Locution: Grading Your Selenium Tests
summary
A prototype service called locution attaches to a Selenium test framework via an event listener and a Java SDK, sends JSON about locators and actions to a web service, and returns a quantitative report that grades tests (sample runs yield a B- for one suite). Rules use static analysis: deductions for findElement calls, sleeps, assertions in page objects, oversized page objects; current prototype works but is synchronous and slow and needs async and broader SDKs.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 4:29Tests
- 4:50Bad Tests
- 7:41Good Tests
- 12:57Grading Your Tests
- 15:06Page Objects
- 15:47Bad Page Objects
- 18:16Good Page Objects
- 20:39Page Modules
- 23:48Locators
- 28:00Locator Grading Rubric
- 31:08Testing GPA
- 34:29Hypothesis
related talks
39mSelenium and the Four Rules of Simple Design
40mSelenium and the Four Rules of Simple Design
46mSelenium: State of the Union
38mSelenium tests, the Object Oriented Way
41mFunctional Tests (Selenium) in Practice
46mHow I learned to stop worrying and love record and playback
55mStability in Unstability: Self‑Healing Test Automation
37mSelenium In the Enterprise: What went right, what went wrong (so far)