[llvm-commits] [lld] r172482 - /lld/trunk/lib/ReaderWriter/ELF/WriterELF.cpp

[David Blaikie]

On Mon, Jan 14, 2013 at 2:10 PM, Nick Kledzik <kledzik at apple.com> wrote:
> Author: kledzik
> Date: Mon Jan 14 16:10:22 2013
> New Revision: 172482
>
> URL: http://llvm.org/viewvc/llvm-project?rev=172482&view=rev
> Log:
> Fix shadowed variable warning
>
> Modified:
>     lld/trunk/lib/ReaderWriter/ELF/WriterELF.cpp
>
> Modified: lld/trunk/lib/ReaderWriter/ELF/WriterELF.cpp
> URL: http://llvm.org/viewvc/llvm-project/lld/trunk/lib/ReaderWriter/ELF/WriterELF.cpp?rev=172482&r1=172481&r2=172482&view=diff
> ==============================================================================
> --- lld/trunk/lib/ReaderWriter/ELF/WriterELF.cpp (original)
> +++ lld/trunk/lib/ReaderWriter/ELF/WriterELF.cpp Mon Jan 14 16:10:22 2013
> @@ -1542,10 +1542,10 @@
>
>    /// \brief find a absolute atom pair given a absolute atom name
>    struct FindByName {
> -    const std::string name;
> -    FindByName(const StringRef name) : name(name) {}
> +    const std::string _name;

I'm not sure this is a shadowing variable warning we should bother
fixing. If Clang is warning, we should fix Clang to suppress this case
as it's a common idiom (but perhaps still catch cases where the
parameter is referenced in the body of the ctor where, presumably, the
user meant to mention the member instead). If GCC is warning & Clang
isn't, we should consider just suppressing the GCC warning & relying
on Clang's.

> +    FindByName(const StringRef name) : _name(name) {}

Odd: we don't usually mark value parameters as "const".

>      bool operator()(AbsoluteAtomPair& j) {
> -      return j.absoluteAtom()->name() == name;
> +      return j.absoluteAtom()->name() == _name;
>      }
>    };
>
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits