On Tue, Oct 27, 2015 at 06:10:26PM +0000, Samuel Benzaquen via cfe-commits wrote: > sbenza added a comment. > > In http://reviews.llvm.org/D14096#275902, @xazax.hun wrote: > > > There is already a similar check in the Google package. What are the > > differences between those two checks? What is the reason we can not just > > register that check into the core guidelines module? > > > That other check discourages c-style cast in favor of C++ style casts, even > if it is a reinterpret_cast. It simply replaces the cstyle cast with an > equivalent C++ one. It is basically a stylistic check. > > This check will warn unsafe cstyle casts, while allowing safe ones like > int->uint casts. > This one is a safety related check.
Looking back to the discussion about the C++ style casts, this argument makes no sense. For C++ code, reinterpret_cast is clearly preferable over C-style casts for all but code size reasons. There seems to be no consideration about "safe" uses with reinterpret_cast, so why should C-style casts be different? Joerg _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
