Message-ID: <43189B0F.2000904@beccati.com>
Date: Fri, 02 Sep 2005 20:33:51 +0200
From: Matteo Beccati <php@beccati.com>
User-Agent: Mozilla Thunderbird 1.0.2 (Windows/20050317)
MIME-Version: 1.0
To: Simon Riggs <simon@2ndquadrant.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: ORDER BY and LIMIT not propagated on inherited
References: <43182789.4070307@beccati.com>
	<1125683697.3956.310.camel@localhost.localdomain>
In-Reply-To: <1125683697.3956.310.camel@localhost.localdomain>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Simon Riggs wrote:
> Returning to Matteo's original query, what we are saying is that the new
> optimization for MIN/MAX queries doesn't work with inherited tables. 
> 
> It could do, by running optimize_minmax_aggregates() for each query that
> gets planned to see if a better plan exists for each child table.
> 
> I think that's a TODO item.

Great. Of course I'm using ORDER BY ... LIMIT as a workaround to get the 
index scan on pre-8.1, and because I'm used to it insted of the 
previously not optimized MIN/MAX aggregates.

> Optimizing ORDER BY and LIMIT down looks like it would be harder to do
> in the general case, even if Matteo's simple transform looks good. I'm
> not sure it's a very common query type though...

If I can find some time, I'll try to write some hacks... I just need to 
find out where to start ;)


Best regards
--
Matteo Beccati
http://phpadsnew.com/
http://phppgads.com/