Bugfixing for mysql_fdw #239
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
added 3 commits
March 29, 2022 15:14
**Describe the bug**
mysql_fdw deparse the Subscripting Array to be the same as postgres_fdw.
However, MySQL have no type as ARRAY in Postgresql, so we don't pushdown Array Subscripting.
**To Reproduce**
1. Create table on MySQL server
mysql -h $MYSQL_HOST -u $MYSQL_USER_NAME -D $MYSQL_PORT -D mysql_fdw_regress -e "CREATE TABLE s6(id int PRIMARY KEY, c1 int, c2 int, c3 text);"
2. Execute SQL
CREATE FOREIGN TABLE ft5 (id int, c1 int, c2 int, c3 text)
SERVER mysql_svr OPTIONS (dbname 'mysql_fdw_regress', table_name 's6');
INSERT INTO ft5
SELECT id,
id,
id % 5,
to_char(id, 'FM00000')
FROM generate_series(1, 10) id;
3. Result before fixing
EXPLAIN VERBOSE
SELECT * FROM ft5 t1 WHERE c1 = (ARRAY[c1,c2,3])[2];
ERROR: failed to prepare the MySQL query:
You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near '[`c1`, `c2`, 3])[2])))' at line 1
**Result after fixing**
EXPLAIN VERBOSE
SELECT * FROM ft5 t1 WHERE c1 = (ARRAY[c1,c2,3])[2];
QUERY PLAN
--------------------------------------------------------------------------------------------------------------------
Foreign Scan on public.ft5 t1 (cost=100.00..125.17 rows=6 width=44)
Output: id, c1, c2, c3
Local server startup cost: 10
Remote query: SELECT `id`, `c1`, `c2`, `c3` FROM `mysql_fdw_regress`.`s6` WHERE ((`c1` = ELT(2, `c1`, `c2`, 3)))
(4 rows)
**Additional context**
Use Mysql ELT(N,str1,str2,str3,...) function to support array subscripting function:
+ Support 1D array subscription.
+ Do not support multiple dimension array
+ Do not support slice function
+ Do not support Var subscription.
**Describe the bug**
Because MySQL does not support tableoid, ctid,...
So not pushdown system column instead of raise error.
**To Reproduce**
1. Execute SQL
CREATE FOREIGN TABLE ft1 (
c0 int,
c1 int NOT NULL,
c2 int NOT NULL,
c3 text,
c4 timestamptz,
c5 timestamp,
c6 varchar(10),
c7 char(10) default 'ft1',
c8 user_enum
) SERVER mysql_svr OPTIONS (dbname 'mysql_fdw_post');
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 t1 ORDER BY t1.c3, t1.c1, t1.tableoid OFFSET 100 LIMIT 10;
2.Result before fixing
EXPLAIN (VERBOSE, COSTS OFF) SELECT * FROM ft1 t1 ORDER BY t1.c3, t1.c1, t1.tableoid OFFSET 100 LIMIT 10;
ERROR: system attribute tableoid can't be fetched from remote relation
**Result after fixing**
QUERY PLAN
----------------------------------------------------------------------------------------------------------------
Limit
Output: c1, c2, c3, c4, c5, c6, c7, c8, tableoid
-> Sort
Output: c1, c2, c3, c4, c5, c6, c7, c8, tableoid
Sort Key: t1.c3, t1.c1, t1.tableoid
-> Foreign Scan on public.ft1 t1
Output: c1, c2, c3, c4, c5, c6, c7, c8, tableoid
Remote query: SELECT `C 1`, `c2`, `c3`, `c4`, `c5`, `c6`, `c7`, `c8` FROM `mysql_fdw_post`.`T 1`
(8 rows)
**Describe the bug** Failed to bind the MySQL query when foreign scan will select NULL from an inner join. **To Reproduce** 1. On MySQL server DROP DATABASE IF EXISTS join_limit; CREATE DATABASE join_limit; USE join_limit; CREATE TABLE J1_TBL ( i integer, j integer, t text); CREATE TABLE J2_TBL ( i integer, k integer); CREATE TABLE J3_TBL ( i integer, t text); INSERT INTO J1_TBL VALUES (1, 4, 'one'); INSERT INTO J1_TBL VALUES (2, 3, 'two'); INSERT INTO J1_TBL VALUES (3, 2, 'three'); INSERT INTO J1_TBL VALUES (4, 1, 'four'); INSERT INTO J1_TBL VALUES (5, 0, 'five'); INSERT INTO J1_TBL VALUES (6, 6, 'six'); INSERT INTO J1_TBL VALUES (7, 7, 'seven'); INSERT INTO J1_TBL VALUES (8, 8, 'eight'); INSERT INTO J1_TBL VALUES (0, NULL, 'zero'); INSERT INTO J1_TBL VALUES (NULL, NULL, 'null'); INSERT INTO J1_TBL VALUES (NULL, 0, 'zero'); INSERT INTO J2_TBL VALUES (1, -1); INSERT INTO J2_TBL VALUES (2, 2); INSERT INTO J2_TBL VALUES (3, -3); INSERT INTO J2_TBL VALUES (2, 4); INSERT INTO J2_TBL VALUES (5, -5); INSERT INTO J2_TBL VALUES (5, -5); INSERT INTO J2_TBL VALUES (0, NULL); INSERT INTO J2_TBL VALUES (NULL, NULL); INSERT INTO J2_TBL VALUES (NULL, 0); INSERT INTO J3_TBL VALUES (1, 'first'); INSERT INTO J3_TBL VALUES (2, NULL); INSERT INTO J3_TBL VALUES (2, 'second'); INSERT INTO J3_TBL VALUES (3, 'third'); INSERT INTO J3_TBL VALUES (4, 'forth'); INSERT INTO J3_TBL VALUES (5, 'five'); 2. Execute SQL CREATE FOREIGN TABLE J1_TBL__mysql_svr__0 ( i integer, j integer, t text) SERVER mysql_svr OPTIONS (dbname 'join_limit', table_name 'J1_TBL'); CREATE FOREIGN TABLE J2_TBL__mysql_svr__0 ( i integer, k integer) SERVER mysql_svr OPTIONS (dbname 'join_limit', table_name 'J2_TBL'); CREATE FOREIGN TABLE J3_TBL__mysql_svr__0 ( i integer, t text) SERVER mysql_svr OPTIONS (dbname 'join_limit', table_name 'J3_TBL'); SELECT * FROM j2_tbl__mysql_svr__0 where exists (select t3.i, t from j3_tbl__mysql_svr__0 t3 join j2_tbl__mysql_svr__0 t2 on t2.i = t3.i); 3. Result before fixing SELECT * FROM j2_tbl__mysql_svr__0 where exists (select t3.i, t from j3_tbl__mysql_svr__0 t3 join j2_tbl__mysql_svr__0 t2 on t2.i = t3.i); ERROR: failed to bind the MySQL query **Result after fixing** i | k ---+---- 1 | -1 2 | 2 3 | -3 2 | 4 5 | -5 5 | -5 0 | | | 0 (9 rows) **Additional context** Because with tupleDescriptor->natts = 0 is used to palloc for table->mysql_bind. But it will return invalid memory like address: 0x7f7f7f7f7f7f7f7f. With this allocated memory, the binding will fail and it causes error "failed to bind the MySQL query". When natts is 0, it means the PostgreSQL/PGSpider core does not need any result in tuple slot. Therefore, no need to allocate memory for festate->table in this case.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Support Subscripting Array syntax
Please check describe in commmit:
(MinhLA1410@1152ff0)
Not pushdown system column instead of raise error
Please check describe in commmit:
(MinhLA1410@e4cc9ff)
Do not allocate when natts is 0
Please check describe in commmit:
MinhLA1410@4d42bba