Eh, I’d argue that Java and C# are in the niche of having few features. While I don’t like this niche, Java having even less features makes it stand out more in this niche. If you’re looking for a language with more features than that, then there’s so many more feature-rich choices than C# that I just don’t feel like you’d choose C#, unless you want Java with integration into the Microsoft ecosystem.

Hmm, Fedora should have the newest version of KDE. Yeah, hard to say why it’s not working then…

Hmm, which distro did you try it with? I believe, KDE should support auto-rotating the screen in the newer versions¹. If you tried it with Kubuntu LTS, for example, that would’ve still been an older version, which does not use Wayland by default.

¹) More precisely: it should support it when it’s being run under Wayland, which is the default since KDE Plasma 6.

¯\_(ツ)_/¯

Oh yeah, we only want to pre-configure it with a static IP address on its Ethernet port, so that we can SSH into it in a controlled manner and then we intend to do the rest with a deployment tool.

Yeah, that is a very good question. It’s one of the last commands in the script and initially I thought they had set up the script so that it would abort, if any of the commands before it would fail.

But then a colleague pointed out that it’s actually the opposite. So, you can tell the shell to abort execution on error by running set -e. But what they had written at the top was set +e, which explicitly turns that off (even though it should be off by default).
The last command in the script is also exit 0, so it always indicates success.
So, yeah, they seem to have knowingly made it so that if the script fails, then it doesn’t retry or anything. It tries to plough through as many of the commands as it can manage (ignoring any that fail on the way) and then it always deletes itself.

I guess, it’s not as egregious of an assumption, because it only runs on a fresh OS. That’s a pretty controlled environment to be executing in, so the chance of something going wrong is rather low.
Well, and the other question is what else would you do? If the script fails and you don’t delete it, it’s going to re-run on the next boot. What’s going to be different on the next boot to make it succeed then? Might as well do as much as you can and then quit…

renaming symbols, presenting documentation, formatting files,

Yes, these are supported via the Language Server Protocol (LSP). I’ve mostly been using it with the Rust LSP server (rust-analyzer) and well, it typically works, but sometimes you have to tell it to restart the LSP server and stuff (which isn’t a huge ordeal, but don’t expect everything to always work as well as in a full-fledged IDE).
I believe, for formatting, there’s also some non-LSP support.

showing code diagnostics beyond syntax errors (for example code smells or so),

This is supported in principle via LSP, too, but it depends on the specific LSP server, how much info it provides. The Rust compiler gives out relatively much on its own, which is passed on by the LSP server, but you can apparently also configure it to use the linter on save.

have AI integration (explain this, rewrite this, replace this with prompt output, …),

Not out of the box. There’s a way to define “External Tools”, which basically allows you to run commands and pass arguments to them and then use their output. For example, you should be able to define an External Tool, where you can select some text, then press your keyboard shortcut for that tool, so it sends the selected text to that tool and then it takes the command output and inserts it instead of the selected text.
While this is a powerful concept, I don’t know, if you hit limitations at some point.

specific framework integrations (reactjs, django, actix, …),

Nope, except where this might be covered by LSP. But there’s no obvious way to just install additional plugins, for example. You get about thirty built-in plugins and that’s it.

and stuff like expanding macros in C/C++ and Rust?

Well, expanding macros is also possible with the Rust LSP server. Don’t know about other languages.

We’ve been mucking around with how to pre-configure Raspberry Pis after flashing at $DAYJOB and basically, the way the Raspberry Pi Imager works is that it writes a firstrun.sh onto the SD card which gets run during first boot.

How does it know to not run that script from the second boot onwards? Well, one of the lines in that script is:

rm -f firstrun.sh

…it deletes itself while it’s running. 🙃

openSUSE has the best integration of KDE, but I wouldn’t expect to see issues like yours on any distro, really…

However, Codeberg seems to push for excluding proprietary software dependencies, which might limit the kinds of projects I can do.

Well, they ask for software projects to be themselves under a FOSS license. If you can isolate the proprietary dependencies, you could still open-source the code.

Of course, you could also consider using Codeberg for all your open-source projects and then self-host a Git server for your proprietary projects.

There’s also an official guide for this: https://git-scm.com/book/en/v2/Git-on-the-Server-The-Protocols

Yeah, good point. It also doesn’t update when the content of a file changes. So, in order to detect a change in a directory, you have to walk all the files and sub-directories and the directory itself to get the last-modified timestamp for each of them. Then determine the highest last-modified and compare it to what you measured in a previous run. If they differ, a change happened.

Yeah, I’m building more-or-less an alternative to make. Major difference is that I’m not using shell commands, but rather users will define their build code in Rust …because it’s intended to be a build tool for Rust applications (beyond what cargo does).

Thanks for the comment, though. So far, I haven’t limited inputs to just be files, so I don’t actually assume to have a last-modified timestamp. Rather, my assumption is that I can get some value which changes when the input changes. In the case of a file, that’s the last-modified timestamp, but theoretically, it could also be a hash. But that means I have to store these values to be able to detect a change. Being able to just say that one thing is newer than the other without storing anything, that is pretty cool and might be worth changing my assumption for.

I don’t think, inotify works for me, because I don’t have a continuously running process. My users rather just run some build command and then I go and check, if any input files changed since the last run.

openSUSE is now one big btrfs partition. They dropped the separate xfs for the home partition maybe two years ago or so. It makes it less likely to run into a situation where the snapshots fill up the root partition (which is really ugly to recover from, because users will try to uninstall packages, which doesn’t help, since the files are still contained in a previous snapshot)…

To be more precise, you don’t really want to use the snapshotting in the home-directory. You can still use btrfs itself and for example, openSUSE sets it up so the home-directory is in a btrfs subvolume that’s excluded from snapshots.
At the very least, you’d want the snapshots in the home-directory to be independent from the rest of the OS, so that you don’t end up rolling back what you’ve worked on when you want to roll back a faulty OS update.

Well, and you also just want proper backups of your home-directory, so the snapshots are not as useful…

Frankly, I would be surprised, if anything uses groff for displaying --help, unless it shows the man page for that.
The most basic implementation of --help is a manually formatted multi-line string written into the source code, which gets printed as-is.
For dynamic layouting, you do need more logic, but rendering it to groff source code first does not make that easier. For tabbing, you print an appropriate number of \t.

It does show those for brightness for me, too:

Thanks, I was looking for a solution for that, too. 🙂

openSUSE, because of the snapshotting. It’s zero-setup and just gives peace of mind when doing upgrades, as I can roll back even from the bootloader.

I am one of the rare few users who actually enjoys PowerShell; I prefer piping typed, structured data over piping streams of bytes. I also really hate sh/zsh/bash syntax.

You might enjoy Nushell.