Access rights and index dataset without file

udata/core/dataset/api_fields.py

@@ -8,7 +8,8 @@
 
 from .constants import (
     UPDATE_FREQUENCIES, RESOURCE_FILETYPES, DEFAULT_FREQUENCY,
-    CHECKSUM_TYPES, DEFAULT_CHECKSUM_TYPE, DEFAULT_LICENSE, RESOURCE_TYPES
+    CHECKSUM_TYPES, DEFAULT_CHECKSUM_TYPE, DEFAULT_LICENSE,
+    RESOURCE_TYPES, ACCESS_RIGHTS, DEFAULT_ACCESS_RIGHTS
 )
 
 
@@ -250,6 +251,9 @@
     'quality': fields.Raw(description='The dataset quality', readonly=True),
     'last_update': fields.ISODateTime(
         description='The resources last modification date', required=True),
+    'access_right': fields.String(
+        description='The dataset access right', required=True,
+        enum=list(ACCESS_RIGHTS), default=DEFAULT_ACCESS_RIGHTS),
     'schema': fields.Nested(
         schema_fields, allow_null=True, description='Reference to the associated schema'),
     'internal': fields.Nested(

udata/core/dataset/constants.py

@@ -70,6 +70,8 @@
 
 PIVOTAL_DATA = 'pivotal-data'
 CLOSED_FORMATS = ('pdf', 'doc', 'docx', 'word', 'xls', 'excel', 'xlsx')
+ACCESS_RIGHTS = ('public', 'non-public', 'restricted')
+DEFAULT_ACCESS_RIGHTS = 'public'
 
 # Maximum acceptable Damerau-Levenshtein distance
 # used to guess license

udata/core/dataset/models.py

@@ -23,7 +23,7 @@
 from udata.utils import get_by, hash_url, to_naive_datetime
 from udata.uris import ValidationError, endpoint_for
 from udata.uris import validate as validate_url
-from .constants import CHECKSUM_TYPES, CLOSED_FORMATS, DEFAULT_LICENSE, LEGACY_FREQUENCIES, MAX_DISTANCE, PIVOTAL_DATA, RESOURCE_FILETYPES, RESOURCE_TYPES, SCHEMA_CACHE_DURATION, UPDATE_FREQUENCIES
+from .constants import CHECKSUM_TYPES, CLOSED_FORMATS, DEFAULT_LICENSE, LEGACY_FREQUENCIES, MAX_DISTANCE, PIVOTAL_DATA, RESOURCE_FILETYPES, RESOURCE_TYPES, SCHEMA_CACHE_DURATION, UPDATE_FREQUENCIES, ACCESS_RIGHTS, DEFAULT_ACCESS_RIGHTS
 
 from .preview import get_preview_url
 from .exceptions import (
@@ -34,6 +34,7 @@
 
 NON_ASSIGNABLE_SCHEMA_TYPES = ['datapackage']
 
+
 log = logging.getLogger(__name__)
 
 
@@ -491,6 +492,8 @@ class Dataset(WithMetrics, BadgeMixin, db.Owned, db.Document):
     deleted = db.DateTimeField()
     archived = db.DateTimeField()
 
+    access_rights = db.StringField(choices=list(ACCESS_RIGHTS), default=DEFAULT_ACCESS_RIGHTS, required=True)
+
     def __str__(self):
         return self.title or ''
 

udata/core/dataset/rdf.py

@@ -200,6 +200,7 @@ def dataset_to_rdf(dataset, graph=None):
     d.set(DCT.description, Literal(dataset.description))
     d.set(DCT.issued, Literal(dataset.created_at))
     d.set(DCT.modified, Literal(dataset.last_modified))
+    d.set(DCT.accessRights, Literal(dataset.access_rights))
 
     if dataset.acronym:
         d.set(SKOS.altLabel, Literal(dataset.acronym))
@@ -610,6 +611,7 @@ def dataset_from_rdf(graph: Graph, dataset=None, node=None):
     default_license = dataset.license or License.default()
     dataset_license = rdf_value(d, DCT.license)
     dataset.license = License.guess(dataset_license, *licenses, default=default_license)
+    dataset.access_rights = rdf_value(d, DCT.accessRights)
 
     identifier = rdf_value(d, DCT.identifier)
     uri = d.identifier.toPython() if isinstance(d.identifier, URIRef) else None

udata/core/dataset/search.py

@@ -45,8 +45,11 @@ class DatasetSearch(ModelSearchAdapter):
 
     @classmethod
     def is_indexable(cls, dataset):
+        valid_access_rights = (dataset.access_rights == 'public'
+                               or dataset.access_rights == 'non-public'
+                               or dataset.access_rights == 'restricted')
         return (dataset.deleted is None and dataset.archived is None and
-                not dataset.private)
+                not dataset.private and valid_access_rights)
 
     @classmethod
     def mongo_search(cls, args):

udata/tests/search/test_adapter.py

@@ -183,3 +183,12 @@ def test_index_model_from_datetime(self, mock_req):
         }
         url = f"{current_app.config['SEARCH_SERVICE_API_URL']}/datasets/index"
         mock_req.assert_called_with(url, json=expected_value)
+
+    def test_dataset_adapter(self):
+        public_dataset = DatasetFactory(id='61fd30cb29ea95c7bc0e1211', access_rights='public')
+        non_public_dataset = DatasetFactory(id='61fd30cb29ea95c7bc0e1211', access_rights='non-public')
+        restricted_dataset = DatasetFactory(id='61fd30cb29ea95c7bc0e1211', access_rights='restricted')
+
+        assert DatasetSearch.is_indexable(public_dataset)
+        assert DatasetSearch.is_indexable(non_public_dataset)
+        assert DatasetSearch.is_indexable(restricted_dataset)