Simon Riggs wrote:
> On Tue, 2008-04-22 at 18:13 -0400, Alvaro Herrera wrote:
> > Simon Riggs wrote:
> > 
> > > OK, so it can;t be copied to a longer lived memory context?
> > 
> > If you need that ability, please explain.
> No, I wish to prevent that, not enable it.

I see.  Sure, we don't have that problem.  In fact, we didn't have it
before either, so I'm not sure I see your point :-)

> Perhaps put the TransactionId on each snapshot and then an Assert can
> check it before its used. 

There's no need for that -- all snapshots go away at transaction end.
An attempt to use one would cause a prompt crash (at least on an
assert-enabled build.)

Alvaro Herrera                      
PostgreSQL Replication, Consulting, Custom Development, 24x7 support

Sent via pgsql-patches mailing list (
To make changes to your subscription:

Reply via email to