newsGNU Core Utilities - News: coreutils-8.21 released [stable]

 
 
Latest News
coreutils-9.5 released [stable] posted by pixelbeat, Thu 28 Mar 2024 03:39:50 PM UTC
coreutils-9.4 released [stable] posted by pixelbeat, Tue 29 Aug 2023 03:16:59 PM UTC
coreutils-9.3 released [stable] posted by pixelbeat, Tue 18 Apr 2023 03:11:21 PM UTC
coreutils-9.2 released [stable] posted by pixelbeat, Mon 20 Mar 2023 03:53:57 PM UTC
coreutils-9.1 released [stable] posted by pixelbeat, Fri 15 Apr 2022 10:34:37 PM UTC

coreutils-8.21 released [stable]

Item posted by Pádraig Brady <pixelbeat> on Thu 14 Feb 2013 05:23:34 PM UTC.


This is to announce coreutils-8.21, a stable release.

There have been 121 commits by 18 poeple in the 16 weeks since 8.20.

Executive summary: 8.21 is mainly a bug fix release, including fixes
for recent regressions in cp, factor and seq.  cut has received fixes for
many long standing issues.  df is updated to better handle newer systems
that link /etc/mtab to /proc/mounts, and also provides a new --output
option to control which fields to display.  A new numfmt utility was
included to provide various number formating and conversion functions.

See the NEWS below for a brief summary.

Thanks to everyone who has contributed!
The following people contributed changes to this release:

  Assaf Gordon (7):
  Benno Schulenberg (6):
  Bernhard Voelker (24):
  Cojocaru Alexandru (2):
  Colin Watson (1):
  Daniel Schepler (1):
  Jakob Truelsen (1):
  Jim Meyering (13):
  Karl Berry (2):
  Mike Frysinger (2):
  Ondrej Oprala (2):
  Ondřej Vašík (1):
  Paul Eggert (12):
  Pádraig Brady (46):
  Stefano Lattarini (1):
  Stephan Krempel (1):
  Zartaj Majeed (1):
  Ángel González (1):

Pádraig [on behalf of the coreutils maintainers]

==================================================================

Here is the GNU coreutils home page:
    http://gnu.org/s/coreutils/

For a summary of changes and contributors, see:
  http://git.sv.gnu.org/gitweb/?p=coreutils.git;a=shortlog;h=v8.21
or run this command from a git-cloned coreutils directory:
  git shortlog v8.20..v8.21

To summarize the 173 gnulib-related changes, run these commands
From a git-cloned coreutils directory:
  git checkout v8.21
  git submodule summary v8.20

==================================================================

Here are the compressed sources and a GPG detached signature[*]:
  http://ftp.gnu.org/gnu/coreutils/coreutils-8.21.tar.xz
  http://ftp.gnu.org/gnu/coreutils/coreutils-8.21.tar.xz.sig

Use a mirror for higher download bandwidth:
  http://ftpmirror.gnu.org/coreutils/coreutils-8.21.tar.xz
  http://ftpmirror.gnu.org/coreutils/coreutils-8.21.tar.xz.sig

[*] Use a .sig file to verify that the corresponding file (without the
.sig suffix) is intact.  First, be sure to download both the .sig file
and the corresponding tarball.  Then, run a command like this:

  gpg --verify coreutils-8.21.tar.xz.sig

If that command fails because you don't have the required public key,
then run this command to import it:

  gpg --keyserver keys.gnupg.net --recv-keys DF6FD971306037D9

and rerun the 'gpg --verify' command.

This release was bootstrapped with the following tools:
  Autoconf 2.68
  Automake 1.11.6
  Gnulib v0.0-7848-g4a82904
  Bison 2.4.3

NEWS

* Noteworthy changes in release 8.21 (2013-02-14) [stable]

** New programs

  numfmt: reformat numbers

** New features

  df now accepts the --output[=FIELD_LIST] option to define the list of columns
  to include in the output, or all available columns if the FIELD_LIST is
  omitted.  Note this enables df to output both block and inode fields together.

  du now accepts the --threshold=SIZE option to restrict the output to entries
  with such a minimum SIZE (or a maximum SIZE if it is negative).
  du recognizes -t SIZE as equivalent, for compatibility with FreeBSD.

** Bug fixes

  cp --no-preserve=mode now no longer exits non-zero.
  [bug introduced in coreutils-8.20]

  cut with a range like "N-" no longer allocates N/8 bytes.  That buffer
  would never be used, and allocation failure could cause cut to fail.
  [bug introduced in coreutils-8.10]

  cut no longer accepts the invalid range 0-, which made it print empty lines.
  Instead, cut now fails and emits an appropriate diagnostic.
  [This bug was present in "the beginning".]

  cut now handles overlapping to-EOL ranges properly.  Before, it would
  interpret "-b2-,3-" like "-b3-".  Now it's treated like "-b2-".
  [This bug was present in "the beginning".]

  cut no longer prints extraneous delimiters when a to-EOL range subsumes
  another range.  Before, "echo 123|cut --output-delim=: -b2-,3" would print
  "2:3".  Now it prints "23".  [bug introduced in 5.3.0]

  cut -f no longer inspects input line N+1 before fully outputting line N,
  which avoids delayed output for intermittent input.
  [bug introduced in TEXTUTILS-1_8b]

  factor no longer loops infinitely on 32 bit powerpc or sparc systems.
  [bug introduced in coreutils-8.20]

  install -m M SOURCE DEST no longer has a race condition where DEST's
  permissions are temporarily derived from SOURCE instead of from M.

  pr -n no longer crashes when passed values >= 32.  Also, line numbers are
  consistently padded with spaces, rather than with zeros for certain widths.
  [bug introduced in TEXTUTILS-1_22i]

  seq -w ensures that for numbers input in scientific notation,
  the output numbers are properly aligned and of the correct width.
  [This bug was present in "the beginning".]

  seq -w ensures correct alignment when the step value includes a precision
  while the start value does not, and the number sequence narrows.
  [This bug was present in "the beginning".]

  seq -s no longer prints an erroneous newline after the first number, and
  outputs a newline after the last number rather than a trailing separator.
  Also seq no longer ignores a specified step value when the end value is 1.
  [bugs introduced in coreutils-8.20]

  timeout now ensures that blocking of ALRM signals is not inherited from
  its parent, which would cause timeouts to be ignored.
  [the bug dates back to the initial implementation]

** Changes in behavior

  df --total now prints '-' into the target column (mount point) of the
  summary line, accommodating the --output option where the target field
  can be in any column.  If there is no source column, then df prints
  'total' in the target column.

  df now properly outputs file system information with bind mounts present on
  the system by skipping duplicate entries (identified by the device number).
  Consequently, df also elides the early-boot pseudo file system type "rootfs".

  nl no longer supports the --page-increment option, which has been
  deprecated since coreutils-7.5.  Use --line-increment instead.

** Improvements

  readlink now supports multiple arguments, and a complementary
  -z, --zero option to delimit output items with the NUL character.

  stat and tail now know about CEPH.  stat -f --format=%T now reports the file
  system type, and tail -f uses polling for files on CEPH file systems.

  stty now supports configuring DTR/DSR hardware flow control where available.

** Build-related

  Perl is now more of a prerequisite.  It has long been required in order
  to run (not skip) a significant percentage of the tests.  Now, it is
  also required in order to generate proper man pages, via help2man.  The
  generated man/*.1 man pages are no longer distributed.  Building without
  perl, you would create stub man pages.  Thus, while perl is not an
  official prerequisite (build and "make check" will still succeed), any
  resulting man pages would be inferior.  In addition, this fixes a bug
  in distributed (not from clone) Makefile.in that could cause parallel
  build failure when building from modified sources, as is common practice
  for a patched distribution package.

  factor now builds on x86_64 with x32 ABI, 32 bit MIPS, and all HPPA systems,
  by avoiding incompatible asm.  [bug introduced in coreutils-8.20]

  A root-only test predicate would always fail.  Its job was to determine
  whether our dummy user, $NON_ROOT_USERNAME, was able to run binaries from
  the build directory.  As a result, all dependent tests were always skipped.
  Now, those tests may be run once again.  [bug introduced in coreutils-8.20]



 

Back to the top

Powered by Savane 3.13-4448.
Corresponding source code