Derek furst/sync component datasets

[DerekFurstPitt]

Changed the status property in provenance schema yaml. Now, its after update trigger method is update_status, which calls set status history, and then calls a new function sync_component_dataset_status. this works by creating new PUT calls to entity on each of the component datasets, updating their status to whatever the multi-assay dataset was. A new validator was also created that disallows directly updating the status of component datasets.

[yuanzhou]

I didn't test, at a quick glance INGEST_API_APP should be referenced using schemaConstants..

[yuanzhou]

Instead of defining this update_status() update_entity() in schema_manager, we should handle the business logic in the trigger method. Because the schema_manager should not care about updating the entity specifically.

We should add a getter method similar to

entity-api/src/schema/schema_manager.py

Lines 1690 to 1701 in 795b63d

	"""
	Get the ingest-api URL to be used by trigger methods
	Returns
	-------
	str
	The ingest-api URL
	"""
	def get_ingest_api_url():
	global _ingest_api_url
	return _ingest_api_url

so other modules can get a handle of the entity-api url and then make request inside the trigger method like this

entity-api/src/schema/schema_triggers.py

Line 1390 in 795b63d

response = requests.post(url = ingest_api_target_url, headers = schema_manager._create_request_headers(user_token), json = json_to_post, verify = False)

[yuanzhou]

Even though this is instructional, we should use localhost URL rather than DEV. Add some comment to explain this is the same URL base where the entity-api is running. Because if someone uses the DEV URL and runs the entity-api locally, it'll end up updating the component datasets in DEV neo4j from the local trigger requests. The key is to keep it consistent. And once deployed on DEV/TEST/PROD, we can still use a localhost:port where the entity-api runs on the VM.

[DerekFurstPitt]

@yuanzhou I applied the requested fixes. Take a look when you get a chance