Neil Conway <neilc@samurai.com> writes:
> On Tue, 2005-01-25 at 02:40 -0500, Tom Lane wrote:
>> Offhand I'd say this should draw a "no such cursor as foo" error.
>> I'm too tired to look into why foo still exists after the rollback...
> I'm confused; I wasn't involved in the design discussions about portals
> and subtransactions this summer, but my understanding is that making
> portals non-transactional was the conclusion. Shouldn't that imply that
> a DECLARE in an aborted subtransaction should persist?
I don't recall the discussions from last summer in detail, but it can't
possibly be rational to allow a cursor created in a failed
subtransaction to persist beyond that subtransaction... your example
in which the cursor uses tables that no longer exist is a fairly
egregious example of why not, but there are others.
regards, tom lane