pytest-apibean

pytest-apibean is a pytest plugin that provides a unified, reusable testing foundation for projects built on the apibean ecosystem.

It is designed to work hand-in-hand with apibean-client to test API-driven services such as:

• apibean-rest – RESTful API layers
• apibean-datacore – backend data services and gateways
• Other apibean-based microservices and integrations

Why pytest-apibean?

Testing API-centric systems often leads to repeated boilerplate: creating HTTP clients, handling authentication, managing tokens, configuring base URLs, and wiring environment-specific settings.

pytest-apibean solves this by providing:

• A preconfigured httpx client for async API testing
• Authentication and token management fixtures
• Autouse setup hooks for consistent test environments
• Centralized configuration via pyproject.toml
• Seamless integration with apibean-client abstractions

All without forcing application-specific logic into your test suite.

apibean-client as the testing backbone

Rather than testing APIs using raw HTTP calls alone, pytest-apibean encourages a client-oriented testing model:

• Business logic is exercised through apibean-client
• Transport details remain encapsulated
• Tests become more stable, expressive, and future-proof

This approach aligns unit tests, integration tests, and contract tests around the same client interface used by real consumers.

Configuration-first design

pytest-apibean follows a pyproject-first philosophy.

All test-related options can be configured via:

[tool.pytest.apibean.options]
base_url = "http://localhost:8080/api/v1"
login_path = "/auth/login"
timeout = 10.0
auto_login = true

Configuration precedence is explicit and predictable:

1. Test fixtures
2. pyproject.toml
3. Environment variables
4. Built-in defaults

Designed for real-world testing

pytest-apibean is built for:

• Async-first APIs
• Token-based authentication
• Multi-service environments
• CI-friendly workflows
• Long-lived OSS projects

It avoids global state, supports namespace packaging, and integrates cleanly with modern tooling such as uv, httpx, and pytest >= 7.

Typical use cases

• Integration testing for apibean-rest
• Contract testing between services and apibean-datacore
• End-to-end testing via apibean-client
• Shared testing infrastructure across multiple apibean projects

Philosophy

pytest-apibean is intentionally minimal and opinionated:

• It provides infrastructure, not test logic
• It favors explicit configuration over magic
• It scales from a single service to a multi-repo ecosystem

If you are building APIs with apibean, pytest-apibean lets you test them consistently, correctly, and with confidence.