mirror of
https://github.com/ClickHouse/ClickHouse.git
synced 2024-09-19 00:00:48 +00:00
Compare commits
83 Commits
9637a5d32e
...
48d738f867
Author | SHA1 | Date | |
---|---|---|---|
|
48d738f867 | ||
|
d793e06860 | ||
|
1986fb1418 | ||
|
f36408a666 | ||
|
de85f5f251 | ||
|
85af661b9c | ||
|
b42c6491e4 | ||
|
1a4c7b7c61 | ||
|
14feba8443 | ||
|
4c4a051d5e | ||
|
a55cc03973 | ||
|
37411bf240 | ||
|
733c57dae7 | ||
|
918ead070a | ||
|
a461d20af9 | ||
|
8fd9345d2d | ||
|
98a2c1c638 | ||
|
bab574d674 | ||
|
57a6a64d8c | ||
|
de78992966 | ||
|
42670a46d4 | ||
|
b55d0b54ea | ||
|
418ef3f8bc | ||
|
b420bbf855 | ||
|
8111a32c72 | ||
|
721e9a7356 | ||
|
1b1db0081f | ||
|
6a7cfd13f7 | ||
|
51f3245030 | ||
|
baf6aaef1d | ||
|
9ca149a487 | ||
|
2fce90ab76 | ||
|
e0c4a88f98 | ||
|
d5c3adabde | ||
|
08fd6c8ab6 | ||
|
f330fdb1bf | ||
|
042194e3f6 | ||
|
e1a206c84d | ||
|
59763a937e | ||
|
7e99f05981 | ||
|
120e38c72a | ||
|
38b5ea9066 | ||
|
fe5e061fff | ||
|
f6b965872f | ||
|
22c3b71196 | ||
|
1da5729f89 | ||
|
8e3ba4bd6c | ||
|
b30aabf635 | ||
|
dca72d2e0b | ||
|
146edca4f7 | ||
|
3556cf92c3 | ||
|
075a85f15c | ||
|
7425d4aa1a | ||
|
cf12e3924f | ||
|
085be6bf59 | ||
|
263cd9e1c3 | ||
|
5539faa54e | ||
|
be55e1d2e1 | ||
|
d089a56272 | ||
|
11478b949f | ||
|
c9aedee24f | ||
|
32cfdc98b2 | ||
|
d17b20705a | ||
|
cfc931160d | ||
|
b2c4b771d8 | ||
|
edf4e09fb2 | ||
|
2c0ddd10a0 | ||
|
bf4b53390b | ||
|
07f44fdb89 | ||
|
34d02304d3 | ||
|
2fcbe2465a | ||
|
29bc7cf5d5 | ||
|
a39a4b1080 | ||
|
229fffcd56 | ||
|
3c586d80c8 | ||
|
0ff90e4a5c | ||
|
5d6d378f24 | ||
|
14dad25adc | ||
|
906a181b97 | ||
|
7b09ec9ccb | ||
|
d5bea37c96 | ||
|
ad24989b31 | ||
|
d36176ad85 |
34
.github/actions/debug/action.yml
vendored
34
.github/actions/debug/action.yml
vendored
@ -4,15 +4,31 @@ description: Prints workflow debug info
|
||||
runs:
|
||||
using: "composite"
|
||||
steps:
|
||||
- name: Print envs
|
||||
- name: Envs, event.json and contexts
|
||||
shell: bash
|
||||
run: |
|
||||
echo "::group::Envs"
|
||||
env
|
||||
echo "::endgroup::"
|
||||
- name: Print Event.json
|
||||
shell: bash
|
||||
run: |
|
||||
echo "::group::Event.json"
|
||||
echo '::group::Environment variables'
|
||||
env | sort
|
||||
echo '::endgroup::'
|
||||
|
||||
echo '::group::event.json'
|
||||
python3 -m json.tool "$GITHUB_EVENT_PATH"
|
||||
echo "::endgroup::"
|
||||
echo '::endgroup::'
|
||||
|
||||
cat << 'EOF'
|
||||
::group::github context
|
||||
${{ toJSON(github) }}
|
||||
::endgroup::
|
||||
|
||||
::group::env context
|
||||
${{ toJSON(env) }}
|
||||
::endgroup::
|
||||
|
||||
::group::runner context
|
||||
${{ toJSON(runner) }}
|
||||
::endgroup::
|
||||
|
||||
::group::job context
|
||||
${{ toJSON(job) }}
|
||||
::endgroup::
|
||||
EOF
|
||||
|
2
.github/workflows/backport_branches.yml
vendored
2
.github/workflows/backport_branches.yml
vendored
@ -27,6 +27,8 @@ jobs:
|
||||
clear-repository: true # to ensure correct digests
|
||||
fetch-depth: 0 # to get version
|
||||
filter: tree:0
|
||||
- name: Debug Info
|
||||
uses: ./.github/actions/debug
|
||||
- name: Labels check
|
||||
run: |
|
||||
cd "$GITHUB_WORKSPACE/tests/ci"
|
||||
|
2
.github/workflows/cherry_pick.yml
vendored
2
.github/workflows/cherry_pick.yml
vendored
@ -33,6 +33,8 @@ jobs:
|
||||
clear-repository: true
|
||||
token: ${{secrets.ROBOT_CLICKHOUSE_COMMIT_TOKEN}}
|
||||
fetch-depth: 0
|
||||
- name: Debug Info
|
||||
uses: ./.github/actions/debug
|
||||
- name: Cherry pick
|
||||
run: |
|
||||
cd "$GITHUB_WORKSPACE/tests/ci"
|
||||
|
4
.github/workflows/create_release.yml
vendored
4
.github/workflows/create_release.yml
vendored
@ -56,13 +56,13 @@ jobs:
|
||||
GH_TOKEN: ${{ secrets.ROBOT_CLICKHOUSE_COMMIT_TOKEN }}
|
||||
runs-on: [self-hosted, release-maker]
|
||||
steps:
|
||||
- name: DebugInfo
|
||||
uses: hmarr/debug-action@f7318c783045ac39ed9bb497e22ce835fdafbfe6
|
||||
- name: Check out repository code
|
||||
uses: ClickHouse/checkout@v1
|
||||
with:
|
||||
token: ${{secrets.ROBOT_CLICKHOUSE_COMMIT_TOKEN}}
|
||||
fetch-depth: 0
|
||||
- name: Debug Info
|
||||
uses: ./.github/actions/debug
|
||||
- name: Prepare Release Info
|
||||
shell: bash
|
||||
run: |
|
||||
|
1
.github/workflows/docker_test_images.yml
vendored
1
.github/workflows/docker_test_images.yml
vendored
@ -11,6 +11,7 @@ name: Build docker images
|
||||
required: false
|
||||
type: boolean
|
||||
default: false
|
||||
|
||||
jobs:
|
||||
DockerBuildAarch64:
|
||||
runs-on: [self-hosted, style-checker-aarch64]
|
||||
|
7
.github/workflows/jepsen.yml
vendored
7
.github/workflows/jepsen.yml
vendored
@ -8,27 +8,28 @@ on: # yamllint disable-line rule:truthy
|
||||
schedule:
|
||||
- cron: '0 */6 * * *'
|
||||
workflow_dispatch:
|
||||
|
||||
jobs:
|
||||
RunConfig:
|
||||
runs-on: [self-hosted, style-checker-aarch64]
|
||||
outputs:
|
||||
data: ${{ steps.runconfig.outputs.CI_DATA }}
|
||||
steps:
|
||||
- name: DebugInfo
|
||||
uses: hmarr/debug-action@f7318c783045ac39ed9bb497e22ce835fdafbfe6
|
||||
- name: Check out repository code
|
||||
uses: ClickHouse/checkout@v1
|
||||
with:
|
||||
clear-repository: true # to ensure correct digests
|
||||
fetch-depth: 0 # to get version
|
||||
filter: tree:0
|
||||
- name: Debug Info
|
||||
uses: ./.github/actions/debug
|
||||
- name: PrepareRunConfig
|
||||
id: runconfig
|
||||
run: |
|
||||
echo "::group::configure CI run"
|
||||
python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --configure --workflow "$GITHUB_WORKFLOW" --outfile ${{ runner.temp }}/ci_run_data.json
|
||||
echo "::endgroup::"
|
||||
|
||||
|
||||
echo "::group::CI run configure results"
|
||||
python3 -m json.tool ${{ runner.temp }}/ci_run_data.json
|
||||
echo "::endgroup::"
|
||||
|
4
.github/workflows/master.yml
vendored
4
.github/workflows/master.yml
vendored
@ -15,14 +15,14 @@ jobs:
|
||||
outputs:
|
||||
data: ${{ steps.runconfig.outputs.CI_DATA }}
|
||||
steps:
|
||||
- name: DebugInfo
|
||||
uses: hmarr/debug-action@f7318c783045ac39ed9bb497e22ce835fdafbfe6
|
||||
- name: Check out repository code
|
||||
uses: ClickHouse/checkout@v1
|
||||
with:
|
||||
clear-repository: true # to ensure correct digests
|
||||
fetch-depth: 0 # to get version
|
||||
filter: tree:0
|
||||
- name: Debug Info
|
||||
uses: ./.github/actions/debug
|
||||
- name: Merge sync PR
|
||||
run: |
|
||||
cd "$GITHUB_WORKSPACE/tests/ci"
|
||||
|
4
.github/workflows/merge_queue.yml
vendored
4
.github/workflows/merge_queue.yml
vendored
@ -14,14 +14,14 @@ jobs:
|
||||
outputs:
|
||||
data: ${{ steps.runconfig.outputs.CI_DATA }}
|
||||
steps:
|
||||
- name: DebugInfo
|
||||
uses: hmarr/debug-action@f7318c783045ac39ed9bb497e22ce835fdafbfe6
|
||||
- name: Check out repository code
|
||||
uses: ClickHouse/checkout@v1
|
||||
with:
|
||||
clear-repository: true # to ensure correct digests
|
||||
fetch-depth: 0 # to get a version
|
||||
filter: tree:0
|
||||
- name: Debug Info
|
||||
uses: ./.github/actions/debug
|
||||
- name: Cancel PR workflow
|
||||
run: |
|
||||
python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --cancel-previous-run
|
||||
|
4
.github/workflows/nightly.yml
vendored
4
.github/workflows/nightly.yml
vendored
@ -15,14 +15,14 @@ jobs:
|
||||
outputs:
|
||||
data: ${{ steps.runconfig.outputs.CI_DATA }}
|
||||
steps:
|
||||
- name: DebugInfo
|
||||
uses: hmarr/debug-action@f7318c783045ac39ed9bb497e22ce835fdafbfe6
|
||||
- name: Check out repository code
|
||||
uses: ClickHouse/checkout@v1
|
||||
with:
|
||||
clear-repository: true # to ensure correct digests
|
||||
fetch-depth: 0 # to get version
|
||||
filter: tree:0
|
||||
- name: Debug Info
|
||||
uses: ./.github/actions/debug
|
||||
- name: PrepareRunConfig
|
||||
id: runconfig
|
||||
run: |
|
||||
|
4
.github/workflows/pull_request.yml
vendored
4
.github/workflows/pull_request.yml
vendored
@ -25,14 +25,14 @@ jobs:
|
||||
outputs:
|
||||
data: ${{ steps.runconfig.outputs.CI_DATA }}
|
||||
steps:
|
||||
- name: DebugInfo
|
||||
uses: hmarr/debug-action@f7318c783045ac39ed9bb497e22ce835fdafbfe6
|
||||
- name: Check out repository code
|
||||
uses: ClickHouse/checkout@v1
|
||||
with:
|
||||
clear-repository: true # to ensure correct digests
|
||||
fetch-depth: 0 # to get a version
|
||||
filter: tree:0
|
||||
- name: Debug Info
|
||||
uses: ./.github/actions/debug
|
||||
- name: Cancel previous Sync PR workflow
|
||||
run: |
|
||||
python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --cancel-previous-run
|
||||
|
2
.github/workflows/release_branches.yml
vendored
2
.github/workflows/release_branches.yml
vendored
@ -24,6 +24,8 @@ jobs:
|
||||
clear-repository: true # to ensure correct digests
|
||||
fetch-depth: 0 # to get version
|
||||
filter: tree:0
|
||||
- name: Debug Info
|
||||
uses: ./.github/actions/debug
|
||||
- name: Labels check
|
||||
run: |
|
||||
cd "$GITHUB_WORKSPACE/tests/ci"
|
||||
|
4
.github/workflows/reusable_simple_job.yml
vendored
4
.github/workflows/reusable_simple_job.yml
vendored
@ -62,8 +62,6 @@ jobs:
|
||||
env:
|
||||
GITHUB_JOB_OVERRIDDEN: ${{inputs.test_name}}
|
||||
steps:
|
||||
- name: DebugInfo
|
||||
uses: hmarr/debug-action@f7318c783045ac39ed9bb497e22ce835fdafbfe6
|
||||
- name: Check out repository code
|
||||
uses: ClickHouse/checkout@v1
|
||||
with:
|
||||
@ -72,6 +70,8 @@ jobs:
|
||||
submodules: ${{inputs.submodules}}
|
||||
fetch-depth: ${{inputs.checkout_depth}}
|
||||
filter: tree:0
|
||||
- name: Debug Info
|
||||
uses: ./.github/actions/debug
|
||||
- name: Set build envs
|
||||
run: |
|
||||
cat >> "$GITHUB_ENV" << 'EOF'
|
||||
|
@ -109,7 +109,7 @@ if [ -n "$CLICKHOUSE_USER" ] && [ "$CLICKHOUSE_USER" != "default" ] || [ -n "$CL
|
||||
<networks>
|
||||
<ip>::/0</ip>
|
||||
</networks>
|
||||
<password>${CLICKHOUSE_PASSWORD}</password>
|
||||
<password><![CDATA[${CLICKHOUSE_PASSWORD//]]>/]]]]><![CDATA[>}]]></password>
|
||||
<quota>default</quota>
|
||||
<access_management>${CLICKHOUSE_ACCESS_MANAGEMENT}</access_management>
|
||||
</${CLICKHOUSE_USER}>
|
||||
|
@ -13,16 +13,17 @@ Here is a complete list of available database engines. Follow the links for more
|
||||
|
||||
- [Atomic](../../engines/database-engines/atomic.md)
|
||||
|
||||
- [MySQL](../../engines/database-engines/mysql.md)
|
||||
- [Lazy](../../engines/database-engines/lazy.md)
|
||||
|
||||
- [MaterializedPostgreSQL](../../engines/database-engines/materialized-postgresql.md)
|
||||
|
||||
- [MaterializedMySQL](../../engines/database-engines/materialized-mysql.md)
|
||||
|
||||
- [Lazy](../../engines/database-engines/lazy.md)
|
||||
- [MySQL](../../engines/database-engines/mysql.md)
|
||||
|
||||
- [PostgreSQL](../../engines/database-engines/postgresql.md)
|
||||
|
||||
- [MaterializedPostgreSQL](../../engines/database-engines/materialized-postgresql.md)
|
||||
|
||||
- [Replicated](../../engines/database-engines/replicated.md)
|
||||
|
||||
- [SQLite](../../engines/database-engines/sqlite.md)
|
||||
|
||||
|
@ -107,6 +107,10 @@ The vector similarity index currently does not work with per-table, non-default
|
||||
[here](https://github.com/ClickHouse/ClickHouse/pull/51325#issuecomment-1605920475)). If necessary, the value must be changed in config.xml.
|
||||
:::
|
||||
|
||||
Vector index creation is known to be slow. To speed the process up, index creation can be parallelized. The maximum number of threads can be
|
||||
configured using server configuration
|
||||
setting [max_build_vector_similarity_index_thread_pool_size](../../../operations/server-configuration-parameters/settings.md#server_configuration_parameters_max_build_vector_similarity_index_thread_pool_size).
|
||||
|
||||
ANN indexes are built during column insertion and merge. As a result, `INSERT` and `OPTIMIZE` statements will be slower than for ordinary
|
||||
tables. ANNIndexes are ideally used only with immutable or rarely changed data, respectively when are far more read requests than write
|
||||
requests.
|
||||
|
@ -1396,6 +1396,7 @@ SELECT * FROM json_each_row_nested
|
||||
- [input_format_json_ignore_unknown_keys_in_named_tuple](/docs/en/operations/settings/settings-formats.md/#input_format_json_ignore_unknown_keys_in_named_tuple) - ignore unknown keys in json object for named tuples. Default value - `false`.
|
||||
- [input_format_json_compact_allow_variable_number_of_columns](/docs/en/operations/settings/settings-formats.md/#input_format_json_compact_allow_variable_number_of_columns) - allow variable number of columns in JSONCompact/JSONCompactEachRow format, ignore extra columns and use default values on missing columns. Default value - `false`.
|
||||
- [input_format_json_throw_on_bad_escape_sequence](/docs/en/operations/settings/settings-formats.md/#input_format_json_throw_on_bad_escape_sequence) - throw an exception if JSON string contains bad escape sequence. If disabled, bad escape sequences will remain as is in the data. Default value - `true`.
|
||||
- [input_format_json_empty_as_default](/docs/en/operations/settings/settings-formats.md/#input_format_json_empty_as_default) - treat empty fields in JSON input as default values. Default value - `false`. For complex default expressions [input_format_defaults_for_omitted_fields](/docs/en/operations/settings/settings-formats.md/#input_format_defaults_for_omitted_fields) must be enabled too.
|
||||
- [output_format_json_quote_64bit_integers](/docs/en/operations/settings/settings-formats.md/#output_format_json_quote_64bit_integers) - controls quoting of 64-bit integers in JSON output format. Default value - `true`.
|
||||
- [output_format_json_quote_64bit_floats](/docs/en/operations/settings/settings-formats.md/#output_format_json_quote_64bit_floats) - controls quoting of 64-bit floats in JSON output format. Default value - `false`.
|
||||
- [output_format_json_quote_denormals](/docs/en/operations/settings/settings-formats.md/#output_format_json_quote_denormals) - enables '+nan', '-nan', '+inf', '-inf' outputs in JSON output format. Default value - `false`.
|
||||
|
@ -491,6 +491,14 @@ Type: Double
|
||||
|
||||
Default: 0.9
|
||||
|
||||
## max_build_vector_similarity_index_thread_pool_size {#server_configuration_parameters_max_build_vector_similarity_index_thread_pool_size}
|
||||
|
||||
The maximum number of threads to use for building vector indexes. 0 means all cores.
|
||||
|
||||
Type: UInt64
|
||||
|
||||
Default: 16
|
||||
|
||||
## cgroups_memory_usage_observer_wait_time
|
||||
|
||||
Interval in seconds during which the server's maximum allowed memory consumption is adjusted by the corresponding threshold in cgroups. (see
|
||||
|
@ -752,6 +752,17 @@ Possible values:
|
||||
|
||||
Default value: 0.
|
||||
|
||||
### input_format_json_empty_as_default {#input_format_json_empty_as_default}
|
||||
|
||||
When enabled, replace empty input fields in JSON with default values. For complex default expressions `input_format_defaults_for_omitted_fields` must be enabled too.
|
||||
|
||||
Possible values:
|
||||
|
||||
+ 0 — Disable.
|
||||
+ 1 — Enable.
|
||||
|
||||
Default value: 0.
|
||||
|
||||
## TSV format settings {#tsv-format-settings}
|
||||
|
||||
### input_format_tsv_empty_as_default {#input_format_tsv_empty_as_default}
|
||||
|
@ -178,6 +178,9 @@
|
||||
M(ObjectStorageAzureThreads, "Number of threads in the AzureObjectStorage thread pool.") \
|
||||
M(ObjectStorageAzureThreadsActive, "Number of threads in the AzureObjectStorage thread pool running a task.") \
|
||||
M(ObjectStorageAzureThreadsScheduled, "Number of queued or active jobs in the AzureObjectStorage thread pool.") \
|
||||
M(BuildVectorSimilarityIndexThreads, "Number of threads in the build vector similarity index thread pool.") \
|
||||
M(BuildVectorSimilarityIndexThreadsActive, "Number of threads in the build vector similarity index thread pool running a task.") \
|
||||
M(BuildVectorSimilarityIndexThreadsScheduled, "Number of queued or active jobs in the build vector similarity index thread pool.") \
|
||||
\
|
||||
M(DiskPlainRewritableAzureDirectoryMapSize, "Number of local-to-remote path entries in the 'plain_rewritable' in-memory map for AzureObjectStorage.") \
|
||||
M(DiskPlainRewritableLocalDirectoryMapSize, "Number of local-to-remote path entries in the 'plain_rewritable' in-memory map for LocalObjectStorage.") \
|
||||
|
@ -609,6 +609,7 @@
|
||||
M(728, UNEXPECTED_DATA_TYPE) \
|
||||
M(729, ILLEGAL_TIME_SERIES_TAGS) \
|
||||
M(730, REFRESH_FAILED) \
|
||||
M(731, QUERY_CACHE_USED_WITH_NON_THROW_OVERFLOW_MODE) \
|
||||
\
|
||||
M(900, DISTRIBUTED_CACHE_ERROR) \
|
||||
M(901, CANNOT_USE_DISTRIBUTED_CACHE) \
|
||||
|
@ -63,6 +63,7 @@ static struct InitFiu
|
||||
REGULAR(keepermap_fail_drop_data) \
|
||||
REGULAR(lazy_pipe_fds_fail_close) \
|
||||
PAUSEABLE(infinite_sleep) \
|
||||
PAUSEABLE(stop_moving_part_before_swap_with_active) \
|
||||
|
||||
|
||||
namespace FailPoints
|
||||
|
@ -57,11 +57,13 @@ namespace ErrorCodes
|
||||
namespace DB
|
||||
{
|
||||
|
||||
static bool supportsAtomicRenameImpl()
|
||||
static std::optional<std::string> supportsAtomicRenameImpl()
|
||||
{
|
||||
VersionNumber renameat2_minimal_version(3, 15, 0);
|
||||
VersionNumber linux_version(Poco::Environment::osVersion());
|
||||
return linux_version >= renameat2_minimal_version;
|
||||
if (linux_version >= renameat2_minimal_version)
|
||||
return std::nullopt;
|
||||
return fmt::format("Linux kernel 3.15+ is required, got {}", linux_version.toString());
|
||||
}
|
||||
|
||||
static bool renameat2(const std::string & old_path, const std::string & new_path, int flags)
|
||||
@ -97,10 +99,14 @@ static bool renameat2(const std::string & old_path, const std::string & new_path
|
||||
ErrnoException::throwFromPath(ErrorCodes::SYSTEM_ERROR, new_path, "Cannot rename {} to {}", old_path, new_path);
|
||||
}
|
||||
|
||||
bool supportsAtomicRename()
|
||||
bool supportsAtomicRename(std::string * out_message)
|
||||
{
|
||||
static bool supports = supportsAtomicRenameImpl();
|
||||
return supports;
|
||||
static auto error = supportsAtomicRenameImpl();
|
||||
if (!error.has_value())
|
||||
return true;
|
||||
if (out_message)
|
||||
*out_message = error.value();
|
||||
return false;
|
||||
}
|
||||
|
||||
}
|
||||
@ -152,16 +158,22 @@ static bool renameat2(const std::string & old_path, const std::string & new_path
|
||||
}
|
||||
|
||||
|
||||
static bool supportsAtomicRenameImpl()
|
||||
static std::optional<std::string> supportsAtomicRenameImpl()
|
||||
{
|
||||
auto fun = dlsym(RTLD_DEFAULT, "renamex_np");
|
||||
return fun != nullptr;
|
||||
if (fun != nullptr)
|
||||
return std::nullopt;
|
||||
return "macOS 10.12 or later is required";
|
||||
}
|
||||
|
||||
bool supportsAtomicRename()
|
||||
bool supportsAtomicRename(std::string * out_message)
|
||||
{
|
||||
static bool supports = supportsAtomicRenameImpl();
|
||||
return supports;
|
||||
static auto error = supportsAtomicRenameImpl();
|
||||
if (!error.has_value())
|
||||
return true;
|
||||
if (out_message)
|
||||
*out_message = error.value();
|
||||
return false;
|
||||
}
|
||||
|
||||
}
|
||||
@ -179,8 +191,10 @@ static bool renameat2(const std::string &, const std::string &, int)
|
||||
return false;
|
||||
}
|
||||
|
||||
bool supportsAtomicRename()
|
||||
bool supportsAtomicRename(std::string * out_message)
|
||||
{
|
||||
if (out_message)
|
||||
*out_message = "only Linux and macOS are supported";
|
||||
return false;
|
||||
}
|
||||
|
||||
|
@ -6,7 +6,7 @@ namespace DB
|
||||
{
|
||||
|
||||
/// Returns true, if the following functions supported by the system
|
||||
bool supportsAtomicRename();
|
||||
bool supportsAtomicRename(std::string * out_message = nullptr);
|
||||
|
||||
/// Atomically rename old_path to new_path. If new_path exists, do not overwrite it and throw exception
|
||||
void renameNoReplace(const std::string & old_path, const std::string & new_path);
|
||||
|
@ -50,7 +50,7 @@ namespace DB
|
||||
M(UInt32, asynchronous_heavy_metrics_update_period_s, 120, "Period in seconds for updating heavy asynchronous metrics.", 0) \
|
||||
M(String, default_database, "default", "Default database name.", 0) \
|
||||
M(String, tmp_policy, "", "Policy for storage with temporary data.", 0) \
|
||||
M(UInt64, max_temporary_data_on_disk_size, 0, "The maximum amount of storage that could be used for external aggregation, joins or sorting., ", 0) \
|
||||
M(UInt64, max_temporary_data_on_disk_size, 0, "The maximum amount of storage that could be used for external aggregation, joins or sorting.", 0) \
|
||||
M(String, temporary_data_in_cache, "", "Cache disk name for temporary data.", 0) \
|
||||
M(UInt64, aggregate_function_group_array_max_element_size, 0xFFFFFF, "Max array element size in bytes for groupArray function. This limit is checked at serialization and help to avoid large state size.", 0) \
|
||||
M(GroupArrayActionWhenLimitReached, aggregate_function_group_array_action_when_limit_is_reached, GroupArrayActionWhenLimitReached::THROW, "Action to execute when max array element size is exceeded in groupArray: `throw` exception, or `discard` extra values", 0) \
|
||||
@ -65,6 +65,7 @@ namespace DB
|
||||
M(UInt64, async_insert_threads, 16, "Maximum number of threads to actually parse and insert data in background. Zero means asynchronous mode is disabled", 0) \
|
||||
M(Bool, async_insert_queue_flush_on_shutdown, true, "If true queue of asynchronous inserts is flushed on graceful shutdown", 0) \
|
||||
M(Bool, ignore_empty_sql_security_in_create_view_query, true, "If true, ClickHouse doesn't write defaults for empty SQL security statement in CREATE VIEW queries. This setting is only necessary for the migration period and will become obsolete in 24.4", 0) \
|
||||
M(UInt64, max_build_vector_similarity_index_thread_pool_size, 16, "The maximum number of threads to use to build vector similarity indexes. 0 means all cores.", 0) \
|
||||
\
|
||||
/* Database Catalog */ \
|
||||
M(UInt64, database_atomic_delay_before_drop_table_sec, 8 * 60, "The delay during which a dropped table can be restored using the UNDROP statement. If DROP TABLE ran with a SYNC modifier, the setting is ignored.", 0) \
|
||||
|
@ -1144,6 +1144,7 @@ class IColumn;
|
||||
M(Bool, input_format_try_infer_variants, false, "Try to infer the Variant type in text formats when there is more than one possible type for column/array elements", 0) \
|
||||
M(Bool, type_json_skip_duplicated_paths, false, "When enabled, during parsing JSON object into JSON type duplicated paths will be ignored and only the first one will be inserted instead of an exception", 0) \
|
||||
M(UInt64, input_format_json_max_depth, 1000, "Maximum depth of a field in JSON. This is not a strict limit, it does not have to be applied precisely.", 0) \
|
||||
M(Bool, input_format_json_empty_as_default, false, "Treat empty fields in JSON input as default values.", 0) \
|
||||
M(Bool, input_format_try_infer_integers, true, "Try to infer integers instead of floats while schema inference in text formats", 0) \
|
||||
M(Bool, input_format_try_infer_dates, true, "Try to infer dates from string fields while schema inference in text formats", 0) \
|
||||
M(Bool, input_format_try_infer_datetimes, true, "Try to infer datetimes from string fields while schema inference in text formats", 0) \
|
||||
|
@ -71,6 +71,7 @@ static std::initializer_list<std::pair<ClickHouseVersion, SettingsChangesHistory
|
||||
},
|
||||
{"24.9",
|
||||
{
|
||||
{"input_format_json_empty_as_default", false, false, "Added new setting to allow to treat empty fields in JSON input as default values."},
|
||||
{"input_format_try_infer_variants", false, false, "Try to infer Variant type in text formats when there is more than one possible type for column/array elements"},
|
||||
{"join_output_by_rowlist_perkey_rows_threshold", 0, 5, "The lower limit of per-key average rows in the right table to determine whether to output by row list in hash join."},
|
||||
{"create_if_not_exists", false, false, "New setting."},
|
||||
|
@ -11,6 +11,7 @@
|
||||
#include <IO/WriteBufferFromString.h>
|
||||
|
||||
#include <Formats/FormatSettings.h>
|
||||
#include <Formats/JSONUtils.h>
|
||||
|
||||
namespace DB
|
||||
{
|
||||
@ -615,28 +616,49 @@ void SerializationArray::serializeTextJSONPretty(const IColumn & column, size_t
|
||||
}
|
||||
|
||||
|
||||
void SerializationArray::deserializeTextJSON(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const
|
||||
template <typename ReturnType>
|
||||
ReturnType SerializationArray::deserializeTextJSONImpl(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const
|
||||
{
|
||||
deserializeTextImpl(column, istr,
|
||||
[&](IColumn & nested_column)
|
||||
auto deserialize_nested = [&settings, this](IColumn & nested_column, ReadBuffer & buf) -> ReturnType
|
||||
{
|
||||
if constexpr (std::is_same_v<ReturnType, void>)
|
||||
{
|
||||
if (settings.null_as_default && !isColumnNullableOrLowCardinalityNullable(nested_column))
|
||||
SerializationNullable::deserializeNullAsDefaultOrNestedTextJSON(nested_column, istr, settings, nested);
|
||||
SerializationNullable::deserializeNullAsDefaultOrNestedTextJSON(nested_column, buf, settings, nested);
|
||||
else
|
||||
nested->deserializeTextJSON(nested_column, istr, settings);
|
||||
}, false);
|
||||
nested->deserializeTextJSON(nested_column, buf, settings);
|
||||
}
|
||||
else
|
||||
{
|
||||
if (settings.null_as_default && !isColumnNullableOrLowCardinalityNullable(nested_column))
|
||||
return SerializationNullable::tryDeserializeNullAsDefaultOrNestedTextJSON(nested_column, buf, settings, nested);
|
||||
return nested->tryDeserializeTextJSON(nested_column, buf, settings);
|
||||
}
|
||||
};
|
||||
|
||||
if (settings.json.empty_as_default)
|
||||
return deserializeTextImpl<ReturnType>(column, istr,
|
||||
[&deserialize_nested, &istr](IColumn & nested_column) -> ReturnType
|
||||
{
|
||||
return JSONUtils::deserializeEmpyStringAsDefaultOrNested<ReturnType>(nested_column, istr, deserialize_nested);
|
||||
}, false);
|
||||
else
|
||||
return deserializeTextImpl<ReturnType>(column, istr,
|
||||
[&deserialize_nested, &istr](IColumn & nested_column) -> ReturnType
|
||||
{
|
||||
return deserialize_nested(nested_column, istr);
|
||||
}, false);
|
||||
}
|
||||
|
||||
|
||||
void SerializationArray::deserializeTextJSON(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const
|
||||
{
|
||||
deserializeTextJSONImpl<void>(column, istr, settings);
|
||||
}
|
||||
|
||||
bool SerializationArray::tryDeserializeTextJSON(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const
|
||||
{
|
||||
auto read_nested = [&](IColumn & nested_column)
|
||||
{
|
||||
if (settings.null_as_default && !isColumnNullableOrLowCardinalityNullable(nested_column))
|
||||
return SerializationNullable::tryDeserializeNullAsDefaultOrNestedTextJSON(nested_column, istr, settings, nested);
|
||||
return nested->tryDeserializeTextJSON(nested_column, istr, settings);
|
||||
};
|
||||
|
||||
return deserializeTextImpl<bool>(column, istr, std::move(read_nested), false);
|
||||
return deserializeTextJSONImpl<bool>(column, istr, settings);
|
||||
}
|
||||
|
||||
|
||||
|
@ -82,6 +82,10 @@ public:
|
||||
SerializationPtr create(const SerializationPtr & prev) const override;
|
||||
ColumnPtr create(const ColumnPtr & prev) const override;
|
||||
};
|
||||
|
||||
private:
|
||||
template <typename ReturnType>
|
||||
ReturnType deserializeTextJSONImpl(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const;
|
||||
};
|
||||
|
||||
}
|
||||
|
@ -6,6 +6,7 @@
|
||||
#include <Columns/ColumnMap.h>
|
||||
#include <Core/Field.h>
|
||||
#include <Formats/FormatSettings.h>
|
||||
#include <Formats/JSONUtils.h>
|
||||
#include <Common/assert_cast.h>
|
||||
#include <Common/quoteString.h>
|
||||
#include <IO/WriteHelpers.h>
|
||||
@ -316,28 +317,52 @@ void SerializationMap::serializeTextJSONPretty(const IColumn & column, size_t ro
|
||||
}
|
||||
|
||||
|
||||
void SerializationMap::deserializeTextJSON(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const
|
||||
template <typename ReturnType>
|
||||
ReturnType SerializationMap::deserializeTextJSONImpl(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const
|
||||
{
|
||||
deserializeTextImpl(column, istr,
|
||||
[&settings](ReadBuffer & buf, const SerializationPtr & subcolumn_serialization, IColumn & subcolumn)
|
||||
auto deserialize_nested = [&settings](IColumn & subcolumn, ReadBuffer & buf, const SerializationPtr & subcolumn_serialization) -> ReturnType
|
||||
{
|
||||
if constexpr (std::is_same_v<ReturnType, void>)
|
||||
{
|
||||
if (settings.null_as_default && !isColumnNullableOrLowCardinalityNullable(subcolumn))
|
||||
SerializationNullable::deserializeNullAsDefaultOrNestedTextJSON(subcolumn, buf, settings, subcolumn_serialization);
|
||||
else
|
||||
subcolumn_serialization->deserializeTextJSON(subcolumn, buf, settings);
|
||||
});
|
||||
}
|
||||
else
|
||||
{
|
||||
if (settings.null_as_default && !isColumnNullableOrLowCardinalityNullable(subcolumn))
|
||||
return SerializationNullable::tryDeserializeNullAsDefaultOrNestedTextJSON(subcolumn, buf, settings, subcolumn_serialization);
|
||||
return subcolumn_serialization->tryDeserializeTextJSON(subcolumn, buf, settings);
|
||||
}
|
||||
};
|
||||
|
||||
if (settings.json.empty_as_default)
|
||||
return deserializeTextImpl<ReturnType>(column, istr,
|
||||
[&deserialize_nested](ReadBuffer & buf, const SerializationPtr & subcolumn_serialization, IColumn & subcolumn) -> ReturnType
|
||||
{
|
||||
return JSONUtils::deserializeEmpyStringAsDefaultOrNested<ReturnType>(subcolumn, buf,
|
||||
[&deserialize_nested, &subcolumn_serialization](IColumn & subcolumn_, ReadBuffer & buf_) -> ReturnType
|
||||
{
|
||||
return deserialize_nested(subcolumn_, buf_, subcolumn_serialization);
|
||||
});
|
||||
});
|
||||
else
|
||||
return deserializeTextImpl<ReturnType>(column, istr,
|
||||
[&deserialize_nested](ReadBuffer & buf, const SerializationPtr & subcolumn_serialization, IColumn & subcolumn) -> ReturnType
|
||||
{
|
||||
return deserialize_nested(subcolumn, buf, subcolumn_serialization);
|
||||
});
|
||||
}
|
||||
|
||||
void SerializationMap::deserializeTextJSON(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const
|
||||
{
|
||||
deserializeTextJSONImpl<void>(column, istr, settings);
|
||||
}
|
||||
|
||||
bool SerializationMap::tryDeserializeTextJSON(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const
|
||||
{
|
||||
auto reader = [&settings](ReadBuffer & buf, const SerializationPtr & subcolumn_serialization, IColumn & subcolumn)
|
||||
{
|
||||
if (settings.null_as_default && !isColumnNullableOrLowCardinalityNullable(subcolumn))
|
||||
return SerializationNullable::tryDeserializeNullAsDefaultOrNestedTextJSON(subcolumn, buf, settings, subcolumn_serialization);
|
||||
return subcolumn_serialization->tryDeserializeTextJSON(subcolumn, buf, settings);
|
||||
};
|
||||
|
||||
return deserializeTextImpl<bool>(column, istr, reader);
|
||||
return deserializeTextJSONImpl<bool>(column, istr, settings);
|
||||
}
|
||||
|
||||
void SerializationMap::serializeTextXML(const IColumn & column, size_t row_num, WriteBuffer & ostr, const FormatSettings & settings) const
|
||||
|
@ -74,6 +74,9 @@ private:
|
||||
|
||||
template <typename ReturnType = void, typename Reader>
|
||||
ReturnType deserializeTextImpl(IColumn & column, ReadBuffer & istr, Reader && reader) const;
|
||||
|
||||
template <typename ReturnType>
|
||||
ReturnType deserializeTextJSONImpl(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const;
|
||||
};
|
||||
|
||||
}
|
||||
|
@ -5,6 +5,7 @@
|
||||
#include <Core/Field.h>
|
||||
#include <Columns/ColumnTuple.h>
|
||||
#include <Common/assert_cast.h>
|
||||
#include <Formats/JSONUtils.h>
|
||||
#include <IO/WriteHelpers.h>
|
||||
#include <IO/ReadHelpers.h>
|
||||
#include <IO/ReadBufferFromString.h>
|
||||
@ -313,27 +314,9 @@ void SerializationTuple::serializeTextJSONPretty(const IColumn & column, size_t
|
||||
}
|
||||
|
||||
template <typename ReturnType>
|
||||
ReturnType SerializationTuple::deserializeTextJSONImpl(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const
|
||||
ReturnType SerializationTuple::deserializeTupleJSONImpl(IColumn & column, ReadBuffer & istr, const FormatSettings & settings, auto && deserialize_element) const
|
||||
{
|
||||
static constexpr bool throw_exception = std::is_same_v<ReturnType, void>;
|
||||
|
||||
auto deserialize_element = [&](IColumn & element_column, size_t element_pos)
|
||||
{
|
||||
if constexpr (throw_exception)
|
||||
{
|
||||
if (settings.null_as_default && !isColumnNullableOrLowCardinalityNullable(element_column))
|
||||
SerializationNullable::deserializeNullAsDefaultOrNestedTextJSON(element_column, istr, settings, elems[element_pos]);
|
||||
else
|
||||
elems[element_pos]->deserializeTextJSON(element_column, istr, settings);
|
||||
return true;
|
||||
}
|
||||
else
|
||||
{
|
||||
if (settings.null_as_default && !isColumnNullableOrLowCardinalityNullable(element_column))
|
||||
return SerializationNullable::tryDeserializeNullAsDefaultOrNestedTextJSON(element_column, istr, settings, elems[element_pos]);
|
||||
return elems[element_pos]->tryDeserializeTextJSON(element_column, istr, settings);
|
||||
}
|
||||
};
|
||||
static constexpr auto throw_exception = std::is_same_v<ReturnType, void>;
|
||||
|
||||
if (settings.json.read_named_tuples_as_objects
|
||||
&& have_explicit_names)
|
||||
@ -506,12 +489,51 @@ ReturnType SerializationTuple::deserializeTextJSONImpl(IColumn & column, ReadBuf
|
||||
}
|
||||
}
|
||||
|
||||
void SerializationTuple::deserializeTextJSON(DB::IColumn & column, DB::ReadBuffer & istr, const DB::FormatSettings & settings) const
|
||||
template <typename ReturnType>
|
||||
ReturnType SerializationTuple::deserializeTextJSONImpl(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const
|
||||
{
|
||||
deserializeTextJSONImpl(column, istr, settings);
|
||||
auto deserialize_nested = [&settings](IColumn & nested_column, ReadBuffer & buf, const SerializationPtr & nested_column_serialization) -> ReturnType
|
||||
{
|
||||
if constexpr (std::is_same_v<ReturnType, void>)
|
||||
{
|
||||
if (settings.null_as_default && !isColumnNullableOrLowCardinalityNullable(nested_column))
|
||||
SerializationNullable::deserializeNullAsDefaultOrNestedTextJSON(nested_column, buf, settings, nested_column_serialization);
|
||||
else
|
||||
nested_column_serialization->deserializeTextJSON(nested_column, buf, settings);
|
||||
}
|
||||
else
|
||||
{
|
||||
if (settings.null_as_default && !isColumnNullableOrLowCardinalityNullable(nested_column))
|
||||
return SerializationNullable::tryDeserializeNullAsDefaultOrNestedTextJSON(nested_column, buf, settings, nested_column_serialization);
|
||||
else
|
||||
return nested_column_serialization->tryDeserializeTextJSON(nested_column, buf, settings);
|
||||
}
|
||||
};
|
||||
|
||||
if (settings.json.empty_as_default)
|
||||
return deserializeTupleJSONImpl<ReturnType>(column, istr, settings,
|
||||
[&deserialize_nested, &istr, this](IColumn & nested_column, size_t element_pos) -> ReturnType
|
||||
{
|
||||
return JSONUtils::deserializeEmpyStringAsDefaultOrNested<ReturnType>(nested_column, istr,
|
||||
[&deserialize_nested, element_pos, this](IColumn & nested_column_, ReadBuffer & buf) -> ReturnType
|
||||
{
|
||||
return deserialize_nested(nested_column_, buf, elems[element_pos]);
|
||||
});
|
||||
});
|
||||
else
|
||||
return deserializeTupleJSONImpl<ReturnType>(column, istr, settings,
|
||||
[&deserialize_nested, &istr, this](IColumn & nested_column, size_t element_pos) -> ReturnType
|
||||
{
|
||||
return deserialize_nested(nested_column, istr, elems[element_pos]);
|
||||
});
|
||||
}
|
||||
|
||||
bool SerializationTuple::tryDeserializeTextJSON(DB::IColumn & column, DB::ReadBuffer & istr, const DB::FormatSettings & settings) const
|
||||
void SerializationTuple::deserializeTextJSON(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const
|
||||
{
|
||||
deserializeTextJSONImpl<void>(column, istr, settings);
|
||||
}
|
||||
|
||||
bool SerializationTuple::tryDeserializeTextJSON(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const
|
||||
{
|
||||
return deserializeTextJSONImpl<bool>(column, istr, settings);
|
||||
}
|
||||
|
@ -81,7 +81,10 @@ private:
|
||||
template <typename ReturnType = void>
|
||||
ReturnType deserializeTextImpl(IColumn & column, ReadBuffer & istr, const FormatSettings & settings, bool whole) const;
|
||||
|
||||
template <typename ReturnType = void>
|
||||
template <typename ReturnType>
|
||||
ReturnType deserializeTupleJSONImpl(IColumn & column, ReadBuffer & istr, const FormatSettings & settings, auto && deserialize_element) const;
|
||||
|
||||
template <typename ReturnType>
|
||||
ReturnType deserializeTextJSONImpl(IColumn & column, ReadBuffer & istr, const FormatSettings & settings) const;
|
||||
|
||||
template <typename ReturnType = void>
|
||||
|
@ -197,8 +197,9 @@ void DatabaseAtomic::renameTable(ContextPtr local_context, const String & table_
|
||||
throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Moving tables between databases of different engines is not supported");
|
||||
}
|
||||
|
||||
if (exchange && !supportsAtomicRename())
|
||||
throw Exception(ErrorCodes::NOT_IMPLEMENTED, "RENAME EXCHANGE is not supported");
|
||||
std::string message;
|
||||
if (exchange && !supportsAtomicRename(&message))
|
||||
throw Exception(ErrorCodes::NOT_IMPLEMENTED, "RENAME EXCHANGE is not supported because exchanging files is not supported by the OS ({})", message);
|
||||
|
||||
waitDatabaseStarted();
|
||||
|
||||
|
@ -152,6 +152,7 @@ FormatSettings getFormatSettings(const ContextPtr & context, const Settings & se
|
||||
format_settings.json.try_infer_objects_as_tuples = settings.input_format_json_try_infer_named_tuples_from_objects;
|
||||
format_settings.json.throw_on_bad_escape_sequence = settings.input_format_json_throw_on_bad_escape_sequence;
|
||||
format_settings.json.ignore_unnecessary_fields = settings.input_format_json_ignore_unnecessary_fields;
|
||||
format_settings.json.empty_as_default = settings.input_format_json_empty_as_default;
|
||||
format_settings.json.type_json_skip_duplicated_paths = settings.type_json_skip_duplicated_paths;
|
||||
format_settings.null_as_default = settings.input_format_null_as_default;
|
||||
format_settings.force_null_for_omitted_fields = settings.input_format_force_null_for_omitted_fields;
|
||||
|
@ -237,6 +237,7 @@ struct FormatSettings
|
||||
bool infer_incomplete_types_as_strings = true;
|
||||
bool throw_on_bad_escape_sequence = true;
|
||||
bool ignore_unnecessary_fields = true;
|
||||
bool empty_as_default = false;
|
||||
bool type_json_skip_duplicated_paths = false;
|
||||
} json{};
|
||||
|
||||
|
@ -2,12 +2,14 @@
|
||||
#include <Formats/JSONUtils.h>
|
||||
#include <Formats/ReadSchemaUtils.h>
|
||||
#include <Formats/EscapingRuleUtils.h>
|
||||
#include <IO/PeekableReadBuffer.h>
|
||||
#include <IO/ReadBufferFromString.h>
|
||||
#include <IO/WriteBufferValidUTF8.h>
|
||||
#include <DataTypes/Serializations/SerializationNullable.h>
|
||||
#include <DataTypes/DataTypeNullable.h>
|
||||
#include <DataTypes/DataTypeObjectDeprecated.h>
|
||||
#include <DataTypes/DataTypeFactory.h>
|
||||
#include <Common/assert_cast.h>
|
||||
|
||||
#include <base/find_symbols.h>
|
||||
|
||||
@ -286,11 +288,19 @@ namespace JSONUtils
|
||||
return true;
|
||||
}
|
||||
|
||||
if (as_nullable)
|
||||
return SerializationNullable::deserializeNullAsDefaultOrNestedTextJSON(column, in, format_settings, serialization);
|
||||
auto deserialize = [as_nullable, &format_settings, &serialization](IColumn & column_, ReadBuffer & buf) -> bool
|
||||
{
|
||||
if (as_nullable)
|
||||
return SerializationNullable::deserializeNullAsDefaultOrNestedTextJSON(column_, buf, format_settings, serialization);
|
||||
|
||||
serialization->deserializeTextJSON(column, in, format_settings);
|
||||
return true;
|
||||
serialization->deserializeTextJSON(column_, buf, format_settings);
|
||||
return true;
|
||||
};
|
||||
|
||||
if (format_settings.json.empty_as_default)
|
||||
return JSONUtils::deserializeEmpyStringAsDefaultOrNested<bool, false>(column, in, deserialize);
|
||||
else
|
||||
return deserialize(column, in);
|
||||
}
|
||||
catch (Exception & e)
|
||||
{
|
||||
@ -920,6 +930,78 @@ namespace JSONUtils
|
||||
}
|
||||
}
|
||||
|
||||
template <typename ReturnType, bool default_column_return_value>
|
||||
ReturnType deserializeEmpyStringAsDefaultOrNested(IColumn & column, ReadBuffer & istr, const NestedDeserialize<ReturnType> & deserialize_nested)
|
||||
{
|
||||
static constexpr auto throw_exception = std::is_same_v<ReturnType, void>;
|
||||
|
||||
static constexpr auto EMPTY_STRING = "\"\"";
|
||||
static constexpr auto EMPTY_STRING_LENGTH = std::string_view(EMPTY_STRING).length();
|
||||
|
||||
if (istr.eof() || *istr.position() != EMPTY_STRING[0])
|
||||
return deserialize_nested(column, istr);
|
||||
|
||||
auto do_deserialize = [](IColumn & column_, ReadBuffer & buf, auto && check_for_empty_string, auto && deserialize) -> ReturnType
|
||||
{
|
||||
if (check_for_empty_string(buf))
|
||||
{
|
||||
column_.insertDefault();
|
||||
return ReturnType(default_column_return_value);
|
||||
}
|
||||
return deserialize(column_, buf);
|
||||
};
|
||||
|
||||
if (istr.available() >= EMPTY_STRING_LENGTH)
|
||||
{
|
||||
/// We have enough data in buffer to check if we have an empty string.
|
||||
auto check_for_empty_string = [](ReadBuffer & buf) -> bool
|
||||
{
|
||||
auto * pos = buf.position();
|
||||
if (checkString(EMPTY_STRING, buf))
|
||||
return true;
|
||||
buf.position() = pos;
|
||||
return false;
|
||||
};
|
||||
|
||||
return do_deserialize(column, istr, check_for_empty_string, deserialize_nested);
|
||||
}
|
||||
|
||||
/// We don't have enough data in buffer to check if we have an empty string.
|
||||
/// Use PeekableReadBuffer to make a checkpoint before checking for an
|
||||
/// empty string and rollback if check was failed.
|
||||
|
||||
auto check_for_empty_string = [](ReadBuffer & buf) -> bool
|
||||
{
|
||||
auto & peekable_buf = assert_cast<PeekableReadBuffer &>(buf);
|
||||
peekable_buf.setCheckpoint();
|
||||
SCOPE_EXIT(peekable_buf.dropCheckpoint());
|
||||
if (checkString(EMPTY_STRING, peekable_buf))
|
||||
return true;
|
||||
peekable_buf.rollbackToCheckpoint();
|
||||
return false;
|
||||
};
|
||||
|
||||
auto deserialize_nested_with_check = [&deserialize_nested](IColumn & column_, ReadBuffer & buf) -> ReturnType
|
||||
{
|
||||
auto & peekable_buf = assert_cast<PeekableReadBuffer &>(buf);
|
||||
if constexpr (throw_exception)
|
||||
deserialize_nested(column_, peekable_buf);
|
||||
else if (!deserialize_nested(column_, peekable_buf))
|
||||
return ReturnType(false);
|
||||
|
||||
if (unlikely(peekable_buf.hasUnreadData()))
|
||||
throw Exception(ErrorCodes::LOGICAL_ERROR, "Incorrect state while parsing JSON: PeekableReadBuffer has unread data in own memory: {}", String(peekable_buf.position(), peekable_buf.available()));
|
||||
|
||||
return ReturnType(true);
|
||||
};
|
||||
|
||||
PeekableReadBuffer peekable_buf(istr, true);
|
||||
return do_deserialize(column, peekable_buf, check_for_empty_string, deserialize_nested_with_check);
|
||||
}
|
||||
|
||||
template void deserializeEmpyStringAsDefaultOrNested<void, true>(IColumn & column, ReadBuffer & istr, const NestedDeserialize<void> & deserialize_nested);
|
||||
template bool deserializeEmpyStringAsDefaultOrNested<bool, true>(IColumn & column, ReadBuffer & istr, const NestedDeserialize<bool> & deserialize_nested);
|
||||
template bool deserializeEmpyStringAsDefaultOrNested<bool, false>(IColumn & column, ReadBuffer & istr, const NestedDeserialize<bool> & deserialize_nested);
|
||||
}
|
||||
|
||||
}
|
||||
|
@ -8,6 +8,7 @@
|
||||
#include <IO/Progress.h>
|
||||
#include <Core/NamesAndTypes.h>
|
||||
#include <Common/Stopwatch.h>
|
||||
#include <functional>
|
||||
#include <utility>
|
||||
|
||||
namespace DB
|
||||
@ -146,6 +147,16 @@ namespace JSONUtils
|
||||
|
||||
bool skipUntilFieldInObject(ReadBuffer & in, const String & desired_field_name, const FormatSettings::JSON & settings);
|
||||
void skipTheRestOfObject(ReadBuffer & in, const FormatSettings::JSON & settings);
|
||||
|
||||
template <typename ReturnType>
|
||||
using NestedDeserialize = std::function<ReturnType(IColumn &, ReadBuffer &)>;
|
||||
|
||||
template <typename ReturnType, bool default_column_return_value = true>
|
||||
ReturnType deserializeEmpyStringAsDefaultOrNested(IColumn & column, ReadBuffer & istr, const NestedDeserialize<ReturnType> & deserialize_nested);
|
||||
|
||||
extern template void deserializeEmpyStringAsDefaultOrNested<void, true>(IColumn & column, ReadBuffer & istr, const NestedDeserialize<void> & deserialize_nested);
|
||||
extern template bool deserializeEmpyStringAsDefaultOrNested<bool, true>(IColumn & column, ReadBuffer & istr, const NestedDeserialize<bool> & deserialize_nested);
|
||||
extern template bool deserializeEmpyStringAsDefaultOrNested<bool, false>(IColumn & column, ReadBuffer & istr, const NestedDeserialize<bool> & deserialize_nested);
|
||||
}
|
||||
|
||||
}
|
||||
|
@ -50,13 +50,6 @@ private:
|
||||
return executeNonconstant(input);
|
||||
}
|
||||
|
||||
[[maybe_unused]] String toString() const
|
||||
{
|
||||
WriteBufferFromOwnString buf;
|
||||
buf << "format:" << format << ", rows:" << rows << ", is_literal:" << is_literal << ", input:" << input.dumpStructure() << "\n";
|
||||
return buf.str();
|
||||
}
|
||||
|
||||
private:
|
||||
ColumnWithTypeAndName executeLiteral(std::string_view literal) const
|
||||
{
|
||||
@ -231,9 +224,7 @@ public:
|
||||
const auto & instruction = instructions[i];
|
||||
try
|
||||
{
|
||||
// std::cout << "instruction[" << i << "]:" << instructions[i].toString() << std::endl;
|
||||
concat_args[i] = instruction.execute();
|
||||
// std::cout << "concat_args[" << i << "]:" << concat_args[i].dumpStructure() << std::endl;
|
||||
}
|
||||
catch (const fmt::v9::format_error & e)
|
||||
{
|
||||
@ -358,7 +349,14 @@ private:
|
||||
|
||||
REGISTER_FUNCTION(Printf)
|
||||
{
|
||||
factory.registerFunction<FunctionPrintf>();
|
||||
factory.registerFunction<FunctionPrintf>(
|
||||
FunctionDocumentation{.description=R"(
|
||||
The `printf` function formats the given string with the values (strings, integers, floating-points etc.) listed in the arguments, similar to printf function in C++.
|
||||
The format string can contain format specifiers starting with `%` character.
|
||||
Anything not contained in `%` and the following format specifier is considered literal text and copied verbatim into the output.
|
||||
Literal `%` character can be escaped by `%%`.)", .examples{{"sum", "select printf('%%%s %s %d', 'Hello', 'World', 2024);", "%Hello World 2024"}}, .categories{"String"}
|
||||
});
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
@ -85,7 +85,9 @@ ConcurrentHashJoin::ConcurrentHashJoin(
|
||||
CurrentMetrics::ConcurrentHashJoinPoolThreads,
|
||||
CurrentMetrics::ConcurrentHashJoinPoolThreadsActive,
|
||||
CurrentMetrics::ConcurrentHashJoinPoolThreadsScheduled,
|
||||
slots))
|
||||
/*max_threads_*/ slots,
|
||||
/*max_free_threads_*/ 0,
|
||||
/*queue_size_*/ slots))
|
||||
, stats_collecting_params(stats_collecting_params_)
|
||||
{
|
||||
hash_joins.resize(slots);
|
||||
|
@ -10,6 +10,7 @@
|
||||
#include <Common/SensitiveDataMasker.h>
|
||||
#include <Common/Macros.h>
|
||||
#include <Common/EventNotifier.h>
|
||||
#include <Common/getNumberOfPhysicalCPUCores.h>
|
||||
#include <Common/Stopwatch.h>
|
||||
#include <Common/formatReadable.h>
|
||||
#include <Common/Throttler.h>
|
||||
@ -121,7 +122,6 @@
|
||||
#include <Interpreters/InterpreterSelectWithUnionQuery.h>
|
||||
#include <base/defines.h>
|
||||
|
||||
|
||||
namespace fs = std::filesystem;
|
||||
|
||||
namespace ProfileEvents
|
||||
@ -164,6 +164,9 @@ namespace CurrentMetrics
|
||||
extern const Metric TablesLoaderForegroundThreadsActive;
|
||||
extern const Metric TablesLoaderForegroundThreadsScheduled;
|
||||
extern const Metric IOWriterThreadsScheduled;
|
||||
extern const Metric BuildVectorSimilarityIndexThreads;
|
||||
extern const Metric BuildVectorSimilarityIndexThreadsActive;
|
||||
extern const Metric BuildVectorSimilarityIndexThreadsScheduled;
|
||||
extern const Metric AttachedTable;
|
||||
extern const Metric AttachedView;
|
||||
extern const Metric AttachedDictionary;
|
||||
@ -297,6 +300,8 @@ struct ContextSharedPart : boost::noncopyable
|
||||
mutable std::unique_ptr<ThreadPool> load_marks_threadpool; /// Threadpool for loading marks cache.
|
||||
mutable OnceFlag prefetch_threadpool_initialized;
|
||||
mutable std::unique_ptr<ThreadPool> prefetch_threadpool; /// Threadpool for loading marks cache.
|
||||
mutable OnceFlag build_vector_similarity_index_threadpool_initialized;
|
||||
mutable std::unique_ptr<ThreadPool> build_vector_similarity_index_threadpool; /// Threadpool for vector-similarity index creation.
|
||||
mutable UncompressedCachePtr index_uncompressed_cache TSA_GUARDED_BY(mutex); /// The cache of decompressed blocks for MergeTree indices.
|
||||
mutable QueryCachePtr query_cache TSA_GUARDED_BY(mutex); /// Cache of query results.
|
||||
mutable MarkCachePtr index_mark_cache TSA_GUARDED_BY(mutex); /// Cache of marks in compressed files of MergeTree indices.
|
||||
@ -3297,6 +3302,21 @@ size_t Context::getPrefetchThreadpoolSize() const
|
||||
return config.getUInt(".prefetch_threadpool_pool_size", 100);
|
||||
}
|
||||
|
||||
ThreadPool & Context::getBuildVectorSimilarityIndexThreadPool() const
|
||||
{
|
||||
callOnce(shared->build_vector_similarity_index_threadpool_initialized, [&] {
|
||||
size_t pool_size = shared->server_settings.max_build_vector_similarity_index_thread_pool_size > 0
|
||||
? shared->server_settings.max_build_vector_similarity_index_thread_pool_size
|
||||
: getNumberOfPhysicalCPUCores();
|
||||
shared->build_vector_similarity_index_threadpool = std::make_unique<ThreadPool>(
|
||||
CurrentMetrics::BuildVectorSimilarityIndexThreads,
|
||||
CurrentMetrics::BuildVectorSimilarityIndexThreadsActive,
|
||||
CurrentMetrics::BuildVectorSimilarityIndexThreadsScheduled,
|
||||
pool_size);
|
||||
});
|
||||
return *shared->build_vector_similarity_index_threadpool;
|
||||
}
|
||||
|
||||
BackgroundSchedulePool & Context::getBufferFlushSchedulePool() const
|
||||
{
|
||||
callOnce(shared->buffer_flush_schedule_pool_initialized, [&] {
|
||||
|
@ -1097,6 +1097,8 @@ public:
|
||||
/// and make a prefetch by putting a read task to threadpoolReader.
|
||||
size_t getPrefetchThreadpoolSize() const;
|
||||
|
||||
ThreadPool & getBuildVectorSimilarityIndexThreadPool() const;
|
||||
|
||||
/// Settings for MergeTree background tasks stored in config.xml
|
||||
BackgroundTaskSchedulingSettings getBackgroundProcessingTaskSchedulingSettings() const;
|
||||
BackgroundTaskSchedulingSettings getBackgroundMoveTaskSchedulingSettings() const;
|
||||
|
@ -1236,6 +1236,7 @@ IBlocksStreamPtr HashJoin::getNonJoinedBlocks(const Block & left_sample_block,
|
||||
|
||||
void HashJoin::reuseJoinedData(const HashJoin & join)
|
||||
{
|
||||
have_compressed = join.have_compressed;
|
||||
data = join.data;
|
||||
from_storage_join = true;
|
||||
|
||||
|
@ -968,6 +968,11 @@ void InterpreterCreateQuery::validateMaterializedViewColumnsAndEngine(const ASTC
|
||||
if (database && database->getEngineName() != "Atomic")
|
||||
throw Exception(ErrorCodes::INCORRECT_QUERY,
|
||||
"Refreshable materialized views (except with APPEND) only support Atomic database engine, but database {} has engine {}", create.getDatabase(), database->getEngineName());
|
||||
|
||||
std::string message;
|
||||
if (!supportsAtomicRename(&message))
|
||||
throw Exception(ErrorCodes::NOT_IMPLEMENTED,
|
||||
"Can't create refreshable materialized view because exchanging files is not supported by the OS ({})", message);
|
||||
}
|
||||
|
||||
Block input_block;
|
||||
|
@ -99,6 +99,7 @@ namespace DB
|
||||
namespace ErrorCodes
|
||||
{
|
||||
extern const int QUERY_CACHE_USED_WITH_NONDETERMINISTIC_FUNCTIONS;
|
||||
extern const int QUERY_CACHE_USED_WITH_NON_THROW_OVERFLOW_MODE;
|
||||
extern const int QUERY_CACHE_USED_WITH_SYSTEM_TABLE;
|
||||
extern const int INTO_OUTFILE_NOT_ALLOWED;
|
||||
extern const int INVALID_TRANSACTION;
|
||||
@ -1118,22 +1119,24 @@ static std::tuple<ASTPtr, BlockIO> executeQueryImpl(
|
||||
&& settings.use_query_cache
|
||||
&& !internal
|
||||
&& client_info.query_kind == ClientInfo::QueryKind::INITIAL_QUERY
|
||||
/// Bug 67476: Avoid that the query cache stores truncated results if the query ran with a non-THROW overflow mode and hit a limit.
|
||||
/// This is more workaround than a fix ... unfortunately it is hard to detect from the perspective of the query cache that the
|
||||
/// query result is truncated.
|
||||
&& (settings.read_overflow_mode == OverflowMode::THROW
|
||||
&& settings.read_overflow_mode_leaf == OverflowMode::THROW
|
||||
&& settings.group_by_overflow_mode == OverflowMode::THROW
|
||||
&& settings.sort_overflow_mode == OverflowMode::THROW
|
||||
&& settings.result_overflow_mode == OverflowMode::THROW
|
||||
&& settings.timeout_overflow_mode == OverflowMode::THROW
|
||||
&& settings.set_overflow_mode == OverflowMode::THROW
|
||||
&& settings.join_overflow_mode == OverflowMode::THROW
|
||||
&& settings.transfer_overflow_mode == OverflowMode::THROW
|
||||
&& settings.distinct_overflow_mode == OverflowMode::THROW)
|
||||
&& (ast->as<ASTSelectQuery>() || ast->as<ASTSelectWithUnionQuery>());
|
||||
QueryCache::Usage query_cache_usage = QueryCache::Usage::None;
|
||||
|
||||
/// Bug 67476: If the query runs with a non-THROW overflow mode and hits a limit, the query cache will store a truncated result (if
|
||||
/// enabled). This is incorrect. Unfortunately it is hard to detect from the perspective of the query cache that the query result
|
||||
/// is truncated. Therefore throw an exception, to notify the user to disable either the query cache or use another overflow mode.
|
||||
if (settings.use_query_cache && (settings.read_overflow_mode != OverflowMode::THROW
|
||||
|| settings.read_overflow_mode_leaf != OverflowMode::THROW
|
||||
|| settings.group_by_overflow_mode != OverflowMode::THROW
|
||||
|| settings.sort_overflow_mode != OverflowMode::THROW
|
||||
|| settings.result_overflow_mode != OverflowMode::THROW
|
||||
|| settings.timeout_overflow_mode != OverflowMode::THROW
|
||||
|| settings.set_overflow_mode != OverflowMode::THROW
|
||||
|| settings.join_overflow_mode != OverflowMode::THROW
|
||||
|| settings.transfer_overflow_mode != OverflowMode::THROW
|
||||
|| settings.distinct_overflow_mode != OverflowMode::THROW))
|
||||
throw Exception(ErrorCodes::QUERY_CACHE_USED_WITH_NON_THROW_OVERFLOW_MODE, "use_query_cache and overflow_mode != 'throw' cannot be used together");
|
||||
|
||||
/// If the query runs with "use_query_cache = 1", we first probe if the query cache already contains the query result (if yes:
|
||||
/// return result from cache). If doesn't, we execute the query normally and write the result into the query cache. Both steps use a
|
||||
/// hash of the AST, the current database and the settings as cache key. Unfortunately, the settings are in some places internally
|
||||
|
@ -74,7 +74,8 @@ private:
|
||||
findMySQLFunctionSecretArguments();
|
||||
}
|
||||
else if ((function.name == "s3") || (function.name == "cosn") || (function.name == "oss") ||
|
||||
(function.name == "deltaLake") || (function.name == "hudi") || (function.name == "iceberg"))
|
||||
(function.name == "deltaLake") || (function.name == "hudi") || (function.name == "iceberg") ||
|
||||
(function.name == "gcs"))
|
||||
{
|
||||
/// s3('url', 'aws_access_key_id', 'aws_secret_access_key', ...)
|
||||
findS3FunctionSecretArguments(/* is_cluster_function= */ false);
|
||||
|
@ -1,6 +1,6 @@
|
||||
#include <Parsers/ParserShowColumnsQuery.h>
|
||||
|
||||
#include <Parsers/ASTIdentifier_fwd.h>
|
||||
#include <Parsers/ASTIdentifier.h>
|
||||
#include <Parsers/ASTLiteral.h>
|
||||
#include <Parsers/ASTShowColumnsQuery.h>
|
||||
#include <Parsers/CommonParsers.h>
|
||||
@ -18,7 +18,6 @@ bool ParserShowColumnsQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
|
||||
ASTPtr from1;
|
||||
ASTPtr from2;
|
||||
|
||||
String from1_str;
|
||||
String from2_str;
|
||||
|
||||
auto query = std::make_shared<ASTShowColumnsQuery>();
|
||||
@ -43,25 +42,18 @@ bool ParserShowColumnsQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
|
||||
else
|
||||
return false;
|
||||
|
||||
tryGetIdentifierNameInto(from1, from1_str);
|
||||
|
||||
bool abbreviated_form = from1_str.contains("."); // FROM database.table
|
||||
if (abbreviated_form)
|
||||
{
|
||||
std::vector<String> split;
|
||||
boost::split(split, from1_str, boost::is_any_of("."));
|
||||
query->database = split[0];
|
||||
query->table = split[1];
|
||||
}
|
||||
const auto * table_id = from1->as<ASTIdentifier>();
|
||||
if (!table_id)
|
||||
return false;
|
||||
query->table = table_id->shortName();
|
||||
if (table_id->compound())
|
||||
query->database = table_id->name_parts[0];
|
||||
else
|
||||
{
|
||||
if (ParserKeyword(Keyword::FROM).ignore(pos, expected) || ParserKeyword(Keyword::IN).ignore(pos, expected))
|
||||
if (!ParserIdentifier().parse(pos, from2, expected))
|
||||
return false;
|
||||
|
||||
tryGetIdentifierNameInto(from2, from2_str);
|
||||
|
||||
query->table = from1_str;
|
||||
query->database = from2_str;
|
||||
}
|
||||
|
||||
|
@ -1,6 +1,6 @@
|
||||
#include <Parsers/ParserShowIndexesQuery.h>
|
||||
|
||||
#include <Parsers/ASTIdentifier_fwd.h>
|
||||
#include <Parsers/ASTIdentifier.h>
|
||||
#include <Parsers/ASTLiteral.h>
|
||||
#include <Parsers/ASTShowIndexesQuery.h>
|
||||
#include <Parsers/CommonParsers.h>
|
||||
@ -17,7 +17,6 @@ bool ParserShowIndexesQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
|
||||
ASTPtr from1;
|
||||
ASTPtr from2;
|
||||
|
||||
String from1_str;
|
||||
String from2_str;
|
||||
|
||||
auto query = std::make_shared<ASTShowIndexesQuery>();
|
||||
@ -39,25 +38,18 @@ bool ParserShowIndexesQuery::parseImpl(Pos & pos, ASTPtr & node, Expected & expe
|
||||
else
|
||||
return false;
|
||||
|
||||
tryGetIdentifierNameInto(from1, from1_str);
|
||||
|
||||
bool abbreviated_form = from1_str.contains("."); // FROM database.table
|
||||
if (abbreviated_form)
|
||||
{
|
||||
std::vector<String> split;
|
||||
boost::split(split, from1_str, boost::is_any_of("."));
|
||||
query->database = split[0];
|
||||
query->table = split[1];
|
||||
}
|
||||
const auto * table_id = from1->as<ASTIdentifier>();
|
||||
if (!table_id)
|
||||
return false;
|
||||
query->table = table_id->shortName();
|
||||
if (table_id->compound())
|
||||
query->database = table_id->name_parts[0];
|
||||
else
|
||||
{
|
||||
if (ParserKeyword(Keyword::FROM).ignore(pos, expected) || ParserKeyword(Keyword::IN).ignore(pos, expected))
|
||||
if (!ParserIdentifier().parse(pos, from2, expected))
|
||||
return false;
|
||||
|
||||
tryGetIdentifierNameInto(from2, from2_str);
|
||||
|
||||
query->table = from1_str;
|
||||
query->database = from2_str;
|
||||
}
|
||||
|
||||
|
@ -2009,33 +2009,6 @@ void ReadFromMergeTree::initializePipeline(QueryPipelineBuilder & pipeline, cons
|
||||
{
|
||||
auto result = getAnalysisResult();
|
||||
|
||||
if (is_parallel_reading_from_replicas && context->canUseParallelReplicasOnInitiator()
|
||||
&& context->getSettingsRef().parallel_replicas_local_plan)
|
||||
{
|
||||
CoordinationMode mode = CoordinationMode::Default;
|
||||
switch (result.read_type)
|
||||
{
|
||||
case ReadFromMergeTree::ReadType::Default:
|
||||
mode = CoordinationMode::Default;
|
||||
break;
|
||||
case ReadFromMergeTree::ReadType::InOrder:
|
||||
mode = CoordinationMode::WithOrder;
|
||||
break;
|
||||
case ReadFromMergeTree::ReadType::InReverseOrder:
|
||||
mode = CoordinationMode::ReverseOrder;
|
||||
break;
|
||||
case ReadFromMergeTree::ReadType::ParallelReplicas:
|
||||
throw Exception(ErrorCodes::LOGICAL_ERROR, "Read type can't be ParallelReplicas on initiator");
|
||||
}
|
||||
|
||||
chassert(number_of_current_replica.has_value());
|
||||
chassert(all_ranges_callback.has_value());
|
||||
|
||||
/// initialize working set from local replica
|
||||
all_ranges_callback.value()(
|
||||
InitialAllRangesAnnouncement(mode, result.parts_with_ranges.getDescriptions(), number_of_current_replica.value()));
|
||||
}
|
||||
|
||||
if (enable_remove_parts_from_snapshot_optimization)
|
||||
{
|
||||
/// Do not keep data parts in snapshot.
|
||||
|
@ -5,9 +5,11 @@
|
||||
#include <Columns/ColumnArray.h>
|
||||
#include <Common/BitHelpers.h>
|
||||
#include <Common/formatReadable.h>
|
||||
#include <Common/getNumberOfPhysicalCPUCores.h>
|
||||
#include <Common/logger_useful.h>
|
||||
#include <Common/typeid_cast.h>
|
||||
#include <Core/Field.h>
|
||||
#include <Core/ServerSettings.h>
|
||||
#include <DataTypes/DataTypeArray.h>
|
||||
#include <IO/ReadHelpers.h>
|
||||
#include <IO/WriteHelpers.h>
|
||||
@ -29,7 +31,6 @@ namespace DB
|
||||
|
||||
namespace ErrorCodes
|
||||
{
|
||||
extern const int CANNOT_ALLOCATE_MEMORY;
|
||||
extern const int FORMAT_VERSION_TOO_OLD;
|
||||
extern const int ILLEGAL_COLUMN;
|
||||
extern const int INCORRECT_DATA;
|
||||
@ -131,8 +132,7 @@ void USearchIndexWithSerialization::deserialize(ReadBuffer & istr)
|
||||
/// See the comment in MergeTreeIndexGranuleVectorSimilarity::deserializeBinary why we throw here
|
||||
throw Exception(ErrorCodes::INCORRECT_DATA, "Could not load vector similarity index. Please drop the index and create it again. Error: {}", String(result.error.release()));
|
||||
|
||||
if (!try_reserve(limits()))
|
||||
throw Exception(ErrorCodes::CANNOT_ALLOCATE_MEMORY, "Could not reserve memory for usearch index");
|
||||
try_reserve(limits());
|
||||
}
|
||||
|
||||
USearchIndexWithSerialization::Statistics USearchIndexWithSerialization::getStatistics() const
|
||||
@ -270,20 +270,49 @@ void updateImpl(const ColumnArray * column_array, const ColumnArray::Offsets & c
|
||||
throw Exception(ErrorCodes::INCORRECT_DATA, "All arrays in column with vector similarity index must have equal length");
|
||||
|
||||
/// Reserving space is mandatory
|
||||
if (!index->try_reserve(roundUpToPowerOfTwoOrZero(index->size() + rows)))
|
||||
throw Exception(ErrorCodes::CANNOT_ALLOCATE_MEMORY, "Could not reserve memory for vector similarity index");
|
||||
size_t max_thread_pool_size = Context::getGlobalContextInstance()->getServerSettings().max_build_vector_similarity_index_thread_pool_size;
|
||||
if (max_thread_pool_size == 0)
|
||||
max_thread_pool_size = getNumberOfPhysicalCPUCores();
|
||||
unum::usearch::index_limits_t limits(roundUpToPowerOfTwoOrZero(index->size() + rows), max_thread_pool_size);
|
||||
index->reserve(limits);
|
||||
|
||||
for (size_t row = 0; row < rows; ++row)
|
||||
/// Vector index creation is slooooow. Add the new rows in parallel. The threadpool is global to avoid oversubscription when multiple
|
||||
/// indexes are build simultaneously (e.g. multiple merges run at the same time).
|
||||
auto & thread_pool = Context::getGlobalContextInstance()->getBuildVectorSimilarityIndexThreadPool();
|
||||
|
||||
auto add_vector_to_index = [&](USearchIndex::vector_key_t key, size_t row, ThreadGroupPtr thread_group)
|
||||
{
|
||||
if (auto result = index->add(static_cast<USearchIndex::vector_key_t>(index->size()), &column_array_data_float_data[column_array_offsets[row - 1]]); !result)
|
||||
SCOPE_EXIT_SAFE(
|
||||
if (thread_group)
|
||||
CurrentThread::detachFromGroupIfNotDetached();
|
||||
);
|
||||
|
||||
if (thread_group)
|
||||
CurrentThread::attachToGroupIfDetached(thread_group);
|
||||
|
||||
/// add is thread-safe
|
||||
if (auto result = index->add(key, &column_array_data_float_data[column_array_offsets[row - 1]]); !result)
|
||||
{
|
||||
throw Exception(ErrorCodes::INCORRECT_DATA, "Could not add data to vector similarity index. Error: {}", String(result.error.release()));
|
||||
}
|
||||
else
|
||||
{
|
||||
ProfileEvents::increment(ProfileEvents::USearchAddCount);
|
||||
ProfileEvents::increment(ProfileEvents::USearchAddVisitedMembers, result.visited_members);
|
||||
ProfileEvents::increment(ProfileEvents::USearchAddComputedDistances, result.computed_distances);
|
||||
}
|
||||
};
|
||||
|
||||
size_t index_size = index->size();
|
||||
|
||||
for (size_t row = 0; row < rows; ++row)
|
||||
{
|
||||
auto key = static_cast<USearchIndex::vector_key_t>(index_size + row);
|
||||
auto task = [group = CurrentThread::getGroup(), &add_vector_to_index, key, row] { add_vector_to_index(key, row, group); };
|
||||
thread_pool.scheduleOrThrowOnError(task);
|
||||
}
|
||||
|
||||
thread_pool.wait();
|
||||
}
|
||||
|
||||
}
|
||||
|
@ -1,6 +1,7 @@
|
||||
#include <Storages/MergeTree/MergeTreeData.h>
|
||||
#include <Storages/MergeTree/MergeTreePartsMover.h>
|
||||
#include <Storages/MergeTree/MergeTreeSettings.h>
|
||||
#include <Common/FailPoint.h>
|
||||
#include <Common/logger_useful.h>
|
||||
|
||||
#include <set>
|
||||
@ -15,6 +16,11 @@ namespace ErrorCodes
|
||||
extern const int DIRECTORY_ALREADY_EXISTS;
|
||||
}
|
||||
|
||||
namespace FailPoints
|
||||
{
|
||||
extern const char stop_moving_part_before_swap_with_active[];
|
||||
}
|
||||
|
||||
namespace
|
||||
{
|
||||
|
||||
@ -226,6 +232,7 @@ MergeTreePartsMover::TemporaryClonedPart MergeTreePartsMover::clonePart(const Me
|
||||
cloned_part.temporary_directory_lock = data->getTemporaryPartDirectoryHolder(part->name);
|
||||
|
||||
MutableDataPartStoragePtr cloned_part_storage;
|
||||
bool preserve_blobs = false;
|
||||
if (disk->supportZeroCopyReplication() && settings->allow_remote_fs_zero_copy_replication)
|
||||
{
|
||||
/// Try zero-copy replication and fallback to default copy if it's not possible
|
||||
@ -253,6 +260,7 @@ MergeTreePartsMover::TemporaryClonedPart MergeTreePartsMover::clonePart(const Me
|
||||
if (zero_copy_part)
|
||||
{
|
||||
/// FIXME for some reason we cannot just use this part, we have to re-create it through MergeTreeDataPartBuilder
|
||||
preserve_blobs = true;
|
||||
zero_copy_part->is_temp = false; /// Do not remove it in dtor
|
||||
cloned_part_storage = zero_copy_part->getDataPartStoragePtr();
|
||||
}
|
||||
@ -272,7 +280,17 @@ MergeTreePartsMover::TemporaryClonedPart MergeTreePartsMover::clonePart(const Me
|
||||
cloned_part.part = std::move(builder).withPartFormatFromDisk().build();
|
||||
LOG_TRACE(log, "Part {} was cloned to {}", part->name, cloned_part.part->getDataPartStorage().getFullPath());
|
||||
|
||||
cloned_part.part->is_temp = data->allowRemoveStaleMovingParts();
|
||||
cloned_part.part->is_temp = false;
|
||||
if (data->allowRemoveStaleMovingParts())
|
||||
{
|
||||
cloned_part.part->is_temp = true;
|
||||
/// Setting it in case connection to zookeeper is lost while moving
|
||||
/// Otherwise part might be stuck in the moving directory due to the KEEPER_EXCEPTION in part's destructor
|
||||
if (preserve_blobs)
|
||||
cloned_part.part->remove_tmp_policy = IMergeTreeDataPart::BlobsRemovalPolicyForTemporaryParts::PRESERVE_BLOBS;
|
||||
else
|
||||
cloned_part.part->remove_tmp_policy = IMergeTreeDataPart::BlobsRemovalPolicyForTemporaryParts::REMOVE_BLOBS;
|
||||
}
|
||||
cloned_part.part->loadColumnsChecksumsIndexes(true, true);
|
||||
cloned_part.part->loadVersionMetadata();
|
||||
cloned_part.part->modification_time = cloned_part.part->getDataPartStorage().getLastModified().epochTime();
|
||||
@ -282,6 +300,8 @@ MergeTreePartsMover::TemporaryClonedPart MergeTreePartsMover::clonePart(const Me
|
||||
|
||||
void MergeTreePartsMover::swapClonedPart(TemporaryClonedPart & cloned_part) const
|
||||
{
|
||||
/// Used to get some stuck parts in the moving directory by stopping moves while pause is active
|
||||
FailPointInjection::pauseFailPoint(FailPoints::stop_moving_part_before_swap_with_active);
|
||||
if (moves_blocker.isCancelled())
|
||||
throw Exception(ErrorCodes::ABORTED, "Cancelled moving parts.");
|
||||
|
||||
|
@ -15,6 +15,7 @@ namespace ErrorCodes
|
||||
{
|
||||
extern const int SUPPORT_IS_DISABLED;
|
||||
extern const int REPLICA_STATUS_CHANGED;
|
||||
extern const int LOGICAL_ERROR;
|
||||
}
|
||||
|
||||
ReplicatedMergeTreeAttachThread::ReplicatedMergeTreeAttachThread(StorageReplicatedMergeTree & storage_)
|
||||
@ -117,6 +118,67 @@ void ReplicatedMergeTreeAttachThread::checkHasReplicaMetadataInZooKeeper(const z
|
||||
}
|
||||
}
|
||||
|
||||
Int32 ReplicatedMergeTreeAttachThread::fixReplicaMetadataVersionIfNeeded(zkutil::ZooKeeperPtr zookeeper)
|
||||
{
|
||||
const String & zookeeper_path = storage.zookeeper_path;
|
||||
const String & replica_path = storage.replica_path;
|
||||
const bool replica_readonly = storage.is_readonly;
|
||||
|
||||
for (size_t i = 0; i != 2; ++i)
|
||||
{
|
||||
String replica_metadata_version_str;
|
||||
const bool replica_metadata_version_exists = zookeeper->tryGet(replica_path + "/metadata_version", replica_metadata_version_str);
|
||||
if (!replica_metadata_version_exists)
|
||||
return -1;
|
||||
|
||||
const Int32 metadata_version = parse<Int32>(replica_metadata_version_str);
|
||||
|
||||
if (metadata_version != 0 || replica_readonly)
|
||||
{
|
||||
/// No need to fix anything
|
||||
return metadata_version;
|
||||
}
|
||||
|
||||
Coordination::Stat stat;
|
||||
zookeeper->get(fs::path(zookeeper_path) / "metadata", &stat);
|
||||
if (stat.version == 0)
|
||||
{
|
||||
/// No need to fix anything
|
||||
return metadata_version;
|
||||
}
|
||||
|
||||
ReplicatedMergeTreeQueue & queue = storage.queue;
|
||||
queue.pullLogsToQueue(zookeeper);
|
||||
if (queue.getStatus().metadata_alters_in_queue != 0)
|
||||
{
|
||||
LOG_DEBUG(log, "No need to update metadata_version as there are ALTER_METADATA entries in the queue");
|
||||
return metadata_version;
|
||||
}
|
||||
|
||||
const Coordination::Requests ops = {
|
||||
zkutil::makeSetRequest(fs::path(replica_path) / "metadata_version", std::to_string(stat.version), 0),
|
||||
zkutil::makeCheckRequest(fs::path(zookeeper_path) / "metadata", stat.version),
|
||||
};
|
||||
Coordination::Responses ops_responses;
|
||||
const auto code = zookeeper->tryMulti(ops, ops_responses);
|
||||
if (code == Coordination::Error::ZOK)
|
||||
{
|
||||
LOG_DEBUG(log, "Successfully set metadata_version to {}", stat.version);
|
||||
return stat.version;
|
||||
}
|
||||
if (code != Coordination::Error::ZBADVERSION)
|
||||
{
|
||||
throw zkutil::KeeperException(code);
|
||||
}
|
||||
}
|
||||
|
||||
/// Second attempt is only possible if metadata_version != 0 or metadata.version changed during the first attempt.
|
||||
/// If metadata_version != 0, on second attempt we will return the new metadata_version.
|
||||
/// If metadata.version changed, on second attempt we will either get metadata_version != 0 and return the new metadata_version or we will get metadata_alters_in_queue != 0 and return 0.
|
||||
/// Either way, on second attempt this method should return.
|
||||
throw Exception(ErrorCodes::LOGICAL_ERROR, "Failed to fix replica metadata_version in ZooKeeper after two attempts");
|
||||
}
|
||||
|
||||
void ReplicatedMergeTreeAttachThread::runImpl()
|
||||
{
|
||||
storage.setZooKeeper();
|
||||
@ -160,11 +222,11 @@ void ReplicatedMergeTreeAttachThread::runImpl()
|
||||
/// Just in case it was not removed earlier due to connection loss
|
||||
zookeeper->tryRemove(replica_path + "/flags/force_restore_data");
|
||||
|
||||
String replica_metadata_version;
|
||||
const bool replica_metadata_version_exists = zookeeper->tryGet(replica_path + "/metadata_version", replica_metadata_version);
|
||||
const Int32 replica_metadata_version = fixReplicaMetadataVersionIfNeeded(zookeeper);
|
||||
const bool replica_metadata_version_exists = replica_metadata_version != -1;
|
||||
if (replica_metadata_version_exists)
|
||||
{
|
||||
storage.setInMemoryMetadata(metadata_snapshot->withMetadataVersion(parse<int>(replica_metadata_version)));
|
||||
storage.setInMemoryMetadata(metadata_snapshot->withMetadataVersion(replica_metadata_version));
|
||||
}
|
||||
else
|
||||
{
|
||||
|
@ -48,6 +48,8 @@ private:
|
||||
void runImpl();
|
||||
|
||||
void finalizeInitialization();
|
||||
|
||||
Int32 fixReplicaMetadataVersionIfNeeded(zkutil::ZooKeeperPtr zookeeper);
|
||||
};
|
||||
|
||||
}
|
||||
|
@ -2222,6 +2222,7 @@ ReplicatedMergeTreeQueue::Status ReplicatedMergeTreeQueue::getStatus() const
|
||||
res.inserts_in_queue = 0;
|
||||
res.merges_in_queue = 0;
|
||||
res.part_mutations_in_queue = 0;
|
||||
res.metadata_alters_in_queue = 0;
|
||||
res.queue_oldest_time = 0;
|
||||
res.inserts_oldest_time = 0;
|
||||
res.merges_oldest_time = 0;
|
||||
@ -2264,6 +2265,11 @@ ReplicatedMergeTreeQueue::Status ReplicatedMergeTreeQueue::getStatus() const
|
||||
res.oldest_part_to_mutate_to = entry->new_part_name;
|
||||
}
|
||||
}
|
||||
|
||||
if (entry->type == LogEntry::ALTER_METADATA)
|
||||
{
|
||||
++res.metadata_alters_in_queue;
|
||||
}
|
||||
}
|
||||
|
||||
return res;
|
||||
|
@ -473,6 +473,7 @@ public:
|
||||
UInt32 inserts_in_queue;
|
||||
UInt32 merges_in_queue;
|
||||
UInt32 part_mutations_in_queue;
|
||||
UInt32 metadata_alters_in_queue;
|
||||
UInt32 queue_oldest_time;
|
||||
UInt32 inserts_oldest_time;
|
||||
UInt32 merges_oldest_time;
|
||||
|
@ -277,7 +277,7 @@ static void extractZooKeeperPathAndReplicaNameFromEngineArgs(
|
||||
|
||||
if (has_valid_arguments)
|
||||
{
|
||||
if (is_replicated_database && local_context->getSettingsRef().database_replicated_allow_replicated_engine_arguments == 0)
|
||||
if (!query.attach && is_replicated_database && local_context->getSettingsRef().database_replicated_allow_replicated_engine_arguments == 0)
|
||||
{
|
||||
throw Exception(ErrorCodes::BAD_ARGUMENTS,
|
||||
"It's not allowed to specify explicit zookeeper_path and replica_name "
|
||||
@ -285,7 +285,7 @@ static void extractZooKeeperPathAndReplicaNameFromEngineArgs(
|
||||
"specify them explicitly, enable setting "
|
||||
"database_replicated_allow_replicated_engine_arguments.");
|
||||
}
|
||||
else if (is_replicated_database && local_context->getSettingsRef().database_replicated_allow_replicated_engine_arguments == 1)
|
||||
else if (!query.attach && is_replicated_database && local_context->getSettingsRef().database_replicated_allow_replicated_engine_arguments == 1)
|
||||
{
|
||||
LOG_WARNING(&Poco::Logger::get("registerStorageMergeTree"), "It's not recommended to explicitly specify "
|
||||
"zookeeper_path and replica_name in ReplicatedMergeTree arguments");
|
||||
@ -305,7 +305,7 @@ static void extractZooKeeperPathAndReplicaNameFromEngineArgs(
|
||||
throw Exception(ErrorCodes::BAD_ARGUMENTS, "Replica name must be a string literal{}", verbose_help_message);
|
||||
|
||||
|
||||
if (is_replicated_database && local_context->getSettingsRef().database_replicated_allow_replicated_engine_arguments == 2)
|
||||
if (!query.attach && is_replicated_database && local_context->getSettingsRef().database_replicated_allow_replicated_engine_arguments == 2)
|
||||
{
|
||||
LOG_WARNING(&Poco::Logger::get("registerStorageMergeTree"), "Replacing user-provided ZooKeeper path and replica name ({}, {}) "
|
||||
"with default arguments", zookeeper_path, replica_name);
|
||||
|
@ -42,6 +42,7 @@
|
||||
<multi_read>1</multi_read>
|
||||
<check_not_exists>1</check_not_exists>
|
||||
<create_if_not_exists>1</create_if_not_exists>
|
||||
<remove_recursive>1</remove_recursive>
|
||||
</feature_flags>
|
||||
</keeper_server>
|
||||
</clickhouse>
|
||||
|
@ -64,6 +64,7 @@ function configure()
|
||||
randomize_config_boolean_value multi_read keeper_port
|
||||
randomize_config_boolean_value check_not_exists keeper_port
|
||||
randomize_config_boolean_value create_if_not_exists keeper_port
|
||||
randomize_config_boolean_value remove_recursive keeper_port
|
||||
fi
|
||||
|
||||
sudo chown clickhouse /etc/clickhouse-server/config.d/keeper_port.xml
|
||||
|
@ -393,6 +393,7 @@ def test_table_functions():
|
||||
f"azureBlobStorageCluster('test_shard_localhost', named_collection_2, connection_string = '{azure_conn_string}', container = 'cont', blob_path = 'test_simple_16.csv', format = 'CSV')",
|
||||
f"azureBlobStorageCluster('test_shard_localhost', named_collection_2, storage_account_url = '{azure_storage_account_url}', container = 'cont', blob_path = 'test_simple_17.csv', account_name = '{azure_account_name}', account_key = '{azure_account_key}')",
|
||||
f"iceberg('http://minio1:9001/root/data/test11.csv.gz', 'minio', '{password}')",
|
||||
f"gcs('http://minio1:9001/root/data/test11.csv.gz', 'minio', '{password}')",
|
||||
]
|
||||
|
||||
def make_test_case(i):
|
||||
|
@ -0,0 +1,33 @@
|
||||
<clickhouse>
|
||||
<remote_servers>
|
||||
<parallel_replicas>
|
||||
<shard>
|
||||
<internal_replication>false</internal_replication>
|
||||
<replica>
|
||||
<host>node0</host>
|
||||
<port>9000</port>
|
||||
</replica>
|
||||
<replica>
|
||||
<host>node1</host>
|
||||
<port>9000</port>
|
||||
</replica>
|
||||
<replica>
|
||||
<host>node2</host>
|
||||
<port>9000</port>
|
||||
</replica>
|
||||
<replica>
|
||||
<host>node3</host>
|
||||
<port>9000</port>
|
||||
</replica>
|
||||
<replica>
|
||||
<host>node4</host>
|
||||
<port>9000</port>
|
||||
</replica>
|
||||
<replica>
|
||||
<host>node5</host>
|
||||
<port>9000</port>
|
||||
</replica>
|
||||
</shard>
|
||||
</parallel_replicas>
|
||||
</remote_servers>
|
||||
</clickhouse>
|
@ -0,0 +1,73 @@
|
||||
import pytest
|
||||
|
||||
from helpers.cluster import ClickHouseCluster
|
||||
|
||||
cluster = ClickHouseCluster(__file__)
|
||||
|
||||
nodes = [
|
||||
cluster.add_instance(
|
||||
f"node{num}", main_configs=["configs/remote_servers.xml"], with_zookeeper=True
|
||||
)
|
||||
for num in range(6)
|
||||
]
|
||||
|
||||
|
||||
@pytest.fixture(scope="module", autouse=True)
|
||||
def start_cluster():
|
||||
try:
|
||||
cluster.start()
|
||||
yield cluster
|
||||
finally:
|
||||
cluster.shutdown()
|
||||
|
||||
|
||||
def _create_tables(table_name):
|
||||
for idx, node in enumerate(nodes):
|
||||
node.query(
|
||||
f"DROP TABLE IF EXISTS {table_name}",
|
||||
settings={"database_atomic_wait_for_drop_and_detach_synchronously": True},
|
||||
)
|
||||
|
||||
node.query(
|
||||
f"""
|
||||
CREATE TABLE {table_name} (value Int64)
|
||||
Engine=ReplicatedMergeTree('/test_parallel_replicas/shard/{table_name}', '{idx}')
|
||||
ORDER BY ()
|
||||
"""
|
||||
)
|
||||
|
||||
nodes[0].query(
|
||||
f"INSERT INTO {table_name} SELECT * FROM numbers(1000)",
|
||||
settings={"insert_deduplicate": 0},
|
||||
)
|
||||
nodes[0].query(f"SYSTEM SYNC REPLICA ON CLUSTER 'parallel_replicas' {table_name}")
|
||||
|
||||
for idx, node in enumerate(nodes):
|
||||
node.query("SYSTEM STOP REPLICATED SENDS")
|
||||
# the same data on all nodes except for a single value
|
||||
node.query(
|
||||
f"INSERT INTO {table_name} VALUES ({idx})",
|
||||
settings={"insert_deduplicate": 0},
|
||||
)
|
||||
|
||||
|
||||
# check that we use the state of data parts from the initiator node (for some sort of determinism of what is been read).
|
||||
# currently it is implemented only when we build local plan for the initiator node (we aim to make this behavior default)
|
||||
def test_initiator_snapshot_is_used_for_reading(start_cluster):
|
||||
table_name = "t"
|
||||
_create_tables(table_name)
|
||||
|
||||
for idx, node in enumerate(nodes):
|
||||
expected = 499500 + idx # sum of all integers 0..999 + idx
|
||||
assert (
|
||||
node.query(
|
||||
f"SELECT sum(value) FROM {table_name}",
|
||||
settings={
|
||||
"allow_experimental_parallel_reading_from_replicas": 2,
|
||||
"max_parallel_replicas": 100,
|
||||
"cluster_for_parallel_replicas": "parallel_replicas",
|
||||
"parallel_replicas_local_plan": True,
|
||||
},
|
||||
)
|
||||
== f"{expected}\n"
|
||||
)
|
46
tests/integration/test_remove_stale_moving_parts/config.xml
Normal file
46
tests/integration/test_remove_stale_moving_parts/config.xml
Normal file
@ -0,0 +1,46 @@
|
||||
<clickhouse>
|
||||
<remote_servers>
|
||||
<cluster>
|
||||
<shard>
|
||||
<replica>
|
||||
<host>ch1</host>
|
||||
<port>9000</port>
|
||||
</replica>
|
||||
</shard>
|
||||
</cluster>
|
||||
</remote_servers>
|
||||
<macros>
|
||||
<shard>01</shard>
|
||||
</macros>
|
||||
<storage_configuration>
|
||||
<disks>
|
||||
<s3>
|
||||
<type>s3</type>
|
||||
<endpoint>http://minio1:9001/root/data/</endpoint>
|
||||
<access_key_id>minio</access_key_id>
|
||||
<secret_access_key>minio123</secret_access_key>
|
||||
</s3>
|
||||
</disks>
|
||||
<policies>
|
||||
<s3>
|
||||
<volumes>
|
||||
<default>
|
||||
<disk>default</disk>
|
||||
<perform_ttl_move_on_insert>False</perform_ttl_move_on_insert>
|
||||
</default>
|
||||
<s3>
|
||||
<disk>s3</disk>
|
||||
<perform_ttl_move_on_insert>False</perform_ttl_move_on_insert>
|
||||
</s3>
|
||||
</volumes>
|
||||
<move_factor>0.0</move_factor>
|
||||
</s3>
|
||||
</policies>
|
||||
</storage_configuration>
|
||||
|
||||
<merge_tree>
|
||||
<allow_remote_fs_zero_copy_replication>true</allow_remote_fs_zero_copy_replication>
|
||||
<storage_policy>s3</storage_policy>
|
||||
</merge_tree>
|
||||
<allow_remove_stale_moving_parts>true</allow_remove_stale_moving_parts>
|
||||
</clickhouse>
|
117
tests/integration/test_remove_stale_moving_parts/test.py
Normal file
117
tests/integration/test_remove_stale_moving_parts/test.py
Normal file
@ -0,0 +1,117 @@
|
||||
from pathlib import Path
|
||||
import time
|
||||
import pytest
|
||||
from helpers.cluster import ClickHouseCluster
|
||||
|
||||
cluster = ClickHouseCluster(__file__)
|
||||
ch1 = cluster.add_instance(
|
||||
"ch1",
|
||||
main_configs=[
|
||||
"config.xml",
|
||||
],
|
||||
macros={"replica": "node1"},
|
||||
with_zookeeper=True,
|
||||
with_minio=True,
|
||||
)
|
||||
|
||||
DATABASE_NAME = "stale_moving_parts"
|
||||
|
||||
|
||||
@pytest.fixture(scope="module")
|
||||
def started_cluster():
|
||||
try:
|
||||
cluster.start()
|
||||
yield cluster
|
||||
|
||||
finally:
|
||||
cluster.shutdown()
|
||||
|
||||
|
||||
def q(node, query):
|
||||
return node.query(database=DATABASE_NAME, sql=query)
|
||||
|
||||
|
||||
# .../disks/s3/store/
|
||||
def get_table_path(node, table):
|
||||
return (
|
||||
node.query(
|
||||
sql=f"SELECT data_paths FROM system.tables WHERE table = '{table}' and database = '{DATABASE_NAME}' LIMIT 1"
|
||||
)
|
||||
.strip('"\n[]')
|
||||
.split(",")[1]
|
||||
.strip("'")
|
||||
)
|
||||
|
||||
|
||||
def exec(node, cmd, path):
|
||||
return node.exec_in_container(
|
||||
[
|
||||
"bash",
|
||||
"-c",
|
||||
f"{cmd} {path}",
|
||||
]
|
||||
)
|
||||
|
||||
|
||||
def wait_part_is_stuck(node, table_moving_path, moving_part):
|
||||
num_tries = 5
|
||||
while q(node, "SELECT part_name FROM system.moves").strip() != moving_part:
|
||||
if num_tries == 0:
|
||||
raise Exception("Part has not started to move")
|
||||
num_tries -= 1
|
||||
time.sleep(1)
|
||||
num_tries = 5
|
||||
while exec(node, "ls", table_moving_path).strip() != moving_part:
|
||||
if num_tries == 0:
|
||||
raise Exception("Part is not stuck in the moving directory")
|
||||
num_tries -= 1
|
||||
time.sleep(1)
|
||||
|
||||
|
||||
def wait_zookeeper_node_to_start(zk_nodes, timeout=60):
|
||||
start = time.time()
|
||||
while time.time() - start < timeout:
|
||||
try:
|
||||
for instance in zk_nodes:
|
||||
conn = cluster.get_kazoo_client(instance)
|
||||
conn.get_children("/")
|
||||
print("All instances of ZooKeeper started")
|
||||
return
|
||||
except Exception as ex:
|
||||
print(("Can't connect to ZooKeeper " + str(ex)))
|
||||
time.sleep(0.5)
|
||||
|
||||
|
||||
def test_remove_stale_moving_parts_without_zookeeper(started_cluster):
|
||||
ch1.query(f"CREATE DATABASE IF NOT EXISTS {DATABASE_NAME}")
|
||||
|
||||
q(
|
||||
ch1,
|
||||
"CREATE TABLE test_remove ON CLUSTER cluster ( id UInt32 ) ENGINE ReplicatedMergeTree() ORDER BY id;",
|
||||
)
|
||||
|
||||
table_moving_path = Path(get_table_path(ch1, "test_remove")) / "moving"
|
||||
|
||||
q(ch1, "SYSTEM ENABLE FAILPOINT stop_moving_part_before_swap_with_active")
|
||||
q(ch1, "INSERT INTO test_remove SELECT number FROM numbers(100);")
|
||||
moving_part = "all_0_0_0"
|
||||
move_response = ch1.get_query_request(
|
||||
sql=f"ALTER TABLE test_remove MOVE PART '{moving_part}' TO DISK 's3'",
|
||||
database=DATABASE_NAME,
|
||||
)
|
||||
|
||||
wait_part_is_stuck(ch1, table_moving_path, moving_part)
|
||||
|
||||
cluster.stop_zookeeper_nodes(["zoo1", "zoo2", "zoo3"])
|
||||
# Stop moves in case table is not read-only yet
|
||||
q(ch1, "SYSTEM STOP MOVES")
|
||||
q(ch1, "SYSTEM DISABLE FAILPOINT stop_moving_part_before_swap_with_active")
|
||||
|
||||
assert "Cancelled moving parts" in move_response.get_error()
|
||||
assert exec(ch1, "ls", table_moving_path).strip() == ""
|
||||
|
||||
cluster.start_zookeeper_nodes(["zoo1", "zoo2", "zoo3"])
|
||||
wait_zookeeper_node_to_start(["zoo1", "zoo2", "zoo3"])
|
||||
q(ch1, "SYSTEM START MOVES")
|
||||
|
||||
q(ch1, f"DROP TABLE test_remove")
|
@ -1549,3 +1549,19 @@ def test_all_groups_cluster(started_cluster):
|
||||
assert "bad_settings_node\ndummy_node\n" == bad_settings_node.query(
|
||||
"select host_name from system.clusters where name='all_groups.db_cluster' order by host_name"
|
||||
)
|
||||
|
||||
|
||||
def test_detach_attach_table(started_cluster):
|
||||
main_node.query("DROP DATABASE IF EXISTS detach_attach_db SYNC")
|
||||
main_node.query(
|
||||
"CREATE DATABASE detach_attach_db ENGINE = Replicated('/clickhouse/databases/detach_attach_db');"
|
||||
)
|
||||
main_node.query(
|
||||
"CREATE TABLE detach_attach_db.detach_attach_table (k UInt64) ENGINE=ReplicatedMergeTree ORDER BY k;"
|
||||
)
|
||||
main_node.query("INSERT INTO detach_attach_db.detach_attach_table VALUES (1);")
|
||||
main_node.query("DETACH TABLE detach_attach_db.detach_attach_table PERMANENTLY;")
|
||||
main_node.query("ATTACH TABLE detach_attach_db.detach_attach_table;")
|
||||
assert (
|
||||
main_node.query("SELECT * FROM detach_attach_db.detach_attach_table;") == "1\n"
|
||||
)
|
||||
|
@ -560,7 +560,6 @@ positionCaseInsensitive
|
||||
positionCaseInsensitiveUTF8
|
||||
positionUTF8
|
||||
pow
|
||||
printf
|
||||
proportionsZTest
|
||||
protocol
|
||||
queryID
|
||||
|
@ -23,23 +23,3 @@ Row 1:
|
||||
x: 1
|
||||
2
|
||||
-- Bug 67476: Queries with overflow mode != throw must not be cached by the query cache
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
0
|
||||
|
@ -43,25 +43,15 @@ DROP TABLE IF EXISTS tab;
|
||||
CREATE TABLE tab(c UInt64) ENGINE = Memory;
|
||||
|
||||
SYSTEM DROP QUERY CACHE;
|
||||
SELECT sum(c) FROM tab SETTINGS read_overflow_mode = 'break', use_query_cache = 1;
|
||||
SELECT count(*) from system.query_cache;
|
||||
SELECT sum(c) FROM tab SETTINGS read_overflow_mode_leaf = 'break', use_query_cache = 1;
|
||||
SELECT count(*) from system.query_cache;
|
||||
SELECT sum(c) FROM tab SETTINGS group_by_overflow_mode = 'break', use_query_cache = 1;
|
||||
SELECT count(*) from system.query_cache;
|
||||
SELECT sum(c) FROM tab SETTINGS sort_overflow_mode = 'break', use_query_cache = 1;
|
||||
SELECT count(*) from system.query_cache;
|
||||
SELECT sum(c) FROM tab SETTINGS result_overflow_mode = 'break', use_query_cache = 1;
|
||||
SELECT count(*) from system.query_cache;
|
||||
SELECT sum(c) FROM tab SETTINGS timeout_overflow_mode = 'break', use_query_cache = 1;
|
||||
SELECT count(*) from system.query_cache;
|
||||
SELECT sum(c) FROM tab SETTINGS set_overflow_mode = 'break', use_query_cache = 1;
|
||||
SELECT count(*) from system.query_cache;
|
||||
SELECT sum(c) FROM tab SETTINGS join_overflow_mode = 'break', use_query_cache = 1;
|
||||
SELECT count(*) from system.query_cache;
|
||||
SELECT sum(c) FROM tab SETTINGS transfer_overflow_mode = 'break', use_query_cache = 1;
|
||||
SELECT count(*) from system.query_cache;
|
||||
SELECT sum(c) FROM tab SETTINGS distinct_overflow_mode = 'break', use_query_cache = 1;
|
||||
SELECT count(*) from system.query_cache;
|
||||
SELECT sum(c) FROM tab SETTINGS read_overflow_mode = 'break', use_query_cache = 1; -- { serverError QUERY_CACHE_USED_WITH_NON_THROW_OVERFLOW_MODE }
|
||||
SELECT sum(c) FROM tab SETTINGS read_overflow_mode_leaf = 'break', use_query_cache = 1; -- { serverError QUERY_CACHE_USED_WITH_NON_THROW_OVERFLOW_MODE }
|
||||
SELECT sum(c) FROM tab SETTINGS group_by_overflow_mode = 'break', use_query_cache = 1; -- { serverError QUERY_CACHE_USED_WITH_NON_THROW_OVERFLOW_MODE }
|
||||
SELECT sum(c) FROM tab SETTINGS sort_overflow_mode = 'break', use_query_cache = 1; -- { serverError QUERY_CACHE_USED_WITH_NON_THROW_OVERFLOW_MODE }
|
||||
SELECT sum(c) FROM tab SETTINGS result_overflow_mode = 'break', use_query_cache = 1; -- { serverError QUERY_CACHE_USED_WITH_NON_THROW_OVERFLOW_MODE }
|
||||
SELECT sum(c) FROM tab SETTINGS timeout_overflow_mode = 'break', use_query_cache = 1; -- { serverError QUERY_CACHE_USED_WITH_NON_THROW_OVERFLOW_MODE }
|
||||
SELECT sum(c) FROM tab SETTINGS set_overflow_mode = 'break', use_query_cache = 1; -- { serverError QUERY_CACHE_USED_WITH_NON_THROW_OVERFLOW_MODE }
|
||||
SELECT sum(c) FROM tab SETTINGS join_overflow_mode = 'break', use_query_cache = 1; -- { serverError QUERY_CACHE_USED_WITH_NON_THROW_OVERFLOW_MODE }
|
||||
SELECT sum(c) FROM tab SETTINGS transfer_overflow_mode = 'break', use_query_cache = 1; -- { serverError QUERY_CACHE_USED_WITH_NON_THROW_OVERFLOW_MODE }
|
||||
SELECT sum(c) FROM tab SETTINGS distinct_overflow_mode = 'break', use_query_cache = 1; -- { serverError QUERY_CACHE_USED_WITH_NON_THROW_OVERFLOW_MODE }
|
||||
|
||||
SYSTEM DROP QUERY CACHE;
|
||||
|
@ -45,3 +45,7 @@ uint64 UInt64 NO PRI SOR \N
|
||||
int32 Int32 NO \N
|
||||
str String NO \N
|
||||
uint64 UInt64 NO PRI SOR \N
|
||||
--- SHOW COLUMNS FROM table with dots
|
||||
int32 Nullable(Int32) YES \N
|
||||
str String NO SOR \N
|
||||
uint64 UInt64 NO PRI SOR \N
|
||||
|
@ -90,3 +90,18 @@ SHOW COLUMNS FROM database_123456789abcde.tab;
|
||||
DROP DATABASE database_123456789abcde;
|
||||
|
||||
DROP TABLE tab;
|
||||
|
||||
DROP TABLE IF EXISTS `tab.with.dots`;
|
||||
CREATE TABLE `tab.with.dots`
|
||||
(
|
||||
`uint64` UInt64,
|
||||
`int32` Nullable(Int32) COMMENT 'example comment',
|
||||
`str` String,
|
||||
INDEX idx str TYPE set(1000)
|
||||
)
|
||||
ENGINE = MergeTree
|
||||
PRIMARY KEY (uint64)
|
||||
ORDER BY (uint64, str);
|
||||
SELECT '--- SHOW COLUMNS FROM table with dots';
|
||||
SHOW COLUMNS FROM `tab.with.dots`;
|
||||
DROP TABLE `tab.with.dots`;
|
||||
|
@ -49,3 +49,10 @@ tbl 1 PRIMARY 1 a A 0 \N \N \N PRIMARY YES
|
||||
--- Short form
|
||||
tbl 1 mmi_idx 1 \N 0 \N \N \N MINMAX YES b
|
||||
tbl 1 PRIMARY 1 a A 0 \N \N \N PRIMARY YES
|
||||
--- SHOW INDEX FROM table with dots
|
||||
tab.with.dots 1 blf_idx 1 \N 0 \N \N \N BLOOM_FILTER YES d, b
|
||||
tab.with.dots 1 mm1_idx 1 \N 0 \N \N \N MINMAX YES a, c, d
|
||||
tab.with.dots 1 mm2_idx 1 \N 0 \N \N \N MINMAX YES c, d, e
|
||||
tab.with.dots 1 PRIMARY 1 c A 0 \N \N \N PRIMARY YES
|
||||
tab.with.dots 1 PRIMARY 2 a A 0 \N \N \N PRIMARY YES
|
||||
tab.with.dots 1 set_idx 1 \N 0 \N \N \N SET YES e
|
||||
|
@ -78,3 +78,22 @@ SHOW INDEX FROM database_123456789abcde.tbl;
|
||||
DROP DATABASE database_123456789abcde;
|
||||
|
||||
DROP TABLE tbl;
|
||||
|
||||
DROP TABLE IF EXISTS `tab.with.dots`;
|
||||
CREATE TABLE `tab.with.dots`
|
||||
(
|
||||
a UInt64,
|
||||
b UInt64,
|
||||
c UInt64,
|
||||
d UInt64,
|
||||
e UInt64,
|
||||
INDEX mm1_idx (a, c, d) TYPE minmax,
|
||||
INDEX mm2_idx (c, d, e) TYPE minmax,
|
||||
INDEX set_idx (e) TYPE set(100),
|
||||
INDEX blf_idx (d, b) TYPE bloom_filter(0.8)
|
||||
)
|
||||
ENGINE = MergeTree
|
||||
PRIMARY KEY (c, a);
|
||||
SELECT '--- SHOW INDEX FROM table with dots';
|
||||
SHOW INDEX FROM `tab.with.dots`;
|
||||
DROP TABLE `tab.with.dots`;
|
||||
|
@ -1,2 +1,2 @@
|
||||
default 127.0.0.1 9181 0 0 0 1 1 ['FILTERED_LIST','MULTI_READ','CHECK_NOT_EXISTS','CREATE_IF_NOT_EXISTS']
|
||||
default 127.0.0.1 9181 0 0 0 1 1 ['FILTERED_LIST','MULTI_READ','CHECK_NOT_EXISTS','CREATE_IF_NOT_EXISTS','REMOVE_RECURSIVE']
|
||||
zookeeper2 localhost 9181 0 0 0 1
|
||||
|
@ -0,0 +1,47 @@
|
||||
-- Simple types
|
||||
-- { echoOn }
|
||||
SELECT x FROM format(JSONEachRow, 'x Date', '{"x":""}');
|
||||
1970-01-01
|
||||
SELECT x FROM format(JSONEachRow, 'x Date32', '{"x":""}');
|
||||
1970-01-01
|
||||
SELECT toTimeZone(x, 'UTC') FROM format(JSONEachRow, 'x DateTime', '{"x":""}');
|
||||
1970-01-01 00:00:00
|
||||
SELECT toTimeZone(x, 'UTC') FROM format(JSONEachRow, 'x DateTime64', '{"x":""}');
|
||||
1970-01-01 00:00:00.000
|
||||
SELECT x FROM format(JSONEachRow, 'x IPv4', '{"x":""}');
|
||||
0.0.0.0
|
||||
SELECT x FROM format(JSONEachRow, 'x IPv6', '{"x":""}');
|
||||
::
|
||||
SELECT x FROM format(JSONEachRow, 'x UUID', '{"x":""}');
|
||||
00000000-0000-0000-0000-000000000000
|
||||
-- { echoOn }
|
||||
SELECT COUNT(DISTINCT col) FROM table1;
|
||||
1
|
||||
-- { echoOn }
|
||||
SELECT * FROM table1 ORDER BY address ASC;
|
||||
::
|
||||
2001:db8:3333:4444:5555:6666:7777:8888
|
||||
-- Nullable
|
||||
-- { echoOn }
|
||||
SELECT x FROM format(JSONEachRow, 'x Nullable(IPv6)', '{"x":""}');
|
||||
\N
|
||||
-- Compound types
|
||||
SELECT x FROM format(JSONEachRow, 'x Array(UUID)', '{"x":["00000000-0000-0000-0000-000000000000","b15f852c-c41a-4fd6-9247-1929c841715e",""]}');
|
||||
['00000000-0000-0000-0000-000000000000','b15f852c-c41a-4fd6-9247-1929c841715e','00000000-0000-0000-0000-000000000000']
|
||||
SELECT x FROM format(JSONEachRow, 'x Array(Nullable(IPv6))', '{"x":["",""]}');
|
||||
[NULL,NULL]
|
||||
SELECT x FROM format(JSONEachRow, 'x Tuple(Date, IPv4, String)', '{"x":["", "", "abc"]}');
|
||||
('1970-01-01','0.0.0.0','abc')
|
||||
SELECT x FROM format(JSONEachRow, 'x Map(String, IPv6)', '{"x":{"abc": ""}}');
|
||||
{'abc':'::'}
|
||||
SELECT x FROM format(JSONEachRow, 'x Variant(Date, UUID)', '{"x":""}');
|
||||
\N
|
||||
-- Deep composition
|
||||
SELECT x FROM format(JSONEachRow, 'x Array(Array(IPv6))', '{"x":[["2001:db8:3333:4444:CCCC:DDDD:EEEE:FFFF", ""], ["", "2001:db8:3333:4444:5555:6666:7777:8888"]]}');
|
||||
[['2001:db8:3333:4444:cccc:dddd:eeee:ffff','::'],['::','2001:db8:3333:4444:5555:6666:7777:8888']]
|
||||
SELECT x FROM format(JSONEachRow, 'x Variant(Date, Array(UUID))', '{"x":["", "b15f852c-c41a-4fd6-9247-1929c841715e"]}');
|
||||
['00000000-0000-0000-0000-000000000000','b15f852c-c41a-4fd6-9247-1929c841715e']
|
||||
SELECT x FROM format(JSONEachRow, 'x Tuple(Array(UUID), Tuple(UUID, Map(String, IPv6)))', '{"x":[[""], ["",{"abc":""}]]}');
|
||||
(['00000000-0000-0000-0000-000000000000'],('00000000-0000-0000-0000-000000000000',{'abc':'::'}))
|
||||
SELECT x FROM format(JSONEachRow, 'x Map(Tuple(Date,IPv4), Variant(UUID,IPv6))', '{"x":{["",""]:""}}');
|
||||
{('1970-01-01','0.0.0.0'):NULL}
|
60
tests/queries/0_stateless/03222_json_empty_as_default.sql
Normal file
60
tests/queries/0_stateless/03222_json_empty_as_default.sql
Normal file
@ -0,0 +1,60 @@
|
||||
SET input_format_json_empty_as_default = 1, allow_experimental_variant_type = 1;
|
||||
|
||||
-- Simple types
|
||||
-- { echoOn }
|
||||
SELECT x FROM format(JSONEachRow, 'x Date', '{"x":""}');
|
||||
SELECT x FROM format(JSONEachRow, 'x Date32', '{"x":""}');
|
||||
SELECT toTimeZone(x, 'UTC') FROM format(JSONEachRow, 'x DateTime', '{"x":""}');
|
||||
SELECT toTimeZone(x, 'UTC') FROM format(JSONEachRow, 'x DateTime64', '{"x":""}');
|
||||
SELECT x FROM format(JSONEachRow, 'x IPv4', '{"x":""}');
|
||||
SELECT x FROM format(JSONEachRow, 'x IPv6', '{"x":""}');
|
||||
SELECT x FROM format(JSONEachRow, 'x UUID', '{"x":""}');
|
||||
-- { echoOff }
|
||||
|
||||
-- Simple type AggregateFunction
|
||||
DROP TABLE IF EXISTS table1;
|
||||
CREATE TABLE table1(col AggregateFunction(uniq, UInt64)) ENGINE=Memory();
|
||||
DROP TABLE IF EXISTS table2;
|
||||
CREATE TABLE table2(UserID UInt64) ENGINE=Memory();
|
||||
|
||||
INSERT INTO table1 SELECT uniqState(UserID) FROM table2;
|
||||
INSERT INTO table1 SELECT x FROM format(JSONEachRow, 'x AggregateFunction(uniq, UInt64)' AS T, '{"x":""}');
|
||||
|
||||
-- { echoOn }
|
||||
SELECT COUNT(DISTINCT col) FROM table1;
|
||||
-- { echoOff }
|
||||
|
||||
DROP TABLE table1;
|
||||
DROP TABLE table2;
|
||||
|
||||
-- The setting input_format_defaults_for_omitted_fields determines the default value if enabled.
|
||||
CREATE TABLE table1(address IPv6 DEFAULT toIPv6('2001:db8:3333:4444:5555:6666:7777:8888')) ENGINE=Memory();
|
||||
|
||||
SET input_format_defaults_for_omitted_fields = 0;
|
||||
INSERT INTO table1 FORMAT JSONEachRow {"address":""};
|
||||
|
||||
SET input_format_defaults_for_omitted_fields = 1;
|
||||
INSERT INTO table1 FORMAT JSONEachRow {"address":""};
|
||||
|
||||
-- { echoOn }
|
||||
SELECT * FROM table1 ORDER BY address ASC;
|
||||
-- { echoOff }
|
||||
|
||||
DROP TABLE table1;
|
||||
|
||||
-- Nullable
|
||||
-- { echoOn }
|
||||
SELECT x FROM format(JSONEachRow, 'x Nullable(IPv6)', '{"x":""}');
|
||||
|
||||
-- Compound types
|
||||
SELECT x FROM format(JSONEachRow, 'x Array(UUID)', '{"x":["00000000-0000-0000-0000-000000000000","b15f852c-c41a-4fd6-9247-1929c841715e",""]}');
|
||||
SELECT x FROM format(JSONEachRow, 'x Array(Nullable(IPv6))', '{"x":["",""]}');
|
||||
SELECT x FROM format(JSONEachRow, 'x Tuple(Date, IPv4, String)', '{"x":["", "", "abc"]}');
|
||||
SELECT x FROM format(JSONEachRow, 'x Map(String, IPv6)', '{"x":{"abc": ""}}');
|
||||
SELECT x FROM format(JSONEachRow, 'x Variant(Date, UUID)', '{"x":""}');
|
||||
|
||||
-- Deep composition
|
||||
SELECT x FROM format(JSONEachRow, 'x Array(Array(IPv6))', '{"x":[["2001:db8:3333:4444:CCCC:DDDD:EEEE:FFFF", ""], ["", "2001:db8:3333:4444:5555:6666:7777:8888"]]}');
|
||||
SELECT x FROM format(JSONEachRow, 'x Variant(Date, Array(UUID))', '{"x":["", "b15f852c-c41a-4fd6-9247-1929c841715e"]}');
|
||||
SELECT x FROM format(JSONEachRow, 'x Tuple(Array(UUID), Tuple(UUID, Map(String, IPv6)))', '{"x":[[""], ["",{"abc":""}]]}');
|
||||
SELECT x FROM format(JSONEachRow, 'x Map(Tuple(Date,IPv4), Variant(UUID,IPv6))', '{"x":{["",""]:""}}');
|
@ -0,0 +1,8 @@
|
||||
Array(UUID)
|
||||
{"x":["00000000-0000-0000-0000-000000000000","b15f852c-c41a-4fd6-9247-1929c841715e","00000000-0000-0000-0000-000000000000"]}
|
||||
{"x":["00000000-0000-0000-0000-000000000000","b15f852c-c41a-4fd6-9247-1929c841715e","00000000-0000-0000-0000-000000000000"]}
|
||||
{"x":["00000000-0000-0000-0000-000000000000","b15f852c-c41a-4fd6-9247-1929c841715e","00000000-0000-0000-0000-000000000000"]}
|
||||
Tuple(Array(UUID), Tuple(UUID, Map(String, IPv6)))
|
||||
{"x":[["00000000-0000-0000-0000-000000000000"],["00000000-0000-0000-0000-000000000000",{"abc":"::"}]]}
|
||||
{"x":[["00000000-0000-0000-0000-000000000000"],["00000000-0000-0000-0000-000000000000",{"abc":"::"}]]}
|
||||
{"x":[["00000000-0000-0000-0000-000000000000"],["00000000-0000-0000-0000-000000000000",{"abc":"::"}]]}
|
31
tests/queries/0_stateless/03222_json_empty_as_default_small_read_buffer.sh
Executable file
31
tests/queries/0_stateless/03222_json_empty_as_default_small_read_buffer.sh
Executable file
@ -0,0 +1,31 @@
|
||||
#!/usr/bin/env bash
|
||||
# Tags: no-parallel
|
||||
|
||||
CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)
|
||||
# shellcheck source=../shell_config.sh
|
||||
. "$CURDIR"/../shell_config.sh
|
||||
|
||||
DATA_FILE=$CLICKHOUSE_TEST_UNIQUE_NAME.json
|
||||
|
||||
# Wrapper for clickhouse-client to always output in JSONEachRow format, that
|
||||
# way format settings will not affect output.
|
||||
function clickhouse_local()
|
||||
{
|
||||
$CLICKHOUSE_LOCAL --output-format JSONEachRow "$@"
|
||||
}
|
||||
|
||||
echo 'Array(UUID)'
|
||||
echo '{"x":["00000000-0000-0000-0000-000000000000","b15f852c-c41a-4fd6-9247-1929c841715e",""]}' > $DATA_FILE
|
||||
# Use increasingly smaller read buffers.
|
||||
clickhouse_local -q "SELECT x FROM file('$DATA_FILE', 'JSONEachRow', 'x Array(UUID)') SETTINGS input_format_json_empty_as_default=1, input_format_parallel_parsing=0, storage_file_read_method='read', max_read_buffer_size=4"
|
||||
clickhouse_local -q "SELECT x FROM file('$DATA_FILE', 'JSONEachRow', 'x Array(UUID)') SETTINGS input_format_json_empty_as_default=1, input_format_parallel_parsing=0, storage_file_read_method='read', max_read_buffer_size=2"
|
||||
clickhouse_local -q "SELECT x FROM file('$DATA_FILE', 'JSONEachRow', 'x Array(UUID)') SETTINGS input_format_json_empty_as_default=1, input_format_parallel_parsing=0, storage_file_read_method='read', max_read_buffer_size=1"
|
||||
|
||||
echo 'Tuple(Array(UUID), Tuple(UUID, Map(String, IPv6)))'
|
||||
echo '{"x":[[""], ["",{"abc":""}]]}' > $DATA_FILE
|
||||
# Use increasingly smaller read buffers.
|
||||
clickhouse_local -q "SELECT x FROM file('$DATA_FILE', 'JSONEachRow', 'x Tuple(Array(UUID), Tuple(UUID, Map(String, IPv6)))') SETTINGS input_format_json_empty_as_default=1, input_format_parallel_parsing=0, storage_file_read_method='read', max_read_buffer_size=16"
|
||||
clickhouse_local -q "SELECT x FROM file('$DATA_FILE', 'JSONEachRow', 'x Tuple(Array(UUID), Tuple(UUID, Map(String, IPv6)))') SETTINGS input_format_json_empty_as_default=1, input_format_parallel_parsing=0, storage_file_read_method='read', max_read_buffer_size=8"
|
||||
clickhouse_local -q "SELECT x FROM file('$DATA_FILE', 'JSONEachRow', 'x Tuple(Array(UUID), Tuple(UUID, Map(String, IPv6)))') SETTINGS input_format_json_empty_as_default=1, input_format_parallel_parsing=0, storage_file_read_method='read', max_read_buffer_size=1"
|
||||
|
||||
rm $DATA_FILE
|
Loading…
Reference in New Issue
Block a user