This script demonstrates a DSS Elasticsearch query for bundles by project uuid.

find_bundles_for_project.py

#!/usr/bin/env python
import argparse

from hca.dss import DSSClient

def get_dss_client(deployment):
    if "prod" == deployment:
        url = "https://dss.data.humancellatlas.org/v1/swagger.json"
    else:
        url = f"https://dss.{deployment}.data.humancellatlas.org/v1/swagger.json"
    return DSSClient(swagger_url=url)

def find_bundles(client, replica, project_uuid):
    q = {
        "query": {
            "bool": {
                "must": [
                    {
                        "match": {
                            "files.project_json.provenance.document_id": project_uuid
                        }
                    }
                ]
            }
        }
    }
    for hit in client.post_search.iterate(replica=replica, es_query=q, output_format="raw"):
        yield hit

if __name__ == "__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument("--deployment", "-d", choices=["dev", "integration", "staging", "prod"], default="dev")
    parser.add_argument("--replica", "-r", choices=["aws", "gcp"], default="aws")
    parser.add_argument("project_uuid")
    args = parser.parse_args()
    client = get_dss_client(args.deployment)
    for hit in find_bundles(client, args.replica, args.project_uuid):
        fqid = "{}.{}".format(hit['metadata']['uuid'], hit['metadata']['manifest']['version'])
        print(fqid)