kazutakahirata wrote: > As a side note, should we create a clang-tidy check to detect these cases > automatically: to use `llvm` algorithms instead of `std` and so on? Is it > documented that we should always prefer the `llvm` version? I see a lot of > work by @kazutakahirata and wish to make it future-proof because any new > contributor could use `std` version as it is the most familiar one.
Yes, a clang-tidy check would be useful. Detecting the case in this particular PR may require some heuristics though. For example, if we see `enumerator_begin()` and `enumerator_end()`, then check for `enumerators()` (the plural form after removing `begin`). https://github.com/llvm/llvm-project/pull/151801 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
