On 3/30/2011 9:49 AM, Robert Haas wrote:
> On Tue, Mar 29, 2011 at 9:40 PM, Stephen Frost<sfrost@snowman.net> wrote:
>> * Robert Haas (robertmhaas@gmail.com) wrote:
>>> I do think we need some kind way of capturing DDL events, though. I wonder if the object-access-hook stuff KaiGai
andI did to support SE-PostgreSQL could be extended to meet this need...
>>
>> My inclination would be 'probably', but it's not likely to really be the
>> way we'd want to provide generalized DDL triggers..
>
> I guess I was imagining that DDL triggers would be primarily important
> for things like Slony, that are already writing C code anyway, but
> maybe that's overly optimistic...
>
Slony is using C code in every performance critical path. Other than
that, we are perfectly happy with PL/pgSQL code.
What I would envision for DDL triggers is that they first don't fire on
an object type, but rather on a command completion code, like "CREATE
TABLE" or "DROP SCHEMA".
To do anything useful with that of course would require that all DDL
does go through tcop's ProcessUtility and actually synthesizes a proper
Utility parsetree. That isn't the case today, so there would be some
previous clean up work to be done.
Jan
--
Anyone who trades liberty for security deserves neither
liberty nor security. -- Benjamin Franklin