This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] NLS back to working order.
- To: "Kaveh R. Ghazi" <ghazi at caip dot rutgers dot edu>
- Subject: Re: [PATCH] NLS back to working order.
- From: Philipp Thomas <pthomas at suse dot de>
- Date: Tue, 6 Jun 2000 20:20:07 +0200
- Cc: gcc-patches at gcc dot gnu dot org
- References: <200006061634.MAA01862@caip.rutgers.edu>
* Kaveh R. Ghazi (ghazi@caip.rutgers.edu) [20000606 18:34]:
> I still think you should make --with-included-gettext the default on
> all systems for reasons I've already explained here:
> Do you have any opinion on this?
Personally I don't like it, but the reasons you present are all very solid.
So yes, I have to agree with you that it is TRT. But doing that would mean
introducing our own version of AM_WITH_NLS that contains an AC_ARG_WITH
which sets nls_cv_force_use_gnu_gettext to yes by default. And we'd have to
do that in order to give user the chance to override this.
> > +AC_C_INLINE
> > +
>
> I'm pretty sure AM_GNU_GETTEXT calls AC_REQUIRE on AC_C_INLINE so the
> above is not necessary (but it doesn't hurt.)
You're prefectly right, it does require AC_C_INLINE. I just forgot to check
what AM_GNU_GETTEXT pulls in by itself.
> You weren't consistent in keeping the order of LIBS. The order should
> be "$(INTLLIBS) $(LIBIBERTY) $(CLIB)" for all cases.
Yes, you're right. I'll do the cleanup when I check in the
--with-included-gettext changes.
> > RCS file: /cvs/gcc/egcs/gcc/intl/dcgettext.c,v
> > -# else
> > +# elif !defined (HAVE_DECL_GETCWD)
>
> I think you are using HAVE_DECL_GETCWD wrong. Its a 3-state macro,
> undefined, 0 or 1. You want to check:
> "defined(HAVE_DECL_GETCWD) && !HAVE_DECL_GETCWD"
Yep, you're right. Shows who's been dealing with that stuff longer than I ;)
> > +#ifndef _GNU_SOURCE
> > +# define _GNU_SOURCE 1
> > +#endif
> >
> > #if defined HAVE_STRING_H || defined _LIBC
> > -# ifndef _GNU_SOURCE
> > -# define _GNU_SOURCE 1
> > -# endif
> Just out of curiosity, why was the above necessary?
Now that I look at it again, it wasn't necessary in this case. I just got
carried away :)
Thanks for your input, I really appreciate it.
Philipp
--
Philipp Thomas <pthomas@suse.de>
Development, SuSE GmbH, Schanzaecker Str. 10, D-90443 Nuremberg, Germany
#define NINODE 50 /* number of in core inodes */
#define NPROC 30 /* max number of processes */
-- Version 7 UNIX fuer PDP 11, /usr/include/sys/param.h