Il 25/12/2009 18:13, Tom Lane ha scritto:
I wrote:
I guess we missed something about when it's safe to do this optimization.
I've applied the attached patch to fix this.
Thanks. Everything's working fine now!
Merry Xmas
--
Matteo Beccati
Development & Consulting - http://www.beccati.com/
I wrote:
> I guess we missed something about when it's safe to do this optimization.
I've applied the attached patch to fix this.
regards, tom lane
Index: joinpath.c
===
RCS file: /cvsroot/pgsql/src/backend/o
"Matteo Beccati" writes:
> With the following data set a SELECT * query returns 1 rows, while SELECT
> COUNT(*) returns 2.
Hm, looks like the join-elimination patch is firing mistakenly. It's not
so much the count(*) that does it as the lack of any select-list
references to a:
regression=# expl
The following bug has been logged online:
Bug reference: 5255
Logged by: Matteo Beccati
Email address: p...@beccati.com
PostgreSQL version: 8.5alpha3
Operating system: NetBSD 5.0.1
Description:COUNT(*) returns wrong result with LEFT JOIN
Details:
Discovered this wh
