On Thursday, January 20, 2011 19:23:17 Karel Zak wrote:
> On Thu, Jan 20, 2011 at 06:50:03PM -0500, Mike Frysinger wrote:
> > On Thursday, January 20, 2011 18:45:28 Karel Zak wrote:
> > > On Tue, Jan 18, 2011 at 05:47:01PM +0100, Fabian Groffen wrote:
> > > > +have_dirfd=
> > > > +AC_MSG_CHECKING([for dirfd])
> > > > +AC_LINK_IFELSE([AC_LANG_PROGRAM([[
> > > > +#include <dirent.h>
> > > > +]],[[
> > > > + DIR *dir = 0;
> > > > + if (dirfd(dir)) ;
> > > > +]])], [have_dirfd=yes], [have_dirfd=no])
> > > > +AC_MSG_RESULT([$have_dirfd])
> > >
> > > What about:
> > >
> > > AC_CHECK_FUNCS(dirfd)
> > > AC_CHECK_DECLS([dirfd], , ,
> > >
> > > [#include <sys/types.h>
> > >
> > > #include <dirent.h>])
> >
> > i think macros would confuse these checks ?
>
> AC_CHECK_DECLS generates:
>
> int
> main ()
> {
> #ifndef dirfd
> #ifdef __cplusplus
> (void) dirfd;
> #else
> (void) dirfd;
> #endif
> #endif
> }
ah, so it should be OK
> BTW gnulib has dirfd.m4, but it seems like overkill for us.
if it were just the .m4, i'd be for it, but importing the whole module does
seem like overkill
-mike
Attachment:
signature.asc
Description: This is a digitally signed message part.
