4ccdd65bf6
[Fix](array) fix mysql_row_buffer may use after free when reserve() delete original address in dynamic_mode ( #11395 )
...
```
if (!_dynamic_mode) {
int8store(_len_pos, _pos - _len_pos - 8);
_len_pos = nullptr;
}
```
_len_pos may be pointed to the pos which already deleted in reserve, int8store will asign value to the freed address,
and lead to use after free when build in ASAN.So I changed _len_pos to the offset of _buf
2022-08-01 22:52:19 +08:00
667689e9ba
[Fix](array) fix array permute ( #11389 )
2022-08-01 22:46:03 +08:00
69961ac19c
fix bug for lower case when rename table name ( #11403 )
2022-08-01 22:30:41 +08:00
6da6ffb836
[feature] (Nereids) add rewrite rule to merge consecutive filter nodes ( #11248 )
...
this rule aims to merge consecutive filters.
For example:
logical plan tree:
```
project
|
filter(a>0)
|
filter(b>0)
|
scan
```
transformed to:
```
project
|
filter(a>0 and b>0)
|
scan
```
2022-08-01 17:29:15 +08:00
69bfbae856
[enhancement](nereids) Normalize expressions before performing plan rewriting ( #11299 )
...
Rules for normalizing expressions should be applied once before do some extra expression transforms.
Normalization rules include:
1. NormalizeBinaryPredicatesRule
2. BetweenToCompoundRule
3. SimplifyNotExprRule
2022-08-01 17:15:04 +08:00
eb778da64a
Add ssb sf0.1 test under unique table with zstd ( #11348 )
...
Co-authored-by: smallhibiscus <844981280>
2022-08-01 17:04:55 +08:00
7ca6f8b6a2
[refactor](Nereids)split rewrite and insert into memo to 2 functions ( #11342 )
...
Split rewrite and insert into memo to 2 functions to make the code easy to read.
2022-08-01 14:44:01 +08:00
8287115ec3
[test]Delete some unnecessary information ( #11382 )
2022-08-01 14:31:39 +08:00
8f85ad1c96
[doc](intro) update the introduce ( #11364 )
2022-08-01 14:22:17 +08:00
8a8e8e8b45
[fix](stop-script) use kill -9 to stop fe as usual ( #11387 )
2022-08-01 14:18:10 +08:00
1cf57a985d
[fix] Fix the query result error caused by the grouping sets statemen… ( #11316 )
...
* [fix] Fix the query result error caused by the grouping sets statement grouping as an expression
2022-08-01 13:52:18 +08:00
4f5e1601df
[bug](scanner) Improve limit query performance on olapScannode and avoid infinite loop ( #11301 )
...
1. Fix a bug that query large column table may cause infinite loop
2. Optimize the query logic with limit, for the case where the limit value is relatively small, reduce the parallelism of the scanner, reduce unnecessary resource consumption, and increase the number of similar queries that the system can carry at the same time, and increase the query speed by more than 60%
2022-08-01 13:50:12 +08:00
b35daf0a04
[improvement](light-schema-change) Support tablet schema cache ( #11131 )
2022-08-01 12:18:00 +08:00
0325fa436e
[fix](agg)Add field of 'is_first_phase' in TAggregationNode ( #11321 )
2022-08-01 11:49:50 +08:00
52460af74b
[Bug][Vectorized] Support the .* in hyperscan to valid the % in SQL ( #11371 )
...
Co-authored-by: lihaopeng <lihaopeng@baidu.com >
2022-08-01 11:00:05 +08:00
34ed40bb71
[Feature](array) doe support array ( #11340 )
2022-08-01 10:56:40 +08:00
76970200ec
[Bugfix](light-schema-change) Fix bf columns message miss when load ( #11367 )
2022-08-01 10:53:25 +08:00
11f6c69a22
[bugfix](start script)fix a bug while JAVA_HOME not set ( #11366 )
2022-08-01 10:44:03 +08:00
547c18d28b
[feature](load) support CLEAN LABEL stmt ( #11362 )
2022-08-01 10:43:33 +08:00
d360974dce
[improvement](agg)Use phmap::flat_hash_set in AggregateFunctionUniq ( #11363 )
...
This reverts commit 688b55053dd1fc5113343a6f565ad732ddd9612a.
2022-08-01 10:36:11 +08:00
61b157abe3
[bug](storage-scanner) fix infinite loop when reading wide table ( #11372 )
...
* fix infinite loop when reading wide table
When a wide table is read, the 1st batch would be exceed raw_bytes_threshold,
so Scanner should read at least 1 row.
Actually, we should adjust batch size automatically to reduce memoery usage.
2022-08-01 10:33:04 +08:00
3f43d7262b
[Bug] (datetimev2) fix bugs for datev2/datetimev2 ( #11358 )
2022-08-01 10:29:56 +08:00
73d8f5901d
fix mem tracker limiter ( #11376 )
2022-08-01 09:44:04 +08:00
e002b55b01
fix be config buffer_pool_limit default value ( #11373 )
...
fix be config buffer_pool_limit default value
2022-08-01 08:46:15 +08:00
388db05ef9
[bugfix](log4j) Upgrade log4j to 2.18.0 ( #11368 )
2022-07-31 22:21:33 +08:00
d1d66f7ffc
[chore](thirdparty) Build script doesn't interrupt when errors occur. ( #11357 )
...
1. Add a comment
2. Remove all &&
2022-07-31 20:53:50 +08:00
e20e6624ab
[feature](nereids): polish property deriver enforcer job ( #11222 )
...
Polish property deriver enforcer job
2022-07-31 17:22:41 +08:00
ab7e65112c
[fix](thirdparty) fix compile error: missing libbrpc.a ( #11353 )
...
fix compile error, add fpic to brpc dependency library
2022-07-31 13:26:08 +08:00
407a6b5c64
[community](readme) update readme and some links ( #11354 )
2022-07-31 10:38:02 +08:00
2c810f577b
[improve](ci): allow blank, -, _ ( #11349 )
2022-07-31 09:51:02 +08:00
688b55053d
Revert "[improvement]Use phmap::flat_hash_set in AggregateFunctionUniq ( #11257 )" ( #11356 )
...
This reverts commit a7199fb98e18b925664b38460b667d04cbee8e01.
2022-07-30 23:15:36 +08:00
2783267599
[feature-wip](unique-key-merge-on-write) update bitmap after compaction, DSIP-018 ( #11289 )
2022-07-30 19:50:47 +08:00
9333e79ae0
[feature-wip](unique-key-merge-on-write) Add support for tablet migration, DSIP-018[5/3] ( #11283 )
2022-07-30 19:50:11 +08:00
1f30e563a7
[refactor][vectorized] refactor first/last value agg functions ( #10661 )
...
* refactor first and last
[refactor][vectorized] refactor first/last value agg functions
* add some change
* remove first/last about always nullable
* remove always nullable and register it
* refactor value remove bool null flag
* refactor win first last to ptr and pos
2022-07-30 18:38:56 +08:00
18864ab7fe
weak relationship between MemTracker and MemTrackerLimiter ( #11347 )
2022-07-30 18:33:54 +08:00
fb90586122
Fix bug in s3 delete directory ( #11344 )
2022-07-30 18:31:32 +08:00
0497d58943
[FIX]DCHECK error of array functions ( #11335 )
2022-07-30 14:15:13 +08:00
1ada6d8641
[feature-wip](unique-key-merge-on-write) fix IndexedColumnReader and add primary key index memory tracker, DSIP-018 ( #11303 )
2022-07-30 10:53:40 +08:00
532395c6a0
[Bug][Function] core dump on sum(distinct) ( #11308 )
...
* fix align size calculate for distinct combinator
2022-07-30 10:24:48 +08:00
9d9fcca40c
[regression-test](array)add test for array_slice ( #11329 )
2022-07-30 09:59:58 +08:00
5999666115
[fix](planner)do not change non-null to nullable for outer join table slot ( #11268 )
...
* remove nullable for outer join table
* fix ut
* remove useless comment
* fix for original engine
2022-07-29 21:34:24 +08:00
c1fbee7fe1
[fix] the nullable info is lost in ifnull expr ( #11212 )
...
ifnull function has defect when processing nullable column or const column in some case
2022-07-29 21:33:58 +08:00
5215d95064
[enhancement](workflow) Use ccache to speed the BE UT (Clang) up ( #11339 )
2022-07-29 21:19:26 +08:00
a46f94320d
[fix](ci): add checkout to fix PR-title-checker ( #11336 )
2022-07-29 21:09:41 +08:00
e7fae413dd
[feature](nereids) add scalar subquery expression ( #11332 )
...
scalar subquery:
A subquery that will return only one row and one column.
A limit has been added, where a = subquery returns a result of 1 row and 1 column.
Here, the first limit is to return 1 column.
TODO: the subsequent return will limit the return to 1 row
2022-07-29 21:05:19 +08:00
8483660fe7
[opt] unify stop script ( #11337 )
2022-07-29 21:04:03 +08:00
e7635e54ee
[doc](unique key) add suggestion for replace_if_not_null ( #11223 )
2022-07-29 20:02:18 +08:00
a7e7df658c
[Doc] update FAQ about ODBC ( #11263 )
2022-07-29 19:18:28 +08:00
d6f937cb01
(performance)[scanner] Isolate local and remote queries using different scanner… ( #11006 )
2022-07-29 19:14:46 +08:00
84ce2a1e98
[feature-wip](multi-catalog)(fix) partition value error when a block contains multiple splits ( #11260 )
...
`FileArrowScanner::get_next` returns a block when full, so it maybe contains multiple
splits in small files or crosses two splits in large files.
However, a block can only fill the partition values from one file. Different splits may be
from different files, causing the error of embed partition values.
2022-07-29 18:48:59 +08:00
