Mark Hubbart
4/29/2005 5:48:00 AM
On 4/28/05, Martin DeMello <martindemello@yahoo.com> wrote:
> I think it'd be useful to define Infinity and NaN as constants (perhaps
> within Math, if they'd "pollute" the top level). The way I see it,
> there's no real reason not to, and right now one has to resort to clumsy
> looking workarounds to get at them.
This has been talked about a few times before on the list... I think I
remember it being suggested that there might not be a portable way to
generate Infinity and NaN on every platform.
cheers,
Mark