lime added a comment.
In D134128#3798127 <https://reviews.llvm.org/D134128#3798127>, @cor3ntin wrote:
> Thanks for working on this.
> I'll be honest though, I still have absolutely no understanding what the use
> cases or intents of this features are. I think we were waiting for core to
> clarify and I'm not sure they did.
> This does seem to implement the wording though...
Perhaps, the intents of this feature are a little confusing, so I add `S5` in
the file `p3-2a.cpp`. If it was necessary to check the constains on the
template template parameter, we could expect an error there.
But one intent might be a mend of the tailing syntax about constains, as a
template parameter like `template <C> typename` is already accepted by Clang.
================
Comment at: clang/lib/Parse/ParseTemplate.cpp:882
/// 'typename' [C++1z]
-NamedDecl *
-Parser::ParseTemplateTemplateParameter(unsigned Depth, unsigned Position) {
+/// template-head: [C++2a]
+/// 'template' '<' template-parameter-list '>'
----------------
This is copied from `Parser::ParseTemplateDeclarationOrSpecialization`.
================
Comment at: clang/lib/Parse/ParseTemplate.cpp:909
+ }
+ }
}
----------------
It is fine to skip the check for the language option of C++20. The parser will
emit an error here and complain the lack of `class` or `typename`, if the
option is not provided.
================
Comment at: clang/test/CXX/temp/temp.arg/temp.arg.template/p3-2a.cpp:36
+
+template<template<typename T> requires C<T> class P> struct S4 { }; //
expected-note 2{{'P' declared here}}
+
----------------
Test whether Clang behaves the same here as it is on `S1`.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D134128/new/
https://reviews.llvm.org/D134128
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
