Feat string to uri

src/spinneret/graph.py

@@ -20,16 +20,166 @@ def create_graph(metadata_files: list = None, vocabulary_files: list = None) ->
     # Load metadata
     if metadata_files is not None:
         for filename in metadata_files:
-            g.parse(filename)
+            g.parse(filename, format="json-ld")
 
     # Load vocabularies
     if vocabulary_files is not None:
         for filename in vocabulary_files:
             g.parse(filename, format=guess_format(filename))
 
+    # Some string literals should be converted to URIRefs to create linked
+    # data with vocabularies. These are often cases where SOSO conventions
+    # recommend connecting the value to an object property (e.g. `url`) rather
+    # than the object @id. The following code makes these conversions based on
+    # where vocabulary URIs can be used in SOSO markup.
+    g = convert_keyword_url_to_uri(g)
+    g = convert_variable_property_id_to_uri(g)
+    g = convert_variable_measurement_technique_to_uri(g)
+    g = convert_variable_unit_code_to_uri(g)
+    g = convert_license_to_uri(g)
+
     return g
 
 
+def convert_keyword_url_to_uri(graph: Graph) -> Graph:
+    """
+    :param graph: Graph of metadata and vocabularies
+    :returns: Graph with keyword URLs converted to URIs
+    :notes: Converts values of `schema:keyword/schema:DefinedTerm/schema:url`
+        to URI references if the value appears to be a URL.
+    """
+    update_request = """
+    PREFIX schema: <https://schema.org/>
+
+    DELETE {
+        ?term schema:url ?value .
+    }
+    INSERT {
+        ?term schema:url ?newURI .
+    }
+    WHERE {
+        ?dataset schema:keywords ?term .
+        ?term a schema:DefinedTerm .
+        ?term schema:url ?value .
+        FILTER (isLiteral(?value) && REGEX(?value, "^https?://", "i")) .
+        BIND (URI(?value) AS ?newURI)
+    }
+    """
+    graph.update(update_request)
+    return graph
+
+
+def convert_variable_property_id_to_uri(graph: Graph) -> Graph:
+    """
+    :param graph: Graph of metadata and vocabularies
+    :returns: Graph with variable property IDs converted to URIs
+    :notes: Converts values of `schema:variableMeasured/schema:PropertyValue/
+        schema:propertyID` to URI references if the value appears to be a URL.
+    """
+    update_request = """
+    PREFIX schema: <https://schema.org/>
+
+    DELETE {
+        ?term schema:propertyID ?value .
+    }
+    INSERT {
+        ?term schema:propertyID ?newURI .
+    }
+    WHERE {
+        ?dataset schema:variableMeasured ?term .
+        ?term a schema:PropertyValue .
+        ?term schema:propertyID ?value .
+        FILTER (isLiteral(?value) && REGEX(?value, "^https?://", "i")) .
+        BIND (URI(?value) AS ?newURI)
+    }
+    """
+    graph.update(update_request)
+    return graph
+
+
+def convert_variable_measurement_technique_to_uri(graph: Graph) -> Graph:
+    """
+    :param graph: Graph of metadata and vocabularies
+    :returns: Graph with variable measurement techniques converted to URIs
+    :notes: Converts values of `schema:variableMeasured/schema:PropertyValue/
+        schema:measurementTechnique` to URI references if the value appears to
+        be a URL.
+    """
+    update_request = """
+    PREFIX schema: <https://schema.org/>
+
+    DELETE {
+        ?term schema:measurementTechnique ?value .
+    }
+    INSERT {
+        ?term schema:measurementTechnique ?newURI .
+    }
+    WHERE {
+        ?dataset schema:variableMeasured ?term .
+        ?term a schema:PropertyValue .
+        ?term schema:measurementTechnique ?value .
+        FILTER (isLiteral(?value) && REGEX(?value, "^https?://", "i")) .
+        BIND (URI(?value) AS ?newURI)
+    }
+    """
+    graph.update(update_request)
+    return graph
+
+
+def convert_variable_unit_code_to_uri(graph: Graph) -> Graph:
+    """
+    :param graph: Graph of metadata and vocabularies
+    :returns: Graph with variable unit codes converted to URIs
+    :notes: Converts values of `schema:variableMeasured/schema:PropertyValue/
+        schema:unitCode` to URI references if the value appears to be a URL.
+    """
+    update_request = """
+    PREFIX schema: <https://schema.org/>
+
+    DELETE {
+        ?term schema:unitCode ?value .
+    }
+    INSERT {
+        ?term schema:unitCode ?newURI .
+    }
+    WHERE {
+        ?dataset schema:variableMeasured ?term .
+        ?term a schema:PropertyValue .
+        ?term schema:unitCode ?value .
+        FILTER (isLiteral(?value) && REGEX(?value, "^https?://", "i")) .
+        BIND (URI(?value) AS ?newURI)
+    }
+    """
+    graph.update(update_request)
+    return graph
+
+
+def convert_license_to_uri(graph: Graph) -> Graph:
+    """
+    :param graph: Graph of metadata and vocabularies
+    :returns: Graph with licenses converted to URIs
+    :notes: Converts values of `schema:license` to URI references if the value
+        appears to be a URL.
+    """
+    update_request = """
+    PREFIX schema: <https://schema.org/>
+
+    DELETE {
+        ?dataset schema:license ?value .
+    }
+    INSERT {
+        ?dataset schema:license ?newURI .
+    }
+    WHERE {
+        ?dataset schema:license ?value .
+        FILTER (isLiteral(?value) && REGEX(?value, "^https?://", "i")) .
+        BIND (URI(?value) AS ?newURI)
+    }
+    """
+    graph.update(update_request)
+    return graph
+
+
 if __name__ == "__main__":
     # Example usage
     WORKING_DIR = "/Users/csmith/Data/soso/all_edi_test_results"