Neil Conway <neilc@samurai.com> writes:
> On Mon, 2007-28-05 at 15:38 -0400, Tom Lane wrote:
>> More generally, I'm really hoping to get rid of bespoke text<->whatever
>> cast functions in favor of using datatypes' I/O functions.
> I don't object, but I'm curious: is there a benefit to this other than
> brevity of implementation? ISTM the spec has the idea that the input to
> a type's constructor is often distinct from the type's text => type
> casting behavior.
Well, (a) it would fill in a whole lot of text-conversion cases that are
currently missing, and (b) it would encourage datatype implementors to
keep the I/O and text-conversion cases behaving alike unless there were
a REALLY good reason not to. IMHO most of the cases that the SQL spec
calls out as behaving differently are pure brain-damage.
regards, tom lane