iswblank failure [was: IRIX failures]

[Eric Blake]

On 09/01/2010 12:48 PM, Tom G. Christensen wrote:
> > Definitely something that can be fixed in gnulib.  Can you help figure
> > out where the earlier iswblank declaration was coming from?
>
> Presumably from one of these system headers<internal/wctype_core.h>  or
> <internal/wchar_core.h>  which both contain this:
> #if defined(__c99)
> extern int      iswblank(wint_t);
> #if !defined(_SGI_COMPILING_LIBC)
> #pragma optional iswblank
> #endif
> #endif /* __c99 */
>
> I've uploaded the pre-processed source here:
> http://jupiterrise.com/tmp/printf-args.i

That shows /usr/include/internal/wctype_core.h as the first culprit in 
your particular case.

> It seems that gcc defines __c99 in -std=gnu99 mode:
> $ gcc -E -dM -std=gnu99 -<  /dev/null|grep c99
> #define __c99 1
>
> > Also, that
> > particular declaration at line 585 of wctype.h (line 282 of wctype.in.h)
> > is guarded by:
> >
> > #if ! @HAVE_ISWCNTRL@ || @REPLACE_ISWCNTRL@
> > #elif ! @HAVE_ISWBLANK@
> >
> > So knowing the value of all the HAVE_IS* and REPLACE_IS* from config.log
> > would be useful to help diagnose what is going wrong here.

HAVE_ISWCNTRL 1
HAVE_ISWBLANK 0
no definition for REPLACE_ISWBLANK

sounds like the m4 test needs to be taught to look for a declaration, 
and if one exists, then line 283 of wctype.in.h needs to key off of 
REPLACE_ISWBLANK and declare rpl_iswblank in that case.

Bruno, is this something you can do quickly?

--
Eric Blake   address@hidden    +1-801-349-2682
Libvirt virtualization library http://libvirt.org