Keystone Python Client

The official Python client for the Keystone API.

A quickstart guide for project developers is provided below. For the full Keystone project documentation, see keystone.bhpc.dev.

Developer Setup

This project uses the Poetry package manager. To initialize a new environment, start by defining your desired Python interpreter:

poetry env use python3

Next, install the project and its dependencies. The dev dependency group is optional, but includes useful utilities for running application tests:

poetry install --extras dev

Common Tasks

Running Tests

Application tests are organized by testing strategy. Unit tests are used to verify components in isolation and require no external services. Function tests validate package integration against a live API server and require a running Keystone API instance.

Unit Tests

Unit tests are executed using the standard unittest library:

python -m unittest discover -s ./tests/unit_tests

Alternatively, the coverage utility can also be used to execute tests and report the resulting coverage:

coverage run -m unittest discover -s ./tests/unit_tests
coverage report

Function Tests

Function tests are used to validate the client against a running API server. The following environmental variables are used to configure connection settings against the upstream API:

Variable	Description	Default
TEST_API_HOST	URL of the Keystone API instance to test against.	http://localhost:8000
TEST_API_USER	Username used to authenticate with the API.	admin
TEST_API_PASSWORD	Password used to authenticate with the API.	quickstart

Once the target API instance is configured and running, function tests are executed using the standard library unittest framework:

python -m unittest discover -s ./tests/function_tests