Re: Plan invalidation vs temp sequences

I wrote:
> Well, we *have* the sequence's Oid in the regclass constant, the problem
> is the difficulty of digging through the plan tree to find it. I did
> consider having the planner extract it and save it aside somewhere, but
> there doesn't seem to be any very convenient place to do that, short of
> an extra traversal of the query tree, which is pretty annoying/expensive
> for data that will probably never be needed for most queries.

Actually ... now that I've consumed a bit more caffeine, it seems this
could be done relatively cheaply in setrefs.c. We could add a
list-of-relation-OIDs to PlannedStmt, and charge setrefs.c with creating
the list, and simplify plancache.c to just use list_member_oid() instead
of groveling over the rangetable for itself. I'll go try that out.

regards, tom lane