Re: C compilation questions
On Fri, Dec 29, 2000 at 06:35:55PM +0000, Mark Brown wrote:
> On Fri, Dec 29, 2000 at 08:29:12AM -0800, Sean 'Shaleh' Perry wrote:
>
> > > So the problem seems to be some conflict over the name "complex", but I
> > > don't understand what the conflict is exactly. gcc supports the keyword
> > > "__complex__" via complex.h, but that's different to just defining
> > > "complex", isn't it? Any ideas?
>
> > I seem to recall something about complex being a valid type in C99 (the new
> > ansi C spec).
>
> Yes, it is - there's a complex type although I believe you're supposed
> to have to include <complex.h> to get it.
>
Yes, the problem is related to complex.h. There's a line in the local
machine.h that says
/* if the system has complex.h */
#ifdef HAVE_COMPLEX_H
#include <complex.h>
#endif
If I comment this out, then I don't get the problem, and I get a full
compile.
The problem then is that the new keyword "complex" is being used
as a local type definition. I guess complex.h didn't exist when meschach was
written (though in that case one does wonder why it's in meschach's machine.h!)
How can I find more about complex.h? There seems to be contradictions been
what info gcc says and what complex.h says.
Looks like the only solution for meschach is to ignore complex.h altogether.
Drew
--
PGP public key available at http://dparsons.webjump.com/drewskey.txt
Fingerprint: A110 EAE1 D7D2 8076 5FE0 EC0A B6CE 7041 6412 4E4A
Reply to: