From: | Artur Rataj <arataj(at)iitis(dot)gliwice(dot)pl> |
---|---|
To: | Jean-Luc Lachance <jllachan(at)nsd(dot)ca> |
Cc: | <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: Boolean product of rows in multiple tables |
Date: | 2002-12-30 19:29:08 |
Message-ID: | Pine.LNX.4.33.0212301933210.23226-100000@linux.iitis.gliwice.pl |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On Mon, 30 Dec 2002, Jean-Luc Lachance wrote:
> I think you are looking for UNION.
>
If there would be one component more:
> SELECT id from table1 t1, table2 t2 where t1.string = 'string1' and
> t2.string='string2' and t1.id=t2.id
> UNION
> SELECT id from table1 t1, table1 t2 where t1.string = 'string1' and
> t2.string='string2' and t1.id=t2.id
> UNION
> SELECT id from table2 t1, table2 t2 where t1.string = 'string1' and
> t2.string='string2' and t1.id=t2.id
UNION
SELECT id from table2 t1, table1 t2 where t1.string = 'string1' and
t2.string='string2' and t1.id=t2.id
> ;
>
it would work in the discussed case. The problem is that there can be
several such tables and there can also be several strings, what could give a
relatively large number of such components, and in effect it could be slow.
A table being a sum of the several searched tables could be used with a
single fast `select', but because an arbitrary subset of the available
tables could possibly be searched, a relatively large number of tables being
sums of the tables in the arbitrary subsets would then be needed.
Best regards,
Artur Rataj
From | Date | Subject | |
---|---|---|---|
Next Message | Andy Kriger | 2002-12-30 19:35:28 | lock table question |
Previous Message | Jean-Luc Lachance | 2002-12-30 19:02:59 | Re: Boolean product of rows in multiple tables |