Re: retriving views name
От | Tom Lane |
---|---|
Тема | Re: retriving views name |
Дата | |
Msg-id | 9777.1003244372@sss.pgh.pa.us обсуждение исходный текст |
Ответ на | Re: retriving views name (Alessio Bragadini <alessio@albourne.com>) |
Список | pgsql-general |
Alessio Bragadini <alessio@albourne.com> writes: > Tom Lane wrote: >> One must wonder why you create views at all, if you don't intend them >> to stick around for long. Why not just write out the SELECTs in full? >> (If you think there's some performance benefit to using a view, you are >> quite mistaken.) > There are no cached query plans in any way? Certainly none associated with views. Views would make a very poor basis for caching plans, since they typically form only a part of the eventual query. Example: suppose you write create view v as select * from foo; select * from v where indexedcol = 42; You'd be fairly unhappy if the system were unable to use an indexscan over foo.indexedcol for this query, no? But a plan associated with the bare view definition couldn't exploit this possibility, because the potential to use that index doesn't arise in the context of the view itself, only in the context of the complete query. regards, tom lane
В списке pgsql-general по дате отправления: