[cfe-commits] r127460 - in /cfe/trunk: lib/Sema/SemaType.cpp test/SemaCXX/return.cpp

[Chandler Carruth]

On Thu, Mar 10, 2011 at 9:27 PM, Eli Friedman <eli.friedman at gmail.com>wrote:

> On Thu, Mar 10, 2011 at 11:56 PM, Rafael Espindola
> <rafael.espindola at gmail.com> wrote:
> > Author: rafael
> > Date: Thu Mar 10 22:56:58 2011
> > New Revision: 127460
> >
> > URL: http://llvm.org/viewvc/llvm-project?rev=127460&view=rev
> > Log:
> > Fix PR9453 by not trying to print a warning about ignored qualifiers
> > in conversion functions.
> >
> > Modified:
> >    cfe/trunk/lib/Sema/SemaType.cpp
> >    cfe/trunk/test/SemaCXX/return.cpp
> >
> > Modified: cfe/trunk/lib/Sema/SemaType.cpp
> > URL:
> http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Sema/SemaType.cpp?rev=127460&r1=127459&r2=127460&view=diff
> >
> ==============================================================================
> > --- cfe/trunk/lib/Sema/SemaType.cpp (original)
> > +++ cfe/trunk/lib/Sema/SemaType.cpp Thu Mar 10 22:56:58 2011
> > @@ -1757,6 +1757,7 @@
> >       // cv-qualifiers on return types are pointless except when the type
> is a
> >       // class type in C++.
> >       if (isa<PointerType>(T) && T.getLocalCVRQualifiers() &&
> > +          (D.getName().getKind() !=
> UnqualifiedId::IK_ConversionFunctionId) &&
> >           (!getLangOptions().CPlusPlus || !T->isDependentType())) {
> >         assert(chunkIndex + 1 < e && "No DeclaratorChunk for the return
> type?");
> >         DeclaratorChunk ReturnTypeChunk = D.getTypeObject(chunkIndex +
> 1);
> >
> > Modified: cfe/trunk/test/SemaCXX/return.cpp
> > URL:
> http://llvm.org/viewvc/llvm-project/cfe/trunk/test/SemaCXX/return.cpp?rev=127460&r1=127459&r2=127460&view=diff
> >
> ==============================================================================
> > --- cfe/trunk/test/SemaCXX/return.cpp (original)
> > +++ cfe/trunk/test/SemaCXX/return.cpp Thu Mar 10 22:56:58 2011
> > @@ -49,3 +49,7 @@
> >     const PCHAR GetName() { return 0; } // expected-warning{{'const' type
> qualifier on return type has no effect}}
> >   };
> >  }
> > +
> > +class foo  {
> > +  operator int * const ();
> > +};
>
> Is there some reason we don't want to warn here?  Or is this just a
> quick hack for 2.9?


If we want to warn on extraneous qualifiers for conversions functions, I
feel like that should be a separate warning tailored to the situation.
Calling it a 'return type' is a bit confusing.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20110310/7553ad5f/attachment.html>