Rust Coreutils 0.4 Boosts GNU Compatibility to 85%

https://linuxiac.com/rust-coreutils-0-4-boosts-gnu-compatibility-to-85-percentage/

The Rust Coreutils project, which aims to provide a full, modern Rust implementation of the GNU Core Utilities — the essential command-line tools found on every Linux and Unix-like operating system — has announced the release of version 0.4.

Notably, the project’s growing maturity has already led to real-world adoption in some Linux distros, such as Ubuntu 25.10 “Questing Quokka” and AerynOS, both of which now utilize Rust Coreutils for select system utilities.

Version 0.4 brings this release a step closer to achieving full GNU Coreutils compatibility. According to devs, the latest test results show 544 passing tests, up from 532 in the previous 0.3 release — an increase that raises total compatibility to 85.8%, while failures dropped from 68 to 56.

71 points · 16 comments · view on lemmy.world

16 Comments

trevor@lemmy.blahaj.zone · 23 pts · 279d (15 replies)

The detractors of this project portray it like it's a far-off pipe dream to be a drop-in replacement for GNUtils. Meanwhile, it's still a relatively young project that already has 85% compatibility. I think we can do it. Lol.

HappyFrog@lemmy.blahaj.zone · 21 pts · 279d (6 replies)

My only issue is the permissive license, but I'm still hope they do well :3

qweertz@programming.dev · 11 pts · 278d (3 replies)

This is my take as well. I'm extremely disappointed they only went with a temporarily open license instead of a proper one, but using MIT is unfortunately to be expected from the Rust ecosystem for whatever reason...

morto@piefed.social · 5 pts · 278d (2 replies)

temporarily open?

qweertz@programming.dev · 1 pts · 276d (1 reply)

MIT is an extremely weak license when it comes to defending free/libre rights; e.g. it allows proprietary forks. i.e. companies stealing the code, making their own bullshit corpo product and not even releasing the source code back

morto@piefed.social · 1 pts · 275d

I understand and share the dislike, but the openly released version will remain free, and no one can change it, so don't you think temporarily open is a bit misleading?

trevor@lemmy.blahaj.zone · 0 pts · 279d (1 reply)

Same.

scroll_responsibly@lemmy.sdf.org · -2 pts · 278d

Same.

qweertz@programming.dev · 14 pts · 278d (4 replies)

Still completely unhinged to ship it in your distro before it's fully compatible cough Ubuntu cough

mesamunefire@piefed.social · 8 pts · 278d

Yep already broke a couple of things and we had to roll back.

trevor@lemmy.blahaj.zone · 2 pts · 278d (2 replies)

Wrong. Have to start adoption somewhere, and doing it in a non-LTS release is a great move.

qweertz@programming.dev · 1 pts · 276d (1 reply)

Last I checked Ubuntu was not an unstable mess of a rolling release, but a distro people rely on for stability.

Their normal non-LTS versions are still considered production ready and acting that rash has only solidified my negative opinion of them more...

trevor@lemmy.blahaj.zone · 0 pts · 276d

If you're expecting stability for any Ubuntu release, that went out the window when Canonical started forcing Snaps.

But non-LTS Ubuntu releases have always been a testing ground for less-than-stable changes. uutils is just one of them, and the only way to make them stable is to see how they're being used in the wild.

bluGill@fedia.io · 11 pts · 278d

I would expect the last 10% to take 90% of the work though. There are a lot of rough edges that just work weird. There is also a question of what is useful to get from that last 10%, or what should be done different despite being incompatible. (BSD utils are also an option to be compatible with instead)

patrick@lemmy.bestiver.se · -2 pts · 278d (1 reply)

Rust coreutils has 17,000 commits and is 12 years old.

spartanatreyu@programming.dev · 5 pts · 278d

Yeah, but it didn't get any serious development until 2021.

Check for yourself: https://github.com/uutils/coreutils/graphs/contributors