Skip to content

merge main#2

Merged
FidelSch merged 109 commits intoFidelSch:mainfrom
uutils:main
Jan 28, 2026
Merged

merge main#2
FidelSch merged 109 commits intoFidelSch:mainfrom
uutils:main

Conversation

@FidelSch
Copy link
Copy Markdown
Owner

No description provided.

sylvestre and others added 30 commits January 18, 2026 21:12
Fixed two bugs in the fuzzing workflow where matrix.test-target.name.should_pass
was used instead of matrix.test-target.should_pass. This caused continue-on-error
to always be true, making CI jobs pass even when fuzzers correctly detected
incompatibilities.
chore(deps): update rust crate divan to v4.3.0
Instead of adding this code to every utility, hoping to make it centralized and then addressing all of the special cases in the individual utilities. There's definitely a chance that a special case is missed and could be a regression, but this should solve a much larger number of issues for all utilities.

Fixes #10325
Fixes #10260
Fixes #10230
Fixes #10214
Fixes #9936
Fixes #8919
Fixes #7252
Fixes #4965
deps: remove unused rust deps
chore(deps): update rust crate proc-macro2 to v1.0.106
* freebsd.yml: Drop useless cache actions

* frebsd.yml: Drop sccache
Reduce duplicated management of supported utils
…sing-for-codspeed-rust-causing-workflow

fix(ci): remove outdated `analysis` mode
* date: handle parentheses as comments like GNU date

* simplify the code

Co-authored-by: Daniel Hofstetter <daniel.hofstetter@42dh.com>

---------

Co-authored-by: Daniel Hofstetter <daniel.hofstetter@42dh.com>
Caused:
  zsh: suspended (tty input)  cargo test
fix(CI): clippy unnecessary unwrap
The test is using a predictable path which implies that running
the test multiple times fails
In order to show the total row we are summing intermediate values as they
are and eventually scaling them.

Now, while this provides a valid output, it also implies that the total value
that we show is not matching the sum of the previously listed values, since
the sum of scaled values may different from the sum of the original values that
gets eventually scaled.

To be consistent with the output:
 - Use a new BytesCell struct to track the bytes values
 - Keep a scaled value tracked
 - Show the sum of the scaled values as total

Closes: #10436
It adds extra (tiny, but still...) computation for no reason
---------

Co-authored-by: Sylvestre Ledru <sylvestre@debian.org>
fix(deps): remove duplicate itertools package
test-mv: Use temporary directory in /dev/shm
oech3 and others added 29 commits January 27, 2026 16:26
Now that #7252 is fixed this workaround is not needed any more:

This is a partial revert of fac6c29 ("util/build-gnu.sh: Bypass timeout/yes SIGPIPE handling bug")
This is a revert of 5004d4b ("build-gnu: replace `timeout` for `tests/tail/follow-stdin.sh`")

Signed-off-by: Etienne Cordonnier <ecordonnier@snap.com>
build-gnu.sh: remove workaround for timeout/yes
GnuTests: Drop cache action outside of VM, use preinstalled rust
Instead of just "Uncategorized error" output:
"base64: read error: Input/output error"
* fix: allow duplicate -o options for same output file in sort

Modify sort command to permit multiple -o/--output flags specifying the same file,
matching GNU sort behavior. Previously, any multiple -o flags would error, but now
only differing output files are rejected. Added test case for duplicate outputs.

* fix(sort): fix case-insensitive sorting to order punctuation after letters

Changed ascii_case_insensitive_cmp to fold characters to uppercase instead of lowercase, ensuring that in ASCII case-insensitive sorting, letters are ordered before punctuation. Added tests to verify the correct behavior.
chore(deps): update rust crate clap to v4.5.55
…st-reverse

cksum/hashsum: Merge common logic under a common crate
@FidelSch FidelSch merged commit 1569267 into FidelSch:main Jan 28, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.