RE: [HACKERS] Bug on complex subselect (was: Bug on complex join)

From: "Jackson, DeJuan" <djackson(at)cpsgroup(dot)com>
To: phd2(at)earthling(dot)net
Cc: PostgreSQL-development <pgsql-hackers(at)postgreSQL(dot)org>
Subject: RE: [HACKERS] Bug on complex subselect (was: Bug on complex join)
Date: 1999-03-11 18:00:32
Message-ID: F10BB1FAF801D111829B0060971D839F70A83E@cpsmail
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

> Hello!
>
> Vadim already gave the idea to use EXISTS. Will try it.
> Thanks to all who replied!
>
> On Wed, 10 Mar 1999, Jackson, DeJuan wrote:
> > Try your query this way:
> > SELECT DISTINCT subsec_id
> > FROM positions p
> > WHERE EXISTS(SELECT 1
> > FROM central c, shops s, districts d
> > WHERE p.pos_id = c.pos_id AND
> > c.shop_id = s.shop_id AND
> > s.distr_id = d.distr_id AND
> > d.city_id = 2);
>
> > Make sure you have indexes on pos_id, shop_id, distr_id,
> and city_id.
>
> All these are primary keys in corresponding tables, and hence have
> UNIQUE indicies. Is it enough?
>
> Oleg.
You should have indexes on both the primary and the referenced table.
(i.e. positions.pos_id and central.pos_id) It gives PostgreSQL more
options on which join methods to use while still having an index to
reference.

-DEJ

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Oleg Broytmann 1999-03-11 18:00:54 RE: [HACKERS] Bug on complex subselect (was: Bug on complex join)
Previous Message Clark Evans 1999-03-11 17:58:35 Parameterized Types and Session Memory