• To: "'Elliotte Rusty Harold'" <elharo@m...>, michael.h.kay@n..., xml-dev@l...
• Subject: RE: JAXP's ID Transform failing to include namespace declaration attributes
• From: Michael Brennan <Michael_Brennan@A...>
• Date: Mon, 7 Jan 2002 13:34:33 -0800

> From: Elliotte Rusty Harold [mailto:elharo@m...]

<snip/>

> I agree this is a major flaw of DOM. In fact, it's even possible to 
> produce DOM documents that not only could not have been parsed from 
> an XML document. It's even possible to create DOM documents that 
> cannot possibly be serialized to a well-formed XML document. 
> Unfortunately, we have to live with it. :-(

I'm not sure I'd agree it is a flaw of the DOM. I think it is a flaw of DTDs
and the screwed up processing model that prevails in the XML world. If the
tasks of parsing, validating, and adding PSVI constructs to the infoset were
clearly defined as separate processing layers, I think things would be much
cleaner. Imagine if you could create a Transformer from a DTD rather than an
XSLT stylesheet, feed a Source into it and get a Result that has been
validated and had PSVI constructs added rather than having this
functionality embedded in a black-box parser, wouldn't that provide a more
sensible and flexible processing model for XML?

Another reason why I like RELAX NG and hate DTDs.
 

• Follow-Ups:
  • RE: JAXP's ID Transform failing to include namespace de claration attributes
    • From: Elliotte Rusty Harold <elharo@m...>

• Prev by Date: RE: Unloading XML file from DOM
• Next by Date: tying attribute values to data types
• Previous by thread: RE: JAXP's ID Transform failing to include namespace declaration attributes
• Next by thread: RE: JAXP's ID Transform failing to include namespace de claration attributes
• Index(es):
  • Date
  • Thread