Search Mailing List Archives
[protege-owl] ValuePartition and Enumeration Patterns Problem
tudorache at stanford.edu
Mon Aug 20 14:41:00 PDT 2007
Thank you for the detailed bug report. The remove class from domain (or
range) was indeed broken. This is fixed in the next Protege build.
David Canos wrote:
> Working with PRotege 3.2.1 and 3.3. Windows and Linux.
> I have a strange behavior after using patterns.
> Whenever I do an enumeration or a value partition using the Pattern
> option in tool menu. After I choose name, items, domain, name of the
> property and accept... Application tells me "succesfull" but I realize
> that the domain of the property created is not properly.
> It appears two classes instead of the class I choose, owl:Thing and the
> domainclass I choose.
> Looking into the owl file it generates I've found:
> <owl:ObjectProperty rdf:ID="hasXXXX">
> <rdfs:range rdf:resource="#XXXX"/>
> <owl:unionOf rdf:parseType="Collection">
> <rdf:Description rdf:about="http://www.w3.org/2002/07/owl#Thing"/>
> <owl:Class rdf:about="#XXXXXXX"/>
> instead of something like:
> <owl:ObjectProperty rdf:ID="hasPollutantType">
> <rdfs:range rdf:resource="#PollutantType"/>
> <owl:Class rdf:about="#PollutionMeasurementRelated"/>
> If you try to fix the problem, removing the domain, by clicking on the
> button. You will get a class cast Internal Error, pasted below:
> Internal Error:java.lang.ClassCastException:
> edu.stanford.smi.protege.model.DefaultCls cannot be cast to
> ...report it to the mailing list...
> Your ontology may now no longer be in a consistent state, and you may
> want to save this version under a different name.
> I indeed, my ontology is now absolutely broken.
> Next time you start Protege you will find some new clases called
> Axiom_X, with different IDs like A01, A02... Sometimes you can't select
> this classes anymore, impossible to delete.
> I fixed it replacing code directly in the owl file, like pasted above.
> But never after clicked in the "remove class from domain" button. It
> completely broke the ontology.
> Thank you
> David Canós
> protege-owl mailing list
> protege-owl at lists.stanford.edu
> Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03
More information about the protege-owl