Re: Bug on NetBSD : redefinition of wint_t

[akim]

On Mon, Aug 06, 2001 at 12:00:29PM -0700, Paul Eggert wrote:
> 
> > Maybe that's something we should handle
> > at configure.in's level, probably checking whether we can include
> > stddef and wchar together.
> 
> How about this idea for a more general fix?  It would fix many of the
> problems that we run into in this area:
> 
> 1.  Modify AC_CHECK_HEADER to always try to compile the header; it
>     should never merely just try to preprocess it.

Agreed, I would very much like to do that but I'm very afraid people
are not used to check the headers passing the prerequisite headers.
I very much doubt that merely including foo.h for compilation would
make it work, but just because of this.

Of course you know that, hence the proposal to...

> 2.  Modify AC_CHECK_HEADER to use the default includes if INCLUDES is
>     empty, just as the other check macros do.
> 
> The idea is to fix autoconf without requiring any fixes to other packages.

I agree the idea is attractive, the hypothesis being that default headers
are always just what we need for headers people are willing to check the
existence of.

But I must confess it scares the hell out of me.  You really want to see
me crucified at the next Autoconf release :(

I need a break :)

Well, we could give this a try at least amongst Autoconf beta testers,
but I fear this change very much.  While if Autoconf was to be deisnged today,
I would sure go that way.

But it seems people don't like when we leave the wrong tracks and try
walking towards light.