JonasToth added a comment. In https://reviews.llvm.org/D54061#1288395, @sammccall wrote:
> In https://reviews.llvm.org/D54061#1286956, @JonasToth wrote: > > > > Theoretically, we could replace `ClangTidyCheck::check` with > > > `ClangTidyCheck::run`, but I'm not sure it is worth, > > > `ClangTidyCheck::check` is a public API, and is widely-used (for all > > > clang-tidy checks), replacing it requires large changes (although it is > > > one-line change), it might break downstream clang-tidy checks. > > > > We can add a deprecation warning and remove the `check` method in the > > next version? > > > As Haojian says, this is a lot of churn. > I don't think the benefit of pushing people to migrate, even with a grace > period, is clear. > > Additionally it's *possible* that future refactorings may mean we want to run > additional "framework" logic when the MatchFinder calls us, and so requiring > `ClangTidy::check` and `MatchCallback::run` to be the same function seems a > little risky. > > I'll add a comment explaining why the two functions exist. Alright. Repository: rCTE Clang Tools Extra https://reviews.llvm.org/D54061 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
