Code Metrics Q&A
summary
Uses Flog, SimpleCov, RuboCop and the runtime tool mutant plus Code Climate integration; Code Climate typically yields an easy ~4.0 score while SimpleCov and mutation metrics tell different stories. One module shows only ~20% mutation coverage because the domain resists mutation testing. A 2,000‑line proof‑of‑concept class is replaced by about fifty narrow classes and the refactor works and improves metrics. Failing builds on metric violations creates friction; trends and colors beat chasing absolute numbers.
AI-written summary · about
chapters (17)
- 0:00Introduction to Q&A
- 1:58Failed attempts at 100% coverage
- 3:45CodeClimate: Hot or Not
- 5:54Interpreting metrics correctly
- 12:57Focusing on metric trends
- 15:23Productivity vs code quality
- 18:16Lint metrics and automation
- 22:55Most important metrics
- 25:50Can metrics mislead developers
- 27:14Metrics for managers
- 29:44Mutant in CodeClimate
- 30:50Problems metrics solve
- 33:20Failing builds on metric violations
- 36:15When is Ruby ready
- 37:34Showing metrics to clients
- 38:17Bad metric trends and performance
- 39:25Final takeaways on metrics
related talks
53mUsing Metrics to Take a Hard look at your code
38mMutants to the rescue: How effective are your unit tests?
36mMutation Testing: Did my test break my code ? 🤔
27mRe-thinking Regression Testing
01h 00mCode Quality in Practice
24mGain more confidence in your code with mutation testing
40mEscaping the Tar Pit
47mMutation Testing: Did my test break my code ? 🤔