r211031 - [C++1z] Implement N4051: 'typename' is permitted instead of 'class' when declaring a template template parameter.

[Richard Smith]

On Thu, Jun 19, 2014 at 8:53 AM, Jordan Rose <jordan_rose at apple.com> wrote:

>
> On Jun 19, 2014, at 4:41 , Richard Smith <richard at metafoo.co.uk> wrote:
>
> On Wed, Jun 18, 2014 at 9:19 AM, Jordan Rose <jordan_rose at apple.com>
> wrote:
>
>>
>> On Jun 16, 2014, at 8:51 , Richard Smith <richard-llvm at metafoo.co.uk>
>> wrote:
>>
>> +def CXXPre1zCompat : DiagGroup<"c++98-c++11-c++14-compat">;
>> +def CXXPre1zCompatPedantic :
>> DiagGroup<"c++98-c++11-c++14-compat-pedantic",
>> +                                       [CXXPre1zCompat]>;
>>
>>
>> I'm not happy with this warning name. Can we say the *compat options are
>> just the newest version? c++14-compat?
>>
>
> That doesn't work: we want -Wc++11-compat to list all incompatibilities
> with C++11, including this one.
>
> My preference is to make this group unnamed, but I don't know if we
> support that, nor what we'd list in the [-W...] in the diagnostic if we did.
>
>
> Then how about
>
> -Wpre-c++1z-compat, which is included by -Wc++98-compat, -Wc++11-compat,
> and -Wc++14-compat.
> -Wpost-c++11-compat, which is included by  -Wc++11-compat, -Wc++14-compat,
> and -Wc++1z-compat.
>
> ?
>
> It's not perfect because "post-" doesn't usually include the reference
> point, but it does provide a very consistent taxonomy.
>

I would expect the latter warning to be called -Wpost-c++98-compat (that
is, warn for things that aren't compatible with C++ after C++98). Other
than that, I think the scheme you suggest is an improvement.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20140620/c3a93dfe/attachment.html>