Various tracking tools for coreutils

Tracking the evolution of https://github.com/uutils/coreutils

GNU testsuite comparison

Below is the evolution of how many GNU tests uutils passes. A more detailed breakdown of the GNU test results of the main branch can be found in the user manual.

Refreshed twice a day by github actions. Changes are documented in the json file (gnu-result.json).

Compares only the Linux execution.

The percentages leave out the tests that can never pass for structural reasons - they intercept glibc internals with LD_PRELOAD (which never fires, since Rust's std issues different syscalls), set gdb breakpoints inside GNU's own C sources, or only run on GNU/Hurd. That list is maintained in the coreutils repo, in util/gnu-unfixable-tests.txt, with a reason per entry; tests we could actually fix stay counted on purpose. See uutils/coreutils#13841.

Based on:

Busybox testsuite comparison

Similar results but using the busybox testsuite: https://github.com/mirror/busybox/tree/master/testsuite

Toybox testsuite comparison

Similar results but using the toybox testsuite: https://github.com/landley/toybox/tree/master/tests

Binary size evolution

Refreshed once a day by github actions.

Compares only the Linux execution.

`unsafe` usage evolution

Tracks how much unsafe Rust the project relies on. The total counts unsafe { … } blocks, unsafe fn/impl/trait/extern items and #[unsafe(...)] attributes (Rust 2024) across all .rs files outside vendor/ and target/.

Data lives in unsafe-result.json. Counting logic is in unsafe_count.py

Development activity

Monthly commits and contributors for both GNU coreutils and uutils coreutils since 2021. Non-merge commits only, authors deduplicated via .mailmap and bots (dependabot, renovate, github-actions, …) excluded. The cumulative contributor panel counts every contributor since each project's first commit, not just those active since 2021.

Regenerated from both git histories once a day by github actions (activity-result.json).