RE: HTML forms with XSLT slideshow question

[Kay Michael <Michael.Kay@xxxxxxx>]

> I love the idea of extension functions, but they are unusable 
> as is to me.
> While I have some SAXON specific XSLT stylesheets, my norm is parser
> independence.  With extension functions parser independence 
> is not possible.

There does seem to be a need for a standard here: perhaps an informal
standard created in the same way as SAX. Any volunteers to propose one and
run the process?

It seems such a standard would have to address:
- how the Java class and method names relate to the XPath function name and
namespace
- the mapping of XPath data types to Java classes (e.g. should a NodeSet be
provided as an Enumeration of DOM Node objects?)
- how to provide context information (this has been my main reservation in
adopting xt's style of interface: it's fine for "pure" functions but doesn't
extend easily to functions that use context, e.g. the current node or the
system id of the containing stylesheet node. In my view if the API isn't
rich enough to implement the system functions defined in XSLT, then it isn't
rich enough for user-defined functions either).

Mike Kay


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