Robert Haas <robertmhaas@gmail.com> writes:
> Now, I think there is a reasonable argument that it would still be
> nice to give an ERROR diagnostic in the cases we can detect, but I do
> not agree with that argument, for all of the reasons stated here: the
> development resources are better spent elsewhere, somebody might be
> creating such a contradictory constraint deliberately for whatever
> purpose, it might annoy or confuse users to get the error in only some
> cases.
It's also arguable that throwing an ERROR would be contrary to spec,
in that it would prevent creation of constraints that the SQL standard
does not forbid.
You could dodge that problem by making the message be just a WARNING
or less. Still, though, the other arguments-against remain valid.
regards, tom lane