> > abstime <-> int4: I think these should not be implicit because they
> > represent different "kinds" of data. (These are binary
> compatible casts,
> > so changing them to not implicit probably won't have any
> effect. I'd have
> > to check this.)
>
> I believe that as of current sources we can mark a binary cast
> non-implicit,
> and I agree with marking these two explicit-only.
Everything in this proposal looks pretty good. With regards to the above
abstime<->int4 thing - what about the 'magic' values in that conversion.
(eg. -infinity, etc.)
Chris