Published

PEP 735 describes dependency groups as sets of optional dependencies that are not shipped with the package but intended for development purposes.

The PEP includes an example for groups that include test, docs, typing, and a combo typing-test.

[dependency-groups] test = ["pytest", "coverage"] docs = ["sphinx", "sphinx-rtd-theme"] typing = ["mypy", "types-requests"] typing-test = [{include-group = "typing"}, {include-group = "test"}, "useful-types"]

This is implemented in uv and can be used by several of their commands.

uv sync --group test uv run --group test uv add --group test pytest uv remove --group test pytest uv export --group test uv tree --group test

Dependency Groups are not Extras #

The docs describe extras as being intended to ship with the application and dependency groups intended for development. The spec allows both to exist with the same name, but care should be taken as tools may have different implementations.

...

Running software applications in production today is crazy. One point release opens up for supply chain attacks. What’s crazier is not running your production applications without a lock file, potentially running dependencies you’ve never ran before for the first time in prod.

I really like how well the local dev is setup to run off of production data here. I’ll use this as a reminder that I need to set up lite stream on a few of my projects that it’s missing from and include a nice sync prod data Posts tagged: justfile recipe.

Litestreams interface always throws me for a loop. It works fantastic, but the global config stored in /etc and some of the commands break my brain. It’s not you it’s me.

Using real data when you can is goated. Fake data is so often a perfect example of what someone thinks the backend should look like and does not include things that users actually do, running pipelines for days, or setting titles to paragraphs worth of text. Obviously this is not possible everywhere and the more sensitive your data the harder that process becomes.

I smell a dependency to python-dotenv dying in my workflow. I originally read the title of the post and thought, “I know how to manage .env and almost skipped it”. I’m leaning more and more on uv run these days, so this should just go in my [[ just file ]] to make it easy to run.

Check out sidekick.nvim by folke. It’s a well-crafted project with great potential.

Your Neovim AI sidekick

I’m really excited about mdserve, an amazing project by jfernandez. It’s worth exploring!

Fast markdown preview server with live reload and theme support.

Interesting catch from the HN discussion over his article that came out yestereday. I scanned it yesterday and it has some really fascinating diagrams showing different phases of the web being open, to being siloed, to somewhere that we are trying to make it easy to publish, and retain ownership. I don’t know enough about bluesky, but the core is build on the AT protocol, you can self host your own instance, you can build different front ends for it. So rather than having siloed instagram, FB, twitter, there are clones of those platforms that read the same data from everyones data, that they have the option of self hosting.

I like this distinction between Mastadon here. Mastadon can also be self host, but its data aggregation is...

This is a super cool movement, I like the idea of giving access to composable components like we have in open source. You want to build a website you have a bunch of options from raw dogging assembly all the way up to predefined templates that just need your content. Idk if the analogy is perfect but there are aspects of it that work. I see where right now we are somewhere in raw dogging c or python. We have cheap nuts and bolts and some low level things, but once someone needs some coupler like this it’s dropping down to drawing it by hand.

Wow, I’ve never seen or thought of multi setup parts this is very thought provoking, not sure how useful it is as we have good adhesives and stuff for printed parts. I definitely want to try this though

I can’t believe this thing is so devicive. I kinda can’t belive that I sit on the same side as Mathes and his always against the grain, non corporate influenced response. So many others have praised Haiden for bringing back the real gladiator bloodsport that SX is, is it that though?? It’s a race to the finish. different than a lot of other racing its very unpredictable takeout moves happen, occasionally as an accident, often taking both riders down at the same time.

Also different than MANY sports we have a huge industry of weekend warriors, Some of which make it into the night show of the biggest race on TV. You see we only bring 20 riders from each class, the top half to top quarter are “Factory” riders, the rest are privateers, sometimes these privateers are completely their on their own.

There is also something called a last chance qualifier. This is your last chance to get into the night show, often fought by these privateers out of box vans with their brother as a mechanic. Often that last spot is filled by sketchy on edge riding and takeout moves from a rider that looks like he is barely making it, but would run circles around anyone at your local race.

...

I’m impressed by fastapi-radar from doganarif.

A powerful debugging dashboard for FastAPI applications. Monitor HTTP requests, SQL queries, and exceptions in real-time with a beautiful React UI. One-line integration, zero configuration needed.