Message-ID: <457FF0CD.4050606@freesurf.fr>
Date: Wed, 13 Dec 2006 13:23:41 +0100
From: Arnaud Lesauvage <thewild@freesurf.fr>
User-Agent: Thunderbird 1.5.0.8 (Windows/20061025)
MIME-Version: 1.0
To: Jens Schipkowski <jens.schipkowski@apus.co.at>
CC: pgsql-performance@postgresql.org
Subject: Re: Slow update with simple query
References: <457FDB1E.6040801@freesurf.fr>
 <op.tkhy4jpg81rjf6@xjens.apus.local>
In-Reply-To: <op.tkhy4jpg81rjf6@xjens.apus.local>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

Jens Schipkowski a écrit :
> the problem is a combination of bad formed SQL and maybe missing indexes.
> try this:
> UPDATE t1
> SET booleanfield = foo.bar
>  FROM (SELECT uid,(field IN ('some','other') AND field2 = 'Y') AS bar FROM  
> t2) AS foo
> WHERE t1.uid=foo.uid;


Hi Jens,
Why is this query better than the other one ? Because it runs the 
"(field IN ('some','other') AND field2 = 'Y')" once and then executes 
the join with the resulting set ?

> and index t1.uid, t2.uid, t2.field, t2.field2

t1.field can only take 3 or 4 values (don't remember exactly), and 
field2 only 2 ('Y' or 'N'). So this fields have a very low cardinality.
Won't the planner chose to do a table scan in such a case ?

Thanks for your advices !

--
Arnaud