Internal subset equivalent in new schema proposals?

[roddey at us.ibm.com]

Hmmm... This is a totally "unencumbered by the thought process" suggestion,
but perhaps something like this could deal with a lot of issues that are
arising in this and related areas.

<XSC:Packet>
    <XSC:Schema>
        schema goes here....
   </XSC:Schema>

   <XSC:Data>
        XML data goes here
   </XSC:Data>
</XSC:Packet>

It would be a way of packing a schema and its data in a single file. Since
its a known format, the application can understand what is happening and
tear it apart and feed it back into the parser. Or, if its an official
concept, the parser itself can understand what is happening and make it all
automagical to a certain degree. Seems like it could be applicable in a
number of ways, though as I said before I pondered it all of 15 seconds
before saying this so it could be totally useless :-)

An extension of this packet would look something like this:

<XSC:Packet>
    <XSC:InternalEntities>
        define internal entities here
    </XSC:InternalEntities>

    <XSC:Schema>
        schema goes here....
   </XSC:Schema>

   <XSC:Data>
        XML data goes here
   </XSC:Data>
</XSC:Packet>

where an internal entities definition section could be provided, and which
must be first, so that entities could be preloaded into the entity pool in
order to affect the upcoming parsing of the schema file? If this were a
known format, it wouldn't be too hard for the parser to feed its own entity
pool and leave it populated for the next step. It would look just like they
were defined via a DTD and all of the same code would be used to handle the
entity expansion and reference replacement, right?

Just a thought anyway...

>>Instead of an internal subset B of an external schema A, would it be
better
>>to make B also an external schema "inheriting" A in some way?
>>(For general processing, stylesheets, reuse etc, wouldn't it be best to
>>externalize as much as possible?)
>
>Interesting question.  I just found out that XSchema does allow inline
>XSchema elements (5.1.2) but I am not sure if it supports schema
>inheritance.
>
>Actually, my questions were more or less regarding existing support.  DCD,
>XML-Data, and RDF does not seem to support inline schema although they do
>allow declaration of datatypes in data instances.
>
>I think inline schema is particularly important in applications like
digital
>television and smart cable boxes where content variation is tremendous.
It
>is possible to do this through multiplexing/tunneling but it would be nice
>to have a single stream of XML data which can handle open variety of
>contents (drama, shows, program schedule, advertisement, closed captions,
>etc.)



xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)