jupyterAuth

A plugin module for authenticate in different scopes using Python Jupyter Notebook as base and a given IDP.

Project structure

The project is structured as follows:

• /jupyterAuth - Contains all source code.
• /tests - Contains all unit tests (unittest).
• requirements.txt - This file contains all the modules required for the application.
• pyproject.toml - This file contains the instructions for build the module when will be imported from pip
• README.md - This file, contains internal developer documentation.

Prerequisites

• Python >= 3.10

Parameters

All parameters for create an Authorizer object are:

• Instantiation argument: <oauth address endpoint> , <scope> , <disable https> , <service host>
  • oauth address endpoint (required - string): The address of the IDP endopoint that provide the authentication
  • scope (required - string): The service that you want to authenticate e.g. "kafka", "opensearch"
  • disable https (required - bool): If True authentication credentials will be transferred by HTTP
  • service host (optional - string): The address of the service that you want to authenticate (only for "opensearch")

All parameters for use an Authorizer with getJWToken and getOauth2 are:

• Function argument: <username> , <password>
  • username (required - string)
  • password (required - string)

Examples

To create an Authorizer object pre-setted within Kafka in local environment (using HTTP) without using Prometheus:

authorizer_object = Authorizer("http://localhost:1852/realms/local-development/protocol/openid-connect/token", "kafka", True)

To use the Authorizer for get a JWToken given the credentials

token = authorizer_object.getJWToken("username", "password")

Warn

• Can work in local environment with HTTP and so the authentication credentials will be transferred in plaintext.