X-Original-To: pgsql-sql-postgresql.org@localhost.postgresql.org
Received: from localhost (unknown [200.46.204.144])
	by svr1.postgresql.org (Postfix) with ESMTP id 817C5528D9
	for <pgsql-sql-postgresql.org@localhost.postgresql.org>;
	Wed, 22 Jun 2005 16:49:28 -0300 (ADT)
Received: from svr1.postgresql.org ([200.46.204.71])
	by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024)
	with ESMTP id 99486-04
	for <pgsql-sql-postgresql.org@localhost.postgresql.org>;
	Wed, 22 Jun 2005 19:49:21 +0000 (GMT)
Received: from sss.pgh.pa.us (sss.pgh.pa.us [66.207.139.130])
	by svr1.postgresql.org (Postfix) with ESMTP id 7F5FD528BB
	for <pgsql-sql@postgresql.org>; Wed, 22 Jun 2005 16:49:20 -0300 (ADT)
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 j5MJkf53020667;
	Wed, 22 Jun 2005 15:46:44 -0400 (EDT)
To: Stephan Szabo <sszabo@megazone.bigpanda.com>
Cc: Russell Simpkins <russellsimpkins@hotmail.com>,
	pgsql-sql@postgresql.org
Subject: Re: after delete trigger behavior 
In-reply-to: <20050622122313.F42989@megazone.bigpanda.com> 
References: <BAY103-F108CECCE4CF4F95511684B5EB0@phx.gbl>
	<20050622092239.T32802@megazone.bigpanda.com>
	<20364.1119467550@sss.pgh.pa.us>
	<20050622122313.F42989@megazone.bigpanda.com>
Comments: In-reply-to Stephan Szabo <sszabo@megazone.bigpanda.com>
	message dated "Wed, 22 Jun 2005 12:27:20 -0700"
Date: Wed, 22 Jun 2005 15:46:41 -0400
Message-ID: <20666.1119469601@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, hits=0.006 tagged_above=0 required=5 tests=AWL
X-Spam-Level: 
X-Archive-Number: 200506/278
X-Sequence-Number: 21879

Stephan Szabo <sszabo@megazone.bigpanda.com> writes:
> Is there anything we have right now that will handle this kind of thing
> without requiring either updating all the counts after a deletion in a
> statement trigger or once per row updating all the counts for records with
> the same "a" (doing something like make a sequence and using it in a
> subselect matching keys)?

The best thing I can think of is your first idea, ie, renumbering all
the rows in a statement-level AFTER DELETE trigger.  Something like
(untested)

	DECLARE
		rec record;
		n integer := 1;
	BEGIN
		FOR rec IN
			SELECT * FROM table
			WHERE <<grouping cols = rec's grouping cols>>
			ORDER BY sort_order
		LOOP
			IF rec.sort_order != n THEN
				UPDATE table SET sort_order = n
				WHERE <<primary key = rec's primary key>>;
			END IF;
			n := n + 1;
		END LOOP;
	END;

Ugly as this is, it's at least linear in the number of rows to be
changed; the originally proposed trigger was O(N^2) in the number of
rows affected, and would surely be intolerably slow for multiple deletes
in a reasonably sized table.  Given an index on the grouping columns
plus sort_order, it could even be reasonably fast (don't forget to make
the ORDER BY match the index).

			regards, tom lane