Re: Explain uses incoherent options
От | Guillaume Lelarge |
---|---|
Тема | Re: Explain uses incoherent options |
Дата | |
Msg-id | 1343490575.2011.9.camel@localhost.localdomain обсуждение исходный текст |
Ответ на | Re: Explain uses incoherent options (Vik Reykja <vikreykja@gmail.com>) |
Ответы |
Re: Explain uses incoherent options
|
Список | pgadmin-support |
On Sat, 2012-07-28 at 01:55 +0200, Vik Reykja wrote: > On Fri, Jul 27, 2012 at 2:50 PM, Vik Reykja <vikreykja@gmail.com> wrote: > > > On Fri, Jul 27, 2012 at 2:47 PM, Guillaume Lelarge <guillaume@lelarge.info > > > wrote: > > > >> On Fri, 2012-07-27 at 14:28 +0200, Vik Reykja wrote: > >> > Steps to reproduce: > >> > * start a freshly installed pgAdmin 1.16.0-beta2 and open a query > >> window on > >> > any database > >> > * type the query: "select 1" > >> > * press F7 or choose Explain from the menu > >> > > >> > Expected: the explain graph > >> > Observed: ERROR: EXPLAIN option TIMING requires ANALYZE > >> > > >> > Since timing is on by default, this means the plain F7 is broken out of > >> the > >> > box. It works as expected if I disable timing in the explain options. > >> > >> You're right. I decided to set it up by default, but it has this issue > >> when you don't use the ANALYZE option. I'll need to find another way. > >> > > > > Can't it just not use timing when it's not using analyze? That would seem > > like the saner solution to me. > > > > Sorry about not replying to the list, I didn't do that on purpose. > Oh, no problem. I didn't reply because I didn't had the time to reply. > Anyway, patch attached. It removes the plain explain blocking when buffers > and timing are checked, and instead prefers to ignore those settings when > not doing an analyze. I thought it was a bit silly that a simple explain > was disabled just because you want to see buffers when doing an explain > analyze (and timing wasn't taken into account which is what spawned the > initial bug report). > I don't like this idea to be honest, but I don't see a better way. The patch is correct IMO, so I commited it. > This is my very first hack on pgAdmin, so if I screwed up on style > guidelines or lack of comments or regression tests or what have you, please > forgive me. I'll learn from my mistakes for next time. No, seems pretty good to me :) Thanks for your patch. -- Guillaume http://blog.guillaume.lelarge.info http://www.dalibo.com
В списке pgadmin-support по дате отправления: