================ @@ -1141,6 +1141,14 @@ void Sema::ActOnLambdaExpressionAfterIntroducer(LambdaIntroducer &Intro, // Determine if we're within a context where we know that the lambda will // be dependent, because there are template parameters in scope. + // + // FIXME: A lambda in the noexcept specifier of an abbreviated function + // template (e.g. `auto&&`) is not marked dependent, even though it captures + // a dependent entity. Could this be because the implicit template parameter + // is not in scope while the specifier is parsed, leaving the lambda's + // dependency kind as LDK_Unknown? See: + // * https://github.com/llvm/llvm-project/issues/232874 + // * https://github.com/llvm/llvm-project/issues/233102 ---------------- steakhal wrote:
I think a comment like this is pretty weak. I think a fix for this issue would deserve a unittest - which also implies that that unittest could also demonstrate the current - allegedly flawed behavior. That would make it a strong enough case that we could escalate to the Clang Area Team and get their opinion. https://github.com/llvm/llvm-project/pull/232986 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
