Merge branch 'multiple-agents' into dcat-us-3

ckanext/dcat/profiles/base.py

@@ -426,7 +426,7 @@ def _insert_or_update_temporal(self, dataset_dict, key, value):
         else:
             dataset_dict["extras"].append({"key": key, "value": value})
 
-    def _agents_details(self, subject, predicate, first_only=False):
+    def _agents_details(self, subject, predicate):
         """
         Returns a list of dicts with details about a foaf:Agent property, e.g.
         dct:publisher or dct:creator entity.
@@ -466,18 +466,18 @@ def _agents_details(self, subject, predicate, first_only=False):
 
     def _contact_details(self, subject, predicate):
         """
-        Returns a dict with details about a vcard expression
+        Returns a list of dicts with details about vcard expressions
 
         Both subject and predicate must be rdflib URIRef or BNode objects
 
         Returns keys for uri, name and email with the values set to
         an empty string if they could not be found
         """
 
-        contact = {}
-
+        contacts = []
         for agent in self.g.objects(subject, predicate):
 
+            contact = {}
             contact["uri"] = str(agent) if isinstance(agent, URIRef) else ""
 
             contact["name"] = self._get_vcard_property_value(
@@ -489,8 +489,9 @@ def _contact_details(self, subject, predicate):
             )
 
             contact["identifier"] = self._get_vcard_property_value(agent, VCARD.hasUID)
+            contacts.append(contact)
 
-        return contact
+        return contacts
 
     def _parse_geodata(self, spatial, datatype, cur_value):
         """

ckanext/dcat/profiles/euro_dcat_ap_base.py

@@ -108,36 +108,43 @@ def _parse_dataset_base(self, dataset_dict, dataset_ref):
                 dataset_dict["extras"].append({"key": key, "value": json.dumps(values)})
 
         # Contact details
-        contact = self._contact_details(dataset_ref, DCAT.contactPoint)
-        if not contact:
-            # adms:contactPoint was supported on the first version of DCAT-AP
-            contact = self._contact_details(dataset_ref, ADMS.contactPoint)
-
-        if contact:
-            for key in ("uri", "name", "email", "identifier"):
-                if contact.get(key):
-                    dataset_dict["extras"].append(
-                        {"key": "contact_{0}".format(key), "value": contact.get(key)}
-                    )
-
-        # Publisher
-        publishers = self._agents_details(dataset_ref, DCT.publisher)
-        if publishers:
-            publisher = publishers[0]
-            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)}
-                    )
+        if self._schema_field("contact"):
+            # This is a scheming field, will be hanlded in a separate profile
+            pass
+        else:
+            contact = self._contact_details(dataset_ref, DCAT.contactPoint)
+            if not contact:
+                # adms:contactPoint was supported on the first version of DCAT-AP
+                contact = self._contact_details(dataset_ref, ADMS.contactPoint)
+            if contact:
+                contact = contact[0]
+                for key in ("uri", "name", "email", "identifier"):
+                    if contact.get(key):
+                        dataset_dict["extras"].append(
+                            {
+                                "key": "contact_{0}".format(key),
+                                "value": contact.get(key)
+                            }
+                        )
 
-        # Creator
-        creators = self._agents_details(dataset_ref, DCT.creator)
-        for creator in creators:
-            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)}
-                    )
+        # Publishers and creators
+        for item in [("publisher", DCT.publisher), ("creator", DCT.creator)]:
+            agent_key, predicate = item
+            if self._schema_field(agent_key):
+                # This is a scheming field, will be hanlded in a separate profile
+                pass
+            else:
+                agents = self._agents_details(dataset_ref, predicate)
+                if agents:
+                    agent = agents[0]
+                    for key in ("uri", "name", "email", "url", "type", "identifier"):
+                        if agent.get(key):
+                            dataset_dict["extras"].append(
+                                {
+                                    "key": f"{agent_key}_{key}",
+                                    "value": agent.get(key)
+                                }
+                            )
 
         # Temporal
         start, end = self._time_interval(dataset_ref, DCT.temporal)

ckanext/dcat/profiles/euro_dcat_ap_scheming.py

@@ -100,6 +100,18 @@ def _parse_list_value(data_dict, field_name):
                     dataset_dict[field_name] = [new_dict]
                     dataset_dict["extras"] = new_extras
 
+        # Contact details
+        contacts = self._contact_details(dataset_ref, DCAT.contactPoint)
+        if contacts:
+            dataset_dict["contact"] = contacts
+
+        # Publishers and creators
+        for item in [("publisher", DCT.publisher), ("creator", DCT.creator)]:
+            key, predicate = item
+            agents = self._agents_details(dataset_ref, predicate)
+            if agents:
+                dataset_dict[key] = agents
+
         # Repeating subfields: resources
         for schema_field in self._dataset_schema["resource_fields"]:
             if "repeating_subfields" in schema_field:
@@ -157,7 +169,7 @@ def _graph_from_dataset_v2_scheming(self, dataset_dict, dataset_ref):
                     _type=URIRefOrLiteral,
                 )
 
-        self._add_agents(dataset_ref, dataset_dict, "publisher", DCT.publisher, first_only=True)
+        self._add_agents(dataset_ref, dataset_dict, "publisher", DCT.publisher)
         self._add_agents(dataset_ref, dataset_dict, "creator", DCT.creator)
 
         temporal = dataset_dict.get("temporal_coverage")

ckanext/dcat/tests/profiles/base/test_base_profile.py

@@ -721,6 +721,8 @@ def test_contact_details(self):
 
         contact = p._contact_details(URIRef('http://example.org'), ADMS.contactPoint)
 
+        contact = contact[0]
+
         assert contact['name'] == 'Point of Contact'
         # mailto gets removed for storage and is added again on output
         assert contact['email'] == '[email protected]'

ckanext/dcat/tests/profiles/dcat_ap_2/test_scheming_support.py

@@ -190,7 +190,7 @@ def test_e2e_ckan_to_dcat(self):
             g,
             publisher[0][2],
             DCT.identifier,
-            URIRef(dataset_dict["publisher"][0]["identifier"])
+            URIRef(dataset_dict["publisher"][0]["identifier"]),
         )
 
         creator = [t for t in g.triples((dataset_ref, DCT.creator, None))]
@@ -221,10 +221,9 @@ def test_e2e_ckan_to_dcat(self):
             g,
             creator[0][2],
             DCT.identifier,
-            URIRef(dataset_dict["creator"][0]["identifier"])
+            URIRef(dataset_dict["creator"][0]["identifier"]),
         )
 
-
         temporal = [t for t in g.triples((dataset_ref, DCT.temporal, None))]
 
         assert len(temporal) == len(dataset["temporal_coverage"])
@@ -275,8 +274,8 @@ def test_e2e_ckan_to_dcat(self):
 
         # Statements
         for item in [
-            ('access_rights', DCT.accessRights),
-            ('provenance', DCT.provenance),
+            ("access_rights", DCT.accessRights),
+            ("provenance", DCT.provenance),
         ]:
             statement = [s for s in g.objects(dataset_ref, item[1])][0]
             assert self._triple(g, statement, RDFS.label, dataset[item[0]])
@@ -402,7 +401,7 @@ def test_e2e_ckan_to_dcat(self):
 
         # Resources: statements
         statement = [s for s in g.objects(distribution_ref, DCT.rights)][0]
-        assert self._triple(g, statement, RDFS.label, resource['rights'])
+        assert self._triple(g, statement, RDFS.label, resource["rights"])
 
     def test_publisher_fallback_org(self):
 
@@ -856,6 +855,116 @@ def test_statement_literal(self):
         assert dataset["notes"] == "This is a dataset"
         assert dataset["access_rights"] == "Some statement"
 
+    def test_multiple_contacts(self):
+
+        data = """
+        @prefix dcat: <http://www.w3.org/ns/dcat#> .
+        @prefix dct: <http://purl.org/dc/terms/> .
+        @prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
+        @prefix vcard: <http://www.w3.org/2006/vcard/ns#> .
+
+        <https://example.com/dataset1>
+          a dcat:Dataset ;
+          dct:title "Dataset 1" ;
+          dct:description "This is a dataset" ;
+            dcat:contactPoint [ a vcard:Kind ;
+                vcard:fn "Test Contact 1" ;
+                vcard:hasEmail <mailto:[email protected]> ],
+            [ a vcard:Kind ;
+                vcard:fn "Test Contact 2" ;
+                vcard:hasEmail <mailto:[email protected]> ] ;
+        .
+        """
+
+        p = RDFParser()
+
+        p.parse(data, _format="ttl")
+        datasets = [d for d in p.datasets()]
+
+        dataset = datasets[0]
+        assert len(dataset["contact"]) == 2
+        assert dataset["contact"][0]["name"] == "Test Contact 1"
+        assert dataset["contact"][0]["email"] == "[email protected]"
+        assert dataset["contact"][1]["name"] == "Test Contact 2"
+        assert dataset["contact"][1]["email"] == "[email protected]"
+
+    def test_multiple_publishers(self):
+
+        data = """
+        @prefix dcat: <http://www.w3.org/ns/dcat#> .
+        @prefix dct: <http://purl.org/dc/terms/> .
+        @prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
+        @prefix org: <http://www.w3.org/ns/org#> .
+        @prefix skos: <http://www.w3.org/2004/02/skos/core#> .
+        @prefix foaf: <http://xmlns.com/foaf/0.1/> .
+        @prefix vcard: <http://www.w3.org/2006/vcard/ns#> .
+
+        <https://example.com/dataset1>
+          a dcat:Dataset ;
+          dct:title "Dataset 1" ;
+          dct:description "This is a dataset" ;
+          dct:publisher [ a org:Organization ;
+                    skos:prefLabel "Test Publisher 1" ;
+                    vcard:hasEmail <mailto:[email protected]> ;
+                    foaf:name "Test Publisher 1" ],
+                    [ a org:Organization ;
+                    skos:prefLabel "Test Publisher 2" ;
+                    vcard:hasEmail <mailto:[email protected]> ;
+                    foaf:name "Test Publisher 2" ] ;
+        .
+        """
+
+        p = RDFParser()
+
+        p.parse(data, _format="ttl")
+        datasets = [d for d in p.datasets()]
+
+        dataset = datasets[0]
+        assert len(dataset["publisher"]) == 2
+        assert dataset["publisher"][0]["name"] == "Test Publisher 1"
+        assert dataset["publisher"][0]["email"] == "[email protected]"
+        assert dataset["publisher"][1]["name"] == "Test Publisher 2"
+        assert dataset["publisher"][1]["email"] == "[email protected]"
+
+    def test_multiple_creators(self):
+
+        data = """
+        @prefix dcat: <http://www.w3.org/ns/dcat#> .
+        @prefix dct: <http://purl.org/dc/terms/> .
+        @prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
+        @prefix org: <http://www.w3.org/ns/org#> .
+        @prefix skos: <http://www.w3.org/2004/02/skos/core#> .
+        @prefix foaf: <http://xmlns.com/foaf/0.1/> .
+        @prefix vcard: <http://www.w3.org/2006/vcard/ns#> .
+
+        <https://example.com/dataset1>
+          a dcat:Dataset ;
+          dct:title "Dataset 1" ;
+          dct:description "This is a dataset" ;
+          dct:creator [ a org:Organization ;
+                    skos:prefLabel "Test Creator 1" ;
+                    vcard:hasEmail <mailto:[email protected]> ;
+                    foaf:name "Test Creator 1" ],
+                    [ a org:Organization ;
+                    skos:prefLabel "Test Creator 2" ;
+                    vcard:hasEmail <mailto:[email protected]> ;
+                    foaf:name "Test Creator 2" ] ;
+        .
+        """
+
+        p = RDFParser()
+
+        p.parse(data, _format="ttl")
+        datasets = [d for d in p.datasets()]
+
+        dataset = datasets[0]
+        assert len(dataset["creator"]) == 2
+        assert dataset["creator"][0]["name"] == "Test Creator 1"
+        assert dataset["creator"][0]["email"] == "[email protected]"
+        assert dataset["creator"][1]["name"] == "Test Creator 2"
+        assert dataset["creator"][1]["email"] == "[email protected]"
+
+
 @pytest.mark.usefixtures("with_plugins", "clean_db", "clean_index")
 @pytest.mark.ckan_config("ckan.plugins", "dcat scheming_datasets")
 @pytest.mark.ckan_config(