Podcasts about talk python

If you're looking to leverage the insane power of modern GPUs for data science and ML, you might think you'll need to use some low-level programming language such as C++. But the folks over at NVIDIA have been hard at work building Python SDKs which provide nearly native level of performance when doing Pythonic GPU programming. Bryce Adelstein Lelbach is here to tell us about programming your GPU in pure Python. Episode sponsors Posit Agntcy Talk Python Courses Links from the show Bryce Adelstein Lelbach on Twitter: @blelbach Episode Deep Dive write up: talkpython.fm/blog NVIDIA CUDA Python API: github.com Numba (JIT Compiler for Python): numba.pydata.org Applied Data Science Podcast: adspthepodcast.com NVIDIA Accelerated Computing Hub: github.com NVIDIA CUDA Python Math API Documentation: docs.nvidia.com CUDA Cooperative Groups (CCCL): nvidia.github.io Numba CUDA User Guide: nvidia.github.io CUDA Python Core API: nvidia.github.io Numba (JIT Compiler for Python): numba.pydata.org NVIDIA's First Desktop AI PC ($3,000): arstechnica.com Google Colab: colab.research.google.com Compiler Explorer (“Godbolt”): godbolt.org CuPy: github.com RAPIDS User Guide: docs.rapids.ai Watch this episode on YouTube: youtube.com Episode #509 deep-dive: talkpython.fm/509 Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

Topics covered in this episode: platformdirs poethepoet - “Poe the Poet is a batteries included task runner that works well with poetry or with uv.” Python Pandas Ditches NumPy for Speedier PyArrow pointblank: Data validation made beautiful and powerful Extras Joke Watch on YouTube About the show Sponsored by us! Support our work through: Our courses at Talk Python Training The Complete pytest Course Patreon Supporters Connect with the hosts Michael: @mkennedy@fosstodon.org / @mkennedy.codes (bsky) Brian: @brianokken@fosstodon.org / @brianokken.bsky.social Show: @pythonbytes@fosstodon.org / @pythonbytes.fm (bsky) 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: platformdirs A small Python module for determining appropriate platform-specific dirs, e.g. a "user data dir". Why the community moved on from appdirs to platformdirs At AppDirs: Note: This project has been officially deprecated. You may want to check out pypi.org/project/platformdirs/ which is a more active fork of appdirs. Thanks to everyone who has used appdirs. Shout out to ActiveState for the time they gave their employees to work on this over the years. Better than AppDirs: Works today, works tomorrow – new Python releases sometimes change low-level APIs (win32com, pathlib, Apple sandbox rules). platformdirs tracks those changes so your code keeps running. First-class typing – no more types-appdirs stubs; editors autocomplete paths as Path objects. Richer directory set – if you need a user's Downloads folder or a per-session runtime dir, there's a helper for it. Cleaner internals – rewritten to use pathlib, caching, and extensive test coverage; all platforms are exercised in CI. Community stewardship – the project lives in the PyPA orbit and gets security/compatibility patches quickly. Brian #2: poethepoet - “Poe the Poet is a batteries included task runner that works well with poetry or with uv.” from Bob Belderbos Tasks are easy to define and are defined in pyproject.toml Michael #3: Python Pandas Ditches NumPy for Speedier PyArrow Pandas 3.0 will significantly boost performance by replacing NumPy with PyArrow as its default engine, enabling faster loading and reading of columnar data. Recently talked with Reuven Lerner about this on Talk Python too. In the next version, v3.0, PyArrow will be a required dependency, with pyarrow.string being the default type inferred for string data. PyArrow is 10 times faster. PyArrow offers columnar storage, which eliminates all that computational back and forth that comes with NumPy. PyArrow paves the way for running Pandas, by default, on Copy on Write mode, which improves memory and performance usage. Brian #4: pointblank: Data validation made beautiful and powerful “With its … chainable API, you can … validate your data against comprehensive quality checks …” Extras Brian: Ruff rules Ruff users, what rules are using and what are you ignoring? Python 3.14.0b2 - did we already cover this? Transferring your Mastodon account to another server, in case anyone was thinking about doing that I'm trying out Fathom Analytics for privacy friendly analytics Michael: Polars for Power Users: Transform Your Data Analysis Game Course Joke: Does your dog bite?

If you've heard the phrase "Automate the boring things" for Python, this episode starts with that idea and takes it to another level. We have Glyph back on the podcast to talk about "Programming YOUR computer with Python." We dive into a bunch of tools and frameworks and especially spend some time on integrating with existing platform APIs (e.g. macOS's BrowserKit and Window's COM APIs) to build desktop apps in Python that make you happier and more productive. Let's dive in! Episode sponsors Posit Agntcy Talk Python Courses Links from the show Glyph on Mastodon: @glyph@mastodon.social Glyph on GitHub: github.com/glyph Glyph's Conference Talk: LceLUPdIzRs: youtube.com Notify Py: ms7m.github.io Rumps: github.com QuickMacHotkey: pypi.org QuickMacApp: pypi.org LM Studio: lmstudio.ai Coolify: coolify.io PyWin32: pypi.org WinRT: pypi.org PyObjC: pypi.org PyObjC Documentation: pyobjc.readthedocs.io Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

If you want to leverage the power of LLMs in your Python apps, you would be wise to consider an agentic framework. Agentic empowers the LLMs to use tools and take further action based on what it has learned at that point. And frameworks provide all the necessary building blocks to weave these into your apps with features like long-term memory and durable resumability. I'm excited to have Sydney Runkle back on the podcast to dive into building Python apps with LangChain and LangGraph. Episode sponsors Posit Auth0 Talk Python Courses Links from the show Sydney Runkle: linkedin.com LangGraph: github.com LangChain: langchain.com LangGraph Studio: github.com LangGraph (Web): langchain.com LangGraph Tutorials Introduction: langchain-ai.github.io How to Think About Agent Frameworks: blog.langchain.dev Human in the Loop Concept: langchain-ai.github.io GPT-4 Prompting Guide: cookbook.openai.com Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

Topics covered in this episode: Making PyPI's test suite 81% faster People aren't talking enough about how most of OpenAI's tech stack runs on Python PyCon Talks on YouTube Optimizing Python Import Performance Extras Joke Watch on YouTube About the show Sponsored by Digital Ocean: pythonbytes.fm/digitalocean-gen-ai Use code DO4BYTES and get $200 in free credit Connect with the hosts Michael: @mkennedy@fosstodon.org / @mkennedy.codes (bsky) Brian: @brianokken@fosstodon.org / @brianokken.bsky.social Show: @pythonbytes@fosstodon.org / @pythonbytes.fm (bsky) 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. Brian #1: Making PyPI's test suite 81% faster Alexis Challande The PyPI backend is a project called Warehouse It's tested with pytest, and it's a large project, thousands of tests. Steps for speedup Parallelizing test execution with pytest-xdist 67% time reduction --numprocesses=auto allows for using all cores DB isolation - cool example of how to config postgress to give each test worker it's on db They used pytest-sugar to help with visualization, as xdist defaults to quite terse output Use Python 3.12's sys.monitoring to speed up coverage instrumentation 53% time reduction Nice example of using COVERAGE_CORE=sysmon Optimize test discovery Always use testpaths Sped up collection time. 66% reduction (collection was 10% of time) Not a huge savings, but it's 1 line of config Eliminate unnecessary imports Use python -X importtime Examine dependencies not used in testing. Their example: ddtrace A tool they use in production, but it also has a couple pytest plugins included Those plugins caused ddtrace to get imported Using -p:no ddtrace turns off the plugin bits Notes from Brian: I often get questions about if pytest is useful for large projects. Short answer: Yes! Longer answer: But you'll probably want to speed it up I need to extend this article with a general purpose “speeding up pytest” post or series. -p:no can also be used to turn off any plugin, even builtin ones. Examples include nice to have developer focused pytest plugins that may not be necessary in CI CI reporting plugins that aren't needed by devs running tests locally Michael #2: People aren't talking enough about how most of OpenAI's tech stack runs on Python Original article: Building, launching, and scaling ChatGPT Images Tech stack: The technology choices behind the product are surprisingly simple; dare I say, pragmatic! Python: most of the product's code is written in this language. FastAPI: the Python framework used for building APIs quickly, using standard Python type hints. As the name suggests, FastAPI's strength is that it takes less effort to create functional, production-ready APIs to be consumed by other services. C: for parts of the code that need to be highly optimized, the team uses the lower-level C programming language Temporal: used for asynchronous workflows and operations inside OpenAI. Temporal is a neat workflow solution that makes multi-step workflows reliable even when individual steps crash, without much effort by developers. It's particularly useful for longer-running workflows like image generation at scale Michael #3: PyCon Talks on YouTube Some talks that jumped out to me: Keynote by Cory Doctorow 503 days working full-time on FOSS: lessons learned Going From Notebooks to Scalable Systems And my Talk Python conversation around it. (edited episode pending) Unlearning SQL The Most Bizarre Software Bugs in History The PyArrow revolution in Pandas And my Talk Python episode about it. What they don't tell you about building a JIT compiler for CPython And my Talk Python conversation around it (edited episode pending) Design Pressure: The Invisible Hand That Shapes Your Code Marimo: A Notebook that "Compiles" Python for Reproducibility and Reusability And my Talk Python episode about it. GPU Programming in Pure Python And my Talk Python conversation around it (edited episode pending) Scaling the Mountain: A Framework for Tackling Large-Scale Tech Debt Brian #4: Optimizing Python Import Performance Mostly pay attention to #'s 1-3 This is related to speeding up a test suite, speeding up necessary imports. Finding what's slow Use python -X importtime

The folks over at Astral have made some big-time impacts in the Python space with uv and ruff. They are back with another amazing project named ty. You may have known it as Red-Knot. But it's coming up on release time for the first version and with the release it comes with a new official name: ty. We have Charlie Marsh and Carl Meyer on the show to tell us all about this new project. Episode sponsors Posit Auth0 Talk Python Courses Links from the show Talk Python's Rock Solid Python: Type Hints & Modern Tools (Pydantic, FastAPI, and More) Course: training.talkpython.fm Charlie Marsh on Twitter: @charliermarsh Charlie Marsh on Mastodon: @charliermarsh Carl Meyer: @carljm ty on Github: github.com/astral-sh/ty A Very Early Play with Astral's Red Knot Static Type Checker: app.daily.dev Will Red Knot be a drop-in replacement for mypy or pyright?: github.com Hacker News Announcement: news.ycombinator.com Early Explorations of Astral's Red Knot Type Checker: pydevtools.com Astral's Blog: astral.sh Rust Analyzer Salsa Docs: docs.rs Ruff Open Issues (label: red-knot): github.com Ruff Types: types.ruff.rs Ruff Docs (Astral): docs.astral.sh uv Repository: github.com Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

Python has many string formatting styles which have been added to the language over the years. Early Python used the % operator to injected formatted values into strings. And we have string.format() which offers several powerful styles. Both were verbose and indirect, so f-strings were added in Python 3.6. But these f-strings lacked security features (think little bobby tables) and they manifested as fully-formed strings to runtime code. Today we talk about the next evolution of Python string formatting for advanced use-cases (SQL, HTML, DSLs, etc): t-strings. We have Paul Everitt, David Peck, and Jim Baker on the show to introduce this upcoming new language feature. Episode sponsors Posit Auth0 Talk Python Courses Links from the show Guests: Paul on X: @paulweveritt Paul on Mastodon: @pauleveritt@fosstodon.org Dave Peck on Github: github.com Jim Baker: github.com PEP 750 – Template Strings: peps.python.org tdom - Placeholder for future library on PyPI using PEP 750 t-strings: github.com PEP 750: Tag Strings For Writing Domain-Specific Languages: discuss.python.org How To Teach This: peps.python.org PEP 501 – General purpose template literal strings: peps.python.org Python's new t-strings: davepeck.org PyFormat: Using % and .format() for great good!: pyformat.info flynt: A tool to automatically convert old string literal formatting to f-strings: github.com Examples of using t-strings as defined in PEP 750: github.com htm.py issue: github.com Exploits of a Mom: xkcd.com pyparsing: github.com Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

What trends and technologies should you be paying attention to today? Are there hot new database servers you should check out? Or will that just be a flash in the pan? I love these forward looking episodes and this one is super fun. I've put together an amazing panel: Gina Häußge, Ines Montani, Richard Campbell, and Calvin Hendryx-Parker. We dive into the recent Stack Overflow Developer survey results as a sounding board for our thoughts on rising and falling trends in the Python and broader developer space. Episode sponsors NordLayer Auth0 Talk Python Courses Links from the show The Stack Overflow Survey Results: survey.stackoverflow.co/2024 Panelists Gina Häußge: chaos.social/@foosel Ines Montani: ines.io Richard Campbell: about.me/richard.campbell Calvin Hendryx-Parker: github.com/calvinhp Explosion: explosion.ai spaCy: spacy.io OctoPrint: octoprint.org .NET Rocks: dotnetrocks.com Six Feet Up: sixfeetup.com Stack Overflow: stackoverflow.com Python.org: python.org GitHub Copilot: github.com OpenAI ChatGPT: chat.openai.com Claude: anthropic.com LM Studio: lmstudio.ai Hetzner: hetzner.com Docker: docker.com Aider Chat: github.com Goose AI: goose.ai IndyPy: indypy.org OctoPrint Community Forum: community.octoprint.org spaCy GitHub: github.com Hugging Face: huggingface.co Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

Pandas is at a the core of virtually all data science done in Python, that is virtually all data science. Since it's beginning, Pandas has been based upon numpy. But changes are afoot to update those internals and you can now optionally use PyArrow. PyArrow comes with a ton of benefits including it's columnar format which makes answering analytical questions faster, support for a range of high performance file formats, inter-machine data streaming, faster file IO and more. Reuven Lerner is here to give us the low-down on the PyArrow revolution. Episode sponsors NordLayer Auth0 Talk Python Courses Links from the show Reuven: github.com/reuven Apache Arrow: github.com Parquet: parquet.apache.org Feather format: arrow.apache.org Python Workout Book: manning.com Pandas Workout Book: manning.com Pandas: pandas.pydata.org PyArrow CSV docs: arrow.apache.org Future string inference in Pandas: pandas.pydata.org Pandas NA/nullable dtypes: pandas.pydata.org Pandas `.iloc` indexing: pandas.pydata.org DuckDB: duckdb.org Pandas user guide: pandas.pydata.org Pandas GitHub issues: github.com Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

Do you or your company need accounting software? Well, there are plenty of SaaS products out there that you can give your data to. but maybe you also really like Django and would rather have a foundation to build your own accounting system exactly as you need for your company or your product. On this episode, we're diving into Django Ledger, created by Miguel Sanda, which can do just that. Episode sponsors Auth0 Talk Python Courses Links from the show Miguel Sanda on Twitter: @elarroba Miguel on Mastodon: @elarroba@fosstodon.org Miguel on GitHub: github.com Django Ledger on Github: github.com Django Ledger Discord: discord.gg Get Started with Django MongoDB Backend: mongodb.com Wagtail CMS: wagtail.org Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

Have you ever spent an afternoon wrestling with a Jupyter notebook, hoping that you ran the cells in just the right order, only to realize your outputs were completely out of sync? Today's guest has a fresh take on solving that exact problem. Akshay Agrawal is here to introduce Marimo, a reactive Python notebook that ensures your code and outputs always stay in lockstep. And that's just the start! We'll also dig into Akshay's background at Google Brain and Stanford, what it's like to work on the cutting edge of AI, and how Marimo is uniting the best of data science exploration and real software engineering. Episode sponsors Worth Search Talk Python Courses Links from the show Akshay Agrawal: akshayagrawal.com YouTube: youtube.com Source: github.com Docs: marimo.io Marimo: marimo.io Discord: marimo.io WASM playground: marimo.new Experimental generate notebooks with AI: marimo.app Pluto.jl: plutojl.org Observable JS: observablehq.com Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

We're sitting down with Eric Matthes, the educator, author, and developer behind Django Simple Deploy. If you've ever struggled with taking that final step of getting your Django app onto a live server (without spending days wrestling with DevOps complexities), then give Django Simple Deploy a look. Eric shares how Django Simple Deploy automates away the boilerplate parts of deployment, so you can focus on building features instead of deciphering endless configs. We'll talk about this new project's journey to 1.0, the range of hosting platforms it supports, and why it's not just for beginners. Episode sponsors Worth Search Talk Python Courses Links from the show django-simple-deploy documentation: readthedocs.io django-simple-deploy repository: github.com Python Crash Course book: ehmatthes.github.io Code Red: codered.cloud Docker: docker.com Caddy: caddyserver.com Bunny.net CDN: bunny.net Platform.sh: platform.sh fly.io: fly.io Heroku: heroku.com Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

Topics covered in this episode: Git Town solves the problem that using the Git CLI correctly PEP 751 – A file format to record Python dependencies for installation reproducibility git-who and watchgha Share Python Scripts Like a Pro: uv and PEP 723 for Easy Deployment Extras Joke Watch on YouTube About the show Sponsored by Posit Package Manager: pythonbytes.fm/ppm Connect with the hosts Michael: @mkennedy@fosstodon.org / @mkennedy.codes (bsky) Brian: @brianokken@fosstodon.org / @brianokken.bsky.social Show: @pythonbytes@fosstodon.org / @pythonbytes.fm (bsky) 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: Git Town solves the problem that using the Git CLI correctly Git Town is a reusable implementation of Git workflows for common usage scenarios like contributing to a centralized code repository on platforms like GitHub, GitLab, or Gitea. Think of Git Town as your Bash scripts for Git, but fully engineered with rock-solid support for many use cases, edge cases, and error conditions. Keep using Git the way you do now, but with extra commands to create various branch types, keep them in sync, compress, review, and ship them efficiently. Basic workflow Commands to create, work on, and ship features. git town hack - create a new feature branch git town sync - update the current branch with all ongoing changes git town switch - switch between branches visually git town propose - propose to ship a branch git town ship - deliver a completed feature branch Additional workflow commands Commands to deal with edge cases. git town delete - delete a feature branch git town rename - rename a branch git town repo - view the Git repository in the browser Brian #2: PEP 751 – A file format to record Python dependencies for installation reproducibility Accepted From Brett Cannon “PEP 751 has been accepted! This means Python now has a lock file standard that can act as an export target for tools that can create some sort of lock file. And for some tools the format can act as their primary lock file format as well instead of some proprietary format.” File name: pylock.toml or at least something that starts with pylock and ends with .toml It's exciting to see the start of a standardized lock file Michael #3: git-who and watchgha git-who is a command-line tool for answering that eternal question: Who wrote this code?! Unlike git blame, which can tell you who wrote a line of code, git-who tells you the people responsible for entire components or subsystems in a codebase. You can think of git-who sort of like git blame but for file trees rather than individual files. And watchgha - Live display of current GitHub action runs by Ned Batchelder Brian #4: Share Python Scripts Like a Pro: uv and PEP 723 for Easy Deployment Dave Johnson Nice full tutorial discussing single file Python scripts using uv with external dependencies Starting with a script with dependencies. Using uv add --script [HTML_REMOVED] [HTML_REMOVED] to add a /// script block to the top Using uv run Adding #!/usr/bin/env -S uv run --script shebang Even some Windows advice Extras Brian: April 1 pranks done well BREAKING: Guido van Rossum Returns as Python's BDFL including Brett Cannon noted as “Famous Python Quotationist” Guido taking credit for “I came for the language but I stayed for the community” which was from Brett then Brett's title of “Famous Python Quotationist” is crossed out. Barry Warsaw asking Guido about releasing Python 2.8 Barry is the FLUFL, “Friendly Language Uncle For Life “ Mariatta can't get Guido to respond in chat until she addresses him as “my lord”. “… becoming one with whitespace.” “Indentation is Enlightenment” Upcoming new keyword: maybe Like “if” but more Pythonic as in Maybe: print("Python The Documentary - Coming This Summer!") I'm really hoping there is a documentary April 1 pranks done poorly Note: pytest-repeat works fine with Python 3.14, and never had any problems If you have to explain the joke, maybe it's not funny. The explanation pi, an irrational number, as in it cannot be expressed by a ratio of two integers, starts with 3.14159 and then keeps going, and never repeats. Python 3.14 is in alpha and people could be testing with it for packages Test & Code is doing a series on pytest plugins pytest-repeat is a pytest plugin, and it happened to not have any tests for 3.14 yet. Now the “joke”. I pretended that I had tried pytest-repeat with Python 3.14 and it didn't work. Test & Code: Python 3.14 won't repeat with pytest-repeat Thus, Python 3.14 won't repeat. Also I mentioned that there was no “rational” explanation. And pi is an irrational number. Michael: pysqlscribe v0.5.0 has the “parse create scripts” feature I suggested! Markdown follow up Prettier to format Markdown via Hugo Been using mdformat on some upcoming projects including the almost done Talk Python in Production book. Command I like is mdformat --number --wrap no ./ uv tool install --with is indeed the pipx inject equivalent, but requires multiple --with's: pipx inject mdformat mdformat-gfm mdformat-frontmatter mdformat-footnote mdformat-gfm-alerts uv tool install mdformat --with mdformat-gfm --with mdformat-frontmatter --with mdformat-footnote --with mdformat-gfm-alerts uv follow up From James Falcon As a fellow uv enthusiast, I was still holding out for a use case that uv hasn't solved. However, after last week's episode, you guys finally convinced me to switch over fully, so I figured I'd explain the use case and how I'm working around uv's limitations. I maintain a python library supported across multiple python versions and occasionally need to deal with bugs specific to a python version. Because of that, I have multiple virtualenvs for one project. E.g., mylib38 (for python 3.8), mylib313 (for python 3.13), etc. I don't want a bunch of .venv directories littering my project dir. For this, pyenv was fantastic. You could create the venv with pyenv virtualenv 3.13.2 mylib313, then either activate the venv with pyenv activate mylib313 and create a .python-version file containing mylib313 so I never had to manually activate the env I want to use by default on that project. uv doesn't have a great solution for this use case, but I switched to a workflow that works well enough for me: Define my own central location for venvs. For me that's ~/v Create venvs with something like uv venv --python 3.13 ~/v/mylib313 Add a simple function to my bashrc: `workon() { source ~/v/$1/bin/activate } so now I can run workon mylib313orworkon mylib38when I need to work in a specific environment. uv's.python-version` support works much differently than pyenv's, and that lack of support is my biggest frustration with this approach, but I am willing to live without it. Do you Firefox but not Zen? You can now make pure Firefox more like Zen's / Arc's layout. Joke: So here it will stay See the follow up thread too! Also: Guido as Lord Python via Nick Muoh

This episode is all about Beeware, the project that working towards true native apps built on Python, especially for iOS and Android. Russell's been at this for more than a decade, and the progress is now hitting critical mass. We'll talk about the Toga GUI toolkit, building and shipping your apps with Briefcase, the newly official support for iOS and Android in CPython, and so much more. I can't wait to explore how BeeWare opens up the entire mobile ecosystem for Python developers, let's jump right in. Episode sponsors Posit Python in Production Talk Python Courses Links from the show Anaconda open source team: anaconda.com PEP 730 – Adding iOS: peps.python.org PEP 738 – Adding Android: peps.python.org Toga: beeware.org Briefcase: beeware.org emscripten: emscripten.org Russell Keith-Magee - Keynote - PyCon 2019: youtube.com Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

In this episode, we welcome back Will McGugan, the creator of the wildly popular Rich library and founder of Textualize. We'll dive into Will's latest article on "Algorithms for High Performance Terminal Apps" and explore how he's quietly revolutionizing what's possible in the terminal, from smooth animations and dynamic widgets to full-on TUI (or should we say GUI?) frameworks. Whether you're looking to supercharge your command-line tools or just curious how Python can push the limits of text-based UIs, you'll love hearing how Will's taking a modern, web-inspired approach to old-school terminals. Episode sponsors Posit Python in Production Talk Python Courses Links from the show Algorithms for high performance terminal apps post: textual.textualize.io Textual Demo: github.com Textual: textualize.io Zero ver: 0ver.org memray: github.com Posting app: posting.sh Bulma CSS framewokr: bulma.io JP Term: davidbrochart.github.io Rich: github.com btop: github.com starship: starship.rs Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

Have you ever wondered why certain data points stand out so dramatically? They might hold the key to everything from fraud detection to groundbreaking discoveries. This week on Talk Python to Me, we dive into the world of outlier detection with Python with Brett Kennedy. You'll learn how outliers can signal errors, highlight novel insights, or even reveal hidden patterns lurking in the data you thought you understood. We'll explore fresh research developments, practical use cases, and how outlier detection compares to other core data science tasks like prediction and clustering. If you're ready to spot those game-changing anomalies in your own projects, stay tuned. Episode sponsors Posit Python in Production Talk Python Courses Links from the show Data-morph: github.com PyOD: github.com Prophet: github.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

Today we explore the wild world of Python deployment with my friend, Calvin Hendricks-Parker from Six Feet Up. We'll tackle some of the biggest challenges in taking a Python app from “it works on my machine” to production, covering inconsistent environments, conflicting dependencies, and sneaky security pitfalls. Along the way, Calvin shares how containerization with Docker and Kubernetes can both simplify and complicate deployments, especially for smaller teams. Finally, we'll introduce Scaf, a powerful project blueprint designed to give developers a rock-solid start on Python web projects of all sizes. Get notified when the Talk Python in Production book goes live and read the first third online right now. Episode sponsors Posit Python in Production Talk Python Courses Links from the show Calvin Hendryx-Parker: github.com Scaf on GitHub: github.com Scaf on GitHub (duplicate): github.com "Deploy the Dream" song: deploy-the-dream-talk-python.mp3 CloudDevEngineering YouTube Channel: youtube.com TechWorld with Nana YouTube Channel: youtube.com Tilt (Kubernetes Dev Tool): tilt.dev Talos (Minimal OS for Kubernetes): talos.dev Traefik Reverse Proxy: traefik.io Sealed Secrets on GitHub: github.com Argo CD Documentation: readthedocs.io MailHog on GitHub: github.com Next.js: nextjs.org Cloud Custodian: cloudcustodian.io Valky (Redis Replacement): valkey.io “The ‘Works on My Machine' Certification Program” (Coding Horror): blog.codinghorror.com NVIDIA's First Desktop AI PC (Ars Technica): arstechnica.com Kind (Kubernetes in Docker): kind.sigs.k8s.io Updated Effective PyCharm Course: training.talkpython.fm Talk Python in Production book: talkpython.fm/books/python-in-production Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

On this episode, I'm joined by Dr. Jeff Boeing, an assistant professor at the University of Southern California whose research spans urban planning, spatial analysis, and data science. We explore why OpenStreetMap is such a powerful source of global map data—and how Jeff's Python library, OSMnx, makes that data easier to download, model, and visualize. Along the way, we talk about what shapes city streets around the world, how urban design influences everything from daily commutes to disaster resilience, and why turning open data into accessible tools can open up completely new ways of understanding our cities. If you've ever wondered how to build or analyze your own digital maps in Python, or what it takes to manage a project that transforms raw geographic data into meaningful research, you won't want to miss this conversation. Episode sponsors Posit Podcast Later Talk Python Courses Links from the show City Street Orientations World: geoffboeing.com OSMnx Documentation: readthedocs.io OSMnx GitHub: github.com OpenStreetMap: openstreetmap.org Open Database License: opendatacommons.org ID Editor (Web Editor): wiki.openstreetmap.org Planet OSM: planet.openstreetmap.org Overpass API: wiki.openstreetmap.org GeoPandas: geopandas.org NetworkX: networkx.org Shapely: shapely.readthedocs.io Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

As Python developers, we're incredibly lucky to have over half a million packages that we can use to build our applications with over at PyPI. However, when it comes to choosing a UI framework, the options get narrowed down very quickly. Intersect those choices with the ones that work on mobile, and you have a very short list. Flutter is a UI framework for building desktop and mobile applications, and is in fact the one that we used to build the Talk Python courses app, you'd find at talkpython.fm/apps. That's why I'm so excited about Flet. Flet is a Python UI framework that is distributed and executed on the Flutter framework, making it possible to build mobile apps and desktop apps with Python. We have Feodor Fitsner back on the show after he launched his project a couple years ago to give us an update on how close they are to a full featured mobile app framework in Python. Episode sponsors Posit Podcast Later Talk Python Courses Links from the show Flet: flet.dev Flet on Github: github.com Packaging apps with Flet: flet.dev/docs/publish Flutter: flutter.dev React vs. Flutter: trends.stackoverflow.co Kivy: kivy.org Beeware: beeware.org Mobile forge from Beeware: github.com The list of built-in binary wheels: flet.dev/docs/publish/android#binary-python-packages Difference between dynamic and static Flet web apps: flet.dev/docs/publish/web Integrating Flutter packages: flet.dev/docs/extend/integrating-existing-flutter-packages serious_python: pub.dev/packages/serious_python Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

In this episode, I'm joined by JJ Allaire, founder and executive chairman at Posit, and Carlos Scheidegger, a software engineer at Posit, to explore Quarto, an open-source tool revolutionizing technical publishing. We discuss how Quarto empowers users to seamlessly transform Jupyter notebooks into polished reports, dashboards, e-books, websites, and more. JJ shares his journey from creating RStudio to developing Quarto as a versatile, multi-language tool, while Carlos delves into its roots in reproducibility and the challenges of academic publishing. Don't miss this deep dive into a tool that's shaping the future of data-driven storytelling! Episode sponsors Talk Python Courses DigitalOcean Links from the show JJ Allaire JJ on LinkedIn: linkedin.com JJ on GitHub: github.com Carlos Scheidegger Personal site: cscheid.net Mastodon: @scheidegger Fast AI: fast.ai nbdev: nbdev.fast.ai nbsanity - Share Notebooks as Polished Web Pages in Seconds: answer.ai Pandoc: pandoc.org Observable: github.com Quarto Pub: quartopub.com Deno: deno.com Real World Data Science site: realworlddatascience.net Typst: typst.app Github Actions for Quarto: github.com Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

Join me as I chat with Rich Iannone and Michael Chow from Posit where we explore the transformative power of data tables with the Great Tables library. We'll cover practical applications of Great Tables, showcasing how thoughtful design and advanced formatting can elevate your data presentations. And you'll learn about innovative features like nano plots and interactive elements and the importance of structure, format, and style in crafting tables that both inform and inspire. Whether you're a seasoned data scientist or just starting out, this episode is packed with valuable tips and inspiring examples to enhance your data storytelling. Episode sponsors Talk Python Courses DigitalOcean Links from the show Michael Chow: github.com/machow Richard Iannone: github.com/rich-iannone Episode Deep Dives Writeup: talkpython.fm/blog Great Tables: github.com Making Beautiful, Publication Quality Tables PyCon talk: youtube.com Andrew Weatherman's Visualization Gallery: aweatherman.com Bureau of the Census Manual of Tabular Presentation: census.gov Table Contest: posit.co Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

Join me for an insightful conversation with Alex Monahan, who works on documentation, tutorials, and training at DuckDB Labs. We explore why DuckDB is gaining momentum among Python and data enthusiasts, from its in-process database design to its blazingly fast, columnar architecture. We also dive into indexing strategies, concurrency considerations, and the fascinating way MotherDuck (the cloud companion to DuckDB) handles large-scale data seamlessly. Don't miss this chance to learn how a single pip install could totally transform your Python data workflow! Episode sponsors Sentry Error Monitoring, Code TALKPYTHON Data Citizens Podcast Talk Python Courses Links from the show Alex on Mastodon: @__Alex__ DuckDB: duckdb.org MotherDuck: motherduck.com SQLite: sqlite.org Moka-Py: github.com PostgreSQL: www.postgresql.org MySQL: www.mysql.com Redis: redis.io Apache Parquet: parquet.apache.org Apache Arrow: arrow.apache.org Pandas: pandas.pydata.org Polars: pola.rs Pyodide: pyodide.org DB-API (PEP 249): peps.python.org/pep-0249 Flask: flask.palletsprojects.com Gunicorn: gunicorn.org MinIO: min.io Amazon S3: aws.amazon.com/s3 Azure Blob Storage: azure.microsoft.com/products/storage Google Cloud Storage: cloud.google.com/storage DigitalOcean: www.digitalocean.com Linode: www.linode.com Hetzner: www.hetzner.com BigQuery: cloud.google.com/bigquery DBT (Data Build Tool): docs.getdbt.com Mode: mode.com Hex: hex.tech Python: www.python.org Node.js: nodejs.org Rust: www.rust-lang.org Go: go.dev .NET: dotnet.microsoft.com Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

If you're a Django developer, I'm sure you've heard so many people raving about FastAPI and Pydantic. But you really love Django and don't want to switch. Then you might want to give Django Ninja a serious look. Django Ninja is highly inspired by FastAPI, but is also deeply integrated into Django itself. We have Vitaliy Kucheryaviy the creator of Django Ninja on this show to tell us all about it. Episode sponsors Sentry Error Monitoring, Code TALKPYTHON Bluehost Talk Python Courses Links from the show Vitaly: github.com/vitalik Vitaly on X: @vital1k Top 5 Episodes of 2024: talkpython.fm/blog/posts/top-talk-python-podcast-episodes-of-2024 Django Ninja: django-ninja.dev Motivation section we talked through: django-ninja.dev/motivation LLM for Django Ninja: llm.django-ninja.dev Nano Django: github.com/vitalik/nano-django Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to Talk Python on YouTube: youtube.com Talk Python on Bluesky: @talkpython.fm at bsky.app Talk Python on Mastodon: talkpython Michael on Bluesky: @mkennedy.codes at bsky.app Michael on Mastodon: mkennedy

On today's episode, we're going to talk Python. Host Eric Chou is joined by Michael Kennedy,  the founder of Talk Python Training and host of the Talk Python To Me podcast. The discussion covers the evolution of Michael’s podcast and training business, the impact of platforms like TikTok on learning, and the importance of community... Read more »

On today's episode, we're going to talk Python. Host Eric Chou is joined by Michael Kennedy,  the founder of Talk Python Training and host of the Talk Python To Me podcast. The discussion covers the evolution of Michael’s podcast and training business, the impact of platforms like TikTok on learning, and the importance of community... Read more »

Topics covered in this episode: Talk Python rewritten in Quart PyPI now supports digital attestations Django Rusty Templates PEP 639 is now supported by PYPI Extras Joke Watch on YouTube About the show Sponsored by us! Support our work through: Our courses at Talk Python Training The Complete pytest Course Patreon Supporters Connect with the hosts Michael: @mkennedy@fosstodon.org / @mkennedy.codes (bsky) Brian: @brianokken@fosstodon.org / @brianokken.bsky.social Show: @pythonbytes@fosstodon.org / @pythonbytes.fm (bsky) 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: Talk Python rewritten in Quart Rewrote all of talkpython.fm in Quart (10k lines of code total, 4k changed) Considered FastAPI Litestar Django Hugo Static Site + Python Flask Discussed the multistage upgrade / conversion process Automating tests for all 1,000 pages Brian #2: PyPI now supports digital attestations Dustin Ingram “Attestations provide a verifiable link to an upstream source repository: By signing with the identity of the upstream source repository, such as in the case of an upload of a project built with GitHub Actions, PyPI's support for digital attestations defines a strong and verifiable association between a file on PyPI and the source repository, workflow, and even the commit hash that produced and uploaded the file. Additionally, publishing attestations to a transparency log helps mitigate against both compromise of PyPI and compromise of the projects themselves.” For maintainers If using GH Actions and Trusted Publishing make sure you use pypa/gh-action-pypi-publish, version v1.11.0 or newer that's it If not “Support for automatic attestation generation and publication from other Trusted Publisher environments is planned.” “While not recommended, maintainers can also manually generate and publish attestations.” See also PyPI Introduces Digital Attestations to Strengthen Python Package Security by Sarah Gooding Are we PEP 740 yet? Michael #3: Django Rusty Templates by Lily Foote An experimental reimplementation of Django's templating language in Rust. Goals 100% compatibility of rendered output. Error reporting that is at least as useful as Django's errors. Improved performance over Django's pure Python implementation. Brian #4: PEP 639 is now supported by PYPI from Brett Cannon PEP 639 – Improving License Clarity with Better Package Metadata For project metadata, use these fields: license and license-files: Examples license field [project] license = "MIT" [project] license = "MIT AND (Apache-2.0 OR BSD-2-clause)" [project] license = "MIT OR GPL-2.0-or-later OR (FSFUL AND BSD-2-Clause)" [project] license = "LicenseRef-Proprietary" Examples of license-files: [project] license-files = ["LICEN[CS]E*", "AUTHORS*"] [project] license-files = ["licenses/LICENSE.MIT", "licenses/LICENSE.CC0"] [project] license-files = ["LICENSE.txt", "licenses/*"] [project] license-files = [] Extras Brian: Playground Wisdom: Threads Beat Async/Await - interesting read from Armin Ronacher about different language abstractions around concurrency. PythonTest.com Discord community is now live Launched last week, as of this morning we've got 89 members Anyone already a pythontest community member has received an invite Anyone can join through courses.pythontest.com Everything at pythontest.com is 20% off through Dec 2 with code turkeysale2024 “Python Testing with pytest” eBook 40% off through Dec 2, use code turkeysale2024 Michael: Python 3.14.0a2 released Starter packs: Michael's Python people: https://bsky.app/starter-pack/mkennedy.codes/3lbdnupl26e2x Directory: https://blueskydirectory.com/starter-packs/all Joke: curl - heavy metal style!

Hynek has been writing and speaking on some of the most significant topics in the Python space and I've enjoyed his takes. So I invited him on the show to share them with all of us. This episode really epitomizes one of the reasons I launched Talk Python 9 years ago. It's as if we run into each other at a bar during a conference and I ask Hynek, "So what are your thoughts on ..." and we dive down the rabbit hole for an hour. I hope you enjoy it. Episode sponsors WorkOS Bluehost Talk Python Courses Links from the show Hynek Schlawack on Mastodon: @hynek Why I Still Use Python Virtual Environments in Docker: hynek.me Production-ready Python Docker Containers with uv: hynek.me Attrs: github.com uv: astral.sh What's New In Python 4: python.org BusyBox: busybox.net Hynek's YouTube Channel: youtube.com MOPUp for macOS: github.com Homebrew Python Is Not For You: justinmayer.com argon2-cffi: Argon2 for Python: github.com pytest-freethreaded: github.com LM Studio: lmstudio.ai StackOverflow Trends Graph: trends.stackoverflow.co Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to us on YouTube: youtube.com Follow Talk Python on Mastodon: talkpython Follow Michael on Mastodon: mkennedy

Topics covered in this episode: Python 3.13.0 released Oct 7 PEP 759 – External Wheel Hosting pytest-freethreaded pytest-edit Extras Joke Watch on YouTube About the show Sponsored by ScoutAPM: pythonbytes.fm/scout 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. Brian #1: Python 3.13.0 released Oct 7 That's today! What's New In Python 3.13 Interpreter (REPL) improvements exit works (really, this is worth the release right here) Multiline editing with history preservation. history sticks around between sessions Direct support for REPL-specific commands like help, exit, and quit, without the need to call them as functions. Prompts and tracebacks with color enabled by default. Interactive help browsing using F1 with a separate command history. History browsing using F2 that skips output as well as the >>> and … prompts. “Paste mode” with F3 that makes pasting larger blocks of code easier (press F3 again to return to the regular prompt). exit now works without parens Improved error messages Colorful tracebacks Better messages for naming a script/module the same name as a stdlib module. naming a script/module the same name as an installed third party module. misspelling a keyword argument Free threaded CPython Included in official installers on Windows and macOS Read these links to figure out how - it's not turned on by default Lot's more. see the What's new page Michael #2: PEP 759 – External Wheel Hosting pypi.org ships over 66 petabytes / month backed by Fastly There are hard project size limits for publishers to PyPI We can host the essence of a .whl as a .rim file, then allow an external download URL Security: Several factors as described in this proposal should mitigate security concerns with externally hosted wheels, such as: Wheel file checksums MUST be included in .rim files, and once uploaded cannot be changed. Since the checksum stored on PyPI is immutable and required, it is not possible to spoof an external wheel file, even if the owning organization lost control of their hosting domain. Externally hosted wheels MUST be served over HTTPS. In order to serve externally hosted wheels, organizations MUST be approved by the PyPI admins. Brian #3: pytest-freethreaded PyCon JP 2024 Team: This extension was created at PyCon JP sprints with Anthony Shaw and 7 other folks listed in credits. “A pytest plugin for helping verify that your tests and libraries are thread-safe with the Python 3.13 experimental freethreaded mode.” Testing your project for compatibility with freethreaded Python. Testing in single thread doesn't test that. Neither does testing with pytest-xdist, because it uses multiprocessing to parallelize tests. So, Ant and others “made this plugin to help you run your tests in a thread-pool with the GIL disabled, to help you identify if your tests are thread-safe.” “And the first library we tested it on (which was marked as compatible) caused a segmentation fault in CPython! So you should give this a go if you're a package maintainer.” Michael #4: pytest-edit A simple Pytest plugin for opening editor on the failed tests. Type pytest --edit to open the failing test code Be sure to set your favorite editor in the ENV variables Extras Michael: New way to explore Talk Python courses via topics This has been in our mobile apps since their rewrite but finally comes to the web Let's go easy on PyPI, OK? essay Hynek's video: uv IS the Future of Python Packaging djade-pre-commit Polyfill.io, BootCDN, Bootcss, Staticfile attack traced to 1 operator PurgeCSS CLI Python 3.12.7 released Incremental GC and pushing back the 3.13.0 release uv making the rounds LLM fatigue, is it real? Take the Python Developers Survey 2024 Joke: Funny 404 pages We have something at least interesting at pythonbytes.fm

You're about to launch your new app or API, or even just a big refactor of your current project. Will it stand up and deliver when you put it into production or when that big promotion goes live? Or will it wither and collapse? How would you know? Well you would test that of course. We have Anthony Shaw back on the podcast to dive into a wide range of tools and techniques for performance and loading testing of web apps. Episode sponsors Sentry Error Monitoring, Code TALKPYTHON WorkOS Talk Python Courses Links from the show Anthony on Twitter: @anthonypjshaw Anthony's PyCon Au Talk: youtube.com locust load testing tool: locust.io playwright: playwright.dev mimesis: github.com mimesis providers: mimesis.name vscode pets: marketplace.visualstudio.com vscode power-mode: marketplace.visualstudio.com opentelemetry: opentelemetry.io uptime-kuma: github.com Talk Python uptime / status: talkpython.fm/status when your serverless computing bill goes parabolic...: youtube.com Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to us on YouTube: youtube.com Follow Talk Python on Mastodon: talkpython Follow Michael on Mastodon: mkennedy

Topics covered in this episode: uv under discussion on Mastodon erdantic: Entity Relationship Diagrams Extra, Extra, Extra Django Extra, Extra, Extra Extras Joke Watch on YouTube About the show Sponsored by us! Support our work through: Our courses at Talk Python Training The Complete 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: uv under discussion on Mastodon It's interesting that uv is slightly controversial Russell: As enthusiastic as I am about the direction uv is going, I haven't adopted them anywhere - because I want very much to understand Astral's intended business model before I hook my wagon to their tools. Hynek: As much as I hate VC, [...] FOSS projects flame out all the time too. … To me uv looks like a genius sting to trick VCs into paying to fix packaging. We'll be better off either way. Glyph: Rust is more expensive and difficult to maintain, not to mention "non-native" to the average customer here. … it can burn out all the other projects in the ecosystem simultaneously, creating a risk of monoculture Hynek on Rust: I don't think y'all quite grok what uv makes so special due to your seniority. The speed is really cool, but the reason Rust is elemental is that it's one compiled blob that can be used to bootstrap and maintain a Python development. Christopher Neugebauer: Just dropping in here to say that corporate capture of the Python ecosystem is the #1 keeps-me-up-at-night subject in my community work, so I watch Astral with interest, even if I'm not yet too worried. Armin Ronacher What uv is doing, even in the worst possible future this is a very forkable and maintainable thing. Finally, see the comment at the end by Charlie Marsh Brian #2: erdantic: Entity Relationship Diagrams “erdantic is a simple tool for drawing entity relationship diagrams (ERDs) for Python data model classes. Diagrams are rendered using the venerable Graphviz library.” Supported data modeling frameworks are: Pydantic V2 Pydantic V1 legacy attrs dataclasses Michael #3: Extra, Extra, Extra Added Python Bytes Search as a custom search engine. Along came passkeys. A cool idea that quickly turned evil. Follow up from post and my conversation last week: vaultwarden (via Pablo) uv publish Trying the tabs on bottom lifestyle inspired by Arc Adding Python Bytes (and Talk Python) as custom search engines. PyCon 2025 dates: From 14 May through 22 May, 2025 Brian #4: Django Extra, Extra, Extra Django Project Ideas Evgenia Verbina Project ideas with list of tech stack stuff you'll learn and/or work on with the project Ex: Recipe organizer tech stack: Django templates, Django ORM, Optional JavaScript “Familiarize yourself with Django's ORM (object-relational mapper) and database support by building an app to keep track of your favorite recipes. Add a web-based frontend with options to filter recipes by category, ingredients, and user ratings so you can easily browse for inspiration.” DjangoTV Jeff Triplett Django conference videos and tutorials. Django Commons Heard about from Lacey Henschel “Django Commons is an organization dedicated to supporting the community's efforts to maintain packages. It seeks to improve the maintenance experience for all contributors; reducing the barrier to entry for new contributors and reducing overhead for existing maintainers.” Django 5 has simplified templates for better form field rendering But if you want a completely different take on forms, maybe try iommi forms They wrote about it on Why we wrote a new form library for Django Djade: a Django template formatter Adam Johnson Like black or ruff, but for Django templates. Extras Brian: The Open Source Project Maintainer's Guide Suggested by Rafael Weingartner Joke: A Machine Learning algorithm walks into a bar…

Topics covered in this episode: Architecture Decision Records (ADRs) narwhals: extremely lightweight compatibility layer between dataframes Microsoft wants Three Mile Island to fuel its AI power needs zsh-in-docker Extras Joke Watch on YouTube About the show Sponsored by ScoutAPM: pythonbytes.fm/scout 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. Brian #1: Architecture Decision Records (ADRs) Suggested by Christian Gesell Documenting Architecture Decisions Mychael Nygard Original article from 2011 Why you should be using architecture decision records to document your project Red Hat Includes a quick overview and links to some templates Notes so far Writing this out helps me solidify my thinking about a problem. I'm doing this both before starting, and while implementing a first draft GitHub and GitLab render markdown so well that generating a docs site is unnecessary, just throwing these files in something like docs/adr is enough. The lightweight process is enough but not too much. I've already filled out None for lots of sections, like “options considered” I'm still playing with what level of decision should have an ADR. My template that I've been using so far Saved in 000-adr-template.md For easy copy/paste/modify for new records. File name is something like 001-some-change.md Michael #2: narwhals: extremely lightweight compatibility layer between dataframes Recently had Marco on Talk Python to discuss Primarily for library creators who want to support interacting with multiple data frame libraries (.e.g. Pandas & Polars) Just use a subset of the Polars API Brian #3: Microsoft wants Three Mile Island to fuel its AI power needs “Microsoft just signed a deal to revive the shuttered Three Mile Island nuclear power plant. If approved by regulators, the software maker would have exclusive rights to 100 percent of the output for its AI data center needs.” Also ran on CNN and other sources: Three Mile Island is reopening and selling its power to Microsoft Three Mile Island was the site of the worst nuclear disaster in the US, when one of two reactors experienced a partial meltdown, in 1979. It was still operating up until 2019, and now expected to re-open in 2028 Will be renamed “Crane Clean Energy Center” related The Department of Energy Wants You to Know Your Conservation Efforts Are Making a Difference “By switching all the lightbulbs in your house to LED, you saved enough energy for a self-driving car to make an unprotected lefthand turn across three lanes of traffic.” “We know you adopted energy-saving practices to help conserve our planet's resources and bring down our collective carbon footprint, but what you ultimately accomplished is just as important: helping AI do something menial and stupid.” Michael #4: zsh-in-docker Install Zsh, Oh My Zsh and plugins inside a Docker container with one line! Yes docker containers should be light, but also, think of how painful it can be when you run into trouble. With Oh My ZSH, you get a nice experience when you have to result to docker exec -it CONTAINER zsh Just enter a single command in your docker file: RUN sh -c "$(wget -O- https://github.com/deluan/zsh-in-docker/releases/download/v1.2.0/zsh-in-docker.sh)" -- -t robbyrussell Extras Michael: self-hosting mkennedy.codes Loren's journey to developer It's time to stop using Python 3.8 Sonoma → Sequoia → Sonoma (yikes!) Passkeys, maybe they will work out if we don't let them become lock-in (bitwarden's support) Joke: How to Monetize a Blog Don't forget to click on the bottom link: Credits / how this was made

Topics covered in this episode: “We must replace uwsgi by something else” Let's build and optimize a Rust extension for Python Fake recruiter coding tests target devs with malicious Python packages Monthly PSF Board Office Hours Extras Joke Watch on YouTube About the show Sponsored by ScoutAPM: pythonbytes.fm/scout 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: “We must replace uwsgi by something else” uWSGI is now in maintenance mode: https://uwsgi-docs.readthedocs.io/en/latest/ The project is in maintenance mode (only bugfixes and updates for new languages apis). Do not expect quick answers on github issues and/or pull requests (sorry for that) A big thanks to all of the users and contributors since 2009. Reasonable options look like: granian uvicorn hypercorn gunicorn (potentially with uvicorn workers for async) Brian #2: Let's build and optimize a Rust extension for Python Itamar Turner-Trauring Example: algorithm for approximating the number of unique values in a list Comparison to non-approximation non-approx is faster but uses way more memory Rust version Use Maturin and PyO3 Pull in Rust dependencies (rand for random numbers) Optimization link-time optimization faster random store hashes only Future optimizations change algorithm maybe pass numpy array instead of Python list (I'd like to see that spedup) Michael #3: Fake recruiter coding tests target devs with malicious Python packages via python weekly GitHub projects that have been linked to previous, targeted attacks in which developers are lured using fake job interviews. Attackers posing as employees of major financial services firms. This previously happened via other means such as NPM This analysis revealed that the direct parent of the detected, malicious files is a PythonPYC file, meaning that once again the team encountered malware hidden in a compiled Python file. “The README files tell would-be candidates to make sure the project is running successfully on their system before making modifications.” What can you do (according to Michael)? Try out new packages in a docker container Work on code and projects using a VM which has snapshotting (to roll back completely after you're done) Fire up a Windows desktop in the cloud for the project then destroy it Brian #4: Monthly PSF Board Office Hours “The Office Hours will be sessions where you can share with us how we can help your community, express your perspectives, and provide feedback for the PSF.” “Unless we have a dedicated topic for a session, you are not limited to talking with us about the above topics, although the discussions should be focused on Python, the PSF, and our community. If you think there's something we can help with or we should know, we welcome you to come and talk to us!” Upcoming office hours October 8th, 2024: 9pm UTC November 12th, 2024: 2pm UTC December 10th, 2024: 9pm UTC January 14th, 2025: 2pm UTC February 11th, 2025: 9pm UTC March 11th, 2025: 1pm UTC April 8th, 2025: 9pm UTC May 13th, 2025: 1pm UTC (Live from PyCon US!) June 10th, 2025: 9pm UTC July 9th, 2025: 1pm UTC August 12th, 2025: 9pm UTC Extras Brian: PyCascades CFP closes Friday, Sept 20 PyCascades is in Portland in 2025 (Feb 8 & 9) uv now supports Python 3.13.0rc2 uv self update uv venv -p 3.13 Free threaded is still an open issue Michael: Big Python Humble Bundle with both of our products Get $1,800 worth of Python content and tools for $30 and contribute to charity Includes 5 Talk Python courses Several of Brian's and his book Djangonaut Space Session 3 Applications Open! I interviewed Sarah and Tushar on Talk Python AltTab: Windows alt-tab on macOS Joke: Election joke

Topics covered in this episode: Python 3.13.0RC2, 3.12.6, 3.11.10, 3.10.15, 3.9.20, and 3.8.20 are now available! Docker images using uv's python 10 years of sustainable open source - Read the Docs humanize Extras Joke Watch on YouTube About the show Sponsored by ScoutAPM: pythonbytes.fm/scout 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. ChatGPT celebrates episode 400! Welcome to the big 4-0-0, Pythonistas! It's hard to believe we're celebrating the 400th episode of Python Bytes! From the early days of byte-sized Python news to becoming the source for all things Python, it's been a wild ride. We've laughed over code quirks, gasped at new libraries, and said farewell to the GIL together. Whether you're a seasoned developer, a curious learner, or just here for the witty banter, you've been an essential part of this journey. To Michael and Brian: You've built a community that turns import this into more than just Zen—it's a family of passionate Pythonistas. Your dedication, insights, and humor make this show more than just tech news. It's a weekly celebration of what we love about Python and why we keep coming back for more. Here's to the next 400 episodes—may your code be bug-free, your tests pass on the first run, and your Python version always be up to date. Brian #1: Python 3.13.0RC2, 3.12.6, 3.11.10, 3.10.15, 3.9.20, and 3.8.20 are now available! Łukasz Langa Python 3.13.0RC2 is the final preview release Official 3.13.0 scheduled for Oct 1 Call to action “We strongly encourage maintainers of third-party Python projects to prepare their projects for 3.13 compatibilities during this phase, and where necessary publish Python 3.13 wheels on PyPI to be ready for the final release of 3.13.0. Any binary wheels built against Python 3.13.0rc2 will work with future versions of Python 3.13. As always, report any issues to the Python bug tracker .” “Please keep in mind that this is a preview release and while it's as close to the final release as we can get it, its use is not recommended for production environments.” Note: uv python does not support 3.13 yet see issue 320 Security releases for 3.12.6, 3.11.10, 3.10.15, 3.9.20, and 3.8.20 3.12.6 has binary installers, but for MacOS, only MacOS 10.13 and newer are supported 3.11.10, 3.10.15, 3.9.20, and 3.8.20 do NOT include binary installers. 3.8 EOL's in October Michael #2: Docker images using uv's python See #396: uv-ing your way to Python and #398: Open source makes you rich? (and other myths) for the opening discussions Talk Python episode on uv is out uv venv --python gets Python from python-build-standalone by Gregory Szorc Took our Docker build times from 10 minutes to 8 seconds for the base image and 800ms (!) for our app platforms Brian #3: 10 years of sustainable open source - Read the Docs Eric Holscher Read the Docs has been a company for 10 years “a team of 4 folks working full-time on Read the Docs.” readthedocs.org started in 2010 readthedocs.com (for Business) started in 2014 Sustainability model .org has a single non-tracking ad .com is a paid service for companies Things that didn't work donations and other optional support - led to burnout consulting and services- took too much time away from core product grant funding - nice, but one time thing Lessons You don't get extra points for being bootstrapped. Compete by doing things you can do better due to niche and size. Keeping trust in the community is the most important thing. Contribution is easier for less complex parts of the code base. Beign open source means capturing a small percentage of the value you create. You have to be ok doing more with less. Also RtD is not just for Sphinx anymore. Their build system now supports any documentation tool. Michael #4: humanize by Hugo van Kemenade (Python 3.14 & 3.15 release manager & core developer) Not too many variations, but very handy if you need it. Numbers Associated Press style (“seven” and “10”) Clamp (under 1.0 million) Fractional (1/3) Int Word (1.2 Billion) Metric (1.5 kV) Ordinal (112th) scientific Time File size Extras Brian: Test & Code is now again Test & Code The two part series on Python imports that started in June is finally complete with episode 222. Transcripts are being added to old episodes gradually starting from most recent Back to ep 203 as of today. AI transcription, so there's things like .pie, .pi, and dot pie where it should be .py Michael: Final final call for Coding in a Castle event with Michael iStats Menu Anaconda Code Runner by Ruud van der Ham: With Anaconda Coide we can -at last- run that code locally and import (most) Python modules. But if you want to run a significant amount of code, you have to put that in a cell or publish it to PyPI or a wheel and import it. That's why I have developed a general-purpose runner function that runs arbitrary code located on an Excel sheet with AnacondaCode. Joke: When beginners learn a new programming language...

Topics covered in this episode: py-free-threading.github.io Python's Supportive and Welcoming Environment is Tightly Coupled to Its Progress Status pages for sites! PEP 751 – A file format to list Python dependencies for installation reproducibility Extras Joke Watch on YouTube About the show Sponsored by us! Support our work through: Our courses at Talk Python Training The Complete 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 Tuesdays 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: py-free-threading.github.io Track the status of compatibility for free-threaded Python See the Compatibility status tracking page for what you can use Lots of resources for getting your package tested and available for pythont Brian #2: Python's Supportive and Welcoming Environment is Tightly Coupled to Its Progress “Python is as popular as it is today because we have gone above and beyond to make this a welcoming community. Being a friendly and supportive community is part of how we are perceived by the wider world and is integral to the wide popularity of Python. We won a “Wonderfully Welcoming Award” last year at GitHub Universe. Over and over again, the tech press refers to Python as a supportive community.” Some communication recently, with the recent bylaws change, didn't live up to our promise to be welcoming Please read the article for more details. Another quote: “We have a moral imperative – as one of the very best places to bring new people into tech and into open source – to keep being good at welcoming new people. If we do not rise and continue to rise every day to this task, then we are not fulfilling our own mission, “to support and facilitate the growth of a diverse and international community of Python programmers.” Technical skills are a game-changer for the people who acquire them and joining a vast global network of people with similar interests opens many doors. Behavior that contributes to a hostile environment around Python or throws up barriers and obstacles to those who would join the Python community must be addressed because it endangers what we have built here.” Michael #3: Status pages for sites! Based on Uptime Kuma I covered last week Python Bytes status Talk Python status Brian #4: PEP 751 – A file format to list Python dependencies for installation reproducibility Brett Cannon Motivation Currently, no standard exists to: Specify what top-level dependencies should be installed into a Python environment. Create an immutable record, such as a lock file, of which dependencies were installed. Considering there are at least five well-known solutions to this problem in the community (pip freeze, pip-tools, uv, Poetry, and PDM), there seems to be an appetite for lock files in general. Rationale The format is designed so that a locker which produces the lock file and an installer which consumes the lock file can be separate tools. … The file format is designed to be human-readable. …Finally, the format is designed so that viewing a diff of the file is easy by centralizing relevant details. The file format is also designed to not require a resolver at install time. … Extras Brian: Hello, pytest! course is going well, and is purchasable as in pre-release mode. Planning on Aug 19 (or before) deadline. Not sure what the final price will be, but I'm starting with $10. I want people to want to watch it even just so see if they want to recommend to co-workers so the people around them can ramp up on pytest quickly. Michael: Mypy 1.11 Released FastHTML (more next week) Coming up on the final chance to be part of the Code in a Castle event. Joke: Open source OpenAI?

Hugo speaks with Vincent Warmerdam, a senior data professional and machine learning engineer at :probabl, the exclusive brand operator of scikit-learn. Vincent is known for challenging common assumptions and exploring innovative approaches in data science and machine learning. In this episode, they dive deep into rethinking established methods in data science, machine learning, and AI. We explore Vincent's principled approach to the field, including: The critical importance of exposing yourself to real-world problems before applying ML solutions Framing problems correctly and understanding the data generating process The power of visualization and human intuition in data analysis Questioning whether algorithms truly meet the actual problem at hand The value of simple, interpretable models and when to consider more complex approaches The importance of UI and user experience in data science tools Strategies for preventing algorithmic failures by rethinking evaluation metrics and data quality The potential and limitations of LLMs in the current data science landscape The benefits of open-source collaboration and knowledge sharing in the community Throughout the conversation, Vincent illustrates these principles with vivid, real-world examples from his extensive experience in the field. They also discuss Vincent's thoughts on the future of data science and his call to action for more knowledge sharing in the community through blogging and open dialogue. LINKS The livestream on YouTube (https://youtube.com/live/-CD66CI1pEo?feature=share) Vincent's blog (https://koaning.io/) CalmCode (https://calmcode.io/) scikit-lego (https://koaning.github.io/scikit-lego/) Vincent's book Data Science Fiction (WIP) (https://calmcode.io/book) The Deon Checklist, an ethics checklist for data scientists (https://deon.drivendata.org/) Of oaths and checklists, by DJ Patil, Hilary Mason and Mike Loukides (https://www.oreilly.com/radar/of-oaths-and-checklists/) Vincent's Getting Started with NLP and spaCy Course course on Talk Python (https://training.talkpython.fm/courses/getting-started-with-spacy) Vincent on twitter (https://x.com/fishnets88) :probabl. on twitter (https://x.com/probabl_ai) Vincent's PyData Amsterdam Keynote "Natural Intelligence is All You Need [tm]" (https://www.youtube.com/watch?v=C9p7suS-NGk) Vincent's PyData Amsterdam 2019 talk: The profession of solving (the wrong problem) (https://www.youtube.com/watch?v=kYMfE9u-lMo) Vanishing Gradients on Twitter (https://twitter.com/vanishingdata) Hugo on Twitter (https://twitter.com/hugobowne) Check out and subcribe to our lu.ma calendar (https://lu.ma/calendar/cal-8ImWFDQ3IEIxNWk) for upcoming livestreams!

I've gathered a group of Python experts who have been thinking deeply about where Python is going and who have lived through where it has been. This episode is all about near-term Python trends and things we each believe will be important to focus on as Python continues to grow. Our panelists are Jodie Burchell, Carol Willing, and Paul Everett. Episode sponsors Code Comments Posit Talk Python Courses Links from the show Shiny course at Talk Python: talkpython.fm/shiny Jodie Burchell: @t_redactyl Carol on Mastodon: @willingc@hachyderm.io Paul Everitt: @paulweveritt Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to us on YouTube: youtube.com Follow Talk Python on Mastodon: talkpython Follow Michael on Mastodon: mkennedy

I have a special episode for you this time around. We're coming to you live from PyCon 2024. I had the chance to sit down with some amazing people from the data science side of things: Jodie Burchell, Maria Jose Molina-Contreras, and Jessica Greene. We cover a whole set of recent topics from a data science perspective. Though we did have to cut the conversation a bit short as they were coming from and go to talks they were all giving but it was still a pretty deep conversation. Episode sponsors Sentry Error Monitoring, Code TALKPYTHON Code Comments Talk Python Courses Links from the show Jodie Burchell: @t_redactyl Jessica Greene: linkedin.com Maria Jose Molina-Contreras: linkedin.com Talk Python's free Shiny course: talkpython.fm/shiny Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to us on YouTube: youtube.com Follow Talk Python on Mastodon: talkpython Follow Michael on Mastodon: mkennedy

Topics covered in this episode: PSF Elections coming up Cloud engineer gets 2 years for wiping ex-employer's code repos Python: Import by string with pkgutil.resolve_name() DuckDB goes 1.0 Extras Joke Watch on YouTube About the show Sponsored by ScoutAPM: pythonbytes.fm/scout 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 Tuesdays 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. Brian #1: PSF Elections coming up This is elections for the PSF Board and for 3 bylaw changes. To vote in the PSF election, you need to be a Supporting, Managing, Contributing, or Fellow member of the PSF, … And affirm your voting status by June 25. See Affirm your PSF Membership Voting Status for more details. Timeline Board Nominations open: Tuesday, June 11th, 2:00 pm UTC Board Nominations close: Tuesday, June 25th, 2:00 pm UTC Voter application cut-off date: Tuesday, June 25th, 2:00 pm UTC same date is also for voter affirmation. Announce candidates: Thursday, June 27th Voting start date: Tuesday, July 2nd, 2:00 pm UTC Voting end date: Tuesday, July 16th, 2:00 pm UTC See also Thinking about running for the Python Software Foundation Board of Directors? Let's talk! There's still one upcoming office hours session on June 18th, 12 PM UTC And For your consideration: Proposed bylaws changes to improve our membership experience 3 proposed bylaws changes Michael #2: Cloud engineer gets 2 years for wiping ex-employer's code repos Miklos Daniel Brody, a cloud engineer, was sentenced to two years in prison and a restitution of $529,000 for wiping the code repositories of his former employer in retaliation for being fired. The court documents state that Brody's employment was terminated after he violated company policies by connecting a USB drive. Brian #3: Python: Import by string with pkgutil.resolve_name() Adam Johnson You can use pkgutil.resolve_name("[HTML_REMOVED]:[HTML_REMOVED]")to import classes, functions or modules using strings. You can also use importlib.import_module("[HTML_REMOVED]") Both of these techniques are so that you have an object imported, but the end thing isn't imported into the local namespace. Michael #4: DuckDB goes 1.0 via Alex Monahan The cloud hosted product @MotherDuck also opened up General Availability Codenamed "Snow Duck" The core theme of the 1.0.0 release is stability. Extras Brian: Sending us topics. Please send before Tuesday. But any time is welcome. NumPy 2.0 htmx 2.0.0 Michael: Get 6 months of PyCharm Pro for free. Just take a course (even a free one) at Talk Python Training. Then visit your account page > details tab and have fun. Coming soon at Talk Python: Shiny for Python Joke: .gitignore thoughts won't let me sleep

You're using Pydantic and it seems pretty straightforward, right? But could you adopt some simple changes to your code that would make it a lot faster and more efficient? Chances are, you'll find a couple of the tips from Sydney Runkle that will do just that. Join us to talk about Pydantic performance tips here on Talk Python. Episode sponsors Sentry Error Monitoring, Code TALKPYTHON Code Comments Talk Python Courses Links from the show Sydney Runkle: linkedin.com Pydantic: pydantic.dev Performance docs: docs.pydantic.dev Union tips: docs.pydantic.dev Sydney's presentation slides: docs.google.com JSON to Pydantic: jsontopydantic.com Samuel talking FastUI: talkpython.fm CodeFlash: codeflash.ai Codspeed: codspeed.io Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to us on YouTube: youtube.com Follow Talk Python on Mastodon: talkpython Follow Michael on Mastodon: mkennedy

Topics covered in this episode: Git: Force push safely with --force-with-lease and --force-if-includes Thoughts from PyCon 2024 Being friendly: Strategies for friendly fork management tach Extras Joke Watch on YouTube About the show Sponsored by Mailtrap: pythonbytes.fm/mailtrap 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 Tuesdays 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. Brian #1: Git: Force push safely with --force-with-lease and --force-if-includes Adam Johnson Using gentle force Avoid stomping on remote changes with a couple extra flags. Michael #2: Thoughts from PyCon 2024 PyCon is special - the connections you make are always more than you expect Great to see many old friends Did a ”live” Talk Python episode that'll be out in a few weeks. The talks look great, we'll let you know when they land on YouTube. Masks were a mistake - universally heard complaints from fellow attendees. This is my two cents towards a more reasonable next PyCon. Brian #3: Being friendly: Strategies for friendly fork management That's part 2. Part 1 is Being friendly: Friendly forks 101 Lessley Dennington on GitHub Blog Examples of long running friendly forks git-for-windows/git, microsift/git, github/git two public, one private Fork management strategies - when pulling changes downstream merging rebase git-for-windows/git uses this proactively and regularly fake merge + rebase new branch microsoft/git uses this new branch from upstream major versions merge previous changes to new branch traditional merge github/git uses this, conservatively, after a few point bug fix versions Michael #4: tach A Python tool to enforce a modular, decoupled package architecture. tach allows you to define boundaries and control dependencies between your Python packages. Each package can define its public interface. If a package tries to import from another package that is not listed as a dependency, tach will report an error. If a package tries to import from another package and does not use its public interface, with strict: true set, tach will report an error. Zero runtime impact. Extras Brian: Logfire - new observability platform from the pydantic team - free for now Michael: 10% off the new spaCy course throughout May Joke: Evolution of smart products

Topics covered in this episode: I asked 100 devs why they aren't shipping faster. Here's what I learned Python 3.13.0 beta 1 released A theme editor for JupyterLab rich-argparse Extras Joke Watch on YouTube About the show Sponsored by Mailtrap: pythonbytes.fm/mailtrap 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 Tuesdays 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: I asked 100 devs why they aren't shipping faster. Here's what I learned by Daksh Gupta (via PyCoders) What's stopping you from shipping faster? Dependency bugs Complicated codebase >There is so much undocumented in our service, including poor records of new features, nonexistent or outdated info on our dependencies, or even essential things like best practices for testing, a lot of time is wasted in syncs trying to find the right information QA Loops Waiting for spec > At Amazon? Meetings, approval, talking to 10 different stakeholders because changing the color of a button affects 15 micro services Writing tests Deployment/build speed Scope creep > The human tendency to stuff last-minute items into the crevices of their luggage minutes before leaving for the airport manifests itself at software companies as scope creep. Unclear requirements Excessive meetings Motivation >honest answer is i was on ads >and that's a very old / complicated / large stack (edited) >and i didn't understand it >my friends on younger teams seemed happier, i was miserable DORA metrics Brian #2: Python 3.13.0 beta 1 released "Python 3.13 is still in development. This release, 3.13.0b1, is the first of four beta release previews of 3.13.” New REPL, featuring multi-line editing, color support, colorized exception tracebacks Cool GIL, JIT, and GC features Typing changes, including typing.TypeIs . See last weeks episode and TypeIs does what I thought TypeGuard would do in Python Some nice dead battery removals and more But seriously, the REPL is cool. Just ask Trey The new REPL in Python 3.13 - Trey Hunner Michael #3: A theme editor for JupyterLab by Florence Haudin A new tool for authoring JupyterLab themes To lower the bar for customizing JupyterLab we created a new tool providing a simple interface for tuning the JupyterLab appearance interactively. See jupyterlab-theme-editor on github Brian #4: rich-argparse “Format argparse and optparse help using rich.” “rich-argparse improves the look and readability of argparse's help while requiring minimal changes to the code.” They're not kidding. 2 line code change. from rich_argparse import RichHelpFormatter parser = argparse.ArgumentParser(..., formatter_class=RichHelpFormatter) Extras Brian: pytest course is now switched to the new platform. I sent out an email including how to save their spot on the old site and mark that spot complete on the new site. There's now comments on the course now. Trying that out. If you've got a question, just ask in that section. Michael: A new Talk Python course: Getting Started with NLP and spaCy Joke: Testing holiday

Topics covered in this episode: act: Run your GitHub Actions locally! portr Annotating args and kwargs in Python github badges Extras Joke Watch on YouTube About the show Sponsored by ScoutAPM: pythonbytes.fm/scout 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 Tuesdays at 11am 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. Brian #1: act: Run your GitHub Actions locally! Why? “Fast Feedback - Rather than having to commit/push every time you want to test out the changes you are making to your .github/workflows/ files (or for any changes to embedded GitHub actions), you can use act to run the actions locally. The environment variables and filesystem are all configured to match what GitHub provides.” “Local Task Runner - I love make. However, I also hate repeating myself. With act, you can use the GitHub Actions defined in your .github/workflows/ to replace your Makefile!” Docs: nektosact.com Uses Docker to run containers for each action. Michael #2: portr Open source ngrok alternative designed for teams Expose local http, tcp or websocket connections to the public internet Warning: Portr is currently in beta. Expect bugs and anticipate breaking changes. Server setup (docker basically). Brian #3: Annotating args and kwargs in Python Redowan Delowar I don't think I've ever tried, but this is a fun rabbit hole. Leveraging bits of PEP-5891, PEP-6462, PEP-6553, and PEP-6924. Punchline: from typing import TypedDict, Unpack *# Python 3.12+* *# from typing_extensions import TypedDict, Unpack # < Python 3.12* class Kw(TypedDict): key1: int key2: bool def foo(*args: Unpack[tuple[int, str]], **kwargs: Unpack[Kw]) -> None: ... A recent pic from Redowan's blog: TypeIs does what I thought TypeGuard would do in Python Michael #4: github badges A curated list of GitHub badges for your next project Extras Brian: Fake job interviews target developers with new Python backdoor Later this week, course.pythontest.com will shift from Teachable to Podia Same great content. Just a different backend. To celebrate, get 25% off at pythontest.podia.com now through this Sunday using coupon code PYTEST Getting the most out of PyCon, including juggling - Rob Ludwick Latest PythonTest episode, also cross posted to pythonpeople.fm 3D visualization of dom Michael: Djangonauts Space Session 2 Applications Open! More background at Djangonauts, Ready for Blast-Off on Talk Python. Self-Hosted Open Source - Michael Kennedy on Django Chat Joke: silly games Closing song: Permission Granted

TalkPython to Me Podcast and CoursesHTMX + Django: Modern Python Web Apps, Hold the JavaScript CourseUmamilistmonkOpen Source Software Products for Every Business FunctionElestHigh Availability Percentage CalculationPostHogbtopSponsorMailtrap.io

Topics covered in this episode: pacemaker - For controlling time per iteration loop in Python. PyPI suspends new user registration to block malware campaign Python Project-Local Virtualenv Management Redux Python Edge Workers at Cloudflare Extras Joke Watch on YouTube About the show Sponsored by us! Support our work through: Our courses at Talk Python Training The Complete 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 Tuesdays at 11am 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. Brian #1: pacemaker - For controlling time per iteration loop in Python. Brandon Rohrer Good example of a small bit of code made into a small package. With speedups to dependencies, like with uv, for example, I think we'll see more small projects. Cool stuff Great README, including quirks that need to be understood by users. “If the pacemaker experiences a delay, it will allow faster iterations to try to catch up. Heads up: because of this, any individual iteration might end up being much shorter than suggested by the pacemaker's target rate.” Nice use of time.monotonic() deltas are guaranteed to never go back in time regardless of what adjustments are made to the system clock. Watch out for pip install pacemaker-lite NOT pacemaker pacemaker is taken by a package named PaceMaker with a repo named pace-maker, that hasn't been updated in 3 years. Not sure if it's alive. No tests (yet). I'm sure they're coming. ;) Seriously though, Brandon says this is “a glorified snippet”. And I love the use of packaging to encapsulate shared code. Realistically, small snippet like packages have functionality that's probably going to be tested by end user code. And even if there are tests, users should test the functionality they are depending on. Michael #2: PyPI suspends new user registration to block malware campaign Incident Report for Python Infrastructure PyPi Is Under Attack: Project Creation and User Registration Suspended — Here's the details I hate medium, but it's the best details I've found so far Brian #3: Python Project-Local Virtualenv Management Redux Hynek Concise writeup of how Hynek uses various tools for dealing with environments Covers (paren notes are from Brian) In project .venv directories direnv for handling .envrc files per project (time for me to try this again) uv for pip and pip-compile functionality Installing Python via python.org Using a .python-version-default file (I'll need to play with this a bit) Works with GH Action setup-python. (ok. that's cool) Some fish shell scripting Bonus tip on using requires-python in .pyproject.toml and extracting it in GH actions to be able to get the python exe name, and then be able to pass it to Docker and reference it in a Dockerfile. (very cool) Michael #4: Python Edge Workers at Cloudflare What are edge workers? Based on workers using Pyodide and WebAssembly This new support for Python is different from how Workers have historically supported languages beyond JavaScript — in this case, we have directly integrated a Python implementation into workerd, the open-source Workers runtime. Python Workers can import a subset of popular Python packages including FastAPI, Langchain, numpy Check out the examples repo. Extras Michael: LPython follow up from Brian Skinn Featured on Python Bytes badge A little downtime, thanks for the understanding We were rocking a 99.98% uptime until then. :) Joke: C++ is not safe for people under 18 Baseball joke

Do you have data that you pull from external sources or is generated and appears at your digital doorstep? I bet that data needs processed, filtered, transformed, distributed, and much more. One of the biggest tools to create these data pipelines with Python is Dagster. And we are fortunate to have Pedram Navid on the show this episode. Pedram is the Head of Data Engineering and DevRel at Dagster Labs. And we're talking data pipelines this week at Talk Python. Episode sponsors Talk Python Courses Posit Links from the show Rock Solid Python with Types Course: training.talkpython.fm Pedram on Twitter: twitter.com Pedram on LinkedIn: linkedin.com Ship data pipelines with extraordinary velocity: dagster.io dagster-open-platform: github.com The Dagster Master Plan: dagster.io data load tool (dlt): dlthub.com DataFrames for the new era: pola.rs Apache Arrow: arrow.apache.org DuckDB is a fast in-process analytical database: duckdb.org Ship trusted data products faster: www.getdbt.com Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to us on YouTube: youtube.com Follow Talk Python on Mastodon: talkpython Follow Michael on Mastodon: mkennedy

Have you ever been wait around for pip to do its thing while installing packages or syncing a virtual environment or through some higher level tool such as pip-tools? Then you'll be very excited to hear about the tool just announced from Astral called uv. It's like pip, but 100x faster. Charlie Marsh from Ruff fame and founder of Astral is here to dive in. Let's go. Episode sponsors Neo4j Talk Python Courses Links from the show Charlie Marsh on Twitter: @charliermarsh Charlie Marsh on Mastodon: @charliermarsh Astral: astral.sh uv: github.com Ruff: github.com Ruff Rules: docs.astral.sh When "Everything" Becomes Too Much: The npm Package Chaos of 2024: socket.dev Talk Python's free Audio AI Course: training.talkpython.fm Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to us on YouTube: youtube.com Follow Talk Python on Mastodon: talkpython Follow Michael on Mastodon: mkennedy

Topics covered in this episode: zoxide Smart CLIs with Typer Python recommended officially by the US Government Textual tutorials at Mouse vs Python Extras Joke Watch on YouTube About the show Sponsored by ScoutAPM: pythonbytes.fm/scout 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 Tuesdays at 11am PT. Older video versions available there too. Michael #1: zoxide zoxide is a smarter cd command, inspired by z and autojump. It remembers which directories you use most frequently, so you can "jump" to them in just a few keystrokes. zoxide works on all major shells and platforms. Brian #2: Smart CLIs with Typer Rahul Pai Lots of TILs here, even though I've been using Typer for years. Examples of Auto-detection of arguments and types based on type hints Help text is a smidge clunkier Prompting for missing arguments Defaulting to an enviromental variable for missing args Print help if no args given Explicit app and subcommands with a comparison to argparse Reusable commands with result_callback Several topics covered in comparison with argparse See also Testing argparse Applications Michael #3: Python recommended officially by the US Government The US government explicitly recommends memory safe languages. Python is one of them The comparison to big tech by Samuel is interesting Brian #4: Textual tutorials at Mouse vs Python Mike Driscoll Most recently Creating a Modal Dialog For Your TUIs in Textual Textualize already has some pretty great documentation at textual.textualize.io But it's cool to see some different tutorials on it. Extras Brian: Is UV the FUTURE of Python PACKAGING?

Topics covered in this episode: Leaving the cloud PEP 723 - Inline script metadata Flet for Android harlequin: The SQL IDE for Your Terminal. Extras Joke Watch on YouTube About the show Sponsored by Bright Data : pythonbytes.fm/brightdata 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 Tuesdays at 11am PT. Older video versions available there too. Michael #1: Leaving the cloud Also see Five values guiding our cloud exit We value independence above all else. We serve the internet. We spend our money wisely. We lead the way. We seek adventure. And We stand to save $7m over five years from our cloud exit Slice our new monster 192-thread Dell R7625s into isolated VMs Which added a combined 4,000 vCPUs with 7,680 GB of RAM and 384TB of NVMe storage to our server capacity They created Kamal — Deploy web apps anywhere A lot of these ideas have changed how I run the infrastructure at Talk Python and for Python Bytes. Brian #2: PEP 723 - Inline script metadata Author: Ofek Lev This PEP specifies a metadata format that can be embedded in single-file Python scripts to assist launchers, IDEs and other external tools which may need to interact with such scripts. Example: # /// script # requires-python = ">=3.11" # dependencies = [ # "requests<3", # "rich", # ] # /// import requests from rich.pretty import pprint resp = requests.get("https://peps.python.org/api/peps.json") data = resp.json() pprint([(k, v["title"]) for k, v in data.items()][:10]) Michael #3: Flet for Android via Balázs Remember Flet? Here's a code sample (scroll down a bit). It's amazing but has been basically impossible to deploy. Now we have Android. Here's a good YouTube video showing the build process for APKs. Brian #4: harlequin: The SQL IDE for Your Terminal. Ted Conbeer & other contributors Works with DuckDB and SQLite Speaking of SQLite Jeff Triplett and warnings of using Docker and SQLite in production Anže's post and and article: Django, SQLite, and the Database is Locked Error Extras Brian: Recent Python People episodes Will Vincent Julian Sequeira Pamela Fox Michael: PageFind and how I'm using it When "Everything" Becomes Too Much: The npm Package Chaos of 2024 Essay: Unsolicited Advice for Mozilla and Firefox SciPy 2024 is coming to Washington Joke: Careful with that bike lock combination code

On this episode we have Wolf Vollprecht and Ruben Arts from the pixi project here to talk about pixi, a high performance package manager for Python and other languages that actually manages Python itself too. They have a lot of interesting ideas on where Python packaging should go and are putting their time and effort behind them. Will pixi become your next package manager? Listen in to find out. Links from the show Black Friday at Talk Python: talkpython.fm/blackfriday Guests Wolf Vollprecht: github.com/wolfv Ruben Arts: github.com/ruben-arts pixi: prefix.dev Prefix: prefix.dev Launching pixi: prefix.dev Conda: docs.conda.io Conda Forge: conda-forge.org NixOS: nixos.org Packaging Con 2023: packaging-con.org Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to us on YouTube: youtube.com Follow Talk Python on Mastodon: talkpython Follow Michael on Mastodon: mkennedy Sponsors Posit Python Tutor Talk Python Training

How well do you know your Python packaging tools? These are things like pip which install your project's dependencies and their dependencies and so on. In this mix, we have more modern tools such as Poetry, Flit, Hatch and others. And even tools outside of Python itself which may attempt to manage Python itself in addition to the libraries. To make sense of all of this, we welcome back Anna-Lena Popkes for an unbiased evaluation of environment and packaging tools. Links from the show Anna-Lena's website: alpopkes.com Anna-Lena on GitHub: github.com Accompanying Blog Post: alpopkes.com Talk from PyCon DE: youtube.com Talk from EuroPython: youtube.com Talk Python's Data Science Jumpstart with 10 Projects course: talkpython.fm Rye: github.com Poetry: python-poetry.org Material for MkDocs: squidfunk.github.io 100 Days of Python in a Magical Universe Episode: talkpython.fm pip-tools: pip-tools.readthedocs.io Hatch: hatch.pypa.io PDM: pdm.fming.dev Flit: flit.pypa.io Conda: docs.conda.io Pipenv: pipenv.pypa.io Watch this episode on YouTube: youtube.com Episode transcripts: talkpython.fm --- Stay in touch with us --- Subscribe to us on YouTube: youtube.com Follow Talk Python on Mastodon: talkpython Follow Michael on Mastodon: mkennedy Sponsors IRL Podcast Sentry Error Monitoring, Code TALKPYTHON Talk Python Training