RE: [xml-dev] xs:assert and Schematron

> That is what the status comments says.
> 
> But the normative text does not reflect that.

Thanks for pointing this out. To be honest, I'm not sure what the current WG
decision is on this one: it's been a point of heated debate where some WG
members have wanted a small subset of XPath and others (including me) have
wanted the whole caboodle. The current unpublished status quo text still has
the inconsistency and I will raise an issue on it.

The Saxon 9.1 implementation allows the whole of XPath 2.0 both for
assertions and for conditional type assignment. The spec will definitely
allow that, and it's my hope that users will demand it, in which case the
subset will become irrelevant (just as the subsetting of the XPath axes in
XQuery has become irrelevant.)

Michael Kay
http://www.saxonica.com/