=?utf-8?q?André?= Brand <andre.br...@mailbox.org> Message-ID: In-Reply-To: <llvm.org/llvm/llvm-project/pull/121...@github.com>
erichkeane wrote: > @erichkeane I am not sure what you mean. As I pointed out earlier, the > examples @cor3ntin brought up actually conform to the standard as in > [[temp.spec.general]](https://wg21.link/temp.spec#temp.inst-3): > > > The implicit instantiation of a class template specialization causes the > > implicit instantiation of the declarations, but not of the definitions of > > [...] scoped member enumerations, [...] > > The issues are correctly diagnosed by clang if the struct is _explicitly > instantiated_ as demonstrated here in Compiler Explorer: > https://godbolt.org/z/GzYq9jzjo. My quick look through the ActOnEnum and this function looked like it showed some misses. If Corentin is comfortable with our checking here, then I'm happy too. > > Nevertheless, I investigated further and managed to discover some real bugs > with diagnostics on template instantiation. These affect not only enums but > also nested classes. If you are interested, you can have a look: > https://godbolt.org/z/vKG6vqWhs. I have already started working on these > issues but so far they seem unrelated. Thats great, thanks for finding that and working on it! Those look like some pretty unfortunate misses. > > I plan to keep working on it but I can't promise much given the facts that I > only have the weekends to spend time on it and my limited familiarity with > the codebase (this would be my very first commit to LLVM). But, of course, I > promise to try my best! đŸ™‚ Definitely appreciate it! I'm happy when Corentin is here, I have no concerns. https://github.com/llvm/llvm-project/pull/121039 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
