To: Louis-David Mitterrand <vindex+lists-pgsql-sql@apartia.org>
cc: pgsql-sql@postgresql.org
Subject: Re: Does IMMUTABLE property propagate? 
In-reply-to: <20100325075405.GA17176@apartia.fr> 
References: <4B91C184.4090906@venaver.info>
	<407d949e1003060645i408e0648te67dfd722172f76a@mail.gmail.com>
	<4B927FED.60209@venaver.info> <4150.1267911115@sss.pgh.pa.us>
	<20100325061616.GA4629@apartia.fr>
	<162867791003250027q3cfc5ba9g59d56768aa4c4e5e@mail.gmail.com>
	<20100325075405.GA17176@apartia.fr>
Comments: In-reply-to Louis-David Mitterrand
	<vindex+lists-pgsql-sql@apartia.org>
	message dated "Thu, 25 Mar 2010 08:54:05 +0100"
Date: Thu, 25 Mar 2010 10:35:26 -0400
Message-ID: <16800.1269527726@sss.pgh.pa.us>
From: Tom Lane <tgl@sss.pgh.pa.us>

Louis-David Mitterrand <vindex+lists-pgsql-sql@apartia.org> writes:
> Hmm, that's interesting. So for simple functions (like my example) it is
> better to write them in plain sql? And in that case no 'immutable' flag
> is necessary?

If it's just a simple SQL expression, then yes write it as a SQL
function.  The planner can "inline" those, eliminating the call overhead
that you'll pay with a plpgsql function.

When you're intending to have a SQL function be inlined, it's probably
best not to mark it as either IMMUTABLE or STRICT --- that doesn't buy
anything and it can complicate matters as to whether inlining is legal.

			regards, tom lane