diff options
| -rw-r--r-- | scripts/load_phenotypes_to_db.py | 76 |
1 files changed, 74 insertions, 2 deletions
diff --git a/scripts/load_phenotypes_to_db.py b/scripts/load_phenotypes_to_db.py index d5d2d16..18f9792 100644 --- a/scripts/load_phenotypes_to_db.py +++ b/scripts/load_phenotypes_to_db.py @@ -3,14 +3,16 @@ import uuid import json import logging import argparse +import datetime from pathlib import Path from zipfile import ZipFile from typing import Any, Union +from urllib.parse import urljoin from functools import reduce, partial from MySQLdb.cursors import Cursor, DictCursor -from gn_libs import jobs, mysqldb, sqlite3 +from gn_libs import jobs, mysqldb, sqlite3, monadic_requests as mrequests from r_qtl import r_qtl2 as rqtl2 from uploader.species.models import species_by_id @@ -227,7 +229,77 @@ def cross_reference_phenotypes_publications_and_data( def update_auth(authserver, token, species, population, dataset, xrefdata): """Grant the user access to their data.""" - raise NotImplemented("Please implement this!") + # TODO Call into the auth server to: + # 1. Link the phenotypes with a user group + # - fetch group: http://localhost:8081/auth/user/group + # - link data to group: http://localhost:8081/auth/data/link/phenotype + # - *might need code update in gn-auth: remove restriction, perhaps* + # 2. Create resource (perhaps?) + # - Get resource categories: http://localhost:8081/auth/resource/categories + # - Create a new resource: http://localhost:80host:8081/auth/resource/create + # - single resource for all phenotypes + # - resource name from user, species, population, dataset, datetime? + # - User will have "ownership" of resource by default + # 3. Link data to the resource: http://localhost:8081/auth/resource/data/link + # - Update code to allow linking multiple items in a single request + _tries = 0 # TODO use this to limit how many tries before quiting and bailing + _delay = 1 + headers = { + "Authorization": f"Bearer {token}", + "Content-Type": "application/json" + } + def authserveruri(endpoint): + return urljoin(authserver, endpoint) + + def __fetch_user_details__(): + return mrequests.get( + authserveruri("/auth/user/"), + headers=headers + ) + + def __link_data__(user, group): + return mrequests.post( + authserveruri("/auth/data/link/phenotype"), + headers=headers, + data={ + "group_id": user["group"]["group_id"], + "traits": []# TODO: process these here + }).then(lambda ld_results: (user, ld_results)) + + def __fetch_phenotype_category_details__(user, linkeddata): + return mrequests.get( + authserveruri("/auth/resource/categories"), + headers=headers + ).then( + lambda categories: ( + user, + linkeddata, + next(category for category in categories + if category["resource_category_key"] == "phenotype")) + ) + + def __create_resource__(user, linkeddata, category): + now = datetime.datetime.now().isoformat() + return mrequests.post( + authserveruri("/auth/resource/create"), + headers=headers, + data={ + "resource_category": category["resource_category_id"], + "resource_name": (f"{user['email']}—{dataset['name']}—{now}—" + f"{len(xrefdata)} phenotypes"), + "public": "off" + }).then(lambda cr_results: (user, linkeddata, resource)) + + def __attach_data_to_resource__(user, linkeddata, resource): + return mrequests.post( + authserveruri("/auth/resource/data/link"), + headers=headers, + data={ + "dataset_type": "phenotype", + "resource_id": resource["resource_id"], + "data_link_ids": [item["data_link_id"] for item in linkeddata] + }).then(lambda attc: (user, linkeddata, resource, attc)) + raise NotImplementedError("Please implement this!") def load_data(conn: mysqldb.Connection, job: dict) -> int: |