Re: [C++ Patch] PR 58980

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

On 01/23/2014 04:01 PM, Paolo Carlini wrote:
> Hi,
>
> On 01/23/2014 03:05 PM, Jason Merrill wrote:
> > On 01/22/2014 04:29 PM, Paolo Carlini wrote:
> > > On 01/22/2014 10:10 PM, Jason Merrill wrote:
> > > > Yep, that's along the lines I was thinking of. But again, prev_scope
> > > > is irrelevant here, so the new code shouldn't mention it at all.
> > > Well, in practice I have to mention it in the error_at itself.
> > Why?  prev_scope is the context where the name is written, i.e. 
> > current_class_type.  Why should the diagnostic treat it as an 
> > explicit qualifier?
> Jason, no problem, you choose. I only wanted to use the same compact 
> form 'A::B' used in the non-template case. Which form do you prefer?
To be clear if we use %qT with nested_name_specifier we get:

58980.C:5:8: error: âtypename A< <template-parameter-1-1> >::Bâ has not 
been declared

which frankly seems suboptimal to me, both vs the non-template case and 
the use of typename.

Thus, as far as I can see, either what I posted, which is consistent 
with the '... has not been declared' message of the non-template case (I 
understand that we want to focus on the 'has not been declared' issue) 
or just '%<%E%>' or even just '%E' and nested_name_specifier.

Paolo.