Add multiple users to a Faculty platform project

01-README.md

Add multiple users by email to a Faculty project

This script adds multiple users by email to a Faculty project. The users need to exist in the platform: it does not invite users.

To use this:

1. Create the project you want to invite users to.
2. Create a Jupyter server in that project.
3. Download add-users-by-email.py in the project workspace by running, e.g.

wget 'https://gist.githubusercontent.com/pbugnion/5edc437c1a031f4a28e03d5d16ce4495/raw/add-users-by-email.py'

4. Create a file with the emails to add, with one email per line.
5. Run python add-users-by-email.py /PATH/TO/EMAIL_FILE --verbose to add the users.

Notes:

• If any user is an admin in the project, their role will be overwritten to data scientists.
• Do not include yourself in the list of emails, since this will remove your admin privilege.
• If any emails in the list do not correspond to users, you will see a warning

add-users-by-email.py

import argparse
import logging
import os

import faculty

# Roles to give to users as they are added
ROLES = ["project-data-scientist", "project-observer", "project-data-provider"]


DESCRIPTION = """\
Add users to a Faculty project in bulk.

This script reads a file containing emails of users, with one email per line.
It then adds these users as data scientists in the current project, provided
the users exist in the platform.
"""


def parse_arguments():
    parser = argparse.ArgumentParser(description=DESCRIPTION)
    parser.add_argument(
        "email_file", 
        type=argparse.FileType(mode="r"), 
        help="file containing emails to add, with one email per file."
    )
    
    parser.add_argument("-v", "--verbose", action="store_true", help="increase verbosity")
    args = parser.parse_args()
    return args


def get_users_for_emails(user_emails):
    user_client = faculty.client("user")
    all_users = user_client.get_all_users(is_system=False, enabled=True)
    email_to_user = {user.email: user for user in all_users}
    found_users = [email_to_user[email] for email in user_emails if email in email_to_user]
    missing_emails = [email for email in user_emails if email not in email_to_user]
    return found_users, missing_emails
            
            
def add_user_to_project(user_id, project_id):
    project_client = faculty.client("project")
    response = project_client._post_raw(f"/project/{project_id}/member/{user_id}", json={"roles": ROLES})
    response.raise_for_status()


def main(email_file, project_id):
    emails = [line.strip() for line in email_file]
    logging.info(f"Found {len(emails)} emails in input file.")
    found_users, missing_emails = get_users_for_emails(emails)
    logging.info(f"Out of the emails in the input, {len(found_users)} were in Faculty database.")
    for email in missing_emails:
        logging.warn(f"No user with email {email} found in Faculty database.")
    for user in found_users:
        add_user_to_project(user.id, project_id)
        logging.info(f"Added user {user.email} to project {project_id}.")


if __name__ == "__main__":
    args = parse_arguments()
    if args.verbose:
        logging.basicConfig(level=logging.INFO)
    project_id = os.getenv("FACULTY_PROJECT_ID")
    if project_id is None:
        logging.error(
            "Missing environment variable FACULTY_PROJECT_ID. "
            "This script should be run in the project you want to add users to."
        )
    else:
        logging.info(f"Adding users to project {project_id}.")
    logging.info(f"Using {args.email_file.name} as input.")
    main(args.email_file, project_id)