Updated status pages

[Nathan Sidwell]

Jim Dehnert wrote:
> Nathan, have you had a chance to look at the RTTI structure
> definitions? And Alex was going to go over the mangling examples.
finally had some time to do that. A few more nits -- pretty tiny in
the big scheme, but we may as well try and get consistency!

1) 2.9.5/6.c The definition of __vmi_class_type_info contains
the base array definition as,
        __base_type_info const base_info[1];
That should be __base_class_type_info. Why is it const qualified?
The other data members are not, and this is not a pointer to a const
object, which would justify a qualification. I beleive that qualifier
should be removed.

2) 2.9.5/6.b has the base pointer called `base_type' 2.9.5/6.c has
__base_class_info's base pointer called `base'. These should have the
same name.

3) 2.9.5/6.c has a `flags' member of __vmi_class_type_info but
an `offset_flags' member of __base_class_info. Again, these should have
the same name, (and so should their associated enumerations).

4) 2.9.5/6.c. The enumeration should include the bit shift to obtain the
offset. I recommend,
        offset_shift = 8

5) 2.9.5/7. The __pbase_type_info has a `qualifier_flags'
member, which now contains more than just type qualifiers. In line
with __vmi_class_type_info, I suggest renaming to simply `flags'.

6) 2.9.5/7 has a `pointee' member, but the text describing the qualifier
flag meanings uses the word `target'. These should be consistent (I have
no preference).

7) 2.9.5/9 The `context_class' member should be a const qualified pointer
        const __class_type_info *context_class;
Maybe shortening the name to simply `context' would be less verbose too.

nathan

-- 
Dr Nathan Sidwell   ::   http://www.codesourcery.com   ::   CodeSourcery LLC
         'But that's a lie.' - 'Yes it is. What's your point?'
nathan at codesourcery.com : http://www.cs.bris.ac.uk/~nathan/ : nathan at acm.org