Re: Datatype test cases: important ones (please have a look)

At 14:33 20/11/2002 +0000, Brian McBride wrote:

[...]

>If we do decide to keep it (grrrr) then I'd suggest formulating it 
>differently.  The way this is put suggests that
>
>  <a> <b> "10"^^xsd:integer .
>
>entails all other datatype representations of the same value.  Suggest 
>instead:
>
>  <a> <b> "10"^^xsd:integer .
>  <c> <d> "1E1"^^xsd:decimal .
>
>entails
>
>  <a> <b> _:v .
>  <c> <d> _:v .
>
>which is more constrained.


I just realized this applies to some of the earlier tests as well, e.g.

  <a> <b> "10"^^xsd:integer .
  <c> <d> "010"^^xsd:integer .

entails

  <a> <b> _:v .
  <c> <d> _:v .

Brian

Received on Wednesday, 20 November 2002 10:40:50 UTC