RE: [xml-dev] The <any/> element: bane of security or savior ofversionin

[Rick Jelliffe <rjelliffe@allette.com.au>]

On Mon, 2007-10-22 at 12:52 -0400, G. Ken Holman wrote:
> At 2007-10-22 12:41 -0400, Costello, Roger L. wrote:
> >Rick Jelliffe: when is Schematron going to have the ability to do
> >datatype assertions, e.g. "The value of the <Value> element is of
> >datatype xs:nonNegativeInteger"?
> 
> If you use an XPath 2.0 binding to Schematron you could say:
> 
>    <rule context="X">
>      <assert test=". castable as xs:nonNegativeInteger">
>        The value of the X element is of datatype xs:nonNegativeInteger
>      </assert>
>    </rule>

Indeed, this is what we are using for our XSD implementation in
Schematron. The rub is that the free version of SAXON, which is
excellent, only does castable for the built-in primitives, not the
built-in derived types, from what I am told.

I have recently decided to make the XSLT2-based implementation of
Schematron a separate script from the XSLT1-based implementation. Some
users had already forked a really nice XSLT2 version off, so it is
pretty much just a matter of adopting that. But it will allow the XSLT2
version and the XSLT1 version to have cleaner code. (Both versions will
continue generate code for XSLT1, EXSLT and XSLT2 systems.) See
www.schematron.com

In order to use XSLT2 expressions, you 

1) have @queryBinding="xslt2" on the sch:schema element, 
2) use the the latest version of the skeleton
3) use an implementation of XSLT2 (SAXON, Gestalt)

Cheers
Rick