Mutation Testing with PIT
summary
Explains mutation testing as introducing artificial bugs and checking whether tests kill them, using PIT to generate and report mutants automatically. Shows a failing demo (two mutants survive) then a fixed test that kills them. Runs PIT on a real project (5000 lines / 12 classes), wires it into GitHub Actions, publishes HTML reports to GitHub Pages, and completes full mutation coverage in about 2.5 minutes. Notes equivalent-mutant risk and PIT’s heuristics to avoid slow, useless mutations.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:40NASA Mars Orbiter disaster
- 3:23Code coverage limitations
- 5:07Mutation testing explained
- 6:30Mutants and operations
- 8:15PIT for Java
- 9:20Demo: Calculator example
- 12:29Scaling to real projects
- 13:55Demo: Hazelcast plugin
- 16:01Why adoption took decades
- 17:25Performance challenges and PIT
- 18:27Convincing teams
- 19:52Danger of coverage thresholds
- 20:35PIT in CI/CD
- 21:18Industry adoption
