Robert Haas <robertmhaas@gmail.com> writes:
> On Mon, Jun 19, 2017 at 9:57 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> Hm, patch looks okay, but while eyeballing it I started to wonder
>> why in the world is pg_get_publication_tables marked prosecdef?
>> If that has any consequences at all, they're probably bad.
>> There are exactly no other built-in functions that have that set.
> Should we add that to the opr_sanity tests?
Yeah, I was wondering about that too. I can imagine that someday
there will be prosecdef built-in functions ... but probably, there
would never be so many that maintaining the expected-results list
would be hard.
regards, tom lane