Gregory Stark <stark@enterprisedb.com> writes:
> Hmm.... I wonder if this doesn't have the same problems you're describing with
> the toaster. If someone has a cursor WITH HOLD against the table they don't
> get a session level lock against the tables which fed the cursor do
> they?
Hmm, interesting point. The tuples are read out and saved in a
tuplestore before we release locks, but I'm not sure if there is
anything in that codepath that would detoast toast references.
Seems like you're right that there would need to be.
regards, tom lane