X-Original-To: pgsql-sql@postgresql.org
Received: from news.hub.org (news.hub.org [64.49.215.80])
	by postgresql.org (Postfix) with ESMTP id F408A476082
	for <pgsql-sql@postgresql.org>; Wed,  5 Mar 2003 09:49:35 -0500 (EST)
Received: by news.hub.org (Postfix, from userid 8)
	id B221543A896; Wed,  5 Mar 2003 09:49:34 -0500 (EST)
Message-ID: <3E660A41.7040509@kyborg.dk>
Date: Wed, 05 Mar 2003 15:31:29 +0100
From: Kim Petersen <kp@kyborg.dk>
Organization: Kyborg A/S
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.1) Gecko/20021003
X-Accept-Language: da, en-us
MIME-Version: 1.0
X-Newsgroups: comp.databases.postgresql.sql
Subject: problematic query (for me ;-)
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Lines: 24
X-Complaints-To: abuse@post.tele.dk
To: pgsql-sql@postgresql.org
X-Archive-Number: 200303/295
X-Sequence-Number: 12335

I'm trying to get a hint as to how i can solve this problem:

i have a table:

  n | t
---+---
  1 | 2
  1 | 5
  2 | 3
  2 | 5
  3 | 4
  3 | 3
(6 rows)

now i want to find the pairs (n1,n2) where no t's collide - eg in this 
table it would be (1,3) and (3,1).

Note - this is not homework ;-)

what i'm actually trying to do is to extract n=(countyid,roadid) 
t=housenumber - and find out which roads that *cannot* cross county 
border because their housenumbers collide (to this i've already narrowed 
the n's depending on roadname).