Remove logic mapping Publication entities into Dataset entities, and …

…support for annotation related to "subclassing". Revise ERD and database generation scripts to remove annotation data related to "subclassing".

sql/README.md

@@ -32,6 +32,9 @@ After setting the connection parameters and clicking "Next", check the following
 ````
 Omit schema qualifier in object names
 ````
+````
+Include model attached scripts
+````
 and uncheck the following option on the same screen
 ````
 Generate USE statements
@@ -44,7 +47,15 @@ and
 ````
 Export MySQL Routine Objects
 ````
-Click "Next".  Examine the generated SQL script to assure it is independent of the database it is being executed in. Use the buttons or clipboard to save this file as `uuid-api.sql`.
+Click "Next".  Examine the generated SQL script to assure it is independent of the database it is being executed in. For example, statements should look like
+
+````CREATE TABLE IF NOT EXISTS `uuids`...````
+
+and **_not_**
+
+````CREATE TABLE IF NOT EXISTS `hm_dev_uuids`.`uuids`...````
+
+Use the buttons or clipboard to save this file as `uuid-api.sql`.
 
 Click "Cancel" to abandon the rest of the "forward engineer" process.
 

sql/uuid-api.sql

@@ -115,7 +115,6 @@ CREATE TABLE IF NOT EXISTS `uuids_attributes` (
   `BASE_ID` VARCHAR(10) NULL DEFAULT NULL,
   `SUBMISSION_ID` VARCHAR(170) NULL DEFAULT NULL,
   `DESCENDANT_COUNT` INT NULL DEFAULT '0',
-  `ENTITY_CLASS_METADATA` JSON NULL DEFAULT NULL,
   PRIMARY KEY (`UUID`),
   UNIQUE INDEX `UNQ_HM_SUBMISSION_ID` (`SUBMISSION_ID` ASC) VISIBLE,
   UNIQUE INDEX `UNQ_BASE_ID` (`BASE_ID` ASC) VISIBLE,

src/instance/app.cfg.example

@@ -16,14 +16,11 @@ APP_ID = 'hubmap_id'
 APP_BASE_ID = 'base_id'
 
 BASE_DIR_TYPES = ['DATA_UPLOAD', 'INGEST_PORTAL_UPLOAD']
-BASE_ID_ENTITY_TYPES = ['ACTIVITY', 'SAMPLE', 'DONOR', 'DATASET', 'COLLECTION', 'UPLOAD', 'REFERENCE', 'AVR']
+# Entity types which have a uuids_attributes.BASE_ID associated with them
+BASE_ID_ENTITY_TYPES = ['ACTIVITY', 'SAMPLE', 'DONOR', 'DATASET', 'COLLECTION', 'UPLOAD', 'REFERENCE', 'AVR', 'PUBLICATION']
+# A full list of values found in uuids.ENTITY_TYPE, currently partitioned into those with a BASE_ID and those without.
 KNOWN_ENTITY_TYPES = BASE_ID_ENTITY_TYPES + ['FILE', 'LAB']
-MAPPED_ENTITY_TYPES = {
-    'PUBLICATION': {    'entity_type': 'DATASET'
-                        ,'entity_class': 'Publication'
-                        ,'entity_class_JSON': '{"class": "Publication"}'}
-}
-ANCESTOR_REQUIRED_ENTITY_TYPES = ['SAMPLE', 'DONOR', 'DATASET', 'FILE', 'UPLOAD']
+ANCESTOR_REQUIRED_ENTITY_TYPES = ['SAMPLE', 'DONOR', 'DATASET', 'FILE', 'UPLOAD', 'PUBLICATION']
 MULTIPLE_ALLOWED_ORGANS = ['LV', 'LY', 'SK', 'BD', 'BM']
 
 # Only for HubMap, not SenNet

src/instance/sennet.app.cfg.example

@@ -16,11 +16,11 @@ APP_ID = 'sennet_id'
 APP_BASE_ID = 'base_id'
 
 BASE_DIR_TYPES = ['DATA_UPLOAD', 'INGEST_PORTAL_UPLOAD']
-BASE_ID_ENTITY_TYPES = ['ACTIVITY', 'SAMPLE', 'SOURCE', 'DATASET', 'COLLECTION', 'UPLOAD', 'REFERENCE', 'AVR']
+# Entity types which have a uuids_attributes.BASE_ID associated with them
+BASE_ID_ENTITY_TYPES = ['ACTIVITY', 'SAMPLE', 'SOURCE', 'DATASET', 'COLLECTION', 'UPLOAD', 'REFERENCE', 'AVR', 'PUBLICATION']]
+# A full list of values found in uuids.ENTITY_TYPE, currently partitioned into those with a BASE_ID and those without.
 KNOWN_ENTITY_TYPES = BASE_ID_ENTITY_TYPES + ['FILE', 'LAB']
-MAPPED_ENTITY_TYPES = {
-}
-ANCESTOR_REQUIRED_ENTITY_TYPES = ['SAMPLE', 'SOURCE', 'DATASET', 'FILE', 'UPLOAD']
+ANCESTOR_REQUIRED_ENTITY_TYPES = ['SAMPLE', 'SOURCE', 'DATASET', 'FILE', 'UPLOAD', 'PUBLICATION']
 MULTIPLE_ALLOWED_ORGANS = ['LY', 'SK', 'BD', 'BM']
 
 # Only for HubMap, not SenNet

src/uuid_worker.py

@@ -50,9 +50,9 @@ class EntityTypeUUIDPrefix(Enum):
 
 SQL_INSERT_UUIDS_ATTRIBUTES_DEFAULT_DESC_COUNT = \
     ("INSERT INTO uuids_attributes "
-     " (UUID, BASE_ID, SUBMISSION_ID, ENTITY_CLASS_METADATA)"
+     " (UUID, BASE_ID, SUBMISSION_ID)"
      " VALUES"
-     " (%s, %s, %s, %s)"
+     " (%s, %s, %s)"
      )
 
 SQL_INSERT_ANCESTORS = \
@@ -284,7 +284,6 @@ def __init__(self, app_config=None):
         global BASE_DIR_TYPES
         global BASE_ID_ENTITY_TYPES
         global KNOWN_ENTITY_TYPES
-        global MAPPED_ENTITY_TYPES
         global ANCESTOR_REQUIRED_ENTITY_TYPES
         global MULTIPLE_ALLOWED_ORGANS
         global SUBMISSION_ID_ENTITY_TYPES
@@ -310,7 +309,6 @@ def __init__(self, app_config=None):
             BASE_DIR_TYPES = app_config['BASE_DIR_TYPES']
             BASE_ID_ENTITY_TYPES = app_config['BASE_ID_ENTITY_TYPES']
             KNOWN_ENTITY_TYPES = app_config['KNOWN_ENTITY_TYPES']
-            MAPPED_ENTITY_TYPES = app_config['MAPPED_ENTITY_TYPES']
             ANCESTOR_REQUIRED_ENTITY_TYPES = app_config['ANCESTOR_REQUIRED_ENTITY_TYPES']
             MULTIPLE_ALLOWED_ORGANS = app_config['MULTIPLE_ALLOWED_ORGANS']
             SUBMISSION_ID_ENTITY_TYPES = app_config['SUBMISSION_ID_ENTITY_TYPES']
@@ -345,15 +343,6 @@ def __init__(self, app_config=None):
                 raise Exception("Configuration parameter APP_CLIENT_ID not valid.")
             if not clientSecret:
                 raise Exception("Configuration parameter APP_CLIENT_SECRET not valid.")
-
-            # Verify the JSON elements of the configuration, which will be stored in
-            # MySQL rows of type 'json', contain valid JSON.
-            for entity_type in MAPPED_ENTITY_TYPES.keys():
-                try:
-                    json.loads(MAPPED_ENTITY_TYPES[entity_type]['entity_class_JSON'])
-                except (ValueError, TypeError):
-                    raise Exception(f"MAPPED_ENTITY_TYPES[{entity_type}] configured with invalid JSON:"
-                                    f" {MAPPED_ENTITY_TYPES[entity_type]['entity_class_JSON']}.")
         except KeyError as ke:
             self.logger.error("Expected configuration failed to load %s from app_config=%s.",ke,app_config)
             raise Exception("Expected configuration failed to load. See the logs.")
@@ -445,19 +434,9 @@ def uuidPost(self, req, nIds):
 
         # Verify we know how to handle the entity type provided
         entityType = content['entity_type'].upper().strip()
-        if entityType not in KNOWN_ENTITY_TYPES and entityType not in MAPPED_ENTITY_TYPES.keys():
+        if entityType not in KNOWN_ENTITY_TYPES:
             return Response(f"Unrecognized entity type {content['entity_type']}.", 400)
 
-        # If this is an entity type which requires transformation, it is done here so
-        # all the business logic associated with the actual entity type and with the
-        # entity class are executed before persisted to the database.
-        # E.G. "publication" entities become "dataset" entities of class "publication"
-        entity_class = MAPPED_ENTITY_TYPES[entityType]['entity_class'] if entityType in MAPPED_ENTITY_TYPES.keys() else None
-        entity_class_JSON = MAPPED_ENTITY_TYPES[entityType]['entity_class_JSON'] if entityType in MAPPED_ENTITY_TYPES.keys() else None
-        # now that entity_class and entity_class_JSON have been set based on entityType, remap entityType to
-        # database value, if needed.
-        entityType = MAPPED_ENTITY_TYPES[entityType]['entity_type'] if entityType in MAPPED_ENTITY_TYPES.keys() else entityType
-
         organ_code = None
         if 'organ_code' in content and not isBlank(content['organ_code']):
             if not entityType == 'SAMPLE':
@@ -523,8 +502,7 @@ def uuidPost(self, req, nIds):
                 return (Response("Parent id " + parentId + " does not exist", 400))
 
         return self.newUUIDs(ancestor_ids, entityType, userId, userEmail, nIds, organ_code=organ_code,
-                             lab_code=lab_code, file_info_array=file_info, base_dir_type=base_dir,
-                             entity_class=entity_class, entity_class_JSON=entity_class_JSON)
+                             lab_code=lab_code, file_info_array=file_info, base_dir_type=base_dir)
 
     '''
     def uuidPut(self, req):
@@ -659,7 +637,7 @@ def __create_submission_ids(self, num_to_gen, parent_id, entity_type, organ_code
     # generate multiple ids, one for each display id in the displayIds array
 
     def newUUIDs(self, parentIDs, entityType, userId, userEmail, nIds, organ_code=None, lab_code=None,
-                 file_info_array=None, base_dir_type=None, entity_class=None, entity_class_JSON=None):
+                 file_info_array=None, base_dir_type=None):
 
         gen_base_ids = entityType in BASE_ID_ENTITY_TYPES
         returnIds = []
@@ -682,8 +660,7 @@ def newUUIDs(self, parentIDs, entityType, userId, userEmail, nIds, organ_code=No
             # the UUID row that will be created.
             # N.B. descendant_count is tied to gen_submission_ids i.e. will not need a
             #      row to support descendant_count unless submission_id is populated.
-            insert_attributes_for_uuid = gen_base_ids or gen_submission_ids or \
-                                         entity_class == 'Publication'
+            insert_attributes_for_uuid = gen_base_ids or gen_submission_ids
 
             # Loop to generate the number of entityType requested in the nIds argument, in batches of
             # size MAX_GEN_IDS.
@@ -748,11 +725,9 @@ def newUUIDs(self, parentIDs, entityType, userId, userEmail, nIds, organ_code=No
                     insertVals.append((insUuid, entityType, awareUTCTimeNow, userId, userEmail))
 
                     if insert_attributes_for_uuid:
-                        insertAttribVals.append((insUuid
-                                                 , ins_app_base_id
-                                                 , submission_ids[n] if gen_submission_ids else None
-                                                 , entity_class_JSON if entity_class_JSON else None))
-
+                        insertAttribVals.append(    (insUuid
+                                                     ,ins_app_base_id
+                                                     ,submission_ids[n] if gen_submission_ids else None))
                     if store_file_info:
                         info_idx = i + n
                         file_path = file_info_array[info_idx]['path']