Re: [PERFORM] Workaround for cross column stats dependency

2008-01-23 Thread Guillaume Smet
On Jan 23, 2008 3:02 AM, Guillaume Smet <[EMAIL PROTECTED]> wrote: > I'll post my results tomorrow morning. It works perfectly well: cityvox_prod=# CREATE OR REPLACE FUNCTION getTypesLieuFromTheme(codeTheme text) returns text[] AS $f$ SELECT ARRAY(SELECT codetylieu::text FROM rubtylieu WHERE codet

Re: [PERFORM] Workaround for cross column stats dependency

2008-01-22 Thread Guillaume Smet
On Jan 23, 2008 2:43 AM, Tom Lane <[EMAIL PROTECTED]> wrote: > regression=# create or replace function getu2(int) returns int[] as $$ > select array(select unique2 from tenk1 where thousand = $1); > $$ language sql immutable; > CREATE FUNCTION > regression=# explain select * from tenk1 where unique

Re: [PERFORM] Workaround for cross column stats dependency

2008-01-22 Thread Tom Lane
"Guillaume Smet" <[EMAIL PROTECTED]> writes: > So the question is: is there any way to improve the results of the > original query, other than doing a first query in the application to > get the list of types and inject them in a second query (the one just > above)? Well, if you're willing to chea

[PERFORM] Workaround for cross column stats dependency

2008-01-22 Thread Guillaume Smet
Hi -performance, While testing 8.3, I found this query which is equally slow on 8.1 and 8.3 and seems to be really slow for a not so complex query. The stats are as good as possible and the behaviour of PostgreSQL seems to be logical considering the stats but I'm looking for a workaround to speed