[cfe-commits] [PATCH] [Review Request] for bug7505 with test cases, for initial review

[Faisal Vali]

*bump*
(no rush - am delighted variadic templates has been getting some
serious attention - just wanted to make sure this patch review wasn't
completely forgotten)

Happy new year to all :)

Faisal Vali





On Sun, Dec 12, 2010 at 11:39 PM, Faisal Vali <faisalv at gmail.com> wrote:
> Once again Doug - sorry for the delay.
>
> To recap, this patch is intended to completely fix bug7505 and allow
> the following code:
> template<class T> void f();
> template<class T, class U> T f(U);
>
> bool b = f<int>;
> int i = (int) f<int>;
>
> I tried two approaches in fixing this bug - patching BuildTemplateID
> vs patching each of the casts and the standard conversions.
> Patching BuildTemplateId did not work because it would bypass overload
> resolution (unless i detect trailing lparens) in certain cases and
> create confusing results.
>
> Please let me know what your initial thoughts are and i can clean the
> patch up and resubmit.
> I tested against SemaCXX, CXX, Sema, SemaTemplate
>
> Thanks!
> -fas
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: first-submission-contains-both-fixes.patch
Type: application/octet-stream
Size: 49250 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20101231/d400c16e/attachment.obj>