[PATCH] D87534: Sema: introduce `__attribute__((__swift_name__))`

[Saleem Abdulrasool via Phabricator via cfe-commits]

compnerd marked 11 inline comments as done.
compnerd added inline comments.


================
Comment at: clang/include/clang/Basic/DiagnosticSemaKinds.td:3977
 
+// Swift attributes.
+def warn_attr_swift_name_function
----------------
aaron.ballman wrote:
> I want to make sure we're clear with the terminology used in the diagnostics, so there's a fair number of "take -> have" suggestions here, but I want to be sure I'm correct before you apply those suggestions.
> 
> I'm used to function declarations having parameters which take arguments from a function call expression. Are the diagnostics about "taking a parameter" talking about "having a parameter" or about "taking an argument"? I believe the string arguments are function signatures rather than function names (perhaps?) and so we should be talking about having a parameter, but I wasn't 100% sure.
Discussed offline.


================
Comment at: clang/include/clang/Basic/DiagnosticSemaKinds.td:3997
+def warn_attr_swift_name_setter_parameters
+  : Warning<"%0 attribute for setter must take one parameter for new value">,
+    InGroup<SwiftNameAttribute>;
----------------
aaron.ballman wrote:
> take -> have
> 
> elsewhere `new value` is spelled `'newValue:`, should that be the same here?
`newValue:` is the argument label spelling and "new value" is the argument value, which is the reason for the difference.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D87534/new/

https://reviews.llvm.org/D87534