Re: New tests in editors draft

I have now run OWLP on the new manifest file.  OWLP reports four divergences.

Parsing http://www.w3.org/2002/03owlt/SymmetricProperty/premises003
Reading DL http://www.w3.org/2002/03owlt/SymmetricProperty/premises003
OWL ERROR: When checking Individual <http://www.w3.org/2002/03owlt/SymmetricProperty/premises003#Ghent>
Illegal value for object property
Required level Lite Determined level Full
FAILED to correctly identify http://www.w3.org/2002/03owlt/SymmetricProperty/premises003

It appears that this file is missing a declaration for Antwerp, and thus is
Full, not Lite.

Parsing http://www.w3.org/2002/03owlt/SymmetricProperty/conclusions003
Reading DL http://www.w3.org/2002/03owlt/SymmetricProperty/conclusions003
OWL ERROR: When checking Individual <http://www.w3.org/2002/03owlt/SymmetricProperty/premises003#Antwerp>
Illegal property on individual
Required level Lite Determined level Full
FAILED to correctly identify http://www.w3.org/2002/03owlt/SymmetricProperty/conclusions003

It appears that this file is missing a declaration for Ghent, and thus is
Full, not Lite.

Parsing http://www.w3.org/2002/03owlt/TransitiveProperty/conclusions002
Reading DL http://www.w3.org/2002/03owlt/TransitiveProperty/conclusions002
OWL ERROR: Cardinality on http://www.w3.org/2002/03owlt/TransitiveProperty/premises002#symProp an (inverse) superclass of a transitive property
Required level Lite Determined level Full
FAILED to correctly identify http://www.w3.org/2002/03owlt/TransitiveProperty/conclusions002

symProp is transitive but has a cardinality restriction on it, and thus the
file is Full.

Trying consistent <http://www.w3.org/2002/03owlt/I5.8/Manifest016#test>
Parsing http://www.w3.org/2002/03owlt/I5.8/consistent016
Reading DL http://www.w3.org/2002/03owlt/I5.8/consistent016
Required level Full Determined level Lite
FAILED to correctly identify http://www.w3.org/2002/03owlt/I5.8/consistent016

OWLP is using the current situation, which makes this document Lite, not full.


From: Jeremy Carroll <jjc@hpl.hp.com>
Subject: New tests in editors draft
Date: Fri, 19 Sep 2003 19:25:04 +0300

> I've added some more tests, following actions from yesterday.
> I haven't updated the change log adequately.
> 
> imports-011 is supposedly fixed.
> 
> Would value implementor feedback particularly Sean and Peter on syntax 
> checker, since mine is currently broke.
> 
> Note I5.8-013 and I5.8-014 are intended to be uncontroversial, but simply to 
> validate the syntax checker agsinst S&AS CR on the undisputed part of user 
> defined datatype URI, then I5.8-015 and I5.9-016 illustrate the issue 
> discussed.
> 
> Fixed
> imports-011

Both premise and conclusion are in Lite as indicated.

> proposed again
> cardinality-005 (contradicting cardinality-007)

Both premise and conclusion are in Full as indicated.

> New
> I5.8-013, I5.8-014, I5.8-015

Lite, Full, and Lite as indicated.

>  I5.8-016

Lite, not Full.

> SymmetricProperty-002

Both premise and conclusion in DL as indicated

> SymmetricProperty-003

Both premise and conclusion in Full, not Lite.

> TransitiveProperty-002

Premise in DL as indicated; conclusion in Full, not Lite.

> Class-001

Full, as indicated.

> Class-002

Premise in Full, conclusion in Lite, as indicated

> Class-003

Premise in Lite, conclusion in Full, as indicated

> Note I5.8-016 has a syntax error in the draft and the zip which is fixed in 
> the CVS copy, and on the direct web copy
> 
> Jeremy


Full results can be temporarily (after about 8:30 PM EST 19 Sep 2003) found in 

http://www-db.research.bell-labs.com/user/pfps/owl/TRACE2.text
http://www-db.research.bell-labs.com/user/pfps/owl/RESULTS2.rdf



peter

Received on Friday, 19 September 2003 20:16:54 UTC