projects / users / c2main / postgres.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: b5d1608) | patch
Go back to using a separate method for doing ILIKE for single byte
authorAndrew Dunstan <andrew@dunslane.net>
Sat, 22 Sep 2007 03:58:34 +0000 (03:58 +0000)
committerAndrew Dunstan <andrew@dunslane.net>
Sat, 22 Sep 2007 03:58:34 +0000 (03:58 +0000)
commite1528933058db8db8e7398c044b1e9fee82cddcf
tree98b77b2335a13ad314fcafb9a893c4a394f08275tree
parentb5d1608b0ae32e47ddfaf4c5e41afc7004171075commit | diff
Go back to using a separate method for doing ILIKE for single byte
character encodings that doesn't involve calling lower(). This should
cure the performance regression in this case complained of by Guillaume
Smet. It still leaves the horrid performance for multi-byte encodings
introduced in 8.2, but there's no obvious solution for that in sight.
src/backend/utils/adt/like.c diff | blob | blame | history
src/backend/utils/adt/like_match.c diff | blob | blame | history
Cedric's Stuff