WIP (sync between work places)

pyproject.toml

@@ -54,16 +54,19 @@ linkml-runtime = "^1.7.2"
 duckdb = "^0.10.1"
 numpy = "<2.0"
 
+Sphinx = { version = ">=4.4.0", optional = true }
+sphinx-pdj-theme = { version = ">=0.2.1", optional = true }
+sphinx-click = ">=3.1.0"
+sphinxcontrib-mermaid = { version = ">=0.9.2", optional = true }
+
 [tool.poetry.dev-dependencies]
 pytest = ">=7.1.1"
-Sphinx = ">=4.4.0"
-sphinx-pdj-theme = ">=0.2.1"
-sphinx-click = ">=3.1.0"
-sphinxcontrib-mermaid = ">=0.9.2"
 myst-parser = "*"
 jupyter = ">=1.0.0"
 lxml = ">=4.9.1"
 
+#mariadb = { version = "^1.3", optional = true }
+
 [tool.poetry.group.llm.dependencies]
 llm = ">=0.12"
 
@@ -82,7 +85,7 @@ extract-schema = "schema_automator.utils.schema_extractor:cli"
 
 [tool.poetry.extras]
 docs = ["Sphinx", "sphinx-pdj-theme", "sphinxcontrib-mermaid"]
-mariadb = ["mariadb"]
+#mariadb = ["mariadb"]
 
 [tool.codespell]
 # Ref: https://github.com/codespell-project/codespell#using-a-config-file

schema_automator/importers/shacl_import_engine.py

@@ -1,13 +1,25 @@
+from collections import defaultdict
 import logging
 
+from dataclasses import dataclass
+from typing import Dict, List, Any
+
+from rdflib import Graph, RDF, OWL, URIRef, RDFS, SKOS, SDO, Namespace
+
+from funowl import Literal
+
 from linkml.utils.schema_builder import SchemaBuilder
 from linkml_runtime import SchemaView
+from linkml_runtime.utils.formatutils import underscore
+from linkml_runtime.utils.introspection import package_schemaview
 from linkml_runtime.linkml_model import (
     SchemaDefinition,
     SlotDefinition,
     ClassDefinition,
 )
+from schema_automator.importers.import_engine import ImportEngine
 
+logger = logging.getLogger(__name__)
 
 HTTP_SDO = Namespace("http://schema.org/")
 
@@ -80,7 +92,7 @@ def convert(
         **kwargs,
     ) -> SchemaDefinition:
         """
-        Converts an OWL schema-style ontology
+        Converts an shacl shapes file
 
         :param file:
         :param name:
@@ -110,6 +122,7 @@ def convert(
             if default_prefix not in schema.prefixes:
                 sb.add_prefix(default_prefix, model_uri, replace_if_present=True)
             schema.id = schema.prefixes[default_prefix].prefix_reference
+
         cls_slots = defaultdict(list)
         props = []
         for rdfs_property_metaclass in self._rdfs_metamodel_iri(
@@ -137,6 +150,7 @@ def convert(
             slot = SlotDefinition(sn, **init_dict)
             slot.slot_uri = str(p.n3(g.namespace_manager))
             sb.add_slot(slot)
+
         rdfs_classes = []
         for rdfs_class_metaclass in self._rdfs_metamodel_iri(ClassDefinition.__name__):
             for s in g.subjects(RDF.type, rdfs_class_metaclass):
@@ -201,7 +215,7 @@ def _element_from_iri(self, iri: URIRef) -> str:
         r = self.reverse_metamodel_mappings.get(iri, [])
         if len(r) > 0:
             if len(r) > 1:
-                logging.debug(f"Multiple mappings for {iri}: {r}")
+                logger.debug(f"Multiple mappings for {iri}: {r}")
             return r[0]
 
     def _object_to_value(self, obj: Any, metaslot: SlotDefinition = None) -> Any:

tests/__init__.py

@@ -1,9 +1,6 @@
 import os
-import pprint
 
 ROOT = os.path.abspath(os.path.dirname(__file__))
 INPUT_DIR = os.path.join(ROOT, 'resources')
 OUTPUT_DIR = os.path.join(ROOT, 'outputs')
 MODEL_DIR = os.path.join(ROOT, 'test_models')
-
-

tests/resources/shacl_simple.ttl

@@ -0,0 +1,34 @@
+# example from http://book.validatingrdf.com/bookHtml011.html#ch050SHACLExample
+
+@prefix schema: <http://schema.org/> .
+@prefix sh: <http://www.w3.org/ns/shacl#> .
+@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .
+@prefix ex: <http://example.org/> .
+
+ex:UserShape a sh:NodeShape;
+   sh:targetClass ex:User ;
+   sh:property [                  # Blank node 1
+    sh:path     schema:name ;
+    sh:minCount 1;
+    sh:maxCount 1;
+    sh:datatype xsd:string ;
+  ] ;
+  sh:property [                   # Blank node 2
+   sh:path schema:gender ;
+   sh:minCount 1;
+   sh:maxCount 1;
+   sh:or (
+    [ sh:in (schema:Male schema:Female) ]
+    [ sh:datatype xsd:string]
+   )
+  ] ;
+  sh:property [                   # Blank node 3  
+   sh:path     schema:birthDate ;
+   sh:maxCount 1;
+   sh:datatype xsd:date ;
+  ] ;
+  sh:property [                   # Blank node 4 
+   sh:path     schema:knows ;
+   sh:nodeKind sh:IRI ;
+   sh:class    ex:User ;
+  ] .

tests/test_importers/test_shacl_importer.py

@@ -11,18 +11,17 @@
 
 # TODO - Write tests (this is a copy of test_rdfs_importer)
 
-REPRO = os.path.join(INPUT_DIR, 'reproschema.ttl')
-OUTSCHEMA = os.path.join(OUTPUT_DIR, 'reproschema-from-ttl.yaml')
-
+REPRO = os.path.join(INPUT_DIR, 'shacl_simple.ttl')
+OUTSCHEMA = os.path.join(OUTPUT_DIR, 'user_from_shacl_simple2.yaml')
 
 
 def test_from_shacl():
     """Test Shacl conversion."""
-    oie = ShaclImportEngine()
+    sie = ShaclImportEngine()
 
-    return
-    schema = oie.convert(REPRO, default_prefix='reproschema', identifier='id')
+    schema = sie.convert(REPRO, default_prefix='usr', identifier='id')
     write_schema(schema, OUTSCHEMA)
+    return
     # roundtrip
     s = YAMLGenerator(OUTSCHEMA).serialize()
     print(s[0:100])
@@ -35,7 +34,3 @@ def test_from_shacl():
     assert len(slots) == 1
     slot = slots[0]
     assert slot.name == "id"
-
-
-
-