[cfe-commits] r155677 - in /cfe/trunk: include/clang/Basic/DiagnosticCommonKinds.td lib/Parse/ParseDecl.cpp lib/Sema/SemaDecl.cpp test/FixIt/fixit.cpp test/Parser/cxx-using-declaration.cpp

[David Blaikie]

On Thu, Apr 26, 2012 at 5:37 PM, Richard Smith <richard at metafoo.co.uk> wrote:
> On Thu, Apr 26, 2012 at 4:54 PM, David Blaikie <dblaikie at gmail.com> wrote:
>>
>> On Thu, Apr 26, 2012 at 4:36 PM, Kaelyn Uhrain <rikka at google.com> wrote:
>>  > +void Bar(int); // expected-note{{non-type 'Bar' shadowing class 'Bar'
>> declared here}}
>>
>> To me this sort of reads strangely - "non-type 'Bar' shadowing (class
>> 'Bar' declared here)" rather than "non-type 'Bar' shadowing (class
>> 'Bar') declared here" - but perhaps I'm being pedantic. "class 'Bar'
>> shadows non-type 'Bar' declared here" feels more clear to me, but does
>> still suffer from the same ambiguity...
>
>
> I suggested earlier on IRC: "class 'Bar' is hidden by a non-type declaration
> of 'Bar' here". Does that read better to you?

Yes, that seems unambiguous.

> I'm torn between 'hidden' and
> 'shadowed' -- I think the former is clearer (and is the standard term), but
> the latter is already used in other diagnostics and -Wshadow.

I'd fall on the side of 'hidden' because I prefer standard wording
where possible. Pity about the existing convention, though.

- David