MyDeveloperDay added a comment. In D69764#2936827 <https://reviews.llvm.org/D69764#2936827>, @HazardyKnusperkeks wrote:
> First off, I think it should be configured in a different way, to prepare the > path for also formatting static, inline, etc. To be honest I think if we wanted to do that we could do so by extending the "CVQualifierOrder" to support more keywords (To be honest I think the original requirement is mainly for const, I concede on the volatile as for the others I think I'd want to see how we get on in order to reduce the complexity. but it would be fun to try I guess!) > If this is kept there should be tests on what happens if there is const or > volatile more than once in the string list, and when there are typos. Should > there be a command line warning on a typo or an unsupported identifier is in > the list? Typos would be ignored by the "if", duplicates well that would impact because the keywords push though themselves hence the reversing based on direction. I guess we COULD validate the options, but I think that would make the .clang-format file not very future proof if I check that it only contains "const volatile", let me check if other options every validate themselves in quite the same way. (but I take your point), as always , thank you for the quick review @HazardyKnusperkeks CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69764/new/ https://reviews.llvm.org/D69764 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
