Re: [bootstrap] Tentative fix for PR 54281

[Diego Novillo <dnovillo at google dot com>]

On 12-08-17 08:32 , Jakub Jelinek wrote:
> Hi!
>
> On Fri, Aug 17, 2012 at 08:23:02AM -0400, Diego Novillo wrote:
>
> > --- a/gcc/intl.h
> > +++ b/gcc/intl.h
> > @@ -27,8 +27,16 @@
> >   # define setlocale(category, locale) (locale)
> >   #endif
> >
> > +/* If libintl.h is available, include it before testing for NLS. If we
> > +   are building with --disable-nls and another header file includes
> > +   libintl.h, the stubs defined down below will cause syntax errors
> > +   when parsing libintl.h. See http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54281
> > +   for details.  */
> > +#ifdef HAVE_LIBINTL_H
> > +# include <libintl.h>
> > +#endif
> > +
> >   #ifdef ENABLE_NLS
> > -#include <libintl.h>
> >   extern void gcc_init_libintl (void);
> >   extern size_t gcc_gettext_width (const char *);
> >   #else
>
> Will that handle even the case where without --disable-nls intl/
> creates its own libintl.h?  Dunno which targets need that, but
> I'd guess configury wouldn't find it in that case.  So perhaps
> it should be #if defined(HAVE_LIBINTL_H) || defined(ENABLE_NLS) ?

Sounds reasonable.  Amended patch attached.

Re-tested with --disable-nls.


OK for trunk?

Attachment: 54281.diff
Description: Text document