> I assume you are looking at the 'setof' processing? Offhand it seems to
> me that this code is broken anyway: use of a relation type should refer
> to the tuple type, but should *not* imply SETOF, at least IMHO.
No, there is another routine (not remembering the name right now) which
is involved, I *think* from within gram.y, which barfs when called with
"opaque" as an argument (among other things). Can look up more info
later (I'm away this weekend).
- Thomas