Inside the Cheeseshop: How Python Packaging Works
summary
Traces concrete history and engineered fixes: why setup.py executes arbitrary code and blocks reliable dependency discovery, why wheels vs eggs matter and how wheel’s spec made binary installs practical, why pip replaced easy_install and why twine is required for secure uploads, how PyPI migrated to Warehouse to scale and add HTTPS, and how PEPs 517/518 introduce pyproject.toml to separate build systems and make reproducible builds possible.
AI-written summary · about
chapters (15)
- 0:00Intro
- 1:19Python Packaging In Just Five Easy Steps!
- 2:26Python Packaging So Easy A Caveman Could Do It
- 5:18Problem: How do I find Python code?
- 7:36Problem: Building takes too long
- 7:56Solution: Built Distributions
- 10:12Problem: How can I specify dependencies?
- 11:06Problem: How can we make installing easier?
- 12:13Problem: The name pyinstall is too long
- 12:47Problem: Application dependencies?
- 14:32Solution: The Wheel Distribution
- 18:58Solution: General care and maintenance
- 20:22Problem: Reproducible Environments
- 21:06Problem: setup.py executes arbitrary code
- 21:37Problem: The distutils/setuptools dance
related talks
24mInside the Cheeseshop:How Python Packaging Works
32mInside the Cheeseshop: How Python Packaging Works
02h 36mPython Packaging Demystified
30mThe Evolutionary Saga of Python Packaging
36mPython Packaging from Init to Deploy
28mModern Python packaging without pain
01h 21mSecure Python Packaging & Release Using Continuous Deployment
44mNobody Expects the Python Packaging Authority