vingeldal added a comment. In D69560#1935071 <https://reviews.llvm.org/D69560#1935071>, @whisperity wrote:
> @aaron.ballman I've gone over LLVM (and a few other projects). Some general > observations: > > - Length of `2` **is vile**. I understand that the C++CG rule says even > lengths of 2 should be matched, but that is industrially infeasible unless > one introduces such a rule incrementally to their project. Findings of length > 2 are **, in general,** an order of magnitude more than... basically the rest > of the findings. > - On big projects, even the current "default" of "length `3`" seems to be > too low. In reality, one should consider (this is likely to be out of scope > for Tidy) how often these functions are called, and various other metrics on > how serious an offender is. Not a problem since existing projects can just use the option and change to whatever level suits them best. My opinion is still that the default shouldn't be set to whatever we think is most useful for the majority of existing projects but to what reflects the actual guideline. Consider how this situation is similar to the guidelines regarding pointers. The guidelines assume that a project isn't using old standards of C++. Where one has a huge legacy code base it will be impractical at best to try to apply the C++ Core Guidelines for pointer handling on the old code. The expectation is that new projects will use the new libraries and language features available to do things differently, in a way that makes it possible and practical to follow the guidelines; while legacy code remains unchecked or incrementally improved. For any new code base this guideline shouldn't be a problem and existing projects can just adapt the usage of this rule to fit their situation by applying it selectively in the code base and/or changing the options. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69560/new/ https://reviews.llvm.org/D69560 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
