Re: PGXS: REGRESS_OPTS=--load-language=plpgsql

Andrew Dunstan <andrew@dunslane.net>

From: Andrew Dunstan <andrew@dunslane.net>

To: Robert Haas <robertmhaas@gmail.com>

Cc: Tom Lane <tgl@sss.pgh.pa.us>, Bruce Momjian <bruce@momjian.us>, "David E. Wheeler" <david@kineticode.com>, Dimitri Fontaine <dfontaine@hi-media.com>, Takahiro Itagaki <itagaki.takahiro@oss.ntt.co.jp>, David Fetter <david@fetter.org>, Euler Taveira de Oliveira <euler@timbira.com>, PG Hackers <pgsql-hackers@postgresql.org>

Date: 2010-02-20T23:18:04Z

Lists: pgsql-hackers

Robert Haas wrote:
>
> IMO, the non-ugly kludges are (1) implement CREATE OR REPLACE LANGUAGE
> and (2) revert the original patch.  Do you want to do one of those
> (which?) or do you have another idea?
>
>
>   

I thought we seemed to be converging on some agreement on CREATE OR 
REPLACE LANGUAGE.

If not, let me add my vote for that.

I also think we need to state explicitly that module authors can not 
expect build files to be version ignorant and always work. Even if we do 
something that handles this particular issue, that is likely to be a 
happy coincidence rather than something that can be expected all the 
time. People need to expect to have to do version-dependent things.

cheers

andrew