Re: [C++ Patch / RFC] PR 50870

[Paolo Carlini <paolo dot carlini at oracle dot com>]

On 10/28/2011 06:07 PM, Jason Merrill wrote:
> On 10/27/2011 08:22 PM, Paolo Carlini wrote:
> > I'm trying to figure out where the very different args argument is
> > coming from.
>
> > Earlier than that things are different: in mainline the same args, as
> > arglist, comes from fixup_template_parm, and earlier we have
> > fixup_template_parms which creates the arglist itself
>
> Right, this is new.  I guess the COMPONENT_REF code needs to be fixed 
> to handle partial instantiation.
I see. Something I didn't tell you yesterday, is that, in 4_5-branch, 
tsubst_template_arg is called like this, by coerce_template_parms:

      /* There must be a default arg in this case.  */
      arg = tsubst_template_arg (TREE_PURPOSE (parm), new_args,
                     complain, in_decl);

which, what can I say, looks right ;)

By the way, your hint that probably we are also passing a wrong first 
argument to qualified_name_lookup_error may be useful for the other 
issue, the ice-on-invalid, which I was trying to fix in the parser ;) 
Let me test something...

Paolo.