Topics covered in this episode:

• Why I Still Use Python Virtual Environments in Docker
• Python Developer Survey Results
• Anaconda Code add-in for Microsoft Excel
• Disabling Scheduled Dependency Updates
• Extras
• Joke

Watch on YouTube

About the show

Sponsored by us! Support our work through

• Our courses at Talk Python Training
• Hello, pytest! Course
• Patreon Supporters

Connect with the hosts

• Michael: @mkennedy@fosstodon.org
• Brian: @brianokken@fosstodon.org
• Show: @pythonbytes@fosstodon.org

Join us on YouTube at pythonbytes.fm/live to be part of the audience. Usually Monday at 10am PT. Older video versions available there too.

Finally, if you want an artisanal, hand-crafted digest of every week of the show notes in email form? Add your name and email to our friends of the show list, we'll never share it.

Michael #1: Why I Still Use Python Virtual Environments in Docker

• by Hynek Schlawack
• I was going to cover Production-ready Docker Containers with uv but decided to take this diversion instead.
• Spend a lot of time thinking about the secondary effects of what you do.
• venvs are well known and well documented. Let’s use them.

Brian #2: Python Developer Survey Results

• “… official Python Developers Survey, conducted as a collaborative effort between the Python Software Foundation and JetBrains.”
• Python w/ Rust rising, but still only 7%
• ““The drop in HTML/CSS/JS might show that data science is increasing its share of Python.” - Paul Everitt
• 37% contribute to open source. Awesome.
• Favorite Resources: Podcasts
  • Lots of familiar faces there. Awesome.
    • Perhaps I shouldn’t have decided to move “Python Test” back to Test & Code
• Usage
  • “Data analysis” down, but I think that’s because “data engineering” is added.
  • Data, Web dev, ML, devops, academic,
  • Testing is down 23%
• Python Versions
  • Still some 2 out there
  • Most folks on 3.10-3.12
• Install from: mostly python.org
• Frameworks
  • web: Flask, Django, Requests, FastAPI …
  • testing: pytest, unittest, mock, doctest, tox, hypothesis, nose (2% might be the Python 2 people)
• Data science
  • 77% use pandas, 72% NumPy
• OS: Windows still at 55%
• Packaging:
  • venv up to 55%
  • I imaging uv will be on the list next year
  • requirements.txt 63%, pyproject.toml 32%
• virtual env in containers? 47% say no

Michael #3: Anaconda Code add-in for Microsoft Excel

• Run their Python-powered projects in Excel locally with the Anaconda Code add-in
• Powered by PyScript, an Anaconda supported open source project that runs Python locally without install and setup
• Features
  • Cells Run Independently
  • Range to Multiple Types
  • init.py file is static and cannot be edited, with Anaconda Code, users have the ability to access and edit imports and definitions, allowing you to write top-level functions and classes and reuse them wherever you need.
  • A Customizable Environment

Brian #4: Disabling Scheduled Dependency Updates

• David Lord
• Interesting discussion of as they happen or batching of upsates to dependencies
• dependencies come in
  • requirements files
  • GH Actions in CI workflows
  • pre-commit hooks
• David was seeing 60 PRs per month when set up on monthly updates (3 ecosystems * 20 projects)
• new tool for updating GH actions: gha-update, allows for local updating of GH dependencies
• New process
  • Run pip-compile, gha-update, and pre-commit locally.
  • Update a project’s dependencies when actively working on the project, not just whenever a dependency updates.
  • Note that this works fine for dev dependencies, less so for security updates from run time dependencies. But for libraries, runtime dependencies are usually not pinned.

Extras

Brian:

• Test & Code coming back this week

Michael:

• Code in a Castle event
• Python Bytes badge spotting
• Guido’s post removed for moderation

Joke: C will watch in silence