Re: argz.m4: error_t handling wrt argz.h and errno.h

2008-02-13 Thread Mike Frysinger
On Wednesday 13 February 2008, Ralf Wildenhues wrote:
> * Mike Frysinger wrote on Wed, Feb 13, 2008 at 07:26:16PM CET:
> > On Wednesday 13 February 2008, Ralf Wildenhues wrote:
> > > * Mike Frysinger wrote on Wed, Feb 13, 2008 at 05:38:05AM CET:
> > > > the argz.m4 header checks to see if error_t is defined, but only does
> > > > so by including the argz.h header.  if you try to build on a system
> > > > that does provide error_t, but not argz.h, the argz replacement
> > > > module fails to build. on glibc systems, error_t is defined in
> > > > errno.h.  perhaps the gl_FUNC_ARGZ should be checking to see if
> > > > errno.h exists and if so, including it.
> > >
> > > I don't quite understand.  If gl_FUNC_ARGZ finds that error_t is not
> > > defined, it defines __error_t_defined in addition to error_t.
> >
> > this must be a semi-recent addition then ... the package i'm looking at
> > does not do that ... here is the snippet from naim:
> > # AC_LTDL_FUNC_ARGZ
>
> The current gnulib module is newer, please try that.  The macro has also
> been renamed to gl_FUNC_ARGZ for gnulib consistency.

i dont maintain naim, i was just trying to build it.  i imagine when upstream 
gets around to updating, they'll do this.

> > > This should keep your errno.h header from defining error_t.
> >
> > shouldnt errno.h also be checked for the error_t type since that is where
> > it actually gets defined ?
>
> Do you know of a system that has error_t that doesn't use
> __error_t_defined?

nope.
-mike


signature.asc
Description: This is a digitally signed message part.
___
Bug-libtool mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-libtool


Re: argz.m4: error_t handling wrt argz.h and errno.h

2008-02-13 Thread Ralf Wildenhues
* Mike Frysinger wrote on Wed, Feb 13, 2008 at 07:26:16PM CET:
> On Wednesday 13 February 2008, Ralf Wildenhues wrote:
> > * Mike Frysinger wrote on Wed, Feb 13, 2008 at 05:38:05AM CET:
> > > the argz.m4 header checks to see if error_t is defined, but only does so
> > > by including the argz.h header.  if you try to build on a system that
> > > does provide error_t, but not argz.h, the argz replacement module fails
> > > to build. on glibc systems, error_t is defined in errno.h.  perhaps the
> > > gl_FUNC_ARGZ should be checking to see if errno.h exists and if so,
> > > including it.
> >
> > I don't quite understand.  If gl_FUNC_ARGZ finds that error_t is not
> > defined, it defines __error_t_defined in addition to error_t.
> 
> this must be a semi-recent addition then ... the package i'm looking at does 
> not do that ... here is the snippet from naim:
> # AC_LTDL_FUNC_ARGZ

The current gnulib module is newer, please try that.  The macro has also
been renamed to gl_FUNC_ARGZ for gnulib consistency.

> > This should keep your errno.h header from defining error_t.
> 
> shouldnt errno.h also be checked for the error_t type since that is where it 
> actually gets defined ?

Do you know of a system that has error_t that doesn't use
__error_t_defined?

Cheers,
Ralf


___
Bug-libtool mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-libtool


Re: argz.m4: error_t handling wrt argz.h and errno.h

2008-02-13 Thread Mike Frysinger
On Wednesday 13 February 2008, Ralf Wildenhues wrote:
> * Mike Frysinger wrote on Wed, Feb 13, 2008 at 05:38:05AM CET:
> > the argz.m4 header checks to see if error_t is defined, but only does so
> > by including the argz.h header.  if you try to build on a system that
> > does provide error_t, but not argz.h, the argz replacement module fails
> > to build. on glibc systems, error_t is defined in errno.h.  perhaps the
> > gl_FUNC_ARGZ should be checking to see if errno.h exists and if so,
> > including it.
>
> I don't quite understand.  If gl_FUNC_ARGZ finds that error_t is not
> defined, it defines __error_t_defined in addition to error_t.

this must be a semi-recent addition then ... the package i'm looking at does 
not do that ... here is the snippet from naim:
# AC_LTDL_FUNC_ARGZ
# -
AC_DEFUN([AC_LTDL_FUNC_ARGZ],
[AC_CHECK_HEADERS([argz.h])

AC_CHECK_TYPES([error_t],
  [],
  [AC_DEFINE([error_t], [int],
[Define to a type to use for `error_t' if it is not otherwise 
available.])],
  [#if HAVE_ARGZ_H
#  include 
#endif])

since it wasnt defining __error_t_defined, errno.h fell apart due to the 
error_t duplication.

> This should keep your errno.h header from defining error_t.

shouldnt errno.h also be checked for the error_t type since that is where it 
actually gets defined ?
-mike


signature.asc
Description: This is a digitally signed message part.
___
Bug-libtool mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-libtool


Re: argz.m4: error_t handling wrt argz.h and errno.h

2008-02-12 Thread Ralf Wildenhues
[ Cc:ing bug-libtool, it's upstream for argz ]

Hi Mike,

* Mike Frysinger wrote on Wed, Feb 13, 2008 at 05:38:05AM CET:
> the argz.m4 header checks to see if error_t is defined, but only does so by 
> including the argz.h header.  if you try to build on a system that does 
> provide error_t, but not argz.h, the argz replacement module fails to build.  
> on glibc systems, error_t is defined in errno.h.  perhaps the gl_FUNC_ARGZ 
> should be checking to see if errno.h exists and if so, including it.

I don't quite understand.  If gl_FUNC_ARGZ finds that error_t is not
defined, it defines __error_t_defined in addition to error_t.  This
should keep your errno.h header from defining error_t.  Can you please
show the relevant config.log snippets, to find out where this goes
wrong?

There are more headers that might define error_t, argp.h for example.

Thanks,
Ralf


___
Bug-libtool mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-libtool