Re: REINDEX filtering in the backend
От | Peter Eisentraut |
---|---|
Тема | Re: REINDEX filtering in the backend |
Дата | |
Msg-id | a81069b1-fdaa-ff40-436e-7840bd639ccf@2ndquadrant.com обсуждение исходный текст |
Ответ на | Re: REINDEX filtering in the backend (Michael Paquier <michael@paquier.xyz>) |
Ответы |
Re: REINDEX filtering in the backend
|
Список | pgsql-hackers |
On 2019-08-30 02:10, Michael Paquier wrote: > On Thu, Aug 29, 2019 at 10:52:55AM +0200, Julien Rouhaud wrote: >> That was already suggested by Thomas and seconded by Peter E., see >> https://www.postgresql.org/message-id/2b1504ac-3d6c-11ec-e1ce-3daf132b3d37%402ndquadrant.com. >> >> I personally think that it's a sensible approach, and I'll be happy to >> propose a patch for that too if no one worked on this yet. > > That may be interesting to sort out first then because we'd likely > want to know what is first in the catalogs before designing the > filtering processing looking at it, no? Right. We should aim to get per-object collation version tracking done. And then we might want to have a REINDEX variant that processes exactly those indexes with an out-of-date version number -- and then updates that version number once the reindexing is done. I think that project is achievable for PG13. I think we can keep this present patch in our back pocket for, say, the last commit fest if we don't make sufficient progress on those other things. Right now, it's hard to review because the target is moving, and it's unclear what guidance to give users. -- Peter Eisentraut http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
В списке pgsql-hackers по дате отправления: