jyknight added a comment. In D153156#4599595 <https://reviews.llvm.org/D153156#4599595>, @rZhBoYao wrote:
> What if a programmer is really trying to call operator""b here (albeit > ill-formed) Because that code is ill-formed, Clang diagnosed it with an error by default. Isn't that preferable to accepting it by default? And disabling the error treats it as string-concat, because that enables users to remain compatible with code written assuming C++98 semantics for macros in string concatenation. (This support was a quite-intentionally added feature, not an accident of implementation!) FWIW, GCC does the same thing, but with an on-by-default-warning that doesn't even default to error severity. I do think it's OK to remove support for this extension if spec-compliance requires it, but AFAICT, that's not the case here, so I think it would be preferable to preserve the previous behavior. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D153156/new/ https://reviews.llvm.org/D153156 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
