database/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2025-07-27 18:02:13 +03:00
Code Activity

MDEV-11563: GROUP_CONCAT(DISTINCT ...) may produce a non-distinct list

Browse Source 
Backported from MYSQL
Bug #25331425: DISTINCT CLAUSE DOES NOT WORK IN GROUP_CONCAT
Issue:
------
The problem occurs when:
1) GROUP_CONCAT (DISTINCT ....) is used in the query.
2) Data size greater than value of system variable:
tmp_table_size.

The result would contain values that are non-unique.

Root cause:
-----------
An in-memory structure is used to filter out non-unique
values. When the data size exceeds tmp_table_size, the
overflow is written to disk as a separate file. The
expectation here is that when all such files are merged,
the full set of unique values can be obtained.

But the Item_func_group_concat::add function is in a bit of
hurry. Even as it is adding values to the tree, it wants to
decide if a value is unique and write it to the result
buffer. This works fine if the configured maximum size is
greater than the size of the data. But since tmp_table_size
is set to a low value, the size of the tree is smaller and
hence requires the creation of multiple copies on disk.

Item_func_group_concat currently has no mechanism to merge
all the copies on disk and then generate the result. This
results in duplicate values.

Solution:
---------
In case of the DISTINCT clause, don't write to the result
buffer immediately. Do the merge and only then put the
unique values in the result buffer. This has be done in
Item_func_group_concat::val_str.

Note regarding result file changes:
-----------------------------------
Earlier when a unique value was seen in
Item_func_group_concat::add, it was dumped to the output.
So result is in the order stored in SE. But with this fix,
we wait until all the data is read and the final set of
unique values are written to output buffer. So the data
appears in the sorted order.

This only fixes the cases when we have DISTINCT without ORDER BY clause
in GROUP_CONCAT.
This commit is contained in:
Varun Gupta
2020-03-21 12:23:44 +05:30
parent fd1755e49d
commit a006e88cac
4 changed files with 65 additions and 38 deletions
Download Patch File Download Diff File Expand all files Collapse all files

53
mysql-test/main/func_gconcat.result
View File

@ -363,8 +363,8 @@ bb,ccc,a,bb,ccc
BB,CCC,A,BB,CCC
select group_concat(distinct b) from t1 group by a;
group_concat(distinct b)
bb,ccc,a
BB,CCC,A
a,bb,ccc
A,BB,CCC
select group_concat(b order by b) from t1 group by a;
group_concat(b order by b)
a,bb,bb,ccc,ccc
@ -383,11 +383,11 @@ Warning 1260 Row 2 was cut by GROUP_CONCAT()
Warning 1260 Row 4 was cut by GROUP_CONCAT()
select group_concat(distinct b) from t1 group by a;
group_concat(distinct b)
bb,c
BB,C
a,bb
A,BB
Warnings:
Warning 1260 Row 2 was cut by GROUP_CONCAT()
Warning 1260 Row 4 was cut by GROUP_CONCAT()
Warning 1260 Row 3 was cut by GROUP_CONCAT()
Warning 1260 Row 6 was cut by GROUP_CONCAT()
select group_concat(b order by b) from t1 group by a;
group_concat(b order by b)
a,bb
@ -413,8 +413,8 @@ bb,ccc,a,bb,ccc,1111111111111111111111111111111111111111111111111111111111111111
BB,CCC,A,BB,CCC,1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111112,1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111112,0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001
select group_concat(distinct b) from t1 group by a;
group_concat(distinct b)
bb,ccc,a,1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111112,0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001
BB,CCC,A,1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111112,0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001
0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001,1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111112,a,bb,ccc
0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001,1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111112,A,BB,CCC
select group_concat(b order by b) from t1 group by a;
group_concat(b order by b)
0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001,1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111112,1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111112,a,bb,bb,ccc,ccc
@ -433,11 +433,11 @@ Warning 1260 Row 7 was cut by GROUP_CONCAT()
Warning 1260 Row 14 was cut by GROUP_CONCAT()
select group_concat(distinct b) from t1 group by a;
group_concat(distinct b)
bb,ccc,a,1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111112,00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
BB,CCC,A,1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111112,00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001,11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111
0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001,11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111
Warnings:
Warning 1260 Row 5 was cut by GROUP_CONCAT()
Warning 1260 Row 10 was cut by GROUP_CONCAT()
Warning 1260 Row 2 was cut by GROUP_CONCAT()
Warning 1260 Row 4 was cut by GROUP_CONCAT()
select group_concat(b order by b) from t1 group by a;
group_concat(b order by b)
0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001,11111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111
@ -520,9 +520,9 @@ a group_concat(b)
NULL 3,4,2,1,2,7,3,3
select a, group_concat(distinct b) from t1 group by a with rollup;
a group_concat(distinct b)
1 3,4,2,1
2 7,3
NULL 3,4,2,1,7
1 1,2,3,4
2 3,7
NULL 1,2,3,4,7
select a, group_concat(b order by b) from t1 group by a with rollup;
a group_concat(b order by b)
1 1,2,2,3,4
@ -745,10 +745,10 @@ CREATE TABLE t1(a TEXT, b CHAR(20));
INSERT INTO t1 VALUES ("one.1","one.1"),("two.2","two.2"),("one.3","one.3");
SELECT GROUP_CONCAT(DISTINCT UCASE(a)) FROM t1;
GROUP_CONCAT(DISTINCT UCASE(a))
ONE.1,TWO.2,ONE.3
ONE.1,ONE.3,TWO.2
SELECT GROUP_CONCAT(DISTINCT UCASE(b)) FROM t1;
GROUP_CONCAT(DISTINCT UCASE(b))
ONE.1,TWO.2,ONE.3
ONE.1,ONE.3,TWO.2
DROP TABLE t1;
CREATE TABLE t1( a VARCHAR( 10 ), b INT );
INSERT INTO t1 VALUES ( repeat( 'a', 10 ), 1),
@ -847,7 +847,7 @@ create table t1(a bit(2) not null);
insert into t1 values (1), (0), (0), (3), (1);
select group_concat(distinct a) from t1;
group_concat(distinct a)
1,0,3
0,1,3
select group_concat(distinct a order by a) from t1;
group_concat(distinct a order by a)
0,1,3
@ -860,13 +860,13 @@ insert into t1 values (1, 'a', 0), (0, 'b', 1), (0, 'c', 0), (3, 'd', 1),
(1, 'e', 1), (3, 'f', 1), (0, 'g', 1);
select group_concat(distinct a, c) from t1;
group_concat(distinct a, c)
10,01,00,31,11
00,01,10,11,31
select group_concat(distinct a, c order by a) from t1;
group_concat(distinct a, c order by a)
00,01,11,10,31
select group_concat(distinct a, c) from t1;
group_concat(distinct a, c)
10,01,00,31,11
00,01,10,11,31
select group_concat(distinct a, c order by a, c) from t1;
group_concat(distinct a, c order by a, c)
00,01,10,11,31
@ -1333,8 +1333,8 @@ select grp,group_concat(c limit 5.5...' at line 1
select grp,group_concat(distinct c limit 1,10 ) from t1 group by grp;
grp group_concat(distinct c limit 1,10 )
1 c
2 b
3 C,D
2 c
3 D,E
select grp,group_concat(c order by a) from t1 group by grp;
grp group_concat(c order by a)
1 b,c
@ -1370,6 +1370,15 @@ grp group_concat(c order by c desc limit 2)
1 c,b
2 c,b
3 E,E
#
# Empty results for group concat as offset is greater than the rows
# for a group
#
select grp,group_concat(distinct c limit 10,1 ) from t1 group by grp;
grp group_concat(distinct c limit 10,1 )
1
2
3
drop table t1;
create table t2 (a int, b varchar(10));
insert into t2 values(1,'a'),(1,'b'),(NULL,'c'),(2,'x'),(2,'y');