Date: Mon, 29 Jun 2009 10:20:58 +0200
From: "A. Kretschmer" <andreas.kretschmer@schollglas.com>
To: pgsql-sql@postgresql.org
Subject: Re: uniqueness constraint with NULLs
Message-ID: <20090629082058.GG24737@a-kretschmer.de>
References: <4A485431.3080707@cs.anu.edu.au>
	<20090629062602.GB24737@a-kretschmer.de>
	<4A486872.5060906@cs.anu.edu.au>
	<1246260231.7296.2.camel@mars.lemaco.ch>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1246260231.7296.2.camel@mars.lemaco.ch>
User-Agent: Mutt/1.5.9i

In response to Leo Mannhart :
> On Mon, 2009-06-29 at 17:08 +1000, Robert Edwards wrote:
> > A. Kretschmer wrote:
> > > In response to Robert Edwards :
> > >> Can anyone suggest a way that I can impose uniqueness on a and b when
> > >> c is NULL?
> > > 
> > > Sure, use a functional index:
> > > 
> > > test=# create table bobtest (a int, b int, c int);
> > > CREATE TABLE
> > > test=*# create unique index idx_bobtest on
> > > bobtest(a,b,coalesce(c::text,'NULL'));
> > > CREATE INDEX
> > > test=*# insert into bobtest (a, b) values (1, 4);
> > > INSERT 0 1
> > > test=*# insert into bobtest (a, b, c) values (1, 4, NULL);
> > > ERROR:  duplicate key value violates unique constraint "idx_bobtest"
> > > test=!#
> > > 
> Why changing the datatype of coulumn "c"? Will this not probably disable
> the use of the index? Why not only:

Maybe, depends on the query.

> 
> lem=# create table bobtest (a int, b int, c int);
> CREATE TABLE
> lem=# create unique index idc_bobtest on bobtest (a, b, coalesce(c, 1));

bad idea if 1 is a valid value for this column.


Andreas
-- 
Andreas Kretschmer
Kontakt:  Heynitz: 035242/47150,   D1: 0160/7141639 (mehr: -> Header)
GnuPG-ID:   0x3FFF606C, privat 0x7F4584DA   http://wwwkeys.de.pgp.net