Re: iso varargs fix

[Sebastian Wilhelmi <wilhelmi ira uka de>]

Hi,

> I just noticed that the latest ISO varargs fix broke the build for the
> non ISO varargs case.  The following was in gmessages.h:
>   #ifdef G_HAVE_ISO_VARARGS
> 
> which now always evaluates true.  I made the obvious fix and checked it
> into CVS:
>   #if G_HAVE_ISO_VARARGS
> 
> I haven't checked if there are other occurences of G_HAVE_ISO_VARARGS
> that need to be fixed because of the change.

I would prefer leaving #ifdef G_HAVE_ISO_VARARGS and defining
G_HAVE_ISO_VARARGS only if it should be. I think, that is more along the lines
of GLib, as there is no other 'boolean' macro defined to be zero to mean
FALSE.

Any objections to changing it to

#if (defined(__cplusplus) ? defined(G_HAVE_ISO_CXX_VARARGS) : \
                            defined(G_HAVE_ISO_C_VARARGS))
# define G_HAVE_ISO_VARARGS 1
#endif

Bye,
Sebastian
-- 
Sebastian Wilhelmi
mailto:wilhelmi ira uka de
http://goethe.ira.uka.de/~wilhelmi