Quuxplusone added a comment. In D113898#3140320 <https://reviews.llvm.org/D113898#3140320>, @kuhnel wrote:
> When trying to revert some of the changes, I just noticed there are a couple > of `if (const auto* ...` in `CodeComplete.cpp` and `AST.cpp` (and maybe other > files as well). So some folks seem to be using this right now. If we want to > be consistent, we would have to remove these `const` as well. Just in case it's news to anyone: (1) Theoretically, the compiler has enough information to tell you when it's deducing a cv-qualified type for a non-forwarding-reference `auto`; and personally I'd like such a warning and would go fix my code every time it fired. I like that `const auto *p` has the same "shape" as `const Widget *p`. That sometimes a non-const `auto` can secretly mean //`const`//`Widget` is //weird// and I don't like it. (2) In template contexts, `const auto *` can be significant, because maybe in some instantiations the `const` doesn't matter and in others it does. https://quuxplusone.github.io/blog/2020/03/04/field-report-on-lifetime-extension/#true-positive-3-conditionally-redundant-lifetime-extension-in-template-code is related. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D113898/new/ https://reviews.llvm.org/D113898 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
