sstwcw added a comment. Allow me another attempt in justifying this patch.
Using the macros, it is easier to see when a keyword is not classified as some language's keyword. The author and reviewers didn't notice `final` and `dollar` being in only one place previously. With the old way, when someone adds a new keyword he is also more likely to forget to include it in the sets. In the commit message when I said the old way turned out to be error-prone, I did debug why some code doesn't get formatted right only to find I forgot to add a keyword in one of the places. With the `isIdentifier` and `isKeyword` functions matching the sets separately, it is more likely one of them will get it wrong. In the original `isCSharpKeyword` function, the check is wrong, it should be `!=`. CSharpExtraKeywords.find(Tok.Tok.getIdentifierInfo()) == CSharpExtraKeywords.end() Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D121753/new/ https://reviews.llvm.org/D121753 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
