Re: Is it bad sorting in UTF ??
От | Craig Ringer |
---|---|
Тема | Re: Is it bad sorting in UTF ?? |
Дата | |
Msg-id | 49E47D40.5070704@postnewspapers.com.au обсуждение исходный текст |
Ответ на | Re: Is it bad sorting in UTF ?? (Richard Huxton <dev@archonet.com>) |
Список | pgsql-general |
Richard Huxton wrote: > wstrzalka wrote: >> Why PG sort's my data in case insensitive manner? > >> masterdb=# select name, setting from pg_settings WHERE name ilike 'lc >> %'; >> name | setting >> -------------+------------- >> lc_collate | en_US.UTF-8 > > Because that's what en_US.UTF-8 does. If you want "computer-style" > sorting choose the "C" locale. You'll need to dump all your databases an > re-initdb to do that I'm afraid. I believe 8.4 is going to allow > different locales for each database though. Even if it were slower, having a way to specify that sorting be WITH a particular collation method/locale would be appealing. I guess this could be done with a custom operator/opclass implementing non-libc-based sorting and collation, which is probably where PostgreSQL will land up heading in the end. -- Craig Ringer
В списке pgsql-general по дате отправления: