Re: do bad datatype literals denote [was Re: Datatype test cases ...]

this is just what we actually found for Jan's datatypes Manifest

<http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/semantic-equivalence-within-type-1>

P+
<http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/language-important-for-non-dt-entailment-2>

NP-
<http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/language-important-for-non-dt-entailment-1>

NP-
<http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/language-ignored-for-numeric-types-3>

P+
<http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/language-ignored-for-numeric-types-2>

P+
<http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/language-ignored-for-numeric-types-1>

P+
<http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/non-well-formed-literal-2>

NP-
** Input string was not in a correct format. in Datatype.Compare
(<http://www.w3.org/2001/XMLSchema#integer>, flargh, flargh)
<http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/non-well-formed-literal-1>

P+
<http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/semantic-equivalence-between-datatypes>

NP+
<http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/semantic-equivalence-within-type-2>

P+


a P+ means proof found for positive entailment test
a NP- means no proof found for negative entailment test (*)
a NP+ means no proof found for positive entailment test

so I think they're all OK (just that we can't find a proof
for the case semantic-equivalence-between-datatypes)

-- ,
Jos De Roo, AGFA http://www.agfa.com/w3c/jdroo/

(*) we actually are not trying to find a proof for a negative entailment
test

PS Jeremy, I had some trouble with Jena generated N3 triple objects such as
   " ... a "badly-formed" datatyped literal is a semantic error."
   so I had to escape the inner quotes as \"



                                                                                                                        
                    Jan Grant                                                                                           
                    <Jan.Grant@bristol.       To:     Brian McBride <bwm@hplb.hpl.hp.com>                               
                    ac.uk>                    cc:     pat hayes <phayes@ai.uwf.edu>, w3c-rdfcore-wg                     
                    Sent by:                   <w3c-rdfcore-wg@w3.org>                                                  
                    w3c-rdfcore-wg-requ       Subject:     Re: do bad datatype literals denote [was Re: Datatype test   
                    est@w3.org                 cases             ...]                                                   
                                                                                                                        
                                                                                                                        
                    2002-11-20 06:27 PM                                                                                 
                                                                                                                        
                                                                                                                        





On Wed, 20 Nov 2002, Brian McBride wrote:

>
> At 10:42 20/11/2002 -0600, pat hayes wrote:
>
> [...]
>
>
> >Oh, I thought that lang tags simply couldn't be attached to datatyped
> >literals other than rdf:XMLLiterals, so this would be a syntax error.
> >That's what the graph syntax rules seem to say. Is that wrong??
>
> Well, I don't think its what they do say.  I think its what they should
> say, but that is not what the WG agreed.
>
> [...]
>
>
> >It works but for a different reason. Perhaps I should spell this out
more
> >in the semantics doc.
> >
> >Making the denotation be something arbitrary in this case (ie not a
> >literal value, but otherwise it could be anything) means that the ONLY
> >entailment you can get is what you would get from basic graph
> >interpretations, which is replacing the bad literal by a new bnode:
>
> Hmmm, are you saying that it does not entail the emtpy graph.  I think I
> wasn't clear; I'm trying to clarify precisely the test case proposed by
> jan, non-well-formed-literal-2, in
>
>    http://www.w3.org/2000/10/rdf-tests/rdfcore/datatypes/Manifest.rdf
>
>
>   The test case is (something like)
>
>    <a> <b> "Arggggggg"^^xsd:integer .
>
> does not entail the empty graph.
>
> I think the MT says it does, but I'm appealing to you for confirmation.

It turns out that it does, because "Arggggg"^^xsd:integer is given a
denotation (even if it's a "spurious" one), and I don't think the woolly
text in the TC WD is sufficient.

Righto. It looks like the state of play is that the test cases for
"duff" datatyped literals need a rethink, and I think I'm happier with
why and how, now. Since Pat's given us a concrete "this is all you get"
for the DT[xsd:integer]-closure from a duff datatyped literal, that's
enough to rebuild those test cases. I'll sleep on it and come back
tomorrow.


--
jan grant, ILRT, University of Bristol. http://www.ilrt.bris.ac.uk/
Tel +44(0)117 9287088 Fax +44 (0)117 9287112 http://ioctl.org/jan/
ioctl(2): probably the coolest Unix system call in the world

Received on Thursday, 21 November 2002 04:54:40 UTC