Merge pull request #5 from casework/add_new_single_stub_generator

Add and test single-stub generator script and Facet-by-IRI-pattern support

.github/workflows/ci.yml

@@ -42,3 +42,5 @@ jobs:
       run: |
         pip -q install pre-commit
         pre-commit run --all-files
+    - name: Run tests
+      run: make PYTHON3=python check

.gitignore

@@ -1,3 +1,4 @@
+*.jar
 *.swp
 .*.done.log
 .DS_Store

.pre-commit-config.yaml

@@ -1,4 +1,10 @@
 repos:
+  - repo: https://github.com/casework/rdf-toolkit-action
+    rev: 2.0.3
+    hooks:
+      - id: rdf-toolkit-normalizer
+        args:
+            - --autofix
   - repo: https://github.com/psf/black
     rev: 24.8.0
     hooks:

Makefile

@@ -19,12 +19,17 @@ SHELL := /bin/bash
 PYTHON3 ?= python3
 
 py_srcfiles := \
-  generate.py
+  src/facet_cardinalities_ttl.py \
+  src/generate_single_stub_json.py
 
 all: \
-  .venv-pre-commit/var/.pre-commit-built.log
+  .venv-pre-commit/var/.pre-commit-built.log \
+  all-tests
 
 .PHONY: \
+  all-tests \
+  all-var \
+  check-pytest \
   check-supply-chain \
   check-supply-chain-pre-commit
 
@@ -37,6 +42,20 @@ all: \
 	    $(py_srcfiles)
 	touch $@
 
+.venv.done.log: \
+  requirements.txt
+	rm -rf venv
+	$(PYTHON3) -m venv \
+	  venv
+	source venv/bin/activate \
+	  && pip install \
+	    --upgrade \
+	    pip
+	source venv/bin/activate \
+	  && pip install \
+	    --requirement requirements.txt
+	touch $@
+
 # This virtual environment is meant to be built once and then persist, even through 'make clean'.
 # If a recipe is written to remove this flag file, it should first run `pre-commit uninstall`.
 .venv-pre-commit/var/.pre-commit-built.log:
@@ -60,11 +79,32 @@ all: \
 	  .venv-pre-commit/var
 	touch $@
 
+all-tests: \
+  all-var
+	$(MAKE) \
+	  --directory tests
+
+all-var: \
+  .mypy.done.log
+	$(MAKE) \
+	  --directory var
+
 check: \
-  .venv-pre-commit/var/.pre-commit-built.log
+  .venv-pre-commit/var/.pre-commit-built.log \
+  check-pytest \
+  all-tests
+
+check-pytest: \
+  .venv.done.log
+	source venv/bin/activate \
+	  && pytest \
+	    --doctest-modules \
+	    --log-level=DEBUG \
+	    $(py_srcfiles)
 
 check-supply-chain: \
-  check-supply-chain-pre-commit
+  check-supply-chain-pre-commit \
+ .mypy.done.log
 
 # Update pre-commit configuration and use the updated config file to
 # review code.  Only have Make exit if 'pre-commit run' modifies files.

requirements.txt

@@ -1,2 +1,4 @@
+PyLD
 case-utils
 mypy
+pytest

src/facet_cardinalities_ttl.py

@@ -0,0 +1,114 @@
+#!/usr/bin/env python3
+
+# Portions of this file contributed by NIST are governed by the
+# following statement:
+#
+# This software was developed at the National Institute of Standards
+# and Technology by employees of the Federal Government in the course
+# of their official duties. Pursuant to title 17 Section 105 of the
+# United States Code this software is not subject to copyright
+# protection and is in the public domain. NIST assumes no
+# responsibility whatsoever for its use by other parties, and makes
+# no guarantees, expressed or implied, about its quality,
+# reliability, or any other characteristic.
+#
+# We would appreciate acknowledgement if the software is used.
+
+import argparse
+import importlib.resources
+import logging
+from typing import Set
+
+import case_utils.ontology
+from case_utils.namespace import NS_OWL, NS_RDF, NS_RDFS, NS_UCO_CORE, NS_XSD
+from rdflib import BNode, Graph, Literal, URIRef
+from rdflib.query import ResultRow
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser()
+    parser.add_argument("out_graph")
+    args = parser.parse_args()
+
+    logging.basicConfig(level=logging.INFO)
+
+    in_graph = Graph()
+    out_graph = Graph()
+    ttl_data = importlib.resources.read_text(case_utils.ontology, "case-1.3.0.ttl")
+    in_graph.parse(data=ttl_data)
+    case_utils.ontology.load_subclass_hierarchy(in_graph)
+
+    n_leaf_facet_classes: Set[URIRef] = set()
+    leaf_facet_query = """\
+PREFIX owl: <http://www.w3.org/2002/07/owl#>
+PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
+PREFIX uco-core: <https://ontology.unifiedcyberontology.org/uco/core/>
+SELECT ?nClass
+WHERE {
+  ?nClass
+    a owl:Class ;
+    rdfs:subClassOf* uco-core:Facet ;
+    .
+  FILTER NOT EXISTS {
+    ?nSubClass
+      a owl:Class ;
+      rdfs:subClassOf ?nClass ;
+      .
+  }
+}
+"""
+    for result in in_graph.query(leaf_facet_query):
+        assert isinstance(result, ResultRow)
+        assert isinstance(result[0], URIRef)
+        n_leaf_facet_classes.add(result[0])
+
+    n_uco_object_classes: Set[URIRef] = set()
+    uco_object_query = """\
+PREFIX owl: <http://www.w3.org/2002/07/owl#>
+PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
+PREFIX uco-core: <https://ontology.unifiedcyberontology.org/uco/core/>
+SELECT ?nClass
+WHERE {
+  ?nClass
+    a owl:Class ;
+    rdfs:subClassOf* uco-core:UcoObject ;
+    .
+}
+"""
+    for result in in_graph.query(uco_object_query):
+        assert isinstance(result, ResultRow)
+        assert isinstance(result[0], URIRef)
+        n_uco_object_classes.add(result[0])
+
+    n_leaf_facet_classes_restricted: Set[URIRef] = set()
+    # Determine which Facets are named by the pattern of a corresponding UcoObject subclass, plus "Facet".
+    for n_uco_object_class in n_uco_object_classes:
+        uco_object_class_iri = str(n_uco_object_class)
+        n_maybe_leaf_facet = URIRef(uco_object_class_iri + "Facet")
+        if n_maybe_leaf_facet in n_leaf_facet_classes:
+            n_restriction = BNode()
+            out_graph.add((n_restriction, NS_RDF.type, NS_OWL.Restriction))
+            out_graph.add((n_restriction, NS_OWL.onClass, n_maybe_leaf_facet))
+            out_graph.add((n_restriction, NS_OWL.onProperty, NS_UCO_CORE.hasFacet))
+            out_graph.add(
+                (
+                    n_restriction,
+                    NS_OWL.qualifiedCardinality,
+                    Literal("1", datatype=NS_XSD.nonNegativeInteger),
+                )
+            )
+            out_graph.add((n_uco_object_class, NS_RDFS.subClassOf, n_restriction))
+            n_leaf_facet_classes_restricted.add(n_maybe_leaf_facet)
+
+    if len(n_leaf_facet_classes_restricted) < len(n_leaf_facet_classes):
+        logging.info("These classes had no pattern-matched UcoObject subclasses:")
+        for n_leaf_facet_class in sorted(
+            n_leaf_facet_classes - n_leaf_facet_classes_restricted
+        ):
+            logging.info("* %s", str(n_leaf_facet_class))
+
+    out_graph.serialize(args.out_graph)
+
+
+if __name__ == "__main__":
+    main()