Re: Handling of #undef FOO

[Ismail Dönmez]

Monday 07 January 2008 09:25:10 tarihinde şunları yazmıştınız:
> On Sun, 2008-01-06 at 13:10 +0200, Ismail Dönmez wrote:
> > Sunday 06 January 2008 08:30:14 tarihinde Paul Eggert şunları yazmıştı:
> > > Ismail Dönmez <address@hidden> writes:
> > > > but it should be defined like this,
> > > >
> > > > #ifndef FOO
> > > > #define FOO
> > > > #endif
> > >
> > > Why should it be defined like that?  Typically, config.h is supposed
> > > to define FOO; if something else is defining FOO first, that's a
> > > problem with the "something else", not with config.h.
> >
> > Because otherwise it breaks with gcc 4.3 where FOO is defined multiple
> > times, see http://www.cyrius.com/journal/gcc/gcc-4.3-pedwarn.html . I
> > already hit this in multiple apps due to HAVE_STDLIB_H and such common
> > config defs.
>
> So far, I've only seen this happening with packages which happen commit
> the mistake to install config.h's or which expose supposed-to-be private
> headers.
>
> Can you provide a real world example for which this happens?

Sure,

Breaks lots of apps via mpeg4ip package which installs a config.h and uses it 
through its headers though the file itself is private.

Breaks again many apps via jconfig.h which is installed by jpeg package and 
used by headers.

I am sure more of this bugs are just hiding in the corner.

-- 
Never learn by your mistakes, if you do you may never dare to try again.