On 20/07/12 10:05, Dave wrote:
OntModel ontm =
ModelFactory.createOntologyModel(OntModelSpec.OWL_MEM_MICRO_RULE_INF);
ontm.read("file:data/temp.owl","RDF/XML");
ontm = ModelFactory.createOntologyModel(OntModelSpec.OWL_MEM,
ontm);
This is wrapping the inference closure as a new model. I suspect that
Olivier is doing is:
ontm = ModelFactory.createOntologyModel(OntModelSpec.OWL_MEM,
ontm.getBaseMode() )
which would, indeed, remove the inference.
Olivier - I think what you need to do is keep *both* models around, the
plain base model to work around the namedHierarchyRoots problem, and the
model with the inference engine attached. Then you use the one that you
need for a given purpose. This is not an uncommon pattern: sometimes we
need the inference closure of the triples, and other times not.
Ian
