Re: Feedback on writing extensible modules

On Mon, 2009-06-01 at 12:23 -0400, Alvaro Herrera wrote:
> Dimitri Fontaine wrote:
>
> > Le 31 mai 09 à 18:21, Tom Lane a écrit :
>
> >> You could maybe make this work by executing your own transaction
> >> to do it, but I really have to wonder if it's a good idea. One
> >> point to think about is that elog(ERROR) still means elog(FATAL)
> >> at this point, so any brokenness in the queries you're trying to
> >> prepare will result in locking all users out of the database.
> >
> > Yeah that's a pretty good foot gun, yet another one. But
> > preprepare.at_init is optional and defaults to off. If you broke it all,
> > you can turn it off again and reload.
>
> Maybe you could set a callback to be called during the first transaction
> in InitPostgres ... in full knowledge that if it fails, people will be
> locked out of the database.

Should be possible to define a custom variable that has an assign hook
that does nothing unless called with PGC_S_DATABASE or PGC_S_USER. That
should guarantee the code only runs after connection, rather than at
server startup. Not tried it yet.

--
Simon Riggs www.2ndQuadrant.com
PostgreSQL Training, Services and Support