cor3ntin wrote: In a follow up, the following suggestion was made
> my compromise suggestion for [CWG2020] was that we could, without changing > name lookup itself, say that `this->BaseT<` (wherein we are interpreting > this-> as referring to the current instantiation, missing wording > notwithstanding) was interpreted as beginning a template argument list if > lookup for `BaseT` finds nothing in the current class and there is a direct > dependent base class that (resolving only trivial alias templates, if any) is > a specialization of a class template named `BaseT`. (Note that we might > reject during instantiation thanks to ambiguity.) This seems worth exploring to me https://github.com/llvm/llvm-project/pull/100425 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
