Peter Eisentraut <peter(dot)eisentraut(at)enterprisedb(dot)com> writes:
> Yes please!
> I have occasionally wondered whether just passing the isnull argument as
> NULL would be sufficient, so we don't need a new function.
I thought about that too. I think I prefer Daniel's formulation
with the new function, but I'm not especially set on that.
An advantage of using a new function name is it'd be more obvious
what's wrong if you try to back-patch such code into a branch that
lacks the feature. (Or, of course, we could back-patch the feature.)
regards, tom lane