> Perhaps the expression "single step" was misleading, as what 
> I meant was that the validation of the input against a W3C 
> XML Schema can be done during the XSLT/XQuery processing 
> using a Schema-aware processor, so that you reduce the number 
> of individual steps in the validation process by one.

If you ask Saxon-SA to validate an input document to a transformation, then
you get a push pipeline that feeds events from the XML parser through the
schema validator to the Saxon tree builder. Apart from the crucial fact that
the type information generated by the validator is retained by the tree
builder, this is very similar to what you would get if you put together your
own SAX-based pipeline in JAXP or XProc, which you can do of course without
a schema-aware XSLT processor.

So it all depends what you mean by "steps" - separate steps from the
application perspective, or separate steps within the internal pipeline?

Internally, Saxon's input pipeline is typically much more complicated, it
can involve steps that strip whitespace, that combine adjacent text nodes,
that expand XInclude, that locate the subtree rooted at a particular ID
value, that maintain the namespace context, etc. And the "step" in this
pipeline that does schema validation is itself a push pipeline whose
internal complexity is a wonder to behold... 

Michael Kay