RTTI data member names

[Mark Mitchell]

>>>>> "Jim" == Jim Dehnert <dehnert at baalbek.engr.sgi.com> writes:

    Jim> is very mild, and I'd rather not decide it.  Mark, has your
    Jim> opinion shifted during the discussion? 

I don't feel terribly strongly, but I still feel that it would be
better to give these fields names that start with underscores.  Why
risk a conflict, where we needn't have one?  

Furthermore, I've got to say that at this point we're becoming
frustrated with changes to the ABI that aren't absolutely essential.
The current state is correct.  Every such change means that we'll have
to go through and change two implementations of the library, three
compiler implementations, the test suite, and so forth.

Quite frankly, we've done a lot of work (by implementing the ABI) to
flush out issues, and we feel that the existence of these
implementations ought to be considered at this point.  In other words,
we should transition from the mode of designing an ABI to the mode of
finishing those bits that are still unspecified, clearing up
ambiguities in the document, and so forth.  We now have existing
practice, and we should standardize on that, unless there is a
compelling reason to do otherwise.

My two cents,

--
Mark Mitchell                   mark at codesourcery.com
CodeSourcery, LLC               http://www.codesourcery.com