uml2semantics-python 0.9.31

A refactored Python implementation of the original uml2semantics tool, updated so that OWL 2 semantics are used consistently throughout the generated ontology.

This work would not be possible without Henriette Harmse's https://henrietteharmse.com/uml-vs-owl/uml-class-diagram-to-owl-and-sroiq-reference/ and the reference implementation here https://github.com/henrietteharmse/uml2semantics

Key features

• Converts UML models (via XMI/TSV) into OWL 2 ontologies
• Preserves UML class, property, and enumeration structure
• Generates OWL 2 class, object property, and data property axioms
• Supports datatype restrictions using proper OWL 2 patterns (DatatypeRestriction, facets)
• Choice patterns (inclusive/exclusive) mapped to union/disjointness
• Pluggable profiles for ISO 20022 and generic UML
• Declarative annotations via AnnotationProperties.tsv + Annotations.tsv

Installation

pip install .

or, using pipx:

pipx install .

Command-line usage

The main entry point is:

uml2semantics [OPTIONS]

Supported options (compatible with 0.8.2, extended):

• -i, --input PATH - UML/XMI input file
• -o, --output PATH - Output ontology path (RDF/XML, Turtle, or OWL/XML)
• -f, --format {rdfxml,turtle,nt,jsonld,trig,n3} - Output serialisation format
• -b, --base-iri IRI - Base IRI for generated ontology
• -p, --profile {generic,iso20022} - Mapping profile to apply
• --imports IRI - Additional ontology IRI to import (repeatable)
• --no-imports - Disable automatic imports
• --validate / --no-validate - Turn structural validation on/off
• --log-level {ERROR,WARNING,INFO,DEBUG} - Logging verbosity
• --dry-run - Parse and build in memory, but do not write a file
• --version - Show version information and exit
• --annotation-properties PATH - TSV defining annotation properties
• --annotations PATH - TSV of annotation assertions

Example:

uml2semantics -i examples/simple-model.xmi -o examples/simple-model.owl -f turtle -p iso20022

Python API

from uml2semantics.ontology_builder import build_ontology
from uml2semantics.tsv_loader import load_model

model = load_model(
    classes_tsv="examples/Classes.tsv",
    attributes_tsv="examples/Attributes.tsv",
    datatypes_tsv="examples/Datatypes.tsv",
    enums_tsv="examples/Enumerations.tsv",
    enum_literals_tsv="examples/EnumerationNamedValues.tsv",
    annotation_properties_tsv="examples/AnnotationProperties.tsv",
    annotations_tsv="examples/Annotations.tsv",
)

g = build_ontology(
    model,
    ontology_iri="http://example.org/uml2semantics/demo#",
    prefix_str="ex=http://example.org/;rdfs=http://www.w3.org/2000/01/rdf-schema#;dct=http://purl.org/dc/terms/;xsd=http://www.w3.org/2001/XMLSchema#",
)
g.serialize("examples/demo.ttl", format="turtle")

Annotation TSVs

• AnnotationProperties.tsv columns: Curie, Name, Definition
• Annotations.tsv columns: TargetCurie, AnnotationProperty, Value, Language, Datatype

Choice semantics

• ChoiceOf on a class lists alternative class/attribute tokens.
• ChoiceSemantics = exclusive emits an owl:unionOf plus an AllDisjointClasses covering the choice and its members. Blank or inclusive emits only the union.

Datatypes

• BaseDatatype is required when facets are present; a missing base raises an error to avoid defaulting to xsd:string.

See docs/cli.md and docs/architecture.md for more details.