I'm wondering whether it would be possible to unify character repertoire
constraints and datatype libraries. My idea (far from fully baked) is
something like this:
- if we can specify character repertoire constraints on <text/> and
<mixed/>, shouldn't we also be able to specify them on the string and token
built-in datatypes?
- which makes the constraint seem a bit like a datatype parameter..
- perhaps one datatype library needs to be able to specify parameters on a
datatype provided by another datatype library
- which could be naturally allowed in the compact syntax by making datatype
parameters be qnames
- could we allow a datatype library to specify parameters that can apply to
text?
> It is so natural to specify such constraints
> on character repertoires as part of grammartical schemars.
>
Agreed it's natural.
>
>
> > - There are a couple of cases where there may be subtleties with the
> > semantics: one is with white-space, and the other is with normalization
> (if
> > CREPDL supports this).
>
> CREPDL does not support normalization. I believe that CREPDL should not
> automatically
> normalize characters and that users should explicitly normalize XML if
> they really want to normalize. Normalization (even NFC and NFD) introduces
> surprising changes.
>
Good. That makes things much easier.
>
> Whitespace is indeed tricky. We might want to introduce some
> restrictions on the use of CREPDL from RELAX NG. But this is merely an
> issue rather than a show stopper.
>
Absolutely.
>
> > - This change should only be done as part of a new version of RELAX NG,
> > which also ought to address other issues, such as defining name classes
> and
> > modularity.
>
> Agreed. When should we start to create the next version of RELAX NG?
>
Not sure. I would like to see if we can get some consensus from the wider
community of RELAX NG users in favour of doing a new version.
And should it be started in OASIS and then moved to ISO or should it be done
exclusively in ISO?
James
-- DSDL comments To unsubscribe, please send a message with the command "unsubscribe" to dsdl-comment-request@dsdl.org (mailto:dsdl-comment-request@dsdl.org?Subject=unsubscribe)Received on Wed Jan 21 13:31:27 2009
This archive was generated by hypermail 2.1.8 : Thu Jan 22 2009 - 04:13:03 UTC