Merge pull request #83 from opendata-swiss/feat/harvest-dataset-quali…

…fied-relation

feat: Map qualified_relations onto dataset

ckanext/dcatapchharvest/profiles.py

@@ -224,6 +224,17 @@ def _relations(self, subject):
 
         return relations
 
+    def _qualified_relations(self, subject):
+        qualified_relations = []
+
+        for relation_node in self.g.objects(subject, DCAT.qualifiedRelation):
+            qualified_relations.append({
+                "relation": self._object_value(relation_node, DCT.relation),
+                "role": self._object_value(relation_node, DCAT.hadRole),
+            })
+
+        return qualified_relations
+
     def _license_rights_name(self, subject, predicate):
         for node in self.g.objects(subject, predicate):
             # DCAT-AP CH v1: the license as a literal (should be
@@ -416,6 +427,7 @@ def parse_dataset(self, dataset_dict, dataset_ref):  # noqa
         dataset_dict['resources'] = []
         dataset_dict['relations'] = []
         dataset_dict['see_alsos'] = []
+        dataset_dict['qualified_relations'] = []
 
         # Basic fields
         for key, predicate in (
@@ -501,6 +513,10 @@ def parse_dataset(self, dataset_dict, dataset_ref):  # noqa
         for see_also in see_alsos:
             dataset_dict['see_alsos'].append({'dataset_identifier': see_also})
 
+        dataset_dict["qualified_relations"] = self._qualified_relations(
+            dataset_ref
+        )
+
         # Dataset URI
         dataset_uri = dh.dataset_uri(dataset_dict, dataset_ref)
         dataset_dict['extras'].append({'key': 'uri', 'value': dataset_uri})
@@ -711,7 +727,7 @@ def graph_from_dataset(self, dataset_dict, dataset_ref):  # noqa
         if dataset_dict.get('see_alsos'):
             references = dataset_dict.get('see_alsos')
             for reference in references:
-                # we only excpect dicts here
+                # we only expect dicts here
                 if not isinstance(reference, dict):
                     continue
                 reference_identifier = reference.get('dataset_identifier')
@@ -722,6 +738,32 @@ def graph_from_dataset(self, dataset_dict, dataset_ref):  # noqa
                         Literal(reference_identifier)
                     ))
 
+        if dataset_dict.get("qualified_relations"):
+            for reference in dataset_dict["qualified_relations"]:
+                if not reference.get("relation"):
+                    continue
+
+                qualified_relation = BNode()
+                g.add((qualified_relation, RDF.type, DCAT.Relationship))
+                g.add((
+                    qualified_relation,
+                    DCT.relation,
+                    URIRef(reference["relation"])
+                ))
+
+                if reference.get("role"):
+                    g.add((
+                        qualified_relation,
+                        DCAT.hadRole,
+                        URIRef(reference["role"])
+                    ))
+
+                g.add((
+                    dataset_ref,
+                    DCAT.qualifiedRelation,
+                    qualified_relation
+                ))
+
         # Contact details
         if dataset_dict.get('contact_points'):
             contact_points = self._get_dataset_value(dataset_dict, 'contact_points')  # noqa

ckanext/dcatapchharvest/tests/fixtures/1901.xml

@@ -31,6 +31,18 @@
     </dcat:distribution>
     <dct:language>de</dct:language>
     <rdfs:seeAlso>4682791@bundesamt-fur-statistik-bfs</rdfs:seeAlso>
+    <dcat:qualifiedRelation>
+      <dcat:Relationship>
+        <dct:relation rdf:resource="http://example.org/Original987"/>
+        <dcat:hadRole rdf:resource="http://www.iana.org/assignments/relation/original"/>
+      </dcat:Relationship>
+    </dcat:qualifiedRelation>
+    <dcat:qualifiedRelation>
+      <dcat:Relationship>
+        <dct:relation rdf:resource="http://example.org/Related486"/>
+        <dcat:hadRole rdf:resource="http://www.iana.org/assignments/relation/related"/>
+      </dcat:Relationship>
+    </dcat:qualifiedRelation>
     <dct:issued rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">1900-12-31T00:00:00</dct:issued>
     <dcat:keyword xml:lang="fr">bases-statistiques-et-generalites</dcat:keyword>
     <dcat:keyword xml:lang="de">publikation</dcat:keyword>

ckanext/dcatapchharvest/tests/test_dcatap_ch_parse.py

@@ -18,23 +18,23 @@
 class TestSwissDCATAPProfileParsing(BaseParseTest):
 
     def test_rights_license(self):
-            
+
             contents = self._get_file_contents('dataset-rights.xml')
             p = RDFParser(profiles=['swiss_dcat_ap'])
             p.parse(contents)
-    
+
             datasets = [d for d in p.datasets()]
-    
+
             # Dataset
             eq_(len(datasets), 1)
             dataset = datasets[0]
-            
+
              # Resources
             eq_(len(dataset['resources']), 1)
             resource = dataset['resources'][0]
             eq_(resource['rights'], u'NonCommercialAllowed-CommercialAllowed-ReferenceRequired')
             eq_(resource['license'], u'NonCommercialAllowed-CommercialWithPermission-ReferenceRequired')
-            
+
     def test_dataset_all_fields(self):
 
         contents = self._get_file_contents('1901.xml')
@@ -101,6 +101,23 @@ def test_dataset_all_fields(self):
         see_also = dataset['see_alsos'][0]
         eq_(see_also['dataset_identifier'], u'4682791@bundesamt-fur-statistik-bfs')
 
+        # Qualified relations
+        qualified_relations = sorted(dataset["qualified_relations"])
+        eq_(
+            qualified_relations[0],
+            {
+                "relation": "http://example.org/Original987",
+                "role": "http://www.iana.org/assignments/relation/original"
+            }
+        )
+        eq_(
+            qualified_relations[1],
+            {
+                "relation": "http://example.org/Related486",
+                "role": "http://www.iana.org/assignments/relation/related"
+            }
+        )
+
         #  Lists
         eq_(sorted(dataset['language']), [u'de', u'fr'])
         eq_(sorted(dataset['groups']), [{'name': u'statistical-basis'}])