RE: Schema-aware processors - W3C schemas only?

#     Michael> What's more, the conformance rules are written in an
#     Michael> "all-or-nothing" style.  You either support all these
#     Michael> schema-aware features (import-schema, validate, node
#     Michael> annotations) or you support none of them. The XSL WG
#     Michael> deliberately tried to avoid having products that
#     Michael> cherry-picked the individual features they wanted to
#     Michael> implement.
# 
# You mean like Saxon 7.8 does? :-)

Yep. Saxon 7.8 is work-in-progress, it couldn't claim conformance in its
current state.

# 
# I refer to the annotating of attributes as xs:IDREF, which is
# specifically disallowed by the standard for Basic XSLT processors, and
# seems to make it impossible for a Basic XSLT processor to be fully
# backwards compatible with XSLT 1.0.
# 
# Which I find objectionable.

We've got public comments on the handling of DTD types which we need to
address. However, I don't see the problem with xs:IDREF: there was nothing
in XSLT 1.0 that handled an IDREF specially, only an ID. Could you clarify?

And don't hesitate to ask for clarifications on the public list. The WGs are
now entitled to ignore the comments, but are unlikely to do so if they make
sense.

Michael Kay



 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list