Managing the code quality of your project. Leave the past behind: Focus on new code.
summary
If a linter on a 20k‑line FlaskBB clone spits ~1,800 Pylint messages and coverage is 44%, the recommended approach is to measure quality only on recent changes: run Pylint in CI against git diff, or use tools that filter to changed lines so new PRs show only relevant issues. The clean as you code method plus a GitHub Action plus Darker or a SonarCloud report reduces noise (the quicksort PR shows three fixes) and yields actionable new‑code coverage (about 75.9% here) rather than global metrics.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 0:36Agenda
- 1:35What is static analysis
- 2:17Why static analysis is useful
- 3:04Example project
- 3:32Code quality tool
- 5:00Traditional approach
- 7:00Limitations
- 10:12Pull requests
- 13:21Clean as you code methodology
- 14:05Recap
- 16:16Darker
- 20:57Forget the old code
- 22:45Summary
- 23:28Questions
related talks
21mTeach Your Old Code New Tricks: Automating Code Quality in Large Projects
44mClean as You Code Your Projects
30mAutomating Code Quality
40mCode Review Skills for Pythonistas
25mcode review is too late for code quality
18mBuilding a Culture of Code Quality
54mFixing legacy code, one pull request at a time
33mBuilding Code Quality