[review] add explanation for tests

Signed-off-by: Meret Behrens <meret.behrens@tngtech.com>

tests/spdx/parser/rdf/test_extracted_licensing_info_parser.py

@@ -18,18 +18,16 @@
 
 
 @pytest.mark.parametrize(
-    "parent_node_object, license_id, extracted_text, comment, license_name, cross_references",
+    "license_id, extracted_text, comment, license_name, cross_references",
     [
         (
-            SPDX_NAMESPACE.SpdxDocument,
             "LicenseRef-1",
             "extractedText",
             "licenseComment",
             "licenseName",
             ["https://see.also"],
         ),
         (
-            SPDX_NAMESPACE.Package,
             "LicenseRef-2",
             "extractedText",
             "licenseComment",
@@ -38,9 +36,11 @@
         ),
     ],
 )
-def test_parse_extracted_licensing_info(
-    parent_node_object, license_id, extracted_text, comment, license_name, cross_references
-):
+# In rdf, as a short form, the explicit node for licenseId can be omitted, since the ID is also encoded in the URIRef
+# of the ExtractedLicensingInfo node. The first test case has an explicit licenseId node whereas the second test case
+# does not. This behaviour is similar to the externalDocumentRefId, see the discussion here:
+# https://github.com/spdx/spdx-spec/issues/816
+def test_parse_extracted_licensing_info(license_id, extracted_text, comment, license_name, cross_references):
     graph = Graph().parse(os.path.join(os.path.dirname(__file__), "data/file_to_test_rdf_parser.rdf.xml"))
     doc_namespace = "https://some.namespace"
     extracted_licensing_info_node = get_extracted_licensing_info_node_by_license_id(graph, license_id)