[PATCH] D124286: [modules] Allow parsing a duplicate Obj-C interface if a previous one comes from a hidden [sub]module.

Jan Svoboda via Phabricator via cfe-commits cfe-commits at lists.llvm.org
Tue Apr 26 04:47:27 PDT 2022


jansvoboda11 added a comment.

This looks good IMO. Let's see if others have more suggestions besides the `Parser`/`Sema` layering.



================
Comment at: clang/lib/Parse/ParseObjc.cpp:16
 #include "clang/Basic/CharInfo.h"
+#include "clang/Basic/DiagnosticSema.h"
 #include "clang/Basic/TargetInfo.h"
----------------
vsapsai wrote:
> I'd like to get feedback on Parser/Sema layering. I'll check myself what can be done but at the first glance accessing pieces of Sema from Parser looks questionable.
Yeah, I think introducing new `Sema::ActOnEndClassInterface()` and moving the diagnostic logic there might be cleaner.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D124286/new/

https://reviews.llvm.org/D124286



More information about the cfe-commits mailing list