Horton

Re: proposal: a validator for configuration files

Tom Lane <tgl@sss.pgh.pa.us>

View thread
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Peter Eisentraut <peter_e@gmx.net>
Cc: Robert Haas <robertmhaas@gmail.com>, Alvaro Herrera <alvherre@commandprompt.com>, Alexey Kluykin <alexk@commandprompt.com>, Florian Pflug <fgp@phlo.org>, Selena Deckelmann <selena@chesnok.com>, Pg Hackers <pgsql-hackers@postgresql.org>
Date: 2011-07-19T15:56:37Z
Lists: pgsql-hackers 
Peter Eisentraut <peter_e@gmx.net> writes:
> On sn, 2011-07-17 at 00:59 -0400, Tom Lane wrote:
>> Well, we *do* have a C API for that, of a sort.  The problem is, what
>> do you do in processes that have not loaded the relevant extension?

> Those processes that have the extension loaded check the parameter
> settings in their namespace, those that don't ignore them.

Then you don't have any meaningful reporting of whether you have entered
valid values --- particularly not with the policy that only the
postmaster makes logfile entries about bad values.  It'd work but I
don't think it's tremendously user-friendly.

			regards, tom lane