Received: from localhost (postgresql.org [64.49.215.8])
	by postgresql.org (Postfix) with ESMTP id A2E82476461
	for <pgsql-sql@postgresql.org>; Fri, 22 Nov 2002 16:55:53 -0500 (EST)
Received: from bast.unixathome.org (bast.unixathome.org [66.11.174.150])
	by postgresql.org (Postfix) with ESMTP id 715BF475FC4
	for <pgsql-sql@postgresql.org>; Fri, 22 Nov 2002 16:55:51 -0500 (EST)
Received: from wocker (wocker.unixathome.org [192.168.0.99])
	by bast.unixathome.org (Postfix) with ESMTP
	id D5D923F4B; Fri, 22 Nov 2002 16:55:52 -0500 (EST)
From: "Dan Langille" <dan@langille.org>
To: "Josh Berkus" <josh@agliodbs.com>
Date: Fri, 22 Nov 2002 16:55:52 -0500
MIME-Version: 1.0
Subject: Re: Trees: maintaining pathnames
Cc: pgsql-sql@postgresql.org
Message-ID: <3DDE6198.26130.C4AB8AF0@localhost>
In-reply-to: <3DDBA82F.15041.BA07175D@localhost>
References: <web-1830006@davinci.ethosmedia.com>
X-mailer: Pegasus Mail for Windows (v4.02a)
Content-type: text/plain; charset=US-ASCII
Content-transfer-encoding: 7BIT
Content-description: Mail message body
X-Virus-Scanned: by AMaViS new-20020517
X-Archive-Number: 200211/349
X-Sequence-Number: 10491

On 20 Nov 2002 at 15:20, Dan Langille wrote:

> On 17 Nov 2002 at 14:51, Josh Berkus wrote:
> 
> > Dan,
> > 
> > > My existing tree implementation reflects the files contained on
> > > disk.
> > >  The
> > > full pathname to a particlar file is obtained from the path to the
> > > parent directory.  I am now considering putting this information
> > > into a field in the table.
> > <snip>
> > > Suggestions, comment, open ridicule, most welcome.  thanks.
> > 
> > This is a fine implementation using the adjacency list model of tree
> > design.  However, I think you may find that the string-based tree
> > implementation in /contrib/ltree is more suited to your purposes,
> > and easier to maintain.
> 
> That looks interesting.  I have installed that onto a test server and
> I'm playing around with it.

FWIW, the ltree seems to implement a tree through text manipulation.  
I already have a tree (using a sinble table with id, parent_id).  
Therefore, I think ltree is not an option in this situation.

My creation of the pathname was to save processing time.  I'll talk 
more about that in my next post.
-- 
Dan Langille : http://www.langille.org/