Re: BRIN range operator class
От | Alvaro Herrera |
---|---|
Тема | Re: BRIN range operator class |
Дата | |
Msg-id | 20150513230803.GV2523@alvh.no-ip.org обсуждение исходный текст |
Ответ на | Re: BRIN range operator class (Emre Hasegeli <emre@hasegeli.com>) |
Список | pgsql-hackers |
Emre Hasegeli wrote: > > I pushed patches 04 and 07, as well as adopting some of the changes to > > the regression test in 06. I'm afraid I caused a bit of merge pain for > > you -- sorry about that. > > No problem. I rebased the remaining ones. Thanks! After some back-and-forth between Emre and me, here's an updated patch. My changes are cosmetic; for a detailed rundown, see https://github.com/alvherre/postgres/commits/brin-inclusion Note that datatype point was removed: it turns out that unless we get box_contain_pt changed to use FPlt() et al, indexes created with this opclass would be corrupt. And we cannot simply change box_contain_pt, because that would break existing GiST and SP-GiST indexes that use it today and pg_upgrade to 9.5! So that needs to be considered separately. Also, removing point support means remove the CAST support procedure, because there is no use for it in the supported types. Also, patch 05 in the previous submissions goes away completely because there's no need for those (box,point) operators anymore. There's nothing Earth-shattering here that hasn't been seen in previous submissions by Emre. One item of note is that this patch is blindly removing the assert-only blocks as previously discussed, without any replacement. Need to think more on how to put something back ... -- Álvaro Herrera http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
Вложения
В списке pgsql-hackers по дате отправления: