This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
The comments in test K2-NameTest-71, and the expected result, reflect a belief that an attribute node should not match the SequenceType attribute(a, xs:anyType). I think this is incorrect. xs:anyType is the root of the schema type hierarchy so all attributes (even untyped attributes) will match it.
Yes, should be fixed in CVS. (Seems we had a private email thread on this, with subject "attribute(foo, xs:anyType". I think I confused the type of an attribute with the type of its value.)