[RFC] libstdc++/6720 and libstdc++/6671

[Mark Mitchell]

--On Thursday, May 23, 2002 12:05:05 AM +0200 Gabriel Dos Reis 
<gdr@codesourcery.com> wrote:

>
>
>| I don't know *why* the standard would say that.  If it were up to me,
>| I'd certainly rather just let -I do whatever it normally does.  But,
>| we have to work with what we've got.
>
> Still, we would also have to cope with files with non-standard-name
> nevertheless used _internally_ to implement the library.  Those files
> the standard doesn't grant us the right to transmute semantics.

To be clear, you mean something like this:

=== cmath ===

#include <bits/std_cmath.h>

=============

Now, suppose that the user has a "std_cmath.h" in their -I path.

Is your point that we might pick up the wrong one?

I thought Zack's idea:

  #include "./bits/std_cmath.h"

was going to fix that?

--
Mark Mitchell                   mark@codesourcery.com
CodeSourcery, LLC               http://www.codesourcery.com