Example:
@ -154,9 +152,9 @@ SELECT * FROM test;
`MATERIALIZED expr`
Materialized expression. Such a column can’t be specified for INSERT, because it is always calculated.
For an INSERT without a list of columns, these columns are not considered.
In addition, this column is not substituted when using an asterisk in a SELECT query. This is to preserve the invariant that the dump obtained using `SELECT *` can be inserted back into the table using INSERT without specifying the list of columns.
Materialized expression. Values of such columns are always calculated, they cannot be specified in INSERT queries.
Also, default value columns of this type are not included in the result of `SELECT *`. This is to preserve the invariant that the result of a `SELECT *` can always be inserted back into the table using `INSERT`. This behavior can be disabled with setting `asterisk_include_materialized_columns`.
Example:
@ -192,8 +190,9 @@ SELECT * FROM test SETTINGS asterisk_include_materialized_columns=1;
`EPHEMERAL [expr]`
Ephemeral column. Such a column isn't stored in the table and cannot be SELECTed, but can be referenced in the defaults of CREATE statement. If `expr` is omitted type for column is required.
INSERT without list of columns will skip such column, so SELECT/INSERT invariant is preserved - the dump obtained using `SELECT *` can be inserted back into the table using INSERT without specifying the list of columns.
Ephemeral column. Columns of this type are not stored in the table and it is not possible to SELECT from them. The only purpose of ephemeral columns is to build default value expressions of other columns from them.
An insert without explicitly specified columns will skip columns of this type. This is to preserve the invariant that the result of a `SELECT *` can always be inserted back into the table using `INSERT`.
Example:
@ -205,7 +204,7 @@ CREATE OR REPLACE TABLE test
hexed FixedString(4) DEFAULT unhex(unhexed)
)
ENGINE = MergeTree
ORDER BY id
ORDER BY id;
INSERT INTO test (id, unhexed) Values (1, '5a90b714');
@ -227,9 +226,9 @@ hex(hexed): 5A90B714
`ALIAS expr`
Synonym. Such a column isn’t stored in the table at all.
Its values can’t be inserted in a table, and it is not substituted when using an asterisk in a SELECT query.
It can be used in SELECTs if the alias is expanded during query parsing.
Calculated columns (synonym). Column of this type are not stored in the table and it is not possible to INSERT values into them.
When SELECT queries explicitly reference columns of this type, the value is computed at query time from `expr`. By default, `SELECT *` excludes ALIAS columns. This behavior can be disabled with setting `asteriks_include_alias_columns`.
When using the ALTER query to add new columns, old data for these columns is not written. Instead, when reading old data that does not have values for the new columns, expressions are computed on the fly by default. However, if running the expressions requires different columns that are not indicated in the query, these columns will additionally be read, but only for the blocks of data that need it.
@ -576,7 +575,7 @@ SELECT * FROM base.t1;
You can add a comment to the table when you creating it.
:::note
The comment is supported for all table engines except [Kafka](../../../engines/table-engines/integrations/kafka.md), [RabbitMQ](../../../engines/table-engines/integrations/rabbitmq.md) and [EmbeddedRocksDB](../../../engines/table-engines/integrations/embedded-rocksdb.md).
The comment clause is supported by all table engines except [Kafka](../../../engines/table-engines/integrations/kafka.md), [RabbitMQ](../../../engines/table-engines/integrations/rabbitmq.md) and [EmbeddedRocksDB](../../../engines/table-engines/integrations/embedded-rocksdb.md).
:::

6
docs/en/sql-reference/statements/create/view.md
View File

@ -70,6 +70,12 @@ A materialized view is implemented as follows: when inserting data to the table
Materialized views in ClickHouse use **column names** instead of column order during insertion into destination table. If some column names are not present in the `SELECT` query result, ClickHouse uses a default value, even if the column is not [Nullable](../../data-types/nullable.md). A safe practice would be to add aliases for every column when using Materialized views.
Materialized views in ClickHouse are implemented more like insert triggers. If there’s some aggregation in the view query, it’s applied only to the batch of freshly inserted data. Any changes to existing data of source table (like update, delete, drop partition, etc.) does not change the materialized view.
Materialized views in ClickHouse do not have deterministic behaviour in case of errors. This means that blocks that had been already written will be preserved in the destination table, but all blocks after error will not.
By default if pushing to one of views fails, then the INSERT query will fail too, and some blocks may not be written to the destination table. This can be changed using `materialized_views_ignore_errors` setting (you should set it for `INSERT` query), if you will set `materialized_views_ignore_errors=true`, then any errors while pushing to views will be ignored and all blocks will be written to the destination table.
Also note, that `materialized_views_ignore_errors` set to `true` by default for `system.*_log` tables.
:::
If you specify `POPULATE`, the existing table data is inserted into the view when creating it, as if making a `CREATE TABLE ... AS SELECT ...` . Otherwise, the query contains only the data inserted in the table after creating the view. We **do not recommend** using `POPULATE`, since data inserted in the table during the view creation will not be inserted in it.

15
docs/en/sql-reference/table-functions/file.md
View File

@ -6,21 +6,22 @@ sidebar_label: file
# file
Creates a table from a file. This table function is similar to [url](../../sql-reference/table-functions/url.md) and [hdfs](../../sql-reference/table-functions/hdfs.md) ones.
Creates a table from a file. This table function is similar to [url](/docs/en/sql-reference/table-functions/url.md) and [hdfs](/docs/en/sql-reference/table-functions/hdfs.md) ones.
`file` function can be used in `SELECT` and `INSERT` queries on data in [File](../../engines/table-engines/special/file.md) tables.
`file` function can be used in `SELECT` and `INSERT` queries on data in [File](/docs/en/engines/table-engines/special/file.md) tables.
**Syntax**
``` sql
file(path [,format] [,structure])
file(path [,format] [,structure] [,compression])
```
**Parameters**
- `path` — The relative path to the file from [user_files_path](../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-user_files_path). Path to file support following globs in read-only mode: `*`, `?`, `{abc,def}` and `{N..M}` where `N`, `M` — numbers, `'abc', 'def'` — strings.
- `format` — The [format](../../interfaces/formats.md#formats) of the file.
- `path` — The relative path to the file from [user_files_path](/docs/en/operations/server-configuration-parameters/settings.md#server_configuration_parameters-user_files_path). Path to file support following globs in read-only mode: `*`, `?`, `{abc,def}` and `{N..M}` where `N`, `M` — numbers, `'abc', 'def'` — strings.
- `format` — The [format](/docs/en/interfaces/formats.md#formats) of the file.
- `structure` — Structure of the table. Format: `'column1_name column1_type, column2_name column2_type, ...'`.
- `compression` — The existing compression type when used in a `SELECT` query, or the desired compression type when used in an `INSERT` query. The supported compression types are `gz`, `br`, `xz`, `zst`, `lz4`, and `bz2`.
**Returned value**
@ -53,7 +54,7 @@ SELECT * FROM file('test.csv', 'CSV', 'column1 UInt32, column2 UInt32, column3 U
└─────────┴─────────┴─────────┘
```
Getting the first 10 lines of a table that contains 3 columns of [UInt32](../../sql-reference/data-types/int-uint.md) type from a CSV file:
Getting the first 10 lines of a table that contains 3 columns of [UInt32](/docs/en/sql-reference/data-types/int-uint.md) type from a CSV file:
``` sql
SELECT * FROM file('test.csv', 'CSV', 'column1 UInt32, column2 UInt32, column3 UInt32') LIMIT 10;
@ -143,4 +144,4 @@ SELECT count(*) FROM file('big_dir/**/file002', 'CSV', 'name String, value UInt3
**See Also**
- [Virtual columns](../../engines/table-engines/index.md#table_engines-virtual_columns)
- [Virtual columns](/docs/en/engines/table-engines/index.md#table_engines-virtual_columns)

2
docs/ru/engines/table-engines/mergetree-family/collapsingmergetree.md
View File

@ -89,7 +89,7 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster]
└─────────────────────┴───────────┴──────────┴──────┘
```
Первая строка отменяет предыдущее состояние объекта (пользователя). Она должен повторять все поля из ключа сортировки для отменённого состояния за исключением `Sign`.
Первая строка отменяет предыдущее состояние объекта (пользователя). Она должна повторять все поля из ключа сортировки для отменённого состояния за исключением `Sign`.
Вторая строка содержит текущее состояние.

6
docs/ru/engines/table-engines/mergetree-family/mergetree.md
View File

@ -584,7 +584,7 @@ TTL d + INTERVAL 1 MONTH GROUP BY k1, k2 SET x = max(x), y = min(y);
Данные с истекшим `TTL` удаляются, когда ClickHouse мёржит куски данных.
Когда ClickHouse видит, что некоторые данные устарели, он выполняет внеплановые мёржи. Для управление частотой подобных мёржей, можно задать настройку `merge_with_ttl_timeout`. Если её значение слишком низкое, придется выполнять много внеплановых мёржей, которые могут начать потреблять значительную долю ресурсов сервера.
Когда ClickHouse видит, что некоторые данные устарели, он выполняет внеплановые мёржи. Для управления частотой подобных мёржей, можно задать настройку `merge_with_ttl_timeout`. Если её значение слишком низкое, придется выполнять много внеплановых мёржей, которые могут начать потреблять значительную долю ресурсов сервера.
Если вы выполните запрос `SELECT` между слияниями вы можете получить устаревшие данные. Чтобы избежать этого используйте запрос [OPTIMIZE](../../../engines/table-engines/mergetree-family/mergetree.md#misc_operations-optimize) перед `SELECT`.
@ -679,7 +679,7 @@ TTL d + INTERVAL 1 MONTH GROUP BY k1, k2 SET x = max(x), y = min(y);
- `policy_name_N` — название политики. Названия политик должны быть уникальны.
- `volume_name_N` — название тома. Названия томов должны быть уникальны.
- `disk` — диск, находящийся внутри тома.
- `max_data_part_size_bytes` — максимальный размер куска данных, который может находится на любом из дисков этого тома. Если в результате слияния размер куска ожидается больше, чем max_data_part_size_bytes, то этот кусок будет записан в следующий том. В основном эта функция позволяет хранить новые / мелкие куски на горячем (SSD) томе и перемещать их на холодный (HDD) том, когда они достигают большого размера. Не используйте этот параметр, если политика имеет только один том.
- `max_data_part_size_bytes` — максимальный размер куска данных, который может находиться на любом из дисков этого тома. Если в результате слияния размер куска ожидается больше, чем max_data_part_size_bytes, то этот кусок будет записан в следующий том. В основном эта функция позволяет хранить новые / мелкие куски на горячем (SSD) томе и перемещать их на холодный (HDD) том, когда они достигают большого размера. Не используйте этот параметр, если политика имеет только один том.
- `move_factor` — доля доступного свободного места на томе, если места становится меньше, то данные начнут перемещение на следующий том, если он есть (по умолчанию 0.1). Для перемещения куски сортируются по размеру от большего к меньшему (по убыванию) и выбираются куски, совокупный размер которых достаточен для соблюдения условия `move_factor`, если совокупный размер всех партов недостаточен, будут перемещены все парты.
- `prefer_not_to_merge` — Отключает слияние кусков данных, хранящихся на данном томе. Если данная настройка включена, то слияние данных, хранящихся на данном томе, не допускается. Это позволяет контролировать работу ClickHouse с медленными дисками.
@ -730,7 +730,7 @@ TTL d + INTERVAL 1 MONTH GROUP BY k1, k2 SET x = max(x), y = min(y);
В приведенном примере, политика `hdd_in_order` реализует прицип [round-robin](https://ru.wikipedia.org/wiki/Round-robin_(%D0%B0%D0%BB%D0%B3%D0%BE%D1%80%D0%B8%D1%82%D0%BC)). Так как в политике есть всего один том (`single`), то все записи производятся на его диски по круговому циклу. Такая политика может быть полезна при наличии в системе нескольких похожих дисков, но при этом не сконфигурирован RAID. Учтите, что каждый отдельный диск ненадёжен и чтобы не потерять важные данные это необходимо скомпенсировать за счет хранения данных в трёх копиях.
Если система содержит диски различных типов, то может пригодиться политика `moving_from_ssd_to_hdd`. В томе `hot` находится один SSD-диск (`fast_ssd`), а также задается ограничение на максимальный размер куска, который может храниться на этом томе (1GB). Все куски такой таблицы больше 1GB будут записываться сразу на том `cold`, в котором содержится один HDD-диск `disk1`. Также, при заполнении диска `fast_ssd` более чем на 80% данные будут переносится на диск `disk1` фоновым процессом.
Если система содержит диски различных типов, то может пригодиться политика `moving_from_ssd_to_hdd`. В томе `hot` находится один SSD-диск (`fast_ssd`), а также задается ограничение на максимальный размер куска, который может храниться на этом томе (1GB). Все куски такой таблицы больше 1GB будут записываться сразу на том `cold`, в котором содержится один HDD-диск `disk1`. Также при заполнении диска `fast_ssd` более чем на 80% данные будут переноситься на диск `disk1` фоновым процессом.
Порядок томов в политиках хранения важен, при достижении условий на переполнение тома данные переносятся на следующий. Порядок дисков в томах так же важен, данные пишутся по очереди на каждый из них.

1
docs/ru/interfaces/third-party/client-libraries.md vendored
View File

@ -8,6 +8,7 @@ sidebar_label: "Клиентские библиотеки от сторонни
:::danger "Disclaimer"
Яндекс не поддерживает перечисленные ниже библиотеки и не проводит тщательного тестирования для проверки их качества.
:::
- Python:
- [infi.clickhouse_orm](https://github.com/Infinidat/infi.clickhouse_orm)

9
docs/ru/interfaces/third-party/gui.md vendored
View File

@ -177,19 +177,20 @@ sidebar_label: "Визуальные интерфейсы от сторонни
### Yandex DataLens {#yandex-datalens}
[Yandex DataLens](https://cloud.yandex.ru/services/datalens) — cервис визуализации и анализа данных.
[Yandex DataLens](https://datalens.yandex.ru) — cервис визуализации и анализа данных.
Основные возможности:
- Широкий выбор инструментов визуализации, от простых столбчатых диаграмм до сложных дашбордов.
- Возможность опубликовать дашборды на широкую аудиторию.
- Поддержка множества источников данных, включая ClickHouse.
- Хранение материализованных данных в кластере ClickHouse DataLens.
Для небольших проектов DataLens [доступен бесплатно](https://cloud.yandex.ru/docs/datalens/pricing), в том числе и для коммерческого использования.
DataLens [доступен бесплатно](https://cloud.yandex.ru/docs/datalens/pricing), в том числе и для коммерческого использования.
- [Знакомство с DataLens]((https://youtu.be/57ngi_6BINE).
- [Чат сообщества DataLens](https://t.me/YandexDataLens)
- [Документация DataLens](https://cloud.yandex.ru/docs/datalens/).
- [Пособие по визуализации данных из ClickHouse](https://cloud.yandex.ru/docs/solutions/datalens/data-from-ch-visualization).
- [Сценарий по визуализации данных из ClickHouse](https://cloud.yandex.ru/docs/solutions/datalens/data-from-ch-visualization).
### Holistics Software {#holistics-software}

12
docs/ru/operations/clickhouse-keeper.md
View File

@ -325,21 +325,21 @@ clickhouse-keeper-converter --zookeeper-logs-dir /var/lib/zookeeper/version-2 --
Например, для кластера из 3 нод, алгоритм кворума продолжает работать при отказе не более чем одной ноды.
Конфигурация кластера может быть изменена динамически с некоторыми ограничениями.
Переконфигурация также использует Raft, поэтому для добавление новой ноды кластера или исключения старой ноды из него требуется достижения кворума в рамках текущей конфигурации кластера.
Переконфигурация также использует Raft, поэтому для добавления новой ноды кластера или исключения старой ноды требуется достижение кворума в рамках текущей конфигурации кластера.
Если в вашем кластере произошел отказ большего числа нод, чем допускает Raft для вашей текущей конфигурации и у вас нет возможности восстановить их работоспособность, Raft перестанет работать и не позволит изменить конфигурацию стандартным механизмом.
Тем не менее ClickHousr Keeper имеет возможность запуститься в режиме восстановления, который позволяет переконфигурировать класте используя только одну ноду кластера.
Тем не менее ClickHouse Keeper имеет возможность запуститься в режиме восстановления, который позволяет переконфигурировать кластер используя только одну ноду кластера.
Этот механизм может использоваться только как крайняя мера, когда вы не можете восстановить существующие ноды кластера или запустить новый сервер с тем же идентификатором.
Важно:
- Удостоверьтесь, что отказавшие ноды не смогут в дальнейшем подключиться к кластеру в будущем.
- Не запускайте новые ноды, пока не завешите процедуру ниже.
- Не запускайте новые ноды, пока не завершите процедуру ниже.
После того, как выполнили действия выше выполните следующие шаги.
1. Выберете одну ноду Keeper, которая станет новым лидером. Учтите, что данные которые с этой ноды будут испольщзованы всем кластером, поэтому рекомендуется выбрать ноду с наиболее актуальным состоянием.
1. Выберете одну ноду Keeper, которая станет новым лидером. Учтите, что данные с этой ноды будут использованы всем кластером, поэтому рекомендуется выбрать ноду с наиболее актуальным состоянием.
2. Перед дальнейшими действиям сделайте резервную копию данных из директорий `log_storage_path` и `snapshot_storage_path`.
3. Измените настройки на всех нодах кластера, которые вы собираетесь использовать.
4. Отправьте команду `rcvr` на ноду, которую вы выбрали или остановите ее и запустите заново с аргументом `--force-recovery`. Это переведет ноду в режим восстановления.
4. Отправьте команду `rcvr` на ноду, которую вы выбрали, или остановите ее и запустите заново с аргументом `--force-recovery`. Это переведет ноду в режим восстановления.
5. Запускайте остальные ноды кластера по одной и проверяйте, что команда `mntr` возвращает `follower` в выводе состояния `zk_server_state` перед тем, как запустить следующую ноду.
6. Пока нода работает в режиме восстановления, лидер будет возвращать ошибку на запрос `mntr` пока кворум не будет достигнут с помощью новых нод. Любые запросы от клиентов и постедователей будут возвращать ошибку.
6. Пока нода работает в режиме восстановления, лидер будет возвращать ошибку на запрос `mntr` пока кворум не будет достигнут с помощью новых нод. Любые запросы от клиентов и последователей будут возвращать ошибку.
7. После достижения кворума лидер перейдет в нормальный режим работы и станет обрабатывать все запросы через Raft. Удостоверьтесь, что запрос `mntr` возвращает `leader` в выводе состояния `zk_server_state`.

1
docs/ru/operations/opentelemetry.md
View File

@ -10,6 +10,7 @@ ClickHouse поддерживает [OpenTelemetry](https://opentelemetry.io/)
:::danger "Предупреждение"
Поддержка стандарта экспериментальная и будет со временем меняться.
:::
## Обеспечение поддержки контекста трассировки в ClickHouse

3
docs/ru/operations/server-configuration-parameters/settings.md
View File

@ -26,6 +26,7 @@ ClickHouse перезагружает встроенные словари с з
:::danger "Внимание"
Лучше не использовать, если вы только начали работать с ClickHouse.
:::
Общий вид конфигурации:
@ -1064,6 +1065,7 @@ ClickHouse использует потоки из глобального пул
:::danger "Обратите внимание"
Завершающий слеш обязателен.
:::
**Пример**
@ -1330,6 +1332,7 @@ TCP порт для защищённого обмена данными с кли
:::danger "Обратите внимание"
Завершающий слеш обязателен.
:::
**Пример**

2
docs/ru/operations/storing-data.md
View File

@ -82,7 +82,7 @@ sidebar_label: "Хранение данных на внешних дисках"
- `type` — `encrypted`. Иначе зашифрованный диск создан не будет.
- `disk` — тип диска для хранения данных.
- `key` — ключ для шифрования и расшифровки. Тип: [Uint64](../sql-reference/data-types/int-uint.md). Вы можете использовать параметр `key_hex` для шифрования в шестнадцатеричной форме.
- `key` — ключ для шифрования и расшифровки. Тип: [UInt64](../sql-reference/data-types/int-uint.md). Вы можете использовать параметр `key_hex` для шифрования в шестнадцатеричной форме.
Вы можете указать несколько ключей, используя атрибут `id` (смотрите пример выше).
Необязательные параметры:

2
docs/ru/sql-reference/data-types/aggregatefunction.md
View File

@ -6,7 +6,7 @@ sidebar_label: AggregateFunction
# AggregateFunction {#data-type-aggregatefunction}
Агрегатные функции могут обладать определяемым реализацией промежуточным состоянием, которое может быть сериализовано в тип данных, соответствующий AggregateFunction(…), и быть записано в таблицу обычно посредством [материализованного представления] (../../sql-reference/statements/create/view.md). Чтобы получить промежуточное состояние, обычно используются агрегатные функции с суффиксом `-State`. Чтобы в дальнейшем получить агрегированные данные необходимо использовать те же агрегатные функции с суффиксом `-Merge`.
Агрегатные функции могут обладать определяемым реализацией промежуточным состоянием, которое может быть сериализовано в тип данных, соответствующий AggregateFunction(…), и быть записано в таблицу обычно посредством [материализованного представления](../../sql-reference/statements/create/view.md). Чтобы получить промежуточное состояние, обычно используются агрегатные функции с суффиксом `-State`. Чтобы в дальнейшем получить агрегированные данные необходимо использовать те же агрегатные функции с суффиксом `-Merge`.
`AggregateFunction(name, types_of_arguments…)` — параметрический тип данных.

1
docs/ru/sql-reference/data-types/geo.md
View File

@ -10,6 +10,7 @@ ClickHouse поддерживает типы данных для отображ
:::danger "Предупреждение"
Сейчас использование типов данных для работы с географическими структурами является экспериментальной возможностью. Чтобы использовать эти типы данных, включите настройку `allow_experimental_geo_types = 1`.
:::
**См. также**
- [Хранение географических структур данных](https://ru.wikipedia.org/wiki/GeoJSON).

1
docs/ru/sql-reference/data-types/special-data-types/interval.md
View File

@ -10,6 +10,7 @@ sidebar_label: Interval
:::danger "Внимание"
Нельзя использовать типы данных `Interval` для хранения данных в таблице.
:::
Структура:

2
docs/ru/sql-reference/data-types/tuple.md
View File

@ -34,7 +34,7 @@ SELECT tuple(1,'a') AS x, toTypeName(x)
## Особенности работы с типами данных {#osobennosti-raboty-s-tipami-dannykh}
При создании кортежа «на лету» ClickHouse автоматически определяет тип каждого аргументов как минимальный из типов, который может сохранить значение аргумента. Если аргумент — [NULL](../../sql-reference/data-types/tuple.md#null-literal), то тип элемента кортежа — [Nullable](nullable.md).
При создании кортежа «на лету» ClickHouse автоматически определяет тип всех аргументов как минимальный из типов, который может сохранить значение аргумента. Если аргумент — [NULL](../../sql-reference/data-types/tuple.md#null-literal), то тип элемента кортежа — [Nullable](nullable.md).
Пример автоматического определения типа данных:

4
docs/ru/sql-reference/dictionaries/external-dictionaries/external-dicts-dict-polygon.md
View File

@ -61,7 +61,7 @@ LAYOUT(POLYGON(STORE_POLYGON_KEY_COLUMN 1))
- Мультиполигон. Представляет из себя массив полигонов. Каждый полигон задается двумерным массивом точек — первый элемент этого массива задает внешнюю границу полигона,
последующие элементы могут задавать дырки, вырезаемые из него.
Точки могут задаваться массивом или кортежем из своих координат. В текущей реализации поддерживается только двумерные точки.
Точки могут задаваться массивом или кортежем из своих координат. В текущей реализации поддерживаются только двумерные точки.
Пользователь может [загружать свои собственные данные](../../../sql-reference/dictionaries/external-dictionaries/external-dicts-dict-sources.md) во всех поддерживаемых ClickHouse форматах.
@ -80,7 +80,7 @@ LAYOUT(POLYGON(STORE_POLYGON_KEY_COLUMN 1))
- `POLYGON`. Синоним к `POLYGON_INDEX_CELL`.
Запросы к словарю осуществляются с помощью стандартных [функций](../../../sql-reference/functions/ext-dict-functions.md) для работы со внешними словарями.
Важным отличием является то, что здесь ключами будут являются точки, для которых хочется найти содержащий их полигон.
Важным отличием является то, что здесь ключами являются точки, для которых хочется найти содержащий их полигон.
**Пример**

1
docs/ru/sql-reference/dictionaries/external-dictionaries/external-dicts-dict-structure.md
View File

@ -59,6 +59,7 @@ ClickHouse поддерживает следующие виды ключей:
:::danger "Обратите внимание"
Ключ не надо дополнительно описывать в атрибутах.
:::
### Числовой ключ {#ext_dict-numeric-key}

2
docs/ru/sql-reference/dictionaries/external-dictionaries/external-dicts.md
View File

@ -14,7 +14,7 @@ ClickHouse:
- Периодически обновляет их и динамически подгружает отсутствующие значения.
- Позволяет создавать внешние словари с помощью xml-файлов или [DDL-запросов](../../statements/create/dictionary.md#create-dictionary-query).
Конфигурация внешних словарей может находится в одном или нескольких xml-файлах. Путь к конфигурации указывается в параметре [dictionaries_config](../../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-dictionaries_config).
Конфигурация внешних словарей может находиться в одном или нескольких xml-файлах. Путь к конфигурации указывается в параметре [dictionaries_config](../../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-dictionaries_config).
Словари могут загружаться при старте сервера или при первом использовании, в зависимости от настройки [dictionaries_lazy_load](../../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-dictionaries_lazy_load).

2
docs/ru/sql-reference/functions/introspection.md
View File

@ -22,7 +22,7 @@ sidebar_label: "Функции интроспекции"
ClickHouse сохраняет отчеты профилировщика в [журнал трассировки](../../operations/system-tables/trace_log.md#system_tables-trace_log) в системной таблице. Убедитесь, что таблица и профилировщик настроены правильно.
## addresssToLine {#addresstoline}
## addressToLine {#addresstoline}
Преобразует адрес виртуальной памяти внутри процесса сервера ClickHouse в имя файла и номер строки в исходном коде ClickHouse.

3
docs/ru/sql-reference/operators/exists.md
View File

@ -8,7 +8,8 @@ slug: /ru/sql-reference/operators/exists
`EXISTS` может быть использован в секции [WHERE](../../sql-reference/statements/select/where.md).
:::danger "Предупреждение"
Ссылки на таблицы или столбцы основного запроса не поддерживаются в подзапросе.
Ссылки на таблицы или столбцы основного запроса не поддерживаются в подзапросе.
:::
**Синтаксис**

12
docs/ru/sql-reference/operators/in.md
View File

@ -38,9 +38,9 @@ SELECT '1' IN (SELECT 1);
└──────────────────────┘
```
Если в качестве правой части оператора указано имя таблицы (например, `UserID IN users`), то это эквивалентно подзапросу `UserID IN (SELECT * FROM users)`. Это используется при работе с внешними данными, отправляемым вместе с запросом. Например, вместе с запросом может быть отправлено множество идентификаторов посетителей, загруженное во временную таблицу users, по которому следует выполнить фильтрацию.
Если в качестве правой части оператора указано имя таблицы (например, `UserID IN users`), то это эквивалентно подзапросу `UserID IN (SELECT * FROM users)`. Это используется при работе с внешними данными, отправляемыми вместе с запросом. Например, вместе с запросом может быть отправлено множество идентификаторов посетителей, загруженное во временную таблицу users, по которому следует выполнить фильтрацию.
Если в качестве правой части оператора, указано имя таблицы, имеющий движок Set (подготовленное множество, постоянно находящееся в оперативке), то множество не будет создаваться заново при каждом запросе.
Если в качестве правой части оператора, указано имя таблицы, имеющей движок Set (подготовленное множество, постоянно находящееся в оперативке), то множество не будет создаваться заново при каждом запросе.
В подзапросе может быть указано более одного столбца для фильтрации кортежей.
Пример:
@ -49,9 +49,9 @@ SELECT '1' IN (SELECT 1);
SELECT (CounterID, UserID) IN (SELECT CounterID, UserID FROM ...) FROM ...
```
Типы столбцов слева и справа оператора IN, должны совпадать.
Типы столбцов слева и справа оператора IN должны совпадать.
Оператор IN и подзапрос могут встречаться в любой части запроса, в том числе в агрегатных и лямбда функциях.
Оператор IN и подзапрос могут встречаться в любой части запроса, в том числе в агрегатных и лямбда-функциях.
Пример:
``` sql
@ -122,7 +122,7 @@ FROM t_null
Существует два варианта IN-ов с подзапросами (аналогично для JOIN-ов): обычный `IN` / `JOIN` и `GLOBAL IN` / `GLOBAL JOIN`. Они отличаются способом выполнения при распределённой обработке запроса.
:::note "Attention"
:::note "Внимание"
Помните, что алгоритмы, описанные ниже, могут работать иначе в зависимости от [настройки](../../operations/settings/settings.md) `distributed_product_mode`.
:::
При использовании обычного IN-а, запрос отправляется на удалённые серверы, и на каждом из них выполняются подзапросы в секциях `IN` / `JOIN`.
@ -228,7 +228,7 @@ SELECT CounterID, count() FROM distributed_table_1 WHERE UserID IN (SELECT UserI
SETTINGS max_parallel_replicas=3
```
преобразуются на каждом сервере в
преобразуется на каждом сервере в
```sql
SELECT CounterID, count() FROM local_table_1 WHERE UserID IN (SELECT UserID FROM local_table_2 WHERE CounterID < 100)

1
docs/ru/sql-reference/operators/index.md
View File

@ -263,6 +263,7 @@ SELECT toDateTime('2014-10-26 00:00:00', 'Europe/Moscow') AS time, time + 60 * 6
│ 2014-10-26 00:00:00 │ 2014-10-26 23:00:00 │ 2014-10-27 00:00:00 │
└─────────────────────┴─────────────────────┴─────────────────────┘
```
:::
**Смотрите также**

2
docs/ru/sql-reference/statements/alter/view.md
View File

@ -6,7 +6,7 @@ sidebar_label: VIEW
# Выражение ALTER TABLE … MODIFY QUERY {#alter-modify-query}
Вы можеие изменить запрос `SELECT`, который был задан при создании [материализованного представления](../create/view.md#materialized), с помощью запроса 'ALTER TABLE … MODIFY QUERY'. Используйте его если при создании материализованного представления не использовалась секция `TO [db.]name`. Настройка `allow_experimental_alter_materialized_view_structure` должна быть включена.
Вы можете изменить запрос `SELECT`, который был задан при создании [материализованного представления](../create/view.md#materialized), с помощью запроса 'ALTER TABLE … MODIFY QUERY'. Используйте его если при создании материализованного представления не использовалась секция `TO [db.]name`. Настройка `allow_experimental_alter_materialized_view_structure` должна быть включена.
Если при создании материализованного представления использовалась конструкция `TO [db.]name`, то для изменения отсоедините представление с помощью [DETACH](../detach.md), измените таблицу с помощью [ALTER TABLE](index.md), а затем снова присоедините запрос с помощью [ATTACH](../attach.md).

1
docs/ru/sql-reference/statements/optimize.md
View File

@ -10,6 +10,7 @@ sidebar_label: OPTIMIZE
:::danger "Внимание"
`OPTIMIZE` не устраняет причину появления ошибки `Too many parts`.
:::
**Синтаксис**

2
docs/zh/guides/improving-query-performance/skipping-indexes.md
View File

@ -1,6 +1,6 @@
---
slug: /zh/guides/improving-query-performance/skipping-indexes
sidebar_label: Data Skipping Indexes
sidebar_label: 跳数索引
sidebar_position: 2
---

2
docs/zh/guides/improving-query-performance/sparse-primary-indexes.md
View File

@ -1,6 +1,6 @@
---
slug: /zh/guides/improving-query-performance/sparse-primary-indexes
sidebar_label: Sparse Primary Indexes
sidebar_label: 主键稀疏索引
sidebar_position: 20
---

3
packages/clickhouse-keeper.service
View File

@ -14,7 +14,8 @@ User=clickhouse
Group=clickhouse
Restart=always
RestartSec=30
RuntimeDirectory=%p # %p is resolved to the systemd unit name
# %p is resolved to the systemd unit name
RuntimeDirectory=%p
ExecStart=/usr/bin/clickhouse-keeper --config=/etc/clickhouse-keeper/keeper_config.xml --pid-file=%t/%p/%p.pid
# Minus means that this file is optional.
EnvironmentFile=-/etc/default/%p

4
programs/copier/ClusterCopier.cpp
View File

@ -1867,8 +1867,8 @@ std::set<String> ClusterCopier::getShardPartitions(const ConnectionTimeouts & ti
String query;
{
WriteBufferFromOwnString wb;
wb << "SELECT DISTINCT " << partition_name << " AS partition FROM"
<< " " << getQuotedTable(task_shard.table_read_shard) << " ORDER BY partition DESC";
wb << "SELECT " << partition_name << " AS partition FROM "
<< getQuotedTable(task_shard.table_read_shard) << " GROUP BY partition ORDER BY partition DESC";
query = wb.str();
}

2
programs/install/Install.cpp
View File

@ -20,7 +20,7 @@
#include <Common/formatReadable.h>
#include <Common/Config/ConfigProcessor.h>
#include <Common/OpenSSLHelpers.h>
#include <Common/hex.h>
#include <base/hex.h>
#include <Common/getResource.h>
#include <base/sleep.h>
#include <IO/ReadBufferFromFileDescriptor.h>

1
programs/keeper/CMakeLists.txt
View File

@ -128,6 +128,7 @@ if (BUILD_STANDALONE_KEEPER)
ch_contrib::lz4
ch_contrib::zstd
ch_contrib::cityhash
ch_contrib::jemalloc
common ch_contrib::double_conversion
ch_contrib::dragonbox_to_chars
pcg_random

6
programs/server/Server.cpp
View File

@ -41,6 +41,7 @@
#include <Common/TLDListsHolder.h>
#include <Common/Config/AbstractConfigurationComparison.h>
#include <Core/ServerUUID.h>
#include <IO/BackupsIOThreadPool.h>
#include <IO/ReadHelpers.h>
#include <IO/ReadBufferFromFile.h>
#include <IO/IOThreadPool.h>
@ -773,6 +774,11 @@ try
server_settings.max_io_thread_pool_free_size,
server_settings.io_thread_pool_queue_size);
BackupsIOThreadPool::initialize(
server_settings.max_backups_io_thread_pool_size,
server_settings.max_backups_io_thread_pool_free_size,
server_settings.backups_io_thread_pool_queue_size);
/// Initialize global local cache for remote filesystem.
if (config().has("local_cache_for_remote_fs"))
{

5
src/Access/EnabledRoles.h
View File

@ -44,10 +44,11 @@ private:
friend class RoleCache;
explicit EnabledRoles(const Params & params_);
void setRolesInfo(const std::shared_ptr<const EnabledRolesInfo> & info_, scope_guard * notifications);
const Params params;
/// Called by RoleCache to store `EnabledRolesInfo` in this `EnabledRoles` after the calculation is done.
void setRolesInfo(const std::shared_ptr<const EnabledRolesInfo> & info_, scope_guard * notifications);
std::shared_ptr<const EnabledRolesInfo> info;
mutable std::mutex info_mutex;

91
src/Access/RoleCache.cpp
View File

@ -57,7 +57,9 @@ namespace
RoleCache::RoleCache(const AccessControl & access_control_)
: access_control(access_control_), cache(600000 /* 10 minutes */) {}
: access_control(access_control_), cache(600000 /* 10 minutes */)
{
}
RoleCache::~RoleCache() = default;
@ -70,18 +72,18 @@ RoleCache::getEnabledRoles(const std::vector<UUID> & roles, const std::vector<UU
EnabledRoles::Params params;
params.current_roles.insert(roles.begin(), roles.end());
params.current_roles_with_admin_option.insert(roles_with_admin_option.begin(), roles_with_admin_option.end());
auto it = enabled_roles.find(params);
if (it != enabled_roles.end())
auto it = enabled_roles_by_params.find(params);
if (it != enabled_roles_by_params.end())
{
auto from_cache = it->second.lock();
if (from_cache)
return from_cache;
enabled_roles.erase(it);
if (auto enabled_roles = it->second.enabled_roles.lock())
return enabled_roles;
enabled_roles_by_params.erase(it);
}
auto res = std::shared_ptr<EnabledRoles>(new EnabledRoles(params));
collectEnabledRoles(*res, nullptr);
enabled_roles.emplace(std::move(params), res);
SubscriptionsOnRoles subscriptions_on_roles;
collectEnabledRoles(*res, subscriptions_on_roles, nullptr);
enabled_roles_by_params.emplace(std::move(params), EnabledRolesWithSubscriptions{res, std::move(subscriptions_on_roles)});
return res;
}
@ -90,21 +92,23 @@ void RoleCache::collectEnabledRoles(scope_guard * notifications)
{
/// `mutex` is already locked.
for (auto i = enabled_roles.begin(), e = enabled_roles.end(); i != e;)
for (auto i = enabled_roles_by_params.begin(), e = enabled_roles_by_params.end(); i != e;)
{
auto elem = i->second.lock();
if (!elem)
i = enabled_roles.erase(i);
auto & item = i->second;
if (auto enabled_roles = item.enabled_roles.lock())
{
collectEnabledRoles(*enabled_roles, item.subscriptions_on_roles, notifications);
++i;
}
else
{
collectEnabledRoles(*elem, notifications);
++i;
i = enabled_roles_by_params.erase(i);
}
}
}
void RoleCache::collectEnabledRoles(EnabledRoles & enabled, scope_guard * notifications)
void RoleCache::collectEnabledRoles(EnabledRoles & enabled_roles, SubscriptionsOnRoles & subscriptions_on_roles, scope_guard * notifications)
{
/// `mutex` is already locked.
@ -112,43 +116,57 @@ void RoleCache::collectEnabledRoles(EnabledRoles & enabled, scope_guard * notifi
auto new_info = std::make_shared<EnabledRolesInfo>();
boost::container::flat_set<UUID> skip_ids;
auto get_role_function = [this](const UUID & id) { return getRole(id); };
/// We need to collect and keep not only enabled roles but also subscriptions for them to be able to recalculate EnabledRolesInfo when some of the roles change.
SubscriptionsOnRoles new_subscriptions_on_roles;
new_subscriptions_on_roles.reserve(subscriptions_on_roles.size());
for (const auto & current_role : enabled.params.current_roles)
auto get_role_function = [this, &subscriptions_on_roles](const UUID & id) TSA_NO_THREAD_SAFETY_ANALYSIS { return getRole(id, subscriptions_on_roles); };
for (const auto & current_role : enabled_roles.params.current_roles)
collectRoles(*new_info, skip_ids, get_role_function, current_role, true, false);
for (const auto & current_role : enabled.params.current_roles_with_admin_option)
for (const auto & current_role : enabled_roles.params.current_roles_with_admin_option)
collectRoles(*new_info, skip_ids, get_role_function, current_role, true, true);
/// Remove duplicates from `subscriptions_on_roles`.
std::sort(new_subscriptions_on_roles.begin(), new_subscriptions_on_roles.end());
new_subscriptions_on_roles.erase(std::unique(new_subscriptions_on_roles.begin(), new_subscriptions_on_roles.end()), new_subscriptions_on_roles.end());
subscriptions_on_roles = std::move(new_subscriptions_on_roles);
/// Collect data from the collected roles.
enabled.setRolesInfo(new_info, notifications);
enabled_roles.setRolesInfo(new_info, notifications);
}
RolePtr RoleCache::getRole(const UUID & role_id)
RolePtr RoleCache::getRole(const UUID & role_id, SubscriptionsOnRoles & subscriptions_on_roles)
{
/// `mutex` is already locked.
auto role_from_cache = cache.get(role_id);
if (role_from_cache)
{
subscriptions_on_roles.emplace_back(role_from_cache->second);
return role_from_cache->first;
}
auto subscription = access_control.subscribeForChanges(role_id,
[this, role_id](const UUID &, const AccessEntityPtr & entity)
auto on_role_changed_or_removed = [this, role_id](const UUID &, const AccessEntityPtr & entity)
{
auto changed_role = entity ? typeid_cast<RolePtr>(entity) : nullptr;
if (changed_role)
roleChanged(role_id, changed_role);
else
roleRemoved(role_id);
});
};
auto subscription_on_role = std::make_shared<scope_guard>(access_control.subscribeForChanges(role_id, on_role_changed_or_removed));
auto role = access_control.tryRead<Role>(role_id);
if (role)
{
auto cache_value = Poco::SharedPtr<std::pair<RolePtr, scope_guard>>(
new std::pair<RolePtr, scope_guard>{role, std::move(subscription)});
auto cache_value = Poco::SharedPtr<std::pair<RolePtr, std::shared_ptr<scope_guard>>>(
new std::pair<RolePtr, std::shared_ptr<scope_guard>>{role, subscription_on_role});
cache.add(role_id, cache_value);
subscriptions_on_roles.emplace_back(subscription_on_role);
return role;
}
@ -162,12 +180,17 @@ void RoleCache::roleChanged(const UUID & role_id, const RolePtr & changed_role)
scope_guard notifications;
std::lock_guard lock{mutex};
auto role_from_cache = cache.get(role_id);
if (!role_from_cache)
return;
role_from_cache->first = changed_role;
cache.update(role_id, role_from_cache);
collectEnabledRoles(&notifications);
if (role_from_cache)
{
/// We update the role stored in a cache entry only if that entry has not expired yet.
role_from_cache->first = changed_role;
cache.update(role_id, role_from_cache);
}
/// An enabled role for some users has been changed, we need to recalculate the access rights.
collectEnabledRoles(&notifications); /// collectEnabledRoles() must be called with the `mutex` locked.
}
@ -177,8 +200,12 @@ void RoleCache::roleRemoved(const UUID & role_id)
scope_guard notifications;
std::lock_guard lock{mutex};
/// If a cache entry with the role has expired already, that remove() will do nothing.
cache.remove(role_id);
collectEnabledRoles(&notifications);
/// An enabled role for some users has been removed, we need to recalculate the access rights.
collectEnabledRoles(&notifications); /// collectEnabledRoles() must be called with the `mutex` locked.
}
}

24
src/Access/RoleCache.h
View File

@ -24,15 +24,29 @@ public:
const std::vector<UUID> & current_roles_with_admin_option);
private:
void collectEnabledRoles(scope_guard * notifications);
void collectEnabledRoles(EnabledRoles & enabled, scope_guard * notifications);
RolePtr getRole(const UUID & role_id);
using SubscriptionsOnRoles = std::vector<std::shared_ptr<scope_guard>>;
void collectEnabledRoles(scope_guard * notifications) TSA_REQUIRES(mutex);
void collectEnabledRoles(EnabledRoles & enabled_roles, SubscriptionsOnRoles & subscriptions_on_roles, scope_guard * notifications) TSA_REQUIRES(mutex);
RolePtr getRole(const UUID & role_id, SubscriptionsOnRoles & subscriptions_on_roles) TSA_REQUIRES(mutex);
void roleChanged(const UUID & role_id, const RolePtr & changed_role);
void roleRemoved(const UUID & role_id);
const AccessControl & access_control;
Poco::AccessExpireCache<UUID, std::pair<RolePtr, scope_guard>> cache;
std::map<EnabledRoles::Params, std::weak_ptr<EnabledRoles>> enabled_roles;
Poco::AccessExpireCache<UUID, std::pair<RolePtr, std::shared_ptr<scope_guard>>> TSA_GUARDED_BY(mutex) cache;
struct EnabledRolesWithSubscriptions
{
std::weak_ptr<EnabledRoles> enabled_roles;
/// We need to keep subscriptions for all enabled roles to be able to recalculate EnabledRolesInfo when some of the roles change.
/// `cache` also keeps subscriptions but that's not enough because values can be purged from the `cache` anytime.
SubscriptionsOnRoles subscriptions_on_roles;
};
std::map<EnabledRoles::Params, EnabledRolesWithSubscriptions> TSA_GUARDED_BY(mutex) enabled_roles_by_params;
mutable std::mutex mutex;
};

6
src/Analyzer/ColumnNode.h
View File

@ -3,6 +3,7 @@
#include <Core/NamesAndTypes.h>
#include <Analyzer/IQueryTreeNode.h>
#include <DataTypes/DataTypeNullable.h>
namespace DB
{
@ -117,6 +118,11 @@ public:
return column.type;
}
void convertToNullable() override
{
column.type = makeNullableSafe(column.type);
}
void dumpTreeImpl(WriteBuffer & buffer, FormatState & state, size_t indent) const override;
protected:

34
src/Analyzer/FunctionNode.cpp
View File

@ -2,18 +2,21 @@
#include <Common/SipHash.h>
#include <Common/FieldVisitorToString.h>
#include <DataTypes/IDataType.h>
#include <Analyzer/ConstantNode.h>
#include <IO/WriteBufferFromString.h>
#include <IO/Operators.h>
#include <DataTypes/IDataType.h>
#include <DataTypes/DataTypeSet.h>
#include <Parsers/ASTFunction.h>
#include <Functions/IFunction.h>
#include <AggregateFunctions/IAggregateFunction.h>
#include <Analyzer/Utils.h>
#include <Analyzer/ConstantNode.h>
#include <Analyzer/IdentifierNode.h>
namespace DB
@ -44,17 +47,29 @@ const DataTypes & FunctionNode::getArgumentTypes() const
ColumnsWithTypeAndName FunctionNode::getArgumentColumns() const
{
const auto & arguments = getArguments().getNodes();
size_t arguments_size = arguments.size();
ColumnsWithTypeAndName argument_columns;
argument_columns.reserve(arguments.size());
for (const auto & arg : arguments)
for (size_t i = 0; i < arguments_size; ++i)
{
ColumnWithTypeAndName argument;
argument.type = arg->getResultType();
if (auto * constant = arg->as<ConstantNode>())
argument.column = argument.type->createColumnConst(1, constant->getValue());
argument_columns.push_back(std::move(argument));
const auto & argument = arguments[i];
ColumnWithTypeAndName argument_column;
if (isNameOfInFunction(function_name) && i == 1)
argument_column.type = std::make_shared<DataTypeSet>();
else
argument_column.type = argument->getResultType();
auto * constant = argument->as<ConstantNode>();
if (constant && !isNotCreatable(argument_column.type))
argument_column.column = argument_column.type->createColumnConst(1, constant->getValue());
argument_columns.push_back(std::move(argument_column));
}
return argument_columns;
}
@ -99,7 +114,7 @@ void FunctionNode::dumpTreeImpl(WriteBuffer & buffer, FormatState & format_state
buffer << ", function_type: " << function_type;
if (function)
buffer << ", result_type: " + function->getResultType()->getName();
buffer << ", result_type: " + getResultType()->getName();
const auto & parameters = getParameters();
if (!parameters.getNodes().empty())
@ -177,6 +192,7 @@ QueryTreeNodePtr FunctionNode::cloneImpl() const
*/
result_function->function = function;
result_function->kind = kind;
result_function->wrap_with_nullable = wrap_with_nullable;
return result_function;
}

13
src/Analyzer/FunctionNode.h
View File

@ -8,6 +8,7 @@
#include <Common/typeid_cast.h>
#include <Core/ColumnsWithTypeAndName.h>
#include <Core/IResolvedFunction.h>
#include <DataTypes/DataTypeNullable.h>
#include <Functions/IFunction.h>
namespace DB
@ -187,7 +188,16 @@ public:
throw Exception(ErrorCodes::UNSUPPORTED_METHOD,
"Function node with name '{}' is not resolved",
function_name);
return function->getResultType();
auto type = function->getResultType();
if (wrap_with_nullable)
return makeNullableSafe(type);
return type;
}
void convertToNullable() override
{
chassert(kind == FunctionKind::ORDINARY);
wrap_with_nullable = true;
}
void dumpTreeImpl(WriteBuffer & buffer, FormatState & format_state, size_t indent) const override;
@ -205,6 +215,7 @@ private:
String function_name;
FunctionKind kind = FunctionKind::UNKNOWN;
IResolvedFunctionPtr function;
bool wrap_with_nullable = false;
static constexpr size_t parameters_child_index = 0;
static constexpr size_t arguments_child_index = 1;

5
src/Analyzer/IQueryTreeNode.h
View File

@ -90,6 +90,11 @@ public:
throw Exception(ErrorCodes::UNSUPPORTED_METHOD, "Method getResultType is not supported for {} query node", getNodeTypeName());
}
virtual void convertToNullable()
{
throw Exception(ErrorCodes::UNSUPPORTED_METHOD, "Method convertToNullable is not supported for {} query node", getNodeTypeName());
}
struct CompareOptions
{
bool compare_aliases = true;

17
src/Analyzer/InDepthQueryTreeVisitor.h
View File

@ -99,8 +99,9 @@ class InDepthQueryTreeVisitorWithContext
public:
using VisitQueryTreeNodeType = std::conditional_t<const_visitor, const QueryTreeNodePtr, QueryTreeNodePtr>;
explicit InDepthQueryTreeVisitorWithContext(ContextPtr context)
explicit InDepthQueryTreeVisitorWithContext(ContextPtr context, size_t initial_subquery_depth = 0)
: current_context(std::move(context))
, subquery_depth(initial_subquery_depth)
{}
/// Return true if visitor should traverse tree top to bottom, false otherwise
@ -125,11 +126,17 @@ public:
return current_context->getSettingsRef();
}
size_t getSubqueryDepth() const
{
return subquery_depth;
}
void visit(VisitQueryTreeNodeType & query_tree_node)
{
auto current_scope_context_ptr = current_context;
SCOPE_EXIT(
current_context = std::move(current_scope_context_ptr);
--subquery_depth;
);
if (auto * query_node = query_tree_node->template as<QueryNode>())
@ -137,6 +144,8 @@ public:
else if (auto * union_node = query_tree_node->template as<UnionNode>())
current_context = union_node->getContext();
++subquery_depth;
bool traverse_top_to_bottom = getDerived().shouldTraverseTopToBottom();
if (!traverse_top_to_bottom)
visitChildren(query_tree_node);
@ -145,7 +154,12 @@ public:
if (traverse_top_to_bottom)
visitChildren(query_tree_node);
getDerived().leaveImpl(query_tree_node);
}
void leaveImpl(VisitQueryTreeNodeType & node [[maybe_unused]])
{}
private:
Derived & getDerived()
{
@ -172,6 +186,7 @@ private:
}
ContextPtr current_context;
size_t subquery_depth = 0;
};
template <typename Derived>

6
src/Analyzer/JoinNode.h
View File

@ -106,6 +106,12 @@ public:
return locality;
}
/// Set join locality
void setLocality(JoinLocality locality_value)
{
locality = locality_value;
}
/// Get join strictness
JoinStrictness getStrictness() const
{

130
src/Analyzer/Passes/ArrayExistsToHasPass.cpp
View File

@ -1,3 +1,5 @@
#include <Analyzer/Passes/ArrayExistsToHasPass.h>
#include <Functions/FunctionFactory.h>
#include <Interpreters/Context.h>
@ -8,71 +10,85 @@
#include <Analyzer/InDepthQueryTreeVisitor.h>
#include <Analyzer/LambdaNode.h>
#include "ArrayExistsToHasPass.h"
namespace DB
{
namespace
{
class RewriteArrayExistsToHasVisitor : public InDepthQueryTreeVisitorWithContext<RewriteArrayExistsToHasVisitor>
class RewriteArrayExistsToHasVisitor : public InDepthQueryTreeVisitorWithContext<RewriteArrayExistsToHasVisitor>
{
public:
using Base = InDepthQueryTreeVisitorWithContext<RewriteArrayExistsToHasVisitor>;
using Base::Base;
void visitImpl(QueryTreeNodePtr & node)
{
public:
using Base = InDepthQueryTreeVisitorWithContext<RewriteArrayExistsToHasVisitor>;
using Base::Base;
if (!getSettings().optimize_rewrite_array_exists_to_has)
return;
void visitImpl(QueryTreeNodePtr & node)
auto * array_exists_function_node = node->as<FunctionNode>();
if (!array_exists_function_node || array_exists_function_node->getFunctionName() != "arrayExists")
return;
auto & array_exists_function_arguments_nodes = array_exists_function_node->getArguments().getNodes();
if (array_exists_function_arguments_nodes.size() != 2)
return;
/// lambda function must be like: x -> x = elem
auto * lambda_node = array_exists_function_arguments_nodes[0]->as<LambdaNode>();
if (!lambda_node)
return;
auto & lambda_arguments_nodes = lambda_node->getArguments().getNodes();
if (lambda_arguments_nodes.size() != 1)
return;
const auto & lambda_argument_column_node = lambda_arguments_nodes[0];
if (lambda_argument_column_node->getNodeType() != QueryTreeNodeType::COLUMN)
return;
auto * filter_node = lambda_node->getExpression()->as<FunctionNode>();
if (!filter_node || filter_node->getFunctionName() != "equals")
return;
const auto & filter_arguments_nodes = filter_node->getArguments().getNodes();
if (filter_arguments_nodes.size() != 2)
return;
const auto & filter_lhs_argument_node = filter_arguments_nodes[0];
auto filter_lhs_argument_node_type = filter_lhs_argument_node->getNodeType();
const auto & filter_rhs_argument_node = filter_arguments_nodes[1];
auto filter_rhs_argument_node_type = filter_rhs_argument_node->getNodeType();
QueryTreeNodePtr has_constant_element_argument;
if (filter_lhs_argument_node_type == QueryTreeNodeType::COLUMN &&
filter_rhs_argument_node_type == QueryTreeNodeType::CONSTANT &&
filter_lhs_argument_node->isEqual(*lambda_argument_column_node))
{
if (!getSettings().optimize_rewrite_array_exists_to_has)
return;
auto * function_node = node->as<FunctionNode>();
if (!function_node || function_node->getFunctionName() != "arrayExists")
return;
auto & function_arguments_nodes = function_node->getArguments().getNodes();
if (function_arguments_nodes.size() != 2)
return;
/// lambda function must be like: x -> x = elem
auto * lambda_node = function_arguments_nodes[0]->as<LambdaNode>();
if (!lambda_node)
return;
auto & lambda_arguments_nodes = lambda_node->getArguments().getNodes();
if (lambda_arguments_nodes.size() != 1)
return;
auto * column_node = lambda_arguments_nodes[0]->as<ColumnNode>();
auto * filter_node = lambda_node->getExpression()->as<FunctionNode>();
if (!filter_node || filter_node->getFunctionName() != "equals")
return;
auto filter_arguments_nodes = filter_node->getArguments().getNodes();
if (filter_arguments_nodes.size() != 2)
return;
ColumnNode * filter_column_node = nullptr;
if (filter_arguments_nodes[1]->as<ConstantNode>() && (filter_column_node = filter_arguments_nodes[0]->as<ColumnNode>())
&& filter_column_node->getColumnName() == column_node->getColumnName())
{
/// Rewrite arrayExists(x -> x = elem, arr) -> has(arr, elem)
function_arguments_nodes[0] = std::move(function_arguments_nodes[1]);
function_arguments_nodes[1] = std::move(filter_arguments_nodes[1]);
function_node->resolveAsFunction(
FunctionFactory::instance().get("has", getContext())->build(function_node->getArgumentColumns()));
}
else if (
filter_arguments_nodes[0]->as<ConstantNode>() && (filter_column_node = filter_arguments_nodes[1]->as<ColumnNode>())
&& filter_column_node->getColumnName() == column_node->getColumnName())
{
/// Rewrite arrayExists(x -> elem = x, arr) -> has(arr, elem)
function_arguments_nodes[0] = std::move(function_arguments_nodes[1]);
function_arguments_nodes[1] = std::move(filter_arguments_nodes[0]);
function_node->resolveAsFunction(
FunctionFactory::instance().get("has", getContext())->build(function_node->getArgumentColumns()));
}
/// Rewrite arrayExists(x -> x = elem, arr) -> has(arr, elem)
has_constant_element_argument = filter_rhs_argument_node;
}
};
else if (filter_lhs_argument_node_type == QueryTreeNodeType::CONSTANT &&
filter_rhs_argument_node_type == QueryTreeNodeType::COLUMN &&
filter_rhs_argument_node->isEqual(*lambda_argument_column_node))
{
/// Rewrite arrayExists(x -> elem = x, arr) -> has(arr, elem)
has_constant_element_argument = filter_lhs_argument_node;
}
else
{
return;
}
auto has_function = FunctionFactory::instance().get("has", getContext());
array_exists_function_arguments_nodes[0] = std::move(array_exists_function_arguments_nodes[1]);
array_exists_function_arguments_nodes[1] = std::move(has_constant_element_argument);
array_exists_function_node->resolveAsFunction(has_function->build(array_exists_function_node->getArgumentColumns()));
}
};
}

12
src/Analyzer/Passes/ArrayExistsToHasPass.h
View File

@ -4,8 +4,15 @@
namespace DB
{
/// Rewrite possible 'arrayExists(func, arr)' to 'has(arr, elem)' to improve performance
/// arrayExists(x -> x = 1, arr) -> has(arr, 1)
/** Rewrite possible 'arrayExists(func, arr)' to 'has(arr, elem)' to improve performance.
*
* Example: SELECT arrayExists(x -> x = 1, arr);
* Result: SELECT has(arr, 1);
*
* Example: SELECT arrayExists(x -> 1 = x, arr);
* Result: SELECT has(arr, 1);
*/
class RewriteArrayExistsToHasPass final : public IQueryTreePass
{
public:
@ -15,4 +22,5 @@ public:
void run(QueryTreeNodePtr query_tree_node, ContextPtr context) override;
};
}

5
src/Analyzer/Passes/AutoFinalOnQueryPass.cpp
View File

@ -22,8 +22,7 @@ public:
void visitImpl(QueryTreeNodePtr & node)
{
const auto & context = getContext();
if (!context->getSettingsRef().final)
if (!getSettings().final)
return;
const auto * query_node = node->as<QueryNode>();
@ -43,7 +42,7 @@ private:
return;
const auto & storage = table_node ? table_node->getStorage() : table_function_node->getStorage();
bool is_final_supported = storage && storage->supportsFinal() && !storage->isRemote();
bool is_final_supported = storage && storage->supportsFinal();
if (!is_final_supported)
return;

241
src/Analyzer/Passes/LogicalExpressionOptimizerPass.cpp Normal file
View File

@ -0,0 +1,241 @@
#include <Analyzer/Passes/LogicalExpressionOptimizerPass.h>
#include <Functions/FunctionFactory.h>
#include <Analyzer/InDepthQueryTreeVisitor.h>
#include <Analyzer/FunctionNode.h>
#include <Analyzer/ConstantNode.h>
#include <Analyzer/HashUtils.h>
namespace DB
{
class LogicalExpressionOptimizerVisitor : public InDepthQueryTreeVisitorWithContext<LogicalExpressionOptimizerVisitor>
{
public:
using Base = InDepthQueryTreeVisitorWithContext<LogicalExpressionOptimizerVisitor>;
explicit LogicalExpressionOptimizerVisitor(ContextPtr context)
: Base(std::move(context))
{}
void visitImpl(QueryTreeNodePtr & node)
{
auto * function_node = node->as<FunctionNode>();
if (!function_node)
return;
if (function_node->getFunctionName() == "or")
{
tryReplaceOrEqualsChainWithIn(node);
return;
}
if (function_node->getFunctionName() == "and")
{
tryReplaceAndEqualsChainsWithConstant(node);
return;
}
}
private:
void tryReplaceAndEqualsChainsWithConstant(QueryTreeNodePtr & node)
{
auto & function_node = node->as<FunctionNode &>();
assert(function_node.getFunctionName() == "and");
if (function_node.getResultType()->isNullable())
return;
QueryTreeNodes and_operands;
QueryTreeNodePtrWithHashMap<const ConstantNode *> node_to_constants;
for (const auto & argument : function_node.getArguments())
{
auto * argument_function = argument->as<FunctionNode>();
if (!argument_function || argument_function->getFunctionName() != "equals")
{
and_operands.push_back(argument);
continue;
}
const auto & equals_arguments = argument_function->getArguments().getNodes();
const auto & lhs = equals_arguments[0];
const auto & rhs = equals_arguments[1];
const auto has_and_with_different_constant = [&](const QueryTreeNodePtr & expression, const ConstantNode * constant)
{
if (auto it = node_to_constants.find(expression); it != node_to_constants.end())
{
if (!it->second->isEqual(*constant))
return true;
}
else
{
node_to_constants.emplace(expression, constant);
and_operands.push_back(argument);
}
return false;
};
bool collapse_to_false = false;
if (const auto * lhs_literal = lhs->as<ConstantNode>())
collapse_to_false = has_and_with_different_constant(rhs, lhs_literal);
else if (const auto * rhs_literal = rhs->as<ConstantNode>())
collapse_to_false = has_and_with_different_constant(lhs, rhs_literal);
else
and_operands.push_back(argument);
if (collapse_to_false)
{
auto false_value = std::make_shared<ConstantValue>(0u, function_node.getResultType());
auto false_node = std::make_shared<ConstantNode>(std::move(false_value));
node = std::move(false_node);
return;
}
}
if (and_operands.size() == function_node.getArguments().getNodes().size())
return;
if (and_operands.size() == 1)
{
/// AND operator can have UInt8 or bool as its type.
/// bool is used if a bool constant is at least one operand.
/// Because we reduce the number of operands here by eliminating the same equality checks,
/// the only situation we can end up here is we had AND check where all the equality checks are the same so we know the type is UInt8.
/// Otherwise, we will have > 1 operands and we don't have to do anything.
assert(!function_node.getResultType()->isNullable() && and_operands[0]->getResultType()->equals(*function_node.getResultType()));
node = std::move(and_operands[0]);
return;
}
auto and_function_resolver = FunctionFactory::instance().get("and", getContext());
function_node.getArguments().getNodes() = std::move(and_operands);
function_node.resolveAsFunction(and_function_resolver);
}
void tryReplaceOrEqualsChainWithIn(QueryTreeNodePtr & node)
{
auto & function_node = node->as<FunctionNode &>();
assert(function_node.getFunctionName() == "or");
QueryTreeNodes or_operands;
QueryTreeNodePtrWithHashMap<QueryTreeNodes> node_to_equals_functions;
QueryTreeNodePtrWithHashMap<QueryTreeNodeConstRawPtrWithHashSet> node_to_constants;
for (const auto & argument : function_node.getArguments())
{
auto * argument_function = argument->as<FunctionNode>();
if (!argument_function || argument_function->getFunctionName() != "equals")
{
or_operands.push_back(argument);
continue;
}
/// collect all equality checks (x = value)
const auto & equals_arguments = argument_function->getArguments().getNodes();
const auto & lhs = equals_arguments[0];
const auto & rhs = equals_arguments[1];
const auto add_equals_function_if_not_present = [&](const auto & expression_node, const ConstantNode * constant)
{
auto & constant_set = node_to_constants[expression_node];
if (!constant_set.contains(constant))
{
constant_set.insert(constant);
node_to_equals_functions[expression_node].push_back(argument);
}
};
if (const auto * lhs_literal = lhs->as<ConstantNode>())
add_equals_function_if_not_present(rhs, lhs_literal);
else if (const auto * rhs_literal = rhs->as<ConstantNode>())
add_equals_function_if_not_present(lhs, rhs_literal);
else
or_operands.push_back(argument);
}
auto in_function_resolver = FunctionFactory::instance().get("in", getContext());
for (auto & [expression, equals_functions] : node_to_equals_functions)
{
const auto & settings = getSettings();
if (equals_functions.size() < settings.optimize_min_equality_disjunction_chain_length && !expression.node->getResultType()->lowCardinality())
{
std::move(equals_functions.begin(), equals_functions.end(), std::back_inserter(or_operands));
continue;
}
Tuple args;
args.reserve(equals_functions.size());
/// first we create tuple from RHS of equals functions
for (const auto & equals : equals_functions)
{
const auto * equals_function = equals->as<FunctionNode>();
assert(equals_function && equals_function->getFunctionName() == "equals");
const auto & equals_arguments = equals_function->getArguments().getNodes();
if (const auto * rhs_literal = equals_arguments[1]->as<ConstantNode>())
{
args.push_back(rhs_literal->getValue());
}
else
{
const auto * lhs_literal = equals_arguments[0]->as<ConstantNode>();
assert(lhs_literal);
args.push_back(lhs_literal->getValue());
}
}
auto rhs_node = std::make_shared<ConstantNode>(std::move(args));
auto in_function = std::make_shared<FunctionNode>("in");
QueryTreeNodes in_arguments;
in_arguments.reserve(2);
in_arguments.push_back(expression.node);
in_arguments.push_back(std::move(rhs_node));
in_function->getArguments().getNodes() = std::move(in_arguments);
in_function->resolveAsFunction(in_function_resolver);
or_operands.push_back(std::move(in_function));
}
if (or_operands.size() == function_node.getArguments().getNodes().size())
return;
if (or_operands.size() == 1)
{
/// if the result type of operand is the same as the result type of OR
/// we can replace OR with the operand
if (or_operands[0]->getResultType()->equals(*function_node.getResultType()))
{
assert(!function_node.getResultType()->isNullable());
node = std::move(or_operands[0]);
return;
}
/// otherwise add a stub 0 to make OR correct
or_operands.push_back(std::make_shared<ConstantNode>(static_cast<UInt8>(0)));
}
auto or_function_resolver = FunctionFactory::instance().get("or", getContext());
function_node.getArguments().getNodes() = std::move(or_operands);
function_node.resolveAsFunction(or_function_resolver);
}
};
void LogicalExpressionOptimizerPass::run(QueryTreeNodePtr query_tree_node, ContextPtr context)
{
LogicalExpressionOptimizerVisitor visitor(std::move(context));
visitor.visit(query_tree_node);
}
}

82
src/Analyzer/Passes/LogicalExpressionOptimizerPass.h Normal file
View File

@ -0,0 +1,82 @@
#pragma once
#include <Analyzer/IQueryTreePass.h>
namespace DB
{
/**
* This pass tries to do optimizations on logical expression:
*
* 1. Replaces chains of equality functions inside an OR with a single IN operator.
* The replacement is done if:
* - one of the operands of the equality function is a constant
* - length of chain is at least 'optimize_min_equality_disjunction_chain_length' long OR the expression has type of LowCardinality
*
* E.g. (optimize_min_equality_disjunction_chain_length = 2)
* -------------------------------
* SELECT *
* FROM table
* WHERE a = 1 OR b = 'test' OR a = 2;
*
* will be transformed into
*
* SELECT *
* FROM TABLE
* WHERE b = 'test' OR a IN (1, 2);
* -------------------------------
*
* 2. Removes duplicate OR checks
* -------------------------------
* SELECT *
* FROM table
* WHERE a = 1 OR b = 'test' OR a = 1;
*
* will be transformed into
*
* SELECT *
* FROM TABLE
* WHERE a = 1 OR b = 'test';
* -------------------------------
*
* 3. Replaces AND chains with a single constant.
* The replacement is done if:
* - one of the operands of the equality function is a constant
* - constants are different for same expression
* -------------------------------
* SELECT *
* FROM table
* WHERE a = 1 AND b = 'test' AND a = 2;
*
* will be transformed into
*
* SELECT *
* FROM TABLE
* WHERE 0;
* -------------------------------
*
* 4. Removes duplicate AND checks
* -------------------------------
* SELECT *
* FROM table
* WHERE a = 1 AND b = 'test' AND a = 1;
*
* will be transformed into
*
* SELECT *
* FROM TABLE
* WHERE a = 1 AND b = 'test';
* -------------------------------
*/
class LogicalExpressionOptimizerPass final : public IQueryTreePass
{
public:
String getName() override { return "LogicalExpressionOptimizer"; }
String getDescription() override { return "Transform equality chain to a single IN function or a constant if possible"; }
void run(QueryTreeNodePtr query_tree_node, ContextPtr context) override;
};
}

4
src/Analyzer/Passes/OptimizeGroupByFunctionKeysPass.cpp
View File

@ -69,8 +69,7 @@ private:
for (auto it = function_arguments.rbegin(); it != function_arguments.rend(); ++it)
candidates.push_back({ *it, is_deterministic });
// Using DFS we traverse function tree and try to find if it uses other keys as function arguments.
// TODO: Also process CONSTANT here. We can simplify GROUP BY x, x + 1 to GROUP BY x.
/// Using DFS we traverse function tree and try to find if it uses other keys as function arguments.
while (!candidates.empty())
{
auto [candidate, parents_are_only_deterministic] = candidates.back();
@ -108,6 +107,7 @@ private:
return false;
}
}
return true;
}

230
src/Analyzer/Passes/QueryAnalysisPass.cpp
View File

@ -193,14 +193,9 @@ namespace ErrorCodes
* lookup should not be continued, and exception must be thrown because if lookup continues identifier can be resolved from parent scope.
*
* TODO: Update exception messages
* TODO: JOIN TREE subquery constant columns
* TODO: Table identifiers with optional UUID.
* TODO: Lookup functions arrayReduce(sum, [1, 2, 3]);
* TODO: SELECT (compound_expression).*, (compound_expression).COLUMNS are not supported on parser level.
* TODO: SELECT a.b.c.*, a.b.c.COLUMNS. Qualified matcher where identifier size is greater than 2 are not supported on parser level.
* TODO: Support function identifier resolve from parent query scope, if lambda in parent scope does not capture any columns.
* TODO: Support group_by_use_nulls.
* TODO: Scalar subqueries cache.
*/
namespace
@ -472,6 +467,12 @@ public:
alias_name_to_expressions[node_alias].push_back(node);
}
if (const auto * function = node->as<FunctionNode>())
{
if (AggregateFunctionFactory::instance().isAggregateFunctionName(function->getFunctionName()))
++aggregate_functions_counter;
}
expressions.emplace_back(node);
}
@ -490,6 +491,12 @@ public:
alias_name_to_expressions.erase(it);
}
if (const auto * function = top_expression->as<FunctionNode>())
{
if (AggregateFunctionFactory::instance().isAggregateFunctionName(function->getFunctionName()))
--aggregate_functions_counter;
}
expressions.pop_back();
}
@ -508,6 +515,11 @@ public:
return alias_name_to_expressions.contains(alias);
}
bool hasAggregateFunction() const
{
return aggregate_functions_counter > 0;
}
QueryTreeNodePtr getExpressionWithAlias(const std::string & alias) const
{
auto expression_it = alias_name_to_expressions.find(alias);
@ -554,6 +566,7 @@ public:
private:
QueryTreeNodes expressions;
size_t aggregate_functions_counter = 0;
std::unordered_map<std::string, QueryTreeNodes> alias_name_to_expressions;
};
@ -684,9 +697,15 @@ struct IdentifierResolveScope
}
if (auto * union_node = scope_node->as<UnionNode>())
{
context = union_node->getContext();
}
else if (auto * query_node = scope_node->as<QueryNode>())
{
context = query_node->getContext();
group_by_use_nulls = context->getSettingsRef().group_by_use_nulls &&
(query_node->isGroupByWithGroupingSets() || query_node->isGroupByWithRollup() || query_node->isGroupByWithCube());
}
}
QueryTreeNodePtr scope_node;
@ -734,9 +753,14 @@ struct IdentifierResolveScope
/// Table expression node to data
std::unordered_map<QueryTreeNodePtr, TableExpressionData> table_expression_node_to_data;
QueryTreeNodePtrWithHashSet nullable_group_by_keys;
/// Use identifier lookup to result cache
bool use_identifier_lookup_to_result_cache = true;
/// Apply nullability to aggregation keys
bool group_by_use_nulls = false;
/// JOINs count
size_t joins_count = 0;
@ -1310,6 +1334,9 @@ private:
/// Global resolve expression node to projection names map
std::unordered_map<QueryTreeNodePtr, ProjectionNames> resolved_expressions;
/// Global resolve expression node to tree size
std::unordered_map<QueryTreeNodePtr, size_t> node_to_tree_size;
/// Global scalar subquery to scalar value map
std::unordered_map<QueryTreeNodePtrWithHash, Block> scalar_subquery_to_scalar_value;
@ -1838,7 +1865,10 @@ void QueryAnalyzer::evaluateScalarSubqueryIfNeeded(QueryTreeNodePtr & node, Iden
Block scalar_block;
QueryTreeNodePtrWithHash node_with_hash(node);
auto node_without_alias = node->clone();
node_without_alias->removeAlias();
QueryTreeNodePtrWithHash node_with_hash(node_without_alias);
auto scalar_value_it = scalar_subquery_to_scalar_value.find(node_with_hash);
if (scalar_value_it != scalar_subquery_to_scalar_value.end())
@ -1928,21 +1958,7 @@ void QueryAnalyzer::evaluateScalarSubqueryIfNeeded(QueryTreeNodePtr & node, Iden
*
* Example: SELECT (SELECT 2 AS x, x)
*/
NameSet block_column_names;
size_t unique_column_name_counter = 1;
for (auto & column_with_type : block)
{
if (!block_column_names.contains(column_with_type.name))
{
block_column_names.insert(column_with_type.name);
continue;
}
column_with_type.name += '_';
column_with_type.name += std::to_string(unique_column_name_counter);
++unique_column_name_counter;
}
makeUniqueColumnNamesInBlock(block);
scalar_block.insert({
ColumnTuple::create(block.getColumns()),
@ -2322,7 +2338,13 @@ QueryTreeNodePtr QueryAnalyzer::tryResolveTableIdentifierFromDatabaseCatalog(con
storage_id = context->resolveStorageID(storage_id);
bool is_temporary_table = storage_id.getDatabaseName() == DatabaseCatalog::TEMPORARY_DATABASE;
auto storage = DatabaseCatalog::instance().tryGetTable(storage_id, context);
StoragePtr storage;
if (is_temporary_table)
storage = DatabaseCatalog::instance().getTable(storage_id, context);
else
storage = DatabaseCatalog::instance().tryGetTable(storage_id, context);
if (!storage)
return {};
@ -2888,7 +2910,10 @@ QueryTreeNodePtr QueryAnalyzer::tryResolveIdentifierFromTableExpression(const Id
break;
IdentifierLookup column_identifier_lookup = {qualified_identifier_with_removed_part, IdentifierLookupContext::EXPRESSION};
if (tryBindIdentifierToAliases(column_identifier_lookup, scope) ||
if (tryBindIdentifierToAliases(column_identifier_lookup, scope))
break;
if (table_expression_data.should_qualify_columns &&
tryBindIdentifierToTableExpressions(column_identifier_lookup, table_expression_node, scope))
break;
@ -2992,11 +3017,39 @@ QueryTreeNodePtr QueryAnalyzer::tryResolveIdentifierFromJoin(const IdentifierLoo
resolved_identifier = std::move(result_column_node);
}
else if (scope.joins_count == 1 && scope.context->getSettingsRef().single_join_prefer_left_table)
else if (left_resolved_identifier->isEqual(*right_resolved_identifier, IQueryTreeNode::CompareOptions{.compare_aliases = false}))
{
const auto & identifier_path_part = identifier_lookup.identifier.front();
auto * left_resolved_identifier_column = left_resolved_identifier->as<ColumnNode>();
auto * right_resolved_identifier_column = right_resolved_identifier->as<ColumnNode>();
if (left_resolved_identifier_column && right_resolved_identifier_column)
{
const auto & left_column_source_alias = left_resolved_identifier_column->getColumnSource()->getAlias();
const auto & right_column_source_alias = right_resolved_identifier_column->getColumnSource()->getAlias();
/** If column from right table was resolved using alias, we prefer column from right table.
*
* Example: SELECT dummy FROM system.one JOIN system.one AS A ON A.dummy = system.one.dummy;
*
* If alias is specified for left table, and alias is not specified for right table and identifier was resolved
* without using left table alias, we prefer column from right table.
*
* Example: SELECT dummy FROM system.one AS A JOIN system.one ON A.dummy = system.one.dummy;
*
* Otherwise we prefer column from left table.
*/
if (identifier_path_part == right_column_source_alias)
return right_resolved_identifier;
else if (!left_column_source_alias.empty() &&
right_column_source_alias.empty() &&
identifier_path_part != left_column_source_alias)
return right_resolved_identifier;
}
return left_resolved_identifier;
}
else if (left_resolved_identifier->isEqual(*right_resolved_identifier, IQueryTreeNode::CompareOptions{.compare_aliases = false}))
else if (scope.joins_count == 1 && scope.context->getSettingsRef().single_join_prefer_left_table)
{
return left_resolved_identifier;
}
@ -4440,6 +4493,7 @@ ProjectionNames QueryAnalyzer::resolveFunction(QueryTreeNodePtr & node, Identifi
bool is_special_function_dict_get = false;
bool is_special_function_join_get = false;
bool is_special_function_exists = false;
bool is_special_function_if = false;
if (!lambda_expression_untyped)
{
@ -4447,6 +4501,7 @@ ProjectionNames QueryAnalyzer::resolveFunction(QueryTreeNodePtr & node, Identifi
is_special_function_dict_get = functionIsDictGet(function_name);
is_special_function_join_get = functionIsJoinGet(function_name);
is_special_function_exists = function_name == "exists";
is_special_function_if = function_name == "if";
auto function_name_lowercase = Poco::toLower(function_name);
@ -4545,6 +4600,60 @@ ProjectionNames QueryAnalyzer::resolveFunction(QueryTreeNodePtr & node, Identifi
is_special_function_in = true;
}
if (is_special_function_if && !function_node_ptr->getArguments().getNodes().empty())
{
/** Handle special case with constant If function, even if some of the arguments are invalid.
*
* SELECT if(hasColumnInTable('system', 'numbers', 'not_existing_column'), not_existing_column, 5) FROM system.numbers;
*/
auto & if_function_arguments = function_node_ptr->getArguments().getNodes();
auto if_function_condition = if_function_arguments[0];
resolveExpressionNode(if_function_condition, scope, false /*allow_lambda_expression*/, false /*allow_table_expression*/);
auto constant_condition = tryExtractConstantFromConditionNode(if_function_condition);
if (constant_condition.has_value() && if_function_arguments.size() == 3)
{
QueryTreeNodePtr constant_if_result_node;
QueryTreeNodePtr possibly_invalid_argument_node;
if (*constant_condition)
{
possibly_invalid_argument_node = if_function_arguments[2];
constant_if_result_node = if_function_arguments[1];
}
else
{
possibly_invalid_argument_node = if_function_arguments[1];
constant_if_result_node = if_function_arguments[2];
}
bool apply_constant_if_optimization = false;
try
{
resolveExpressionNode(possibly_invalid_argument_node,
scope,
false /*allow_lambda_expression*/,
false /*allow_table_expression*/);
}
catch (...)
{
apply_constant_if_optimization = true;
}
if (apply_constant_if_optimization)
{
auto result_projection_names = resolveExpressionNode(constant_if_result_node,
scope,
false /*allow_lambda_expression*/,
false /*allow_table_expression*/);
node = std::move(constant_if_result_node);
return result_projection_names;
}
}
}
/// Resolve function arguments
bool allow_table_expressions = is_special_function_in;
@ -5033,7 +5142,7 @@ ProjectionNames QueryAnalyzer::resolveFunction(QueryTreeNodePtr & node, Identifi
/// Do not constant fold get scalar functions
bool disable_constant_folding = function_name == "__getScalar" || function_name == "shardNum" ||
function_name == "shardCount";
function_name == "shardCount" || function_name == "hostName";
/** If function is suitable for constant folding try to convert it to constant.
* Example: SELECT plus(1, 1);
@ -5059,7 +5168,8 @@ ProjectionNames QueryAnalyzer::resolveFunction(QueryTreeNodePtr & node, Identifi
/** Do not perform constant folding if there are aggregate or arrayJoin functions inside function.
* Example: SELECT toTypeName(sum(number)) FROM numbers(10);
*/
if (column && isColumnConst(*column) && (!hasAggregateFunctionNodes(node) && !hasFunctionNode(node, "arrayJoin")))
if (column && isColumnConst(*column) && !typeid_cast<const ColumnConst *>(column.get())->getDataColumn().isDummy() &&
(!hasAggregateFunctionNodes(node) && !hasFunctionNode(node, "arrayJoin")))
{
/// Replace function node with result constant node
Field column_constant_value;
@ -5407,10 +5517,18 @@ ProjectionNames QueryAnalyzer::resolveExpressionNode(QueryTreeNodePtr & node, Id
}
}
validateTreeSize(node, scope.context->getSettingsRef().max_expanded_ast_elements, node_to_tree_size);
if (scope.nullable_group_by_keys.contains(node) && !scope.expressions_in_resolve_process_stack.hasAggregateFunction())
{
node = node->clone();
node->convertToNullable();
}
/** Update aliases after expression node was resolved.
* Do not update node in alias table if we resolve it for duplicate alias.
*/
if (!node_alias.empty() && use_alias_table)
if (!node_alias.empty() && use_alias_table && !scope.group_by_use_nulls)
{
auto it = scope.alias_name_to_expression_node.find(node_alias);
if (it != scope.alias_name_to_expression_node.end())
@ -6418,9 +6536,6 @@ void QueryAnalyzer::resolveQuery(const QueryTreeNodePtr & query_node, Identifier
auto & query_node_typed = query_node->as<QueryNode &>();
const auto & settings = scope.context->getSettingsRef();
if (settings.group_by_use_nulls)
throw Exception(ErrorCodes::UNSUPPORTED_METHOD, "GROUP BY use nulls is not supported");
bool is_rollup_or_cube = query_node_typed.isGroupByWithRollup() || query_node_typed.isGroupByWithCube();
if (query_node_typed.isGroupByWithGroupingSets() && query_node_typed.isGroupByWithTotals())
@ -6556,15 +6671,21 @@ void QueryAnalyzer::resolveQuery(const QueryTreeNodePtr & query_node, Identifier
resolveQueryJoinTreeNode(query_node_typed.getJoinTree(), scope, visitor);
}
scope.use_identifier_lookup_to_result_cache = true;
if (!scope.group_by_use_nulls)
scope.use_identifier_lookup_to_result_cache = true;
/// Resolve query node sections.
auto projection_columns = resolveProjectionExpressionNodeList(query_node_typed.getProjectionNode(), scope);
if (query_node_typed.getProjection().getNodes().empty())
throw Exception(ErrorCodes::EMPTY_LIST_OF_COLUMNS_QUERIED,
"Empty list of columns in projection. In scope {}",
scope.scope_node->formatASTForErrorMessage());
NamesAndTypes projection_columns;
if (!scope.group_by_use_nulls)
{
projection_columns = resolveProjectionExpressionNodeList(query_node_typed.getProjectionNode(), scope);
if (query_node_typed.getProjection().getNodes().empty())
throw Exception(ErrorCodes::EMPTY_LIST_OF_COLUMNS_QUERIED,
"Empty list of columns in projection. In scope {}",
scope.scope_node->formatASTForErrorMessage());
}
if (query_node_typed.hasWith())
resolveExpressionNodeList(query_node_typed.getWithNode(), scope, true /*allow_lambda_expression*/, false /*allow_table_expression*/);
@ -6586,6 +6707,15 @@ void QueryAnalyzer::resolveQuery(const QueryTreeNodePtr & query_node, Identifier
resolveExpressionNodeList(grouping_sets_keys_list_node, scope, false /*allow_lambda_expression*/, false /*allow_table_expression*/);
}
if (scope.group_by_use_nulls)
{
for (const auto & grouping_set : query_node_typed.getGroupBy().getNodes())
{
for (const auto & group_by_elem : grouping_set->as<ListNode>()->getNodes())
scope.nullable_group_by_keys.insert(group_by_elem);
}
}
}
else
{
@ -6593,6 +6723,12 @@ void QueryAnalyzer::resolveQuery(const QueryTreeNodePtr & query_node, Identifier
replaceNodesWithPositionalArguments(query_node_typed.getGroupByNode(), query_node_typed.getProjection().getNodes(), scope);
resolveExpressionNodeList(query_node_typed.getGroupByNode(), scope, false /*allow_lambda_expression*/, false /*allow_table_expression*/);
if (scope.group_by_use_nulls)
{
for (const auto & group_by_elem : query_node_typed.getGroupBy().getNodes())
scope.nullable_group_by_keys.insert(group_by_elem);
}
}
}
@ -6645,6 +6781,15 @@ void QueryAnalyzer::resolveQuery(const QueryTreeNodePtr & query_node, Identifier
convertLimitOffsetExpression(query_node_typed.getOffset(), "OFFSET", scope);
}
if (scope.group_by_use_nulls)
{
projection_columns = resolveProjectionExpressionNodeList(query_node_typed.getProjectionNode(), scope);
if (query_node_typed.getProjection().getNodes().empty())
throw Exception(ErrorCodes::EMPTY_LIST_OF_COLUMNS_QUERIED,
"Empty list of columns in projection. In scope {}",
scope.scope_node->formatASTForErrorMessage());
}
/** Resolve nodes with duplicate aliases.
* Table expressions cannot have duplicate aliases.
*
@ -6708,7 +6853,16 @@ void QueryAnalyzer::resolveQuery(const QueryTreeNodePtr & query_node, Identifier
"ARRAY JOIN",
"in PREWHERE");
validateAggregates(query_node);
validateAggregates(query_node, { .group_by_use_nulls = scope.group_by_use_nulls });
for (const auto & column : projection_columns)
{
if (isNotCreatable(column.type))
throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT,
"Invalid projection column with type {}. In scope {}",
column.type->getName(),
scope.scope_node->formatASTForErrorMessage());
}
/** WITH section can be safely removed, because WITH section only can provide aliases to query expressions
* and CTE for other sections to use.

3
src/Analyzer/Passes/ShardNumColumnToFunctionPass.h
View File

@ -6,6 +6,9 @@ namespace DB
{
/** Rewrite _shard_num column into shardNum() function.
*
* Example: SELECT _shard_num FROM distributed_table;
* Result: SELECT shardNum() FROM distributed_table;
*/
class ShardNumColumnToFunctionPass final : public IQueryTreePass
{

70
src/Analyzer/QueryTreeBuilder.cpp
View File

@ -355,21 +355,67 @@ QueryTreeNodePtr QueryTreeBuilder::buildSelectExpression(const ASTPtr & select_q
if (select_limit_by)
current_query_tree->getLimitByNode() = buildExpressionList(select_limit_by, current_context);
/// Combine limit expression with limit setting
/// Combine limit expression with limit and offset settings into final limit expression
/// The sequence of application is the following - offset expression, limit expression, offset setting, limit setting.
/// Since offset setting is applied after limit expression, but we want to transfer settings into expression
/// we must decrease limit expression by offset setting and then add offset setting to offset expression.
/// select_limit - limit expression
/// limit - limit setting
/// offset - offset setting
///
/// if select_limit
/// -- if offset >= select_limit (expr 0)
/// then (0) (0 rows)
/// -- else if limit > 0 (expr 1)
/// then min(select_limit - offset, limit) (expr 2)
/// -- else
/// then (select_limit - offset) (expr 3)
/// else if limit > 0
/// then limit
///
/// offset = offset + of_expr
auto select_limit = select_query_typed.limitLength();
if (select_limit && limit)
if (select_limit)
{
auto function_node = std::make_shared<FunctionNode>("least");
function_node->getArguments().getNodes().push_back(buildExpression(select_limit, current_context));
function_node->getArguments().getNodes().push_back(std::make_shared<ConstantNode>(limit));
current_query_tree->getLimit() = std::move(function_node);
}
else if (limit)
current_query_tree->getLimit() = std::make_shared<ConstantNode>(limit);
else if (select_limit)
current_query_tree->getLimit() = buildExpression(select_limit, current_context);
/// Shortcut
if (offset == 0 && limit == 0)
{
current_query_tree->getLimit() = buildExpression(select_limit, current_context);
}
else
{
/// expr 3
auto expr_3 = std::make_shared<FunctionNode>("minus");
expr_3->getArguments().getNodes().push_back(buildExpression(select_limit, current_context));
expr_3->getArguments().getNodes().push_back(std::make_shared<ConstantNode>(offset));
/// Combine offset expression with offset setting
/// expr 2
auto expr_2 = std::make_shared<FunctionNode>("least");
expr_2->getArguments().getNodes().push_back(expr_3->clone());
expr_2->getArguments().getNodes().push_back(std::make_shared<ConstantNode>(limit));
/// expr 0
auto expr_0 = std::make_shared<FunctionNode>("greaterOrEquals");
expr_0->getArguments().getNodes().push_back(std::make_shared<ConstantNode>(offset));
expr_0->getArguments().getNodes().push_back(buildExpression(select_limit, current_context));
/// expr 1
auto expr_1 = std::make_shared<ConstantNode>(limit > 0);
auto function_node = std::make_shared<FunctionNode>("multiIf");
function_node->getArguments().getNodes().push_back(expr_0);
function_node->getArguments().getNodes().push_back(std::make_shared<ConstantNode>(0));
function_node->getArguments().getNodes().push_back(expr_1);
function_node->getArguments().getNodes().push_back(expr_2);
function_node->getArguments().getNodes().push_back(expr_3);
current_query_tree->getLimit() = std::move(function_node);
}
}
else if (limit > 0)
current_query_tree->getLimit() = std::make_shared<ConstantNode>(limit);
/// Combine offset expression with offset setting into final offset expression
auto select_offset = select_query_typed.limitOffset();
if (select_offset && offset)
{

Some files were not shown because too many files have changed in this diff Show More