RE: Feedback on XInclude CR Version 1.0 21 February 2002

Responses below:

> -----Original Message-----
> From: Daniel Veillard [mailto:daniel@veillard.com]
> Sent: Wednesday, May 29, 2002 5:48 AM
> To: www-xml-xinclude-comments@w3.org
> Subject: Feedback on XInclude CR Version 1.0 21 February 2002
> 
> 
>   First I would like to report that libxml2 implementation is
> mostly complete and I got feedback it was used for documentation
> maintainance (DocBook based). The main interest was that contrary
> to entities it was possible to keep the included parts valid
> document instances.

Thanks for the report!

>   I think I got only one feedback from users reporting the use
> of XPointer expressions, they were actually xpath expression
> using the xpointer() scheme to select subset of the included document.
> I personally would prefer to keep the dependancy on the full XPointer
> spec but I can hardly justify it with existing use. It is clear that
> it makes the implementation more complex, I don't believe that the
> XInclude spec should itself define a subset, if such a need exist it
> is likely to be defined directly at the fragment identifier definition
> level for XML resources. There is a clear distinction too between the
> complexity of just implementing bare name identifier (pointing to the
> ID) which is an expected feature of such a fragment identifier and
> the full XPointer selection capabilities, I tend to think that if a
> level can be drawn in XInclude XPointer support it's the most natural
> place to make the cut.

The XML Core WG has received a number of comments on XPointer support in
XInclude, most suggesting some sort of XPointer subset.  The recent
factoring of XPointer has given the WG the opportunity to simplify
XInclude implementation without creating our own subset of XPointer. The
WG has decided to require support for the XPointer Framework [1], and
the XPointer element() Scheme [2].  Support for the XPointer
xpointer() Scheme [3] and other schemes developed in the future would be
optional.

[1] http://www.w3.org/TR/xptr-framework/
[2] http://www.w3.org/TR/xptr-element/
[3] http://www.w3.org/TR/xptr-xpointer/

>   Libxml2 recent versions implement ID/IDREF fixups, this was actually
> asked by one user of the XInclude support.
>   Concerning Namespace Fixup, it seems more reasonable to do that
> fixup when the resulting document is to be serialized back (if needed)
> after XInclude processing, this follows the guidelines of XSLT
processing
> and allow implementors to keep the specificities of their namespace
node
> implementation (libxml2 does not represent them in the document tree).

We have removed mandatory namespace fixups.  We warn users that
[namespace attributes] is not guaranteed to be reliable after inclusion,
and therefore recommend against it.  We still allow
implementation-defined fix-ups where necessary.

>   Basically I think XInclude should proceed to Proposed Recommendation
> status with minimal changes,
> 
> Daniel
> 
> --
> Daniel Veillard      | Red Hat Network
http://redhat.com/products/network/
> veillard@redhat.com  | libxml Gnome XML XSLT toolkit
http://xmlsoft.org/
> http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/

Received on Wednesday, 17 July 2002 16:49:33 UTC