[cfe-dev] Type::isIncompleteType()...

[Chris Lattner]

On Feb 11, 2008, at 1:49 AM, Neil Booth wrote:
>> That said, Steve's idea might still be a good one.  In what other
>> cases can a struct decl be marked erroneous?
>
> Actually my cfe does what I said second above: it has foo as a
> member that is flagged erroneous.  I put a lot of effort into
> "clean" recovery; as far as possible I wanted something erroneous
> to only cause a single complaint.

Right, I'm sorry I wasn't clear, but that's what I meant.  In clang we  
try really hard to never create invalid ASTs, even for invalid code.   
To mark the field decl invalid, we first have to force the type to  
something valid, create the decl, then mark the decl as invalid.  I  
completely agree with you that this is the right thing to do in this  
case.

-Chris