[
Lists Home |
Date Index |
Thread Index
]
> > maxInclusive of INF
> > minInclusive of -INF
I thought I would add a little more to this thread-- I was reading the
erratta (all 137 pages) and came across a change in the construction of
double and float:
From E2-40
Any value ·incomparable· with the value used for the four bounding facets
(·minInclusive·, ·maxInclusive·, ·minExclusive·, and ·maxExclusive·) will be
excluded from the resulting restricted ·value space·. In particular, when
"NaN" is used as a facet value for a bounding facet, since no other float
values are ·comparable· with it, the result is a ·value space· either having
NaN as its only member (the inclusive cases) or that is empty (the exclusive
cases). If any other value is used for a bounding facet, NaN will be
excluded from the resulting restricted ·value space·; to add NaN back in
requires union with the NaN-only space.
This is just further proof that you should use the inclusive cases of INF
and -INF. This will help your cmopatibility now-- and the will be usable in
the future. Of course, whether or not this makes more sense than using a
pattern is still debateable.
Cheers,
Jeff Rafter
|
