Millisecond Full Stack Acceptance Tests
This video is also available in the GOTO Play video app! Download it to enjoy offline access to our conference videos while on the move.
Are your full stack acceptance tests slow, non-deterministic and hard to maintain? You're not alone.
Imagine running hundreds of them in a few seconds, giving the same result every time. How do you think a feedback loop that fast would that affect your team's productivity?
In this talk you will see what this workflow looks like. You will learn about the underlying principles and techniques for millisecond full stack acceptance tests.
This is primarily a talk for programmers, as the solution to the problem requires refactoring of the system under test as well as tests. I will demonstrate that the basic assumption of the Test Pyramid model are wrong, and suggest a more useful taxonomy of tests, and how to partition testing efforts.
What will the audience learn from this talk?
- How to identify what makes tests slow
- Design patterns for decoupling application code and test code
- How to run the same acceptance tests against different layers of the app
Does it feature code examples and/or live coding?
Yes, about 10-20% of the presentation is code
Prerequisite attendee experience level:
Creator of Cucumber; Author of "The Cucumber Book, Behaviour-Driven Development for Developers and Testers"