arphaman added inline comments.
================
Comment at: lib/Sema/SemaStmt.cpp:1165
if (!hasCasesNotInSwitch)
SS->setAllEnumCasesCovered();
----------------
ahatanak wrote:
> This function used to call setAllEnumCasesCovered() when parsing a switch
> statement with an opaque enum condition, but this patch changes the behavior.
> I couldn't tell whether isAllEnumCasesCovered should return true or false for
> such switch statements, but was this change intentional?
You're right, good catch! Originally it wasn't intentional, but I've looked
over this change and it seems like the right change to me. We can't know if we
actually covered all enum cases when checking an opaque enum because we don't
know which cases the enum has (unless we have a switch that covers all the of
possible values of the enum's integer type)! It seems that the previous
behaviour was incorrect.
Repository:
rL LLVM
https://reviews.llvm.org/D27299
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
