xgsa added a comment. In https://reviews.llvm.org/D41326#957749, @JVApen wrote:
> I'm missing some documentation to understand the corner cases. How does this > check behave with suppressed warnings for checks which ain't currently > checked. (Using -no-... on a code base or suppressing the warnings via the > pragmas) The check just gathers NOLINT comments and compares them with a list of reported diagnostics for the current configuration. Thus, it will report diagnostics on the NOLINT comments, which are added for the checks, which are disabled in current configuration (e.g. with option -check=-some_check). Note also that in spite NOLINT for clang warnings now works for clang-tidy, it doesn't work for clang itself, so I wouldn't recommended using something like "// NOLINT(clang-diagnostic-unused-variable)". As you mentioned, pragmas or -no-... options should be used instead. The check doesn't perform any processing of those things. https://reviews.llvm.org/D41326 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
