From: | "Dorian Taylor" <leg0(at)hotmail(dot)com> |
---|---|
To: | pgsql-sql(at)postgresql(dot)org |
Subject: | case statement in group by clause |
Date: | 2002-04-15 10:44:05 |
Message-ID: | F264UVvjMa7YflPcTHL00005ea1@hotmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-sql |
Greetings,
I can't remember if I asked the list this before (it's been a very hectic
month), if so, please disregard. I'm trying to find the most elegant way of
integrating SQL into an application I am writing which has a potential for
N^3*M fairly similar queries (Luckily enough N is 3 and M is 6, but could
easily and very soon become more). Rather than explicitly write each of
these out, or concoct some sort of illegible mutant
string-concatenation/bind parameter frankenstein, I would like to simply
create one massive query riddled with case statements triggered by named
bind parameters. What is holding me back from this is the inability to
coerce certain column types in a case statement in a GROUP BY clause. Now,
the obvious solution would be to flatten each test out to return a value of
the same type, although I'm not too sure how to access the original sorting
function/operator that is associated with a GROUP BY clause. I'm sure this
isn't a new problem, but I've run into great difficulty finding anything on
the subject. Please advise.
Regards,
Dorian
_________________________________________________________________
Send and receive Hotmail on your mobile device: http://mobile.msn.com
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2002-04-15 14:35:25 | Re: case statement in group by clause |
Previous Message | jack | 2002-04-15 06:40:03 | Re: data order problem |