TEST: datatypes and cardinality

(Mainly for Ian)
At the telecon last night we suggested that finite datatypes were somehow 
harder than infinite ones.
I tried building a test case, but it ended up in OWL Full - I filed it under 
issue 5.1
http://www.w3.org/2002/03owlt/editors-draft/snapshot#proposedIssue-I5.1-Uniform-treatment-of-literal-data-values

To get the finiteness of the datatype to interact with the domain of discourse 
I seemed to require an InverseFunctional DatatypeProperty.

DatatypeProperty( p, range(xsd:byte), InverseFunctional )
ObjectProperty( q, inverse(invQ) )
Individual( spy, type( restriction( invQ, cardinality=127 ) ) )
EquivalentClasses(
   Restriction( p, someValuesFrom(xsd:unsignedInt) )
   Restriction( p, hasValue( spy ) )
)

(This is consistent, changing the 127 to 128 is not).

Can anyone show me how to get a similar affect in OWL DL?
(I might need to move this test into the extra credit section)

Jeremy