> I agree. I would expect some sort of mapping transform as well. It would
> be good to develop a convention for such a thing, perhaps as a derivation
> of Ucode? Could that be a useful micro-deliverable for this group?
>
There's a mapping defined here:
http://www.w3.org/TR/2003/REC-soap12-part2-20030624/#namemap
But if you want a mapping that maps arbitrary strings to valid names
and the mapping must not lose information, then the mapping cannot leave
all valid names as is. And if a mapping changes valid names, then it
shouldn't be done by the parser. So I think such a mapping is better done
in a post-parsing process. This is easy to do if the parser leaves names as
is.
James