mirror/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2026-05-07 23:54:31 +02:00
Code Issues Projects Releases Packages Wiki Activity

BUG#18233 - Memory tables INDEX USING HASH (a,b) returns 1 row on

Browse source 
SELECT WHERE a= AND b=

Selecting data from memory table with varchar column and hash index over it
returns only first row matched.

Problem was that key length calculation for varchar columns didn't include
number of bytes to store length.

Fixed key length for varchar fields to include number of bytes to store length.


heap/hp_create.c:
  Fixed key length for varchar fields to include number of bytes to store length.
mysql-test/r/heap.result:
  Testcase for BUG#18233.
mysql-test/t/heap.test:
  Testcase for BUG#18233.
This commit is contained in:
unknown 2006-05-10 21:50:04 +05:00
commit 2eeab0b0d8
3 changed files with 27 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

12
mysql-test/t/heap.test
View file

@ -458,4 +458,16 @@ create table t2 (c varchar(10)) engine=memory;
show table status like 't_';
drop table t1, t2;
#
# BUG#18233 - Memory tables INDEX USING HASH (a,b) returns 1 row on
# SELECT WHERE a= AND b=
#
CREATE TABLE t1(a VARCHAR(1), b VARCHAR(2), c VARCHAR(256),
KEY(a), KEY(b), KEY(c)) ENGINE=MEMORY;
INSERT INTO t1 VALUES('a','aa',REPEAT('a', 256)),('a','aa',REPEAT('a',256));
SELECT COUNT(*) FROM t1 WHERE a='a';
SELECT COUNT(*) FROM t1 WHERE b='aa';
SELECT COUNT(*) FROM t1 WHERE c=REPEAT('a',256);
DROP TABLE t1;
# End of 5.0 tests