mirror of
https://github.com/ClickHouse/ClickHouse.git
synced 2024-11-21 23:21:59 +00:00
Merge branch 'master' into check-table-structure-completely-setting
This commit is contained in:
Vladimir Cherkasov
GitHub
commit
c0fa7bca61
2
.github/ISSUE_TEMPLATE/20_feature-request.md
vendored
2
.github/ISSUE_TEMPLATE/20_feature-request.md
vendored
@ -15,7 +15,7 @@ assignees: ''
|
||||
|
||||
**Use case**
|
||||
|
||||
> A clear and concise description of what is the intended usage scenario is.
|
||||
> A clear and concise description of what the intended usage scenario is.
|
||||
|
||||
**Describe the solution you'd like**
|
||||
|
||||
|
||||
@ -11,6 +11,10 @@ option (ARCH_NATIVE "Add -march=native compiler flag. This makes your binaries n
|
||||
if (ARCH_NATIVE)
|
||||
set (COMPILER_FLAGS "${COMPILER_FLAGS} -march=native")
|
||||
|
||||
# Populate the ENABLE_ option flags. This is required for the build of some third-party dependencies, specifically snappy, which
|
||||
# (somewhat weirdly) expects the relative SNAPPY_HAVE_ preprocessor variables to be populated, in addition to the microarchitecture
|
||||
# feature flags being enabled in the compiler. This fixes the ARCH_NATIVE flag by automatically populating the ENABLE_ option flags
|
||||
# according to the current CPU's capabilities, detected using clang.
|
||||
if (ARCH_AMD64)
|
||||
execute_process(
|
||||
COMMAND sh -c "clang -E - -march=native -###"
|
||||
|
||||
@ -1,4 +1,21 @@
|
||||
set (CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} -D_LIBCPP_DEBUG=0") # More checks in debug build.
|
||||
if (CMAKE_BUILD_TYPE_UC STREQUAL "DEBUG")
|
||||
# Enable libcxx debug mode: https://releases.llvm.org/15.0.0/projects/libcxx/docs/DesignDocs/DebugMode.html
|
||||
# The docs say the debug mode violates complexity guarantees, so do this only for Debug builds.
|
||||
# set (CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} -D_LIBCPP_ENABLE_DEBUG_MODE=1")
|
||||
# ^^ Crashes the database upon startup, needs investigation.
|
||||
# Besides that, the implementation looks like a poor man's MSAN specific to libcxx. Since CI tests MSAN
|
||||
# anyways, we can keep the debug mode disabled.
|
||||
|
||||
# Libcxx also provides extra assertions:
|
||||
# --> https://releases.llvm.org/15.0.0/projects/libcxx/docs/UsingLibcxx.html#assertions-mode
|
||||
# These look orthogonal to the debug mode but the debug mode enables them implicitly:
|
||||
# --> https://github.com/llvm/llvm-project/blob/release/15.x/libcxx/include/__assert#L29
|
||||
# They are cheap and straightforward, so enable them in debug builds:
|
||||
set (CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} -D_LIBCPP_ENABLE_ASSERTIONS=1")
|
||||
|
||||
# TODO Once we upgrade to LLVM 18+, reconsider all of the above as they introduced "hardening modes":
|
||||
# https://libcxx.llvm.org/Hardening.html
|
||||
endif ()
|
||||
|
||||
add_subdirectory(contrib/libcxxabi-cmake)
|
||||
add_subdirectory(contrib/libcxx-cmake)
|
||||
|
||||
@ -1,6 +1,9 @@
|
||||
set(ABSL_ROOT_DIR "${ClickHouse_SOURCE_DIR}/contrib/abseil-cpp")
|
||||
set(ABSL_COMMON_INCLUDE_DIRS "${ABSL_ROOT_DIR}")
|
||||
|
||||
# To avoid errors "'X' does not refer to a value" while using `offsetof` function.
|
||||
set(CMAKE_CXX_STANDARD 17)
|
||||
|
||||
# This is a minimized version of the function definition in CMake/AbseilHelpers.cmake
|
||||
|
||||
#
|
||||
|
||||
@ -5,6 +5,9 @@ if(NOT ENABLE_PROTOBUF)
|
||||
return()
|
||||
endif()
|
||||
|
||||
# To avoid errors "'X' does not refer to a value" while using `offsetof` function.
|
||||
set(CMAKE_CXX_STANDARD 17)
|
||||
|
||||
set(Protobuf_INCLUDE_DIR "${ClickHouse_SOURCE_DIR}/contrib/google-protobuf/src")
|
||||
if(OS_FREEBSD AND SANITIZE STREQUAL "address")
|
||||
# ../contrib/protobuf/src/google/protobuf/arena_impl.h:45:10: fatal error: 'sanitizer/asan_interface.h' file not found
|
||||
|
||||
@ -6,6 +6,8 @@ if(NOT ENABLE_GRPC)
|
||||
return()
|
||||
endif()
|
||||
|
||||
set(CMAKE_CXX_STANDARD 17)
|
||||
|
||||
set(_gRPC_SOURCE_DIR "${ClickHouse_SOURCE_DIR}/contrib/grpc")
|
||||
set(_gRPC_BINARY_DIR "${ClickHouse_BINARY_DIR}/contrib/grpc")
|
||||
|
||||
|
||||
@ -22,7 +22,7 @@
|
||||
# limitations under the License.
|
||||
|
||||
# We want to use C++23, but GRPC is not ready
|
||||
set (CMAKE_CXX_STANDARD 20)
|
||||
set (CMAKE_CXX_STANDARD 17)
|
||||
|
||||
set(_gRPC_ZLIB_INCLUDE_DIR "")
|
||||
set(_gRPC_ZLIB_LIBRARIES ch_contrib::zlib)
|
||||
|
||||
@ -9,7 +9,7 @@ RUN CGO_ENABLED=0 go install github.com/wjdp/htmltest@v${HTMLTEST_VERSION} \
|
||||
# nodejs 17 prefers ipv6 and is broken in our environment
|
||||
FROM node:16-alpine
|
||||
|
||||
RUN apk add --no-cache git openssh bash
|
||||
RUN apk add --no-cache git openssh bash curl
|
||||
|
||||
# At this point we want to really update /opt/clickhouse-docs directory
|
||||
# So we reset the cache
|
||||
@ -33,4 +33,7 @@ RUN mkdir /output_path \
|
||||
COPY run.sh /run.sh
|
||||
COPY --from=htmltest-builder /usr/bin/htmltest /usr/bin/htmltest
|
||||
|
||||
# Install ClickHouse Local, which is used to auto-generate some doc pages.
|
||||
RUN curl https://clickhouse.com/ | sh
|
||||
|
||||
ENTRYPOINT ["/run.sh"]
|
||||
|
||||
@ -21,6 +21,78 @@ do
|
||||
fi
|
||||
done
|
||||
|
||||
# Generate pages with settings
|
||||
|
||||
./clickhouse -q "
|
||||
WITH
|
||||
|
||||
'/ClickHouse/src/Core/Settings.cpp' AS cpp_file,
|
||||
|
||||
settings_from_cpp AS
|
||||
(
|
||||
SELECT extract(line, 'M\\(\\w+, (\\w+),') AS name
|
||||
FROM file(cpp_file, LineAsString)
|
||||
WHERE match(line, '^\\s*M\\(')
|
||||
),
|
||||
|
||||
main_content AS
|
||||
(
|
||||
SELECT format('## {} {}\\n\\nType: {}\\n\\nDefault value: {}\\n\\n{}\\n\\n', name, '{#'||name||'}', type, default, trim(BOTH '\\n' FROM description))
|
||||
FROM system.settings WHERE name IN settings_from_cpp
|
||||
ORDER BY name
|
||||
),
|
||||
|
||||
'---
|
||||
sidebar_label: Core Settings
|
||||
sidebar_position: 2
|
||||
slug: /en/operations/settings/settings
|
||||
toc_max_heading_level: 2
|
||||
---
|
||||
|
||||
# Core Settings
|
||||
|
||||
All below settings are also available in table [system.settings](/docs/en/operations/system-tables/settings).
|
||||
|
||||
' AS prefix
|
||||
|
||||
SELECT prefix || (SELECT groupConcat(*) FROM main_content)
|
||||
INTO OUTFILE '/opt/clickhouse-docs/docs/en/operations/settings/settings.md' TRUNCATE FORMAT LineAsString
|
||||
"
|
||||
|
||||
./clickhouse -q "
|
||||
WITH
|
||||
|
||||
'/ClickHouse/src/Core/FormatFactorySettingsDeclaration.h' AS cpp_file,
|
||||
|
||||
settings_from_cpp AS
|
||||
(
|
||||
SELECT extract(line, 'M\\(\\w+, (\\w+),') AS name
|
||||
FROM file(cpp_file, LineAsString)
|
||||
WHERE match(line, '^\\s*M\\(')
|
||||
),
|
||||
|
||||
main_content AS
|
||||
(
|
||||
SELECT format('## {} {}\\n\\nType: {}\\n\\nDefault value: {}\\n\\n{}\\n\\n', name, '{#'||name||'}', type, default, trim(BOTH '\\n' FROM description))
|
||||
FROM system.settings WHERE name IN settings_from_cpp
|
||||
ORDER BY name
|
||||
),
|
||||
|
||||
'---
|
||||
sidebar_label: Format Settings
|
||||
sidebar_position: 52
|
||||
slug: /en/operations/settings/formats
|
||||
toc_max_heading_level: 2
|
||||
---
|
||||
|
||||
# Format settings {#format-settings}
|
||||
|
||||
' AS prefix
|
||||
|
||||
SELECT prefix || (SELECT groupConcat(*) FROM main_content)
|
||||
INTO OUTFILE '/opt/clickhouse-docs/docs/en/operations/settings/settings-formats.md' TRUNCATE FORMAT LineAsString
|
||||
"
|
||||
|
||||
# Force build error on wrong symlinks
|
||||
sed -i '/onBrokenMarkdownLinks:/ s/ignore/error/g' docusaurus.config.js
|
||||
|
||||
|
||||
5
docker/test/integration/runner/misc/openldap/initialized.sh
Executable file
5
docker/test/integration/runner/misc/openldap/initialized.sh
Executable file
@ -0,0 +1,5 @@
|
||||
#!/bin/bash
|
||||
set -e
|
||||
|
||||
# workaround for https://github.com/bitnami/containers/issues/73310
|
||||
touch /tmp/.openldap-initialized
|
||||
@ -196,7 +196,6 @@ When writing docs, you can use prepared templates. Copy the code of a template a
|
||||
Templates:
|
||||
|
||||
- [Function](_description_templates/template-function.md)
|
||||
- [Setting](_description_templates/template-setting.md)
|
||||
- [Server Setting](_description_templates/template-server-setting.md)
|
||||
- [Database or Table engine](_description_templates/template-engine.md)
|
||||
- [System table](_description_templates/template-system-table.md)
|
||||
|
||||
@ -1,27 +0,0 @@
|
||||
## setting_name {#setting_name}
|
||||
|
||||
Description.
|
||||
|
||||
For the switch setting, use the typical phrase: “Enables or disables something ...”.
|
||||
|
||||
Possible values:
|
||||
|
||||
*For switcher setting:*
|
||||
|
||||
- 0 — Disabled.
|
||||
- 1 — Enabled.
|
||||
|
||||
*For another setting (typical phrases):*
|
||||
|
||||
- Positive integer.
|
||||
- 0 — Disabled or unlimited or something else.
|
||||
|
||||
Default value: `value`.
|
||||
|
||||
**Additional Info** (Optional)
|
||||
|
||||
The name of an additional section can be any, for example, **Usage**.
|
||||
|
||||
**See Also** (Optional)
|
||||
|
||||
- [link](#)
|
||||
@ -1,11 +0,0 @@
|
||||
sudo apt-get install -y apt-transport-https ca-certificates dirmngr
|
||||
sudo apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv 8919F6BD2B48D754
|
||||
|
||||
echo "deb https://packages.clickhouse.com/deb stable main" | sudo tee \
|
||||
/etc/apt/sources.list.d/clickhouse.list
|
||||
sudo apt-get update
|
||||
|
||||
sudo apt-get install -y clickhouse-server clickhouse-client
|
||||
|
||||
sudo service clickhouse-server start
|
||||
clickhouse-client # or "clickhouse-client --password" if you've set up a password.
|
||||
@ -1,6 +0,0 @@
|
||||
sudo yum install -y yum-utils
|
||||
sudo yum-config-manager --add-repo https://packages.clickhouse.com/rpm/clickhouse.repo
|
||||
sudo yum install -y clickhouse-server clickhouse-client
|
||||
|
||||
sudo /etc/init.d/clickhouse-server start
|
||||
clickhouse-client # or "clickhouse-client --password" if you set up a password.
|
||||
@ -1,32 +0,0 @@
|
||||
LATEST_VERSION=$(curl -s https://packages.clickhouse.com/tgz/stable/ | \
|
||||
grep -Eo '[0-9]+\.[0-9]+\.[0-9]+\.[0-9]+' | sort -V -r | head -n 1)
|
||||
export LATEST_VERSION
|
||||
|
||||
case $(uname -m) in
|
||||
x86_64) ARCH=amd64 ;;
|
||||
aarch64) ARCH=arm64 ;;
|
||||
*) echo "Unknown architecture $(uname -m)"; exit 1 ;;
|
||||
esac
|
||||
|
||||
for PKG in clickhouse-common-static clickhouse-common-static-dbg clickhouse-server clickhouse-client
|
||||
do
|
||||
curl -fO "https://packages.clickhouse.com/tgz/stable/$PKG-$LATEST_VERSION-${ARCH}.tgz" \
|
||||
|| curl -fO "https://packages.clickhouse.com/tgz/stable/$PKG-$LATEST_VERSION.tgz"
|
||||
done
|
||||
|
||||
tar -xzvf "clickhouse-common-static-$LATEST_VERSION-${ARCH}.tgz" \
|
||||
|| tar -xzvf "clickhouse-common-static-$LATEST_VERSION.tgz"
|
||||
sudo "clickhouse-common-static-$LATEST_VERSION/install/doinst.sh"
|
||||
|
||||
tar -xzvf "clickhouse-common-static-dbg-$LATEST_VERSION-${ARCH}.tgz" \
|
||||
|| tar -xzvf "clickhouse-common-static-dbg-$LATEST_VERSION.tgz"
|
||||
sudo "clickhouse-common-static-dbg-$LATEST_VERSION/install/doinst.sh"
|
||||
|
||||
tar -xzvf "clickhouse-server-$LATEST_VERSION-${ARCH}.tgz" \
|
||||
|| tar -xzvf "clickhouse-server-$LATEST_VERSION.tgz"
|
||||
sudo "clickhouse-server-$LATEST_VERSION/install/doinst.sh" configure
|
||||
sudo /etc/init.d/clickhouse-server start
|
||||
|
||||
tar -xzvf "clickhouse-client-$LATEST_VERSION-${ARCH}.tgz" \
|
||||
|| tar -xzvf "clickhouse-client-$LATEST_VERSION.tgz"
|
||||
sudo "clickhouse-client-$LATEST_VERSION/install/doinst.sh"
|
||||
@ -14,7 +14,12 @@ then
|
||||
HAS_SSE42=$(grep sse4_2 /proc/cpuinfo)
|
||||
if [ "${HAS_SSE42}" ]
|
||||
then
|
||||
DIR="amd64"
|
||||
if ldd --version 2>&1 | grep -q musl
|
||||
then
|
||||
DIR="amd64musl"
|
||||
else
|
||||
DIR="amd64"
|
||||
fi
|
||||
else
|
||||
DIR="amd64compat"
|
||||
fi
|
||||
|
||||
@ -63,7 +63,34 @@ Currently there are 3 ways to authenticate:
|
||||
- `SAS Token` - Can be used by providing an `endpoint`, `connection_string` or `storage_account_url`. It is identified by presence of '?' in the url.
|
||||
- `Workload Identity` - Can be used by providing an `endpoint` or `storage_account_url`. If `use_workload_identity` parameter is set in config, ([workload identity](https://github.com/Azure/azure-sdk-for-cpp/tree/main/sdk/identity/azure-identity#authenticate-azure-hosted-applications)) is used for authentication.
|
||||
|
||||
### Data cache {#data-cache}
|
||||
|
||||
`Azure` table engine supports data caching on local disk.
|
||||
See filesystem cache configuration options and usage in this [section](/docs/en/operations/storing-data.md/#using-local-cache).
|
||||
Caching is made depending on the path and ETag of the storage object, so clickhouse will not read a stale cache version.
|
||||
|
||||
To enable caching use a setting `filesystem_cache_name = '<name>'` and `enable_filesystem_cache = 1`.
|
||||
|
||||
```sql
|
||||
SELECT *
|
||||
FROM azureBlobStorage('DefaultEndpointsProtocol=http;AccountName=devstoreaccount1;AccountKey=Eby8vdM02xNOcqFlqUwJPLlmEtlCDXJ1OUzFT50uSRZ6IFsuFq2UVErCz4I6tq/K1SZFPTOtr/KBHBeksoGMGw==;BlobEndpoint=http://azurite1:10000/devstoreaccount1/;', 'test_container', 'test_table', 'CSV')
|
||||
SETTINGS filesystem_cache_name = 'cache_for_azure', enable_filesystem_cache = 1;
|
||||
```
|
||||
|
||||
1. add the following section to clickhouse configuration file:
|
||||
|
||||
``` xml
|
||||
<clickhouse>
|
||||
<filesystem_caches>
|
||||
<cache_for_azure>
|
||||
<path>path to cache directory</path>
|
||||
<max_size>10Gi</max_size>
|
||||
</cache_for_azure>
|
||||
</filesystem_caches>
|
||||
</clickhouse>
|
||||
```
|
||||
|
||||
2. reuse cache configuration (and therefore cache storage) from clickhouse `storage_configuration` section, [described here](/docs/en/operations/storing-data.md/#using-local-cache)
|
||||
|
||||
## See also
|
||||
|
||||
|
||||
@ -48,6 +48,10 @@ Using named collections:
|
||||
CREATE TABLE deltalake ENGINE=DeltaLake(deltalake_conf, filename = 'test_table')
|
||||
```
|
||||
|
||||
### Data cache {#data-cache}
|
||||
|
||||
`Iceberg` table engine and table function support data caching same as `S3`, `AzureBlobStorage`, `HDFS` storages. See [here](../../../engines/table-engines/integrations/s3.md#data-cache).
|
||||
|
||||
## See also
|
||||
|
||||
- [deltaLake table function](../../../sql-reference/table-functions/deltalake.md)
|
||||
|
||||
@ -63,6 +63,10 @@ CREATE TABLE iceberg_table ENGINE=IcebergS3(iceberg_conf, filename = 'test_table
|
||||
|
||||
Table engine `Iceberg` is an alias to `IcebergS3` now.
|
||||
|
||||
### Data cache {#data-cache}
|
||||
|
||||
`Iceberg` table engine and table function support data caching same as `S3`, `AzureBlobStorage`, `HDFS` storages. See [here](../../../engines/table-engines/integrations/s3.md#data-cache).
|
||||
|
||||
## See also
|
||||
|
||||
- [iceberg table function](/docs/en/sql-reference/table-functions/iceberg.md)
|
||||
|
||||
@ -9,7 +9,7 @@ sidebar_label: MongoDB
|
||||
MongoDB engine is read-only table engine which allows to read data from remote [MongoDB](https://www.mongodb.com/) collection.
|
||||
|
||||
Only MongoDB v3.6+ servers are supported.
|
||||
[Seed list(`mongodb**+srv**`)](https://www.mongodb.com/docs/manual/reference/glossary/#std-term-seed-list) is not yet supported.
|
||||
[Seed list(`mongodb+srv`)](https://www.mongodb.com/docs/manual/reference/glossary/#std-term-seed-list) is not yet supported.
|
||||
|
||||
:::note
|
||||
If you're facing troubles, please report the issue, and try to use [the legacy implementation](../../../operations/server-configuration-parameters/settings.md#use_legacy_mongodb_integration).
|
||||
|
||||
@ -4,12 +4,8 @@ sidebar_position: 138
|
||||
sidebar_label: MySQL
|
||||
---
|
||||
|
||||
import CloudAvailableBadge from '@theme/badges/CloudAvailableBadge';
|
||||
|
||||
# MySQL Table Engine
|
||||
|
||||
<CloudAvailableBadge />
|
||||
|
||||
The MySQL engine allows you to perform `SELECT` and `INSERT` queries on data that is stored on a remote MySQL server.
|
||||
|
||||
## Creating a Table {#creating-a-table}
|
||||
|
||||
@ -26,6 +26,7 @@ SELECT * FROM s3_engine_table LIMIT 2;
|
||||
│ two │ 2 │
|
||||
└──────┴───────┘
|
||||
```
|
||||
|
||||
## Create Table {#creating-a-table}
|
||||
|
||||
``` sql
|
||||
@ -43,6 +44,37 @@ CREATE TABLE s3_engine_table (name String, value UInt32)
|
||||
- `aws_access_key_id`, `aws_secret_access_key` - Long-term credentials for the [AWS](https://aws.amazon.com/) account user. You can use these to authenticate your requests. Parameter is optional. If credentials are not specified, they are used from the configuration file. For more information see [Using S3 for Data Storage](../mergetree-family/mergetree.md#table_engine-mergetree-s3).
|
||||
- `compression` — Compression type. Supported values: `none`, `gzip/gz`, `brotli/br`, `xz/LZMA`, `zstd/zst`. Parameter is optional. By default, it will auto-detect compression by file extension.
|
||||
|
||||
### Data cache {#data-cache}
|
||||
|
||||
`S3` table engine supports data caching on local disk.
|
||||
See filesystem cache configuration options and usage in this [section](/docs/en/operations/storing-data.md/#using-local-cache).
|
||||
Caching is made depending on the path and ETag of the storage object, so clickhouse will not read a stale cache version.
|
||||
|
||||
To enable caching use a setting `filesystem_cache_name = '<name>'` and `enable_filesystem_cache = 1`.
|
||||
|
||||
```sql
|
||||
SELECT *
|
||||
FROM s3('http://minio:10000/clickhouse//test_3.csv', 'minioadmin', 'minioadminpassword', 'CSV')
|
||||
SETTINGS filesystem_cache_name = 'cache_for_s3', enable_filesystem_cache = 1;
|
||||
```
|
||||
|
||||
There are two ways to define cache in configuration file.
|
||||
|
||||
1. add the following section to clickhouse configuration file:
|
||||
|
||||
``` xml
|
||||
<clickhouse>
|
||||
<filesystem_caches>
|
||||
<cache_for_s3>
|
||||
<path>path to cache directory</path>
|
||||
<max_size>10Gi</max_size>
|
||||
</cache_for_s3>
|
||||
</filesystem_caches>
|
||||
</clickhouse>
|
||||
```
|
||||
|
||||
2. reuse cache configuration (and therefore cache storage) from clickhouse `storage_configuration` section, [described here](/docs/en/operations/storing-data.md/#using-local-cache)
|
||||
|
||||
### PARTITION BY
|
||||
|
||||
`PARTITION BY` — Optional. In most cases you don't need a partition key, and if it is needed you generally don't need a partition key more granular than by month. Partitioning does not speed up queries (in contrast to the ORDER BY expression). You should never use too granular partitioning. Don't partition your data by client identifiers or names (instead, make client identifier or name the first column in the ORDER BY expression).
|
||||
|
||||
@ -31,6 +31,10 @@ The table must be enabled in the server configuration, see the `opentelemetry_sp
|
||||
|
||||
The tags or attributes are saved as two parallel arrays, containing the keys and values. Use [ARRAY JOIN](../sql-reference/statements/select/array-join.md) to work with them.
|
||||
|
||||
## Log-query-settings
|
||||
|
||||
ClickHouse allows you to log changes to query settings during query execution. When enabled, any modifications made to query settings will be recorded in the OpenTelemetry span log. This feature is particularly useful in production environments for tracking configuration changes that may affect query performance.
|
||||
|
||||
## Integration with monitoring systems
|
||||
|
||||
At the moment, there is no ready tool that can export the tracing data from ClickHouse to a monitoring system.
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
File diff suppressed because it is too large
Load Diff
@ -177,6 +177,26 @@ When you are ready to insert your files into ClickHouse, startup a ClickHouse se
|
||||
:::
|
||||
|
||||
|
||||
## Format Conversions
|
||||
|
||||
You can use `clickhouse-local` for converting data between different formats. Example:
|
||||
|
||||
``` bash
|
||||
$ clickhouse-local --input-format JSONLines --output-format CSV --query "SELECT * FROM table" < data.json > data.csv
|
||||
```
|
||||
|
||||
Formats are auto-detected from file extensions:
|
||||
|
||||
``` bash
|
||||
$ clickhouse-local --query "SELECT * FROM table" < data.json > data.csv
|
||||
```
|
||||
|
||||
As a shortcut, you can write it using the `--copy` argument:
|
||||
``` bash
|
||||
$ clickhouse-local --copy < data.json > data.csv
|
||||
```
|
||||
|
||||
|
||||
## Usage {#usage}
|
||||
|
||||
By default `clickhouse-local` has access to data of a ClickHouse server on the same host, and it does not depend on the server's configuration. It also supports loading server configuration using `--config-file` argument. For temporary data, a unique temporary data directory is created by default.
|
||||
|
||||
@ -261,9 +261,10 @@ windowFunnel(window, [mode, [mode, ... ]])(timestamp, cond1, cond2, ..., condN)
|
||||
|
||||
- `window` — Length of the sliding window, it is the time interval between the first and the last condition. The unit of `window` depends on the `timestamp` itself and varies. Determined using the expression `timestamp of cond1 <= timestamp of cond2 <= ... <= timestamp of condN <= timestamp of cond1 + window`.
|
||||
- `mode` — It is an optional argument. One or more modes can be set.
|
||||
- `'strict_deduplication'` — If the same condition holds for the sequence of events, then such repeating event interrupts further processing.
|
||||
- `'strict_deduplication'` — If the same condition holds for the sequence of events, then such repeating event interrupts further processing. Note: it may work unexpectedly if several conditions hold for the same event.
|
||||
- `'strict_order'` — Don't allow interventions of other events. E.g. in the case of `A->B->D->C`, it stops finding `A->B->C` at the `D` and the max event level is 2.
|
||||
- `'strict_increase'` — Apply conditions only to events with strictly increasing timestamps.
|
||||
- `'strict_once'` — Count each event only once in the chain even if it meets the condition several times
|
||||
|
||||
**Returned value**
|
||||
|
||||
@ -490,7 +491,7 @@ Where:
|
||||
|
||||
## uniqUpTo(N)(x)
|
||||
|
||||
Calculates the number of different values of the argument up to a specified limit, `N`. If the number of different argument values is greater than `N`, this function returns `N` + 1, otherwise it calculates the exact value.
|
||||
Calculates the number of different values of the argument up to a specified limit, `N`. If the number of different argument values is greater than `N`, this function returns `N` + 1, otherwise it calculates the exact value.
|
||||
|
||||
Recommended for use with small `N`s, up to 10. The maximum value of `N` is 100.
|
||||
|
||||
@ -522,7 +523,7 @@ This function behaves the same as [sumMap](../../sql-reference/aggregate-functio
|
||||
- `keys`: [Array](../data-types/array.md) of keys.
|
||||
- `values`: [Array](../data-types/array.md) of values.
|
||||
|
||||
**Returned Value**
|
||||
**Returned Value**
|
||||
|
||||
- Returns a tuple of two arrays: keys in sorted order, and values summed for the corresponding keys.
|
||||
|
||||
@ -539,10 +540,10 @@ CREATE TABLE sum_map
|
||||
)
|
||||
ENGINE = Log
|
||||
|
||||
INSERT INTO sum_map VALUES
|
||||
('2000-01-01', '2000-01-01 00:00:00', [1, 2, 3], [10, 10, 10]),
|
||||
INSERT INTO sum_map VALUES
|
||||
('2000-01-01', '2000-01-01 00:00:00', [1, 2, 3], [10, 10, 10]),
|
||||
('2000-01-01', '2000-01-01 00:00:00', [3, 4, 5], [10, 10, 10]),
|
||||
('2000-01-01', '2000-01-01 00:01:00', [4, 5, 6], [10, 10, 10]),
|
||||
('2000-01-01', '2000-01-01 00:01:00', [4, 5, 6], [10, 10, 10]),
|
||||
('2000-01-01', '2000-01-01 00:01:00', [6, 7, 8], [10, 10, 10]);
|
||||
```
|
||||
|
||||
@ -572,7 +573,7 @@ This function behaves the same as [sumMap](../../sql-reference/aggregate-functio
|
||||
- `keys`: [Array](../data-types/array.md) of keys.
|
||||
- `values`: [Array](../data-types/array.md) of values.
|
||||
|
||||
**Returned Value**
|
||||
**Returned Value**
|
||||
|
||||
- Returns a tuple of two arrays: keys in sorted order, and values summed for the corresponding keys.
|
||||
|
||||
@ -591,10 +592,10 @@ CREATE TABLE sum_map
|
||||
)
|
||||
ENGINE = Log
|
||||
|
||||
INSERT INTO sum_map VALUES
|
||||
('2000-01-01', '2000-01-01 00:00:00', [1, 2, 3], [10, 10, 10]),
|
||||
INSERT INTO sum_map VALUES
|
||||
('2000-01-01', '2000-01-01 00:00:00', [1, 2, 3], [10, 10, 10]),
|
||||
('2000-01-01', '2000-01-01 00:00:00', [3, 4, 5], [10, 10, 10]),
|
||||
('2000-01-01', '2000-01-01 00:01:00', [4, 5, 6], [10, 10, 10]),
|
||||
('2000-01-01', '2000-01-01 00:01:00', [4, 5, 6], [10, 10, 10]),
|
||||
('2000-01-01', '2000-01-01 00:01:00', [6, 7, 8], [10, 10, 10]);
|
||||
```
|
||||
|
||||
|
||||
@ -1,190 +0,0 @@
|
||||
---
|
||||
slug: /en/sql-reference/ansi
|
||||
sidebar_position: 40
|
||||
sidebar_label: ANSI Compatibility
|
||||
title: "ANSI SQL Compatibility of ClickHouse SQL Dialect"
|
||||
---
|
||||
|
||||
:::note
|
||||
This article relies on Table 38, “Feature taxonomy and definition for mandatory features”, Annex F of [ISO/IEC CD 9075-2:2011](https://www.iso.org/obp/ui/#iso:std:iso-iec:9075:-2:ed-4:v1:en:sec:8).
|
||||
:::
|
||||
|
||||
## Differences in Behaviour
|
||||
|
||||
The following table lists cases when query feature works in ClickHouse, but behaves not as specified in ANSI SQL.
|
||||
|
||||
| Feature ID | Feature Name | Difference |
|
||||
|------------|-----------------------------|-----------------------------------------------------------------------------------------------------------|
|
||||
| E011 | Numeric data types | Numeric literal with period is interpreted as approximate (`Float64`) instead of exact (`Decimal`) |
|
||||
| E051-05 | Select items can be renamed | Item renames have a wider visibility scope than just the SELECT result |
|
||||
| E141-01 | NOT NULL constraints | `NOT NULL` is implied for table columns by default |
|
||||
| E011-04 | Arithmetic operators | ClickHouse overflows instead of checked arithmetic and changes the result data type based on custom rules |
|
||||
|
||||
## Feature Status
|
||||
|
||||
| Feature ID | Feature Name | Status | Comment |
|
||||
|------------|--------------------------------------------------------------------------------------------------------------------------|----------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
|
||||
| **E011** | **Numeric data types** | <span class="text-warning">Partial</span> | |
|
||||
| E011-01 | INTEGER and SMALLINT data types | <span class="text-success">Yes</span> | |
|
||||
| E011-02 | REAL, DOUBLE PRECISION and FLOAT data types data types | <span class="text-success">Yes</span> | |
|
||||
| E011-03 | DECIMAL and NUMERIC data types | <span class="text-success">Yes</span> | |
|
||||
| E011-04 | Arithmetic operators | <span class="text-success">Yes</span> | |
|
||||
| E011-05 | Numeric comparison | <span class="text-success">Yes</span> | |
|
||||
| E011-06 | Implicit casting among the numeric data types | <span class="text-danger">No</span> | ANSI SQL allows arbitrary implicit cast between numeric types, while ClickHouse relies on functions having multiple overloads instead of implicit cast |
|
||||
| **E021** | **Character string types** | <span class="text-warning">Partial</span> | |
|
||||
| E021-01 | CHARACTER data type | <span class="text-success">Yes</span> | |
|
||||
| E021-02 | CHARACTER VARYING data type | <span class="text-success">Yes</span> | |
|
||||
| E021-03 | Character literals | <span class="text-success">Yes</span> | |
|
||||
| E021-04 | CHARACTER_LENGTH function | <span class="text-warning">Partial</span> | No `USING` clause |
|
||||
| E021-05 | OCTET_LENGTH function | <span class="text-danger">No</span> | `LENGTH` behaves similarly |
|
||||
| E021-06 | SUBSTRING | <span class="text-warning">Partial</span> | No support for `SIMILAR` and `ESCAPE` clauses, no `SUBSTRING_REGEX` variant |
|
||||
| E021-07 | Character concatenation | <span class="text-warning">Partial</span> | No `COLLATE` clause |
|
||||
| E021-08 | UPPER and LOWER functions | <span class="text-success">Yes</span> | |
|
||||
| E021-09 | TRIM function | <span class="text-success">Yes</span> | |
|
||||
| E021-10 | Implicit casting among the fixed-length and variable-length character string types | <span class="text-warning">Partial</span> | ANSI SQL allows arbitrary implicit cast between string types, while ClickHouse relies on functions having multiple overloads instead of implicit cast |
|
||||
| E021-11 | POSITION function | <span class="text-warning">Partial</span> | No support for `IN` and `USING` clauses, no `POSITION_REGEX` variant |
|
||||
| E021-12 | Character comparison | <span class="text-success">Yes</span> | |
|
||||
| **E031** | **Identifiers** | <span class="text-warning">Partial</span>| |
|
||||
| E031-01 | Delimited identifiers | <span class="text-warning">Partial</span> | Unicode literal support is limited |
|
||||
| E031-02 | Lower case identifiers | <span class="text-success">Yes</span> | |
|
||||
| E031-03 | Trailing underscore | <span class="text-success">Yes</span> | |
|
||||
| **E051** | **Basic query specification** | <span class="text-warning">Partial</span>| |
|
||||
| E051-01 | SELECT DISTINCT | <span class="text-success">Yes</span> | |
|
||||
| E051-02 | GROUP BY clause | <span class="text-success">Yes</span> | |
|
||||
| E051-04 | GROUP BY can contain columns not in `<select list>` | <span class="text-success">Yes</span> | |
|
||||
| E051-05 | Select items can be renamed | <span class="text-success">Yes</span> | |
|
||||
| E051-06 | HAVING clause | <span class="text-success">Yes</span> | |
|
||||
| E051-07 | Qualified \* in select list | <span class="text-success">Yes</span> | |
|
||||
| E051-08 | Correlation name in the FROM clause | <span class="text-success">Yes</span> | |
|
||||
| E051-09 | Rename columns in the FROM clause | <span class="text-danger">No</span> | |
|
||||
| **E061** | **Basic predicates and search conditions** | <span class="text-warning">Partial</span> | |
|
||||
| E061-01 | Comparison predicate | <span class="text-success">Yes</span> | |
|
||||
| E061-02 | BETWEEN predicate | <span class="text-warning">Partial</span> | No `SYMMETRIC` and `ASYMMETRIC` clause |
|
||||
| E061-03 | IN predicate with list of values | <span class="text-success">Yes</span> | |
|
||||
| E061-04 | LIKE predicate | <span class="text-success">Yes</span> | |
|
||||
| E061-05 | LIKE predicate: ESCAPE clause | <span class="text-danger">No</span> | |
|
||||
| E061-06 | NULL predicate | <span class="text-success">Yes</span> | |
|
||||
| E061-07 | Quantified comparison predicate | <span class="text-danger">No</span> | |
|
||||
| E061-08 | EXISTS predicate | <span class="text-danger">No</span> | |
|
||||
| E061-09 | Subqueries in comparison predicate | <span class="text-success">Yes</span> | |
|
||||
| E061-11 | Subqueries in IN predicate | <span class="text-success">Yes</span> | |
|
||||
| E061-12 | Subqueries in quantified comparison predicate | <span class="text-danger">No</span> | |
|
||||
| E061-13 | Correlated subqueries | <span class="text-danger">No</span> | |
|
||||
| E061-14 | Search condition | <span class="text-success">Yes</span> | |
|
||||
| **E071** | **Basic query expressions** | <span class="text-warning">Partial</span> | |
|
||||
| E071-01 | UNION DISTINCT table operator | <span class="text-success">Yes</span> | |
|
||||
| E071-02 | UNION ALL table operator | <span class="text-success">Yes</span> | |
|
||||
| E071-03 | EXCEPT DISTINCT table operator | <span class="text-danger">No</span> | |
|
||||
| E071-05 | Columns combined via table operators need not have exactly the same data type | <span class="text-success">Yes</span> | |
|
||||
| E071-06 | Table operators in subqueries | <span class="text-success">Yes</span> | |
|
||||
| **E081** | **Basic privileges** | <span class="text-success">Yes</span> |
|
||||
| E081-01 | SELECT privilege at the table level | <span class="text-success">Yes</span> |
|
||||
| E081-02 | DELETE privilege | |
|
||||
| E081-03 | INSERT privilege at the table level | <span class="text-success">Yes</span> |
|
||||
| E081-04 | UPDATE privilege at the table level | <span class="text-success">Yes</span> |
|
||||
| E081-05 | UPDATE privilege at the column level | |
|
||||
| E081-06 | REFERENCES privilege at the table level | | |
|
||||
| E081-07 | REFERENCES privilege at the column level | | |
|
||||
| E081-08 | WITH GRANT OPTION | <span class="text-success">Yes</span> | |
|
||||
| E081-09 | USAGE privilege | | |
|
||||
| E081-10 | EXECUTE privilege | | |
|
||||
| **E091** | **Set functions** |<span class="text-success">Yes</span> |
|
||||
| E091-01 | AVG | <span class="text-success">Yes</span> | |
|
||||
| E091-02 | COUNT | <span class="text-success">Yes</span> | |
|
||||
| E091-03 | MAX | <span class="text-success">Yes</span> | |
|
||||
| E091-04 | MIN | <span class="text-success">Yes</span> | |
|
||||
| E091-05 | SUM | <span class="text-success">Yes</span> | |
|
||||
| E091-06 | ALL quantifier | <span class="text-success">Yes</span> | |
|
||||
| E091-07 | DISTINCT quantifier | <span class="text-success">Yes</span> | Not all aggregate functions supported |
|
||||
| **E101** | **Basic data manipulation** | <span class="text-warning">Partial</span> | |
|
||||
| E101-01 | INSERT statement | <span class="text-success">Yes</span> | Note: primary key in ClickHouse does not imply the `UNIQUE` constraint |
|
||||
| E101-03 | Searched UPDATE statement | <span class="text-warning">Partial</span> | There’s an `ALTER UPDATE` statement for batch data modification |
|
||||
| E101-04 | Searched DELETE statement | <span class="text-warning">Partial</span> | There’s an `ALTER DELETE` statement for batch data removal |
|
||||
| **E111** | **Single row SELECT statement** | <span class="text-danger">No</span> | |
|
||||
| **E121** | **Basic cursor support** | <span class="text-danger">No</span> | |
|
||||
| E121-01 | DECLARE CURSOR | <span class="text-danger">No</span> | |
|
||||
| E121-02 | ORDER BY columns need not be in select list | <span class="text-success">Yes</span> | |
|
||||
| E121-03 | Value expressions in ORDER BY clause | <span class="text-success">Yes</span> | |
|
||||
| E121-04 | OPEN statement | <span class="text-danger">No</span> | |
|
||||
| E121-06 | Positioned UPDATE statement | <span class="text-danger">No</span> | |
|
||||
| E121-07 | Positioned DELETE statement | <span class="text-danger">No</span> | |
|
||||
| E121-08 | CLOSE statement | <span class="text-danger">No</span> | |
|
||||
| E121-10 | FETCH statement: implicit NEXT | <span class="text-danger">No</span> | |
|
||||
| E121-17 | WITH HOLD cursors | <span class="text-danger">No</span> | |
|
||||
| **E131** | **Null value support (nulls in lieu of values)** | <span class="text-success">Yes</span> | Some restrictions apply |
|
||||
| **E141** | **Basic integrity constraints** | <span class="text-warning">Partial</span> | |
|
||||
| E141-01 | NOT NULL constraints | <span class="text-success">Yes</span> | Note: `NOT NULL` is implied for table columns by default |
|
||||
| E141-02 | UNIQUE constraint of NOT NULL columns | <span class="text-danger">No</span> | |
|
||||
| E141-03 | PRIMARY KEY constraints | <span class="text-warning">Partial</span> | |
|
||||
| E141-04 | Basic FOREIGN KEY constraint with the NO ACTION default for both referential delete action and referential update action | <span class="text-danger">No</span> | |
|
||||
| E141-06 | CHECK constraint | <span class="text-success">Yes</span> | |
|
||||
| E141-07 | Column defaults | <span class="text-success">Yes</span> | |
|
||||
| E141-08 | NOT NULL inferred on PRIMARY KEY | <span class="text-success">Yes</span> | |
|
||||
| E141-10 | Names in a foreign key can be specified in any order | <span class="text-danger">No</span> | |
|
||||
| **E151** | **Transaction support** | <span class="text-danger">No</span> | |
|
||||
| E151-01 | COMMIT statement | <span class="text-danger">No</span> | |
|
||||
| E151-02 | ROLLBACK statement | <span class="text-danger">No</span> | |
|
||||
| **E152** | **Basic SET TRANSACTION statement** | <span class="text-danger">No</span> | |
|
||||
| E152-01 | SET TRANSACTION statement: ISOLATION LEVEL SERIALIZABLE clause | <span class="text-danger">No</span> | |
|
||||
| E152-02 | SET TRANSACTION statement: READ ONLY and READ WRITE clauses | <span class="text-danger">No</span> | |
|
||||
| **E153** | **Updatable queries with subqueries** | <span class="text-success">Yes</span> | |
|
||||
| **E161** | **SQL comments using leading double minus** | <span class="text-success">Yes</span> | |
|
||||
| **E171** | **SQLSTATE support** | <span class="text-danger">No</span> | |
|
||||
| **E182** | **Host language binding** | <span class="text-danger">No</span> | |
|
||||
| **F031** | **Basic schema manipulation** | <span class="text-warning">Partial</span>| |
|
||||
| F031-01 | CREATE TABLE statement to create persistent base tables | <span class="text-warning">Partial</span> | No `SYSTEM VERSIONING`, `ON COMMIT`, `GLOBAL`, `LOCAL`, `PRESERVE`, `DELETE`, `REF IS`, `WITH OPTIONS`, `UNDER`, `LIKE`, `PERIOD FOR` clauses and no support for user resolved data types |
|
||||
| F031-02 | CREATE VIEW statement | <span class="text-warning">Partial</span> | No `RECURSIVE`, `CHECK`, `UNDER`, `WITH OPTIONS` clauses and no support for user resolved data types |
|
||||
| F031-03 | GRANT statement | <span class="text-success">Yes</span> | |
|
||||
| F031-04 | ALTER TABLE statement: ADD COLUMN clause | <span class="text-success">Yes</span> | No support for `GENERATED` clause and system time period |
|
||||
| F031-13 | DROP TABLE statement: RESTRICT clause | <span class="text-danger">No</span> | |
|
||||
| F031-16 | DROP VIEW statement: RESTRICT clause | <span class="text-danger">No</span> | |
|
||||
| F031-19 | REVOKE statement: RESTRICT clause | <span class="text-danger">No</span> | |
|
||||
| **F041** | **Basic joined table** | <span class="text-warning">Partial</span> | |
|
||||
| F041-01 | Inner join (but not necessarily the INNER keyword) | <span class="text-success">Yes</span> | |
|
||||
| F041-02 | INNER keyword | <span class="text-success">Yes</span> | |
|
||||
| F041-03 | LEFT OUTER JOIN | <span class="text-success">Yes</span> | |
|
||||
| F041-04 | RIGHT OUTER JOIN | <span class="text-success">Yes</span> | |
|
||||
| F041-05 | Outer joins can be nested | <span class="text-success">Yes</span> | |
|
||||
| F041-07 | The inner table in a left or right outer join can also be used in an inner join | <span class="text-success">Yes</span> | |
|
||||
| F041-08 | All comparison operators are supported (rather than just =) | <span class="text-danger">No</span> | |
|
||||
| **F051** | **Basic date and time** | <span class="text-warning">Partial</span> | |
|
||||
| F051-01 | DATE data type (including support of DATE literal) | <span class="text-success">Yes</span> | |
|
||||
| F051-02 | TIME data type (including support of TIME literal) with fractional seconds precision of at least 0 | <span class="text-danger">No</span> | |
|
||||
| F051-03 | TIMESTAMP data type (including support of TIMESTAMP literal) with fractional seconds precision of at least 0 and 6 | <span class="text-success">Yes</span> | |
|
||||
| F051-04 | Comparison predicate on DATE, TIME, and TIMESTAMP data types | <span class="text-success">Yes</span> | |
|
||||
| F051-05 | Explicit CAST between datetime types and character string types | <span class="text-success">Yes</span> | |
|
||||
| F051-06 | CURRENT_DATE | <span class="text-danger">No</span> | `today()` is similar |
|
||||
| F051-07 | LOCALTIME | <span class="text-danger">No</span> | `now()` is similar |
|
||||
| F051-08 | LOCALTIMESTAMP | <span class="text-danger">No</span> | |
|
||||
| **F081** | **UNION and EXCEPT in views** | <span class="text-warning">Partial</span> | |
|
||||
| **F131** | **Grouped operations** | <span class="text-warning">Partial</span> | |
|
||||
| F131-01 | WHERE, GROUP BY, and HAVING clauses supported in queries with grouped views | <span class="text-success">Yes</span> | |
|
||||
| F131-02 | Multiple tables supported in queries with grouped views | <span class="text-success">Yes</span> | |
|
||||
| F131-03 | Set functions supported in queries with grouped views | <span class="text-success">Yes</span> | |
|
||||
| F131-04 | Subqueries with GROUP BY and HAVING clauses and grouped views | <span class="text-success">Yes</span> | |
|
||||
| F131-05 | Single row SELECT with GROUP BY and HAVING clauses and grouped views | <span class="text-danger">No</span> | |
|
||||
| **F181** | **Multiple module support** | <span class="text-danger">No</span> | |
|
||||
| **F201** | **CAST function** | <span class="text-success">Yes</span> | |
|
||||
| **F221** | **Explicit defaults** | <span class="text-danger">No</span> | |
|
||||
| **F261** | **CASE expression** | <span class="text-success">Yes</span> | |
|
||||
| F261-01 | Simple CASE | <span class="text-success">Yes</span> | |
|
||||
| F261-02 | Searched CASE | <span class="text-success">Yes</span> | |
|
||||
| F261-03 | NULLIF | <span class="text-success">Yes</span> | |
|
||||
| F261-04 | COALESCE | <span class="text-success">Yes</span> | |
|
||||
| **F311** | **Schema definition statement** | <span class="text-warning">Partial</span> | |
|
||||
| F311-01 | CREATE SCHEMA | <span class="text-warning">Partial</span> | See CREATE DATABASE |
|
||||
| F311-02 | CREATE TABLE for persistent base tables | <span class="text-success">Yes</span> | |
|
||||
| F311-03 | CREATE VIEW | <span class="text-success">Yes</span> | |
|
||||
| F311-04 | CREATE VIEW: WITH CHECK OPTION | <span class="text-danger">No</span> | |
|
||||
| F311-05 | GRANT statement | <span class="text-success">Yes</span> | |
|
||||
| **F471** | **Scalar subquery values** | <span class="text-success">Yes</span> | |
|
||||
| **F481** | **Expanded NULL predicate** | <span class="text-success">Yes</span> | |
|
||||
| **F812** | **Basic flagging** | <span class="text-danger">No</span> | |
|
||||
| **S011** | **Distinct data types** | | |
|
||||
| **T321** | **Basic SQL-invoked routines** | <span class="text-danger">No</span> | |
|
||||
| T321-01 | User-defined functions with no overloading | <span class="text-danger">No</span> | |
|
||||
| T321-02 | User-defined stored procedures with no overloading | <span class="text-danger">No</span> | |
|
||||
| T321-03 | Function invocation | <span class="text-danger">No</span> | |
|
||||
| T321-04 | CALL statement | <span class="text-danger">No</span> | |
|
||||
| T321-05 | RETURN statement | <span class="text-danger">No</span> | |
|
||||
| **T631** | **IN predicate with one list element** | <span class="text-success">Yes</span> | |
|
||||
@ -2933,7 +2933,42 @@ The same as ‘today() - 1’.
|
||||
|
||||
## timeSlot
|
||||
|
||||
Rounds the time to the half hour.
|
||||
Round the time to the start of a half-an-hour length interval.
|
||||
|
||||
**Syntax**
|
||||
|
||||
```sql
|
||||
timeSlot(time[, time_zone])
|
||||
```
|
||||
|
||||
**Arguments**
|
||||
|
||||
- `time` — Time to round to the start of a half-an-hour length interval. [DateTime](../data-types/datetime.md)/[Date32](../data-types/date32.md)/[DateTime64](../data-types/datetime64.md).
|
||||
- `time_zone` — A String type const value or an expression representing the time zone. [String](../data-types/string.md).
|
||||
|
||||
:::note
|
||||
Though this function can take values of the extended types `Date32` and `DateTime64` as an argument, passing it a time outside the normal range (year 1970 to 2149 for `Date` / 2106 for `DateTime`) will produce wrong results.
|
||||
:::
|
||||
|
||||
**Return type**
|
||||
|
||||
- Returns the time rounded to the start of a half-an-hour length interval. [DateTime](../data-types/datetime.md).
|
||||
|
||||
**Example**
|
||||
|
||||
Query:
|
||||
|
||||
```sql
|
||||
SELECT timeSlot(toDateTime('2000-01-02 03:04:05', 'UTC'));
|
||||
```
|
||||
|
||||
Result:
|
||||
|
||||
```response
|
||||
┌─timeSlot(toDateTime('2000-01-02 03:04:05', 'UTC'))─┐
|
||||
│ 2000-01-02 03:00:00 │
|
||||
└────────────────────────────────────────────────────┘
|
||||
```
|
||||
|
||||
## toYYYYMM
|
||||
|
||||
|
||||
@ -5261,9 +5261,9 @@ SELECT toFixedString('foo', 8) AS s;
|
||||
Result:
|
||||
|
||||
```response
|
||||
┌─s─────────────┬─s_cut─┐
|
||||
│ foo\0\0\0\0\0 │ foo │
|
||||
└───────────────┴───────┘
|
||||
┌─s─────────────┐
|
||||
│ foo\0\0\0\0\0 │
|
||||
└───────────────┘
|
||||
```
|
||||
|
||||
## toStringCutToZero
|
||||
|
||||
@ -1,2 +0,0 @@
|
||||
# Just an empty yaml file. Keep it alone.
|
||||
{}
|
||||
@ -33,7 +33,7 @@ sidebar_label: "Отличительные возможности ClickHouse"
|
||||
|
||||
## Поддержка SQL {#sql-support}
|
||||
|
||||
ClickHouse поддерживает [декларативный язык запросов на основе SQL](../sql-reference/index.md) и во [многих случаях](../sql-reference/ansi.mdx) совпадающий с SQL-стандартом.
|
||||
ClickHouse поддерживает декларативный язык запросов SQL.
|
||||
|
||||
Поддерживаются [GROUP BY](../sql-reference/statements/select/group-by.md), [ORDER BY](../sql-reference/statements/select/order-by.md), подзапросы в секциях [FROM](../sql-reference/statements/select/from.md), [IN](../sql-reference/operators/in.md), [JOIN](../sql-reference/statements/select/join.md), [функции window](../sql-reference/window-functions/index.mdx), а также скалярные подзапросы.
|
||||
|
||||
|
||||
@ -30,7 +30,7 @@ sidebar_label: "Настройки пользователей"
|
||||
<profile>profile_name</profile>
|
||||
|
||||
<quota>default</quota>
|
||||
<default_database>default<default_database>
|
||||
<default_database>default</default_database>
|
||||
<databases>
|
||||
<database_name>
|
||||
<table_name>
|
||||
|
||||
@ -1,10 +0,0 @@
|
||||
---
|
||||
slug: /ru/sql-reference/ansi
|
||||
sidebar_position: 40
|
||||
sidebar_label: ANSI Compatibility
|
||||
title: "ANSI Compatibility"
|
||||
---
|
||||
|
||||
import Content from '@site/docs/en/sql-reference/ansi.md';
|
||||
|
||||
<Content />
|
||||
@ -37,7 +37,7 @@ ClickHouse会使用服务器上一切可用的资源，从而以最自然的方
|
||||
|
||||
## 支持SQL {#zhi-chi-sql}
|
||||
|
||||
ClickHouse支持一种[基于SQL的声明式查询语言](../sql-reference/index.md)，它在许多情况下与[ANSI SQL标准](../sql-reference/ansi.md)相同。
|
||||
ClickHouse支持一种基于SQL的声明式查询语言。
|
||||
|
||||
支持的查询[GROUP BY](../sql-reference/statements/select/group-by.md), [ORDER BY](../sql-reference/statements/select/order-by.md), [FROM](../sql-reference/statements/select/from.md), [JOIN](../sql-reference/statements/select/join.md), [IN](../sql-reference/operators/in.md)以及非相关子查询。
|
||||
|
||||
|
||||
@ -1,191 +0,0 @@
|
||||
---
|
||||
slug: /zh/sql-reference/ansi
|
||||
sidebar_position: 40
|
||||
sidebar_label: "ANSI\u517C\u5BB9\u6027"
|
||||
---
|
||||
|
||||
# ClickHouse SQL方言 与ANSI SQL的兼容性{#ansi-sql-compatibility-of-clickhouse-sql-dialect}
|
||||
|
||||
:::note
|
||||
本文参考Annex G所著的[ISO/IEC CD 9075-2:2011](https://www.iso.org/obp/ui/#iso:std:iso-iec:9075:-2:ed-4:v1:en:sec:8)标准.
|
||||
:::
|
||||
|
||||
## 行为差异 {#differences-in-behaviour}
|
||||
|
||||
下表列出了ClickHouse能够使用，但与ANSI SQL规定有差异的查询特性。
|
||||
|
||||
| 功能ID | 功能名称 | 差异 |
|
||||
| ------- | --------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
||||
| E011 | 数值型数据类型 | 带小数点的数字被视为近似值 (`Float64`）而不是精确值 (`Decimal`) |
|
||||
| E051-05 | SELECT 的列可以重命名 | 字段重命名的作用范围不限于进行重命名的SELECT子查询（参考[表达式别名](https://clickhouse.com/docs/zh/sql-reference/syntax/#notes-on-usage)） |
|
||||
| E141-01 | NOT NULL（非空）约束 | ClickHouse表中每一列默认为`NOT NULL` |
|
||||
| E011-04 | 算术运算符 | ClickHouse在运算时会进行溢出，而不是四舍五入。此外会根据自定义规则修改结果数据类型（参考[溢出检查](https://clickhouse.com/docs/zh/sql-reference/data-types/decimal/#yi-chu-jian-cha)） |
|
||||
|
||||
## 功能状态 {#feature-status}
|
||||
|
||||
| 功能ID | 功能名称 | 状态 | 注释 |
|
||||
| -------- | ---------------------------------------------------------------------------------------- | ----------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
|
||||
| **E011** | **数值型数据类型** | **部分**{.text-warning} | |
|
||||
| E011-01 | INTEGER （整型）和SMALLINT （小整型）数据类型 | 是 {.text-success} | |
|
||||
| E011-02 | REAL （实数）、DOUBLE PRECISION （双精度浮点数）和FLOAT（单精度浮点数）数据类型数据类型 | 是 {.text-success} | |
|
||||
| E011-03 | DECIMAL （精确数字）和NUMERIC （精确数字）数据类型 | 是 {.text-success} | |
|
||||
| E011-04 | 算术运算符 | 是 {.text-success} | |
|
||||
| E011-05 | 数值比较 | 是 {.text-success} | |
|
||||
| E011-06 | 数值数据类型之间的隐式转换 | 否 {.text-danger} | ANSI SQL允许在数值类型之间进行任意隐式转换，而ClickHouse针对不同数据类型有对应的比较函数和类型转换函数 |
|
||||
| **E021** | **字符串类型** | **部分**{.text-warning} | |
|
||||
| E021-01 | CHARACTER （字符串）数据类型 | 是 {.text-success} | |
|
||||
| E021-02 | CHARACTER VARYING （可变字符串）数据类型 | 是 {.text-success} | |
|
||||
| E021-03 | 字符字面量 | 是 {.text-success} | |
|
||||
| E021-04 | CHARACTER_LENGTH 函数 | 部分 {.text-warning} | 不支持 `using` 从句 |
|
||||
| E021-05 | OCTET_LENGTH 函数 | 否 {.text-danger} | 使用 `LENGTH` 函数代替 |
|
||||
| E021-06 | SUBSTRING | 部分 {.text-warning} | 不支持 `SIMILAR` 和 `ESCAPE` 从句，没有`SUBSTRING_REGEX` 函数 |
|
||||
| E021-07 | 字符串拼接 | 部分 {.text-warning} | 不支持 `COLLATE` 从句 |
|
||||
| E021-08 | 大小写转换 | 是 {.text-success} | |
|
||||
| E021-09 | 裁剪字符串 | 是 {.text-success} | |
|
||||
| E021-10 | 固定长度和可变长度字符串类型之间的隐式转换 | 部分 {.text-warning} | ANSI SQL允许在数据类型之间进行任意隐式转换，而ClickHouse针对不同数据类型有对应的比较函数和类型转换函数 |
|
||||
| E021-11 | POSITION 函数 | 部分 {.text-warning} | 不支持 `IN` 和 `USING` 从句，不支持`POSITION_REGEX`函数 |
|
||||
| E021-12 | 字符串比较 | 是 {.text-success} | |
|
||||
| **E031** | **标识符** | **部分**{.text-warning} | |
|
||||
| E031-01 | 分隔标识符 | 部分 {.text-warning} | Unicode文字支持有限 |
|
||||
| E031-02 | 小写标识符 | 是 {.text-success} | |
|
||||
| E031-03 | 标识符最后加下划线 | 是 {.text-success} | |
|
||||
| **E051** | **基本查询规范** | **部分**{.text-warning} | |
|
||||
| E051-01 | SELECT DISTINCT | 是 {.text-success} | |
|
||||
| E051-02 | GROUP BY 从句 | 是 {.text-success} | |
|
||||
| E051-04 | GROUP BY 从句中的列可以包含不在 `<select list>`中出现的列 | 是 {.text-success} | |
|
||||
| E051-05 | SELECT 的列可以重命名 | 是 {.text-success} | |
|
||||
| E051-06 | HAVING 从句 | 是 {.text-success} | |
|
||||
| E051-07 | SELECT 选择的列中允许出现\* | 是 {.text-success} | |
|
||||
| E051-08 | FROM 从句中的关联名称 | 是 {.text-success} | |
|
||||
| E051-09 | 重命名 FROM 从句中的列 | 否 {.text-danger} | |
|
||||
| **E061** | **基本谓词和搜索条件** | **部分**{.text-warning} | |
|
||||
| E061-01 | 比较谓词 | 是 {.text-success} | |
|
||||
| E061-02 | BETWEEN 谓词 | 部分 {.text-warning} | 不支持 `SYMMETRIC` 和 `ASYMMETRIC` 从句 |
|
||||
| E061-03 | IN 谓词后可接值列表 | 是 {.text-success} | |
|
||||
| E061-04 | LIKE 谓词 | 是 {.text-success} | |
|
||||
| E061-05 | LIKE 谓词后接 ESCAPE 从句 | 否 {.text-danger} | |
|
||||
| E061-06 | NULL 谓词 | 是 {.text-success} | |
|
||||
| E061-07 | 量化比较谓词（ALL、SOME、ANY） | 否 {.text-danger} | |
|
||||
| E061-08 | EXISTS 谓词 | 否 {.text-danger} | |
|
||||
| E061-09 | 比较谓词中的子查询 | 是 {.text-success} | |
|
||||
| E061-11 | IN 谓词中的子查询 | 是 {.text-success} | |
|
||||
| E061-12 | 量化比较谓词（BETWEEN、IN、LIKE）中的子查询 | 否 {.text-danger} | |
|
||||
| E061-13 | 相关子查询 | 否 {.text-danger} | |
|
||||
| E061-14 | 搜索条件 | 是 {.text-success} | |
|
||||
| **E071** | **基本查询表达式** | **部分**{.text-warning} | |
|
||||
| E071-01 | UNION DISTINCT 表运算符 | 是 {.text-success} | |
|
||||
| E071-02 | UNION ALL 表运算符 | 是 {.text-success} | |
|
||||
| E071-03 | EXCEPT DISTINCT 表运算符 | 否 {.text-danger} | |
|
||||
| E071-05 | 通过表运算符组合的列不必具有完全相同的数据类型 | 是 {.text-success} | |
|
||||
| E071-06 | 子查询中的表运算符 | 是 {.text-success} | |
|
||||
| **E081** | **基本权限** | **是**{.text-success} | |
|
||||
| E081-01 | 表级别的SELECT（查询）权限 | 是 {.text-success} | |
|
||||
| E081-02 | DELETE（删除）权限 | 是 {.text-success} | |
|
||||
| E081-03 | 表级别的INSERT（插入）权限 | 是 {.text-success} | |
|
||||
| E081-04 | 表级别的UPDATE（更新）权限 | 是 {.text-success} | |
|
||||
| E081-05 | 列级别的UPDATE（更新）权限 | 是 {.text-success} | |
|
||||
| E081-06 | 表级别的REFERENCES（引用）权限 | 是 {.text-success} | |
|
||||
| E081-07 | 列级别的REFERENCES（引用）权限 | 是 {.text-success} | |
|
||||
| E081-08 | WITH GRANT OPTION | 是 {.text-success} | |
|
||||
| E081-09 | USAGE（使用）权限 | 是 {.text-success} | |
|
||||
| E081-10 | EXECUTE（执行）权限 | 是 {.text-success} | |
|
||||
| **E091** | **集合函数** | **是**{.text-success} | |
|
||||
| E091-01 | AVG | 是 {.text-success} | |
|
||||
| E091-02 | COUNT | 是 {.text-success} | |
|
||||
| E091-03 | MAX | 是 {.text-success} | |
|
||||
| E091-04 | MIN | 是 {.text-success} | |
|
||||
| E091-05 | SUM | 是 {.text-success} | |
|
||||
| E091-06 | ALL修饰词 | 否。 {.text-danger} | |
|
||||
| E091-07 | DISTINCT修饰词 | 是 {.text-success} | 并非所有聚合函数都支持该修饰词 |
|
||||
| **E101** | **基本数据操作** | **部分**{.text-warning} | |
|
||||
| E101-01 | INSERT（插入）语句 | 是 {.text-success} | 注：ClickHouse中的主键并不隐含`UNIQUE` 约束 |
|
||||
| E101-03 | 可指定范围的UPDATE（更新）语句 | 部分 {.text-warning} | `ALTER UPDATE` 语句用来批量更新数据 |
|
||||
| E101-04 | 可指定范围的DELETE（删除）语句 | 部分 {.text-warning} | `ALTER DELETE` 语句用来批量删除数据 |
|
||||
| **E111** | **返回一行的SELECT语句** | **否**{.text-danger} | |
|
||||
| **E121** | **基本游标支持** | **否**{.text-danger} | |
|
||||
| E121-01 | DECLARE CURSOR | 否 {.text-danger} | |
|
||||
| E121-02 | ORDER BY 涉及的列不需要出现在SELECT的列中 | 是 {.text-success} | |
|
||||
| E121-03 | ORDER BY 从句中的表达式 | 是 {.text-success} | |
|
||||
| E121-04 | OPEN 语句 | 否 {.text-danger} | |
|
||||
| E121-06 | 受游标位置控制的 UPDATE 语句 | 否 {.text-danger} | |
|
||||
| E121-07 | 受游标位置控制的 DELETE 语句 | 否 {.text-danger} | |
|
||||
| E121-08 | CLOSE 语句 | 否 {.text-danger} | |
|
||||
| E121-10 | FETCH 语句中包含隐式NEXT | 否 {.text-danger} | |
|
||||
| E121-17 | WITH HOLD 游标 | 否 {.text-danger} | |
|
||||
| **E131** | **空值支持** | **是**{.text-success} | 有部分限制 |
|
||||
| **E141** | **基本完整性约束** | **部分**{.text-warning} | |
|
||||
| E141-01 | NOT NULL（非空）约束 | 是 {.text-success} | 注: 默认情况下ClickHouse表中的列隐含`NOT NULL`约束 |
|
||||
| E141-02 | NOT NULL（非空）列的UNIQUE（唯一）约束 | 否 {.text-danger} | |
|
||||
| E141-03 | 主键约束 | 部分 {.text-warning} | |
|
||||
| E141-04 | 对于引用删除和引用更新操作，基本的FOREIGN KEY（外键）约束默认不进行任何操作（NO ACTION） | 否 {.text-danger} | |
|
||||
| E141-06 | CHECK（检查）约束 | 是 {.text-success} | |
|
||||
| E141-07 | 列默认值 | 是 {.text-success} | |
|
||||
| E141-08 | 在主键上推断非空 | 是 {.text-success} | |
|
||||
| E141-10 | 可以按任何顺序指定外键中的名称 | 否 {.text-danger} | |
|
||||
| **E151** | **事务支持** | **否**{.text-danger} | |
|
||||
| E151-01 | COMMIT（提交）语句 | 否 {.text-danger} | |
|
||||
| E151-02 | ROLLBACK（回滚）语句 | 否 {.text-danger} | |
|
||||
| **E152** | **基本的SET TRANSACTION（设置事务隔离级别）语句** | **否**{.text-danger} | |
|
||||
| E152-01 | SET TRANSACTION语句：ISOLATION LEVEL SERIALIZABLE（隔离级别为串行化）从句 | 否 {.text-danger} | |
|
||||
| E152-02 | SET TRANSACTION语句：READ ONLY（只读）和READ WRITE（读写）从句 | 否 {.text-danger} | |
|
||||
| **E153** | **具有子查询的可更新查询** | **是**{.text-success} | |
|
||||
| **E161** | **使用“--”符号作为SQL注释** | **是**{.text-success} | |
|
||||
| **E171** | **SQLSTATE支持** | **否**{.text-danger} | |
|
||||
| **E182** | **主机语言绑定** | **否**{.text-danger} | |
|
||||
| **F031** | **基本架构操作** | **部分**{.text-warning} | |
|
||||
| F031-01 | 使用 CREATE TABLE 语句创建持久表 | 部分 {.text-warning} | 不支持 `SYSTEM VERSIONING`, `ON COMMIT`, `GLOBAL`, `LOCAL`, `PRESERVE`, `DELETE`, `REF IS`, `WITH OPTIONS`, `UNDER`, `LIKE`, `PERIOD FOR` 从句，不支持用户解析的数据类型 |
|
||||
| F031-02 | CREATE VIEW（创建视图）语句 | 部分 {.text-warning} | 不支持 `RECURSIVE`, `CHECK`, `UNDER`, `WITH OPTIONS` 从句，不支持用户解析的数据类型 |
|
||||
| F031-03 | GRANT（授权）语句 | 是 {.text-success} | |
|
||||
| F031-04 | ALTER TABLE语句：ADD COLUMN从句 | 是 {.text-success} | 不支持 `GENERATED` 从句和以系统时间做参数 |
|
||||
| F031-13 | DROP TABLE语句：RESTRICT从句 | 否 {.text-danger} | |
|
||||
| F031-16 | DROP VIEW语句：RESTRICT子句 | 否 {.text-danger} | |
|
||||
| F031-19 | REVOKE语句：RESTRICT子句 | 否 {.text-danger} | |
|
||||
| **F041** | **基本连接关系** | **部分**{.text-warning} | |
|
||||
| F041-01 | Inner join（但不一定是INNER关键字) | 是 {.text-success} | |
|
||||
| F041-02 | INNER 关键字 | 是 {.text-success} | |
|
||||
| F041-03 | LEFT OUTER JOIN | 是 {.text-success} | |
|
||||
| F041-04 | RIGHT OUTER JOIN | 是 {.text-success} | |
|
||||
| F041-05 | 外连接可嵌套 | 是 {.text-success} | |
|
||||
| F041-07 | 左外部连接或右外连接中的内部表也可用于内部联接 | 是 {.text-success} | |
|
||||
| F041-08 | 支持所有比较运算符（而不仅仅是=) | 否 {.text-danger} | |
|
||||
| **F051** | **基本日期和时间** | **部分**{.text-warning} | |
|
||||
| F051-01 | DATE（日期）数据类型（并支持用于表达日期的字面量) | 是 {.text-success} | |
|
||||
| F051-02 | TIME（时间）数据类型（并支持用于表达时间的字面量），小数秒精度至少为0 | 否 {.text-danger} | |
|
||||
| F051-03 | 时间戳数据类型（并支持用于表达时间戳的字面量），小数秒精度至少为0和6 | 是 {.text-danger} | |
|
||||
| F051-04 | 日期、时间和时间戳数据类型的比较谓词 | 是 {.text-success} | |
|
||||
| F051-05 | DateTime 类型和字符串形式表达的时间之间的显式转换 | 是 {.text-success} | |
|
||||
| F051-06 | CURRENT_DATE | 否 {.text-danger} | 使用`today()`替代 |
|
||||
| F051-07 | LOCALTIME | 否 {.text-danger} | 使用`now()`替代 |
|
||||
| F051-08 | LOCALTIMESTAMP | 否 {.text-danger} | |
|
||||
| **F081** | **视图的UNION和EXCEPT操作** | **部分**{.text-warning} | |
|
||||
| **F131** | **分组操作** | **部分**{.text-warning} | |
|
||||
| F131-01 | 在具有分组视图的查询中支持 WHERE、GROUP BY 和 HAVING 子句 | 是 {.text-success} | |
|
||||
| F131-02 | 在分组视图中支持多张表 | 是 {.text-success} | |
|
||||
| F131-03 | 分组视图的查询中支持集合函数 | 是 {.text-success} | |
|
||||
| F131-04 | 带有 `GROUP BY` 和 `HAVING` 从句，以及分组视图的子查询 | 是 {.text-success} | |
|
||||
| F131-05 | 带有 `GROUP BY` 和 `HAVING` 从句，以及分组视图的仅返回1条记录的SELECT查询 | 否 {.text-danger} | |
|
||||
| **F181** | **多模块支持** | **否**{.text-danger} | |
|
||||
| **F201** | **CAST 函数** | **是**{.text-success} | |
|
||||
| **F221** | **显式默认值** | **否**{.text-danger} | |
|
||||
| **F261** | **CASE 表达式** | **是**{.text-success} | |
|
||||
| F261-01 | 简单 CASE 表达式 | 是 {.text-success} | |
|
||||
| F261-02 | 搜索型 CASE 表达式 | 是 {.text-success} | |
|
||||
| F261-03 | NULLIF | 是 {.text-success} | |
|
||||
| F261-04 | COALESCE | 是 {.text-success} | |
|
||||
| **F311** | **架构定义语句** | **部分**{.text-warning} | |
|
||||
| F311-01 | CREATE SCHEMA | 部分 {.text-warning} | 见`CREATE DATABASE` |
|
||||
| F311-02 | 用于创建持久表的 CREATE TABLE | 是 {.text-success} | |
|
||||
| F311-03 | CREATE VIEW | 是 {.text-success} | |
|
||||
| F311-04 | CREATE VIEW: WITH CHECK OPTION | 否 {.text-danger} | |
|
||||
| F311-05 | GRANT 语句 | 是 {.text-success} | |
|
||||
| **F471** | **标量子查询** | **是**{.text-success} | |
|
||||
| **F481** | **扩展 NULL 谓词** | **是**{.text-success} | |
|
||||
| **F812** | **基本标志位** | **否**{.text-danger} |
|
||||
| **S011** | **用于不重复数据的数据类型** | **否**{.text-danger} |
|
||||
| **T321** | **基本的SQL调用例程** | **否**{.text-danger} | |
|
||||
| T321-01 | 没有重载的用户定义函数 | 否{.text-danger} | |
|
||||
| T321-02 | 没有重载的用户定义存储过程 | 否{.text-danger} | |
|
||||
| T321-03 | 功能调用 | 否 {.text-danger} | |
|
||||
| T321-04 | CALL 语句 | 否 {.text-danger} | |
|
||||
| T321-05 | RETURN 语句 | 否 {.text-danger} | |
|
||||
| **T631** | **IN 谓词后接一个列表** | **是**{.text-success} | |
|
||||
@ -38,18 +38,10 @@ public:
|
||||
String path_to = disk_to.getRelativeFromRoot(getValueFromCommandLineOptionsThrow<String>(options, "path-to"));
|
||||
bool recursive = options.count("recursive");
|
||||
|
||||
if (!disk_from.getDisk()->exists(path_from))
|
||||
{
|
||||
throw Exception(
|
||||
ErrorCodes::BAD_ARGUMENTS,
|
||||
"cannot stat '{}' on disk '{}': No such file or directory",
|
||||
path_from,
|
||||
disk_from.getDisk()->getName());
|
||||
}
|
||||
if (disk_from.getDisk()->isFile(path_from))
|
||||
if (disk_from.getDisk()->existsFile(path_from))
|
||||
{
|
||||
auto target_location = getTargetLocation(path_from, disk_to, path_to);
|
||||
if (!disk_to.getDisk()->exists(target_location) || disk_to.getDisk()->isFile(target_location))
|
||||
if (!disk_to.getDisk()->existsDirectory(target_location))
|
||||
{
|
||||
disk_from.getDisk()->copyFile(
|
||||
path_from,
|
||||
@ -65,7 +57,7 @@ public:
|
||||
ErrorCodes::BAD_ARGUMENTS, "cannot overwrite directory {} with non-directory {}", target_location, path_from);
|
||||
}
|
||||
}
|
||||
else if (disk_from.getDisk()->isDirectory(path_from))
|
||||
else if (disk_from.getDisk()->existsDirectory(path_from))
|
||||
{
|
||||
if (!recursive)
|
||||
{
|
||||
@ -73,11 +65,11 @@ public:
|
||||
}
|
||||
auto target_location = getTargetLocation(path_from, disk_to, path_to);
|
||||
|
||||
if (disk_to.getDisk()->isFile(target_location))
|
||||
if (disk_to.getDisk()->existsFile(target_location))
|
||||
{
|
||||
throw Exception(ErrorCodes::BAD_ARGUMENTS, "cannot overwrite non-directory {} with directory {}", path_to, target_location);
|
||||
}
|
||||
if (!disk_to.getDisk()->exists(target_location))
|
||||
if (!disk_to.getDisk()->existsDirectory(target_location))
|
||||
{
|
||||
disk_to.getDisk()->createDirectory(target_location);
|
||||
}
|
||||
@ -89,6 +81,14 @@ public:
|
||||
/* write_settings= */ {},
|
||||
/* cancellation_hook= */ {});
|
||||
}
|
||||
else
|
||||
{
|
||||
throw Exception(
|
||||
ErrorCodes::BAD_ARGUMENTS,
|
||||
"cannot stat '{}' on disk '{}': No such file or directory",
|
||||
path_from,
|
||||
disk_from.getDisk()->getName());
|
||||
}
|
||||
}
|
||||
};
|
||||
|
||||
|
||||
@ -30,21 +30,21 @@ public:
|
||||
String path_from = disk.getRelativeFromRoot(getValueFromCommandLineOptionsThrow<String>(options, "path-from"));
|
||||
String path_to = disk.getRelativeFromRoot(getValueFromCommandLineOptionsThrow<String>(options, "path-to"));
|
||||
|
||||
if (disk.getDisk()->isFile(path_from))
|
||||
if (disk.getDisk()->existsFile(path_from))
|
||||
{
|
||||
disk.getDisk()->moveFile(path_from, path_to);
|
||||
}
|
||||
else if (disk.getDisk()->isDirectory(path_from))
|
||||
else if (disk.getDisk()->existsDirectory(path_from))
|
||||
{
|
||||
auto target_location = getTargetLocation(path_from, disk, path_to);
|
||||
if (!disk.getDisk()->exists(target_location))
|
||||
if (!disk.getDisk()->existsDirectory(target_location))
|
||||
{
|
||||
disk.getDisk()->createDirectory(target_location);
|
||||
disk.getDisk()->moveDirectory(path_from, target_location);
|
||||
}
|
||||
else
|
||||
{
|
||||
if (disk.getDisk()->isFile(target_location))
|
||||
if (disk.getDisk()->existsFile(target_location))
|
||||
{
|
||||
throw Exception(
|
||||
ErrorCodes::BAD_ARGUMENTS, "cannot overwrite non-directory '{}' with directory '{}'", target_location, path_from);
|
||||
@ -57,7 +57,7 @@ public:
|
||||
disk.getDisk()->moveDirectory(path_from, target_location);
|
||||
}
|
||||
}
|
||||
else if (!disk.getDisk()->exists(path_from))
|
||||
else
|
||||
{
|
||||
throw Exception(
|
||||
ErrorCodes::BAD_ARGUMENTS,
|
||||
|
||||
@ -28,11 +28,7 @@ public:
|
||||
auto disk = client.getCurrentDiskWithPath();
|
||||
const String & path = disk.getRelativeFromRoot(getValueFromCommandLineOptionsThrow<String>(options, "path"));
|
||||
bool recursive = options.count("recursive");
|
||||
if (!disk.getDisk()->exists(path))
|
||||
{
|
||||
throw Exception(ErrorCodes::BAD_ARGUMENTS, "Path {} on disk {} doesn't exist", path, disk.getDisk()->getName());
|
||||
}
|
||||
if (disk.getDisk()->isDirectory(path))
|
||||
if (disk.getDisk()->existsDirectory(path))
|
||||
{
|
||||
if (!recursive)
|
||||
{
|
||||
@ -41,10 +37,12 @@ public:
|
||||
|
||||
disk.getDisk()->removeRecursive(path);
|
||||
}
|
||||
else
|
||||
else if (disk.getDisk()->existsFile(path))
|
||||
{
|
||||
disk.getDisk()->removeFileIfExists(path);
|
||||
}
|
||||
else
|
||||
throw Exception(ErrorCodes::BAD_ARGUMENTS, "Path {} on disk {} doesn't exist", path, disk.getDisk()->getName());
|
||||
}
|
||||
};
|
||||
|
||||
|
||||
@ -1,6 +1,7 @@
|
||||
#include <Interpreters/Context.h>
|
||||
#include "ICommand.h"
|
||||
|
||||
#include <IO/ReadBufferFromEmptyFile.h>
|
||||
#include <IO/ReadBufferFromFile.h>
|
||||
#include <IO/WriteBufferFromFile.h>
|
||||
#include <IO/copyData.h>
|
||||
@ -36,7 +37,11 @@ public:
|
||||
return std::make_unique<ReadBufferFromFileDescriptor>(STDIN_FILENO);
|
||||
|
||||
String relative_path_from = disk.getRelativeFromRoot(path_from.value());
|
||||
return disk.getDisk()->readFile(relative_path_from, getReadSettings());
|
||||
auto res = disk.getDisk()->readFileIfExists(relative_path_from, getReadSettings());
|
||||
if (res)
|
||||
return res;
|
||||
/// For backward compatibility.
|
||||
return std::make_unique<ReadBufferFromEmptyFile>();
|
||||
}();
|
||||
|
||||
auto out = disk.getDisk()->writeFile(path_to);
|
||||
|
||||
@ -29,7 +29,7 @@ DiskWithPath::DiskWithPath(DiskPtr disk_, std::optional<String> path_) : disk(di
|
||||
}
|
||||
|
||||
String relative_path = normalizePathAndGetAsRelative(path);
|
||||
if (disk->isDirectory(relative_path) || (relative_path.empty() && (disk->isDirectory("/"))))
|
||||
if (disk->existsDirectory(relative_path) || (relative_path.empty() && (disk->existsDirectory("/"))))
|
||||
{
|
||||
return;
|
||||
}
|
||||
|
||||
@ -33,7 +33,7 @@ public:
|
||||
|
||||
bool isDirectory(const String & any_path) const
|
||||
{
|
||||
return disk->isDirectory(getRelativeFromRoot(any_path)) || (getRelativeFromRoot(any_path).empty() && (disk->isDirectory("/")));
|
||||
return disk->existsDirectory(getRelativeFromRoot(any_path)) || (getRelativeFromRoot(any_path).empty() && (disk->existsDirectory("/")));
|
||||
}
|
||||
|
||||
std::vector<String> listAllFilesByPath(const String & any_path) const;
|
||||
|
||||
@ -90,7 +90,7 @@ protected:
|
||||
|
||||
String getTargetLocation(const String & path_from, DiskWithPath & disk_to, const String & path_to)
|
||||
{
|
||||
if (!disk_to.getDisk()->isDirectory(path_to))
|
||||
if (!disk_to.getDisk()->existsDirectory(path_to))
|
||||
{
|
||||
return path_to;
|
||||
}
|
||||
|
||||
@ -296,9 +296,14 @@ void LocalServer::tryInitPath()
|
||||
|
||||
global_context->setUserFilesPath(""); /// user's files are everywhere
|
||||
|
||||
std::string user_scripts_path = getClientConfiguration().getString("user_scripts_path", fs::path(path) / "user_scripts/");
|
||||
std::string user_scripts_path = getClientConfiguration().getString("user_scripts_path", fs::path(path) / "user_scripts" / "");
|
||||
global_context->setUserScriptsPath(user_scripts_path);
|
||||
|
||||
/// Set path for filesystem caches
|
||||
String filesystem_caches_path(getClientConfiguration().getString("filesystem_caches_path", fs::path(path) / "cache" / ""));
|
||||
if (!filesystem_caches_path.empty())
|
||||
global_context->setFilesystemCachesPath(filesystem_caches_path);
|
||||
|
||||
/// top_level_domains_lists
|
||||
const std::string & top_level_domains_path = getClientConfiguration().getString("top_level_domains_path", fs::path(path) / "top_level_domains/");
|
||||
if (!top_level_domains_path.empty())
|
||||
@ -853,6 +858,7 @@ void LocalServer::addOptions(OptionsDescription & options_description)
|
||||
{
|
||||
options_description.main_description->add_options()
|
||||
("table,N", po::value<std::string>(), "name of the initial table")
|
||||
("copy", "shortcut for format conversion, equivalent to: --query 'SELECT * FROM table'")
|
||||
|
||||
/// If structure argument is omitted then initial query is not generated
|
||||
("structure,S", po::value<std::string>(), "structure of the initial table (list of column and type names)")
|
||||
@ -925,6 +931,12 @@ void LocalServer::processOptions(const OptionsDescription &, const CommandLineOp
|
||||
getClientConfiguration().setString("send_logs_level", options["send_logs_level"].as<std::string>());
|
||||
if (options.count("wait_for_suggestions_to_load"))
|
||||
getClientConfiguration().setBool("wait_for_suggestions_to_load", true);
|
||||
if (options.count("copy"))
|
||||
{
|
||||
if (!queries.empty())
|
||||
throw Exception(ErrorCodes::BAD_ARGUMENTS, "Options '--copy' and '--query' cannot be specified at the same time");
|
||||
queries.emplace_back("SELECT * FROM table");
|
||||
}
|
||||
}
|
||||
|
||||
void LocalServer::readArguments(int argc, char ** argv, Arguments & common_arguments, std::vector<Arguments> &, std::vector<Arguments> &)
|
||||
|
||||
@ -1496,6 +1496,8 @@ try
|
||||
|
||||
NamedCollectionFactory::instance().loadIfNot();
|
||||
|
||||
FileCacheFactory::instance().loadDefaultCaches(config());
|
||||
|
||||
/// Initialize main config reloader.
|
||||
std::string include_from_path = config().getString("include_from", "/etc/metrika.xml");
|
||||
|
||||
|
||||
@ -117,20 +117,20 @@ bool operator ==(const AuthenticationData & lhs, const AuthenticationData & rhs)
|
||||
}
|
||||
|
||||
|
||||
void AuthenticationData::setPassword(const String & password_)
|
||||
void AuthenticationData::setPassword(const String & password_, bool validate)
|
||||
{
|
||||
switch (type)
|
||||
{
|
||||
case AuthenticationType::PLAINTEXT_PASSWORD:
|
||||
setPasswordHashBinary(Util::stringToDigest(password_));
|
||||
setPasswordHashBinary(Util::stringToDigest(password_), validate);
|
||||
return;
|
||||
|
||||
case AuthenticationType::SHA256_PASSWORD:
|
||||
setPasswordHashBinary(Util::encodeSHA256(password_));
|
||||
setPasswordHashBinary(Util::encodeSHA256(password_), validate);
|
||||
return;
|
||||
|
||||
case AuthenticationType::DOUBLE_SHA1_PASSWORD:
|
||||
setPasswordHashBinary(Util::encodeDoubleSHA1(password_));
|
||||
setPasswordHashBinary(Util::encodeDoubleSHA1(password_), validate);
|
||||
return;
|
||||
|
||||
case AuthenticationType::BCRYPT_PASSWORD:
|
||||
@ -149,12 +149,12 @@ void AuthenticationData::setPassword(const String & password_)
|
||||
throw Exception(ErrorCodes::NOT_IMPLEMENTED, "setPassword(): authentication type {} not supported", toString(type));
|
||||
}
|
||||
|
||||
void AuthenticationData::setPasswordBcrypt(const String & password_, int workfactor_)
|
||||
void AuthenticationData::setPasswordBcrypt(const String & password_, int workfactor_, bool validate)
|
||||
{
|
||||
if (type != AuthenticationType::BCRYPT_PASSWORD)
|
||||
throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot specify bcrypt password for authentication type {}", toString(type));
|
||||
|
||||
setPasswordHashBinary(Util::encodeBcrypt(password_, workfactor_));
|
||||
setPasswordHashBinary(Util::encodeBcrypt(password_, workfactor_), validate);
|
||||
}
|
||||
|
||||
String AuthenticationData::getPassword() const
|
||||
@ -165,7 +165,7 @@ String AuthenticationData::getPassword() const
|
||||
}
|
||||
|
||||
|
||||
void AuthenticationData::setPasswordHashHex(const String & hash)
|
||||
void AuthenticationData::setPasswordHashHex(const String & hash, bool validate)
|
||||
{
|
||||
Digest digest;
|
||||
digest.resize(hash.size() / 2);
|
||||
@ -179,7 +179,7 @@ void AuthenticationData::setPasswordHashHex(const String & hash)
|
||||
throw Exception(ErrorCodes::BAD_ARGUMENTS, "Cannot read password hash in hex, check for valid characters [0-9a-fA-F] and length");
|
||||
}
|
||||
|
||||
setPasswordHashBinary(digest);
|
||||
setPasswordHashBinary(digest, validate);
|
||||
}
|
||||
|
||||
|
||||
@ -195,7 +195,7 @@ String AuthenticationData::getPasswordHashHex() const
|
||||
}
|
||||
|
||||
|
||||
void AuthenticationData::setPasswordHashBinary(const Digest & hash)
|
||||
void AuthenticationData::setPasswordHashBinary(const Digest & hash, bool validate)
|
||||
{
|
||||
switch (type)
|
||||
{
|
||||
@ -217,7 +217,7 @@ void AuthenticationData::setPasswordHashBinary(const Digest & hash)
|
||||
|
||||
case AuthenticationType::DOUBLE_SHA1_PASSWORD:
|
||||
{
|
||||
if (hash.size() != 20)
|
||||
if (validate && hash.size() != 20)
|
||||
throw Exception(ErrorCodes::BAD_ARGUMENTS,
|
||||
"Password hash for the 'DOUBLE_SHA1_PASSWORD' authentication type has length {} "
|
||||
"but must be exactly 20 bytes.", hash.size());
|
||||
@ -231,7 +231,7 @@ void AuthenticationData::setPasswordHashBinary(const Digest & hash)
|
||||
/// However the library we use to encode it requires hash string to be 64 characters long,
|
||||
/// so we also allow the hash of this length.
|
||||
|
||||
if (hash.size() != 59 && hash.size() != 60 && hash.size() != 64)
|
||||
if (validate && hash.size() != 59 && hash.size() != 60 && hash.size() != 64)
|
||||
throw Exception(ErrorCodes::BAD_ARGUMENTS,
|
||||
"Password hash for the 'BCRYPT_PASSWORD' authentication type has length {} "
|
||||
"but must be 59 or 60 bytes.", hash.size());
|
||||
@ -240,10 +240,13 @@ void AuthenticationData::setPasswordHashBinary(const Digest & hash)
|
||||
resized.resize(64);
|
||||
|
||||
#if USE_BCRYPT
|
||||
/// Verify that it is a valid hash
|
||||
int ret = bcrypt_checkpw("", reinterpret_cast<const char *>(resized.data()));
|
||||
if (ret == -1)
|
||||
throw Exception(ErrorCodes::BAD_ARGUMENTS, "Could not decode the provided hash with 'bcrypt_hash'");
|
||||
if (validate)
|
||||
{
|
||||
/// Verify that it is a valid hash
|
||||
int ret = bcrypt_checkpw("", reinterpret_cast<const char *>(resized.data()));
|
||||
if (ret == -1)
|
||||
throw Exception(ErrorCodes::BAD_ARGUMENTS, "Could not decode the provided hash with 'bcrypt_hash'");
|
||||
}
|
||||
#endif
|
||||
|
||||
password_hash = hash;
|
||||
@ -385,7 +388,7 @@ std::shared_ptr<ASTAuthenticationData> AuthenticationData::toAST() const
|
||||
}
|
||||
|
||||
|
||||
AuthenticationData AuthenticationData::fromAST(const ASTAuthenticationData & query, ContextPtr context, bool check_password_rules)
|
||||
AuthenticationData AuthenticationData::fromAST(const ASTAuthenticationData & query, ContextPtr context, bool validate)
|
||||
{
|
||||
if (query.type && query.type == AuthenticationType::NO_PASSWORD)
|
||||
return AuthenticationData();
|
||||
@ -431,7 +434,7 @@ AuthenticationData AuthenticationData::fromAST(const ASTAuthenticationData & que
|
||||
if (!query.type && !context)
|
||||
throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot get default password type without context");
|
||||
|
||||
if (check_password_rules && !context)
|
||||
if (validate && !context)
|
||||
throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot check password complexity rules without context");
|
||||
|
||||
if (query.type == AuthenticationType::BCRYPT_PASSWORD && !context)
|
||||
@ -448,13 +451,13 @@ AuthenticationData AuthenticationData::fromAST(const ASTAuthenticationData & que
|
||||
|
||||
AuthenticationData auth_data(current_type);
|
||||
|
||||
if (check_password_rules)
|
||||
if (validate)
|
||||
context->getAccessControl().checkPasswordComplexityRules(value);
|
||||
|
||||
if (query.type == AuthenticationType::BCRYPT_PASSWORD)
|
||||
{
|
||||
int workfactor = context->getAccessControl().getBcryptWorkfactor();
|
||||
auth_data.setPasswordBcrypt(value, workfactor);
|
||||
auth_data.setPasswordBcrypt(value, workfactor, validate);
|
||||
return auth_data;
|
||||
}
|
||||
|
||||
@ -486,7 +489,7 @@ AuthenticationData AuthenticationData::fromAST(const ASTAuthenticationData & que
|
||||
#endif
|
||||
}
|
||||
|
||||
auth_data.setPassword(value);
|
||||
auth_data.setPassword(value, validate);
|
||||
return auth_data;
|
||||
}
|
||||
|
||||
@ -498,11 +501,11 @@ AuthenticationData AuthenticationData::fromAST(const ASTAuthenticationData & que
|
||||
|
||||
if (query.type == AuthenticationType::BCRYPT_PASSWORD)
|
||||
{
|
||||
auth_data.setPasswordHashBinary(AuthenticationData::Util::stringToDigest(value));
|
||||
auth_data.setPasswordHashBinary(AuthenticationData::Util::stringToDigest(value), validate);
|
||||
return auth_data;
|
||||
}
|
||||
|
||||
auth_data.setPasswordHashHex(value);
|
||||
auth_data.setPasswordHashHex(value, validate);
|
||||
|
||||
|
||||
if (query.type == AuthenticationType::SHA256_PASSWORD && args_size == 2)
|
||||
|
||||
@ -31,17 +31,17 @@ public:
|
||||
AuthenticationType getType() const { return type; }
|
||||
|
||||
/// Sets the password and encrypt it using the authentication type set in the constructor.
|
||||
void setPassword(const String & password_);
|
||||
void setPassword(const String & password_, bool validate);
|
||||
|
||||
/// Returns the password. Allowed to use only for Type::PLAINTEXT_PASSWORD.
|
||||
String getPassword() const;
|
||||
|
||||
/// Sets the password as a string of hexadecimal digits.
|
||||
void setPasswordHashHex(const String & hash);
|
||||
void setPasswordHashHex(const String & hash, bool validate);
|
||||
String getPasswordHashHex() const;
|
||||
|
||||
/// Sets the password in binary form.
|
||||
void setPasswordHashBinary(const Digest & hash);
|
||||
void setPasswordHashBinary(const Digest & hash, bool validate);
|
||||
const Digest & getPasswordHashBinary() const { return password_hash; }
|
||||
|
||||
/// Sets the salt in String form.
|
||||
@ -49,7 +49,7 @@ public:
|
||||
String getSalt() const;
|
||||
|
||||
/// Sets the password using bcrypt hash with specified workfactor
|
||||
void setPasswordBcrypt(const String & password_, int workfactor_);
|
||||
void setPasswordBcrypt(const String & password_, int workfactor_, bool validate);
|
||||
|
||||
/// Sets the server name for authentication type LDAP.
|
||||
const String & getLDAPServerName() const { return ldap_server_name; }
|
||||
@ -77,7 +77,7 @@ public:
|
||||
friend bool operator ==(const AuthenticationData & lhs, const AuthenticationData & rhs);
|
||||
friend bool operator !=(const AuthenticationData & lhs, const AuthenticationData & rhs) { return !(lhs == rhs); }
|
||||
|
||||
static AuthenticationData fromAST(const ASTAuthenticationData & query, ContextPtr context, bool check_password_rules);
|
||||
static AuthenticationData fromAST(const ASTAuthenticationData & query, ContextPtr context, bool validate);
|
||||
std::shared_ptr<ASTAuthenticationData> toAST() const;
|
||||
|
||||
struct Util
|
||||
|
||||
@ -121,6 +121,7 @@ namespace
|
||||
bool allow_no_password,
|
||||
bool allow_plaintext_password)
|
||||
{
|
||||
const bool validate = true;
|
||||
auto user = std::make_shared<User>();
|
||||
user->setName(user_name);
|
||||
String user_config = "users." + user_name;
|
||||
@ -157,17 +158,17 @@ namespace
|
||||
if (has_password_plaintext)
|
||||
{
|
||||
user->authentication_methods.emplace_back(AuthenticationType::PLAINTEXT_PASSWORD);
|
||||
user->authentication_methods.back().setPassword(config.getString(user_config + ".password"));
|
||||
user->authentication_methods.back().setPassword(config.getString(user_config + ".password"), validate);
|
||||
}
|
||||
else if (has_password_sha256_hex)
|
||||
{
|
||||
user->authentication_methods.emplace_back(AuthenticationType::SHA256_PASSWORD);
|
||||
user->authentication_methods.back().setPasswordHashHex(config.getString(user_config + ".password_sha256_hex"));
|
||||
user->authentication_methods.back().setPasswordHashHex(config.getString(user_config + ".password_sha256_hex"), validate);
|
||||
}
|
||||
else if (has_password_double_sha1_hex)
|
||||
{
|
||||
user->authentication_methods.emplace_back(AuthenticationType::DOUBLE_SHA1_PASSWORD);
|
||||
user->authentication_methods.back().setPasswordHashHex(config.getString(user_config + ".password_double_sha1_hex"));
|
||||
user->authentication_methods.back().setPasswordHashHex(config.getString(user_config + ".password_double_sha1_hex"), validate);
|
||||
}
|
||||
else if (has_ldap)
|
||||
{
|
||||
|
||||
@ -28,11 +28,36 @@ namespace ErrorCodes
|
||||
namespace
|
||||
{
|
||||
|
||||
constexpr size_t max_events = 32;
|
||||
constexpr size_t MAX_EVENTS = 32;
|
||||
|
||||
|
||||
template <typename T>
|
||||
void mergeEventsList(T & events_list, size_t prefix_size, bool prefix_sorted, bool suffix_sorted)
|
||||
{
|
||||
/// either sort whole container or do so partially merging ranges afterwards
|
||||
if (!prefix_sorted && !suffix_sorted)
|
||||
std::stable_sort(std::begin(events_list), std::end(events_list));
|
||||
else
|
||||
{
|
||||
const auto begin = std::begin(events_list);
|
||||
const auto middle = std::next(begin, prefix_size);
|
||||
const auto end = std::end(events_list);
|
||||
|
||||
if (!prefix_sorted)
|
||||
std::stable_sort(begin, middle);
|
||||
|
||||
if (!suffix_sorted)
|
||||
std::stable_sort(middle, end);
|
||||
|
||||
std::inplace_merge(begin, middle, end);
|
||||
}
|
||||
}
|
||||
|
||||
template <typename T>
|
||||
struct AggregateFunctionWindowFunnelData
|
||||
{
|
||||
static constexpr bool strict_once_enabled = false;
|
||||
|
||||
using TimestampEvent = std::pair<T, UInt8>;
|
||||
using TimestampEvents = PODArrayWithStackMemory<TimestampEvent, 64>;
|
||||
|
||||
@ -66,24 +91,7 @@ struct AggregateFunctionWindowFunnelData
|
||||
|
||||
events_list.insert(std::begin(other.events_list), std::end(other.events_list));
|
||||
|
||||
/// either sort whole container or do so partially merging ranges afterwards
|
||||
if (!sorted && !other.sorted)
|
||||
std::stable_sort(std::begin(events_list), std::end(events_list));
|
||||
else
|
||||
{
|
||||
const auto begin = std::begin(events_list);
|
||||
const auto middle = std::next(begin, size);
|
||||
const auto end = std::end(events_list);
|
||||
|
||||
if (!sorted)
|
||||
std::stable_sort(begin, middle);
|
||||
|
||||
if (!other.sorted)
|
||||
std::stable_sort(middle, end);
|
||||
|
||||
std::inplace_merge(begin, middle, end);
|
||||
}
|
||||
|
||||
mergeEventsList(events_list, size, sorted, other.sorted);
|
||||
sorted = true;
|
||||
}
|
||||
|
||||
@ -133,6 +141,131 @@ struct AggregateFunctionWindowFunnelData
|
||||
}
|
||||
};
|
||||
|
||||
template <typename T>
|
||||
struct AggregateFunctionWindowFunnelStrictOnceData
|
||||
{
|
||||
static constexpr bool strict_once_enabled = true;
|
||||
struct TimestampEvent
|
||||
{
|
||||
T timestamp;
|
||||
UInt8 event_type;
|
||||
UInt64 unique_id;
|
||||
|
||||
TimestampEvent(T timestamp_, UInt8 event_type_, UInt64 unique_id_)
|
||||
: timestamp(timestamp_), event_type(event_type_), unique_id(unique_id_) {}
|
||||
|
||||
bool operator<(const TimestampEvent & other) const
|
||||
{
|
||||
return std::tie(timestamp, event_type, unique_id) < std::tie(other.timestamp, other.event_type, other.unique_id);
|
||||
}
|
||||
|
||||
bool operator<=(const TimestampEvent & other) const
|
||||
{
|
||||
return std::tie(timestamp, event_type, unique_id) <= std::tie(other.timestamp, other.event_type, other.unique_id);
|
||||
}
|
||||
};
|
||||
|
||||
|
||||
using TimestampEvents = PODArrayWithStackMemory<TimestampEvent, 64>;
|
||||
TimestampEvents events_list;
|
||||
|
||||
/// Next unique identifier for events
|
||||
/// Used to distinguish events with the same timestamp that matches several conditions.
|
||||
UInt64 next_unique_id = 1;
|
||||
bool sorted = true;
|
||||
|
||||
size_t size() const
|
||||
{
|
||||
return events_list.size();
|
||||
}
|
||||
|
||||
void advanceId()
|
||||
{
|
||||
++next_unique_id;
|
||||
}
|
||||
|
||||
void add(T timestamp, UInt8 event_type)
|
||||
{
|
||||
TimestampEvent new_event(timestamp, event_type, next_unique_id);
|
||||
/// Check if the new event maintains the sorted order
|
||||
if (sorted && !events_list.empty())
|
||||
sorted = events_list.back() <= new_event;
|
||||
events_list.push_back(new_event);
|
||||
}
|
||||
|
||||
void merge(const AggregateFunctionWindowFunnelStrictOnceData & other)
|
||||
{
|
||||
if (other.events_list.empty())
|
||||
return;
|
||||
|
||||
const auto current_size = events_list.size();
|
||||
|
||||
UInt64 new_next_unique_id = next_unique_id;
|
||||
events_list.reserve(current_size + other.events_list.size());
|
||||
for (auto other_event : other.events_list)
|
||||
{
|
||||
/// Assign unique IDs to the new events to prevent conflicts
|
||||
other_event.unique_id += next_unique_id;
|
||||
new_next_unique_id = std::max(new_next_unique_id, other_event.unique_id + 1);
|
||||
events_list.push_back(other_event);
|
||||
}
|
||||
next_unique_id = new_next_unique_id;
|
||||
|
||||
mergeEventsList(events_list, current_size, sorted, other.sorted);
|
||||
|
||||
sorted = true;
|
||||
}
|
||||
|
||||
void sort()
|
||||
{
|
||||
if (!sorted)
|
||||
{
|
||||
std::stable_sort(std::begin(events_list), std::end(events_list));
|
||||
sorted = true;
|
||||
}
|
||||
}
|
||||
|
||||
void serialize(WriteBuffer & buf) const
|
||||
{
|
||||
writeBinary(sorted, buf);
|
||||
writeBinary(events_list.size(), buf);
|
||||
|
||||
for (const auto & event : events_list)
|
||||
{
|
||||
writeBinary(event.timestamp, buf);
|
||||
writeBinary(event.event_type, buf);
|
||||
writeBinary(event.unique_id, buf);
|
||||
}
|
||||
}
|
||||
|
||||
void deserialize(ReadBuffer & buf)
|
||||
{
|
||||
readBinary(sorted, buf);
|
||||
|
||||
size_t events_size;
|
||||
readBinary(events_size, buf);
|
||||
|
||||
if (events_size > 100'000'000) /// Arbitrary limit to prevent excessive memory allocation
|
||||
throw Exception(ErrorCodes::TOO_LARGE_ARRAY_SIZE, "Too large size of the state of windowFunnel");
|
||||
|
||||
events_list.clear();
|
||||
events_list.reserve(events_size);
|
||||
|
||||
T timestamp;
|
||||
UInt8 event_type;
|
||||
UInt64 unique_id = 0;
|
||||
|
||||
for (size_t i = 0; i < events_size; ++i)
|
||||
{
|
||||
readBinary(timestamp, buf);
|
||||
readBinary(event_type, buf);
|
||||
readBinary(unique_id, buf);
|
||||
next_unique_id = std::max(next_unique_id, unique_id + 1);
|
||||
events_list.emplace_back(timestamp, event_type, unique_id);
|
||||
}
|
||||
}
|
||||
};
|
||||
|
||||
/** Calculates the max event level in a sliding window.
|
||||
* The max size of events is 32, that's enough for funnel analytics
|
||||
*
|
||||
@ -160,22 +293,15 @@ private:
|
||||
/// The level path must be 1---2---3---...---check_events_size, find the max event level that satisfied the path in the sliding window.
|
||||
/// If found, returns the max event level, else return 0.
|
||||
/// The algorithm works in O(n) time, but the overall function works in O(n * log(n)) due to sorting.
|
||||
UInt8 getEventLevel(Data & data) const
|
||||
UInt8 getEventLevelNonStrictOnce(const AggregateFunctionWindowFunnelData<T>::TimestampEvents & events_list) const
|
||||
{
|
||||
if (data.size() == 0)
|
||||
return 0;
|
||||
if (!strict_order && events_size == 1)
|
||||
return 1;
|
||||
|
||||
data.sort();
|
||||
|
||||
/// events_timestamp stores the timestamp of the first and previous i-th level event happen within time window
|
||||
std::vector<std::optional<std::pair<UInt64, UInt64>>> events_timestamp(events_size);
|
||||
bool first_event = false;
|
||||
for (size_t i = 0; i < data.events_list.size(); ++i)
|
||||
for (size_t i = 0; i < events_list.size(); ++i)
|
||||
{
|
||||
const T & timestamp = data.events_list[i].first;
|
||||
const auto & event_idx = data.events_list[i].second - 1;
|
||||
const T & timestamp = events_list[i].first;
|
||||
const auto & event_idx = events_list[i].second - 1;
|
||||
if (strict_order && event_idx == -1)
|
||||
{
|
||||
if (first_event)
|
||||
@ -189,7 +315,7 @@ private:
|
||||
}
|
||||
else if (strict_deduplication && events_timestamp[event_idx].has_value())
|
||||
{
|
||||
return data.events_list[i - 1].second;
|
||||
return events_list[i - 1].second;
|
||||
}
|
||||
else if (strict_order && first_event && !events_timestamp[event_idx - 1].has_value())
|
||||
{
|
||||
@ -222,6 +348,126 @@ private:
|
||||
return 0;
|
||||
}
|
||||
|
||||
UInt8 getEventLevelStrictOnce(const AggregateFunctionWindowFunnelStrictOnceData<T>::TimestampEvents & events_list) const
|
||||
{
|
||||
/// Stores the timestamp of the first and last i-th level event happen within time window
|
||||
struct EventMatchTimeWindow
|
||||
{
|
||||
UInt64 first_timestamp;
|
||||
UInt64 last_timestamp;
|
||||
std::array<UInt64, MAX_EVENTS> event_path;
|
||||
|
||||
EventMatchTimeWindow() = default;
|
||||
EventMatchTimeWindow(UInt64 first_ts, UInt64 last_ts)
|
||||
: first_timestamp(first_ts), last_timestamp(last_ts) {}
|
||||
};
|
||||
|
||||
/// We track all possible event sequences up to the current event.
|
||||
/// It's required because one event can meet several conditions.
|
||||
/// For example: for events 'start', 'a', 'b', 'a', 'end'.
|
||||
/// The second occurrence of 'a' should be counted only once in one sequence.
|
||||
/// However, we do not know in advance if the next event will be 'b' or 'end', so we try to keep both paths.
|
||||
std::vector<std::list<EventMatchTimeWindow>> event_sequences(events_size);
|
||||
|
||||
bool has_first_event = false;
|
||||
for (size_t i = 0; i < events_list.size(); ++i)
|
||||
{
|
||||
const auto & current_event = events_list[i];
|
||||
auto timestamp = current_event.timestamp;
|
||||
Int64 event_idx = current_event.event_type - 1;
|
||||
UInt64 unique_id = current_event.unique_id;
|
||||
|
||||
if (strict_order && event_idx == -1)
|
||||
{
|
||||
if (has_first_event)
|
||||
break;
|
||||
else
|
||||
continue;
|
||||
}
|
||||
else if (event_idx == 0)
|
||||
{
|
||||
auto & event_seq = event_sequences[0].emplace_back(timestamp, timestamp);
|
||||
event_seq.event_path[0] = unique_id;
|
||||
has_first_event = true;
|
||||
}
|
||||
else if (strict_deduplication && !event_sequences[event_idx].empty())
|
||||
{
|
||||
return events_list[i - 1].event_type;
|
||||
}
|
||||
else if (strict_order && has_first_event && event_sequences[event_idx - 1].empty())
|
||||
{
|
||||
for (size_t event = 0; event < event_sequences.size(); ++event)
|
||||
{
|
||||
if (event_sequences[event].empty())
|
||||
return event;
|
||||
}
|
||||
}
|
||||
else if (!event_sequences[event_idx - 1].empty())
|
||||
{
|
||||
auto & prev_level = event_sequences[event_idx - 1];
|
||||
for (auto it = prev_level.begin(); it != prev_level.end();)
|
||||
{
|
||||
auto first_ts = it->first_timestamp;
|
||||
bool time_matched = timestamp <= first_ts + window;
|
||||
if (!time_matched && prev_level.size() > 1)
|
||||
{
|
||||
// Remove old events that are out of the window, but keep at least one
|
||||
it = prev_level.erase(it);
|
||||
continue;
|
||||
}
|
||||
|
||||
auto prev_path = it->event_path;
|
||||
chassert(event_idx > 0);
|
||||
|
||||
/// Ensure the unique_id hasn't been used in the path already
|
||||
for (size_t j = 0; j < static_cast<size_t>(event_idx); ++j)
|
||||
{
|
||||
if (!time_matched)
|
||||
break;
|
||||
time_matched = prev_path[j] != unique_id;
|
||||
}
|
||||
|
||||
if (time_matched && strict_increase)
|
||||
time_matched = it->last_timestamp < timestamp;
|
||||
|
||||
if (time_matched)
|
||||
{
|
||||
prev_path[event_idx] = unique_id;
|
||||
|
||||
auto & new_seq = event_sequences[event_idx].emplace_back(first_ts, timestamp);
|
||||
new_seq.event_path = std::move(prev_path);
|
||||
if (event_idx + 1 == events_size)
|
||||
return events_size;
|
||||
}
|
||||
++it;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
for (size_t event = event_sequences.size(); event > 0; --event)
|
||||
{
|
||||
if (!event_sequences[event - 1].empty())
|
||||
return event;
|
||||
}
|
||||
return 0;
|
||||
}
|
||||
|
||||
|
||||
UInt8 getEventLevel(Data & data) const
|
||||
{
|
||||
if (data.size() == 0)
|
||||
return 0;
|
||||
if (!strict_order && events_size == 1)
|
||||
return 1;
|
||||
|
||||
data.sort();
|
||||
|
||||
if constexpr (Data::strict_once_enabled)
|
||||
return getEventLevelStrictOnce(data.events_list);
|
||||
else
|
||||
return getEventLevelNonStrictOnce(data.events_list);
|
||||
}
|
||||
|
||||
public:
|
||||
String getName() const override
|
||||
{
|
||||
@ -246,6 +492,9 @@ public:
|
||||
strict_order = true;
|
||||
else if (option == "strict_increase")
|
||||
strict_increase = true;
|
||||
else if (option == "strict_once")
|
||||
/// Checked in factory
|
||||
chassert(Data::strict_once_enabled);
|
||||
else if (option == "strict")
|
||||
throw Exception(ErrorCodes::BAD_ARGUMENTS, "strict is replaced with strict_deduplication in Aggregate function {}", getName());
|
||||
else
|
||||
@ -272,6 +521,9 @@ public:
|
||||
|
||||
if (strict_order && !has_event)
|
||||
this->data(place).add(timestamp, 0);
|
||||
|
||||
if constexpr (Data::strict_once_enabled)
|
||||
this->data(place).advanceId();
|
||||
}
|
||||
|
||||
void merge(AggregateDataPtr __restrict place, ConstAggregateDataPtr rhs, Arena *) const override
|
||||
@ -296,7 +548,6 @@ public:
|
||||
};
|
||||
|
||||
|
||||
template <template <typename> class Data>
|
||||
AggregateFunctionPtr
|
||||
createAggregateFunctionWindowFunnel(const std::string & name, const DataTypes & arguments, const Array & params, const Settings *)
|
||||
{
|
||||
@ -309,7 +560,7 @@ createAggregateFunctionWindowFunnel(const std::string & name, const DataTypes &
|
||||
throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH,
|
||||
"Aggregate function {} requires one timestamp argument and at least one event condition.", name);
|
||||
|
||||
if (arguments.size() > max_events + 1)
|
||||
if (arguments.size() > MAX_EVENTS + 1)
|
||||
throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Too many event arguments for aggregate function {}", name);
|
||||
|
||||
for (const auto i : collections::range(1, arguments.size()))
|
||||
@ -321,16 +572,29 @@ createAggregateFunctionWindowFunnel(const std::string & name, const DataTypes &
|
||||
cond_arg->getName(), toString(i + 1), name);
|
||||
}
|
||||
|
||||
AggregateFunctionPtr res(createWithUnsignedIntegerType<AggregateFunctionWindowFunnel, Data>(*arguments[0], arguments, params));
|
||||
WhichDataType which(arguments.front().get());
|
||||
if (res)
|
||||
return res;
|
||||
if (which.isDate())
|
||||
return std::make_shared<AggregateFunctionWindowFunnel<DataTypeDate::FieldType, Data<DataTypeDate::FieldType>>>(arguments, params);
|
||||
if (which.isDateTime())
|
||||
return std::make_shared<AggregateFunctionWindowFunnel<DataTypeDateTime::FieldType, Data<DataTypeDateTime::FieldType>>>(
|
||||
arguments, params);
|
||||
|
||||
bool strict_once = params.size() > 1 && std::any_of(params.begin() + 1, params.end(), [](const auto & f) { return f.template safeGet<String>() == "strict_once"; });
|
||||
if (strict_once)
|
||||
{
|
||||
AggregateFunctionPtr res(createWithUnsignedIntegerType<AggregateFunctionWindowFunnel, AggregateFunctionWindowFunnelStrictOnceData>(*arguments[0], arguments, params));
|
||||
WhichDataType which(arguments.front().get());
|
||||
if (res)
|
||||
return res;
|
||||
if (which.isDate())
|
||||
return std::make_shared<AggregateFunctionWindowFunnel<DataTypeDate::FieldType, AggregateFunctionWindowFunnelStrictOnceData<DataTypeDate::FieldType>>>(arguments, params);
|
||||
if (which.isDateTime())
|
||||
return std::make_shared<AggregateFunctionWindowFunnel<DataTypeDateTime::FieldType, AggregateFunctionWindowFunnelStrictOnceData<DataTypeDateTime::FieldType>>>(arguments, params);
|
||||
}
|
||||
else
|
||||
{
|
||||
AggregateFunctionPtr res(createWithUnsignedIntegerType<AggregateFunctionWindowFunnel, AggregateFunctionWindowFunnelData>(*arguments[0], arguments, params));
|
||||
WhichDataType which(arguments.front().get());
|
||||
if (res)
|
||||
return res;
|
||||
if (which.isDate())
|
||||
return std::make_shared<AggregateFunctionWindowFunnel<DataTypeDate::FieldType, AggregateFunctionWindowFunnelData<DataTypeDate::FieldType>>>(arguments, params);
|
||||
if (which.isDateTime())
|
||||
return std::make_shared<AggregateFunctionWindowFunnel<DataTypeDateTime::FieldType, AggregateFunctionWindowFunnelData<DataTypeDateTime::FieldType>>>(arguments, params);
|
||||
}
|
||||
throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT,
|
||||
"Illegal type {} of first argument of aggregate function {}, must "
|
||||
"be Unsigned Number, Date, DateTime", arguments.front().get()->getName(), name);
|
||||
@ -340,7 +604,7 @@ createAggregateFunctionWindowFunnel(const std::string & name, const DataTypes &
|
||||
|
||||
void registerAggregateFunctionWindowFunnel(AggregateFunctionFactory & factory)
|
||||
{
|
||||
factory.registerFunction("windowFunnel", createAggregateFunctionWindowFunnel<AggregateFunctionWindowFunnelData>);
|
||||
factory.registerFunction("windowFunnel", createAggregateFunctionWindowFunnel);
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
@ -78,11 +78,6 @@ struct WindowFunction : public IAggregateFunctionHelper<WindowFunction>, public
|
||||
}
|
||||
|
||||
String getName() const override { return name; }
|
||||
void create(AggregateDataPtr __restrict) const override { }
|
||||
void destroy(AggregateDataPtr __restrict) const noexcept override { }
|
||||
bool hasTrivialDestructor() const override { return true; }
|
||||
size_t sizeOfData() const override { return 0; }
|
||||
size_t alignOfData() const override { return 1; }
|
||||
void add(AggregateDataPtr __restrict, const IColumn **, size_t, Arena *) const override { fail(); }
|
||||
void merge(AggregateDataPtr __restrict, ConstAggregateDataPtr, Arena *) const override { fail(); }
|
||||
void serialize(ConstAggregateDataPtr __restrict, WriteBuffer &, std::optional<size_t>) const override { fail(); }
|
||||
@ -90,6 +85,22 @@ struct WindowFunction : public IAggregateFunctionHelper<WindowFunction>, public
|
||||
void insertResultInto(AggregateDataPtr __restrict, IColumn &, Arena *) const override { fail(); }
|
||||
};
|
||||
|
||||
struct StatelessWindowFunction : public WindowFunction
|
||||
{
|
||||
StatelessWindowFunction(
|
||||
const std::string & name_, const DataTypes & argument_types_, const Array & parameters_, const DataTypePtr & result_type_)
|
||||
: WindowFunction(name_, argument_types_, parameters_, result_type_)
|
||||
{
|
||||
}
|
||||
|
||||
size_t sizeOfData() const override { return 0; }
|
||||
size_t alignOfData() const override { return 1; }
|
||||
|
||||
void create(AggregateDataPtr __restrict) const override { }
|
||||
void destroy(AggregateDataPtr __restrict) const noexcept override { }
|
||||
bool hasTrivialDestructor() const override { return true; }
|
||||
};
|
||||
|
||||
template <typename State>
|
||||
struct StatefulWindowFunction : public WindowFunction
|
||||
{
|
||||
@ -100,7 +111,7 @@ struct StatefulWindowFunction : public WindowFunction
|
||||
}
|
||||
|
||||
size_t sizeOfData() const override { return sizeof(State); }
|
||||
size_t alignOfData() const override { return 1; }
|
||||
size_t alignOfData() const override { return alignof(State); }
|
||||
|
||||
void create(AggregateDataPtr __restrict place) const override { new (place) State(); }
|
||||
|
||||
|
||||
@ -34,7 +34,7 @@ namespace ErrorCodes
|
||||
namespace
|
||||
{
|
||||
|
||||
void exctractJoinConditions(const QueryTreeNodePtr & node, QueryTreeNodes & equi_conditions, QueryTreeNodes & other)
|
||||
void extractJoinConditions(const QueryTreeNodePtr & node, QueryTreeNodes & equi_conditions, QueryTreeNodes & other)
|
||||
{
|
||||
auto * func = node->as<FunctionNode>();
|
||||
if (!func)
|
||||
@ -52,7 +52,7 @@ void exctractJoinConditions(const QueryTreeNodePtr & node, QueryTreeNodes & equi
|
||||
else if (func->getFunctionName() == "and")
|
||||
{
|
||||
for (const auto & arg : args)
|
||||
exctractJoinConditions(arg, equi_conditions, other);
|
||||
extractJoinConditions(arg, equi_conditions, other);
|
||||
}
|
||||
else
|
||||
{
|
||||
@ -118,7 +118,7 @@ public:
|
||||
|
||||
QueryTreeNodes equi_conditions;
|
||||
QueryTreeNodes other_conditions;
|
||||
exctractJoinConditions(where_condition, equi_conditions, other_conditions);
|
||||
extractJoinConditions(where_condition, equi_conditions, other_conditions);
|
||||
bool can_convert_cross_to_inner = false;
|
||||
for (auto & condition : equi_conditions)
|
||||
{
|
||||
|
||||
@ -432,6 +432,14 @@ QueryTreeNodePtr IdentifierResolver::tryResolveTableIdentifierFromDatabaseCatalo
|
||||
else
|
||||
storage = DatabaseCatalog::instance().tryGetTable(storage_id, context);
|
||||
|
||||
if (!storage && storage_id.hasUUID())
|
||||
{
|
||||
// If `storage_id` has UUID, it is possible that the UUID is removed from `DatabaseCatalog` after `context->resolveStorageID(storage_id)`
|
||||
// We try to get the table with the database name and the table name.
|
||||
auto database = DatabaseCatalog::instance().tryGetDatabase(storage_id.getDatabaseName());
|
||||
if (database)
|
||||
storage = database->tryGetTable(table_name, context);
|
||||
}
|
||||
if (!storage)
|
||||
return {};
|
||||
|
||||
|
||||
@ -20,7 +20,7 @@ BackupReaderDisk::~BackupReaderDisk() = default;
|
||||
|
||||
bool BackupReaderDisk::fileExists(const String & file_name)
|
||||
{
|
||||
return disk->exists(root_path / file_name);
|
||||
return disk->existsFile(root_path / file_name);
|
||||
}
|
||||
|
||||
UInt64 BackupReaderDisk::getFileSize(const String & file_name)
|
||||
@ -68,7 +68,7 @@ BackupWriterDisk::~BackupWriterDisk() = default;
|
||||
|
||||
bool BackupWriterDisk::fileExists(const String & file_name)
|
||||
{
|
||||
return disk->exists(root_path / file_name);
|
||||
return disk->existsFile(root_path / file_name);
|
||||
}
|
||||
|
||||
UInt64 BackupWriterDisk::getFileSize(const String & file_name)
|
||||
@ -91,7 +91,7 @@ std::unique_ptr<WriteBuffer> BackupWriterDisk::writeFile(const String & file_nam
|
||||
void BackupWriterDisk::removeFile(const String & file_name)
|
||||
{
|
||||
disk->removeFileIfExists(root_path / file_name);
|
||||
if (disk->isDirectory(root_path) && disk->isDirectoryEmpty(root_path))
|
||||
if (disk->existsDirectory(root_path) && disk->isDirectoryEmpty(root_path))
|
||||
disk->removeDirectory(root_path);
|
||||
}
|
||||
|
||||
@ -99,7 +99,7 @@ void BackupWriterDisk::removeFiles(const Strings & file_names)
|
||||
{
|
||||
for (const auto & file_name : file_names)
|
||||
disk->removeFileIfExists(root_path / file_name);
|
||||
if (disk->isDirectory(root_path) && disk->isDirectoryEmpty(root_path))
|
||||
if (disk->existsDirectory(root_path) && disk->isDirectoryEmpty(root_path))
|
||||
disk->removeDirectory(root_path);
|
||||
}
|
||||
|
||||
|
||||
@ -244,6 +244,7 @@ add_object_library(clickhouse_storages Storages)
|
||||
add_object_library(clickhouse_storages_mysql Storages/MySQL)
|
||||
add_object_library(clickhouse_storages_distributed Storages/Distributed)
|
||||
add_object_library(clickhouse_storages_mergetree Storages/MergeTree)
|
||||
add_object_library(clickhouse_storages_mergetree_merge_selectors Storages/MergeTree/MergeSelectors)
|
||||
add_object_library(clickhouse_storages_statistics Storages/Statistics)
|
||||
add_object_library(clickhouse_storages_liveview Storages/LiveView)
|
||||
add_object_library(clickhouse_storages_windowview Storages/WindowView)
|
||||
|
||||
@ -15,7 +15,7 @@ namespace DB
|
||||
{
|
||||
struct ConnectionParameters
|
||||
{
|
||||
std::string host;
|
||||
String host;
|
||||
UInt16 port{};
|
||||
std::string default_database;
|
||||
std::string user;
|
||||
@ -30,8 +30,8 @@ struct ConnectionParameters
|
||||
ConnectionTimeouts timeouts;
|
||||
|
||||
ConnectionParameters() = default;
|
||||
ConnectionParameters(const Poco::Util::AbstractConfiguration & config, std::string host);
|
||||
ConnectionParameters(const Poco::Util::AbstractConfiguration & config, std::string host, std::optional<UInt16> port);
|
||||
ConnectionParameters(const Poco::Util::AbstractConfiguration & config, String host);
|
||||
ConnectionParameters(const Poco::Util::AbstractConfiguration & config, String host, std::optional<UInt16> port);
|
||||
|
||||
static UInt16 getPortFromConfig(const Poco::Util::AbstractConfiguration & config, const std::string & connection_host);
|
||||
|
||||
|
||||
@ -291,9 +291,14 @@
|
||||
M(CacheWarmerBytesInProgress, "Total size of remote file segments waiting to be asynchronously loaded into filesystem cache.") \
|
||||
M(DistrCacheOpenedConnections, "Number of open connections to Distributed Cache") \
|
||||
M(DistrCacheUsedConnections, "Number of currently used connections to Distributed Cache") \
|
||||
M(DistrCacheAllocatedConnections, "Number of currently allocated connections to Distributed Cache connection pool") \
|
||||
M(DistrCacheBorrowedConnections, "Number of currently borrowed connections to Distributed Cache connection pool") \
|
||||
M(DistrCacheReadRequests, "Number of executed Read requests to Distributed Cache") \
|
||||
M(DistrCacheWriteRequests, "Number of executed Write requests to Distributed Cache") \
|
||||
M(DistrCacheServerConnections, "Number of open connections to ClickHouse server from Distributed Cache") \
|
||||
M(DistrCacheRegisteredServers, "Number of distributed cache registered servers") \
|
||||
M(DistrCacheRegisteredServersCurrentAZ, "Number of distributed cache registered servers in current az") \
|
||||
M(DistrCacheServerS3CachedClients, "Number of distributed cache S3 cached clients") \
|
||||
\
|
||||
M(SchedulerIOReadScheduled, "Number of IO reads are being scheduled currently") \
|
||||
M(SchedulerIOWriteScheduled, "Number of IO writes are being scheduled currently") \
|
||||
@ -314,6 +319,20 @@
|
||||
M(FilteringMarksWithSecondaryKeys, "Number of threads currently doing filtering of mark ranges by secondary keys") \
|
||||
\
|
||||
M(DiskS3NoSuchKeyErrors, "The number of `NoSuchKey` errors that occur when reading data from S3 cloud storage through ClickHouse disks.") \
|
||||
\
|
||||
M(SharedCatalogStateApplicationThreads, "Number of threads in the threadpool for state application in Shared Catalog.") \
|
||||
M(SharedCatalogStateApplicationThreadsActive, "Number of active threads in the threadpool for state application in Shared Catalog.") \
|
||||
M(SharedCatalogStateApplicationThreadsScheduled, "Number of queued or active jobs in the threadpool for state application in Shared Catalog.") \
|
||||
\
|
||||
M(SharedCatalogDropLocalThreads, "Number of threads in the threadpool for drop of local tables in Shared Catalog.") \
|
||||
M(SharedCatalogDropLocalThreadsActive, "Number of active threads in the threadpool for drop of local tables in Shared Catalog.") \
|
||||
M(SharedCatalogDropLocalThreadsScheduled, "Number of queued or active jobs in the threadpool for drop of local tables in Shared Catalog.") \
|
||||
\
|
||||
M(SharedCatalogDropZooKeeperThreads, "Number of threads in the threadpool for drop of object in ZooKeeper in Shared Catalog.") \
|
||||
M(SharedCatalogDropZooKeeperThreadsActive, "Number of active threads in the threadpool for drop of object in ZooKeeper in Shared Catalog.") \
|
||||
M(SharedCatalogDropZooKeeperThreadsScheduled, "Number of queued or active jobs in the threadpool for drop of object in ZooKeeper in Shared Catalog.") \
|
||||
\
|
||||
M(SharedDatabaseCatalogTablesInLocalDropDetachQueue, "Number of tables in the queue for local drop or detach in Shared Catalog.") \
|
||||
|
||||
#ifdef APPLY_FOR_EXTERNAL_METRICS
|
||||
#define APPLY_FOR_METRICS(M) APPLY_FOR_BUILTIN_METRICS(M) APPLY_FOR_EXTERNAL_METRICS(M)
|
||||
|
||||
@ -452,7 +452,7 @@
|
||||
M(553, LZMA_STREAM_ENCODER_FAILED) \
|
||||
M(554, LZMA_STREAM_DECODER_FAILED) \
|
||||
M(555, ROCKSDB_ERROR) \
|
||||
M(556, SYNC_MYSQL_USER_ACCESS_ERROR)\
|
||||
M(556, SYNC_MYSQL_USER_ACCESS_ERROR) \
|
||||
M(557, UNKNOWN_UNION) \
|
||||
M(558, EXPECTED_ALL_OR_DISTINCT) \
|
||||
M(559, INVALID_GRPC_QUERY_INFO) \
|
||||
@ -578,7 +578,7 @@
|
||||
M(697, CANNOT_RESTORE_TO_NONENCRYPTED_DISK) \
|
||||
M(698, INVALID_REDIS_STORAGE_TYPE) \
|
||||
M(699, INVALID_REDIS_TABLE_STRUCTURE) \
|
||||
M(700, USER_SESSION_LIMIT_EXCEEDED) \
|
||||
M(700, USER_SESSION_LIMIT_EXCEEDED) \
|
||||
M(701, CLUSTER_DOESNT_EXIST) \
|
||||
M(702, CLIENT_INFO_DOES_NOT_MATCH) \
|
||||
M(703, INVALID_IDENTIFIER) \
|
||||
@ -610,15 +610,17 @@
|
||||
M(729, ILLEGAL_TIME_SERIES_TAGS) \
|
||||
M(730, REFRESH_FAILED) \
|
||||
M(731, QUERY_CACHE_USED_WITH_NON_THROW_OVERFLOW_MODE) \
|
||||
\
|
||||
M(733, TABLE_IS_BEING_RESTARTED) \
|
||||
\
|
||||
M(900, DISTRIBUTED_CACHE_ERROR) \
|
||||
M(901, CANNOT_USE_DISTRIBUTED_CACHE) \
|
||||
\
|
||||
M(902, PROTOCOL_VERSION_MISMATCH) \
|
||||
\
|
||||
M(999, KEEPER_EXCEPTION) \
|
||||
M(1000, POCO_EXCEPTION) \
|
||||
M(1001, STD_EXCEPTION) \
|
||||
M(1002, UNKNOWN_EXCEPTION) \
|
||||
/* See END */
|
||||
/* See END */
|
||||
|
||||
#ifdef APPLY_FOR_EXTERNAL_ERROR_CODES
|
||||
#define APPLY_FOR_ERROR_CODES(M) APPLY_FOR_BUILTIN_ERROR_CODES(M) APPLY_FOR_EXTERNAL_ERROR_CODES(M)
|
||||
|
||||
@ -49,11 +49,21 @@ static struct InitFiu
|
||||
ONCE(smt_commit_write_zk_fail_before_op) \
|
||||
ONCE(smt_commit_merge_change_version_before_op) \
|
||||
ONCE(smt_merge_mutate_intention_freeze_in_destructor) \
|
||||
ONCE(smt_add_part_sleep_after_add_before_commit) \
|
||||
ONCE(smt_sleep_in_constructor) \
|
||||
ONCE(meta_in_keeper_create_metadata_failure) \
|
||||
ONCE(smt_insert_retry_timeout) \
|
||||
ONCE(smt_insert_fake_hardware_error) \
|
||||
ONCE(smt_sleep_after_hardware_in_insert) \
|
||||
ONCE(smt_throw_keeper_exception_after_successful_insert) \
|
||||
REGULAR(smt_dont_merge_first_part) \
|
||||
REGULAR(smt_sleep_in_schedule_data_processing_job) \
|
||||
REGULAR(cache_warmer_stall) \
|
||||
REGULAR(check_table_query_delay_for_part) \
|
||||
REGULAR(dummy_failpoint) \
|
||||
REGULAR(prefetched_reader_pool_failpoint) \
|
||||
REGULAR(shared_set_sleep_during_update) \
|
||||
REGULAR(smt_outdated_parts_exception_response) \
|
||||
PAUSEABLE_ONCE(replicated_merge_tree_insert_retry_pause) \
|
||||
PAUSEABLE_ONCE(finish_set_quorum_failed_parts) \
|
||||
PAUSEABLE_ONCE(finish_clean_quorum_failed_parts) \
|
||||
|
||||
@ -196,7 +196,7 @@ void FileChecker::load()
|
||||
|
||||
bool FileChecker::fileReallyExists(const String & path_) const
|
||||
{
|
||||
return disk ? disk->exists(path_) : fs::exists(path_);
|
||||
return disk ? disk->existsFile(path_) : fs::exists(path_);
|
||||
}
|
||||
|
||||
size_t FileChecker::getRealFileSize(const String & path_) const
|
||||
|
||||
@ -222,6 +222,8 @@
|
||||
M(SelectedBytes, "Number of bytes (uncompressed; for columns as they stored in memory) SELECTed from all tables.", ValueType::Bytes) \
|
||||
M(RowsReadByMainReader, "Number of rows read from MergeTree tables by the main reader (after PREWHERE step).", ValueType::Number) \
|
||||
M(RowsReadByPrewhereReaders, "Number of rows read from MergeTree tables (in total) by prewhere readers.", ValueType::Number) \
|
||||
M(LoadedDataParts, "Number of data parts loaded by MergeTree tables during initialization.", ValueType::Number) \
|
||||
M(LoadedDataPartsMicroseconds, "Microseconds spent by MergeTree tables for loading data parts during initialization.", ValueType::Microseconds) \
|
||||
\
|
||||
M(WaitMarksLoadMicroseconds, "Time spent loading marks", ValueType::Microseconds) \
|
||||
M(BackgroundLoadingMarksTasks, "Number of background tasks for loading marks", ValueType::Number) \
|
||||
@ -241,6 +243,8 @@
|
||||
M(MergeVerticalStageExecuteMilliseconds, "Total busy time spent for execution of vertical stage of background merges", ValueType::Milliseconds) \
|
||||
M(MergeProjectionStageTotalMilliseconds, "Total time spent for projection stage of background merges", ValueType::Milliseconds) \
|
||||
M(MergeProjectionStageExecuteMilliseconds, "Total busy time spent for execution of projection stage of background merges", ValueType::Milliseconds) \
|
||||
M(MergePrewarmStageTotalMilliseconds, "Total time spent for prewarm stage of background merges", ValueType::Milliseconds) \
|
||||
M(MergePrewarmStageExecuteMilliseconds, "Total busy time spent for execution of prewarm stage of background merges", ValueType::Milliseconds) \
|
||||
\
|
||||
M(MergingSortedMilliseconds, "Total time spent while merging sorted columns", ValueType::Milliseconds) \
|
||||
M(AggregatingSortedMilliseconds, "Total time spent while aggregating sorted columns", ValueType::Milliseconds) \
|
||||
@ -639,6 +643,8 @@ The server successfully detected this situation and will download merged part fr
|
||||
M(MetadataFromKeeperBackgroundCleanupTransactions, "Number of times old transaction idempotency token was cleaned up by background task", ValueType::Number) \
|
||||
M(MetadataFromKeeperBackgroundCleanupErrors, "Number of times an error was encountered in background cleanup task", ValueType::Number) \
|
||||
\
|
||||
M(SharedMergeTreeMetadataCacheHintLoadedFromCache, "Number of times metadata cache hint was found without going to Keeper", ValueType::Number) \
|
||||
\
|
||||
M(KafkaRebalanceRevocations, "Number of partition revocations (the first stage of consumer group rebalance)", ValueType::Number) \
|
||||
M(KafkaRebalanceAssignments, "Number of partition assignments (the final stage of consumer group rebalance)", ValueType::Number) \
|
||||
M(KafkaRebalanceErrors, "Number of failed consumer group rebalances", ValueType::Number) \
|
||||
@ -742,29 +748,51 @@ The server successfully detected this situation and will download merged part fr
|
||||
M(ConnectionPoolIsFullMicroseconds, "Total time spent waiting for a slot in connection pool.", ValueType::Microseconds) \
|
||||
M(AsyncLoaderWaitMicroseconds, "Total time a query was waiting for async loader jobs.", ValueType::Microseconds) \
|
||||
\
|
||||
M(DistrCacheServerSwitches, "Number of server switches between distributed cache servers in read/write-through cache", ValueType::Number) \
|
||||
M(DistrCacheReadMicroseconds, "Time spent reading from distributed cache", ValueType::Microseconds) \
|
||||
M(DistrCacheFallbackReadMicroseconds, "Time spend reading from fallback buffer instead of distribted cache", ValueType::Microseconds) \
|
||||
M(DistrCachePrecomputeRangesMicroseconds, "Time spent to precompute read ranges", ValueType::Microseconds) \
|
||||
M(DistrCacheNextImplMicroseconds, "Time spend in ReadBufferFromDistributedCache::nextImpl", ValueType::Microseconds) \
|
||||
M(DistrCacheOpenedConnections, "The number of open connections to distributed cache", ValueType::Number) \
|
||||
M(DistrCacheReusedConnections, "The number of reused connections to distributed cache", ValueType::Number) \
|
||||
M(DistrCacheHoldConnections, "The number of used connections to distributed cache", ValueType::Number) \
|
||||
M(DistrCacheServerSwitches, "Distributed Cache read buffer event. Number of server switches between distributed cache servers in read/write-through cache", ValueType::Number) \
|
||||
M(DistrCacheReadMicroseconds, "Distributed Cache read buffer event. Time spent reading from distributed cache", ValueType::Microseconds) \
|
||||
M(DistrCacheFallbackReadMicroseconds, "Distributed Cache read buffer event. Time spend reading from fallback buffer instead of distributed cache", ValueType::Microseconds) \
|
||||
M(DistrCachePrecomputeRangesMicroseconds, "Distributed Cache read buffer event. Time spent to precompute read ranges", ValueType::Microseconds) \
|
||||
M(DistrCacheNextImplMicroseconds, "Distributed Cache read buffer event. Time spend in ReadBufferFromDistributedCache::nextImpl", ValueType::Microseconds) \
|
||||
M(DistrCacheStartRangeMicroseconds, "Distributed Cache read buffer event. Time spent to start a new read range with distributed cache", ValueType::Microseconds) \
|
||||
M(DistrCacheIgnoredBytesWhileWaitingProfileEvents, "Distributed Cache read buffer event. Ignored bytes while waiting for profile events in distributed cache", ValueType::Number) \
|
||||
M(DistrCacheRangeChange, "Distributed Cache read buffer event. Number of times we changed read range because of seek/last_position change", ValueType::Number) \
|
||||
\
|
||||
M(DistrCacheGetResponseMicroseconds, "Time spend to wait for response from distributed cache", ValueType::Microseconds) \
|
||||
M(DistrCacheStartRangeMicroseconds, "Time spent to start a new read range with distributed cache", ValueType::Microseconds) \
|
||||
M(DistrCacheLockRegistryMicroseconds, "Time spent to take DistributedCacheRegistry lock", ValueType::Microseconds) \
|
||||
M(DistrCacheUnusedPackets, "Number of skipped unused packets from distributed cache", ValueType::Number) \
|
||||
M(DistrCachePackets, "Total number of packets received from distributed cache", ValueType::Number) \
|
||||
M(DistrCacheUnusedPacketsBytes, "The number of bytes in Data packets which were ignored", ValueType::Bytes) \
|
||||
M(DistrCacheRegistryUpdateMicroseconds, "Time spent updating distributed cache registry", ValueType::Microseconds) \
|
||||
M(DistrCacheRegistryUpdates, "Number of distributed cache registry updates", ValueType::Number) \
|
||||
M(DistrCacheGetResponseMicroseconds, "Distributed Cache client event. Time spend to wait for response from distributed cache", ValueType::Microseconds) \
|
||||
M(DistrCacheReadErrors, "Distributed Cache client event. Number of distributed cache errors during read", ValueType::Number) \
|
||||
M(DistrCacheMakeRequestErrors, "Distributed Cache client event. Number of distributed cache errors when making a request", ValueType::Number) \
|
||||
M(DistrCacheReceiveResponseErrors, "Distributed Cache client event. Number of distributed cache errors when receiving response a request", ValueType::Number) \
|
||||
\
|
||||
M(DistrCacheConnectMicroseconds, "The time spent to connect to distributed cache", ValueType::Microseconds) \
|
||||
M(DistrCacheConnectAttempts, "The number of connection attempts to distributed cache", ValueType::Number) \
|
||||
M(DistrCacheGetClient, "Number of client access times", ValueType::Number) \
|
||||
M(DistrCachePackets, "Distributed Cache client event. Total number of packets received from distributed cache", ValueType::Number) \
|
||||
M(DistrCachePacketsBytes, "Distributed Cache client event. The number of bytes in Data packets which were not ignored", ValueType::Bytes) \
|
||||
M(DistrCacheUnusedPackets, "Distributed Cache client event. Number of skipped unused packets from distributed cache", ValueType::Number) \
|
||||
M(DistrCacheUnusedPacketsBytes, "Distributed Cache client event. The number of bytes in Data packets which were ignored", ValueType::Bytes) \
|
||||
M(DistrCacheUnusedPacketsBufferAllocations, "Distributed Cache client event. The number of extra buffer allocations in case we could not reuse existing buffer", ValueType::Number) \
|
||||
\
|
||||
M(DistrCacheServerProcessRequestMicroseconds, "Time spent processing request on DistributedCache server side", ValueType::Microseconds) \
|
||||
M(DistrCacheLockRegistryMicroseconds, "Distributed Cache registry event. Time spent to take DistributedCacheRegistry lock", ValueType::Microseconds) \
|
||||
M(DistrCacheRegistryUpdateMicroseconds, "Distributed Cache registry event. Time spent updating distributed cache registry", ValueType::Microseconds) \
|
||||
M(DistrCacheRegistryUpdates, "Distributed Cache registry event. Number of distributed cache registry updates", ValueType::Number) \
|
||||
M(DistrCacheHashRingRebuilds, "Distributed Cache registry event. Number of distributed cache hash ring rebuilds", ValueType::Number) \
|
||||
\
|
||||
M(DistrCacheReadBytesFromCache, "Distributed Cache read buffer event. Bytes read from distributed cache", ValueType::Bytes) \
|
||||
M(DistrCacheReadBytesFromFallbackBuffer, "Distributed Cache read buffer event. Bytes read from fallback buffer", ValueType::Number) \
|
||||
\
|
||||
M(DistrCacheRangeResetBackward, "Distributed Cache read buffer event. Number of times we reset read range because of seek/last_position change", ValueType::Number) \
|
||||
M(DistrCacheRangeResetForward, "Distributed Cache read buffer event. Number of times we reset read range because of seek/last_position change", ValueType::Number) \
|
||||
\
|
||||
M(DistrCacheOpenedConnections, "Distributed Cache connection event. The number of open connections to distributed cache", ValueType::Number) \
|
||||
M(DistrCacheReusedConnections, "Distributed Cache connection event. The number of reused connections to distributed cache", ValueType::Number) \
|
||||
M(DistrCacheOpenedConnectionsBypassingPool, "Distributed Cache connection event. The number of open connections to distributed cache bypassing pool", ValueType::Number) \
|
||||
M(DistrCacheConnectMicroseconds, "Distributed Cache connection event. The time spent to connect to distributed cache", ValueType::Microseconds) \
|
||||
M(DistrCacheConnectAttempts, "Distributed Cache connection event. The number of connection attempts to distributed cache", ValueType::Number) \
|
||||
M(DistrCacheGetClientMicroseconds, "Distributed Cache connection event. Time spent getting client for distributed cache", ValueType::Microseconds) \
|
||||
\
|
||||
M(DistrCacheServerProcessRequestMicroseconds, "Distributed Cache server event. Time spent processing request on DistributedCache server side", ValueType::Microseconds) \
|
||||
M(DistrCacheServerStartRequestPackets, "Distributed Cache server event. Number of StartRequest packets in DistributedCacheServer", ValueType::Number) \
|
||||
M(DistrCacheServerContinueRequestPackets, "Distributed Cache server event. Number of ContinueRequest packets in DistributedCacheServer", ValueType::Number) \
|
||||
M(DistrCacheServerEndRequestPackets, "Distributed Cache server event. Number of EndRequest packets in DistributedCacheServer", ValueType::Number) \
|
||||
M(DistrCacheServerAckRequestPackets, "Distributed Cache server event. Number of AckRequest packets in DistributedCacheServer", ValueType::Number) \
|
||||
M(DistrCacheServerNewS3CachedClients, "Distributed Cache server event. The number of new cached s3 clients", ValueType::Number) \
|
||||
M(DistrCacheServerReusedS3CachedClients, "Distributed Cache server event. The number of reused cached s3 clients", ValueType::Number) \
|
||||
\
|
||||
M(LogTest, "Number of log messages with level Test", ValueType::Number) \
|
||||
M(LogTrace, "Number of log messages with level Trace", ValueType::Number) \
|
||||
@ -788,15 +816,38 @@ The server successfully detected this situation and will download merged part fr
|
||||
M(InterfacePostgreSQLReceiveBytes, "Number of bytes received through PostgreSQL interfaces", ValueType::Bytes) \
|
||||
\
|
||||
M(ParallelReplicasUsedCount, "Number of replicas used to execute a query with task-based parallel replicas", ValueType::Number) \
|
||||
M(ParallelReplicasAvailableCount, "Number of replicas available to execute a query with task-based parallel replicas", ValueType::Number) \
|
||||
M(ParallelReplicasUnavailableCount, "Number of replicas which was chosen, but found to be unavailable during query execution with task-based parallel replicas", ValueType::Number) \
|
||||
\
|
||||
M(SharedMergeTreeVirtualPartsUpdates, "Virtual parts update count", ValueType::Number) \
|
||||
M(SharedMergeTreeVirtualPartsUpdatesByLeader, "Virtual parts updates by leader", ValueType::Number) \
|
||||
M(SharedMergeTreeVirtualPartsUpdateMicroseconds, "Virtual parts update microseconds", ValueType::Microseconds) \
|
||||
M(SharedMergeTreeVirtualPartsUpdatesFromZooKeeper, "Virtual parts updates count from ZooKeeper", ValueType::Number) \
|
||||
M(SharedMergeTreeVirtualPartsUpdatesFromZooKeeperMicroseconds, "Virtual parts updates from ZooKeeper microseconds", ValueType::Microseconds) \
|
||||
M(SharedMergeTreeVirtualPartsUpdatesPeerNotFound, "Virtual updates from peer failed because no one found", ValueType::Number) \
|
||||
M(SharedMergeTreeVirtualPartsUpdatesFromPeer, "Virtual parts updates count from peer", ValueType::Number) \
|
||||
M(SharedMergeTreeVirtualPartsUpdatesFromPeerMicroseconds, "Virtual parts updates from peer microseconds", ValueType::Microseconds) \
|
||||
M(SharedMergeTreeVirtualPartsUpdatesForMergesOrStatus, "Virtual parts updates from non-default background job", ValueType::Number) \
|
||||
M(SharedMergeTreeVirtualPartsUpdatesLeaderFailedElection, "Virtual parts updates leader election failed", ValueType::Number) \
|
||||
M(SharedMergeTreeVirtualPartsUpdatesLeaderSuccessfulElection, "Virtual parts updates leader election successful", ValueType::Number) \
|
||||
M(SharedMergeTreeMergeMutationAssignmentAttempt, "How many times we tried to assign merge or mutation", ValueType::Number) \
|
||||
M(SharedMergeTreeMergeMutationAssignmentFailedWithNothingToDo, "How many times we tried to assign merge or mutation and failed because nothing to merge", ValueType::Number) \
|
||||
M(SharedMergeTreeMergeMutationAssignmentFailedWithConflict, "How many times we tried to assign merge or mutation and failed because of conflict in Keeper", ValueType::Number) \
|
||||
M(SharedMergeTreeMergeMutationAssignmentSuccessful, "How many times we tried to assign merge or mutation", ValueType::Number) \
|
||||
M(SharedMergeTreeMergePartsMovedToOudated, "How many parts moved to oudated directory", ValueType::Number) \
|
||||
M(SharedMergeTreeMergePartsMovedToCondemned, "How many parts moved to condemned directory", ValueType::Number) \
|
||||
M(SharedMergeTreeOutdatedPartsConfirmationRequest, "How many ZooKeeper requests were used to config outdated parts", ValueType::Number) \
|
||||
M(SharedMergeTreeOutdatedPartsConfirmationInvocations, "How many invocations were made to confirm outdated parts", ValueType::Number) \
|
||||
M(SharedMergeTreeOutdatedPartsHTTPRequest, "How many HTTP requests were send to confirm outdated parts", ValueType::Number) \
|
||||
M(SharedMergeTreeOutdatedPartsHTTPResponse, "How many HTTP responses were send to confirm outdated parts", ValueType::Number) \
|
||||
M(SharedMergeTreeCondemnedPartsKillRequest, "How many ZooKeeper requests were used to remove condemned parts", ValueType::Number) \
|
||||
M(SharedMergeTreeCondemnedPartsLockConfict, "How many times we failed to acquite lock because of conflict", ValueType::Number) \
|
||||
M(SharedMergeTreeCondemnedPartsRemoved, "How many condemned parts were removed", ValueType::Number) \
|
||||
M(KeeperLogsEntryReadFromLatestCache, "Number of log entries in Keeper being read from latest logs cache", ValueType::Number) \
|
||||
M(KeeperLogsEntryReadFromCommitCache, "Number of log entries in Keeper being read from commit logs cache", ValueType::Number) \
|
||||
M(KeeperLogsEntryReadFromFile, "Number of log entries in Keeper being read directly from the changelog file", ValueType::Number) \
|
||||
M(KeeperLogsPrefetchedEntries, "Number of log entries in Keeper being prefetched from the changelog file", ValueType::Number) \
|
||||
\
|
||||
M(ParallelReplicasAvailableCount, "Number of replicas available to execute a query with task-based parallel replicas", ValueType::Number) \
|
||||
M(ParallelReplicasUnavailableCount, "Number of replicas which was chosen, but found to be unavailable during query execution with task-based parallel replicas", ValueType::Number) \
|
||||
\
|
||||
M(StorageConnectionsCreated, "Number of created connections for storages", ValueType::Number) \
|
||||
M(StorageConnectionsReused, "Number of reused connections for storages", ValueType::Number) \
|
||||
M(StorageConnectionsReset, "Number of reset connections for storages", ValueType::Number) \
|
||||
@ -828,6 +879,9 @@ The server successfully detected this situation and will download merged part fr
|
||||
M(ReadWriteBufferFromHTTPRequestsSent, "Number of HTTP requests sent by ReadWriteBufferFromHTTP", ValueType::Number) \
|
||||
M(ReadWriteBufferFromHTTPBytes, "Total size of payload bytes received and sent by ReadWriteBufferFromHTTP. Doesn't include HTTP headers.", ValueType::Bytes) \
|
||||
\
|
||||
M(SharedDatabaseCatalogFailedToApplyState, "Number of failures to apply new state in SharedDatabaseCatalog", ValueType::Number) \
|
||||
M(SharedDatabaseCatalogStateApplicationMicroseconds, "Total time spend on application of new state in SharedDatabaseCatalog", ValueType::Microseconds) \
|
||||
\
|
||||
M(GWPAsanAllocateSuccess, "Number of successful allocations done by GWPAsan", ValueType::Number) \
|
||||
M(GWPAsanAllocateFailed, "Number of failed allocations done by GWPAsan (i.e. filled pool)", ValueType::Number) \
|
||||
M(GWPAsanFree, "Number of free operations done by GWPAsan", ValueType::Number) \
|
||||
|
||||
@ -38,6 +38,9 @@ namespace ProfileEvents
|
||||
};
|
||||
Timer(Counters & counters_, Event timer_event_, Resolution resolution_);
|
||||
Timer(Counters & counters_, Event timer_event_, Event counter_event, Resolution resolution_);
|
||||
Timer(Timer && other) noexcept
|
||||
: counters(other.counters), timer_event(std::move(other.timer_event)), watch(std::move(other.watch)), resolution(std::move(other.resolution))
|
||||
{}
|
||||
~Timer() { end(); }
|
||||
void cancel() { watch.reset(); }
|
||||
void restart() { watch.restart(); }
|
||||
|
||||
@ -31,7 +31,7 @@ std::string RemoteProxyHostFetcherImpl::fetch(const Poco::URI & endpoint, const
|
||||
endpoint.toString(),
|
||||
response.getStatus(),
|
||||
response.getReason(),
|
||||
"");
|
||||
/* body_length = */ 0);
|
||||
|
||||
std::string proxy_host;
|
||||
Poco::StreamCopier::copyToString(response_body_stream, proxy_host);
|
||||
|
||||
@ -159,6 +159,8 @@ public:
|
||||
const std::string & getLastKeeperErrorMessage() const { return keeper_error.message; }
|
||||
|
||||
/// action will be called only once and only after latest failed retry
|
||||
/// NOTE: this one will be called only in case when retries finishes with Keeper exception
|
||||
/// if it will be some other exception this function will not be called.
|
||||
void actionAfterLastFailedRetry(std::function<void()> f) { action_after_last_failed_retry = std::move(f); }
|
||||
|
||||
const std::string & getName() const { return name; }
|
||||
|
||||
@ -25,15 +25,11 @@ namespace
|
||||
* `curl` strips leading dot and accepts url gitlab.com as a match for no_proxy .gitlab.com,
|
||||
* while `wget` does an exact match.
|
||||
* */
|
||||
std::string buildPocoRegexpEntryWithoutLeadingDot(const std::string & host)
|
||||
std::string buildPocoRegexpEntryWithoutLeadingDot(std::string_view host)
|
||||
{
|
||||
std::string_view view_without_leading_dot = host;
|
||||
if (host[0] == '.')
|
||||
{
|
||||
view_without_leading_dot = std::string_view {host.begin() + 1u, host.end()};
|
||||
}
|
||||
|
||||
return RE2::QuoteMeta(view_without_leading_dot);
|
||||
if (host.starts_with('.'))
|
||||
host.remove_prefix(1);
|
||||
return RE2::QuoteMeta(host);
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
@ -1890,7 +1890,7 @@ void Changelog::removeExistingLogs(ChangelogIter begin, ChangelogIter end)
|
||||
{
|
||||
auto & changelog_description = itr->second;
|
||||
|
||||
if (!disk->exists(timestamp_folder))
|
||||
if (!disk->existsDirectory(timestamp_folder))
|
||||
{
|
||||
LOG_WARNING(log, "Moving broken logs to {}", timestamp_folder);
|
||||
disk->createDirectories(timestamp_folder);
|
||||
|
||||
@ -964,7 +964,7 @@ static uint64_t getTotalSize(const DiskPtr & disk, const std::string & path = ""
|
||||
uint64_t size = 0;
|
||||
for (auto it = disk->iterateDirectory(path); it->isValid(); it->next())
|
||||
{
|
||||
if (disk->isFile(it->path()))
|
||||
if (disk->existsFile(it->path()))
|
||||
size += disk->getFileSize(it->path());
|
||||
else
|
||||
size += getTotalSize(disk, it->path());
|
||||
|
||||
@ -913,7 +913,7 @@ SnapshotFileInfoPtr KeeperSnapshotManager<Storage>::getLatestSnapshotInfo() cons
|
||||
|
||||
try
|
||||
{
|
||||
if (disk->exists(path))
|
||||
if (disk->existsFile(path))
|
||||
return std::make_shared<SnapshotFileInfo>(path, disk);
|
||||
}
|
||||
catch (...)
|
||||
|
||||
@ -334,7 +334,7 @@ void KeeperStateManager::save_state(const nuraft::srv_state & state)
|
||||
|
||||
auto disk = getStateFileDisk();
|
||||
|
||||
if (disk->exists(server_state_file_name))
|
||||
if (disk->existsFile(server_state_file_name))
|
||||
{
|
||||
auto buf = disk->writeFile(copy_lock_file);
|
||||
buf->finalize();
|
||||
@ -422,7 +422,7 @@ nuraft::ptr<nuraft::srv_state> KeeperStateManager::read_state()
|
||||
}
|
||||
};
|
||||
|
||||
if (disk->exists(server_state_file_name))
|
||||
if (disk->existsFile(server_state_file_name))
|
||||
{
|
||||
auto state = try_read_file(server_state_file_name);
|
||||
|
||||
@ -435,9 +435,9 @@ nuraft::ptr<nuraft::srv_state> KeeperStateManager::read_state()
|
||||
disk->removeFile(server_state_file_name);
|
||||
}
|
||||
|
||||
if (disk->exists(old_path))
|
||||
if (disk->existsFile(old_path))
|
||||
{
|
||||
if (disk->exists(copy_lock_file))
|
||||
if (disk->existsFile(copy_lock_file))
|
||||
{
|
||||
disk->removeFile(old_path);
|
||||
disk->removeFile(copy_lock_file);
|
||||
@ -453,7 +453,7 @@ nuraft::ptr<nuraft::srv_state> KeeperStateManager::read_state()
|
||||
disk->removeFile(old_path);
|
||||
}
|
||||
}
|
||||
else if (disk->exists(copy_lock_file))
|
||||
else if (disk->existsFile(copy_lock_file))
|
||||
{
|
||||
disk->removeFile(copy_lock_file);
|
||||
}
|
||||
|
||||
@ -170,6 +170,9 @@ Avoid reordering rows when reading from Parquet files. Usually makes it much slo
|
||||
)", 0) \
|
||||
M(Bool, input_format_parquet_filter_push_down, true, R"(
|
||||
When reading Parquet files, skip whole row groups based on the WHERE/PREWHERE expressions and min/max statistics in the Parquet metadata.
|
||||
)", 0) \
|
||||
M(Bool, input_format_parquet_bloom_filter_push_down, false, R"(
|
||||
When reading Parquet files, skip whole row groups based on the WHERE expressions and bloom filter in the Parquet metadata.
|
||||
)", 0) \
|
||||
M(Bool, input_format_parquet_use_native_reader, false, R"(
|
||||
When reading Parquet files, to use native reader instead of arrow reader.
|
||||
@ -190,6 +193,9 @@ When reading ORC files, skip whole stripes or row groups based on the WHERE/PREW
|
||||
)", 0) \
|
||||
M(String, input_format_orc_reader_time_zone_name, "GMT", R"(
|
||||
The time zone name for ORC row reader, the default ORC row reader's time zone is GMT.
|
||||
)", 0) \
|
||||
M(Bool, input_format_orc_dictionary_as_low_cardinality, true, R"(
|
||||
Treat ORC dictionary encoded columns as LowCardinality columns while reading ORC files.
|
||||
)", 0) \
|
||||
M(Bool, input_format_parquet_allow_missing_columns, true, R"(
|
||||
Allow missing columns while reading Parquet input formats
|
||||
@ -604,6 +610,9 @@ See also:
|
||||
- [Interval](../../sql-reference/data-types/special-data-types/interval.md)
|
||||
)", 0) \
|
||||
\
|
||||
M(Bool, date_time_64_output_format_cut_trailing_zeros_align_to_groups_of_thousands, false, R"(
|
||||
Dynamically trim the trailing zeros of datetime64 values to adjust the output scale to [0, 3, 6],
|
||||
corresponding to 'seconds', 'milliseconds', and 'microseconds')", 0) \
|
||||
M(Bool, input_format_ipv4_default_on_conversion_error, false, R"(
|
||||
Deserialization of IPv4 will use default values instead of throwing exception on conversion error.
|
||||
|
||||
|
||||
13
src/Core/MergeSelectorAlgorithm.h
Normal file
13
src/Core/MergeSelectorAlgorithm.h
Normal file
@ -0,0 +1,13 @@
|
||||
#pragma once
|
||||
|
||||
#include <cstdint>
|
||||
namespace DB
|
||||
{
|
||||
|
||||
enum class MergeSelectorAlgorithm : uint8_t
|
||||
{
|
||||
SIMPLE,
|
||||
STOCHASTIC_SIMPLE,
|
||||
};
|
||||
|
||||
}
|
||||
@ -700,6 +700,9 @@ Move more conditions from WHERE to PREWHERE and do reads from disk and filtering
|
||||
)", 0) \
|
||||
M(Bool, move_primary_key_columns_to_end_of_prewhere, true, R"(
|
||||
Move PREWHERE conditions containing primary key columns to the end of AND chain. It is likely that these conditions are taken into account during primary key analysis and thus will not contribute a lot to PREWHERE filtering.
|
||||
)", 0) \
|
||||
M(Bool, allow_reorder_prewhere_conditions, true, R"(
|
||||
When moving conditions from WHERE to PREWHERE, allow reordering them to optimize filtering
|
||||
)", 0) \
|
||||
\
|
||||
M(UInt64, alter_sync, 1, R"(
|
||||
@ -2700,7 +2703,7 @@ The maximum read speed in bytes per second for particular backup on server. Zero
|
||||
Log query performance statistics into the query_log, query_thread_log and query_views_log.
|
||||
)", 0) \
|
||||
M(Bool, log_query_settings, true, R"(
|
||||
Log query settings into the query_log.
|
||||
Log query settings into the query_log and OpenTelemetry span log.
|
||||
)", 0) \
|
||||
M(Bool, log_query_threads, false, R"(
|
||||
Setting up query threads logging.
|
||||
@ -4812,6 +4815,9 @@ Max attempts to read with backoff
|
||||
)", 0) \
|
||||
M(Bool, enable_filesystem_cache, true, R"(
|
||||
Use cache for remote filesystem. This setting does not turn on/off cache for disks (must be done via disk config), but allows to bypass cache for some queries if intended
|
||||
)", 0) \
|
||||
M(String, filesystem_cache_name, "", R"(
|
||||
Filesystem cache name to use for stateless table engines or data lakes
|
||||
)", 0) \
|
||||
M(Bool, enable_filesystem_cache_on_write_operations, false, R"(
|
||||
Write into cache on write operations. To actually work this setting requires be added to disk config too
|
||||
@ -5151,7 +5157,7 @@ SELECT * FROM test_table
|
||||
Rewrite count distinct to subquery of group by
|
||||
)", 0) \
|
||||
M(Bool, throw_if_no_data_to_insert, true, R"(
|
||||
Allows or forbids empty INSERTs, enabled by default (throws an error on an empty insert)
|
||||
Allows or forbids empty INSERTs, enabled by default (throws an error on an empty insert). Only applies to INSERTs using [`clickhouse-client`](/docs/en/interfaces/cli) or using the [gRPC interface](/docs/en/interfaces/grpc).
|
||||
)", 0) \
|
||||
M(Bool, compatibility_ignore_auto_increment_in_create_table, false, R"(
|
||||
Ignore AUTO_INCREMENT keyword in column declaration if true, otherwise return error. It simplifies migration from MySQL
|
||||
@ -5376,7 +5382,7 @@ Result:
|
||||
If enabled, server will ignore all DROP table queries with specified probability (for Memory and JOIN engines it will replcase DROP to TRUNCATE). Used for testing purposes
|
||||
)", 0) \
|
||||
M(Bool, traverse_shadow_remote_data_paths, false, R"(
|
||||
Traverse shadow directory when query system.remote_data_paths
|
||||
Traverse frozen data (shadow directory) in addition to actual table data when query system.remote_data_paths
|
||||
)", 0) \
|
||||
M(Bool, geo_distance_returns_float64_on_float64_arguments, true, R"(
|
||||
If all four arguments to `geoDistance`, `greatCircleDistance`, `greatCircleAngle` functions are Float64, return Float64 and use double precision for internal calculations. In previous ClickHouse versions, the functions always returned Float32.
|
||||
@ -5498,8 +5504,8 @@ Replace external dictionary sources to Null on restore. Useful for testing purpo
|
||||
M(Bool, create_if_not_exists, false, R"(
|
||||
Enable `IF NOT EXISTS` for `CREATE` statement by default. If either this setting or `IF NOT EXISTS` is specified and a table with the provided name already exists, no exception will be thrown.
|
||||
)", 0) \
|
||||
M(Bool, enable_secure_identifiers, false, R"(
|
||||
If enabled, only allow secure identifiers which contain only underscore and alphanumeric characters
|
||||
M(Bool, enforce_strict_identifier_format, false, R"(
|
||||
If enabled, only allow identifiers containing alphanumeric characters and underscores.
|
||||
)", 0) \
|
||||
M(Bool, mongodb_throw_on_unsupported_query, true, R"(
|
||||
If enabled, MongoDB tables will return an error when a MongoDB query cannot be built. Otherwise, ClickHouse reads the full table and processes it locally. This option does not apply to the legacy implementation or when 'allow_experimental_analyzer=0'.
|
||||
@ -6199,6 +6205,16 @@ std::vector<std::string_view> Settings::getUnchangedNames() const
|
||||
return setting_names;
|
||||
}
|
||||
|
||||
std::vector<std::string_view> Settings::getChangedNames() const
|
||||
{
|
||||
std::vector<std::string_view> setting_names;
|
||||
for (const auto & setting : impl->allChanged())
|
||||
{
|
||||
setting_names.emplace_back(setting.getName());
|
||||
}
|
||||
return setting_names;
|
||||
}
|
||||
|
||||
void Settings::dumpToSystemSettingsColumns(MutableColumnsAndConstraints & params) const
|
||||
{
|
||||
MutableColumns & res_columns = params.res_columns;
|
||||
|
||||
@ -134,6 +134,7 @@ struct Settings
|
||||
std::vector<std::string_view> getAllRegisteredNames() const;
|
||||
std::vector<std::string_view> getChangedAndObsoleteNames() const;
|
||||
std::vector<std::string_view> getUnchangedNames() const;
|
||||
std::vector<std::string_view> getChangedNames() const;
|
||||
|
||||
void dumpToSystemSettingsColumns(MutableColumnsAndConstraints & params) const;
|
||||
void dumpToMapColumn(IColumn * column, bool changed_only = true) const;
|
||||
|
||||
@ -69,17 +69,18 @@ static std::initializer_list<std::pair<ClickHouseVersion, SettingsChangesHistory
|
||||
{"24.10",
|
||||
{
|
||||
{"check_table_structure_completely", true, false, "Add new setting to allow attach when source table's projections and secondary indices is a subset of those in the target table."},
|
||||
{"enforce_strict_identifier_format", false, false, "New setting."},
|
||||
{"enable_parsing_to_custom_serialization", false, true, "New setting"},
|
||||
{"mongodb_throw_on_unsupported_query", false, true, "New setting."},
|
||||
{"enable_parallel_replicas", false, false, "Parallel replicas with read tasks became the Beta tier feature."},
|
||||
{"parallel_replicas_mode", "read_tasks", "read_tasks", "This setting was introduced as a part of making parallel replicas feature Beta"},
|
||||
{"filesystem_cache_name", "", "", "Filesystem cache name to use for stateless table engines or data lakes"},
|
||||
{"restore_replace_external_dictionary_source_to_null", false, false, "New setting."},
|
||||
{"show_create_query_identifier_quoting_rule", "when_necessary", "when_necessary", "New setting."},
|
||||
{"show_create_query_identifier_quoting_style", "Backticks", "Backticks", "New setting."},
|
||||
{"output_format_native_write_json_as_string", false, false, "Add new setting to allow write JSON column as single String column in Native format"},
|
||||
{"output_format_binary_write_json_as_string", false, false, "Add new setting to write values of JSON type as JSON string in RowBinary output format"},
|
||||
{"input_format_binary_read_json_as_string", false, false, "Add new setting to read values of JSON type as JSON string in RowBinary input format"},
|
||||
{"enable_secure_identifiers", false, false, "New setting."},
|
||||
{"min_free_disk_bytes_to_perform_insert", 0, 0, "New setting."},
|
||||
{"min_free_disk_ratio_to_perform_insert", 0.0, 0.0, "New setting."},
|
||||
{"cloud_mode_database_engine", 1, 1, "A setting for ClickHouse Cloud"},
|
||||
@ -98,8 +99,12 @@ static std::initializer_list<std::pair<ClickHouseVersion, SettingsChangesHistory
|
||||
{"distributed_cache_read_alignment", 0, 0, "A setting for ClickHouse Cloud"},
|
||||
{"distributed_cache_max_unacked_inflight_packets", 10, 10, "A setting for ClickHouse Cloud"},
|
||||
{"distributed_cache_data_packet_ack_window", 5, 5, "A setting for ClickHouse Cloud"},
|
||||
{"input_format_orc_dictionary_as_low_cardinality", false, true, "Treat ORC dictionary encoded columns as LowCardinality columns while reading ORC files"},
|
||||
{"allow_experimental_refreshable_materialized_view", false, true, "Not experimental anymore"},
|
||||
{"max_parts_to_move", 1000, 1000, "New setting"},
|
||||
{"allow_reorder_prewhere_conditions", false, true, "New setting"},
|
||||
{"input_format_parquet_bloom_filter_push_down", false, true, "When reading Parquet files, skip whole row groups based on the WHERE/PREWHERE expressions and bloom filter in the Parquet metadata."},
|
||||
{"date_time_64_output_format_cut_trailing_zeros_align_to_groups_of_thousands", false, false, "Dynamically trim the trailing zeros of datetime64 values to adjust the output scale to (0, 3, 6), corresponding to 'seconds', 'milliseconds', and 'microseconds'."}
|
||||
}
|
||||
},
|
||||
{"24.9",
|
||||
@ -112,7 +117,7 @@ static std::initializer_list<std::pair<ClickHouseVersion, SettingsChangesHistory
|
||||
{"allow_materialized_view_with_bad_select", true, true, "Support (but not enable yet) stricter validation in CREATE MATERIALIZED VIEW"},
|
||||
{"parallel_replicas_mark_segment_size", 128, 0, "Value for this setting now determined automatically"},
|
||||
{"database_replicated_allow_replicated_engine_arguments", 1, 0, "Don't allow explicit arguments by default"},
|
||||
{"database_replicated_allow_explicit_uuid", 0, 0, "Added a new setting to disallow explicitly specifying table UUID"},
|
||||
{"database_replicated_allow_explicit_uuid", 1, 0, "Added a new setting to disallow explicitly specifying table UUID"},
|
||||
{"parallel_replicas_local_plan", false, false, "Use local plan for local replica in a query with parallel replicas"},
|
||||
{"join_to_sort_minimum_perkey_rows", 0, 40, "The lower limit of per-key average rows in the right table to determine whether to rerange the right table by key in left or inner join. This setting ensures that the optimization is not applied for sparse table keys"},
|
||||
{"join_to_sort_maximum_table_rows", 0, 10000, "The maximum number of rows in the right table to determine whether to rerange the right table by key in left or inner join"},
|
||||
|
||||
@ -1,7 +1,6 @@
|
||||
#include <Core/SettingsEnums.h>
|
||||
#include <magic_enum.hpp>
|
||||
#include <Access/Common/SQLSecurityDefs.h>
|
||||
|
||||
#include <boost/range/adaptor/map.hpp>
|
||||
|
||||
|
||||
@ -273,4 +272,11 @@ IMPLEMENT_SETTING_ENUM(
|
||||
{{"user_display", IdentifierQuotingRule::UserDisplay},
|
||||
{"when_necessary", IdentifierQuotingRule::WhenNecessary},
|
||||
{"always", IdentifierQuotingRule::Always}})
|
||||
|
||||
IMPLEMENT_SETTING_ENUM(
|
||||
MergeSelectorAlgorithm,
|
||||
ErrorCodes::BAD_ARGUMENTS,
|
||||
{{"Simple", MergeSelectorAlgorithm::SIMPLE},
|
||||
{"StochasticSimple", MergeSelectorAlgorithm::STOCHASTIC_SIMPLE}})
|
||||
|
||||
}
|
||||
|
||||
@ -14,6 +14,7 @@
|
||||
#include <Parsers/IdentifierQuotingStyle.h>
|
||||
#include <QueryPipeline/SizeLimits.h>
|
||||
#include <Common/ShellCommandSettings.h>
|
||||
#include <Core/MergeSelectorAlgorithm.h>
|
||||
|
||||
|
||||
namespace DB
|
||||
@ -363,4 +364,6 @@ enum class GroupArrayActionWhenLimitReached : uint8_t
|
||||
};
|
||||
DECLARE_SETTING_ENUM(GroupArrayActionWhenLimitReached)
|
||||
|
||||
DECLARE_SETTING_ENUM(MergeSelectorAlgorithm)
|
||||
|
||||
}
|
||||
|
||||
@ -36,8 +36,8 @@ public:
|
||||
|
||||
auto findByValue(const T & value) const
|
||||
{
|
||||
const auto it = value_to_name_map.find(value);
|
||||
if (it == std::end(value_to_name_map))
|
||||
auto it = value_to_name_map.find(value);
|
||||
if (it == value_to_name_map.end())
|
||||
throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unexpected value {} in enum", toString(value));
|
||||
|
||||
return it;
|
||||
@ -58,7 +58,7 @@ public:
|
||||
bool getNameForValue(const T & value, StringRef & result) const
|
||||
{
|
||||
const auto it = value_to_name_map.find(value);
|
||||
if (it == std::end(value_to_name_map))
|
||||
if (it == value_to_name_map.end())
|
||||
return false;
|
||||
|
||||
result = it->second;
|
||||
|
||||
@ -321,6 +321,8 @@ bool isUInt8(TYPE data_type) { return WhichDataType(data_type).isUInt8(); } \
|
||||
bool isUInt16(TYPE data_type) { return WhichDataType(data_type).isUInt16(); } \
|
||||
bool isUInt32(TYPE data_type) { return WhichDataType(data_type).isUInt32(); } \
|
||||
bool isUInt64(TYPE data_type) { return WhichDataType(data_type).isUInt64(); } \
|
||||
bool isUInt128(TYPE data_type) { return WhichDataType(data_type).isUInt128(); } \
|
||||
bool isUInt256(TYPE data_type) { return WhichDataType(data_type).isUInt256(); } \
|
||||
bool isNativeUInt(TYPE data_type) { return WhichDataType(data_type).isNativeUInt(); } \
|
||||
bool isUInt(TYPE data_type) { return WhichDataType(data_type).isUInt(); } \
|
||||
\
|
||||
@ -328,6 +330,8 @@ bool isInt8(TYPE data_type) { return WhichDataType(data_type).isInt8(); } \
|
||||
bool isInt16(TYPE data_type) { return WhichDataType(data_type).isInt16(); } \
|
||||
bool isInt32(TYPE data_type) { return WhichDataType(data_type).isInt32(); } \
|
||||
bool isInt64(TYPE data_type) { return WhichDataType(data_type).isInt64(); } \
|
||||
bool isInt128(TYPE data_type) { return WhichDataType(data_type).isInt128(); } \
|
||||
bool isInt256(TYPE data_type) { return WhichDataType(data_type).isInt256(); } \
|
||||
bool isNativeInt(TYPE data_type) { return WhichDataType(data_type).isNativeInt(); } \
|
||||
bool isInt(TYPE data_type) { return WhichDataType(data_type).isInt(); } \
|
||||
\
|
||||
|
||||
@ -457,7 +457,9 @@ struct WhichDataType
|
||||
bool isUInt8(TYPE data_type); \
|
||||
bool isUInt16(TYPE data_type); \
|
||||
bool isUInt32(TYPE data_type); \
|
||||
bool isUInt64(TYPE data_type); \
|
||||
bool isUInt64(TYPE data_type);\
|
||||
bool isUInt128(TYPE data_type);\
|
||||
bool isUInt256(TYPE data_type); \
|
||||
bool isNativeUInt(TYPE data_type); \
|
||||
bool isUInt(TYPE data_type); \
|
||||
\
|
||||
@ -465,6 +467,8 @@ bool isInt8(TYPE data_type); \
|
||||
bool isInt16(TYPE data_type); \
|
||||
bool isInt32(TYPE data_type); \
|
||||
bool isInt64(TYPE data_type); \
|
||||
bool isInt128(TYPE data_type); \
|
||||
bool isInt256(TYPE data_type); \
|
||||
bool isNativeInt(TYPE data_type); \
|
||||
bool isInt(TYPE data_type); \
|
||||
\
|
||||
|
||||
@ -26,7 +26,10 @@ void SerializationDateTime64::serializeText(const IColumn & column, size_t row_n
|
||||
switch (settings.date_time_output_format)
|
||||
{
|
||||
case FormatSettings::DateTimeOutputFormat::Simple:
|
||||
writeDateTimeText(value, scale, ostr, time_zone);
|
||||
if (settings.date_time_64_output_format_cut_trailing_zeros_align_to_groups_of_thousands)
|
||||
writeDateTimeTextCutTrailingZerosAlignToGroupOfThousands(value, scale, ostr, time_zone);
|
||||
else
|
||||
writeDateTimeText(value, scale, ostr, time_zone);
|
||||
return;
|
||||
case FormatSettings::DateTimeOutputFormat::UnixTimestamp:
|
||||
writeDateTimeUnixTimestamp(value, scale, ostr);
|
||||
|
||||
@ -370,7 +370,7 @@ void DatabaseOnDisk::dropTable(ContextPtr local_context, const String & table_na
|
||||
|
||||
for (const auto & [disk_name, disk] : getContext()->getDisksMap())
|
||||
{
|
||||
if (disk->isReadOnly() || !disk->exists(table_data_path_relative))
|
||||
if (disk->isReadOnly() || !disk->existsDirectory(table_data_path_relative))
|
||||
continue;
|
||||
|
||||
LOG_INFO(log, "Removing data directory from disk {} with path {} for dropped table {} ", disk_name, table_data_path_relative, table_name);
|
||||
|
||||
@ -31,22 +31,22 @@ public:
|
||||
|
||||
ReservationPtr reserve(UInt64 bytes) override;
|
||||
|
||||
bool exists(const String & path) const override
|
||||
bool existsFile(const String & path) const override
|
||||
{
|
||||
auto wrapped_path = wrappedPath(path);
|
||||
return delegate->exists(wrapped_path);
|
||||
return delegate->existsFile(wrapped_path);
|
||||
}
|
||||
|
||||
bool isFile(const String & path) const override
|
||||
bool existsDirectory(const String & path) const override
|
||||
{
|
||||
auto wrapped_path = wrappedPath(path);
|
||||
return delegate->isFile(wrapped_path);
|
||||
return delegate->existsDirectory(wrapped_path);
|
||||
}
|
||||
|
||||
bool isDirectory(const String & path) const override
|
||||
bool existsFileOrDirectory(const String & path) const override
|
||||
{
|
||||
auto wrapped_path = wrappedPath(path);
|
||||
return delegate->isDirectory(wrapped_path);
|
||||
return delegate->existsFileOrDirectory(wrapped_path);
|
||||
}
|
||||
|
||||
size_t getFileSize(const String & path) const override;
|
||||
|
||||
@ -71,7 +71,7 @@ std::unique_ptr<WriteBufferFromFileBase> DiskEncryptedTransaction::writeFile( //
|
||||
FileEncryption::Header header;
|
||||
String key;
|
||||
UInt64 old_file_size = 0;
|
||||
if (mode == WriteMode::Append && delegate_disk->exists(wrapped_path))
|
||||
if (mode == WriteMode::Append && delegate_disk->existsFile(wrapped_path))
|
||||
{
|
||||
size_t size = delegate_disk->getFileSize(wrapped_path);
|
||||
old_file_size = size > FileEncryption::Header::kSize ? (size - FileEncryption::Header::kSize) : 0;
|
||||
|
||||
@ -262,17 +262,17 @@ std::optional<UInt64> DiskLocal::getUnreservedSpace() const
|
||||
return available_space;
|
||||
}
|
||||
|
||||
bool DiskLocal::exists(const String & path) const
|
||||
bool DiskLocal::existsFileOrDirectory(const String & path) const
|
||||
{
|
||||
return fs::exists(fs::path(disk_path) / path);
|
||||
}
|
||||
|
||||
bool DiskLocal::isFile(const String & path) const
|
||||
bool DiskLocal::existsFile(const String & path) const
|
||||
{
|
||||
return fs::is_regular_file(fs::path(disk_path) / path);
|
||||
}
|
||||
|
||||
bool DiskLocal::isDirectory(const String & path) const
|
||||
bool DiskLocal::existsDirectory(const String & path) const
|
||||
{
|
||||
return fs::is_directory(fs::path(disk_path) / path);
|
||||
}
|
||||
@ -369,8 +369,11 @@ void DiskLocal::removeFile(const String & path)
|
||||
void DiskLocal::removeFileIfExists(const String & path)
|
||||
{
|
||||
auto fs_path = fs::path(disk_path) / path;
|
||||
if (0 != unlink(fs_path.c_str()) && errno != ENOENT)
|
||||
ErrnoException::throwFromPath(ErrorCodes::CANNOT_UNLINK, fs_path, "Cannot unlink file {}", fs_path);
|
||||
if (0 != unlink(fs_path.c_str()))
|
||||
{
|
||||
if (errno != ENOENT)
|
||||
ErrnoException::throwFromPath(ErrorCodes::CANNOT_UNLINK, fs_path, "Cannot unlink file {}", fs_path);
|
||||
}
|
||||
}
|
||||
|
||||
void DiskLocal::removeDirectory(const String & path)
|
||||
@ -638,7 +641,7 @@ void DiskLocal::setup()
|
||||
|
||||
try
|
||||
{
|
||||
if (exists(disk_checker_path))
|
||||
if (existsFile(disk_checker_path))
|
||||
{
|
||||
auto magic_number = readDiskCheckerMagicNumber();
|
||||
if (magic_number)
|
||||
|
||||
@ -42,11 +42,9 @@ public:
|
||||
|
||||
UInt64 getKeepingFreeSpace() const override { return keep_free_space_bytes; }
|
||||
|
||||
bool exists(const String & path) const override;
|
||||
|
||||
bool isFile(const String & path) const override;
|
||||
|
||||
bool isDirectory(const String & path) const override;
|
||||
bool existsFile(const String & path) const override;
|
||||
bool existsDirectory(const String & path) const override;
|
||||
bool existsFileOrDirectory(const String & path) const override;
|
||||
|
||||
size_t getFileSize(const String & path) const override;
|
||||
|
||||
|
||||
@ -43,6 +43,18 @@ void IDisk::copyFile( /// NOLINT
|
||||
out->finalize();
|
||||
}
|
||||
|
||||
std::unique_ptr<ReadBufferFromFileBase> IDisk::readFileIfExists( /// NOLINT
|
||||
const String & path,
|
||||
const ReadSettings & settings,
|
||||
std::optional<size_t> read_hint,
|
||||
std::optional<size_t> file_size) const
|
||||
{
|
||||
if (existsFile(path))
|
||||
return readFile(path, settings, read_hint, file_size);
|
||||
else
|
||||
return {};
|
||||
}
|
||||
|
||||
DiskTransactionPtr IDisk::createTransaction()
|
||||
{
|
||||
return std::make_shared<FakeDiskTransaction>(*this);
|
||||
@ -96,7 +108,7 @@ void asyncCopy(
|
||||
const WriteSettings & write_settings,
|
||||
const std::function<void()> & cancellation_hook)
|
||||
{
|
||||
if (from_disk.isFile(from_path))
|
||||
if (from_disk.existsFile(from_path))
|
||||
{
|
||||
runner(
|
||||
[&from_disk, from_path, &to_disk, to_path, &read_settings, &write_settings, &cancellation_hook] {
|
||||
@ -149,7 +161,7 @@ void IDisk::copyDirectoryContent(
|
||||
const WriteSettings & write_settings,
|
||||
const std::function<void()> & cancellation_hook)
|
||||
{
|
||||
if (!to_disk->exists(to_dir))
|
||||
if (!to_disk->existsDirectory(to_dir))
|
||||
to_disk->createDirectories(to_dir);
|
||||
|
||||
copyThroughBuffers(from_dir, to_disk, to_dir, /* copy_root_dir= */ false, read_settings, write_settings, cancellation_hook);
|
||||
|
||||
@ -154,14 +154,12 @@ public:
|
||||
/// Amount of bytes which should be kept free on the disk.
|
||||
virtual UInt64 getKeepingFreeSpace() const { return 0; }
|
||||
|
||||
/// Return `true` if the specified file exists.
|
||||
virtual bool exists(const String & path) const = 0;
|
||||
/// Return `true` if the specified file/directory exists.
|
||||
virtual bool existsFile(const String & path) const = 0;
|
||||
virtual bool existsDirectory(const String & path) const = 0;
|
||||
|
||||
/// Return `true` if the specified file exists and it's a regular file (not a directory or special file type).
|
||||
virtual bool isFile(const String & path) const = 0;
|
||||
|
||||
/// Return `true` if the specified file exists and it's a directory.
|
||||
virtual bool isDirectory(const String & path) const = 0;
|
||||
/// This method can be less efficient than the above.
|
||||
virtual bool existsFileOrDirectory(const String & path) const = 0;
|
||||
|
||||
/// Return size of the specified file.
|
||||
virtual size_t getFileSize(const String & path) const = 0;
|
||||
@ -223,6 +221,14 @@ public:
|
||||
std::optional<size_t> read_hint = {},
|
||||
std::optional<size_t> file_size = {}) const = 0;
|
||||
|
||||
/// Returns nullptr if the file does not exist, otherwise opens it for reading.
|
||||
/// This method can save a request. The default implementation will do a separate `exists` call.
|
||||
virtual std::unique_ptr<ReadBufferFromFileBase> readFileIfExists( /// NOLINT
|
||||
const String & path,
|
||||
const ReadSettings & settings = ReadSettings{},
|
||||
std::optional<size_t> read_hint = {},
|
||||
std::optional<size_t> file_size = {}) const;
|
||||
|
||||
/// Open the file for write and return WriteBufferFromFileBase object.
|
||||
virtual std::unique_ptr<WriteBufferFromFileBase> writeFile( /// NOLINT
|
||||
const String & path,
|
||||
@ -308,6 +314,13 @@ public:
|
||||
getDataSourceDescription().toString());
|
||||
}
|
||||
|
||||
virtual std::optional<StoredObjects> getStorageObjectsIfExist(const String & path) const
|
||||
{
|
||||
if (existsFile(path))
|
||||
return getStorageObjects(path);
|
||||
return std::nullopt;
|
||||
}
|
||||
|
||||
/// For one local path there might be multiple remote paths in case of Log family engines.
|
||||
struct LocalPathWithObjectStoragePaths
|
||||
{
|
||||
@ -385,8 +398,8 @@ public:
|
||||
|
||||
/// Check file exists and ClickHouse has an access to it
|
||||
/// Overrode in remote FS disks (s3/hdfs)
|
||||
/// Required for remote disk to ensure that replica has access to data written by other node
|
||||
virtual bool checkUniqueId(const String & id) const { return exists(id); }
|
||||
/// Required for remote disk to ensure that the replica has access to data written by other node
|
||||
virtual bool checkUniqueId(const String & id) const { return existsFile(id); }
|
||||
|
||||
/// Invoked on partitions freeze query.
|
||||
virtual void onFreeze(const String &) { }
|
||||
|
||||
@ -111,9 +111,9 @@ std::future<IAsynchronousReader::Result> ThreadPoolReader::submit(Request reques
|
||||
/// RWF_NOWAIT flag may return 0 even when not at end of file.
|
||||
/// It can't be distinguished from the real eof, so we have to
|
||||
/// disable pread with nowait.
|
||||
static std::atomic<bool> has_pread_nowait_support = !hasBugInPreadV2();
|
||||
static const bool has_pread_nowait_support = !hasBugInPreadV2();
|
||||
|
||||
if (has_pread_nowait_support.load(std::memory_order_relaxed))
|
||||
if (has_pread_nowait_support)
|
||||
{
|
||||
/// It reports real time spent including the time spent while thread was preempted doing nothing.
|
||||
/// And it is Ok for the purpose of this watch (it is used to lower the number of threads to read from tables).
|
||||
@ -161,7 +161,8 @@ std::future<IAsynchronousReader::Result> ThreadPoolReader::submit(Request reques
|
||||
if (errno == ENOSYS || errno == EOPNOTSUPP)
|
||||
{
|
||||
/// No support for the syscall or the flag in the Linux kernel.
|
||||
has_pread_nowait_support.store(false, std::memory_order_relaxed);
|
||||
/// It shouldn't happen because we check the kernel version but let's
|
||||
/// fallback to the thread pool.
|
||||
break;
|
||||
}
|
||||
if (errno == EAGAIN)
|
||||
|
||||
@ -31,7 +31,7 @@ CachedObjectStorage::CachedObjectStorage(
|
||||
|
||||
FileCache::Key CachedObjectStorage::getCacheKey(const std::string & path) const
|
||||
{
|
||||
return cache->createKeyForPath(path);
|
||||
return FileCacheKey::fromPath(path);
|
||||
}
|
||||
|
||||
ObjectStorageKey
|
||||
@ -71,7 +71,7 @@ std::unique_ptr<ReadBufferFromFileBase> CachedObjectStorage::readObject( /// NOL
|
||||
{
|
||||
if (cache->isInitialized())
|
||||
{
|
||||
auto cache_key = cache->createKeyForPath(object.remote_path);
|
||||
auto cache_key = FileCacheKey::fromPath(object.remote_path);
|
||||
auto global_context = Context::getGlobalContextInstance();
|
||||
auto modified_read_settings = read_settings.withNestedBuffer();
|
||||
|
||||
|
||||
@ -91,15 +91,19 @@ StoredObjects DiskObjectStorage::getStorageObjects(const String & local_path) co
|
||||
}
|
||||
|
||||
|
||||
bool DiskObjectStorage::exists(const String & path) const
|
||||
bool DiskObjectStorage::existsFile(const String & path) const
|
||||
{
|
||||
return metadata_storage->exists(path);
|
||||
return metadata_storage->existsFile(path);
|
||||
}
|
||||
|
||||
|
||||
bool DiskObjectStorage::isFile(const String & path) const
|
||||
bool DiskObjectStorage::existsDirectory(const String & path) const
|
||||
{
|
||||
return metadata_storage->isFile(path);
|
||||
return metadata_storage->existsDirectory(path);
|
||||
}
|
||||
|
||||
bool DiskObjectStorage::existsFileOrDirectory(const String & path) const
|
||||
{
|
||||
return metadata_storage->existsFileOrDirectory(path);
|
||||
}
|
||||
|
||||
|
||||
@ -175,7 +179,7 @@ void DiskObjectStorage::moveFile(const String & from_path, const String & to_pat
|
||||
|
||||
void DiskObjectStorage::replaceFile(const String & from_path, const String & to_path)
|
||||
{
|
||||
if (exists(to_path))
|
||||
if (existsFile(to_path))
|
||||
{
|
||||
auto transaction = createObjectStorageTransaction();
|
||||
transaction->replaceFile(from_path, to_path);
|
||||
@ -258,12 +262,6 @@ void DiskObjectStorage::setReadOnly(const String & path)
|
||||
}
|
||||
|
||||
|
||||
bool DiskObjectStorage::isDirectory(const String & path) const
|
||||
{
|
||||
return metadata_storage->isDirectory(path);
|
||||
}
|
||||
|
||||
|
||||
void DiskObjectStorage::createDirectory(const String & path)
|
||||
{
|
||||
auto transaction = createObjectStorageTransaction();
|
||||
@ -554,6 +552,18 @@ std::unique_ptr<ReadBufferFromFileBase> DiskObjectStorage::readFile(
|
||||
return impl;
|
||||
}
|
||||
|
||||
std::unique_ptr<ReadBufferFromFileBase> DiskObjectStorage::readFileIfExists(
|
||||
const String & path,
|
||||
const ReadSettings & settings,
|
||||
std::optional<size_t> read_hint,
|
||||
std::optional<size_t> file_size) const
|
||||
{
|
||||
if (auto storage_objects = metadata_storage->getStorageObjectsIfExist(path))
|
||||
return readFile(path, settings, read_hint, file_size);
|
||||
else
|
||||
return {};
|
||||
}
|
||||
|
||||
std::unique_ptr<WriteBufferFromFileBase> DiskObjectStorage::writeFile(
|
||||
const String & path,
|
||||
size_t buf_size,
|
||||
|
||||
@ -58,9 +58,9 @@ public:
|
||||
|
||||
UInt64 getKeepingFreeSpace() const override { return 0; }
|
||||
|
||||
bool exists(const String & path) const override;
|
||||
|
||||
bool isFile(const String & path) const override;
|
||||
bool existsFile(const String & path) const override;
|
||||
bool existsDirectory(const String & path) const override;
|
||||
bool existsFileOrDirectory(const String & path) const override;
|
||||
|
||||
void createFile(const String & path) override;
|
||||
|
||||
@ -108,8 +108,6 @@ public:
|
||||
|
||||
void setReadOnly(const String & path) override;
|
||||
|
||||
bool isDirectory(const String & path) const override;
|
||||
|
||||
void createDirectory(const String & path) override;
|
||||
|
||||
void createDirectories(const String & path) override;
|
||||
@ -142,6 +140,12 @@ public:
|
||||
std::optional<size_t> read_hint,
|
||||
std::optional<size_t> file_size) const override;
|
||||
|
||||
std::unique_ptr<ReadBufferFromFileBase> readFileIfExists(
|
||||
const String & path,
|
||||
const ReadSettings & settings,
|
||||
std::optional<size_t> read_hint,
|
||||
std::optional<size_t> file_size) const override;
|
||||
|
||||
std::unique_ptr<WriteBufferFromFileBase> writeFile(
|
||||
const String & path,
|
||||
size_t buf_size,
|
||||
|
||||
@ -117,7 +117,7 @@ void DiskObjectStorageRemoteMetadataRestoreHelper::migrateToRestorableSchemaRecu
|
||||
bool dir_contains_only_files = true;
|
||||
for (auto it = disk->iterateDirectory(path); it->isValid(); it->next())
|
||||
{
|
||||
if (disk->isDirectory(it->path()))
|
||||
if (disk->existsDirectory(it->path()))
|
||||
{
|
||||
dir_contains_only_files = false;
|
||||
break;
|
||||
@ -138,7 +138,7 @@ void DiskObjectStorageRemoteMetadataRestoreHelper::migrateToRestorableSchemaRecu
|
||||
{
|
||||
for (auto it = disk->iterateDirectory(path); it->isValid(); it->next())
|
||||
{
|
||||
if (disk->isDirectory(it->path()))
|
||||
if (disk->existsDirectory(it->path()))
|
||||
{
|
||||
migrateToRestorableSchemaRecursive(it->path(), pool);
|
||||
}
|
||||
@ -161,7 +161,7 @@ void DiskObjectStorageRemoteMetadataRestoreHelper::migrateToRestorableSchema()
|
||||
ThreadPool pool{CurrentMetrics::LocalThread, CurrentMetrics::LocalThreadActive, CurrentMetrics::LocalThreadScheduled};
|
||||
|
||||
for (const auto & root : data_roots)
|
||||
if (disk->exists(root))
|
||||
if (disk->existsDirectory(root))
|
||||
migrateToRestorableSchemaRecursive(root + '/', pool);
|
||||
|
||||
pool.wait();
|
||||
@ -180,7 +180,7 @@ void DiskObjectStorageRemoteMetadataRestoreHelper::restore(const Poco::Util::Abs
|
||||
{
|
||||
LOG_INFO(disk->log, "Restore operation for disk {} called", disk->name);
|
||||
|
||||
if (!disk->exists(RESTORE_FILE_NAME))
|
||||
if (!disk->existsFile(RESTORE_FILE_NAME))
|
||||
{
|
||||
LOG_INFO(disk->log, "No restore file '{}' exists, finishing restore", RESTORE_FILE_NAME);
|
||||
return;
|
||||
@ -228,7 +228,7 @@ void DiskObjectStorageRemoteMetadataRestoreHelper::restore(const Poco::Util::Abs
|
||||
|
||||
bool cleanup_s3 = information.source_path != disk->object_key_prefix;
|
||||
for (const auto & root : data_roots)
|
||||
if (disk->exists(root))
|
||||
if (disk->existsDirectory(root))
|
||||
disk->removeSharedRecursive(root + '/', !cleanup_s3, {});
|
||||
|
||||
LOG_INFO(disk->log, "Old metadata removed, restoring new one");
|
||||
@ -326,7 +326,7 @@ static std::tuple<UInt64, String> extractRevisionAndOperationFromKey(const Strin
|
||||
|
||||
void DiskObjectStorageRemoteMetadataRestoreHelper::moveRecursiveOrRemove(const String & from_path, const String & to_path, bool send_metadata)
|
||||
{
|
||||
if (disk->exists(to_path))
|
||||
if (disk->existsFileOrDirectory(to_path))
|
||||
{
|
||||
if (send_metadata)
|
||||
{
|
||||
@ -337,7 +337,7 @@ void DiskObjectStorageRemoteMetadataRestoreHelper::moveRecursiveOrRemove(const S
|
||||
};
|
||||
createFileOperationObject("rename", revision, object_metadata);
|
||||
}
|
||||
if (disk->isDirectory(from_path))
|
||||
if (disk->existsDirectory(from_path))
|
||||
{
|
||||
for (auto it = disk->iterateDirectory(from_path); it->isValid(); it->next())
|
||||
moveRecursiveOrRemove(it->path(), fs::path(to_path) / it->name(), false);
|
||||
@ -490,13 +490,13 @@ void DiskObjectStorageRemoteMetadataRestoreHelper::restoreFileOperations(IObject
|
||||
{
|
||||
auto from_path = object_attributes["from_path"];
|
||||
auto to_path = object_attributes["to_path"];
|
||||
if (disk->exists(from_path))
|
||||
if (disk->existsFileOrDirectory(from_path))
|
||||
{
|
||||
moveRecursiveOrRemove(from_path, to_path, send_metadata);
|
||||
|
||||
LOG_TRACE(disk->log, "Revision {}. Restored rename {} -> {}", revision, from_path, to_path);
|
||||
|
||||
if (restore_information.detached && disk->isDirectory(to_path))
|
||||
if (restore_information.detached && disk->existsDirectory(to_path))
|
||||
{
|
||||
/// Sometimes directory paths are passed without trailing '/'. We should keep them in one consistent way.
|
||||
if (!from_path.ends_with('/'))
|
||||
@ -517,7 +517,7 @@ void DiskObjectStorageRemoteMetadataRestoreHelper::restoreFileOperations(IObject
|
||||
{
|
||||
auto src_path = object_attributes["src_path"];
|
||||
auto dst_path = object_attributes["dst_path"];
|
||||
if (disk->exists(src_path))
|
||||
if (disk->existsFile(src_path))
|
||||
{
|
||||
disk->createDirectories(directoryPath(dst_path));
|
||||
disk->createHardLink(src_path, dst_path, send_metadata);
|
||||
@ -564,7 +564,7 @@ void DiskObjectStorageRemoteMetadataRestoreHelper::restoreFileOperations(IObject
|
||||
to_path /= from_path.filename();
|
||||
|
||||
/// to_path may exist and non-empty in case for example abrupt restart, so remove it before rename
|
||||
if (disk->metadata_storage->exists(to_path))
|
||||
if (disk->metadata_storage->existsFileOrDirectory(to_path))
|
||||
tx->removeRecursive(to_path);
|
||||
|
||||
disk->createDirectories(directoryPath(to_path));
|
||||
|
||||
@ -22,7 +22,6 @@ namespace ErrorCodes
|
||||
extern const int CANNOT_READ_ALL_DATA;
|
||||
extern const int CANNOT_OPEN_FILE;
|
||||
extern const int FILE_DOESNT_EXIST;
|
||||
extern const int BAD_FILE_TYPE;
|
||||
extern const int CANNOT_PARSE_INPUT_ASSERTION_FAILED;
|
||||
extern const int LOGICAL_ERROR;
|
||||
}
|
||||
@ -126,17 +125,14 @@ struct RemoveObjectStorageOperation final : public IDiskObjectStorageOperation
|
||||
|
||||
void execute(MetadataTransactionPtr tx) override
|
||||
{
|
||||
if (!metadata_storage.exists(path))
|
||||
if (!metadata_storage.existsFile(path))
|
||||
{
|
||||
if (if_exists)
|
||||
return;
|
||||
|
||||
throw Exception(ErrorCodes::FILE_DOESNT_EXIST, "Metadata path '{}' doesn't exist", path);
|
||||
throw Exception(ErrorCodes::FILE_DOESNT_EXIST, "Metadata path '{}' doesn't exist or isn't a regular file", path);
|
||||
}
|
||||
|
||||
if (!metadata_storage.isFile(path))
|
||||
throw Exception(ErrorCodes::BAD_FILE_TYPE, "Path '{}' is not a regular file", path);
|
||||
|
||||
try
|
||||
{
|
||||
auto objects = metadata_storage.getStorageObjects(path);
|
||||
@ -211,17 +207,14 @@ struct RemoveManyObjectStorageOperation final : public IDiskObjectStorageOperati
|
||||
{
|
||||
for (const auto & [path, if_exists] : remove_paths)
|
||||
{
|
||||
if (!metadata_storage.exists(path))
|
||||
if (!metadata_storage.existsFile(path))
|
||||
{
|
||||
if (if_exists)
|
||||
continue;
|
||||
|
||||
throw Exception(ErrorCodes::FILE_DOESNT_EXIST, "Metadata path '{}' doesn't exist", path);
|
||||
throw Exception(ErrorCodes::FILE_DOESNT_EXIST, "Metadata path '{}' doesn't exist or isn't a regular file", path);
|
||||
}
|
||||
|
||||
if (!metadata_storage.isFile(path))
|
||||
throw Exception(ErrorCodes::BAD_FILE_TYPE, "Path '{}' is not a regular file", path);
|
||||
|
||||
try
|
||||
{
|
||||
auto objects = metadata_storage.getStorageObjects(path);
|
||||
@ -318,7 +311,7 @@ struct RemoveRecursiveObjectStorageOperation final : public IDiskObjectStorageOp
|
||||
{
|
||||
checkStackSize(); /// This is needed to prevent stack overflow in case of cyclic symlinks.
|
||||
|
||||
if (metadata_storage.isFile(path_to_remove))
|
||||
if (metadata_storage.existsFile(path_to_remove))
|
||||
{
|
||||
try
|
||||
{
|
||||
@ -367,7 +360,7 @@ struct RemoveRecursiveObjectStorageOperation final : public IDiskObjectStorageOp
|
||||
void execute(MetadataTransactionPtr tx) override
|
||||
{
|
||||
/// Similar to DiskLocal and https://en.cppreference.com/w/cpp/filesystem/remove
|
||||
if (metadata_storage.exists(path))
|
||||
if (metadata_storage.existsFileOrDirectory(path))
|
||||
removeMetadataRecursive(tx, path);
|
||||
}
|
||||
|
||||
@ -433,7 +426,7 @@ struct ReplaceFileObjectStorageOperation final : public IDiskObjectStorageOperat
|
||||
|
||||
void execute(MetadataTransactionPtr tx) override
|
||||
{
|
||||
if (metadata_storage.exists(path_to))
|
||||
if (metadata_storage.existsFile(path_to))
|
||||
{
|
||||
objects_to_remove = metadata_storage.getStorageObjects(path_to);
|
||||
tx->replaceFile(path_from, path_to);
|
||||
@ -583,13 +576,8 @@ struct TruncateFileObjectStorageOperation final : public IDiskObjectStorageOpera
|
||||
|
||||
void execute(MetadataTransactionPtr tx) override
|
||||
{
|
||||
if (metadata_storage.exists(path))
|
||||
{
|
||||
if (!metadata_storage.isFile(path))
|
||||
throw Exception(ErrorCodes::LOGICAL_ERROR, "Path {} is not a file", path);
|
||||
|
||||
if (metadata_storage.existsFile(path))
|
||||
truncate_outcome = tx->truncateFile(path, size);
|
||||
}
|
||||
}
|
||||
|
||||
void undo() override
|
||||
@ -663,7 +651,7 @@ void DiskObjectStorageTransaction::clearDirectory(const std::string & path)
|
||||
{
|
||||
for (auto it = metadata_storage.iterateDirectory(path); it->isValid(); it->next())
|
||||
{
|
||||
if (metadata_storage.isFile(it->path()))
|
||||
if (metadata_storage.existsFile(it->path()))
|
||||
removeFile(it->path());
|
||||
}
|
||||
}
|
||||
@ -769,7 +757,7 @@ std::unique_ptr<WriteBufferFromFileBase> DiskObjectStorageTransaction::writeFile
|
||||
{
|
||||
/// Otherwise we will produce lost blobs which nobody points to
|
||||
/// WriteOnce storages are not affected by the issue
|
||||
if (!tx->object_storage.isPlain() && tx->metadata_storage.exists(path))
|
||||
if (!tx->object_storage.isPlain() && tx->metadata_storage.existsFile(path))
|
||||
tx->object_storage.removeObjectsIfExist(tx->metadata_storage.getStorageObjects(path));
|
||||
|
||||
tx->metadata_transaction->createMetadataFile(path, key_, count);
|
||||
@ -802,7 +790,7 @@ std::unique_ptr<WriteBufferFromFileBase> DiskObjectStorageTransaction::writeFile
|
||||
{
|
||||
/// Otherwise we will produce lost blobs which nobody points to
|
||||
/// WriteOnce storages are not affected by the issue
|
||||
if (!object_storage_tx->object_storage.isPlain() && object_storage_tx->metadata_storage.exists(path))
|
||||
if (!object_storage_tx->object_storage.isPlain() && object_storage_tx->metadata_storage.existsFile(path))
|
||||
{
|
||||
object_storage_tx->object_storage.removeObjectsIfExist(object_storage_tx->metadata_storage.getStorageObjects(path));
|
||||
}
|
||||
@ -876,7 +864,7 @@ void DiskObjectStorageTransaction::writeFileUsingBlobWritingFunction(
|
||||
{
|
||||
/// Otherwise we will produce lost blobs which nobody points to
|
||||
/// WriteOnce storages are not affected by the issue
|
||||
if (!object_storage.isPlain() && metadata_storage.exists(path))
|
||||
if (!object_storage.isPlain() && metadata_storage.existsFile(path))
|
||||
object_storage.removeObjectsIfExist(metadata_storage.getStorageObjects(path));
|
||||
|
||||
metadata_transaction->createMetadataFile(path, std::move(object_key), object_size);
|
||||
|
||||
@ -182,14 +182,19 @@ public:
|
||||
|
||||
/// ==== General purpose methods. Define properties of object storage file based on metadata files ====
|
||||
|
||||
virtual bool exists(const std::string & path) const = 0;
|
||||
|
||||
virtual bool isFile(const std::string & path) const = 0;
|
||||
|
||||
virtual bool isDirectory(const std::string & path) const = 0;
|
||||
virtual bool existsFile(const std::string & path) const = 0;
|
||||
virtual bool existsDirectory(const std::string & path) const = 0;
|
||||
virtual bool existsFileOrDirectory(const std::string & path) const = 0;
|
||||
|
||||
virtual uint64_t getFileSize(const std::string & path) const = 0;
|
||||
|
||||
virtual std::optional<uint64_t> getFileSizeIfExists(const std::string & path) const
|
||||
{
|
||||
if (existsFile(path))
|
||||
return getFileSize(path);
|
||||
return std::nullopt;
|
||||
}
|
||||
|
||||
virtual Poco::Timestamp getLastModified(const std::string & path) const = 0;
|
||||
|
||||
virtual time_t getLastChanged(const std::string & /* path */) const
|
||||
@ -242,6 +247,13 @@ public:
|
||||
/// object_storage_path is absolute.
|
||||
virtual StoredObjects getStorageObjects(const std::string & path) const = 0;
|
||||
|
||||
virtual std::optional<StoredObjects> getStorageObjectsIfExist(const std::string & path) const
|
||||
{
|
||||
if (existsFile(path))
|
||||
return getStorageObjects(path);
|
||||
return std::nullopt;
|
||||
}
|
||||
|
||||
protected:
|
||||
[[noreturn]] static void throwNotImplemented()
|
||||
{
|
||||
|
||||
@ -116,7 +116,7 @@ void registerPlainMetadataStorage(MetadataStorageFactory & factory)
|
||||
ObjectStoragePtr object_storage) -> MetadataStoragePtr
|
||||
{
|
||||
auto key_compatibility_prefix = getObjectKeyCompatiblePrefix(*object_storage, config, config_prefix);
|
||||
return std::make_shared<MetadataStorageFromPlainObjectStorage>(object_storage, key_compatibility_prefix);
|
||||
return std::make_shared<MetadataStorageFromPlainObjectStorage>(object_storage, key_compatibility_prefix, config.getUInt64(config_prefix + ".file_sizes_cache_size", 0));
|
||||
});
|
||||
}
|
||||
|
||||
@ -130,7 +130,7 @@ void registerPlainRewritableMetadataStorage(MetadataStorageFactory & factory)
|
||||
ObjectStoragePtr object_storage) -> MetadataStoragePtr
|
||||
{
|
||||
auto key_compatibility_prefix = getObjectKeyCompatiblePrefix(*object_storage, config, config_prefix);
|
||||
return std::make_shared<MetadataStorageFromPlainRewritableObjectStorage>(object_storage, key_compatibility_prefix);
|
||||
return std::make_shared<MetadataStorageFromPlainRewritableObjectStorage>(object_storage, key_compatibility_prefix, config.getUInt64(config_prefix + ".file_sizes_cache_size", 0));
|
||||
});
|
||||
}
|
||||
|
||||
|
||||
@ -20,19 +20,19 @@ const std::string & MetadataStorageFromDisk::getPath() const
|
||||
return disk->getPath();
|
||||
}
|
||||
|
||||
bool MetadataStorageFromDisk::exists(const std::string & path) const
|
||||
bool MetadataStorageFromDisk::existsFile(const std::string & path) const
|
||||
{
|
||||
return disk->exists(path);
|
||||
return disk->existsFile(path);
|
||||
}
|
||||
|
||||
bool MetadataStorageFromDisk::isFile(const std::string & path) const
|
||||
bool MetadataStorageFromDisk::existsDirectory(const std::string & path) const
|
||||
{
|
||||
return disk->isFile(path);
|
||||
return disk->existsDirectory(path);
|
||||
}
|
||||
|
||||
bool MetadataStorageFromDisk::isDirectory(const std::string & path) const
|
||||
bool MetadataStorageFromDisk::existsFileOrDirectory(const std::string & path) const
|
||||
{
|
||||
return disk->isDirectory(path);
|
||||
return disk->existsFileOrDirectory(path);
|
||||
}
|
||||
|
||||
Poco::Timestamp MetadataStorageFromDisk::getLastModified(const std::string & path) const
|
||||
|
||||
@ -15,7 +15,7 @@ namespace DB
|
||||
struct UnlinkMetadataFileOperationOutcome;
|
||||
using UnlinkMetadataFileOperationOutcomePtr = std::shared_ptr<UnlinkMetadataFileOperationOutcome>;
|
||||
|
||||
/// Store metadata on a separate disk
|
||||
/// Stores metadata on a separate disk
|
||||
/// (used for object storages, like S3 and related).
|
||||
class MetadataStorageFromDisk final : public IMetadataStorage
|
||||
{
|
||||
@ -35,11 +35,9 @@ public:
|
||||
|
||||
MetadataStorageType getType() const override { return MetadataStorageType::Local; }
|
||||
|
||||
bool exists(const std::string & path) const override;
|
||||
|
||||
bool isFile(const std::string & path) const override;
|
||||
|
||||
bool isDirectory(const std::string & path) const override;
|
||||
bool existsFile(const std::string & path) const override;
|
||||
bool existsDirectory(const std::string & path) const override;
|
||||
bool existsFileOrDirectory(const std::string & path) const override;
|
||||
|
||||
uint64_t getFileSize(const String & path) const override;
|
||||
|
||||
|
||||
@ -109,7 +109,7 @@ void CreateDirectoryRecursiveOperation::execute(std::unique_lock<SharedMutex> &)
|
||||
{
|
||||
namespace fs = std::filesystem;
|
||||
fs::path p(path);
|
||||
while (!disk.exists(p))
|
||||
while (!disk.existsFileOrDirectory(p))
|
||||
{
|
||||
paths_created.push_back(p);
|
||||
if (!p.has_parent_path())
|
||||
@ -151,26 +151,26 @@ RemoveRecursiveOperation::RemoveRecursiveOperation(const std::string & path_, ID
|
||||
|
||||
void RemoveRecursiveOperation::execute(std::unique_lock<SharedMutex> &)
|
||||
{
|
||||
if (disk.isFile(path))
|
||||
if (disk.existsFile(path))
|
||||
disk.moveFile(path, temp_path);
|
||||
else if (disk.isDirectory(path))
|
||||
else if (disk.existsDirectory(path))
|
||||
disk.moveDirectory(path, temp_path);
|
||||
}
|
||||
|
||||
void RemoveRecursiveOperation::undo(std::unique_lock<SharedMutex> &)
|
||||
{
|
||||
if (disk.isFile(temp_path))
|
||||
if (disk.existsFile(temp_path))
|
||||
disk.moveFile(temp_path, path);
|
||||
else if (disk.isDirectory(temp_path))
|
||||
else if (disk.existsDirectory(temp_path))
|
||||
disk.moveDirectory(temp_path, path);
|
||||
}
|
||||
|
||||
void RemoveRecursiveOperation::finalize()
|
||||
{
|
||||
if (disk.exists(temp_path))
|
||||
if (disk.existsFileOrDirectory(temp_path))
|
||||
disk.removeRecursive(temp_path);
|
||||
|
||||
if (disk.exists(path))
|
||||
if (disk.existsFileOrDirectory(path))
|
||||
disk.removeRecursive(path);
|
||||
}
|
||||
|
||||
@ -246,7 +246,7 @@ ReplaceFileOperation::ReplaceFileOperation(const std::string & path_from_, const
|
||||
|
||||
void ReplaceFileOperation::execute(std::unique_lock<SharedMutex> &)
|
||||
{
|
||||
if (disk.exists(path_to))
|
||||
if (disk.existsFile(path_to))
|
||||
disk.moveFile(path_to, temp_path_to);
|
||||
|
||||
disk.replaceFile(path_from, path_to);
|
||||
@ -272,10 +272,9 @@ WriteFileOperation::WriteFileOperation(const std::string & path_, IDisk & disk_,
|
||||
|
||||
void WriteFileOperation::execute(std::unique_lock<SharedMutex> &)
|
||||
{
|
||||
if (disk.exists(path))
|
||||
if (auto buf = disk.readFileIfExists(path, ReadSettings{}))
|
||||
{
|
||||
existed = true;
|
||||
auto buf = disk.readFile(path, ReadSettings{});
|
||||
readStringUntilEOF(prev_data, *buf);
|
||||
}
|
||||
auto buf = disk.writeFile(path);
|
||||
@ -299,7 +298,7 @@ void WriteFileOperation::undo(std::unique_lock<SharedMutex> &)
|
||||
void AddBlobOperation::execute(std::unique_lock<SharedMutex> & metadata_lock)
|
||||
{
|
||||
DiskObjectStorageMetadataPtr metadata;
|
||||
if (metadata_storage.exists(path))
|
||||
if (metadata_storage.existsFile(path))
|
||||
metadata = metadata_storage.readMetadataUnlocked(path, metadata_lock);
|
||||
else
|
||||
metadata = std::make_unique<DiskObjectStorageMetadata>(disk.getPath(), path);
|
||||
@ -351,7 +350,7 @@ void UnlinkMetadataFileOperation::undo(std::unique_lock<SharedMutex> & lock)
|
||||
|
||||
void TruncateMetadataFileOperation::execute(std::unique_lock<SharedMutex> & metadata_lock)
|
||||
{
|
||||
if (metadata_storage.exists(path))
|
||||
if (metadata_storage.existsFile(path))
|
||||
{
|
||||
auto metadata = metadata_storage.readMetadataUnlocked(path, metadata_lock);
|
||||
while (metadata->getTotalSizeBytes() > target_size)
|
||||
|
||||
@ -3,6 +3,7 @@
|
||||
#include <Disks/ObjectStorages/InMemoryPathMap.h>
|
||||
#include <Disks/ObjectStorages/MetadataStorageFromPlainObjectStorageOperations.h>
|
||||
#include <Disks/ObjectStorages/StaticDirectoryIterator.h>
|
||||
#include <Disks/ObjectStorages/StoredObject.h>
|
||||
|
||||
#include <Common/filesystemHelpers.h>
|
||||
|
||||
@ -10,9 +11,15 @@
|
||||
#include <tuple>
|
||||
#include <unordered_set>
|
||||
|
||||
|
||||
namespace DB
|
||||
{
|
||||
|
||||
namespace ErrorCodes
|
||||
{
|
||||
extern const int FILE_DOESNT_EXIST;
|
||||
}
|
||||
|
||||
namespace
|
||||
{
|
||||
|
||||
@ -23,10 +30,12 @@ std::filesystem::path normalizeDirectoryPath(const std::filesystem::path & path)
|
||||
|
||||
}
|
||||
|
||||
MetadataStorageFromPlainObjectStorage::MetadataStorageFromPlainObjectStorage(ObjectStoragePtr object_storage_, String storage_path_prefix_)
|
||||
MetadataStorageFromPlainObjectStorage::MetadataStorageFromPlainObjectStorage(ObjectStoragePtr object_storage_, String storage_path_prefix_, size_t file_sizes_cache_size)
|
||||
: object_storage(object_storage_)
|
||||
, storage_path_prefix(std::move(storage_path_prefix_))
|
||||
{
|
||||
if (file_sizes_cache_size)
|
||||
file_sizes_cache.emplace(file_sizes_cache_size);
|
||||
}
|
||||
|
||||
MetadataTransactionPtr MetadataStorageFromPlainObjectStorage::createTransaction()
|
||||
@ -39,35 +48,61 @@ const std::string & MetadataStorageFromPlainObjectStorage::getPath() const
|
||||
return storage_path_prefix;
|
||||
}
|
||||
|
||||
bool MetadataStorageFromPlainObjectStorage::exists(const std::string & path) const
|
||||
bool MetadataStorageFromPlainObjectStorage::existsFile(const std::string & path) const
|
||||
{
|
||||
/// NOTE: exists() cannot be used here since it works only for existing
|
||||
/// key, and does not work for some intermediate path.
|
||||
auto object_key = object_storage->generateObjectKeyForPath(path, std::nullopt /* key_prefix */);
|
||||
return object_storage->existsOrHasAnyChild(object_key.serialize());
|
||||
ObjectStorageKey object_key = object_storage->generateObjectKeyForPath(path, std::nullopt /* key_prefix */);
|
||||
StoredObject object(object_key.serialize(), path);
|
||||
if (!object_storage->exists(object))
|
||||
return false;
|
||||
|
||||
/// The path does not correspond to a directory.
|
||||
auto directory = std::filesystem::path(object_key.serialize()) / "";
|
||||
return !object_storage->existsOrHasAnyChild(directory);
|
||||
}
|
||||
|
||||
bool MetadataStorageFromPlainObjectStorage::isFile(const std::string & path) const
|
||||
{
|
||||
/// NOTE: This check is inaccurate and has excessive API calls
|
||||
return exists(path) && !isDirectory(path);
|
||||
}
|
||||
|
||||
bool MetadataStorageFromPlainObjectStorage::isDirectory(const std::string & path) const
|
||||
bool MetadataStorageFromPlainObjectStorage::existsDirectory(const std::string & path) const
|
||||
{
|
||||
auto key_prefix = object_storage->generateObjectKeyForPath(path, std::nullopt /* key_prefix */).serialize();
|
||||
auto directory = std::filesystem::path(std::move(key_prefix)) / "";
|
||||
|
||||
return object_storage->existsOrHasAnyChild(directory);
|
||||
}
|
||||
|
||||
bool MetadataStorageFromPlainObjectStorage::existsFileOrDirectory(const std::string & path) const
|
||||
{
|
||||
/// NOTE: exists() cannot be used here since it works only for existing
|
||||
/// key, and does not work for some intermediate path.
|
||||
auto key_prefix = object_storage->generateObjectKeyForPath(path, std::nullopt /* key_prefix */).serialize();
|
||||
return object_storage->existsOrHasAnyChild(key_prefix);
|
||||
}
|
||||
|
||||
|
||||
uint64_t MetadataStorageFromPlainObjectStorage::getFileSize(const String & path) const
|
||||
{
|
||||
auto object_key = object_storage->generateObjectKeyForPath(path, std::nullopt /* key_prefix */);
|
||||
auto metadata = object_storage->tryGetObjectMetadata(object_key.serialize());
|
||||
if (metadata)
|
||||
return metadata->size_bytes;
|
||||
return 0;
|
||||
if (auto res = getFileSizeIfExists(path))
|
||||
return *res;
|
||||
throw Exception(ErrorCodes::FILE_DOESNT_EXIST, "File {} does not exist on plain object storage", path);
|
||||
}
|
||||
|
||||
std::optional<uint64_t> MetadataStorageFromPlainObjectStorage::getFileSizeIfExists(const String & path) const
|
||||
{
|
||||
auto get = [&] -> std::shared_ptr<uint64_t>
|
||||
{
|
||||
auto object_key = object_storage->generateObjectKeyForPath(path, std::nullopt /* key_prefix */);
|
||||
auto metadata = object_storage->tryGetObjectMetadata(object_key.serialize());
|
||||
if (metadata)
|
||||
return std::make_shared<uint64_t>(metadata->size_bytes);
|
||||
return nullptr;
|
||||
};
|
||||
|
||||
std::shared_ptr<uint64_t> res;
|
||||
if (file_sizes_cache)
|
||||
res = file_sizes_cache->getOrSet(path, get).first;
|
||||
else
|
||||
res = get();
|
||||
|
||||
if (res)
|
||||
return *res;
|
||||
return std::nullopt;
|
||||
}
|
||||
|
||||
std::vector<std::string> MetadataStorageFromPlainObjectStorage::listDirectory(const std::string & path) const
|
||||
@ -75,18 +110,18 @@ std::vector<std::string> MetadataStorageFromPlainObjectStorage::listDirectory(co
|
||||
auto key_prefix = object_storage->generateObjectKeyForPath(path, std::nullopt /* key_prefix */).serialize();
|
||||
|
||||
RelativePathsWithMetadata files;
|
||||
std::string abs_key = key_prefix;
|
||||
if (!abs_key.ends_with('/'))
|
||||
abs_key += '/';
|
||||
std::string absolute_key = key_prefix;
|
||||
if (!absolute_key.ends_with('/'))
|
||||
absolute_key += '/';
|
||||
|
||||
object_storage->listObjects(abs_key, files, 0);
|
||||
object_storage->listObjects(absolute_key, files, 0);
|
||||
|
||||
std::unordered_set<std::string> result;
|
||||
for (const auto & elem : files)
|
||||
{
|
||||
const auto & p = elem->relative_path;
|
||||
chassert(p.find(abs_key) == 0);
|
||||
const auto child_pos = abs_key.size();
|
||||
chassert(p.find(absolute_key) == 0);
|
||||
const auto child_pos = absolute_key.size();
|
||||
/// string::npos is ok.
|
||||
const auto slash_pos = p.find('/', child_pos);
|
||||
if (slash_pos == std::string::npos)
|
||||
@ -114,6 +149,16 @@ StoredObjects MetadataStorageFromPlainObjectStorage::getStorageObjects(const std
|
||||
return {StoredObject(object_key.serialize(), path, object_size)};
|
||||
}
|
||||
|
||||
std::optional<StoredObjects> MetadataStorageFromPlainObjectStorage::getStorageObjectsIfExist(const std::string & path) const
|
||||
{
|
||||
if (auto object_size = getFileSizeIfExists(path))
|
||||
{
|
||||
auto object_key = object_storage->generateObjectKeyForPath(path, std::nullopt /* key_prefix */);
|
||||
return StoredObjects{StoredObject(object_key.serialize(), path, *object_size)};
|
||||
}
|
||||
return std::nullopt;
|
||||
}
|
||||
|
||||
const IMetadataStorage & MetadataStorageFromPlainObjectStorageTransaction::getStorageForNonTransactionalReads() const
|
||||
{
|
||||
return metadata_storage;
|
||||
|
||||
@ -5,11 +5,13 @@
|
||||
#include <Disks/ObjectStorages/InMemoryPathMap.h>
|
||||
#include <Disks/ObjectStorages/MetadataOperationsHolder.h>
|
||||
#include <Disks/ObjectStorages/MetadataStorageTransactionState.h>
|
||||
#include <Common/CacheBase.h>
|
||||
|
||||
#include <map>
|
||||
#include <string>
|
||||
#include <unordered_set>
|
||||
|
||||
|
||||
namespace DB
|
||||
{
|
||||
|
||||
@ -31,6 +33,7 @@ class MetadataStorageFromPlainObjectStorage : public IMetadataStorage
|
||||
{
|
||||
private:
|
||||
friend class MetadataStorageFromPlainObjectStorageTransaction;
|
||||
mutable std::optional<CacheBase<String, uint64_t>> file_sizes_cache;
|
||||
|
||||
protected:
|
||||
ObjectStoragePtr object_storage;
|
||||
@ -39,7 +42,7 @@ protected:
|
||||
mutable SharedMutex metadata_mutex;
|
||||
|
||||
public:
|
||||
MetadataStorageFromPlainObjectStorage(ObjectStoragePtr object_storage_, String storage_path_prefix_);
|
||||
MetadataStorageFromPlainObjectStorage(ObjectStoragePtr object_storage_, String storage_path_prefix_, size_t file_sizes_cache_size);
|
||||
|
||||
MetadataTransactionPtr createTransaction() override;
|
||||
|
||||
@ -47,13 +50,12 @@ public:
|
||||
|
||||
MetadataStorageType getType() const override { return MetadataStorageType::Plain; }
|
||||
|
||||
bool exists(const std::string & path) const override;
|
||||
|
||||
bool isFile(const std::string & path) const override;
|
||||
|
||||
bool isDirectory(const std::string & path) const override;
|
||||
bool existsFile(const std::string & path) const override;
|
||||
bool existsDirectory(const std::string & path) const override;
|
||||
bool existsFileOrDirectory(const std::string & path) const override;
|
||||
|
||||
uint64_t getFileSize(const String & path) const override;
|
||||
std::optional<uint64_t> getFileSizeIfExists(const String & path) const override;
|
||||
|
||||
std::vector<std::string> listDirectory(const std::string & path) const override;
|
||||
|
||||
@ -62,6 +64,7 @@ public:
|
||||
DiskPtr getDisk() const { return {}; }
|
||||
|
||||
StoredObjects getStorageObjects(const std::string & path) const override;
|
||||
std::optional<StoredObjects> getStorageObjectsIfExist(const std::string & path) const override;
|
||||
|
||||
Poco::Timestamp getLastModified(const std::string & /* path */) const override
|
||||
{
|
||||
|
||||
@ -7,9 +7,8 @@
|
||||
#include <IO/ReadHelpers.h>
|
||||
#include <IO/S3Common.h>
|
||||
#include <IO/SharedThreadPools.h>
|
||||
#include "Common/SharedLockGuard.h"
|
||||
#include "Common/SharedMutex.h"
|
||||
#include <Common/ErrorCodes.h>
|
||||
#include <Common/SharedLockGuard.h>
|
||||
#include <Common/SharedMutex.h>
|
||||
#include <Common/logger_useful.h>
|
||||
#include "CommonPathPrefixKeyGenerator.h"
|
||||
|
||||
@ -181,8 +180,8 @@ void getDirectChildrenOnDiskImpl(
|
||||
}
|
||||
|
||||
MetadataStorageFromPlainRewritableObjectStorage::MetadataStorageFromPlainRewritableObjectStorage(
|
||||
ObjectStoragePtr object_storage_, String storage_path_prefix_)
|
||||
: MetadataStorageFromPlainObjectStorage(object_storage_, storage_path_prefix_)
|
||||
ObjectStoragePtr object_storage_, String storage_path_prefix_, size_t file_sizes_cache_size)
|
||||
: MetadataStorageFromPlainObjectStorage(object_storage_, storage_path_prefix_, file_sizes_cache_size)
|
||||
, metadata_key_prefix(DB::getMetadataKeyPrefix(object_storage))
|
||||
, path_map(loadPathPrefixMap(metadata_key_prefix, object_storage))
|
||||
{
|
||||
@ -211,9 +210,9 @@ MetadataStorageFromPlainRewritableObjectStorage::~MetadataStorageFromPlainRewrit
|
||||
CurrentMetrics::sub(metric, path_map->map.size());
|
||||
}
|
||||
|
||||
bool MetadataStorageFromPlainRewritableObjectStorage::exists(const std::string & path) const
|
||||
bool MetadataStorageFromPlainRewritableObjectStorage::existsFileOrDirectory(const std::string & path) const
|
||||
{
|
||||
if (MetadataStorageFromPlainObjectStorage::exists(path))
|
||||
if (MetadataStorageFromPlainObjectStorage::existsFileOrDirectory(path))
|
||||
return true;
|
||||
|
||||
if (useSeparateLayoutForMetadata())
|
||||
@ -225,14 +224,19 @@ bool MetadataStorageFromPlainRewritableObjectStorage::exists(const std::string &
|
||||
return false;
|
||||
}
|
||||
|
||||
bool MetadataStorageFromPlainRewritableObjectStorage::isDirectory(const std::string & path) const
|
||||
bool MetadataStorageFromPlainRewritableObjectStorage::existsFile(const std::string & path) const
|
||||
{
|
||||
return MetadataStorageFromPlainObjectStorage::existsFile(path);
|
||||
}
|
||||
|
||||
bool MetadataStorageFromPlainRewritableObjectStorage::existsDirectory(const std::string & path) const
|
||||
{
|
||||
if (useSeparateLayoutForMetadata())
|
||||
{
|
||||
auto directory = std::filesystem::path(object_storage->generateObjectKeyForPath(path, getMetadataKeyPrefix()).serialize()) / "";
|
||||
return object_storage->existsOrHasAnyChild(directory);
|
||||
}
|
||||
return MetadataStorageFromPlainObjectStorage::isDirectory(path);
|
||||
return MetadataStorageFromPlainObjectStorage::existsDirectory(path);
|
||||
}
|
||||
|
||||
std::vector<std::string> MetadataStorageFromPlainRewritableObjectStorage::listDirectory(const std::string & path) const
|
||||
@ -240,21 +244,12 @@ std::vector<std::string> MetadataStorageFromPlainRewritableObjectStorage::listDi
|
||||
auto key_prefix = object_storage->generateObjectKeyForPath(path, "" /* key_prefix */).serialize();
|
||||
|
||||
RelativePathsWithMetadata files;
|
||||
auto abs_key = std::filesystem::path(object_storage->getCommonKeyPrefix()) / key_prefix / "";
|
||||
auto absolute_key = std::filesystem::path(object_storage->getCommonKeyPrefix()) / key_prefix / "";
|
||||
|
||||
object_storage->listObjects(abs_key, files, 0);
|
||||
object_storage->listObjects(absolute_key, files, 0);
|
||||
|
||||
std::unordered_set<std::string> directories;
|
||||
getDirectChildrenOnDisk(abs_key, files, std::filesystem::path(path) / "", directories);
|
||||
/// List empty directories that are identified by the `prefix.path` metadata files. This is required to, e.g., remove
|
||||
/// metadata along with regular files.
|
||||
if (useSeparateLayoutForMetadata())
|
||||
{
|
||||
auto metadata_key = std::filesystem::path(getMetadataKeyPrefix()) / key_prefix / "";
|
||||
RelativePathsWithMetadata metadata_files;
|
||||
object_storage->listObjects(metadata_key, metadata_files, 0);
|
||||
getDirectChildrenOnDisk(metadata_key, metadata_files, std::filesystem::path(path) / "", directories);
|
||||
}
|
||||
getDirectChildrenOnDisk(absolute_key, files, std::filesystem::path(path) / "", directories);
|
||||
|
||||
return std::vector<std::string>(std::make_move_iterator(directories.begin()), std::make_move_iterator(directories.end()));
|
||||
}
|
||||
|
||||
Some files were not shown because too many files have changed in this diff Show More
Loading…
Reference in New Issue
Block a user