Running (snippets of) Python in the browser
summary
Three approaches to run Python in the browser are compared and concrete claims are given: PyScript is a transpiler that generates JavaScript, installs via pip/conda, and trades some Python semantics for speed so it is faster than most other Python-in-the-browser solutions; JavaScript-based interpreters behave more like real Python but are about 100x slower; compiled interpreters (PyPy.js style) yield near-native speed but load a VM of about 13 megabytes. PyScript is used for Bokeh client callbacks, Flexx widgets, and a Clinical Graphics image-annotation app and those demos work.
AI-written summary · about
chapters (13)
- 0:00Running Python in browser
- 1:51Approaches to Python in browser
- 3:41JavaScript Python interpreters
- 4:46WebAssembly Python interpreters
- 6:57Frameworks and history
- 8:02Comparing Python-in-browser solutions
- 10:13Choosing the right approach
- 12:17PyScript as transpiler
- 14:48PyScript with Bokeh callbacks
- 16:40PyScript with Flex sync
- 18:37Medical image annotation with PyScript
- 20:02Summary and future outlook
- 20:50Q&A
related talks
39mPyScript - Python in the Browser
33mPython in the browser
32mPyScript - what you need to know
02h 55mWriting Serverless Python Web Apps with PyScript
45mThe marvelous world of Python in Browser: a nice surprise
01h 39mPyScript - Writing a Python application in the browser
30mNext Level Python Applications with PyScript
42mPyScript & Data Science- PyData stack on the Browser