the datamodel in the rwapi dir has two purposes: the first one is to
represent XPath expressions such as "5 instance of xs:integer" in a more
accurate way than just representing xs:integer as a QName (but representing
it as an AtomicType in this case).
The second purpose is to represent xpath results in term of its datamodel
as defined in the XQuery and XPath data model workking draft (and so
independently of any internal source document models, DOM, DTM, XDM,
etc..). For now, I don't know what is the best way to do it between 1)
defining a bunch of new interfaces implementing the XPath data model and
providing one evaluator or 2) providing as much evaluators than existing
source document models. Need more neurons on it....
Lionel
Joseph
Kesselman/Watson/ To: [EMAIL PROTECTED]
[EMAIL PROTECTED] cc: [EMAIL PROTECTED]
Subject: Re: [xslt20][rwapi]
Refactoring base 'Node' classes - comments?
03/07/2003 03:01
PM
Please respond to
xalan-dev
>org.apache.xpath.rwapi.datamodel.Node
I haven't been keeping up with the rwapi stuff, but... Which data model
are we talking about here? Source documents? If so, there's a question
here about whether you want to return your own flavor of Node, or (if we
decide this proposal makes sense) a Cursor (which may proxy more than one
node), or (if these are values to be returned to the user) perhaps a proxy
which honors the DOM Node API...
______________________________________
Joe Kesselman, IBM Next-Generation Web Technologies: XML, XSL and more.
"may'ron DaroQbe'chugh vaj bIrIQbej" ("Put down the squeezebox and nobody
gets hurt.")
