On Fri, 17 Sep 2010, John Peterson wrote:
> On Fri, Sep 17, 2010 at 9:58 AM, Roy Stogner <roystgnr@...> wrote:
>> On Fri, 17 Sep 2010, Ming Q wrote:
>>> I saw the same problem of dynamic casting. I have to work around by using GNU compiler instead.
>> One stupid question and one serious one, for everybody:
>> Is there any place in XCode where they're turning on -fno-rtti by
>> default, and if so can that be user-disabled?
>> Should we go through the library and change the rest of our
>> dynamic_cast operations to libmesh_cast_*?
> I don't think we should change all of them! There are cases in the
> numerics classes where the intent is to downcast to a more derived
> type, e.g. a PetscVector. You'd have to look at each case by hand,
> which sounds like a pain.
The downcasts are some of the ones I'd already changed: casting a
NumericVector-which-we-know-is-a-PetscVector into a PetscVector is
exactly the sort of scenario where libmesh_cast_* is useful. And
checking less than 50 cases by hand won't be any more of a pain than
the namespacing was. The only downside is removing some more error
checks from opt mode, but we already have a lot of checks that aren't