ls: honor LC_TIME for --time-style=locale by sylvestre · Pull Request #11669 · uutils/coreutils

sylvestre · 2026-04-05T21:39:07Z

on my system

$ export L="en_US fr_FR ru_RU.KOI8-R fa_IR.UTF-8 am_ET.UTF-8 th_TH.UTF-8 zh_CN.GB18030"; for f in $L; do echo "$f:"; LANG=$f cargo run -q --features unix -- ls -al --time-style=locale Cargo.toml LICENSE; done
en_US:
-rw-rw-r--+  1 sylvestre sylvestre 27798 Apr  5 22:05 Cargo.toml
-rw-r--r--+  1 sylvestre sylvestre  1056 Mar 20  2025 LICENSE
fr_FR:
-rw-rw-r--+  1 sylvestre sylvestre 27798 avr  5 22:05 Cargo.toml
-rw-r--r--+  1 sylvestre sylvestre  1056 mars 20  2025 LICENSE
ru_RU.KOI8-R:
-rw-rw-r--+  1 sylvestre sylvestre 27798 апр  5 22:05 Cargo.toml
-rw-r--r--+  1 sylvestre sylvestre  1056 март 20  2025 LICENSE
fa_IR.UTF-8:
-rw-rw-r--+  1 sylvestre sylvestre 27798 فروردین 16 22:05 Cargo.toml
-rw-r--r--+  1 sylvestre sylvestre  1056 اسفند 30  1403 LICENSE
am_ET.UTF-8:
-rw-rw-r--+  1 sylvestre sylvestre 27798 ኤፕሪ 27 22:05 Cargo.toml
-rw-r--r--+  1 sylvestre sylvestre  1056 ማርች 11  2017 LICENSE
th_TH.UTF-8:
-rw-rw-r--+  1 sylvestre sylvestre 27798 เม.ย  5 22:05 Cargo.toml
-rw-r--r--+  1 sylvestre sylvestre  1056 มี.ค 20  2568 LICENSE
zh_CN.GB18030:
-rw-rw-r--+  1 sylvestre sylvestre 27798 4月  5 22:05 Cargo.toml
-rw-r--r--+  1 sylvestre sylvestre  1056 3月 20  2025 LICENSE

github-actions · 2026-04-05T22:02:10Z

GNU testsuite comparison:

GNU test failed: tests/ls/abmon-align. tests/ls/abmon-align is passing on 'main'. Maybe you have to rebase?
Congrats! The gnu test tests/seq/seq-epipe is now passing!

sylvestre · 2026-04-05T22:05:04Z

GNU test failed: tests/ls/abmon-align. tests/ls/abmon-align is passing on 'main'. Maybe you have to rebase?

looking at this

github-actions · 2026-04-06T08:08:23Z

GNU testsuite comparison:

GNU test failed: tests/date/date-thailand. tests/date/date-thailand is passing on 'main'. Maybe you have to rebase?
Skip an intermittent issue tests/date/date-locale-hour (fails in this run but passes in the 'main' branch)
Skipping an intermittent issue tests/pr/bounded-memory (passes in this run but fails in the 'main' branch)
Note: The gnu test tests/basenc/bounded-memory is now being skipped but was previously passing.
Note: The gnu test tests/expand/bounded-memory is now being skipped but was previously passing.
Congrats! The gnu test tests/cut/cut-huge-range is now passing!

github-actions · 2026-04-06T08:24:34Z

GNU testsuite comparison:

GNU test failed: tests/date/date-thailand. tests/date/date-thailand is passing on 'main'. Maybe you have to rebase?
Skipping an intermittent issue tests/pr/bounded-memory (passes in this run but fails in the 'main' branch)
Skipping an intermittent issue tests/tail/follow-name (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/cut/cut-huge-range is now passing!

github-actions · 2026-04-06T09:26:55Z

GNU testsuite comparison:

Skipping an intermittent issue tests/tty/tty-eof (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/pr/bounded-memory is now passing!
Congrats! The gnu test tests/seq/seq-epipe is now passing!

github-actions · 2026-04-06T09:55:32Z

GNU testsuite comparison:

Skipping an intermittent issue tests/tail/follow-name (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/pr/bounded-memory is now passing!

LevitatingBusinessMan · 2026-04-06T11:51:38Z

You picked this issue up fast :D

github-actions · 2026-04-06T14:21:21Z

GNU testsuite comparison:

Skipping an intermittent issue tests/date/date-locale-hour (passes in this run but fails in the 'main' branch)
Skipping an intermittent issue tests/tty/tty-eof (passes in this run but fails in the 'main' branch)

sylvestre · 2026-04-06T14:37:15Z

@cakebaker
spoiler! After this one, i have --quoting-style=locale/clocale built on top :)

tests/by-util/test_ls.rs

cakebaker · 2026-04-07T15:29:57Z

tests/by-util/test_ls.rs

+    /// Substring that MUST appear in the (lossy UTF-8) stdout. Used for the
+    /// C-locale sanity case to pin the English output.
+    must_contain: Option<&'static str>,
+    /// Substring that must NOT appear in the (lossy UTF-8) stdout. Used to
+    /// guarantee non-C locales don't fall back to the English format.
+    must_not_contain: Option<&'static str>,


I'm not sure about those properties. You only introduce them because of the C-locale. It might make sense to separate the handling of the C-locale from the handling of the special cases, instead of using a "one size fits all" approach.

The C locale is already handled separately (lines 2500-2522) with its own dedicated block and C_LOCALE_OUTPUT assertion. The must_contain/must_not_contain properties you were referring to have been removed — the remaining struct fields (script_range, require_high_byte, allow_gregorian_year) are all about differentiating between non-C locales (e.g. KOI8-R vs Persian vs Thai). I think this is addressed now?

tests/by-util/test_ls.rs

cakebaker · 2026-04-07T17:09:59Z

I will finish the review of test_ls.rs and the two uucore files (hopefully) tomorrow.

tests/by-util/test_ls.rs

src/uucore/src/lib/features/i18n/datetime.rs

github-actions · 2026-04-08T10:05:53Z

GNU testsuite comparison:

Skipping an intermittent issue tests/date/date-locale-hour (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/printf/printf-surprise is now passing!

cakebaker · 2026-04-08T13:30:36Z

src/uucore/src/lib/features/i18n/datetime.rs

+struct CachedLocaleNames {
+    /// `%B` — full month names, raw
+    month_long: Option<[String; 12]>,
+    /// `%B` — full month names, padded to uniform display width
+    month_long_padded: Option<[String; 12]>,
+    /// `%b` / `%h` — abbreviated month names (trailing dots stripped), raw
+    month_abbrev: Option<[String; 12]>,
+    /// `%b` / `%h` — abbreviated month names, padded
+    month_abbrev_padded: Option<[String; 12]>,
+    /// `%A` — full weekday names, raw
+    weekday_long: Option<[String; 7]>,
+    /// `%A` — full weekday names, padded
+    weekday_long_padded: Option<[String; 7]>,
+    /// `%a` — abbreviated weekday names, raw
+    weekday_short: Option<[String; 7]>,
+    /// `%a` — abbreviated weekday names, padded
+    weekday_short_padded: Option<[String; 7]>,
+}


This struct doesn't feel right. If I understand it correctly, only half of the properties are used, the others are set but never read.

I would remove the "padded" properties so that the remaining properties contain either the padded or raw data, depending on how the struct was filled.

Both sets are actually read — localize_format_string() takes a NamePadding parameter and selects between _padded and raw variants at each substitution point (see lines 250-290). date calls with Raw, ls calls with Padded, so all 8 fields get used depending on the caller.

That said, I could avoid storing both by keeping only the raw names and applying pad_names() on-the-fly when Padded is requested. The padding cost is negligible (12 or 7 string scans). Would you prefer that approach?

cakebaker · 2026-04-08T13:38:06Z

Hm, it looks like one of my previous suggestions doesn't work on OpenBSD as test_ls_abmon_align fails :|

Add format_system_time_locale_aware in uucore::time that runs the format string through i18n::datetime::localize_format_string when a non-C LC_TIME locale is active, and route ls's display_date through it. Fixes month names and alternate-calendar years (Persian, Buddhist, Ethiopian) matching GNU. Also generalize uutests' is_locale_available to derive the expected charmap from the locale suffix so non-UTF-8 locales work, and consolidate the ls locale tests into a single data-driven case.

The ls-specific column alignment padding (from abmon-align) was leaking into date output. Add a `pad` parameter to `localize_format_string` so ls gets padded names and date gets raw names.

github-actions · 2026-04-09T09:55:35Z

GNU testsuite comparison:

Skip an intermittent issue tests/cut/bounded-memory (fails in this run but passes in the 'main' branch)
Skip an intermittent issue tests/date/date-locale-hour (fails in this run but passes in the 'main' branch)
Skip an intermittent issue tests/tty/tty-eof (fails in this run but passes in the 'main' branch)
Congrats! The gnu test tests/cp/link-heap is now passing!
Congrats! The gnu test tests/dd/no-allocate is now passing!
Congrats! The gnu test tests/seq/seq-epipe is now passing!

sylvestre requested a review from cakebaker April 5, 2026 22:02

sylvestre force-pushed the time-style_locale branch from e28448c to c302650 Compare April 6, 2026 08:10

sylvestre force-pushed the time-style_locale branch from c302650 to 7856aaf Compare April 6, 2026 09:12

sylvestre force-pushed the time-style_locale branch from 7856aaf to 0a9df65 Compare April 6, 2026 09:38

sylvestre mentioned this pull request Apr 6, 2026

ls: Compatibility Tracking Issue #1872

Open

57 tasks

sylvestre marked this pull request as ready for review April 6, 2026 11:39

sylvestre force-pushed the time-style_locale branch from 0a9df65 to 893c2ae Compare April 6, 2026 13:48

cakebaker reviewed Apr 7, 2026

View reviewed changes

tests/by-util/test_ls.rs Outdated Show resolved Hide resolved

cakebaker reviewed Apr 7, 2026

View reviewed changes

tests/by-util/test_ls.rs Outdated Show resolved Hide resolved

cakebaker reviewed Apr 8, 2026

View reviewed changes

tests/by-util/test_ls.rs Outdated Show resolved Hide resolved

cakebaker reviewed Apr 8, 2026

View reviewed changes

tests/by-util/test_ls.rs Outdated Show resolved Hide resolved

cakebaker reviewed Apr 8, 2026

View reviewed changes

src/uucore/src/lib/features/i18n/datetime.rs Show resolved Hide resolved

sylvestre force-pushed the time-style_locale branch from 893c2ae to 36be3f5 Compare April 8, 2026 09:47

cakebaker reviewed Apr 8, 2026

View reviewed changes

sylvestre added 3 commits April 9, 2026 10:58

ls: add tests for --time-style=locale across locales

811a755

ls: pad locale month/weekday names to uniform width (GNU abmon-align)

ea7c339

date: don't pad locale month/weekday names with trailing spaces

4486cc1

The ls-specific column alignment padding (from abmon-align) was leaking into date output. Add a `pad` parameter to `localize_format_string` so ls gets padded names and date gets raw names.

sylvestre force-pushed the time-style_locale branch from 36be3f5 to 4486cc1 Compare April 9, 2026 09:42

Uh oh!

Conversation

sylvestre commented Apr 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented Apr 5, 2026

Uh oh!

sylvestre commented Apr 5, 2026

Uh oh!

github-actions bot commented Apr 6, 2026

Uh oh!

github-actions bot commented Apr 6, 2026

Uh oh!

github-actions bot commented Apr 6, 2026

Uh oh!

github-actions bot commented Apr 6, 2026

Uh oh!

LevitatingBusinessMan commented Apr 6, 2026

Uh oh!

github-actions bot commented Apr 6, 2026

Uh oh!

sylvestre commented Apr 6, 2026

Uh oh!

Uh oh!

cakebaker Apr 7, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

sylvestre Apr 9, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

cakebaker commented Apr 7, 2026

Uh oh!

Uh oh!

Uh oh!

Uh oh!

github-actions bot commented Apr 8, 2026

Uh oh!

cakebaker Apr 8, 2026

Choose a reason for hiding this comment

Uh oh!

sylvestre Apr 9, 2026

Choose a reason for hiding this comment

Uh oh!

cakebaker commented Apr 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented Apr 9, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

sylvestre commented Apr 5, 2026 •

edited

Loading

cakebaker Apr 7, 2026 •

edited

Loading

cakebaker commented Apr 8, 2026 •

edited

Loading