Message-ID: <4AE2C8E3.8000906@hsolutions.ch>
Date: Sat, 24 Oct 2009 11:29:07 +0200
From: Denis BUCHER <dbucherml@hsolutions.ch>
Organization: Horus Digital Solutions =?ISO-8859-1?Q?S=E0rl?=
User-Agent: Thunderbird 2.0.0.23 (Windows/20090812)
MIME-Version: 1.0
To: Tom Lane <tgl@sss.pgh.pa.us>
CC: pgsql-sql@postgresql.org
Subject: Re: Problem with return type of function ??? (corrected)
References: <4AE046E4.8010205@hsolutions.ch> <29935.1256222015@sss.pgh.pa.us>
	<4AE07708.7020506@hsolutions.ch> <1268.1256226694@sss.pgh.pa.us>
In-Reply-To: <1268.1256226694@sss.pgh.pa.us>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit

Tom Lane a écrit :
> Denis BUCHER <dbucherml@hsolutions.ch> writes:
>> To do this it will be a little complicated because of table
>> dependencies... And it could bug again at the next DROP COLUMN... Is
>> there a way to change my function (RETURN SETOF part) to specify the
>> column names/types ?
> 
> No, not really.  You could maybe un-drop the columns with some manual
> surgery on pg_attribute, but it doesn't seem like that's going to lead
> to a nice solution.
> 
> If you were really desperate you could try back-porting the patch:
> http://archives.postgresql.org/pgsql-committers/2009-08/msg00068.php
> but I wouldn't want to vouch for its safety, considering it hasn't
> been through a beta test cycle yet.

Finally there was no real dependencies of that table, and I've found out
that the "dependency" was my function ! Using SETOF (table) makes
impossible to DROP the table. Therefore I DROP my function and was able
to follow you advice, and it worked perfectly !

Thanks a lot for your help (as well as Richard's)

Denis