X-Original-To: pgsql-sql-postgresql.org@localhost.postgresql.org
Received: from localhost (av.hub.org [200.46.204.144])
	by postgresql.org (Postfix) with ESMTP id 4411A9DC9E6
	for <pgsql-sql-postgresql.org@localhost.postgresql.org>;
	Tue, 27 Dec 2005 20:25:41 -0400 (AST)
Received: from postgresql.org ([200.46.204.71])
	by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024)
	with ESMTP id 23750-03
	for <pgsql-sql-postgresql.org@localhost.postgresql.org>;
	Tue, 27 Dec 2005 20:25:43 -0400 (AST)
X-Greylist: from auto-whitelisted by SQLgrey-
Received: from sss.pgh.pa.us (sss.pgh.pa.us [66.207.139.130])
	by postgresql.org (Postfix) with ESMTP id 152239DC891
	for <pgsql-sql@postgresql.org>; Tue, 27 Dec 2005 20:25:37 -0400 (AST)
Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1])
	by sss.pgh.pa.us (8.13.1/8.13.1) with ESMTP id jBS0PeWW009417;
	Tue, 27 Dec 2005 19:25:40 -0500 (EST)
To: elein <elein@varlena.com>
cc: George Pavlov <gpavlov@mynewplace.com>, pgsql-sql@postgresql.org
Subject: Re: "large" IN/NOT IN subqueries result in query returning wrong data
In-reply-to: <20051228001251.GB20674@varlena.com> 
References: 
 <CCB89282FCE1024EA3DCE687A96A5164039EBD13@ehost010-6.exch010.intermedia.net>
	<3697.1135712691@sss.pgh.pa.us>
	<20051228001251.GB20674@varlena.com>
Comments: In-reply-to elein <elein@varlena.com>
	message dated "Tue, 27 Dec 2005 16:12:51 -0800"
Date: Tue, 27 Dec 2005 19:25:40 -0500
Message-ID: <9416.1135729540@sss.pgh.pa.us>
From: Tom Lane <tgl@sss.pgh.pa.us>
X-Virus-Scanned: by amavisd-new at hub.org
X-Spam-Status: No, score=0.049 required=5 tests=[AWL=0.049]
X-Spam-Score: 0.049
X-Spam-Level: 
X-Archive-Number: 200512/215
X-Sequence-Number: 23757

elein <elein@varlena.com> writes:
> In 8.0 we get:

>    elein=# select 1 in (NULL, 1, 2);
>     ?column?
>    ----------
>     t
>    (1 row)
   
>    elein=# select 3 not in (NULL, 1, 2);
>     ?column?
>    ----------
   
>    (1 row)
 
> For consistency, either both should return NULL or
> both return true/false.

The above behavior is correct per spec.  Feel free to argue its
consistency with the SQL committee ;-)

Note that the above are not inverses because you changed the lefthand
input.  You do get consistent results when you just add or omit NOT:

regression=# select 1 not in (NULL, 1, 2);
 ?column?
----------
 f
(1 row)

regression=# select 3 in (NULL, 1, 2);
 ?column?
----------

(1 row)

			regards, tom lane