Re: XInclude implementation experience

On Fri, 2002-06-14 at 10:57, Paul Grosso wrote:
> At 10:25 2002 06 14 -0600, Uche Ogbuji wrote:
> >On Mon, 2002-06-10 at 10:13, Paul Grosso wrote:
> >> At 15:28 2002 05 02 -0500, Paul Grosso wrote:
> >> >Uche,
> >> >
> >> >Though I'm not subscribed to xml-dev any more (it got to
> >> >be too much for me about a year ago), I've been forwarded
> >> >an April 29th message you sent there about XInclude (it 
> >> >appears that the OASIS archives don't include it).
> >> >
> >> >Anyway, the purpose of my message is to ask you to send
> >> >a message with any implementation experience you may have
> >> >(comments, problems, and levels of implementation) to the
> >> >XInclude comment list [1].  The W3C XML Core WG is going
> >> >to be working over the next couple months to figure out
> >> >exactly how to make further progress on XInclude, and we'd
> >> >like to have a record of feedback for all implementations
> >> >out there.
> >
> >I'm sorry.  I've been hoping we can get 4Suite 0.12.0 to beta, so I can
> >base my comments on a beta rather than alpha implementation.
> >
> >I'll try to write up general comments this weekend.  Then you can decide
> >whether to have me refine them when it goes beta or full 0.12.0 release.
> 
> 
> Thanks.
> 
> We are more interested in feedback you might have on the spec
> than on the exact state of your implementation.  Were there parts
> you didn't implement, were there parts you thought shouldn't
> have been in the spec (and why), were there parts of the spec
> that were unclear when you went to implement it?  That sort
> of thing.

OK.  Finally have time to do this some justice.

First of all, to the main 2 questions you pose:

   1. Would a subset of XPointer simplfy XInclude implementation? Which
features should be available in this subset?

Yes.  I would start with simple child axis + predicates in absolute
location paths.  Perhaps more could be included, but this is what we
implemented, because more was too much of a burden.


   2. Would a subset of XPointer assist in building streaming XInclude
processors? Which features should be available in this subset?

Same answer as (1)


We implemented XInclude in 4Suite (http://4Suite.org) as an optional
part of the process of parsing XML.  XInclude processing is the default,
which has been controversial among some observers, but not any users. 
We've updated the support in 4Suite 0.12.0, currently in alpha test. 
These comments are applicable to this version.

We support the xi:include element with href, parse and encoding
attributes.  We do not implement xi:fallback, though we may add support
for this.  All children of xi:include are silently ignored.  We
approximate the descriptions of infosets in our DOM-like internal XML
data model.  We do also support xml:base.


-- 
Uche Ogbuji                                    Fourthought, Inc.
http://uche.ogbuji.net    http://4Suite.org    http://fourthought.com
Track chair, XML/Web Services One (San Jose, Boston):
http://www.xmlconference.com/
DAML Reference - http://www.xml.com/pub/a/2002/05/01/damlref.html
The Languages of the Semantic Web -
http://www.newarchitectmag.com/documents/s=2453/new1020218556549/index.html
XML, The Model Driven Architecture, and RDF @ XML Europe -
http://www.xmleurope.com/2002/kttrack.asp#themodel

Received on Wednesday, 26 June 2002 02:11:53 UTC