Re: Why no pg_has_role(..., 'ADMIN')?

Robert Haas <robertmhaas(at)gmail(dot)com> writes:
> On Fri, Sep 20, 2024 at 2:34 PM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> I'm now inclined to add wording within the pg_has_role entry, along
>> the lines of
>>
>> WITH ADMIN OPTION or WITH GRANT OPTION can be added to any of
>> these privilege types to test whether ADMIN privilege is held
>> (all six spellings test the same thing).

> I don't have an opinion about the details, but +1 for documenting it
> somehow. I also think it's weird that we have six spellings that test
> the same thing, none of which are $SUBJECT. pg_has_role seems a little
> half-baked to me...

Yeah. I think the original idea was to make it as parallel to
has_table_privilege and friends as we could (but why did we then
stick a pg_ prefix on it?). So that led to MEMBER WITH GRANT OPTION,
and then the other spellings seem to have come along later.

regards, tom lane