26_numerics/c99_classification_macros_c.cc

[Loren James Rittle]

>So what I understand you to be saying is that the test still fails
>(as expected) on freebsd5, but for some reason PCH is making it
>succeed when it shouldn't.

Yes, PCH is making it succeed when it shouldn't.

>Why is the test failing in the first place?  I confess I don't
>understand what it is trying to test.

math.h on C99 contains:

fpclassify(a) (a macro)

that should conflict with:

void fpclassify() { }
(what is in c99_classification_macros_c.cc)

Some library maintainers believe that there is some purpose in
checking that this conflict occurs since it is related to "clean
namespaces", etc.  I'm less sure that it is worth doing but
agree to the party line.

Loren