Neil Conway <neilc(at)samurai(dot)com> writes:
> Personally, I'm inclined to change both of these cases to result in an
> error...
No strong objection from me, but perhaps you ought to toss out a query
on pgsql-sql or pgsql-general to see if anyone wants to complain. Not
all the folks who might be upset read pghackers.
regards, tom lane