http://archives.postgresql.org/message-id/[EMAIL PROTECTED]
1) Sometimes index doesn't find all matching rows:
postgres=# SELECT * FROM qq WHERE t ='asd';
i | t
---+-----
2 | asd
1 | asd
2 | asd
(3 rows)
postgres=# SET enable_seqscan=off;
SET
postgres=# SELECT * FROM qq WHERE t ='asd';
i | t
---+-----
2 | asd
(1 row)
How to reproduce:
DROP TABLE IF EXISTS qq;
CREATE TABLE qq ( i int, t text );
INSERT INTO qq VALUES (1, 'qwe');
INSERT INTO qq VALUES (2, 'asd');
CREATE INDEX qqidx ON qq USING bitmap (i,t);
INSERT INTO qq VALUES (1, 'asd');
INSERT INTO qq VALUES (2, 'asd');
SELECT * FROM qq;
SELECT * FROM qq WHERE t ='asd';
SET enable_seqscan=off;
SELECT * FROM qq WHERE t ='asd';
2) Why is pg_am.amstrategies set to 5 while index supports only equal operation?
3) Typo in bmbulkdelete:
/* allocate stats if first time through, else re-use existing struct */
if (result == NULL)
result = (IndexBulkDeleteResult *)
palloc0(sizeof(IndexBulkDeleteResult));
result = (IndexBulkDeleteResult *) palloc0(sizeof(IndexBulkDeleteResult));
'result' is allocated twice.
4) Bitmap index is marked with pg_am.amcanorder = 'f', so you don't need to
support ammarkpos/amrestrpos - see
http://archives.postgresql.org/pgsql-hackers/2008-10/msg00862.php
--
Teodor Sigaev E-mail: [EMAIL PROTECTED]
WWW: http://www.sigaev.ru/
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
