Paul Schlie <schlie_at_comcast.net> writes:
>> There are naturally three disjoint cases:
>> * finite?
>> * infinite?
>> * nan?
>
> as (fl< +nan.0 +inf.0) => #f, and considered a valid numerical
> comparison, it would appear they are not numerically disjoint?
They are disjoint in the sense that every real number satisfies
exactly one of these predicates.
> [although the draft is seemingly silent on (fl<= +nan.0 +inf.0)]
IEEE arithmetic says #f, and I consider bropken any language which
tries to do otherwise.
> and why indefinite? may be a more appropriate name than infinite?
You can use (not (finite? x)). No need for a separate predicate.
Strictly speaking any two of the three predicates would be sufficient,
but providing all three is convenient.
--
__("< Marcin Kowalczyk
\__/ qrczak_at_knm.org.pl
^^ http://qrnik.knm.org.pl/~qrczak/
Received on Sun Oct 08 2006 - 15:21:23 UTC