[PATCH] D88445: Use "default member initializer" instead of "in-class initializer" for diagnostics

[Richard Smith - zygoloid via Phabricator via cfe-commits]

rsmith added a comment.

Thanks!



================
Comment at: clang/include/clang/Basic/DiagnosticParseKinds.td:864-867
+  "default member initialization of non-static data member is a C++11 "
+  "extension">, InGroup<CXX11>;
 def warn_cxx98_compat_nonstatic_member_init : Warning<
+  "default member initialization of non-static data members is incompatible "
----------------
Consider "initialization of" -> "initializer for", so we use the standard term in full.


================
Comment at: clang/include/clang/Basic/DiagnosticSemaKinds.td:8506-8524
+def err_default_member_initializer_non_const : Error<
   "non-const static data member must be initialized out of line">;
-def err_in_class_initializer_volatile : Error<
+def err_default_member_initializer_volatile : Error<
   "static const volatile data member must be initialized out of line">;
-def err_in_class_initializer_bad_type : Error<
+def err_default_member_initializer_bad_type : Error<
   "static data member of type %0 must be initialized out of line">;
+def ext_default_member_initializer_float_type : ExtWarn<
----------------
For a static member, it's just an initializer, not a default member initializer. I think "in-class initializer" is probably the best we can say here.


================
Comment at: clang/include/clang/Basic/DiagnosticSemaKinds.td:8533-8536
+def ext_default_member_initializer_non_constant : Extension<
+  "default member initializer for static data member is not a constant "
+  "expression; folding it to a constant is a GNU extension">,
+  InGroup<GNUFoldingConstant>;
----------------
Likewise, this case is an initializer for a static member, and is not a default member initializer.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D88445/new/

https://reviews.llvm.org/D88445