Refactor RdfsImportEngine

pyproject.toml

@@ -33,25 +33,29 @@ packages = [
 [tool.poetry.dependencies]
 python = "^3.9"
 linkml = "^1.7.4"
-mkdocs = ">=1.2.3"
 pandas = ">=1.3.5"
 python-dateutil = ">=2.8.2"
-jsonpatch = ">=1.32"
 quantulum3 = ">=0.7.9"
 funowl = ">=0.2.3"
 click-log = ">=0.4.0"
-psycopg2-binary = "^2.9.2"
+psycopg2-binary = { version = "^2.9.2", optional = true }
 strsimpy = ">=0.2.1"
 requests = ">=2.26.0"
 oaklib = ">=0.5.25"
 pandera = ">=0.12.0"
 tomlkit = ">=0.11.4"
 inflect = ">=6.0.0"
 schemasheets = ">=0.1.24"
-xmltodict = "^0.13.0"
-click-default-group = "^1.2.4"
 linkml-runtime = "^1.7.2"
-duckdb = "^0.10.1"
+duckdb = { version = "^0.10.1", optional = true }
+click = "^8.1.7"
+deprecated = "^1.2.15"
+mariadb = {version = "^1.1.11", optional = true}
+sqlalchemy = "^2.0.36"
+ruamel-yaml = "^0.18.6"
+rdflib = "^7.1.1"
+jsonasobj2 = "^1.0.4"
+deprecation = "^2.1.0"
 numpy = "<2.0"
 pydbml = "^1.1.2"
 
@@ -84,6 +88,7 @@ extract-schema = "schema_automator.utils.schema_extractor:cli"
 [tool.poetry.extras]
 docs = ["Sphinx", "sphinx-pdj-theme", "sphinxcontrib-mermaid"]
 mariadb = ["mariadb"]
+postgres = ["psycopg2-binary"]
 
 [tool.codespell]
 # Ref: https://github.com/codespell-project/codespell#using-a-config-file

schema_automator/cli.py

@@ -473,15 +473,15 @@ def import_owl(owlfile, output, **args):
 @click.argument('rdfsfile')
 @output_option
 @schema_name_option
-@click.option('--input-type', '-I',
+@click.option('--format', '-f',
               default='turtle',
               help="Input format, eg. turtle")
 @click.option('--identifier', '-I', help="Slot to use as identifier")
 @click.option('--model-uri', help="Model URI prefix")
 @click.option('--metamodel-mappings',
               help="Path to metamodel mappings YAML dictionary")
 @click.option('--output', '-o', help="Path to saved yaml schema")
-def import_rdfs(rdfsfile, output, metamodel_mappings, **args):
+def import_rdfs(rdfsfile: str, output: str, metamodel_mappings: str, schema_name: str, **args):
     """
     Import an RDFS schema to LinkML
 
@@ -494,7 +494,7 @@ def import_rdfs(rdfsfile, output, metamodel_mappings, **args):
         with open(metamodel_mappings) as f:
             mappings_obj = yaml.safe_load(f)
     sie = RdfsImportEngine(initial_metamodel_mappings=mappings_obj)
-    schema = sie.convert(rdfsfile, **args)
+    schema = sie.convert(rdfsfile, name=schema_name, **args)
     write_schema(schema, output)
 
 @main.command()

tests/resources/foaf_snippet.ttl

@@ -0,0 +1,19 @@
+@prefix foaf: <http://xmlns.com/foaf/0.1/> .
+@prefix owl: <http://www.w3.org/2002/07/owl#> .
+@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
+@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
+
+foaf:knows rdf:type owl:ObjectProperty ;
+           rdfs:domain foaf:Person ;
+           rdfs:range foaf:Person ;
+           rdfs:comment "A person known by this person (indicating some level of reciprocated interaction between the parties)." ;
+           rdfs:isDefinedBy <http://xmlns.com/foaf/0.1/> ;
+           rdfs:label "knows" .
+
+foaf:Person rdf:type owl:Class ;
+            rdfs:subClassOf <http://www.w3.org/2003/01/geo/wgs84_pos#SpatialThing> ,
+                            foaf:Agent ;
+            owl:disjointWith foaf:Project ;
+            rdfs:comment "A person." ;
+            rdfs:isDefinedBy <http://xmlns.com/foaf/0.1/> ;
+            rdfs:label "Person" .

tests/test_importers/test_rdfs_importer.py

@@ -2,6 +2,7 @@
 
 """Test the module can be imported."""
 
+from io import StringIO
 import unittest
 import os
 import yaml
@@ -16,8 +17,55 @@
 
 REPRO = os.path.join(INPUT_DIR, 'reproschema.ttl')
 OUTSCHEMA = os.path.join(OUTPUT_DIR, 'reproschema-from-ttl.yaml')
+FOAF = os.path.join(INPUT_DIR, 'foaf_snippet.ttl')
 
 
+def test_import_foaf():
+    engine = RdfsImportEngine()
+    schema = engine.convert(FOAF)
+    sv = SchemaView(schema)
+    assert len(sv.all_classes()) == 3
+    assert len(sv.all_slots()) == 1
+    assert sv.get_slot("knows").range == "Person"
+    assert sv.schema.default_prefix == "foaf"
+    assert "foaf" in sv.schema.prefixes
+
+def test_comment_description():
+    """
+    rdfs:comment should be converted to description
+    """
+    rdf = StringIO("""
+    @prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
+
+    <http://example.org/Class> a rdfs:Class ;
+        rdfs:comment "A class." .
+    """)
+    engine = RdfsImportEngine()
+    schema = engine.convert(rdf)
+    sv = SchemaView(schema)
+    cls = sv.get_class("Class")
+    assert cls.description == "A class."
+
+def test_infer_prefix():
+    """
+    If the schema has no name, id or default prefix, the importer should infer them from prefix usage in the schema.
+    """
+    rdf = StringIO("""
+    @prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
+    @prefix foo: <https://foo.com> .
+
+    foo:Class a rdfs:Class ;
+        rdfs:comment "A class." .
+
+    foo:prop a rdfs:Property ;
+        rdfs:comment "A property." .
+    """)
+    engine = RdfsImportEngine()
+    schema = engine.convert(rdf)
+    # Although not explicitly provided, the importer should realise that the prefix is "foo"
+    assert schema.default_prefix == "foo"
+    assert schema.id == "https://foo.com"
+    assert schema.name == "foo"
 
 def test_from_rdfs():
     """Test OWL conversion."""
@@ -26,7 +74,6 @@ def test_from_rdfs():
     write_schema(schema, OUTSCHEMA)
     # roundtrip
     s = YAMLGenerator(OUTSCHEMA).serialize()
-    print(s[0:100])
     sv = SchemaView(OUTSCHEMA)
     activity = sv.get_class("Activity")
     assert activity
@@ -36,7 +83,3 @@ def test_from_rdfs():
     assert len(slots) == 1
     slot = slots[0]
     assert slot.name == "id"
-
-
-
-