XInclude is Rolling Along

  • Article

I was working through the XInclude class hierarchy today, and ran across some interesting questions.

  • Should there be an XIncludeNodeList implementation that is the product of the merged Infosets?  Or is this irrelevant since it would only apply to a fully-loaded DOM instance, which should already have been loaded?
  • Should loading an XIncludeDocument be in any part asynchronous?
  • Should the first version of XIncludeReader support XPointer?  If so, to what degree?  Should we only support the XPointer elements() scheme? 
  • XPointer itself does not seem recursive as XPath is (there is no reverse axis to contend with, only ranges with a defined start and end relative to the document root).  Is supporting a fully compliant XPointer syntax within the scope of XInclude?

I guess this all boils down to answering "How complete should revision 1 be?" 

Answers to these questions will have a direct bearing on another really cool project I have been working on... details (and hopefully downloadable source) within the week.