1
InnoDB's estimate for the index cardinality depends on a pseudo random
2
number generator (it picks up random pages to sample). After an
3
optimization that was made in r2625 the following EXPLAINs started
4
returning a different number of rows (3 instead of 4).
6
This patch adjusts the result file.
8
This patch cannot be proposed to MySQL because the failures occur only
9
in this tree and do not occur in the standard InnoDB 5.1. Furthermore,
10
the file index_merge2.inc is used by other engines too.
12
--- mysql-test/r/index_merge_innodb.result.orig 2008-09-30 18:32:13.000000000 +0300
13
+++ mysql-test/r/index_merge_innodb.result 2008-09-30 18:33:01.000000000 +0300
15
explain select count(*) from t1 where
16
key1a = 2 and key1b is null and key2a = 2 and key2b is null;
17
id select_type table type possible_keys key key_len ref rows Extra
18
-1 SIMPLE t1 index_merge i1,i2 i1,i2 10,10 NULL 4 Using intersect(i1,i2); Using where; Using index
19
+1 SIMPLE t1 index_merge i1,i2 i1,i2 10,10 NULL 3 Using intersect(i1,i2); Using where; Using index
20
select count(*) from t1 where
21
key1a = 2 and key1b is null and key2a = 2 and key2b is null;
24
explain select count(*) from t1 where
25
key1a = 2 and key1b is null and key3a = 2 and key3b is null;
26
id select_type table type possible_keys key key_len ref rows Extra
27
-1 SIMPLE t1 index_merge i1,i3 i1,i3 10,10 NULL 4 Using intersect(i1,i3); Using where; Using index
28
+1 SIMPLE t1 index_merge i1,i3 i1,i3 10,10 NULL 3 Using intersect(i1,i3); Using where; Using index
29
select count(*) from t1 where
30
key1a = 2 and key1b is null and key3a = 2 and key3b is null;