nix VS cue

Thinking about it like rpm/deb is where you're getting hung up. Think of it more along the lines of npm, but for your OS (or just your profile in the case of nixpkgs+home-manager).

At the most trivial level, you can set up some packages to install [1][2]. You'll generally come out ahead of classical package managers if you do that and nothing else.

Nix also acknowledges the configuration issue. Think of this like copying some config files using a dockerfile. You can either use literals[3], or use the nix language to generate the config [4] (provided that someone has created the required projection from nix to config).

What this ends up becoming is a single git repo with your entire system setup. My repo gets a bit fancier: I have my home desktop (currently Nixos, but it distrohops a lot), my personal laptop running Ubuntu+intune+nixpkgs for work, and then my work Mac machine (which I am aiming to get rid of). The single repo contains nix configs for all 3, and shares config where appropriate.

I really need to write a blog post about Nix in 2023. The main issue with getting started right now is that the installer requires some convincing to use flakes (not to mention that flakes are disabled by default), and you really should be using flakes.

[1]: https://gitlab.com/jcdickinson/nix/-/blob/main/system/jono-d...

Using these with home-manager is also really simple. You simply enable nix-direnv[0], and then `use flake` in your envrc[1]. Finally, set up your flake.nix with a dev shell[2] (I'm definitely going to take a look at numtide's devshell).

You'll also need flakes and the nix command enabled first [4] (add that line to `/etc/nix/nix.conf` if you aren't using nixos).

Why use flakes? Mostly because it has a lockfile: there's a really good chance that "works on my machine" is "works on my team's machines." Flakes are also much cleaner than vanilla Nix.

[0]: https://gitlab.com/jcdickinson/nix/-/blob/main/home/general.... [1]: https://gitlab.com/jcdickinson/nix/-/blob/main/.envrc#L3 [2]: https://gitlab.com/jcdickinson/nix/-/blob/main/flake.nix#L65 [4]: https://gitlab.com/jcdickinson/nix/-/blob/70844981d5cd63c839...

The nice thing with flake-based NixOS is that it's trivial to cherry-pick unstable onto a stable base. I do a bunch of that in my nixconfigs: https://gitlab.com/jcdickinson/nix

If you are in a situation where you have a backend and you want to expose an API and then you would eventually want a client, you would need format specs as the starting point where server and clients are generated from that one source.

At the moment, OpenAPI with YAML is the only way to go but you can't easily split the spec into separate files as you would do any program with packages, modules and what not.

There are third party tools[0] which are archived and the libraries they depend upon are up for adoption.

In that space, either you can use something like cue language 1] or something like TypeSpec which is purpose built for this so yet, this seems like a great tool although I have not tried it yet myself.

[0]. https://github.com/APIDevTools/swagger-cli

[1]. https://cuelang.org/

EDIT: formating

Where `kube.cue` sets reasonable defaults (e.g. image is /). The "cluster" runs on a mini PC in my basement, and I have a small Digital Ocean VM with a static IP acting as an ingress (networking via Tailscale). Backups to cloud storage with restic, alerting/monitoring with Prometheus/Grafana, Caddy/Tailscale for local ingress.

[1] https://www.talos.dev/

[2] https://cuelang.org/

I've been somewhat surprised that CUE bills itself as "tooling friendly" and doesn't yet have a language server- the number one bit of tooling most devs use for a particular language.

I'm assuming it's becaus CUE is still unstable?

Anyway, if others are interested in CUE's LSP work, I think https://github.com/cue-lang/cue/issues/142 is the issue to subscribe to

This is where I usually pitch in with "Have your heard of CUELang, our lord and savior?": https://cuelang.org/

- Not turing complete

CUE: The core problem CUE solves is "type checking", which is mainly used in configuration constraint verification scenarios and simple cloud native configuration scenarios.

If you really want executable configurations please consider a newer language like https://dascript.org or https://cuelang.org which provide better type safety.

1- https://news.ycombinator.com/item?id=38030778

Markdown and XML are nice, but what about more advanced documentation formats like OpenAPI? For one recent project, I set up automatic generation of the OpenAPI docs from (much more compact and flexible) CUE definitions (https://cuelang.org/) - which has the bonus of also being able to test the API against the definitions. JetBrains has a CUE plugin, but it's really barebones (doesn't even support jumping from the usage of a schema to its definition). Of course the possibilities when generating docs are endless (just think of the various syntaxes for doc comments, embedding examples/tests in source code etc.)...

It doesn't include validators for TOML and INI, but if you're doing JSON and YAML, I would take a look at using or building upon CUE (https://cuelang.org/). It is a different take on schema definition (plus more), and is surprising terse and powerful model.