Re: [C++ PATCH for 3.4/4.0/4.1] Fix PR20240 (Accepting invalid usingdecl involving variable declaration)

[Mark Mitchell <mark at codesourcery dot com>]

Kriang Lerdsuwanakij wrote:
> Mark Mitchell wrote:
>
> > Why aren't we issuing an error in supplement_binding, when we try to 
> > add the USING_DECL for "a" to the existing binding for "a"?
> Because do_toplevel_using_decl never calls supplement_binding. From what 
> I see, supplement_binding is mainly intended for struct stat hack (with 
> a few other case thrown in).  And then there is DR400.  I could try a 
> different approach, using
> supplement_binding and such in do_toplevel_using_decl.

Oh, I missed that this was a toplevel binding.

The real problem here, of course, is that we have such different code 
paths for top-level bindings from local bindings; there's nothing in the 
language that implies that these things should be so different.  But, 
that's a big project.

> By the way, the patch was already approved and committed:
>
>  http://gcc.gnu.org/ml/gcc-patches/2005-03/msg01771.html
>
> So it has to be taken out and replaced if we want to fix it
> in a different way.

I think it's OK.  It might result in some duplicate diagnostics, in the 
worst case.

Thanks,

--
Mark Mitchell
CodeSourcery, LLC
mark@codesourcery.com
(916) 791-8304