In this week’s Python Report: Sly tricks for setting up Python on a machine with no network, snappier Python code with asyncand await, swifter Python programs with Zig, and slicker web apps with HTMX and Django.
Top picks for Python readers on InfoWorld
Air-gapped Python: Setting up Python without a net(work)
Flaky connection? No connection at all? Air gapped by design? You can still get Python, and Python packages, set up and running with a little juggling.
Get started with async in Python
Speaking of juggling… Do you want to interleave tasks in Python more efficiently without threads or multiple processes? Python’s async and await help you get more done in less time.
How to boost Python program performance with Zig
Old and venerable Python meets the hot new Zig, and both win. See how you can write fast, machine-native code with Zig, and wrap it in Python for ease and convenience.
Dynamic web apps with HTMX, Python, and Django
Use HTMX to give HTML the easy interactivity it always needed. Combine HTMX with Python and its super-powered Django framework, and you’ve got a web stack to beat them all.
More good reads and Python updates elsewhere
Proposed Linux kernel patches would allow access to libperf from Python
Imagine having access to a low-level Linux subsystem through a Python library—one supported natively by the Linux kernel. IBM’s proposal prepares to make that happen.
Fastplotlib: Interactive plotting in Python powered by the GPU
Create snappy, live-updating graphs and plots that can run in a variety of contexts (Jupyter notebooks, PyQt-powered windows, and more). The current release is considered a late alpha, but you’re encouraged to give it a whirl outside production.
A map of Python
A highly granular, interactive map of the package dependencies on PyPI, along with details about how to generate the same sort of interactive graph from a similarly sprawling data set. (Caution: Don’t try to plot everything!)
An oral history of Bank Python
How Python has been used, in proprietary forks, by various investment firms. Prepare to be stupefied at the flagrant abuse of everything from Python’s object system to proprietary, pre-Pandas tables.