extensions: creating nodes for return from custom extension elements

[Simon Kitching <si...@ecnetwork.co.nz>]

Hi All,

I need to write an extension element which returns a node with various
children, attributes, etc.

Does anyone know how I create nodes in my extension code? None of the
examples or standard Xalan extensions appear to do this, ie I can't find
any code to steal.

Normally, nodes are created via factory methods on the Document object
to which they are later going to be attached. However I can't see any
way to get at the Document object associated with the Result object to
which the results of the transformation are being written. 

Looking at the XSLProcessorContext.outputToResultTree method, it looks
like the nodes that are returned from the extension are actually
expected to reside in the source document (ie are cloned when output to
the result), but I can't see any way to create new nodes in the source
document either.

The ElemExtensionCall object passed to the extension method inherits
from UnImplNode, which has a Node interface, but whose methods (like
getOwnerDocument) are not very useful :-).

I'm probably missing something very obvious....

Thanks,

Simon

Re: XPath 2.0

[Joseph Kesselman <ke...@us.ibm.com>]

There is an experimental and still-incomplete version of XPath/XSLT 2.0 
checked into CVS on Xalan's XPATH20 branch. I wouldn't really recommend 
trying to use it yet unless you want to get involved in the development 
process.

When it's ready for more general use, we'll make an announcement.


______________________________________
Joe Kesselman  / IBM Research

XPath 2.0

[Richard Emberson <re...@outerharbor.com>]

Is there an implementation of XQuery1/XPath2 available?
What is the xalan ETA for XPath2 ?

Thanks

Richard

Re: extensions: creating nodes for return from custom extension elements

[Joseph Kesselman <ke...@us.ibm.com>]

Create the nodes using your favorite DOM implementation and pass those 
back. Xalan should recognize that they implement org.w3c.dom.Node and do 
the right things with them. They don't have to be in the same document; 
we're going to extract the data from them and rebuild it into the output 
document anyway.

There's an outstanding enhancement suggestion that it might be nice to be 
able to do something similar with SAX, but we haven't had time to look at 
it.

______________________________________
Joe Kesselman  / IBM Research