agourakis82 wrote: Thanks for the fix! One question: the test expects a second diagnostic, "cannot compile this scalar expression yet", which comes from CodeGen. Doesn't that mean the RecoveryExpr still reaches CodeGen after the Sema error? I'd have expected CodeGen not to run at all here. Would a `-fsyntax-only` test in SemaObjC (plus checking that no CodeGen diagnostic is emitted) be a better fit?
https://github.com/llvm/llvm-project/pull/224849 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
