[PATCH] D101572: Make `hasTypeLoc` matcher support more node types.

Weston Carvalho via Phabricator via cfe-commits cfe-commits at lists.llvm.org
Wed May 5 15:53:56 PDT 2021


SilensAngelusNex added a comment.

Here's my branch on Github: https://github.com/SilensAngelusNex/llvm-project/tree/has-type-info



================
Comment at: clang/include/clang/ASTMatchers/ASTMatchersInternal.h:138-150
+template <typename... T> struct disjunction;
+template <typename T> struct disjunction<T> : public T {};
+template <typename Head, typename... Tail> struct disjunction<Head, Tail...> {
+  using type =
+      typename std::conditional<Head::value, Head, disjunction<Tail...>>::type;
+  static constexpr bool value = type::value;
+};
----------------
steveire wrote:
> SilensAngelusNex wrote:
> > Is there a better way to express this? I was originally using
> > 
> > ```
> > template <typename... Expected> struct is_one_of {
> >   template <typename T>
> >   static constexpr bool value = (std::is_base_of_v<Expected, T> || ...);
> > };
> > ```
> > but that didn't compile because `is_base_of_v` and fold expressions are C++17 features.
> > 
> > Maybe it would be better to just explicitly write out an overload of `GetTypeSourceInfo` for each type?
> 
> I like your c++17 solution too, but given that we have `TypeListContainsSuperOf` in this file used for this kind of thing, I think it makes sense to use that. I think this will work:
> 
> ```
> template <typename T,
>           std::enable_if_t<ast_matchers::internal::TypeListContainsSuperOf<
>             ast_matchers::internal::TypeList<
>               CXXBaseSpecifier, CXXCtorInitializer, CXXTemporaryObjectExpr,
>               CXXUnresolvedConstructExpr, CompoundLiteralExpr, DeclaratorDecl,
>               ObjCPropertyDecl, TemplateArgumentLoc, TypedefNameDecl>, T>::value>
>               * = nullptr>
> ```
> 
Perfect, thanks!


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D101572/new/

https://reviews.llvm.org/D101572



More information about the cfe-commits mailing list