pwin/owlready2

Parsing error in Turtle/N3 ontology

rpgoldman opened this issue · 2 comments

The ontology file here https://github.com/rpgoldman/container-ontology/blob/80a1a88d056bcdeac0eda77698348ff9f41acb82/owl/strateos-catalog-individuals.ttl causes a parsing error in Owlready 2. When I look at the source line in question, it's near the bottom:

#################################################################
#    General axioms
#################################################################

[ rdf:type owl:AllDisjointClasses ;
  owl:members ( cont:CatalogEntry
                cont:CoatingMaterial
                cont:Color
                cont:LabEquipment
                cont:VendorFirm
                cont:WellShape
              )
] .

The error message and backtrace is as follows:

---------------------------------------------------------------------------

ValueError                                Traceback (most recent call last)

~/.virtualenvs/container-ontology/lib/python3.9/site-packages/owlready2/driver.py in parse(self, f, format, delete_existing_triples, default_base)
    153               if not line.endswith("\n"): line = "%s\n" % line
--> 154               s,p,o = splitter.split(line[:-3], 2)
    155 

ValueError: not enough values to unpack (expected 3, got 1)


The above exception was the direct cause of the following exception:

OwlReadyOntologyParsingError              Traceback (most recent call last)

/var/folders/5s/cx04gv9j1mq4gvcv2lw70s400000gn/T/ipykernel_79391/2588599199.py in <module>
----> 1 container_onto: owl.Ontology = owl.get_ontology("https://www.dropbox.com/s/s1e2dzw64m01f9n/container-ontology.ttl").load()

~/.virtualenvs/container-ontology/lib/python3.9/site-packages/owlready2/namespace.py in load(self, only_local, fileobj, reload, reload_if_newer, url, **args)
    909         if _LOG_LEVEL: print("* Owlready2 *     ...loading ontology %s from %s..." % (self.name, f), file = sys.stderr)
    910         fileobj = open(f, "rb")
--> 911         try:     new_base_iri = self.graph.parse(fileobj, default_base = self.base_iri, **args)
    912         finally: fileobj.close()
    913       else:

~/.virtualenvs/container-ontology/lib/python3.9/site-packages/owlready2/driver.py in parse(self, f, format, delete_existing_triples, default_base)
    186           self._add_obj_triple_raw_spo(self.onto.storid, rdf_type, owl_ontology)
    187         if current_line:
--> 188           raise OwlReadyOntologyParsingError("NTriples parsing error (or unrecognized file format) in %s, line %s." % (getattr(f, "name", getattr(f, "url", "???")), current_line)) from e
    189         else:
    190           raise OwlReadyOntologyParsingError("NTriples parsing error (or unrecognized file format) in %s." % getattr(f, "name", getattr(f, "url", "???"))) from e

OwlReadyOntologyParsingError: NTriples parsing error (or unrecognized file format) in ../owl/container-ontology.ttl, line 1209.

It looks like it is trying to break the line up into three components and is not finding three, perhaps because there is a top level block that (IIUC) corresponds to a blank node?

I had a cursory look at driver.py and I don't see anything in there that checks for the possibility of a line starting with a [, so maybe that is the problem?

Both Protege and RDFLib read this file successfully.

Hi,

I had a similar issue. I read the file into a graph using rdflib and serialized it into a format accepted by owlready.
In my case, the NTriples format worked for me.

Good luck

Hi,

I had a similar issue. I read the file into a graph using rdflib and serialized it into a format accepted by owlready. In my case, the NTriples format worked for me.

Good luck

In which format you have converted the file after serialization , and what was your file format before and after serialization, could you please tell me?