#1 turtle parses object property as annotation

[Ignazio Palmisano]

I just found a problem if I try to load e.g.

http://dbpedia.org/ontology/numberOfPads
http://www.w3.org/1999/02/22-rdf-syntax-ns#type
http://www.w3.org/2002/07/owl#ObjectProperty ;
http://www.w3.org/2000/01/rdf-schema#domain
http://dbpedia.org/ontology/Settlement .

with latest OWL API.

I don't know why, but the created axiom is a
AnnotationPropertyDomain(http://dbpedia.org/ontology/numberOfPads
http://dbpedia.org/ontology/Settlement) axioms, although it is a
owl:ObjectProperty.
Seems that this is a general problem with domain axioms

[Matthew Horridge]

It might be because Settlement isn't typed as a class.

[Matthew Horridge]

Strictly speaking that's not valid OWL. In these cases the parser falls back to annotation properties. The parser could be a bit smarter here. Although, this might screw up parsing of dublin core and other vocabularies.


[Ignazio Palmisano]

That's possible, but it should take into account the explicit object property type, right? As it is it looks like an illegal pun is introduced, while it should be only an undeclared class violation.

[Ignazio Palmisano]

I haven't tried this code yet. I'll try adding the declaration and seeing what comes of it.

[Matthew Horridge]

I seem to recall some weird problems with ontologies generated as W3C test cases and ontology generated by Protege 2/3 (lots of them) where properties were typed as object properties and annotation properties but the intention is that they come out as annotation properties.

The parser defaults to annotations if it doesn't have all the typing info it wants - not sure this is helpful in this case though.

Let me know how the extra declaration goes.

[Ignazio Palmisano]

Ticket moved from /p/owlapi/bugs/366/

Can't be converted:

• _milestone: v1.0_(example)
• _priority: 5

[Ignazio Palmisano]

pfffft a long explanation of the issue and why I moved it here just went lost.

Ontology fixes should have a priority and be rerun after each change is applied.