Ontophora

Typed Pydantic models for the OWL 2 structural specification.

Ontophora represents OWL entities, expressions, axioms, and ontology documents as validated records, so an OWL-aware application gets construct-level data that checks itself and travels as plain JSON — without parsing an OWL surface syntax. Each record has a UID and a fixed construct kind; records refer to one another by UID rather than nesting an object graph.

Scope

Ontophora owns:

• Typed models for the OWL 2 structural construct catalog.
• Construct validation, a registry-derived JSON Schema, and reference shapes.
• JSON envelopes for sets of construct records and content fingerprints.

It does not parse or serialize OWL document formats, store a graph, execute queries, reason over an ontology, or resolve references. Applications that own a complete record set perform those operations.

Relationship to the stack

Ontophora is the typed-model library in the wider ontology tooling family. It is independent of its sister projects:

• Ontoplexis maps OWL/XML to a Cypher-queryable graph and back.
• Ontopoiesis provides operator workflows such as build, lint, test, diff, and render on Ontoplexis projections.

Neither project currently depends on Ontophora at runtime.

Install

pip install ontophora

from ontophora import Klass, SubClassOf, records_to_json

animal = Klass(uid="0x1", iri="https://example.org/animals#Animal")
dog = Klass(uid="0x2", iri="https://example.org/animals#Dog")
axiom = SubClassOf(
    uid="0x3",
    sub_class_expression=dog.uid,
    super_class_expression=animal.uid,
)

print(records_to_json([animal, dog, axiom]))

Documentation

Read the documentation for a tutorial, JSON package guidance, the model rationale, and the API reference.

License

Ontophora is licensed under the Apache License 2.0.