Derek furst/get associated organs from dataset #254
Conversation
get_associated_organs_from_dataset(id). Added the associated method in app_neo4j_queries.py that implements the neo4j query.
individual organ. An organ will always be public if a dataset created from it is public.
| return jsonify(results) | ||
|
|
||
|
|
||
| @app.route('/datasets/<id>/organs', methods=['GET']) |
There was a problem hiding this comment.
Can you add a comment block above this function like the rest with description, input parameters, and return?
src/app.py
Outdated
|
|
||
| # If a dataset is published/public, then the organs associated to it will by definition also be public. So no | ||
| # further validation is needed on the individual organs. | ||
| return jsonify(associated_organs) |
There was a problem hiding this comment.
@shirey do we just need to return a list of Sample organ uuids? If a list of normalized Sample dicts are supposed to be the result, @DerekFurstPitt should following the result handling in get_ancestor_organs(id).
src/app_neo4j_queries.py
Outdated
| """ | ||
| def get_associated_organs_from_dataset(neo4j_driver, dataset_uuid): | ||
| query = (f"MATCH (ds:Dataset {{uuid:'{dataset_uuid}'}})<-[*]-(organ:Sample {{specimen_type:'organ'}}) " | ||
| f"RETURN distinct organ.uuid") |
There was a problem hiding this comment.
If there are multiple associated Sample organs found, then the Cypher query should return a Set (Collection with no duplicates). Either a set of uuids or nodes (need to be converted to dicts). And the return data is determined by the need of that API call -- either a list of uuids or normalized dicts of sample details. Probably dicts since there won't be too many organs associated with a given dataset.
src/app_neo4j_queries.py
Outdated
| query = (f"MATCH (ds:Dataset {{uuid:'{dataset_uuid}'}})<-[*]-(organ:Sample {{specimen_type:'organ'}}) " | ||
| f"RETURN distinct organ.uuid") | ||
|
|
||
| logger.debug("======get_associated_organs_from_sample() query======") |
There was a problem hiding this comment.
Need use the same function name to be consistent.
…aks to app.py get_associated_organs accordingly. Also added comment block above that method.
|
Requested changes made |
|
@DerekFurstPitt there are still two things need to be addressed:
|
Applied normalization to response before returning in app.py get_associated_organs_from_dataset(id)
Implemented method get_asociated_organs_from_dataset(id) at endpoint /datasets//organs.
Takes an id for a dataset and runs a neo4j query to return every organ associated with this dataset. If no organ is found, given id is not to a dataset, user has bad token, or user has insufficient permissions, it fails.