| From: | "Fip" <Filippo(dot)Bernante(at)gmail(dot)com> |
|---|---|
| To: | pgsql-general(at)postgresql(dot)org |
| Subject: | grouping excluding some text results |
| Date: | 2006-10-26 07:25:58 |
| Message-ID: | 1161847558.199150.294690@i3g2000cwc.googlegroups.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Hi,
ho can I select some one restult for type, excluding the rest?
I have something like, as result of a join:
| ID | score | description |
-----------------------------------------------------------------
MG01 56 "textual description1..."
MG02 47 "another text ..."
MG02 55 "textual description, text...." <----
note this
MG02 55 "textual description, text...." <----
note this
MG01 35 "this is a different text...."
MG02 61 "random chars..."
(...)
I know that is possible selecting by grouping if I use an aggregate
function:
> select ID,max(score) by table group by ID
but I want the description too, only one description, and I have a very
large dataset, I cannot cycle with more subselections, this require too
much time; also I want to exclude duplicates: only one ID.
What can I do?
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Stuart Grimshaw | 2006-10-26 08:17:26 | Re: Enableing contrib modules on Debian |
| Previous Message | Chris | 2006-10-26 06:20:25 | Re: Reducing pg_dump & pg_restore times |