On Feb 12, 2009, at 12:47 PM, Chris Lattner wrote:
>
> On Feb 12, 2009, at 7:55 AM, Steve Naroff wrote:
>
>> +Sema::ActOnObjCAtThrowStmt(SourceLocation AtLoc, ExprArg
>> expr,Scope *CurScope) {
>> Expr *ThrowExpr = static_cast<Expr*>(expr.release());
>> if (!ThrowExpr) {
>> // @throw without an expression designates a rethrow (which much
>> occur
>> @@ -998,8 +997,7 @@
>> if (!Context.isObjCObjectPointerType(ThrowType)) {
>> const PointerType *PT = ThrowType->getAsPointerType();
>> if (!PT || !PT->getPointeeType()->isVoidType())
>> - // This should be an error, however GCC only yields a
>> warning.
>> - Diag(AtLoc, diag::warn_objc_throw_expects_object)
>> + Diag(AtLoc, diag::error_objc_throw_expects_object)
>> << ThrowExpr->getType() << ThrowExpr-
>> >getSourceRange();
>> }
>>
>
> I think this should "return true;" after the error so that we don't
> create and return an invalid AST.
>
Oops. Good catch - thanks,
snaroff
> -Chris
_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
