Objectionable Content
summary
Clear claim: use the Python object model to build production systems (not ML), with concrete patterns and tradeoffs. Demos include a resistor-network builder that extracts arrays into numpy/pandas, a memory probe showing ~48 bytes per plain Python object, a slots example that can be counterintuitively larger, a dataclasses refactor, a range-lookup (getitem) flagger and a chainmap undo pattern. Covers tools: trace malloc, namedtuple, pathlib, enum, metaclass hooks and why classmethods are the right factories; the examples mostly work and expose real-world pitfalls.
AI-written summary · about
chapters (8)
- 0:00Python object model intro
- 1:04Python object model in production
- 20:20Tuples to named tuples
- 36:30Python as orchestration
- 51:39Rearranging complexity with protocols
- 1:08:09Object model hooks deep dive
- 1:24:39Custom object systems
- 1:26:22Summary
related talks
46mOn Being Pythonic
19mObjects and classes in Python (and JavaScript)
01h 07mWhat Does It All Really Mean
34mTeaching Object-Oriented Programming with Python
47mA Deep-Dive Into Python Classes
01h 18mPython 103: Mmmm... Understanding Python's Memory Model, Mutability, and Methods
01h 18mPython 103: Mmmm... Understanding Python's Memory Model, Mutability, and Methods
36mPractical Python Design Patterns