Test Junkie tutorials

Each tutorial teaches one Test Junkie concept with examples you can run, from core features to running in CI and a full Selenium WebDriver framework.

Concepts

Core features

The capabilities that set Test Junkie apart, each explained with working examples.

Intermediate10 min

Parallel execution

Thread limits for suites and tests, opting out with parallelized=False, and running parameter variants side by side.

Intermediate12 min

Multi-layer parametrization

Suite-level and test-level parameters= multiply into tracked variants: 2 environments × 3 browsers = 6 runs from one method.

Beginner8 min

Retrying flaky tests

retry_on and no_retry_on retry flaky infrastructure, never real bugs. With a real API example.

Beginner10 min

Metadata and targeting

Owner, feature, component, tags and priority, then run only the matching subset from Python or tj run.

Intermediate10 min

Shared setup and teardown

Write setup, checks and cleanup once in a Rules subclass and attach it to any suite with rules=.

Advanced15 min

Test results and listeners

Read every result after a run, or react as each test finishes with a Listener: Slack alerts, a results database, custom reports.

Pipelines

CI and the command line

Running Test Junkie where it matters: in your pipeline, with results you can act on.

Intermediate10 min

Running Test Junkie in CI

A committed tj.cfg, exit codes, CI-only retries, reports as artifacts and failures marked on the pull request.

Beginner6 min

Random order and --seed

Shuffle tests to find hidden dependencies between them, and repeat the exact order of a failing run.

Patterns

Recipes

Short, focused patterns for everyday problems.

Beginner6 min

Data-driven testing

Feed tests from a CSV, a JSON file or a database: parameters= takes a function.

Beginner5 min

Fast feedback loops

Run only the slice of a large suite you need right now, in parallel, and get an answer in seconds.

Series

Selenium WebDriver series

5 partsIntermediateAbout an hour

Build a small UI test framework from scratch, then test a real website with it. Each part adds one layer to the framework.

Start with part 1 → Read in order: each part builds on the last.
What you’ll buildpart
src/
├─ utils/Settings.py              P1
└─ pages/
   ├─ Browser.py                  P1
   ├─ UiObject.py                 P2
   ├─ BasePage.py                 P3
   ├─ HomePage.py                 P3
   ├─ DocumentationPage.py        P3
   └─ components/
      ├─ PageHeader.py            P3
      └─ PageFooter.py            P3
tests/
├─ suites/NavigationSuite.py      P4
├─ suites/DocumentationSuite.py   P4
└─ TestRules.py                   P5

  1. Part 1 · 10 min

    Managing browsers

    A Browser class that gives every test thread its own WebDriver, so tests can run in parallel.

  2. Part 2 · 10 min

    Wrapping UI elements

    A UiObject that waits, finds and acts, so tests never sleep.

  3. Part 3 · 15 min

    Page objects

    Pages and shared header/footer components, with locators that survive a redesign.

  4. Part 4 · 15 min

    Writing tests

    33 parameterized navigation checks from two methods, run in parallel with tj run.

  5. Part 5 · 10 min

    Catching JavaScript errors

    One rule that fails any test that causes a JavaScript error.