David Fetter <david@fetter.org> writes:
> On Thu, Feb 21, 2019 at 04:04:41PM -0500, Tom Lane wrote:
>> I still say we should reject this and invent array_greatest/array_least
>> functions instead.
> Might other array_* functions of this type be in scope for this patch?
Uh ... no, I wouldn't expect that. Why would we insist on more
functionality than is there now? (I'm only arguing about how we
present the functionality, not what it does.)
regards, tom lane