[GRASS5] r.flow: strange compile error M_PI

[Markus Neteler]

On Tue, Jun 28, 2005 at 12:27:55PM +0100, Glynn Clements wrote:
> 
> Brad Douglas wrote:
> 
> > Currently, "gmath.h" doesn't seem like the best place to define it, but
> > it's not the worst, either.
> 
> It shouldn't go in gmath.h; that's for more "heavyweight" stuff.
> 
> > Personally, I would work this out via autoconf by doing something like:
> > 
> > ...
> > AC_MSG_CHECKING(for M_PI) 
> > mpi=no 
> > AC_TRY_COMPILE([#include <math.h>], 
> > [ double f = M_PI; ], 
> > mpi=yes, AC_DEFINE(M_PI, 3.14159265358979323846)) 
> > AC_MSG_RESULT($mpi)
> 
> Personally, I would just conditionalise the local definitions, e.g.:
> 
> 	#ifndef M_PI
> 	#define M_PI 3.14159265358979323846
> 	#endif
> 

What about moving this condition to include/gis.h to avoid that (different)
values are scattered around?

Markus