RE: How to: Returning HTML Elements From a Saxon Exten

You should probably be asking this question on the Saxon list rather
than here, but here goes.

This is actually a lot easier in 7.4 than in 6.5.2. In 7.4 there are two
very easy solutions:

(1) call XPathEvaluator, you can give this any JAXP Source object, and
it will give you back a NodeInfo which you can return from your
extension function

(2) return the XML string to the stylesheet and call saxon:parse to
parse it into a tree.

In 6.5.2 you have to probe a little bit deeper to find the right
interfaces. I think the simplest way (because it involves least use of
Saxon-specific methods) is to do a JAXP identity transformation from any
Source object to a DOMResult, where the DOMResult is primed using
setNode() on a document node created using new
com.icl.saxon.TinyDocumentImpl(); you can then return this document node
as the result of your extension function.

Happy Easter,

Michael Kay
Software AG
home: Michael.H.Kay@xxxxxxxxxxxx
work: Michael.Kay@xxxxxxxxxxxxxx 

> -----Original Message-----
> From: owner-xsl-list@xxxxxxxxxxxxxxxxxxxxxx 
> [mailto:owner-xsl-list@xxxxxxxxxxxxxxxxxxxxxx] On Behalf Of 
> W. Eliot Kimber
> Sent: 17 April 2003 19:39
> To: xsl-list
> Subject:  How to: Returning HTML Elements From a Saxon 
> Extension Function (6.5.x)
> 
> 
> I want to make sure I'm thinking about a small XSLT puzzle 
> correctly: my 
> requirement is to provide a Saxon extension function that returns a 
> restult tree (or a string that will become a result tree). What I'm 
> doing is building a navigation bar based on configuration information 
> held in an external file and accessed through an external 
> Java library.
> 
> If I just return the markup as a string (e.g., return "<a 
> href="#foo">bar</a>";) then it gets escaped, as you would expect.
> 
> Looking through the docs for extension functions, it appears 
> that I need 
> to return a NodeInfo node that contains nodes for the markup. 
> NodeInfo 
> is a Saxon-specific DOM implementation. The Saxon DOM 
> implementation is 
> read-only, and if I'm understanding the docs correctly, I 
> have to use a 
> SAX process to construct new nodes.
> 
> This leads to two questions:
> 
> 1. Is my analysis correct or have I missed some easy way to 
> return this 
> markup?
> 
> 2. If my analysis is correct, does anyone have an example of 
> doing this 
> sort of off-the-cuff SAX processing I could crib from? I 
> didn't find any 
>   examples in the Saxon-supplied samples.
> 
> Thanks,
> 
> Eliot
> -- 
> W. Eliot Kimber, eliot@xxxxxxxxxx
> Consultant, ISOGEN International
> 
> 1016 La Posada Dr., Suite 240
> Austin, TX  78752 Phone: 512.656.4139
> 
> 
>  XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list
> 


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