Hello, I ran into a build failure when compiling coreutils from
master. On FreeBSD amd64 14.0-RELEASE-p4 with clang version 16.0.6 and
no options given to ./configure I see the following error:
src/od.c:510:1: error: passing 'bfloat16' (aka '__bf16') to parameter of
incompatible type 'float'
PRIN
mailing list. Would that
be a better place to send patches in the future? Thanks.From 09aff2babb8323ec125f52494874a33be8987346 Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: Wed, 7 Feb 2024 03:58:02 -0800
Subject: [PATCH] env: Fix errors caused by -Wshadow.
* src/env.c (parse_signal_
Hello, I noticed an old FIXME note in coreutils.texi and TODO about
info pages for these programs. Is there still a need/desire for them?
I notice that `man [' opens the manpage for `test' so I assume there
isn't much need for a separate one. When opening the man page for
pinky the SEE ALSO section
From: Collin Funk
Date: Sun, 18 Feb 2024 12:23:07 -0800
Subject: [PATCH] maint: Prefer #include <...> for gnulib substitute headers
* src/shuf.c: Change #include "getopt.h" to #include .
* src/stat.c: Change #include "getopt.h" to #include .
* src/system.h: Change #inc
the gnulib readutmp module. I've left it out for now since I am
unsure.
CollinFrom ab1a38bb2e47b0987597a54604de287d2bfa27d6 Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: Tue, 12 Mar 2024 06:30:52 -0700
Subject: [PATCH] doc: add basic documentation for 'pinky'
* doc/coreut
I noticed that src/comm.c #define's min. Since system.h includes
gnulib's minmax.h we can just remove it and use that.
Pretty pointless, but might as well remove the duplication. :)
CollinFrom 66af203c4da9834da22787b46be9cc610f7f8b6c Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: T
:blank:]]*include[[:blank:]]+'
CollinFrom 31a9bc521f495613e3a911e3b040cf668af01f2f Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: Tue, 16 Apr 2024 13:38:06 -0700
Subject: [PATCH] maint: Remove unnecessary stdbool.h include
* src/temp-stream.c: Don't include since it is defined in
config
yone has a better idea.
CollinFrom e179a4432343914868410227a98157b90e82a93a Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: Sat, 27 Apr 2024 19:29:30 -0700
Subject: [PATCH 1/2] maint: pacify GCC 14 -Wmissing-variable-declarations,
part 1
* src/local.mk: Add an extern decl for 'Version'.
---
sr
On 5/1/24 6:41 AM, Pádraig Brady wrote:
> I merged your attached two patches, adjusted them a little and pushed at:
> https://github.com/coreutils/coreutils/commit/ba8a43c24
Oops, I didn't see the code for generating the CRC table. Thanks!
> That's a bit tricky as full prototypes would be a bit o
Hi,
Hongyi Zhao writes:
> Using env command:
>
> env PATH="/custom/path:$PATH" my_command
>
> Directly setting the environment variable:
>
> PATH="/custom/path:$PATH" my_command
[...]
> In what scenarios would it be more appropriate to use env versus
> directly setting the environment variable?
a look at that.
Collin
>From ce463172d9340ae1b078808d8318fd893ae1445b Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: Sat, 29 Jun 2024 04:23:11 -0700
Subject: [PATCH 1/2] build: update gnulib submodule to latest
---
gnulib | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --g
lin
>From 81ee20f2fd459b5c138fc3411540f32c4a5faecc Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: Sun, 30 Jun 2024 20:54:58 -0700
Subject: [PATCH] maint: Silence deprecated module warning
* bootstrap.conf (gnulib_modules): Use putenv-gnu instead of putenv.
---
bootstrap.conf | 2 +-
1 file changed, 1 insertion(+
00:00:00 2001
From: Collin Funk
Date: Wed, 3 Jul 2024 17:50:14 -0700
Subject: [PATCH] maint: import tests/init.sh from Gnulib during bootstrap
* bootstrap.conf (bootstrap_post_import_hook): Use gnulib-tool
--copy-file to import tests/init.sh.
* tests/init.sh: Remove file.
* .gitignore (/tests/i
Jim Meyering writes:
> Thanks. I've done the same for gzip, sed, diffutils, grep and cppi.
> It needed a tiny adaptation for sed's use of "testsuite" as the
> directory name rather than the more common "tests".
> Also, there were two new warnings exposed in gzip (because I built
> with bleeding e
ing like in 'ls'.
Collin
>From 0c2bd6a6840d6f46e01e2caeaad9d42d10c34367 Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: Mon, 29 Jul 2024 21:16:36 -0700
Subject: [PATCH] maint: remove unnecessary inttostr usage in printf
* src/cksum.c (output_crc): Use '%ju' instead of um
Hi,
These patches get rid of the use of deprecated function declarations due
to Bruno's changes today [1].
Collin
[1] https://lists.gnu.org/archive/html/bug-gnulib/2024-08/msg00146.html
>From 87231767771aed7c699c540f12efd5699b752cb7 Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: Sat
Hi,
Gnulib and Coreutils assume C89 escape sequences everywhere else, so I
think these #if __STDC__ conditionals can be removed.
>From 2702190bc135340fad044fd4f36f322ad7bbce11 Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: Tue, 3 Sep 2024 19:45:13 -0700
Subject: [PATCH] maint: assume
I noticed these two calls to error assume that uid_t fits in int and
unsigned long int. Attached patch fixes it.
CollinFrom 31aa496133ded6132ca94538ad9388eba589b3b3 Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: Sun, 6 Oct 2024 22:19:51 -0700
Subject: [PATCH] chroot,whoami: use uintmax_t for
oying since they have to be quoted for the shell to
not interpret them as redirections. But now that they are standardized
they should probably be implemented. :)
Collin
[1] https://pubs.opengroup.org/onlinepubs/9799919799/utilities/test.html
>From 2a76b314be161eb6f2e0915e84401e7e04a0ca4b Mon S
Pádraig Brady writes:
> Excellent.
> I added a NEWS entry and s/coreutils.txt/coreutils.texi/ in the commit
> message,
> and pushed.
Thanks, I always forget to update NEWS.
> This also has the advantage of matching the behavior of bash and dash at
> least.
Ah, cool. I meant to check bash but
ies/stty.html
>From 49a15c27690e8d4ce19863fdbff55d6b4b3f414a Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: Thu, 31 Oct 2024 22:43:18 -0700
Subject: [PATCH] stty: adjust --help to match POSIX 2024
* src/stty.c (usage): Mark cols, rows, and size as standardized by
POSIX.
* doc/coreutils.texi (Speci
Hi,
2 simple patches.
First to use mbszero over memset like the rest of the code.
Second to ignore a generated CRC table.
Collin
>From c70ccaeb7422c2f075930746ee76502b39baed8b Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: Tue, 19 Nov 2024 18:43:19 -0800
Subject: [PATCH 1/2] maint: pre
Collin Funk writes:
> It is harmless to use it twice, but the attached patch
Forgot attachment...
>From 4d76e2641e9c1597245c30026e02c9eaa6209a9b Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: Sun, 19 Jan 2025 21:15:35 -0800
Subject: [PATCH] maint: tests: remove duplicate uses of &
In Perl the 'my' keyword only needs to be once per block.
It is harmless to use it twice, but the attached patch removes all the
warnings like this in the test suite:
"my" variable $arg1 masks earlier declaration in same scope at
./tests/factor/factor.pl line 105.
Collin
GNU/Hurd fails a test for env because LD_ORIGIN_PATH is defined. If it
is unset it passes. I attached a patch.
Collin
>From a8e2a18459bb5ab34064e2da2135a48fec141d66 Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: Mon, 13 Jan 2025 19:43:12 -0800
Subject: [PATCH] tests: env-S.pl: unset GNU/H
Bruno Haible via GNU coreutils General Discussion
writes:
> I propose to ignore it, because
> - AIX 7.1 is end-of-life since 2023-07-28,
> - Newer AIX releases come with newer compilers.
I ran into something strange on this platform that I came to the same
conclusion for.
The AIX 'make' com
Collin
>From 910cc6d312278fc4afc30abe5f232ba4051ea93c Mon Sep 17 00:00:00 2001
From: Collin Funk
Date: Fri, 4 Apr 2025 12:35:15 -0700
Subject: [PATCH] maint: ensure that new "make syntax-check"-run sc_codespell
passes
* cfg.mk (codespell_ignore_words_list): Ignore false-positives.
(exclude_file_name_regexp--s
Pádraig Brady writes:
> I split an over-long line as follows and pushed.
I wasn't sure how to split that line and now feel a bit silly for not
using make variables... Thanks!
Collin
recommend GNULIB_REFDIR since it will
always create a submodule and checkout the proper Gnulib commit that
Coreutils specifies.
Collin
[1]
https://github.com/coreutils/gnulib/commit/2122284380cc0d1b3b6f11d92c04652616da79c7
>From 790d6677f2bb9e1304ffe41180194283e65954ee Mon Sep 17 00:00:00 2001
From
t would only matter if you were using a
cross-compiler on a machine without a host compiler, though.
Collin
>From c805c00b65ae3a931f7b389cabe249a45069fe18 Mon Sep 17 00:00:00 2001
Message-ID:
From: Collin Funk
Date: Sat, 17 May 2025 21:02:20 -0700
Subject: [PATCH 1/2] build: fix 'make
l let you decide.
Collin
[1]
https://www.ibm.com/docs/en/aix/7.3.0?topic=techniques-preloaded-shared-libraries
>From ab3a7c93c5e4dd40f998e944b52e7de1d2f50390 Mon Sep 17 00:00:00 2001
Message-ID:
From: Collin Funk
Date: Sun, 22 Jun 2025 18:13:36 -0700
Subject: [PATCH] stdbuf: support AI
:00 2001
Message-ID: <9439e4eace278baaa0a4b9d090831b67f8e07947.1750827159.git.collin.fu...@gmail.com>
From: Collin Funk
Date: Tue, 24 Jun 2025 21:39:20 -0700
Subject: [PATCH] maint: prefer UINTMAX_MAX to TYPE_MAXIMUM (uintmax_t)
* src/df.c (get_dev): Use UINTMAX_MAX.
* src/shred.c (dopass):
Hi Shou-Chi,
Shou-Chi Chen writes:
> I've subscribed to the coreutils mailing lists and am still observing how
> things work.
> I'm curious about where to send patches. I read the guide in `README-hacking`
> at the
> project root, which states that patches for fixes or new features should be
083e62727a74ae624e2778333c Mon Sep 17 00:00:00 2001
Message-ID: <10e69e75b23247083e62727a74ae624e2778333c.1750571130.git.collin.fu...@gmail.com>
From: Collin Funk
Date: Sat, 21 Jun 2025 21:50:27 -0700
Subject: [PATCH 1/2] maint: use _GL_CMP instead of handwriting three-valued
comparisons
l push in a bit.
[1]
https://github.com/freebsd/freebsd-src/commit/b52c534bffcc40e5be49f6d012c9d4005f0bbd81
>From f37b23dafb1ac79556ab1afa9c3f7917065fe108 Mon Sep 17 00:00:00 2001
Message-ID:
From: Collin Funk
Date: Thu, 26 Jun 2025 18:48:51 -0700
Subject: [PATCH] doc: mention that POSI
Hi Mike,
Mike Hoye writes:
> Hello, coreutils people -
>
> I am wondering if there is any appetite for adding jsonlines as an output
> format in the various core utilities. A common structured-output format in
> low level tools would make tool chaining and constructive exploration
> significantl
Pádraig Brady writes:
> Formatting is good.
>
> +1
Thanks for checking. Pushed.
Collin
Pádraig Brady writes:
> On 27/06/2025 02:56, Collin Funk wrote:
>> POSIX.1-2024 added the 'iflags=fullblock' option to 'dd'. This patch
>> documents that.
>> Although it is a part of POSIX, it is not widely supported. FreeBSD
>> added it in 2019,
Hi Berny,
Bernhard Voelker writes:
> That is 2x bash syntax, and at least the == is flagged by:
>
> maint.mk: use "test x = x", not "test x == x"
> make: *** [maint.mk:1218: sc_prohibit_test_double_equal] Error 1
Oops, right.
> Better use expr(1) instead of $(( )).
I assumed $(( )) was fi
www.bipm.org/en/measurement-units/si-prefixes
[4]
https://web.archive.org/web/20190325185816/http://www.bipm.org/en/publications/si-brochure/chapter3.html
[5] https://en.wikipedia.org/wiki/Link_rot
>From a8645e387147823e82bb08827f2dc75068134825 Mon Sep 17 00:00:00 2001
Message-ID:
From: Collin Funk
e7ddf60 Mon Sep 17 00:00:00 2001
Message-ID: <9cb89ff084ef4414cacd78bc2ee579252e7ddf60.1751508571.git.collin.fu...@gmail.com>
From: Collin Funk
Date: Wed, 2 Jul 2025 18:41:53 -0700
Subject: [PATCH 1/2] doc: remove outdated TODO/FIXME items
* TODO: Remove note about missing documentation for 'pin
04f898e9 Mon Sep 17 00:00:00 2001
Message-ID: <1dc6de03bb8cd6b1a7f313ec81ee935004f898e9.1751598116.git.collin.fu...@gmail.com>
From: Collin Funk
Date: Thu, 3 Jul 2025 19:48:44 -0700
Subject: [PATCH] maint: prefer endian.h macros to WORDS_BIGENDIAN
* src/od.c: Include endian.h.
(WORDS_BIGE
Pádraig Brady writes:
> I think it's better to link to the sites directly when possible.
> The top level link [2] presents almost the same info as [1]
> so your patch is best.
Thanks for checking. I pushed the patch after rebasing [1].
Collin
[1]
https://git.savannah.gnu.org/gitweb/?p=coreuti
Pádraig Brady writes:
> +1 on the od changesince blake2 is copied in we try to minimize changes to
> that source.
I figured the blake2 changes were okay since the original lines were not
from upstream.
I guess we can keep it though since it my changes would add an extra
line to the diffs.
Pus
Rob Landley writes:
> Do you also need a specific tool to check if a file is zero length?
> Because "touch file; [ -f file -a 0 -eq "$(stat -c%s file)" ] && echo
> boom" is a similar level of complexity, since merely going stat -c%s
> file doesn't prove that file exits and isn't a directory...
t
Pádraig Brady writes:
> Cool. I see no more instances matched by the following at least:
> grep -E '^[^@#].{5,} "[[:alpha:]]{1,}"' doc/coreutils.texi
> That's a bit too fuzzy match though to put in a syntax check I think.
Yep, I think it would be too much of a pain to write a syntax check.
Sin
enough to commit right away.
Collin
>From 1535ac272e20c6f2deb84e037d95a4148515d566 Mon Sep 17 00:00:00 2001
Message-ID: <1535ac272e20c6f2deb84e037d95a4148515d566.1751338609.git.collin.fu...@gmail.com>
From: Collin Funk
Date: Mon, 30 Jun 2025 19:50:58 -0700
Subject: [PATCH] doc: use ``
:00 2001
Message-ID: <2a08ab2bd538aed5144522a45eb168174288a334.1751087822.git.collin.fu...@gmail.com>
From: Collin Funk
Date: Fri, 27 Jun 2025 22:15:28 -0700
Subject: [PATCH] doc: list signals added by POSIX.1-2024
* doc/coreutils.texi (Signal specifications): Mention that POSIX.1-2024
added
Hi Bruno,
Bruno Haible via GNU coreutils General Discussion
writes:
> +for specifiying your general locale, and
s/specifiying/specifying/
> +For more information on how to set your locale, see the GNU gettext manual at
> +@ifinfo
> +@ref{Users,,The User's View,gettext}.
> +@end ifinfo
> +@ifno
riting the others until I get input on that. They
shouldn't be too difficult.
Collin
[1] https://lists.gnu.org/archive/html/bug-gnulib/2025-07/msg00102.html
>From 75389decab580d913abac8ead9b257bfbf4d959e Mon Sep 17 00:00:00 2001
Message-ID: <75389decab580d913abac8ead9b257bfbf4d959e.17
for checking. I'll write tests for
the other two locales tomorrow.
Collin
>From 5ab05e5a36a45f6efe0118eb115746830cc1859c Mon Sep 17 00:00:00 2001
Message-ID: <5ab05e5a36a45f6efe0118eb115746830cc1859c.1753158930.git.collin.fu...@gmail.com>
From: Collin Funk
Date: Thu, 17 Jul 2025 1
Collin Funk writes:
>> thanks for working on this
>
> No problem, thanks to you and Berny for checking. I'll write tests for
> the other two locales tomorrow.
Done, two patches attached. I'll push them tomorrow unless there is any
issues. I ran 'make syntax-chec
Hi Sylvestre,
Sylvestre Ledru writes:
> I am wondering if this is an undocumented behavior or a bug in du:
>
> $ mkdir dir
>
> $ /usr/bin/du dir dir dir dir
> 0 dir
>
> $ /usr/bin/du -l dir dir dir dir
> 0 dir
> 0 dir
> 0 dir
> 0 dir
>
> And -l means "count sizes many times if h
rftime
changes that implement this feature, and to add a NEWS entry for the
improvement.
Collin
>From 1b98590037c8abdaa9c8cc7e149a990d22e9d422 Mon Sep 17 00:00:00 2001
Message-ID: <1b98590037c8abdaa9c8cc7e149a990d22e9d422.1753135775.git.collin.fu...@gmail.com>
From: Collin Funk
Date: M
9e3a9d057cae06ecf1af3189 Mon Sep 17 00:00:00 2001
Message-ID: <192bcefdc9f3a0b29e3a9d057cae06ecf1af3189.1753515191.git.collin.fu...@gmail.com>
From: Collin Funk
Date: Sat, 26 Jul 2025 00:30:51 -0700
Subject: [PATCH] tests: avoid a false failure on Cygwin
* tests/cksum/md5sum-bsd.sh:
/bootstrap: Regenerated.
I pushed the attached patch to update HACKING. The rest of the file
looks up-to-date.
Collin
[1] https://github.com/git/git/commit/0a53e9ddeaddad63ad106860237bbf53411d11a7
>From afd899cb47c450488d9f9d518c85105d91d5c587 Mon Sep 17 00:00:00 2001
Message-ID:
From: Coll
Bruno Haible writes:
>> +if ! echo '' > 'backslash\is\not\dir\sep'; then
>
> Negating the exit status of a command with ! is not portable [1].
Thanks for letting me know even though as Pádraig said gl_POSIX_SHELL
checks this. I was not aware of this limitation in some shells.
> Suggestion: Swap
>From 93304acc1063b760c3e3fe229bd972749d055533 Mon Sep 17 00:00:00 2001
Message-ID: <93304acc1063b760c3e3fe229bd972749d055533.1753555163.git.collin.fu...@gmail.com>
From: Collin Funk
Date: Sat, 26 Jul 2025 11:29:02 -0700
Subject: [PATCH] tests: avoid a false failure on Cygwin
* tests/cksum
0:00:00 2001
Message-ID: <49a3697d1fbd08f3ec826b074982dd01965f5b71.1753575177.git.collin.fu...@gmail.com>
From: Collin Funk
Date: Sat, 26 Jul 2025 17:10:01 -0700
Subject: [PATCH] doc: emit better output for --option=[+]count in man pages
* man/help2man: Check for the '+' character like we
Brendan O'Dea writes:
> On Sat, Jul 26, 2025 at 05:20:35PM -0700, Collin Funk wrote:
>>In the Coreutils man pages we have occurrences of:
>>
>>--option=[+]count
>>
> [...]
>>The attached patch fixes it. WDYT?
>
> Looks good, I'll ad
Pádraig Brady writes:
> * cfg.mk (sc_standard_outputs): A new syntax check to
> enforce standard references.
> * doc/coreutils.texi: s/stderr/standard error/ etc.
> * src/date.c: Likewise.
> * src/dd.c: Likewise.
> * src/env.c: Likewise.
> * src/sort.c: Likewise.
> * src/split.c: Likewise.
> * sr
Grisha Levit writes:
> On macOS, using a default configure (without --with-libintl-prefix) causes
> link errors due to missing frameworks used by localename-unsafe:
>
> $ ./configure
> ...
>
> $ grep -w -e LIBINTL -e USE_NLS config.status
> S["USE_NLS"]="no"
> S["LIBINTL"]=""
a50 Mon Sep 17 00:00:00 2001
Message-ID: <72d483dab7cabcf4b18fbfa7429edc96967a4a50.1753918934.git.collin.fu...@gmail.com>
From: Collin Funk
Date: Wed, 30 Jul 2025 16:33:24 -0700
Subject: [PATCH] build: don't use '$<' in non-suffix rules
* src/local.mk (src/speedli
Pádraig Brady writes:
> Hi Bruno,
>
> Your two date patches look good.
> I'm replying directly since I can only see the messages in the archive,
> but they've not been delivered to me as of yet.
>
> (Generally GNU infra has been problematic over the last few months.
> Apparently this is due to on
Bruno Haible via GNU coreutils General Discussion
writes:
> The source code src/date.c is written mostly bottom-up, a style which does
> not need forward declarations. Here's a patch to make it entirely bottom-up,
> removing the last needed forward declarations.
Yep, looks good in this simple ca
Bruno Haible writes:
> Collin Funk wrote:
>> This is because of the following change in one of the patches:
>>
>> diff --git a/gnulib b/gnulib
>> index 84ddfc7bd..caf768863 16
>> --- a/gnulib
>> +++ b/gnulib
>&
66 matches
Mail list logo
