new snapshot available: coreutils-6.11.85-c4a95
We're on track to make a stable release in the next week or so. There is still a pending portability problem involving Mac OS ACLs. It's nothing new, so I may defer the fix until after coreutils-6.12. The most recent change was required to work around a new test failure induced by a new version of libc on some systems: tests: skip when a debian libc6-2.7-11 bug makes printf segfault http://git.savannah.gnu.org/gitweb/?p=coreutils.git;a=commitdiff;h=c4a9551eee make check passes for me on Fedora rawhide, Debian unstable, Solaris 10 and FreeBSD 6.1, but given the changes in the testing framework, I'm sure to have introduced a few new problems there, so please give this a try and report any failures. coreutils snapshot: http://meyering.net/cu/coreutils-ss.tar.gz8.5 MB http://meyering.net/cu/coreutils-ss.tar.lzma 3.6 MB http://meyering.net/cu/coreutils-ss.tar.gz.sig http://meyering.net/cu/coreutils-ss.tar.lzma.sig aka http://meyering.net/cu/coreutils-6.11.85-c4a95.tar.gz http://meyering.net/cu/coreutils-6.11.85-c4a95.tar.lzma Changes since 6.11.17-d30ac: Bernhard Marx (1): doc: improve description of niceness values Bo Borgerson (6): Only cleanup test dirs from the process that created them. tests: don't chmod after a failed chdir in cleanup tests: remove references to tests/wc from bootstrap Add Daniel Dunbar's lcov instructions to HACKING base64 module: adjust API so it's compatible with gnulib's base64: remove some unused/redundant getopt code Bruno Haible (2): Speed up wc -m and wc -w in multibyte case. doc: some Unicode characters cannot be specified via \u or \U Jim Meyering (59): tests: do define built_programs tweak HACKING advice tests: cp/perm (usually not run) was failing on systems with SELinux tests: move another file (expensive) into test-lib.sh tests: put root-only (usually skipped) tests at the end tests: reorder some tests in the long list tests: improve perl-based tempdir handling tests: ensure at least one failure when $built_programs is empty tests: don't source envvar-check manually, test-lib.sh does it tests: remove temporary log file upon catchable signal move wc tests from own subdir into a single script tests: slightly relax sc_cast_of_argument_to_free syntax check tests: don't hard-code coreutils list of tests/ SUBDIRS in bootstrap chcon, runcon: make --help print the bug-reporting address tests: avoid spurious make check-root failure tests: avoid a make check-root failure when mcstransd is running tac: avoid segfault for e.g., echo x; tac -r x x * tests/misc/tac: Set execute bit. help2man: avoid failure with Debian unstable's Perl 5.10.0 help2man: fix perl 5.10 problem properly avoid problems with sign-extended char operand to is* functions tests: detect and check more uses of strcmp * tests/misc/wc: Remove an unused variable. tests: test split more thoroughly * tests/touch/Makefile.am: Remove now-unused file. fix typo in comments: s/ouput/output/ tests: hoist envvar-check so it is run for every test tests: hoist the sourcing of lang-default tests: save and restore TERM around use of TESTS_ENVIRONMENT, tests: use printf+sed rather than yes+head+tr tests: move group-names into test-lib.sh tests: move sparse-file into test-lib.sh tests: Coreutils.pm improvements tests: remove directory, tests/head/ tests: remove directory, tests/cut/ tests: remove directory, tests/uniq/ tests: translate uniq-z-test-adding code to new framework tests: skip another test if mcstransd is running tests: remove directory, tests/tr/ * TODO: Bo Borgerson is rewriting support for cp --recursive tests: Coreutils.pm: support running a program that is a shell built-in tests: remove directory, tests/test/ tests: remove directory, tests/tac/ tests: remove directory, tests/sort/ tests: remove directory, tests/tail/ tests: convert pr tests * tests/misc/tac: Also perform stdin and piped tests. tests: remove directory, tests/join/ now that the last of the mk-script-using tests is gone... move sha256 and sha512 modules to gnulib make HACKING slightly more generic tests: improve coverage of printf.c tests: env-related clean up tests: allow to run Perl tests more cleanly tests: remove ugly /bin/sh wrapper around each perl-based test script test invalid-option handling in all programs du, wc: merge improved --files0-from=F-related diagnostics tests: sync and update wc and du --files0-from tests tests: skip when a debian libc6-2.7-11 bug makes printf segfault ___ Bug-coreutils mailing list Bug-coreutils@gnu.org
Feat-Req: uniq -c delimiter should be changable
Hi gurus, uniq -c uses space to separate counts from lines since 2003, as this is required by posix. I understand that you want to stay compatible with posix, but why is there not option to change this behaviour, to make uniq behave like cut or sort? This would make combining uniq -c with tools like gawk, sort or cut much easier. Patrick Tufts already proposed this (I only found it here: http://www.opensubscriber.com/message/bug-coreutils@gnu.org/5107663.html) and Bob Proulx wanted examples, so here is my standard example: Let's say I only want the 50 most common lines of a file: cat textfile | sort | uniq -c | sort -n | tail -n 50 | tr -s ' ' | cut -f2 This will only print the first word of each line with the current uniq version though if uniq -c understood -d/-t it would print the whole This is just a proposition for a feature that would make uniq more homogenous. Debian documents that --separator was actually patched into the debian version of uniq in the coreutils package, but the patch has been removed, I guess the coreutils developers at the time didn't find this change useful enough. Is there any good reason not to include such an option? thanks Max ___ Bug-coreutils mailing list Bug-coreutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-coreutils
Re: Feat-Req: uniq -c delimiter should be changable
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Maximilian Haeussler on 5/17/2008 8:36 AM: | | Let's say I only want the 50 most common lines of a file: | cat textfile | sort | uniq -c | sort -n | tail -n 50 | tr -s ' ' | cut -f2 Unrelated to your report, but this is a useless use of cat. Why not: sort textfile | uniq -c ... - -- Don't work too hard, make some time for fun as well! Eric Blake [EMAIL PROTECTED] -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (Cygwin) Comment: Public key at home.comcast.net/~ericblake/eblake.gpg Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkgu8HwACgkQ84KuGfSFAYCT0QCdHo9+eVgsK6XUrHjMuyHe3xkD mVMAoNd7Sdskvz7IhNmPcejguvneCeH5 =ed2u -END PGP SIGNATURE- ___ Bug-coreutils mailing list Bug-coreutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-coreutils
Re: Feat-Req: uniq -c delimiter should be changable
Hi, On Sat, May 17, 2008 at 04:36:30PM +0200, Maximilian Haeussler wrote: Let's say I only want the 50 most common lines of a file: cat textfile | sort | uniq -c | sort -n | tail -n 50 | tr -s ' ' | cut -f2 This will only print the first word of each line with the current uniq version though if uniq -c understood -d/-t it would print the whole You should tell `cut' to output every field starting after the number field inserted by `uniq': sort textfile | uniq -c | sort -n | tail -n50 | tr -s ' ' | cut -f3- -d' ' Erik ___ Bug-coreutils mailing list Bug-coreutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-coreutils
Re: Problème sous linux
On 13/05/08 at 10:11 +0100, James Youngman wrote: In any case, you have asked for help in the wrong place. This mailing list is devoted to the support and improvements of the GNU Core Utilities (http://www.gnu.org/software/coreutils/). It seems to me that your problem is probably not related to these. I think that the best thing for you to do would be to ask for help on a French-language mailing list devoted to user-support on the Linux distribution of your choice.This one isn't such a list. I wonder if he got that email from the output of command --help: $ LC_ALL=fr_FR.UTF8 ls --help [..] Rapporter toutes anomalies à bug-coreutils@gnu.org. Which translates to report all bugs, not just report bugs. -- | Lucas Nussbaum | [EMAIL PROTECTED] http://www.lucas-nussbaum.net/ | | jabber: [EMAIL PROTECTED] GPG: 1024D/023B3F4F | ___ Bug-coreutils mailing list Bug-coreutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-coreutils
sort --version-sort
Here is the usage text from my latest --sort=version patch (attached),
now that paperwork is in order.
Cheers - Bruce
$ ./sort --help
Usage: ./sort [OPTION]... [FILE]...
Write sorted concatenation of all FILE(s) to standard output.
Mandatory arguments to long options are mandatory for short options too.
Ordering options:
-b, --ignore-leading-blanks ignore leading blanks
-d, --dictionary-order consider only blanks and alphanumeric characters
-f, --ignore-case fold lower case to upper case characters
-g, --general-numeric-sort compare according to general numerical value
-i, --ignore-nonprintingconsider only printable characters
-M, --month-sortcompare (unknown) `JAN' ... `DEC'
-n, --numeric-sort compare according to string numerical value
-R, --random-sort sort by random hash of keys
-V, --version-sort sort by numeric version (see strverscmp(3C))
--random-source=FILEget random bytes from FILE (default /dev/urandom)
--sort=WORD sort according to WORD:
general-numeric -g, month -M, numeric -n,
random -R, version -V
-r, --reverse reverse the result of comparisons
Other options:
-c, --check, --check=diagnose-first check for sorted input; do not sort
-C, --check=quiet, --check=silent like -c, but do not report first bad line
--compress-program=PROG compress temporaries with PROG;
decompress them with PROG -d
-k, --key=POS1[,POS2] start a key at POS1, end it at POS2 (origin 1)
-m, --merge merge already sorted files; do not sort
-o, --output=FILE write result to FILE instead of standard output
-s, --stable stabilize sort by disabling last-resort comparison
-S, --buffer-size=SIZEuse SIZE for main memory buffer
-t, --field-separator=SEP use SEP instead of non-blank to blank transition
-T, --temporary-directory=DIR use DIR for temporaries, not $TMPDIR or /tmp;
multiple options specify multiple directories
-u, --unique with -c, check for strict ordering;
without -c, output only the first of an equal run
-z, --zero-terminated end lines with 0 byte, not newline
--help display this help and exit
--version output version information and exit
POS is F[.C][OPTS], where F is the field number and C the character position
in the field; both are origin 1. If neither -t nor -b is in effect, characters
in a field are counted from the beginning of the preceding whitespace. OPTS is
one or more single-letter ordering options, which override global ordering
options for that key. If no key is given, use the entire line as the key.
SIZE may be followed by the following multiplicative suffixes:
% 1% of memory, b 1, K 1024 (default), and so on for M, G, T, P, E, Z, Y.
With no FILE, or when FILE is -, read standard input.
*** WARNING ***
The locale specified by the environment affects sort order.
Set LC_ALL=C to get the traditional sort order that uses
native byte values.
Report bugs to bug-coreutils@gnu.org.
diff --git a/src/sort.c b/src/sort.c
index 8b2eec5..adb4589 100644
--- a/src/sort.c
+++ b/src/sort.c
@@ -172,6 +172,7 @@ struct keyfield
Handle numbers in exponential notation. */
bool month; /* Flag for comparison by month name. */
bool reverse; /* Reverse the sense of comparison. */
+ bool version; /* sort by version number */
struct keyfield *next; /* Next keyfield to try. */
};
@@ -328,10 +329,11 @@ Ordering options:\n\
-M, --month-sortcompare (unknown) `JAN' ... `DEC'\n\
-n, --numeric-sort compare according to string numerical value\n\
-R, --random-sort sort by random hash of keys\n\
+ -V, --version-sort sort by numeric version (see strverscmp(3C))\n\
--random-source=FILEget random bytes from FILE (default /dev/urandom)\n\
--sort=WORD sort according to WORD:\n\
general-numeric -g, month -M, numeric -n,\n\
-random -R\n\
+random -R, version -V\n\
-r, --reverse reverse the result of comparisons\n\
\n\
), stdout);
@@ -398,7 +400,7 @@ enum
SORT_OPTION
};
-static char const short_options[] = -bcCdfgik:mMno:rRsS:t:T:uy:z;
+static char const short_options[] = -bcCdfgik:mMno:rRsS:t:T:uVy:z;
static struct option const long_options[] =
{
@@ -423,31 +425,50 @@ static struct option const long_options[] =
{field-separator, required_argument, NULL, 't'},
{temporary-directory, required_argument, NULL, 'T'},
{unique, no_argument, NULL, 'u'},
+ {compare-version, no_argument, NULL, 'V'},
{zero-terminated, no_argument, NULL, 'z'},
{GETOPT_HELP_OPTION_DECL},
{GETOPT_VERSION_OPTION_DECL},
{NULL, 0,
Fwd: Problème sous linux
Dear French translation team, We seem to get a lot of questions, in French, about non-coreutils problems sent to the bug-coreutils@gnu.org mailing list. Lucas (CC'ed) pointed out that it's possible that the translation of the --help message for various coreutils programs could be a contributory factor to this pattern. I notice that other coreutils tools feature the same translation of the message (cat, for example). If this usage really is misleading it might be worth fixing the translation of the help message.Hence forwarding this message to you. James. -- Forwarded message -- From: Lucas Nussbaum [EMAIL PROTECTED] Date: Sat, May 17, 2008 at 5:43 PM Subject: Re: Problème sous linux To: James Youngman [EMAIL PROTECTED] Cc: nel natou [EMAIL PROTECTED], bug-coreutils@gnu.org On 13/05/08 at 10:11 +0100, James Youngman wrote: In any case, you have asked for help in the wrong place. This mailing list is devoted to the support and improvements of the GNU Core Utilities (http://www.gnu.org/software/coreutils/). It seems to me that your problem is probably not related to these. I think that the best thing for you to do would be to ask for help on a French-language mailing list devoted to user-support on the Linux distribution of your choice.This one isn't such a list. I wonder if he got that email from the output of command --help: $ LC_ALL=fr_FR.UTF8 ls --help [..] Rapporter toutes anomalies à bug-coreutils@gnu.org. Which translates to report all bugs, not just report bugs. -- | Lucas Nussbaum | [EMAIL PROTECTED] http://www.lucas-nussbaum.net/ | | jabber: [EMAIL PROTECTED] GPG: 1024D/023B3F4F | ___ Bug-coreutils mailing list Bug-coreutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-coreutils
Re: Problème sous linux
Perhaps it should be useful to mention that although the tools are i18n'zed, all reports should be in english to be taken into account. If so, we would have to display such a notice in a very obvious place, like near the mailing-list address, not hidden in some man pages. Seriously, I would be glad to fix the french translations if needed (I can do the german and japanese one as well). About the bogus bug report, I don't think that user even know the ls command exists, or knows about the console commands (he/she is a basic user, not a developer, asking general technical questions like he/she would do on a dedicated tech support address) . There is obviously easy to access graphical tool displaying this list address as a direct support contact in the Ubuntu release. This has to be something very noticeable. I have contacted the user to get more information on this. Gabriel On 05/17/2008 6:43:18 PM +0200, Lucas Nussbaum [EMAIL PROTECTED] wrote: On 13/05/08 at 10:11 +0100, James Youngman wrote: In any case, you have asked for help in the wrong place. This mailing list is devoted to the support and improvements of the GNU Core Utilities (http://www.gnu.org/software/coreutils/). It seems to me that your problem is probably not related to these. I think that the best thing for you to do would be to ask for help on a French-language mailing list devoted to user-support on the Linux distribution of your choice.This one isn't such a list. I wonder if he got that email from the output of command --help: $ LC_ALL=fr_FR.UTF8 ls --help [..] Rapporter toutes anomalies à bug-coreutils@gnu.org. Which translates to report all bugs, not just report bugs. ___ Bug-coreutils mailing list Bug-coreutils@gnu.org http://lists.gnu.org/mailman/listinfo/bug-coreutils
