On 10/13/12 3:15 PM, Jim Nasby wrote:
FWIW, it's definitely an issue of not being able to push down past the GROUP BY:
I take that back... GROUP BY doesn't matter. It's an issue of having the EXISTS
in the inner query. I realize the examples have gotten a bit silly, but this
seems to break it
Jim Nasby writes:
> FWIW, it's definitely an issue of not being able to push down past the GROUP
> BY:
I think it's not that so much as the EXISTS inside a LEFT JOIN.
regards, tom lane
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make chang
On 10/13/12 2:45 PM, Tom Lane wrote:
Jim Nasby writes:
Just upgraded to 8.4 (I know, I know…) and ran across this. Unfortunately I
have no way to test this on 9.x, so I don't know if it's been fixed or not. I'm
hoping that someone *cough*Tom*cough* would quickly recognize whether this push
i
On 10/13/12 2:45 PM, Tom Lane wrote:
BTW, your workaround looks wrong --- you need to constrain the outside
of the left join not the inside, no?
Ugh, yes, you're correct. :(
--
Jim C. Nasby, Database Architect j...@nasby.net
512.569.9461 (cell) http://j
Jim Nasby writes:
> Just upgraded to 8.4 (I know, I know
) and ran across this. Unfortunately I
> have no way to test this on 9.x, so I don't know if it's been fixed or not.
> I'm hoping that someone *cough*Tom*cough* would quickly recognize whether
> this push into subquery issue has been fixe
Just upgraded to 8.4 (I know, I know…) and ran across this. Unfortunately I
have no way to test this on 9.x, so I don't know if it's been fixed or not. I'm
hoping that someone *cough*Tom*cough* would quickly recognize whether this push
into subquery issue has been fixed or not, so I haven't incl
