Typing
summary
Conference session collects concrete work: Python 3.10 gains union-with-| and teams debate a postfix ? for Optional; a PyPI tool tricast validates JSON against TypedDicts today and proposes TypeForm to erase runtime casts; a self-type shorthand and simpler type-variable syntax reduce boilerplate; a CPython fork compiles “static python” modules, replaces ~20 Cython modules and recovers ~0.7% global CPU (1.7% from five hot modules); Pyre incremental check cuts full-checks from ~5 minutes to ~3.9 seconds; Typeshed tooling auto-generates baseline stubs and publishes per-package wheels; experimental tensor-shape types + type arithmetic catch dimension errors before runtime.
AI-written summary · about
chapters (9)
- 0:00Start
- 0:06Type Syntax Simplifications (Maggie Moss)
- 29:45Validating JSON with TypedDict, trycast, and TypeForm (David Foster)
- 54:20Type Variables for All (Pradeep Kumar Srinivasan)
- 1:25:16Static Python: Types in Bytecode Compilation & Runtime (Carl Meyer)
- 1:50:55Incremental Check in Pyre (Jia Chen)
- 2:17:01Scaling Typeshed to 1000 Packages (Jukka Lehtosalo)
- 2:43:06Catching Tensor Shape Errors Using the Type Checker (Pradeep Kumar Srinivasan, Matthew Rahtz)
- 3:11:10Type Arithmetic (Alfonso Castaño)
related talks
03h 04mTyping Summit - at PyCon US 2022
03h 38mPython Types for Fun and Profit
57mWhy You Should Care About Types
29mStatic Typing in Python
32mStatic Typing in Python
28mMind the gap! Why static typing requires more than just adding annotations
30mTyping at scale: statically type-checking a large codebase
30mClearer Code at Scale: Static Types at Zulip and Dropbox