Commit Graph
31776 Commits
Author SHA1 Message Date
Collin Funk b41cbfd6cc tests: readlink: check the output with multiple files
* tests/readlink/multi.sh: Add pwd to the print_ver_ call. Check the
output of the existing tests.
2026-09-23 19:42:11 -07:00
Kamila Szewczyk d8d99b5cd8 dircolors: register the .bz3 compression format
* src/dircolors.hin: Register .bz3 and .tbz3 formats.
2026-09-23 13:53:41 +01:00
Pádraig Brady 965657d686 dircolors: remove obsolete compression formats
* src/dircolors.hin: Remove .ace, .alz, .apk, .arc, .arj, .bz, .zoo.
2026-09-23 13:45:53 +01:00
Sylvestre Ledru 9b37456b93 tests: date: check --reference with a non-UTF-8 file name
Nothing covered passing --reference a file name that is not valid in the
current locale, which is fine since the name is passed through to stat
as-is.
Identified here:  https://github.com/uutils/coreutils/pull/14558

* tests/date/reference.sh: Add a test case, skipped when the file system
rejects such a name.

Link: https://github.com/coreutils/coreutils/pull/357
2026-09-22 14:36:10 +01:00
Sylvestre Ledru 1d1f670e5a tests: fmt: cover the width derived from --goal
When only --goal is given, fmt sets the maximum width to the goal plus
ten (src/fmt.c), but nothing exercised that.
identified by https://github.com/uutils/coreutils/pull/14275

* tests/fmt/goal-option.sh: Check that "-g G" formats as "-g G -w G+10"
does, that the derived width is exactly G+10, that a goal of zero is
accepted, and that without -w the goal is still limited to 75.

Link: https://github.com/coreutils/coreutils/pull/358
2026-09-22 14:26:05 +01:00
Sylvestre Ledru 885849853d tests: echo: check that \u, \U and \" are not escapes
* tests/misc/echo.sh: Ensure 'echo -e' outputs \", \u and \U verbatim,
with their backslash, as they are not escape sequences for echo
(unlike printf).  Also cover truncated, non-hexadecimal, surrogate and
out-of-range payloads after \u and \U.
  Reported https://github.com/uutils/coreutils/issues/14414

Link: https://github.com/coreutils/coreutils/pull/346
2026-09-22 12:03:30 +01:00
Pádraig Brady 6bfc90019f tests: od: avoid large mem allocation due to MALLOC_PERTURB_
$ testmem() {
    env time -f %M \
    make TESTS="$*"RUN_VERY_EXPENSIVE_TESTS=yes SUBDIRS=. check
   }

 # Before change
 $ testmem tests/od/big-w.sh
 2965404 # 3GB

 # After change
 $ testmem tests/od/big-w.sh
 14136

* tests/od/big-w.sh: Disable MALLOC_PERTURB_.
Also remove the redundant setting of LC_ALL.
2026-09-21 15:06:48 +01:00
Pádraig Brady c445c7580c tests: od: halve the invocations of od
* tests/od/big-w.sh: Avoid rerunning the expensive
od invocations just to count bytes, instead using
tee to count the large output in parallel.
2026-09-21 14:59:41 +01:00
Pádraig Brady 98e9bed688 tests: od: halve the memory requirement by using -v
* tests/od/big-w.sh: Add the -v option to use half the mem.
2026-09-21 14:59:41 +01:00
Pádraig Brady c12f1296e0 od: halve the memory used when not suppressing duplicates
* src/od.c (dump): Only use a single memory block
when there is no need to compare against the previous block.
2026-09-21 14:59:36 +01:00
Pádraig Brady cecd945aa9 tests: cp: avoid confusing partial failure
* tests/cp/sparse-perf.sh: On file systems without SEEK_HOLE support,
this test would confusingly partially fail before being skipped.
2026-09-20 22:07:35 +01:00
Collin Funk 9977a36277 tests: cp: avoid a false failure in SELinux disabled containers as root
* tests/cp/capability.sh: Add a getlimits_ invocation. Allow 'cp' to
fail with ENOTSUP, which occurs when trying to copy the security.selinux
extended attribute.
2026-09-19 21:53:31 -07:00
Pádraig Brady f76b1d7ca5 maint: move tests/misc/xattr.h to tests/cp/
* tests/misc/xattr.sh: Rename to ...
* tests/cp/xattr.sh: ... here.
* tests/local.mk: Update the reference.
2026-09-18 17:53:00 +01:00
Pádraig Brady 182225b05a tests: avoid false failures due to xattr ENOTSUP warnings
E.g. with podman we get default SELinux context mount options
for /dev/shm and /, resulting in all files getting context like
the following when created:
  security.selinux="system_u:object_r:container_file_t:s0:c458,c677"

However you can't explicitly set that xattr:

  $ touch file
  $ setfattr -n "security.selinux" \
    -v "system_u:object_r:container_file_t:s0:c458,c677" file
  setfattr: file: Operation not supported

* tests/mv/part-symlink.sh: Avoid the issue.  Note this is new
since v9.12 as mv now warns about xattr ENOTSUP failures.
* tests/misc/xattr.sh: Avoid the issue (which would have
existed before v9.12).
2026-09-18 17:48:14 +01:00
Pádraig Brady 1b89826f51 tests: avoid false failures on SELinux disabled containers
* init.cfg (require_selinux_): Ensure we call is_selinux_enabled().
* src/id.c (main): Add a note about the coupling of error messages.
2026-09-18 16:27:54 +01:00
Pádraig Brady b257228543 maint: move tests/misc/printenv.h to tests/env/
* tests/misc/printenv.sh: Rename to ...
* tests/env/printenv.sh: ... here.
* tests/local.mk: Update the reference.
2026-09-17 08:52:12 +01:00
Pádraig Brady f799b2f48a env,printenv: support --quoting-style to control quoting
Quoting is currently only enabled by default when writing to terminals,
but it is usually desired to constrain the output format for processing.
E.g. for processing by a shell you can now:

  env --quoting-style=shell-escape | grep ^FOO

We use an option rather than reyling on $QUOTING_STYLE so
that $QUOTING_STYLE is restricted to be user config for terminals
and does not impact redirected processing.
Also it avoids an extra getenv() call if not outputting to tty.

* src/env.c (main): Parse --quoting-style and use that for the output
format for all outputs.
* src/printenv.c (main): Likewise.
* doc/coreutils.texi (env invocation): Describe --quoting-style.
(printenv invocation): Likewise.
* tests/env/env.sh: Reinstate testing of various output modes.
* tests/misc/printenv.sh: Likewise.
* NEWS: Mention the new feature.  Also mention the change from
the previous commit not done there so that builders would have
a clean patch against the previous release.
2026-09-17 08:52:12 +01:00
Pádraig Brady 782a1e5bc2 env,printenv: only quote when outputting to terminals
To support albeit non-robust but existing use cases like:

  env | grep SPARK_JAVA_OPT_ | sort -t_ -k4 -n |
   sed 's/[^=]*=\(.*\)/\1/g' > java_opts.txt
  readarray -t SPARK_EXECUTOR_JAVA_OPTS < java_opts.txt
  ...
  CMD=("${JAVA_HOME}/bin/java" "${SPARK_EXECUTOR_JAVA_OPTS[@]}" ...)

* src/env.c (main): Restrict quoting to terminals.
* src/printenv.c (main): Likewise.
* tests/env/env.sh: Adjust accordingly.
* tests/misc/printenv.sh: Likewise.
* doc/coreutils.texi: Likewise.

Link: https://github.com/coreutils/coreutils/issues/355
2026-09-17 08:52:12 +01:00
Collin Funk 5f6c02d344 tests: ln: test --force with a missing target
* tests/ln/misc.sh: Use getlimits_. Check the behavior of 'ln -f' with a
missing target an existing file and a file where the link would be
placed.
2026-09-15 18:27:17 -07:00
Pádraig Brady 064473118c doc: HACKING: mention the paperless assignment option
* HACKING: Mention electronic signing is now supported.
2026-09-15 20:33:21 +01:00
Sylvestre Ledru 7b970665d8 tests: csplit: cover the trailing empty split
csplit always creates the split holding the remainder of the input,
even when the patterns have already consumed everything and that
split is therefore empty.  Two ways of reaching that state were not
exercised: a regexp with an offset that lands past the last line,
and --suppress-matched with a mix of a line number and a regexp
matching the last line.  Only the all-line-numbers case was covered.
identified here: https://github.com/uutils/coreutils/pull/14565

* tests/csplit/csplit.sh: Test an offset reaching past end of input,
with and without -z.
* tests/csplit/csplit-suppress-matched.pl (mixed-1, mixed-2): Test a
line number combined with a regexp matching the last line.

Link: https://github.com/coreutils/coreutils/pull/353
2026-09-15 13:09:29 +01:00
Sylvestre Ledru 1ac51052c4 tests: ls: cover -u/-c sorting in a long listing
In a long listing, -u and -c only choose which timestamp is shown;
the entries stay sorted by name, and time order is selected by -t
(or by asking for a format that is not long).  This was only ever
exercised for short listings, so the rule went untested for every
long listing, including the ones requested with something other
than -l.
Identified here: https://github.com/uutils/coreutils/pull/14573

* tests/ls/ls-time.sh: Check that -u and -c keep name order for each
of -l, -g, -o, -n, --full-time and --dired, and that -t still gives
time order in a long listing.

Link: https://github.com/coreutils/coreutils/pull/354
2026-09-15 12:58:48 +01:00
Zachary Winnerman 32ba811b75 doc: paste: improve the wording of --help
* src/paste.c (usage): Simplify the description in --help to make it
easier to understand.
Fixes https://bugs.gnu.org/81859
Copyright-paperwork-exempt: Yes
2026-09-14 20:25:17 -07:00
Collin Funk 9d7b86ed23 maint: commit-msg: handle reparenting by Linux subreapers
* scripts/git-hooks/commit-msg ($git_pid): Define the parent process ID
at the start of the script. Use it to check if we were reparented.
2026-09-14 18:17:40 -07:00
Pádraig Brady 75f734b075 maint: add a note about dired handling of ansi codes
* src/ls.c (put_indicator): Add a note as to why we
don't update dired_pos here.

Link: https://github.com/coreutils/coreutils/pull/348
2026-09-14 22:33:15 +01:00
Paul Eggert 3474738f18 doc: File timestamps and POSIX.1-2024
* doc/coreutils.texi (File timestamps):
Update in the light of <https://bugs.gnu.org/13601#46>,
<https://austingroupbugs.net/view.php?id=1997>, and
POSIX.1-2024 Base Specifications 4.12 “File Times Update”.
2026-09-14 10:40:25 -07:00
Pádraig Brady 738bd25727 maint: post-release administrivia
* NEWS: Add header line for next release.
* .prev-version: Record previous version.
* cfg.mk (old_NEWS_hash): Auto-update.
2026-09-14 16:30:16 +01:00
Pádraig Brady c0f8514d98 version 9.12
* NEWS: Record release date.
v9.12
2026-09-14 15:55:07 +01:00
Ismail Ramzi 25792e3763 ls: fix use-after-free reading LS_COLORS ln=target
parse_ls_color stores color_indicator[C_LINK].string as a pointer into
the color_buf buffer, then frees color_buf on the unparsable-value path.
The "ln=target" check after the parse loop still read that freed buffer,
e.g. with LS_COLORS='ln=target:x'.  Skip it on the failure path, where
color output is disabled anyway and the buffer is gone.
* src/ls.c (parse_ls_color): Return after freeing color_buf on failure,
so the color_indicator[C_LINK] check only runs when the buffer is live.
* tests/ls/color-symlink-target.sh: New test.
* tests/local.mk (all_tests): Add it.
* NEWS: Mention the fix.

Link: https://github.com/coreutils/coreutils/pull/352
2026-09-14 03:07:54 +01:00
Collin Funk 6ff3889887 tests: cp: avoid a false failure on GNU/Hurd
* tests/cp/sparse-to-pipe.sh: Use 'cp -T' to avoid opening the fifo as a
target directory, which causes GNU/Hurd to wait for a writer.
2026-09-11 23:39:57 -07:00
Collin Funk 4be9843dc9 tests: dd: fix a test that would never finish on GNU/Hurd
On GNU/Hurd, this test would get stuck in an infinite loop like the
following:

    + kill -s INFO 16535
    + sleep .01
    + kill -s INFO 16535
    + sleep .01
    + kill -s INFO 16535
    + sleep .01
    [...]

The signals being sent that frequently seem to prevent 'dd' from
performing writes, since it outputs the following during that time:

    0+0 records in
    0+0 records out
    0 bytes copied, 7161.11 s, 0.0 kB/s
    0+0 records in
    0+0 records out
    0 bytes copied, 7161.88 s, 0.0 kB/s
    [...]

* tests/dd/stats.sh: Use a larger interval between signals on GNU/Hurd.
2026-09-10 22:22:37 -07:00
Collin Funk e4a3653bc3 tests: printenv: fix a false failure on GNU/Hurd
* tests/misc/printenv.sh: Also filter out LD_ORIGIN_PATH.
2026-09-10 21:00:04 -07:00
Collin Funk 80d6d6f4e3 tests: split: avoid a hang on GNU/Hurd
* tests/split/suffix-length.sh: Don't run a test on GNU/Hurd.
2026-09-10 20:19:15 -07:00
Pádraig Brady b3cdc09344 tests: avoid hangs with strace on cygwin
`strace -e tee true` will run the tee command on cygwin,
rather than filtering for the tee syscall.
Reported by Bruno Haible on Cygwin 2.9.0

* init.cfg: Use -e trace=foo where foo might be a command,
so that strace probes will fail rather than invoke foo on Cygwin.
* tests/cat/cat-distinct-err.sh: Likewise.
* tests/cat/splice.sh: Likewise.
* tests/ls/no-cap.sh: Likewise.
* tests/misc/yes.sh: Likewise.
* tests/mv/atomic.sh: Likewise.
* tests/mv/atomic2.sh: Likewise.
* tests/nproc/nproc-quota.sh: Likewise.
* tests/pwd/pwd-long.sh: Likewise.
* tests/stty/stty.sh: Likewise.
* tests/tail/debug.sh: Likewise.
* tests/tail/inotify-only-regular.sh: Likewise.
2026-09-10 16:40:06 +01:00
Pádraig Brady 83009c56fe tests: sort: avoid a false failure on very slow systems
* tests/sort/sort-NaN-infloop.sh: Up the timeout to about 240x
a "standard" system, as this was seen to timeout on a
mipsel VM using QEMU emulation.
Reported by Bruno Haible.
2026-09-10 16:12:19 +01:00
Pádraig Brady 1ea33dafe9 tests: dd: avoid false failure due to restricted memory
* tests/dd/no-allocate.sh: Expand memory restriction
(and allocated buffers) by 6MiB, to give more head-room
to the shell.
Reported and tested by Bruno Haible on OpenIndiana (from 2025).
2026-09-10 14:24:28 +01:00
Collin Funk 9e4b42982a tests: sort: avoid a failure on GNU/Hurd
On GNU/Hurd, file descriptor limits set via 'ulimit -n' seem to be
ineffective when set below the default:

    $ (ulimit -n; ulimit -n 512; ulimit -n; \
         tee $(seq 1024) < /dev/null; echo $?)
    1024
    512
    tee: 1022: Too many open files
    tee: 1023: Too many open files
    tee: 1024: Too many open files
    1

* tests/sort/sort-merge-fdlimit.sh: Don't fail if using 'ulimit -n' to
decrease the file descriptor limit is ineffective.
2026-09-08 23:53:25 -07:00
Collin Funk 1859316a9f tests: env: fix a false failure on GNU/Hurd
* tests/env/env.sh: Also filter out LD_ORIGIN_PATH.
2026-09-08 21:19:39 -07:00
Pádraig Brady 8317648768 tests: install: avoid failure on macOS
* tests/install/stdin.sh: Avoid failure on macOS for now.
In future we may support this feature on macOS as discussed at:
https://lists.gnu.org/r/coreutils/2026-06/msg00004.html
2026-09-08 19:11:27 +01:00
Pádraig Brady 4fc469f850 tests: dd: avoid false failure with older strace
* tests/dd/fail-ftruncate-fstat.sh: Ensure strace intercepts
the call, which was not the case on strace 6.19 on Alpine 3.24.1.
2026-09-08 18:35:09 +01:00
Pádraig Brady 7659e41552 tests: stat: avoid false failure with LeakSanitizer
* tests/stat/stat-mount.sh: LeakSanitizer requires /proc access
for instrumented binaries, so ensure we're checking coreutils
binaries, rather than shell builtins.
2026-09-08 17:50:18 +01:00
Pádraig Brady 7f973c4e83 stat,tail: recognize failfs file system type
* src/stat.c (human_fstype): Add "failfs" (0x4641494C).
* NEWS: Mention the improvement.
2026-09-07 18:15:24 +01:00
Pádraig Brady a551119cf7 doc: remove older ChangeLog items
* Makefile.am: Update the oldest documented version
to 9.0 which is now about 5 years old.
2026-09-07 17:46:24 +01:00
Pádraig Brady 8d6242ac3e build: update gnulib submodule to latest
* gnulib: Update to latest.
2026-09-07 17:11:50 +01:00
Collin Funk 712fe02df1 tests: df: skip a test on WSL
* tests/df/skip-rootfs.sh: Test for /init which is listed as rootfs in
/proc/self/mountinfo, but tmpfs according to statfs.
Fixes https://bugs.gnu.org/65331
2026-09-07 16:23:49 +01:00
Pádraig Brady bff0e5412d build: sort: explicitly tag libcrypto dependency
This adds an ELF note as per the UAPI Group's .note.dlopen spec:

  $ readelf -p .note.dlopen src/sort
  String dump of section '.note.dlopen':
    [     4]  y
    [     a]  |@FDO
    [    10]  [{"feature":"random-sort",
                "description":"Support for random sorting",
                "priority":"required",
                "soname":["libcrypto.so.3"]}]

* bootstrap.conf: Add sd-dlopen.
* src/sort.c (DLOPEN_LIBCRYPTO && HAVE_OPENSSL_MD5): Add an ELF note
with SD_ELF_NOTE_DLOPEN() so that packagers can determine the libcrypto
dependency, needed when --random-sort is specified.
* NEWS: Mention the Build-related change.
2026-09-06 23:16:55 +01:00
Collin Funk b2fc0baf2c doc: NEWS: fix mismatched quotation marks
* NEWS: Use '...' instead of '...`.
2026-09-04 13:57:55 -07:00
Pádraig Brady 92a8d8fe40 cut: simplify multi-byte field processing
Simplify cut_fields_mb_any() used with:

  -w in a multi-byte locale, including UTF-8
  --w=trimmed in any locale, including C
  -f -d CHAR in non utf-8 multi-byte locale
  -f -d CHAR in utf-8 with certain invalid CHARs

This is a net reducation of 24 lines,
and is performance neutral.

(cut_fields_mb_any): Use persistent first field and output state.
This allows using the existing field output helpers and
avoiding separate first field and line completion handling.
(cut_fields_bytesearch): Handle blanks explicitly.
(finish_current_line): Remove redundant condition.
2026-09-04 21:10:45 +01:00
Pádraig Brady 71419a36f7 cut: -w: optimize extracting the start of a line
There were no significant performance changes from the test cases
in commit a325c9978, but adding additional whitespace delimited cases
shows significant improvement.  Below, with -f1, typical space delimited
input shows 4x more throughput, while the long line test shows 17x more.

Performance was tested with GCC 16 on an i7-5600U with
CFLAGS='-march=native -O3 -flto' with the following setup:

$ yes | head -n10M > sl.in
$ yes $(yes eeeaae | head -n10K | paste -s -d' ') | head -n10K > llw.in
$ yes $(yes eeeaae | head -n9 | paste -s -d' ') | head -n1M > asw.in
$ yes $(yes éééááé | head -n9 | paste -s -d$'\xe2\x80\x83') |
   head -n1M > mbw.in

for type in sl llw asw mbw; do
  cat $type.in >/dev/null;
  for imp in '-before' '-after'; do
    echo ============ "${imp:-base}" $type ==============;
    for d in -w; do
      fields='-f1 -f10 -f100'
      test "$d" = "-b" && { fields='-b1 -b10 -b100'; d=''; }
      test "$d" = "-c" && { fields='-c1 -c10 -c100'; d=''; }
      for f in $fields; do
        for loc in C.UTF-8; do
          # Skip -b for UTF-8 as no different
          test "$loc" = C.UTF-8 && echo "$f" | grep -q -- -b \
           && continue
          # Skip multi-byte delimiter for C and not allowed
          test "$loc" = C && test $(echo -n "$d" | wc -c) -ge 4 \
           && continue
          bin=src/cut${imp}
          LC_ALL=$loc $bin $f $d /dev/null 2>/dev/null &&
          hyperfine --warmup=2 -m2 -M6 \
           "LC_ALL=$loc $bin $f $d $type.in >/dev/null" ||
          printf 'Benchmark 1: %s\n  unsupported\n\n' \
           "LC_ALL=$loc $bin $f $d $type.in >/dev/null"
        done;
      done;
    done;
  done;
done

After a little post-processing of the results, we get:

-- src/cut-before

| command         |       sl |      llw |      asw |      mbw |
| --------------- | -------- | -------- | -------- | -------- |
| UTF8 -f1 -w     |  96.2 ms |  1.916 s | 190.6 ms | 906.5 ms |
| UTF8 -f10 -w    | 104.6 ms |  1.921 s | 187.7 ms | 845.6 ms |
| UTF8 -f100 -w   | 104.4 ms |  1.916 s | 187.4 ms | 846.4 ms |

-- src/cut-after

| command         |       sl |      llw |      asw |      mbw |
| --------------- | -------- | -------- | -------- | -------- |
| UTF8 -f1 -w     |  88.7 ms | 112.4 ms |  43.5 ms | 146.3 ms |
| UTF8 -f10 -w    |  95.2 ms | 114.4 ms | 193.5 ms | 848.4 ms |
| UTF8 -f100 -w   |  94.8 ms | 130.6 ms | 194.4 ms | 848.1 ms |

* src/cut.c (scan_mb_line_end): A new function to skip
to EOL once we have exhausted the field selection.
(cut_fields_mb_any): Only skip after recognizing a field delimiter.
* tests/cut/cut.pl: Add test cases.
* tests/cut/mb-non-utf8.sh: Likewise.
* NEWS: Mention the improvement.
2026-09-04 21:10:45 +01:00
Pádraig Brady f665fdffca cut: -w: avoid per character function calls
Adding a couple of inlining hints to the compiler
gives a 30% boost in raw character processing performance
in cut_fields_mb_any().

  $ yes $(yes eeeaae | head -n10K | paste -s -d,) | head -n10K > ll.in

  $ time LC_ALL=C.UTF-8 src/cut-before -f1 -w ll.in >/dev/null
  real	0m2.279s

  $ time LC_ALL=C.UTF-8 src/cut-after -f1 -w ll.in >/dev/null
  real	0m1.619s

* src/cut.c (scan_mb_blank_field): Tag as inline.
(scan_mb_delim_field): Likewise.
2026-09-04 21:10:45 +01:00