Behat for characterisation on tests

on YouTube ↗45m

" summary

Shows how Behat captures existing API and UI behavior so teams can refactor big, running systems with confidence — used to refactor 5+ year old apps by first recording current outputs, then changing code. Uses Behat + Gherkin at the HTTP layer for characterization tests, Mountebank for HTTP mocks to simulate downstream services, and Mink + Selenium (Chrome) for browser tests. Recommends snapshot testing (PHPUnit extension) for large outputs, one test per boundary, and keeping characterization tests short-lived.

AI-written summary · about

" chapters (22)

1. 0:00Intro
2. 0:53No tests
3. 1:28Without fear
4. 1:59Hello
5. 3:20Overview
6. 3:45What is legacy code
7. 6:19Unintended changes
8. 7:23Characterization tests
9. 13:24Do not change the card
10. 15:07Minimum number of tests
11. 16:09Dont overstate it
12. 17:46Identify pinch points
13. 18:52Behatgherkin
14. 19:51Installation
15. 20:36HTTP API
16. 24:34Different types of tests
17. 26:32HTTP endpoints
18. 27:35HTTP mocks
19. 29:45Browser automation
20. 31:57Custom steps
21. 38:29Snapshot testing
22. 43:40Recap

" related talks

• 45mBehat from 0 to Hero: a Practical GuideSamuele Lilli, Matteo Moretti · SymfonyDay 2017
• 27mAutomated Web Acceptance Testing with BehatShashikant Jagtap · SeleniumConf 2012 London
• 45mBehaviour Driven Development and Behat: Telling Stories Through CodeMatt Brunt · PHP UK Conference 2018
• 31mTesting API's written in PHP using the Behat feature testing toolTimo Bakx · Dutch PHP Conference 2024
• 34mTesting APIs written in PHP using the Behat feature testing toolTimo Bakx · phpday 2025
• 49mBending Behat's Benefits: A live coding adventureSteve Persch · MidCamp 2018
• 26mBehaviour Driven Development with LaravelMarcus Moore · Laracon AU 2019
• 34mWriting end to end tests with a smileEkaterina Budnikov, Kim Hogeling · GDG DevFest Hamburg 2017