Checking entities visibility for collections

src/app.py

@@ -4618,14 +4618,13 @@ def get_collections(id):
     # Get the entity dict from cache if exists
     # Otherwise query against uuid-api and neo4j to get the entity dict if the id exists
     entity_dict = query_target_entity(id)
-    normalized_entity_type = entity_dict['entity_type']
     uuid = entity_dict['uuid']
     public_entity = True
 
-    if not schema_manager.entity_type_instanceof(normalized_entity_type, 'Dataset'):
-        abort_bad_req(f"Unsupported entity type of id {id}: {normalized_entity_type}")
+    entity_scope = _get_entity_visibility(normalized_entity_type=entity_dict['entity_type'],
+                                          entity_dict=entity_dict)
 
-    if entity_dict['status'].lower() != DATASET_STATUS_PUBLISHED:
+    if entity_scope == DataVisibilityEnum.NONPUBLIC:
         # Token is required and the user must belong to HuBMAP-READ group
         token = get_user_token(request, non_public_access_required=True)
         public_entity = False
@@ -5605,7 +5604,7 @@ def delete_cache(id):
         upload_dataset_uuids = schema_neo4j_queries.get_upload_datasets(neo4j_driver_instance, entity_uuid , 'uuid')
 
         # If the target entity is Datasets/Publication, delete the associated Collections cache, Upload cache
-        collection_uuids = schema_neo4j_queries.get_dataset_collections(neo4j_driver_instance, entity_uuid , 'uuid')
+        collection_uuids = schema_neo4j_queries.get_entity_collections(neo4j_driver_instance, entity_uuid , 'uuid')
         collection_dict = schema_neo4j_queries.get_publication_associated_collection(neo4j_driver_instance, entity_uuid)
         upload_dict = schema_neo4j_queries.get_dataset_upload(neo4j_driver_instance, entity_uuid)
 

src/schema/provenance_schema.yaml

@@ -444,8 +444,8 @@ ENTITIES:
         transient: true
         generated: true
         description: "A list of collections that this dataset belongs to. Will be returned in response"
-        on_read_trigger: get_dataset_collections
-        on_index_trigger: get_dataset_collections
+        on_read_trigger: get_entity_collections
+        on_index_trigger: get_entity_collections
         # No on_index_trigger to include collections in the OpenSearch document for a Dataset (from HM, we might need this for SN)
       upload:
         type: json_string # dict
@@ -888,6 +888,13 @@ ENTITIES:
         type: string
         required_on_create: true
         description: "A high level description of where this source originates from."
+      collections:
+        type: list
+        transient: true
+        generated: true
+        description: "A list of collections that this source belongs to. Will be returned in response"
+        on_read_trigger: get_entity_collections
+        on_index_trigger: get_entity_collections
 
 
   ############################################# Sample #############################################
@@ -1149,6 +1156,13 @@ ENTITIES:
         immutable: true
         description: "A boolean indicating if the Sample contains any Datasets in its descendants."
         on_index_trigger: get_contains_data
+      collections:
+        type: list
+        transient: true
+        generated: true
+        description: "A list of collections that this sample belongs to. Will be returned in response"
+        on_read_trigger: get_entity_collections
+        on_index_trigger: get_entity_collections
 
   ############################################# Upload #############################################
   # Eventually we might allow folks to bulk register Samples or Sources through the same mechanism.

src/schema/schema_neo4j_queries.py

@@ -662,14 +662,14 @@ def get_next_revision_uuid(neo4j_driver, uuid):
 
 
 """
-Get a list of associated collection uuids for a given dataset
+Get a list of associated collection uuids for a given entity
 
 Parameters
 ----------
 neo4j_driver : neo4j.Driver object
     The neo4j database connection pool
 uuid : str
-    The uuid of dataset
+    The uuid of entity
 property_key : str
     A target property key for result filtering
 
@@ -680,7 +680,7 @@ def get_next_revision_uuid(neo4j_driver, uuid):
 """
 
 
-def get_dataset_collections(neo4j_driver, uuid, property_key=None):
+def get_entity_collections(neo4j_driver, uuid, property_key=None):
     results = []
 
     if property_key:
@@ -692,7 +692,7 @@ def get_dataset_collections(neo4j_driver, uuid, property_key=None):
                  f"WHERE e.uuid = '{uuid}' "
                  f"RETURN apoc.coll.toSet(COLLECT(c)) AS {record_field_name}")
 
-    logger.info("======get_dataset_collections() query======")
+    logger.info("======get_entity_collections() query======")
     logger.info(query)
 
     with neo4j_driver.session() as session:
@@ -1996,7 +1996,8 @@ def get_tuplets(neo4j_driver, uuid, property_key=None):
 
 
 """
-Get all collections by uuid
+Get all collections by for a given entity uuid
+
 Parameters
 ----------
 neo4j_driver : neo4j.Driver object
@@ -2005,6 +2006,7 @@ def get_tuplets(neo4j_driver, uuid, property_key=None):
     The uuid of target entity 
 property_key : str
     A target property key for result filtering
+
 Returns
 -------
 list
@@ -2014,14 +2016,14 @@ def get_collections(neo4j_driver, uuid, property_key = None):
     results = []
 
     if property_key:
-        query = (f"MATCH (c:Collection)<-[:IN_COLLECTION]-(ds:Dataset) "
-                 f"WHERE ds.uuid='{uuid}' "
+        query = (f"MATCH (c:Collection)<-[:IN_COLLECTION]-(e:Entity) "
+                 f"WHERE e.uuid='{uuid}' "
                  # COLLECT() returns a list
                  # apoc.coll.toSet() reruns a set containing unique nodes
                  f"RETURN apoc.coll.toSet(COLLECT(c.{property_key})) AS {record_field_name}")
     else:
-        query = (f"MATCH (c:Collection)<-[:IN_COLLECTION]-(ds:Dataset) "
-                 f"WHERE ds.uuid='{uuid}' "
+        query = (f"MATCH (c:Collection)<-[:IN_COLLECTION]-(e:Entity) "
+                 f"WHERE e.uuid='{uuid}' "
                  # COLLECT() returns a list
                  # apoc.coll.toSet() reruns a set containing unique nodes
                  f"RETURN apoc.coll.toSet(COLLECT(c)) AS {record_field_name}")

src/schema/schema_triggers.py

@@ -799,7 +799,7 @@ def set_dataset_status_new(property_key, normalized_type, user_token, existing_d
     return property_key, 'New'
 
 
-def get_dataset_collections(property_key, normalized_type, user_token, existing_data_dict, new_data_dict):
+def get_entity_collections(property_key, normalized_type, user_token, existing_data_dict, new_data_dict):
     """Trigger event method of getting a list of collections for this new Dataset.
 
     Parameters
@@ -825,15 +825,15 @@ def get_dataset_collections(property_key, normalized_type, user_token, existing_
 
     if 'uuid' not in existing_data_dict:
         msg = create_trigger_error_msg(
-            "Missing 'uuid' key in 'existing_data_dict' during calling 'get_dataset_collections()' trigger method.",
+            "Missing 'uuid' key in 'existing_data_dict' during calling 'get_entity_collections()' trigger method.",
             existing_data_dict, new_data_dict
         )
         raise KeyError(msg)
 
     # No property key needs to filter the result
     # Get back the list of collection dicts
-    collections_list = schema_neo4j_queries.get_dataset_collections(schema_manager.get_neo4j_driver_instance(),
-                                                                    existing_data_dict['uuid'])
+    collections_list = schema_neo4j_queries.get_entity_collections(schema_manager.get_neo4j_driver_instance(),
+                                                                   existing_data_dict['uuid'])
     if collections_list:
         # Exclude datasets from each resulting collection
         # We don't want to show too much nested information