Skip to content

Module 04: Integration Testing

← Mocking and Test Doubles | Topic Home | Next → End-to-End Testing

Status Difficulty Time

Testing across component boundaries — TestContainers for real databases in tests, HTTP client testing, file system integration, in-memory vs. real dependencies.

Overview

This module is a stub. Content will be added in a future iteration.

Topics covered in this module:

  • What integration testing is and why it complements (not replaces) unit testing
  • Testing with real databases using TestContainers (Python and JavaScript) — spinning up PostgreSQL, MySQL, Redis in Docker for tests
  • pytest fixtures for database setup and teardown — transaction rollback patterns
  • Testing HTTP clients — using httpx.AsyncClient with a test app, supertest in Node.js
  • In-memory vs. real dependency trade-offs — when SQLite is fine, when you need the real database
  • File system testing — using tmp_path in pytest, os.tmpdir() in Node.js
  • Test data management — factories, fixtures, and seed data strategies
  • Testing Django and Flask applications — the test_client() pattern

Prerequisites

  • Module 03: Mocking and Test Doubles in this topic
  • Basic understanding of databases (SQL), HTTP (REST), or file I/O

Objectives

By the end of this module, you will be able to:

  1. Write integration tests that use a real PostgreSQL database in a Docker container
  2. Test a REST API endpoint with a real HTTP client against a running test application
  3. Apply pytest fixtures with appropriate scope for database setup/teardown
  4. Distinguish between scenarios where in-memory substitutes are appropriate vs. where real dependencies are required
  5. Manage test data effectively across a test suite using factories and seed data
  • [[qa-testing/modules/03_mocking-and-test-doubles]] — The contrast between mocking (unit) and real dependencies (integration) is the central distinction
  • [[qa-testing/modules/06_api-testing]] — API testing extends integration testing to HTTP interfaces specifically
  • [[django-fastapi-flask]] — Django's TestCase and FastAPI's TestClient are integration testing tools