Iván Ezequiel Rodriguez <[email protected]> writes:
> The chmod "preserve ctime on no-op" idea conflicts with
> tests/chmod/only-op.sh, which requires always calling chmod(2)
> so operations like 'chmod + /' still fail for non-root.
True. That test is relatively new. I added it a few months ago, but
didn't see that TODO note [1]. Pádraig's note about other side effects,
such as removing ACLs and TOCTOU issues, is probably a better rationale,
though [2][3].
Not saying your commit message is wrong or needs changing, just sharing
those links for your reference.
> -printf:
> - Now that gnulib supports *printf("%a"), import one of the
> - *printf-posix modules so that printf(1) will support %a even on
> - platforms where the native *printf(3) is deficient.
> - Suggestion from Eric Blake.
> -
This one looks good as well.
> -ls: add --format=FORMAT option that controls how each line is printed.
> +ls: custom per-entry output template (not --format=WORD, which already
> + selects across/commas/long/…). Would need a new option name
> + (e.g. --printf) and a design discussion; do not overload --format.
Makes sense. Not really related to your change, but I can't tell whether
I have an unreasonable dislike of the Unicode HORIZONTAL ELLIPSIS
character or my preferred font just renders it too squished.
I think I'll push your patch in a bit, but simplify that part to just
say:
ls: add --printf=FORMAT option that controls how each line is
printed.
It is a smaller change, and I can't see a more reasonable name for that
option. E.g., we already use --printf in 'stat'.
Collin
[1] https://lists.gnu.org/archive/html/coreutils/2026-02/msg00080.html
[2] https://lists.gnu.org/archive/html/coreutils/2026-02/msg00082.html
[3] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=497514
