https://github.com/zygoloid approved this pull request.
Thanks, looks good. Our overall approach here seems error-prone, and I wonder if there's a better way to model the reversal of the conversion sequence. For example, perhaps we could change `ConversionSequenceList` into a class that tracks its `OverloadCandidateParamOrder`, and provides both accessors in candidate order (optionally the object parameter, then the remaining declared parameters in order) and in comparison order (potentially reversed). But let's land this as-is before thinking about that :) https://github.com/llvm/llvm-project/pull/68999 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
