Peter Eisentraut <peter_e@gmx.net> writes:
> Am Mittwoch, 9. Juli 2008 schrieb Peter Eisentraut:
>> I propose that we relax these two checks to test for binary-coercibility
>> instead, which is effectively what is expected and required here anyway.
> Here is the corresponding patch.
Looks good, but you might want to add a sentence to the documentation
pointing out that the function result type has to match the cast target
type; that seems not to be explicitly stated anywhere.
I also notice that the CREATE CAST page is still worded as if "binary
compatible" were a symmetric relationship, which it is not. We probably
need to rework it a bit; but that's not really related to this patch.
regards, tom lane