Merge pull request #302 from Health-RI/feature/add-creator-support

Feature/add creator support

ckanext/dcat/converters.py

@@ -29,15 +29,31 @@ def dcat_to_ckan(dcat_dict):
     elif isinstance(dcat_publisher, dict) and dcat_publisher.get('name'):
         package_dict['extras'].append({'key': 'dcat_publisher_name', 'value': dcat_publisher.get('name')})
 
-        if dcat_publisher.get('mbox'):
-            package_dict['extras'].append({'key': 'dcat_publisher_email', 'value': dcat_publisher.get('mbox')})
+        if dcat_publisher.get('email'):
+            package_dict['extras'].append({'key': 'dcat_publisher_email', 'value': dcat_publisher.get('email')})
 
         if dcat_publisher.get('identifier'):
             package_dict['extras'].append({
                 'key': 'dcat_publisher_id',
                 'value': dcat_publisher.get('identifier')  # This could be a URI like https://ror.org/05wg1m734
             })
 
+    dcat_creator = dcat_dict.get('creator')
+    if isinstance(dcat_creator, basestring):
+        package_dict['extras'].append({'key': 'dcat_creator_name', 'value': dcat_creator})
+    elif isinstance(dcat_creator, dict) and dcat_creator.get('name'):
+        if dcat_creator.get('name'):
+            package_dict['extras'].append({'key': 'dcat_creator_name', 'value': dcat_creator.get('name')})
+
+        if dcat_creator.get('email'):
+            package_dict['extras'].append({'key': 'dcat_creator_email', 'value': dcat_creator.get('email')})
+
+        if dcat_creator.get('identifier'):
+            package_dict['extras'].append({
+                'key': 'dcat_creator_id',
+                'value': dcat_creator.get('identifier')
+            })
+
     package_dict['extras'].append({
         'key': 'language',
         'value': ','.join(dcat_dict.get('language', []))
@@ -63,20 +79,20 @@ def dcat_to_ckan(dcat_dict):
 
 
 def ckan_to_dcat(package_dict):
-
     dcat_dict = {}
 
     dcat_dict['title'] = package_dict.get('title')
     dcat_dict['description'] = package_dict.get('notes')
     dcat_dict['landingPage'] = package_dict.get('url')
 
-
+    # Keywords
     dcat_dict['keyword'] = []
     for tag in package_dict.get('tags', []):
         dcat_dict['keyword'].append(tag['name'])
 
-
+    # Publisher
     dcat_dict['publisher'] = {}
+    dcat_dict['creator'] = {}
 
     for extra in package_dict.get('extras', []):
         if extra['key'] in ['dcat_issued', 'dcat_modified']:
@@ -85,19 +101,41 @@ def ckan_to_dcat(package_dict):
         elif extra['key'] == 'language':
             dcat_dict['language'] = extra['value'].split(',')
 
+        # Publisher fields
         elif extra['key'] == 'dcat_publisher_name':
             dcat_dict['publisher']['name'] = extra['value']
 
         elif extra['key'] == 'dcat_publisher_email':
-            dcat_dict['publisher']['mbox'] = extra['value']
+            dcat_dict['publisher']['email'] = extra['value']
+
+        elif extra['key'] == 'dcat_publisher_id':
+            dcat_dict['publisher']['identifier'] = extra['value']
+
+        # Creator fields
+        elif extra['key'] == 'dcat_creator_name':
+            dcat_dict['creator']['name'] = extra['value']
 
+        elif extra['key'] == 'dcat_creator_email':
+            dcat_dict['creator']['email'] = extra['value']
+
+        elif extra['key'] == 'dcat_creator_id':
+            dcat_dict['creator']['identifier'] = extra['value']
+
+        # Identifier
         elif extra['key'] == 'guid':
             dcat_dict['identifier'] = extra['value']
 
+    # Fallback for publisher (if no name in extras, use maintainer)
     if not dcat_dict['publisher'].get('name') and package_dict.get('maintainer'):
         dcat_dict['publisher']['name'] = package_dict.get('maintainer')
         if package_dict.get('maintainer_email'):
-            dcat_dict['publisher']['mbox'] = package_dict.get('maintainer_email')
+            dcat_dict['publisher']['email'] = package_dict.get('maintainer_email')
+
+    # Fallback for creator (if no name in extras, optionally use author)
+    if not dcat_dict['creator'].get('name') and package_dict.get('author'):
+        dcat_dict['creator']['name'] = package_dict.get('author')
+        if package_dict.get('author_email'):
+            dcat_dict['creator']['email'] = package_dict.get('author_email')
 
     dcat_dict['distribution'] = []
     for resource in package_dict.get('resources', []):

ckanext/dcat/profiles/base.py

@@ -419,58 +419,37 @@ def _insert_or_update_temporal(self, dataset_dict, key, value):
         else:
             dataset_dict["extras"].append({"key": key, "value": value})
 
-    def _publisher(self, subject, predicate):
+    def _agent_details(self, subject, predicate):
         """
-        Returns a dict with details about a dct:publisher entity, a foaf:Agent
+        Returns a dict with details about a dct:publisher or dct:creator entity, a foaf:Agent
 
         Both subject and predicate must be rdflib URIRef or BNode objects
 
         Examples:
 
-        <dct:publisher>
+        <dct:publisher> or <dct:creator>
             <foaf:Organization rdf:about="http://orgs.vocab.org/some-org">
                 <foaf:name>Publishing Organization for dataset 1</foaf:name>
                 <foaf:mbox>[email protected]</foaf:mbox>
                 <foaf:homepage>http://some.org</foaf:homepage>
                 <dct:type rdf:resource="http://purl.org/adms/publishertype/NonProfitOrganisation"/>
             </foaf:Organization>
-        </dct:publisher>
-
-        {
-            'uri': 'http://orgs.vocab.org/some-org',
-            'name': 'Publishing Organization for dataset 1',
-            'email': '[email protected]',
-            'url': 'http://some.org',
-            'type': 'http://purl.org/adms/publishertype/NonProfitOrganisation',
-        }
-
-        <dct:publisher rdf:resource="http://publications.europa.eu/resource/authority/corporate-body/EURCOU" />
-
-        {
-            'uri': 'http://publications.europa.eu/resource/authority/corporate-body/EURCOU'
-        }
 
-        Returns keys for uri, name, email, url and type with the values set to
-        an empty string if they could not be found
+        Returns keys for uri, name, email, url, type, and identifier with the values set to
+        an empty string if they could not be found.
         """
 
-        publisher = {}
+        agent_details = {}
 
         for agent in self.g.objects(subject, predicate):
+            agent_details["uri"] = str(agent) if isinstance(agent, term.URIRef) else ""
+            agent_details["name"] = self._object_value(agent, FOAF.name)
+            agent_details["email"] = self._object_value(agent, FOAF.mbox)
+            agent_details["url"] = self._object_value(agent, FOAF.homepage)
+            agent_details["type"] = self._object_value(agent, DCT.type)
+            agent_details['identifier'] = self._object_value(agent, DCT.identifier)
 
-            publisher["uri"] = str(agent) if isinstance(agent, term.URIRef) else ""
-
-            publisher["name"] = self._object_value(agent, FOAF.name)
-
-            publisher["email"] = self._object_value(agent, FOAF.mbox)
-
-            publisher["url"] = self._object_value(agent, FOAF.homepage)
-
-            publisher["type"] = self._object_value(agent, DCT.type)
-
-            publisher['identifier'] = self._object_value(agent, DCT.identifier)
-
-        return publisher
+        return agent_details
 
     def _contact_details(self, subject, predicate):
         """
@@ -1136,7 +1115,7 @@ def _extract_catalog_dict(self, catalog_ref):
         out.append(
             {
                 "key": "source_catalog_publisher",
-                "value": json.dumps(self._publisher(catalog_ref, DCT.publisher)),
+                "value": json.dumps(self._agent_details(catalog_ref, DCT.publisher)),
             }
         )
         return out

ckanext/dcat/profiles/euro_dcat_ap_base.py

@@ -34,7 +34,6 @@
 
 config = toolkit.config
 
-
 DISTRIBUTION_LICENSE_FALLBACK_CONFIG = "ckanext.dcat.resource.inherit.license"
 
 
@@ -122,13 +121,21 @@ def _parse_dataset_base(self, dataset_dict, dataset_ref):
                     )
 
         # Publisher
-        publisher = self._publisher(dataset_ref, DCT.publisher)
+        publisher = self._agent_details(dataset_ref, DCT.publisher)
         for key in ("uri", "name", "email", "url", "type", "identifier"):
             if publisher.get(key):
                 dataset_dict["extras"].append(
                     {"key": "publisher_{0}".format(key), "value": publisher.get(key)}
                 )
 
+        # Creator
+        creator = self._agent_details(dataset_ref, DCT.creator)
+        for key in ("uri", "name", "email", "url", "type", "identifier"):
+            if creator.get(key):
+                dataset_dict["extras"].append(
+                    {"key": "creator_{0}".format(key), "value": creator.get(key)}
+                )
+
         # Temporal
         start, end = self._time_interval(dataset_ref, DCT.temporal)
         if start:
@@ -250,7 +257,6 @@ def _parse_dataset_base(self, dataset_dict, dataset_ref):
                     "publisher_name",
                     "publisher_email",
                 ):
-
                     extra["key"] = "dcat_" + extra["key"]
 
                 if extra["key"] == "language":
@@ -412,6 +418,48 @@ def _graph_from_dataset_base(self, dataset_dict, dataset_ref):
             ]
             self._add_triples_from_dict(publisher_details, publisher_ref, items)
 
+        # Creator
+        creator_ref = None
+
+        if dataset_dict.get("creator"):
+            # Scheming publisher field: will be handled in a separate profile
+            pass
+        elif any(
+            [
+                self._get_dataset_value(dataset_dict, "creator_uri"),
+                self._get_dataset_value(dataset_dict, "creator_name"),
+            ]
+        ):
+            # Legacy creator_* extras
+            creator_uri = self._get_dataset_value(dataset_dict, "creator_uri")
+            creator_name = self._get_dataset_value(dataset_dict, "creator_name")
+            if creator_uri:
+                creator_ref = CleanedURIRef(creator_uri)
+            else:
+                # No creator_uri
+                creator_ref = BNode()
+
+            creator_details = {
+                "name": creator_name,
+                "email": self._get_dataset_value(dataset_dict, "creator_email"),
+                "url": self._get_dataset_value(dataset_dict, "creator_url"),
+                "type": self._get_dataset_value(dataset_dict, "creator_type"),
+                "identifier": self._get_dataset_value(dataset_dict, "creator_identifier"),
+            }
+
+        # Add to graph
+        if creator_ref:
+            g.add((creator_ref, RDF.type, FOAF.Agent))
+            g.add((dataset_ref, DCT.creator, creator_ref))  # Use DCT.creator for creator
+            items = [
+                ("name", FOAF.name, None, Literal),
+                ("email", FOAF.mbox, None, Literal),
+                ("url", FOAF.homepage, None, URIRef),
+                ("type", DCT.type, None, URIRefOrLiteral),
+                ("identifier", DCT.identifier, None, URIRefOrLiteral),
+            ]
+            self._add_triples_from_dict(creator_details, creator_ref, items)
+
         # Temporal
         start = self._get_dataset_value(dataset_dict, "temporal_start")
         end = self._get_dataset_value(dataset_dict, "temporal_end")