Tom Lane wrote:
> If only ANYELEMENT and not ANYARRAY appears in a function declaration,
> then it can stand for any type, because only rule 2 applies. (The
> difference from ANY is that multiple occurences of ANYELEMENT are all
> constrained to stand for the same type.)
Hmmm, I don't remember that nuance, hence the code deficiency. I'm
should be able to finish up the plpgsql hash table stuff in the next
couple of hours, then I'll get back to thinking about this one. Do I
have until midnite PDT, or EDT?
Joe