On 20.01.2012, at 17:17, David Blaikie wrote:
> I've had the same thought but never got around to trying to implement
it. Does everything compile for you if it's protected? If so, then a patch
would probably be happily accepted
Did that in r148550. It found a really dubious piece of code in clang but
compiled fine otherwise.
- Ben
> From: Vane, Edwin
> Sent: 1/20/2012 7:13 AM
> To: llvmdev at cs.uiuc.edu
> Subject: [LLVMdev] Public SmallVectorImpl constructor?
>
> Hi all,
>
> Just finished debugging a memory clobbering bug resulting from using
SmallVectorImpl directly without realizing this is a bad idea (aside: I was
using it directly because llvm::sys::path::append()’s first argument is a
SmallVectorImpl<char>). A note in the docs about not using SmallVectorImpl
directly would be nice but could we go further and make SmallVectorImpl’s
constructors not public? Is there a reason why they’re public right now?
>
> --
> Edwin Vane
> Software Developer
> SSG/DPD/ECDL/ArBB
> Phone: +1 519 772 2567
> iNET: 87722567
>
> _______________________________________________
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev