To: Michael Fuhr <mike@fuhr.org>
cc: Greg Wittel <gwittel@proofpoint.com>, pgsql-sql@postgresql.org
Subject: Re: Odd PL/PgSQL Error -- relation "X" does not exist when using
	index expression
In-reply-to: <20070208155504.GA78152@winnie.fuhr.org> 
References: <45CA5601.1090007@proofpoint.com>
	<20070208112331.GA75899@winnie.fuhr.org>
	<20070208141433.GD24069@alvh.no-ip.org>
	<20070208150746.GA77230@winnie.fuhr.org>
	<8064.1170948745@sss.pgh.pa.us>
	<20070208155504.GA78152@winnie.fuhr.org>
Comments: In-reply-to Michael Fuhr <mike@fuhr.org>
	message dated "Thu, 08 Feb 2007 08:55:04 -0700"
Date: Thu, 08 Feb 2007 10:59:16 -0500
Message-ID: <8487.1170950356@sss.pgh.pa.us>
From: Tom Lane <tgl@sss.pgh.pa.us>

Michael Fuhr <mike@fuhr.org> writes:
> On Thu, Feb 08, 2007 at 10:32:25AM -0500, Tom Lane wrote:
>> My advice is not to try to execute multiple commands in the same EXECUTE
>> string --- if we were going to do anything to "fix" this, I think it
>> would be along the lines of enforcing that advice.  Trying to make the
>> world safe for it doesn't sound productive.

> The SPI_execute() documentation does mention that multiple commands
> are allowed:

Well, the point here is that there's one pass of parsing and one of
execution, and you won't get far if the parsing pass requires an earlier
command to have already been executed.  So maybe the appropriate warning
is something about not using interdependent DDL commands.  Feel free to
draft up a docs patch.

			regards, tom lane