Code Quality in Python - tools and reasons
summary
Claims that full correctness is undecidable (Rice theorem) but practical checks catch lots of problems. Lists concrete tools and tradeoffs: flake8 (combines pep8 pyflakes mccabe), pylint (many violations, needs config), vulture (dead‑code finder with false positives), isort/yapf/autopep8 for formatting, coverage+diff‑cover to measure lines touched, CI hooks (tox, Travis/Jenkins) that block PRs when coverage drops. Recommends gradual rule enabling and aiming for 90% coverage before risky refactors.
AI-written summary · about
chapters (14)
- 0:00Start
- 0:43Rice's theorem
- 1:23Examples
- 2:45What's this talk about?
- 3:43Checkers
- 6:52radon
- 10:01The lure of Pylama
- 11:59Vulture
- 13:34Extensions
- 15:32Formatters
- 16:00Autopep8
- 17:18Test coverage
- 20:27Automation
- 22:26Benefits?
related talks
27mStatic Code Analysis with Python
27mAn Overview of the Python Code Tool Landscape 2023
51mTaking Python beyond scripting
21mTeach Your Old Code New Tricks: Automating Code Quality in Large Projects
43mIdiomatic Python: Tools and Tips
41mEcosystem for Reliable Python Programming
29mSix QA Tools that will make your Python code better
22mPublishing well formed Python packages