Re: Oddity in EXPLAIN for foreign/custom join pushdown plans
От | Robert Haas |
---|---|
Тема | Re: Oddity in EXPLAIN for foreign/custom join pushdown plans |
Дата | |
Msg-id | CA+TgmoaKRcbBA16bjTxJfL7fRZb3=qEjstiQ8cd6weoFWppf8A@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: Oddity in EXPLAIN for foreign/custom join pushdown plans (Etsuro Fujita <fujita.etsuro@lab.ntt.co.jp>) |
Ответы |
Re: Oddity in EXPLAIN for foreign/custom join pushdown
plans
|
Список | pgsql-hackers |
On Tue, Aug 23, 2016 at 11:18 PM, Etsuro Fujita <fujita.etsuro@lab.ntt.co.jp> wrote: >> Yes, it seems what we are doing now is not consistent with what >> happens for plain tables; that should probably be fixed. > > OK, I think we should fix the issue that postgres_fdw produces incorrect > aliases for joining relations shown in the Relations line in EXPLAIN for a > join pushdown query like the above) in advance of the 9.6 release, so I'll > add this to the 9.6 open items. Isn't it a bit late for that? I'm not eager to have 9.6 get further delayed while we work on this issue, and I definitely don't believe that this is a sufficiently important issue to justify reverting join pushdown in its entirety. We're talking about a minor detail of the EXPLAIN output that we'd like to fix, but for which we have no consensus on exactly how to fix it, and the fix may involve some redesign. That does not seem like a good thing to the week before rc1. Let's just leave this well enough alone and work on it for v10. -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise PostgreSQL Company
В списке pgsql-hackers по дате отправления: