test_n_ops.md

Testing and Operations

Testing

Unit tests

Testing a module/class/function in isolation (dependencies mocked). Tooling:

• pytest
• pytest_mock

Functional tests

Testing the whole microservice functionality and still mocking network interactions.

Tooling:

• Flask provides FlaskClient class that simulates HTTP interaction
• OR WebTest can be used instead

import json
from flask_basic import app as tested_app

class TestApp():
    def test_help(self):
    # creating a FlaskClient instance to interact with the app
    app = tested_app.test_client()
    # calling /api/ endpoint
    hello = app.get('/api')
    # asserting the body
    body = json.loads(str(hello.data, 'utf8'))
    assert body['Hello'] == 'World'

Integration tests

Testing microservices integration with each other and other network resources.

Load tests

Testing performance and maximum workload.

Tooling:

• boom (simple load test)
• molotov (load tests with scenarios)
• flask-profiler (collects metrics while load-testing)

End-to-end tests

Testing the whole system from the user perspective.

Tooling:

• Selenium (scripts for automating browser)

Additional testing tools

• pytest-cov (measuring Coverage)
• pytest-flake8 (style checks)
• tox (testing on multiple interpreter versions)

Documentation

Tooling:

• Sphinx (tool for generating documentation)
• Autodoc (extension for Sphinx that injects docstrings into docs)
• ReadTheDocs (free docs hosting for Sphinx docs; maybe hooked to CI)

Continuous Integration

CI tools:

• Travis CI
• GitLab CI
• GitHub Actions

Additional tools

Coveralls: hosts your coverage data to show a nice green button in GitHub.