Re: Solaris 2.6 finite() prototype

["Andrew S . Halper" <ashalper acm org>]

On 2001.08.10 23:49 Cyrille Chepelov wrote:
> Le ven, aoû 10, 2001, à 03:37:26 -0700, Andrew Halper a écrit:
>
> > >         (check that autoheader really includes a HAVE_IEEEFP variable
> then),
> > > and then #ifdef HAVE_IEEEFP (_H ?), do the right thing (it can't hurt to
> > > include both ieeefp.h and math.h on platform which have both but have
> > > finite() in math.h, anyway)
> >
> > Patch attached.  Let me know if there are problems.
>
> Patch looks good (except that it's in the reverse order), it goes in (in my
> "new" StdProp tree), with #include "config.h" guarded by HAVE_CONFIG_H.

OK, it goes <newtree> <oldtree> instead of <oldtree> <newtree> I guess?  Sorry
about that.  It's been ages.

After clearing up those compiler warnings, I noticed the isinf() function in
lib/dia_xml.c is another one that is not in Solaris math.h.  Solaris isinf() is
prototyped in sunmath.h, but the patch is going to be more complex, because I
think it also involves linking libsunmath.

Andy