Skip to main content

Enforce coverage and report on tests that honor constraints

Project description

PyPI version Python versions See Build Status on Travis CI See Build Status on AppVeyor

Report on tests that honor constraints, and guard against regressions


Intro

pytest-honors is a pytest plugin for automatically generating reports showing which of your project’s unit tests “honor” (or “prove” or “demonstrate” or “support”) various constraints that you care about. It can also be integrated with your build pipeline to make sure that you never accidentally remove the tests that are most important to you.

Below is a summary of features. More information is available at the pytest-honors documentation page.

Features

  • Lightning fast! Since it piggybacks onto pytest’s own test discovery and reporting system, it adds almost no measurable overhead to your testing process.
  • Generate nicely human-readable Markdown reports, suitable for handing to an auditor.
  • Automatically ensure that your test coverage only ever increases with time. Fixed things stay fixed!
  • Comes with ISO 27001 control definitions (but it’s super easy to add your own).

Requirements

  • pytest 3.5.0 or newer.

Installation

You can install “pytest-honors” via pip from PyPI:

$ pip install pytest-honors

Usage

You’ve written several thousand unit tests, but you don’t know which are actually important to you. Your team could laboriously struggle to keep docs up to date, but realistically that almost never works out as hoped. Even if your documentation is perfect, developers in the middle of a large refactoring project don’t want to flip between static text and their code.

pytest-honors wants to help you. For example, given this code:

import pytest
from pytest_honors.constraints import ConstraintsGroup

class MyControls(ConstraintsGroup):
    PasswordsMustBeGood = "We don't want bad passwords"
    EmailAddressesMustBeUnique = "No two users may have the same email"

@pytest.mark.honors(MyControls.PasswordsMustBeGood)
def test_password_strength():
    with pytest.raises(ValueError):
        check_password("12345")

@pytest.mark.honors(MyControls.EmailAddressesMustBeUnique)
def test_unique_email():
    add_account("User One", "spam@example.com")
    with pytest.raises(ValueError):
        add_account("User Two", "spam@example.com")

In the language of pytest-honors, we say that test_password_string “honors” the PasswordsMustBeGood constraint and that test_unique_email honors the EmailAddressesMustBeUnique constraint. This is valuable on its own as developers can tell at a glance that each test actually matters to the overall design of the system, and they’re not just there because a new boss wants everyone to reach 100% test coverage. By moving important documentation to a machine-readable format that lives next to the code, we can put that information to work to give you some very useful tools.

When run like pytest --honors-report-markdown report.md, we can get nice, human-readable documentation like:

# MyControls - An enumeration.

## EmailAddressesMustBeUnique: No two users may have the same email

Supporting evidence:

- Name: test_unique_email
  Explanation: "None"
  Path: tests/test_important_stuff.py::test_unique_email
  Result: passed

## PasswordsMustBeGood: We don't want bad passwords

Supporting evidence:

- Name: test_password_strength
  Explanation: "None"
  Path: tests/test_important_stuff.py::test_password_strength
  Result: passed

This shows us all controls that are honored by the tests that we ran. Want to show your auditor that you’re checking important controls in your code? Now you have evidence.

When run like pytest --honors-store-counts, pytest-honors saves the number of tests honoring each constraint. Later you can run pytest --honors-regression-fail, and if the coverage for any controls has decreased since the last test run, then pytest fails. Suppose an intern deletes the test_unique_email unit test. That results in the error:

ValueError: ['Constraint MyControls.EmailAddressesMustBeUnique honorers count dropped from 1 to 0']

You can integrate this in your CI pipeline and know that a rogue developer isn’t deleting the constraints you care about.

Contributing

Contributions are very welcome. Tests can be run with tox, please ensure the coverage at least stays the same before you submit a pull request.

Especially appreciated, and requiring the least amount of coding experience, would be other constraint definitions so that new users have a pleasant “batteries included” experience.

License

Distributed under the terms of the MIT license, “pytest-honors” is free and open source software.

Issues

If you encounter any problems, please file an issue along with a detailed description.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for pytest-honors, version 0.1.4
Filename, size File type Python version Upload date Hashes
Filename, size pytest_honors-0.1.4-py3-none-any.whl (10.5 kB) File type Wheel Python version py3 Upload date Hashes View hashes
Filename, size pytest-honors-0.1.4.tar.gz (12.5 kB) File type Source Python version None Upload date Hashes View hashes

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN SignalFx SignalFx Supporter DigiCert DigiCert EV certificate StatusPage StatusPage Status page