Commit Graph

3811 Commits

Author SHA1 Message Date
4e783afa7a [feature] add Generic debug timer for debugging or profiling (#7923)
add a group of debug-timer for the purpose of profiling or testing
you can use these timers for custom meaning purpose unlike the specific named timer
2022-01-31 22:15:43 +08:00
a6962af30f [typo](doc) rename DDL Statements to SQL Statements in sidebars of docs in english (#7922) 2022-01-31 22:15:10 +08:00
b1b1a8c64d [chore] Fix the return code of function download_func (#7919) 2022-01-31 22:14:46 +08:00
358bd79fb1 [improvement](vec)(Join) Mem reuse to speed up join operator (#7905)
1. Reuse the mem of output block in vec join node
2. Add the function `replicate` in column
2022-01-31 22:14:12 +08:00
68c4bb34e6 [chore] Build script add build broker option (#7885) 2022-01-31 22:13:25 +08:00
14c209c7cf [refactor] remove useless if statement in segment_writer.cpp (#7864) 2022-01-31 22:12:54 +08:00
4ada8e4854 [fix](httpv2) make http v2 and v1 interface compatible (#7848)
http v2 TableSchemaAction adds the return value of aggregation_type,
and modifies the corresponding code of Flink/Spark Connector
2022-01-31 22:12:34 +08:00
cacc29470b [demo] Add Flink oracle cdc demo (#7845) 2022-01-31 22:11:38 +08:00
Pxl
2003da7cf9 [fix](ut) fix abs function ut (#7938) 2022-01-31 14:58:29 +08:00
3f221e1d0b [fix](memory-leak) using unique_ptr to refactor some fields (#7933)
Using unique_ptr to refactor some class members.
Fix mem leak for `SegmentIterator`'s `_pre_eval_block_predicate`.
2022-01-30 16:49:04 +08:00
7a1ad65d88 [fix](fe-ui) fix home page spin bug #7170 (#7912)
The Spin(loading circle) will disappear after hardware info loaded
2022-01-30 16:47:58 +08:00
Pxl
3ee000c13c [chore] support build with libc++ && add some build config (#7903)
support LIBCPP/LDD/BUILD_META_TOOL for build.sh
2022-01-30 16:47:22 +08:00
c1fef37399 [improvement](runtime-filter) Support adaptive runtime filter(#7546) (#7645)
Change 1: Support an adaptive runtime filter: IN_OR_BLOOM_FILTER
    The processing logic is
    If the number of rows in the right table < runtime_filter_max_in_num, then IN predicate will work
    If the number of rows in the right table >= runtime_filter_max_in_num, then Bloom filter can take effect

Change 2: The default runtime filter is changed to filter: IN_OR_BLOOM_FILTER
2022-01-30 16:46:52 +08:00
7325e9d7f7 [chore] Remove BUILD_TYPE bcc (#7893)
no need anymore
2022-01-30 15:55:20 +08:00
3904447db8 [fix](toolchain) Allow building when system contains libunwind (#7866) 2022-01-29 12:33:32 +08:00
a72eaa2b2e [fix](Vectorized) optinmize dict page decoder init (#7917)
this may cause mem leak
2022-01-29 11:47:57 +08:00
4c7525cf2c [improvement](show) Support that user can use show data skew statement instead of admin (#7914)
* [improvement](show) Support that user can use show data skew statement instead of admin
This PR mainly do two things:
1. Support that user can use show data skew statement instead of admin
2. Fix fe ut failed caused by pr [improvement](rewrite) Make RewriteDateLiteralRule to be compatible with mysql #7876 and pr [feature-wip](iceberg) Step1: Support create Iceberg external table #7391

Co-authored-by: caiconghui1 <caiconghui1@jd.com>
2022-01-29 10:45:03 +08:00
6a1a2a2ed5 [fix](query) Add init function for result_file_sink (#7927)
Add init function in `result_file_sink` to fix the error "Empty partition info",
which is occasional reported when using SELECT INFO OUTFILE.
2022-01-29 10:08:57 +08:00
1d900d8605 (fix)[planner] Fix the right tuple ids in empty set node (#7931)
The tuple ids of the empty set node must be exactly the same as the tuple ids of the origin root node.
In the issue, we found that once the tree where the root node is located has a window function,
the tuple ids of the empty set node cannot be calculated correctly.

This pr mostly fixes the problem.
In order to calculate the correct tuple ids,
the tuple ids obtained from the SelectStmt.getMaterializedTupleIds() function in the past
are changed to directly use the tuple ids of the origin root node.

Although we tried to fix #7929 by modifying the SelectStmt.getMaterializedTupleIds() function,
this method can't get the tuple of the last correct window function.
So we use other ways to construct tupleids of empty nodes.
2022-01-29 09:46:05 +08:00
fb6e22f4ca [Fix] fix memory leak in be unit test (#7857)
1. fix be unit test memory leak
2. ignore mindump test with ASAN test
2022-01-29 01:00:38 +08:00
2d99041ec0 [docs] fix typo (#7907) 2022-01-28 22:32:50 +08:00
071be928f9 [fix](vectorized) fix bug multi distinct function get wrong type (#7900) 2022-01-28 22:31:41 +08:00
6a89b893a3 [refactor](benchmark) Change SSB create table column to NOT NULL (#7898) 2022-01-28 22:26:55 +08:00
f9ac302807 [docs] add function substring document (#7895) 2022-01-28 22:26:11 +08:00
3a7bb7e144 [improvement](fe-meta-version)Some if conditions do not use the FeMetaVersion constant (#7879) 2022-01-28 22:25:17 +08:00
f93ac89a67 [fix](lateral-view) fix bugs of lateral view with CTE or where clause (#7865)
fix bugs of lateral view with CTE or where clause.
The error case can be found in newly added tests in `TableFunctionPlanTest.java`
But there are still some bugs not being fixed, so the unit test is annotated with @Ignore

This PR contains the change is #7824 :

> Issue Number: close #7823
> 
> After the subquery is rewritten, the rewritten stmt needs to be reset
> (that is, the content of the first analyze semantic analysis is cleared),
> and then the rewritten stmt can be reAnalyzed.
> 
> The lateral view ref in the previous implementation forgot to implement the reset function.
> This caused him to keep the first error message in the second analyze.
> Eventually, two duplicate tupleIds appear in the new stmt and are marked with different tuple.
> From the explain string, the following syntax will have an additional wrong join predicate.
> ```
> Query: explain select k1 from test_explode lateral view explode_split(k2, ",") tmp as e1  where k1 in (select k3 from tbl1);
> Error equal join conjunct: `k3` = `k3`
> ```
> 
> This pr mainly adds the reset function of the lateral view
> to avoid possible errors in the second analyze
> when the lateral view and subquery rewrite occur at the same time.
2022-01-28 22:24:23 +08:00
22830ea498 [feature](show) add new statement show proc '/current_query_stmts' (#7487)
To show the the query statement at first level.
2022-01-28 22:23:13 +08:00
1ba20b1dbb [improvement](storage) improving Column inserter (#7855)
* optimize Column inserter

* DCHECK

* DCHECK

Co-authored-by: zuochunwei <zuochunwei@meituan.com>
2022-01-27 14:18:15 +08:00
dee79d98a8 [improvement](explain) Displays cast information with implicit conversions in verbose (#7851)
Displays cast information with implicit conversions in verbose.
2022-01-27 10:37:38 +08:00
0d3fe8f07b [typo](docs) fix document backup-restore.md typo (#7868)
Refactor the document format and improve the content of the English version
2022-01-27 10:34:34 +08:00
ea96242529 [chore] add set -e to download_thirdparty.sh (#7867)
Let the build process stop immediately if some command execute failed.
For example, patch failed due to command not found, otherwise it will stop until compile error.
2022-01-27 10:33:08 +08:00
d2386dd85d [improvement](rewrite) Make RewriteDateLiteralRule to be compatible with mysql (#7876) 2022-01-27 10:32:18 +08:00
d69b7bff2e [feature](meta) Support show compactionTooSlowTablets and oversizeTablets (#7821)
Add more columns in `show proc "/statistic"`
2022-01-27 10:26:41 +08:00
3b8d48f08b [feature-wip](iceberg) Step1: Support create Iceberg external table (#7391)
Close related #7389

Support create Iceberg external table in Doris. 

This is the first step to support Iceberg external table.

### Create Iceberg external table
This pr describes two ways to create Iceberg external tables. Both ways do not require explicitly specifying column definitions, Doris automatically converts them based on Iceberg's column definitions.

1. Create an Iceberg external table directly

```sql
    CREATE [EXTERNAL] TABLE table_name 
    ENGINE = ICEBERG
    [COMMENT "comment"]
    PROPERTIES (
    "iceberg.database" = "iceberg_db_name",
    "iceberg.table" = "icberg_table_name",
    "iceberg.hive.metastore.uris"  =  "thrift://192.168.0.1:9083",
    "iceberg.catalog.type"  =  "HIVE_CATALOG"
    );
```

2. Create an Iceberg database and automatically create all the tables under that db.

```sql
    CREATE DATABASE db_name 
    [COMMENT "comment"]
    PROPERTIES (
    "iceberg.database" = "iceberg_db_name",
    "iceberg.hive.metastore.uris" = "thrift://192.168.0.1:9083",
    "iceberg.catalog.type" = "HIVE_CATALOG"
    );
```

### Show table creation

1. For individual tables you can view them with `help show create table`.

```sql 
mysql> show create table iceberg_db.logs_1;
+--------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Table  | Create Table                                                                                                                                                                                                                                                                                                                                                 |
+--------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| logs_1 | CREATE TABLE `logs_1` (
  `level` varchar(-1) NOT NULL COMMENT "null",
  `event_time` datetime NOT NULL COMMENT "null",
  `message` varchar(-1) NOT NULL COMMENT "null"
) ENGINE=ICEBERG
COMMENT "ICEBERG"
PROPERTIES (
"iceberg.database" = "doris",
"iceberg.table" = "logs_1",
"iceberg.hive.metastore.uris"  =  "thrift://10.10.10.10:9087",
"iceberg.catalog.type"  =  "HIVE_CATALOG"
) |
+--------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
```

2. For Iceberg database, you can view it with `help show table creation`.

```sql
mysql> show table creation from iceberg_db;
+--------+---------+---------------------+---------------------------------------------------------+
| Table  | Status  | Create Time         | Error Msg                                               |
+--------+---------+---------------------+---------------------------------------------------------+
| logs   | fail    | 2021-12-14 13:50:10 | Cannot convert unknown type to Doris type: list<string> |
| logs_1 | success | 2021-12-14 13:50:10 |                                                         |
+--------+---------+---------------------+---------------------------------------------------------+
2 rows in set (0.00 sec)
```

  This is a new syntax.
  
  Show table creation records in Iceberg database:
  
  Syntax:
  ```sql
      SHOW TABLE CREATION [FROM db] [LIKE mask]
  ```
2022-01-27 10:22:47 +08:00
df76a5b34c refactor SegmentIterator (#7852)
Co-authored-by: zuochunwei <zuochunwei@meituan.com>
2022-01-26 16:44:02 +08:00
ec5ecd1604 handle conflict (#7836)
Co-authored-by: zuochunwei <zuochunwei@meituan.com>
2022-01-26 16:33:37 +08:00
015371ac72 [fix](grouping-set) Fix the bug of grouping set core in both vec and non vec query engine (#7800) 2022-01-26 16:15:30 +08:00
38660d5095 [docs]correct SeaTunnel website hyperlink and PPMC into the Member List (#7886)
* [docs]correct SeaTunnel website hyperlink
* [docs]Add the new PPMC into the Member List

* Update members.md
2022-01-26 15:24:52 +08:00
f227472db2 [chore] fix error while compiling with -O3 (#7890) 2022-01-26 12:53:56 +08:00
Pxl
cd73a6b84b [chore] fix clang compile error (#7883) 2022-01-26 12:53:35 +08:00
e3bc232578 [docs] OS Installation Requirements (#7830) 2022-01-26 09:11:41 +08:00
4bdeef3b64 [chore][fix][doc](fe-plugin)(mysqldump) fix build auditlog plugin error (#7804)
1. fix problems when build fe_plugins
2. format
3. add docs about dump data using mysql dump
2022-01-26 09:11:23 +08:00
b435a54304 [fix] Consider backend status when more than one backends exists in same host (#7784) 2022-01-26 09:10:34 +08:00
461b352d3e [fix](function) Change digital_masking function arg type to BIGINT (#7888)
Change digital_masking function arg type to BIGINT to fix the wrong result.
2022-01-25 22:28:05 +08:00
a6831535e9 [Vectorized][Bug] fix bug of coalesce function (#7827) 2022-01-25 20:44:16 +08:00
ee0037e1af [fix](ldap) fix ldap password logic (#7862)
1. write ldap info to image;
2. optimizing LdapClient class thread safety.
2022-01-25 09:59:24 +08:00
40f993ca15 [docs][seatunnel] add seatunnel flink doris sink doc (#7844) 2022-01-24 21:13:06 +08:00
8aa9faa7cb [chore](docker) Add docker dev image with ldb-toolchain (#7838)
Add docker images `apache/incubator-doris:build-env-ldb-toolchain-latest`,
which is built with ldb-toolchain
2022-01-24 21:12:15 +08:00
be9ebbc14d [fix] Fix bug that wrong exception message returned by insert statement (#7832)
`insert` statement may return exception message `Execute timeout` after load data failed.
But the real reason is that there exists unhealthy backend, not execute timeout.

```
MySQL [ssb]> insert into lineorder_flat select * from lineorder_flat;
ERROR 1105 (HY000): errCode = 2, detailMessage = Execute timeout
```
2022-01-24 21:11:45 +08:00
4e9bc5cb65 [doc] add documents for bitwise functions (#7790) 2022-01-24 21:08:41 +08:00