Hi! I have noticed self-check failures of libidn on ArchLinux:
https://gitlab.com/libidn/libidn/-/jobs/932649011
The gnulib self-tests that fail are: test-binary-io.sh test-perror.sh
test-init.sh
Bruno Haible writes:
> Hi Simon,
>
>> Essentialy it boils down to:
>>
>> ../../gltests/test-binary-io.sh: line 8: cmp: command not found
>> test-init.sh: failed test: err not empty: ../../gltests/init.sh:
>> line 648: cmp: command not found
>> ../../gltests/test-perror.sh: line 14: diff:
Hi. I'm setting GNU SASL up for git-version-gen and bootstrap, and
noticed that bootstrap doesn't handle git-version-gen well. See:
jas@latte:~$ cat>configure.ac
AC_INIT([GNU SASL],
m4_esyscmd([build-aux/git-version-gen .tarball-version]),
[bug-gs...@gnu.org],
[gsasl])
lör 2020-12-26 klockan 11:12 -0800 skrev Paul Eggert:
> On 12/26/20 10:49 AM, Bruno Haible wrote:
> > > It would be nice to reduce that set of tools.
> > Well, both 'cmp' and 'diff' are part of POSIX:
>
> Even awk is on that list, and it's much fancier than cmp or diff. And
> I wouldn't support
Hello. I noticed the test-parse-datetime fails on Alpine Linux, and
probably has failed since 2017-04-26 when the "Outlandishly-long time
zone abbreviations" test was added. (It could also be a recent
regression, of course.) I don't have cycles to work on that particular
bug, but I'm wondering
Bruno Haible writes:
> Hi Simon,
>
>> I'm wondering if there is any CI/CD testing of gnulib for
>> different platforms?
>
> There is some automated CI testing done one gitlab.com. Tim, can you please
> answer my question from [1]?
Thanks -- indeed doing this on GitLab would be easy, and I'm
Bruno Haible writes:
> A testdir created by this command:
>
> ./gnulib-tool --create-testdir --dir=../testdir --single-configure clean-temp
> crypto/gc getaddrinfo getlogin getlogin_r gettimeofday isatty link localename
> opendir physmem poll putenv read readdir relocatable-prog rename
Hi. I am experimenting with using bootstrap+gnulib in a project that
has a library and a command-line tool, and thus needs two gnulib
instances. The limitation that --with-tests cannot be used in this case
is helpful to repeat in the relevant section, it caused hard to
understand build errors.
Bruno Haible writes:
>> Right. It would be nice if gnulib's valgrind m4 test was a bit
>> smarter, maybe it should try to build a small program printing
>> something to stdout and see if valgrind is able to run it and that it
>> prints the magic string to stdout. If that doesn't work, disable
Hi. I noticed that 'make syntax-check' never ran syntax checks whose
name contained 'w' on my local machine, but did so on CICD builds. I
tracked it down due to this:
jas@latte:~/src/gsasl$ echo $LANG
sv_SE.UTF-8
jas@latte:~/src/gsasl$ echo foowbar: | sed -n 's/^\([a-z]*\):/\1/p'
tis 2021-01-05 klockan 12:52 -0800 skrev Jim Meyering:
> On Mon, Jan 4, 2021 at 2:44 PM Jim Meyering wrote:
> > On Mon, Jan 4, 2021 at 2:22 PM Simon Josefsson via Gnulib
> > discussion
> > list wrote:
> > > Hi. I noticed that 'make syntax-check' never ran syntax
Paul Eggert writes:
> On 1/6/21 1:55 PM, Simon Josefsson via Gnulib discussion list wrote:
>
>> I'm using the patch below in gsasl, what do you think?
>
> Thanks, looks good to me.
>
> On 1/6/21 3:54 PM, Bernhard Voelker wrote:
>
>> And instead of adding y
I have pushed this, and I also updated the link from
https://www.gnu.org/software/gnulib/
/Simon
* build-aux/bootstrap (default_gnulib_url): Ditto.
* config/srclistvars.sh: Ditto.
* doc/gnulib-readme.texi (Git Checkout): Ditto.
* doc/gnulib-tool.texi (VCS Issues): Ditto.
* top/gitsub.sh: Update
I had a walk and realized it might be better to think of the problem
like this. Consider if someone wants to volunteer to do a new gettext
release, they would go to
https://savannah.gnu.org/git/?group=gettext
which properly suggest to checkout over https or SSH. After reading
HACKING the
Bruno Haible writes:
> Hi Simon,
>
> Can you briefly say, why, please? Is the 'git' protocol unsecure?
> Is it a problem specifically with Savannah? Or what else?
Sorry I should have included this -- I thought it was well-known.
The man page for git-clone https://git-scm.com/docs/git-clone
fre 2021-01-15 klockan 09:55 +0100 skrev Bruno Haible:
> It is an interesting idea. Leaving the question aside how it is
> implemented
> (through an AC_SUBSTed variable or what else), the main question is:
> Would
> some GNU package maintainers want this?
>
> I always thought that GNU package
Hi. I have pushed this, see background here:
https://lists.gnutls.org/pipermail/gnutls-devel/2016-October/008183.html
https://savannah.gnu.org/support/?107874
/Simon
From a8bac4d4940cd7c7eeef85f3c618a78a274f0cb8 Mon Sep 17 00:00:00 2001
From: Simon Josefsson
Date: Wed, 13 Jan 2021 14:38:20
Bruno Haible writes:
> The other reason is that every package maintainer has their preferred set of
> warnings — that's what the 'manywarnings' module is made for —, but it does
> not make sense for package maintainers to enforce the absence of certain
> warnings on code that 1) they don't
>> I think there is room for improvements in this field
>
> From what you are telling, the first improvement should be to clarify
> where to report failing tests. Done as follows:
Great! I think this will help, reporters rarely know anything about
the project and just report a build failure,
Bruno Haible writes:
> Alpine Linux does not have the 'join' program.
> The GCS [1] don't list it among the essential utilities.
I ran into that problem during ./bootstrap -- it seems gnulib-tool
relies on the 'join' tool as well:
./bootstrap: gnulib/gnulib-tool--no-changelog
Hi. I have pushed the attached patch. It should be completely
backwards compatible, or there is a bug, so please test this. It adds
two new variables to allow both developers and users to influence the
valgrind options used. You may want to re-read the manual, I updated it
to assume people use
Further testing quickly found a small bug, pushed.
/Simon
From 784fdea59920d69998b59ab326c11a8a2a93ef88 Mon Sep 17 00:00:00 2001
From: Simon Josefsson
Date: Fri, 14 May 2021 15:03:25 +0200
Subject: [PATCH] valgrind-tests: Fix LOG_VALGRIND when valgrind is missing.
* m4/valgrind-tests.m4: Clear
Bruno Haible writes:
> Simon Josefsson wrote:
>> Was anyone working on setting up CI/CD for gnulib on GitLab?
>> I recall there was a private project for it, but I don't have access.
>> Any reason for this? I have become rather aquinted with GitLab CI/CD
>> lately so it would be easy for me to
Bruce Korb writes:
> On 5/15/21 1:01 AM, Simon Josefsson via Gnulib discussion list wrote:
>> However, running ./bootstrap remains painfully slow.
> Amen!!!
>> Then a second run of ./bootstrap in the directory would avoid the
>> gnulib-tool step, and only run autore
I implemented the tarcache idea:
https://git.savannah.gnu.org/cgit/libidn.git/commit/?id=9ae53e866a6fafa56db26d184ccae9c39dae7446
It is opt-in for any developer, instead of normal ./bootstrap you do:
env GNULIB_BOOTSTRAP_CACHE=/home/jas/.local/gnulib-bootstrap-cache ./bootstrap
...
The first
Bruno Haible writes:
> That is not proper behaviour. Fixing it through the patch below.
Thank you!
It would be nice to write a 'join' replacement for gnulib-tool, as that
is the only thing that needs coreutils for bootstrapping libidn2 on
alpine. But it is not important, and with your patch
Hi! I just made a new release of libidn2, and after posting the
announcement I noticed that hash checksums was missing from my
announcement. It appears maint.mk's announce-gen rules always were like
this, I just didn't use them before. May I suggest the patch below as a
starting point, to at
Bernhard Voelker writes:
> On 5/12/21 7:58 PM, Simon Josefsson via Gnulib discussion list wrote:
>> It would be nice to write a 'join' replacement for gnulib-tool, as that
>> is the only thing that needs coreutils for bootstrapping libidn2 on
>> alpine. But it is not imp
Hi. There are a couple of different ways to use gnulib in projects:
https://www.gnu.org/software/gnulib/manual/html_node/VCS-Issues.html
At some point in some project I've used all variants, but today most of
my projects are converted into not commiting any gnulib files at all and
use gnulib's
Bruno Haible writes:
> Hi Simon,
>
>> +Replace @code{LOG_COMPILER} with @code{TESTS_ENVIRONMENT} if you are
>> +using the old serial test harness.
>
> Assuming I am a GNU package maintainer and I read this paragraph. How do
> I know whether my package uses the "old serial test harness"? Can I
>
Bruno Haible writes:
>> However I think these lists often become outdated. In my view, to claim
>> that a platform is supported by a software project, you should have
>> continous building for the platform. Otherwise support is reactive and
>> tends to be spurious.
>
> Gnulib is a hobbyist,
Eric Blake writes:
> It was pointed out to me that for multiple years, m4's
> 'm4-latest.tar.XX' links were out of date at https://ftp.gnu.org/gnu/m4/
> - the links were created in in 2013 when I released m4-1.4.17, but never
> updated in 2016 when I released m4-1.4.18.
>
> I managed to fix the
arn...@skeeve.com writes:
>> At this point I wouldn't worry about the older clang and gcc versions
>> that complain about {0} as an initializer. We can either let them die
>> off noisily, or use the appropriate -Wno-whatever option when using them
>> to compile.
>
> I've decided to just not
Bruno Haible writes:
> Maxim Cournoyer wrote:
>> I'm discovering this project and attempted to run 'make check'. It
>> seems to be finding problems, failing with:
>>
>> --8<---cut here---start->8---
>> [...]
>> GPL?? lib/xstrtoull.c
Eric Blake writes:
> Speaking of tools, should we include SPDX tags alongside the full text
> of all our licenses, as that is yet another thing that aids
> license-checking tools?
>
> https://spdx.github.io/spdx-spec/appendix-V-using-SPDX-short-identifiers-in-source-files/
I'm not a big fan of
Marc Nieper-Wißkirchen writes:
> Moreover, use cases for a baked-in timeout are not restricted to tests. For
> example, I may want to restrict the build time of certain components in
> situations where a logical error may lead to infinite build times (a simple
> example is that of a Scheme
Bruno Haible writes:
> So, I don't think the "let's treat timeout like valgrind" approach is going
> to work. Instead, you need to design a way to deal with timeouts,
> independently.
Hi! I think Marc's request for functionality to introduce timeouts for
self-tests is a good one. However I
severity 48113 wishlist
retitle 48113 Self-test timeout functionality
thanks
Karl Berry writes:
> What do bug-automake people think?
>
> For myself, I have no objection to sprinkling timeout commands through
> the Automake test infrastructure wherever appropriate. It's not ever
> going to
Hi! I believe it is used by Shishi —
https://www.gnu.org/software/shishi/manual/html_node/Utility-Functions.html —
however I am sure it can be worked around by including the gnulib
implementation directly. What problems does it cause to keep around? It is
rather small.
/Simon
> 2 apr. 2021
Paul Eggert writes:
> On 4/2/21 8:32 AM, Paul Eggert wrote:
>> I can fix it by removing the divergence.
>
> I did that by installing the attached, which causes xgetdomainname.c
> to be a couple of #defines followed by "#include "xgethostname.c".
Thank you!
/Simon
signature.asc
Description:
Paul Eggert writes:
> -extern void base64_encode (const char *restrict in, size_t inlen,
> - char *restrict out, size_t outlen);
> +extern void base64_encode (const char *restrict in, idx_t inlen,
> + char *restrict out, idx_t outlen);
Thanks
Bruno Haible writes:
> Hi Simon,
>
>> I'm not sure I understand why idx_t is better than size_t
>> here, can you elaborate? Why not ssize_t?
>
> You find a detailed explanation in the comments of idx.h.
Thanks for the pointer -- it doesn't say anything about why ssize_t
can't be used though?
Hi. I realized the main reason I don't consistenly enforce code
indentation in the projects I work on is that it isn't tested before
releases (or via CICD builds). The syntax-check framework to the
rescue! Not everyone will like this, but as usual it is simple to
disable on a per-project basis,
Hi. I followed doc/README to update
https://www.gnu.org/software/gnulib/manual/ -- let me know if anything
looks odd. Eventually it would be nice to use gnu-web-doc-update for
this.
/Simon
signature.asc
Description: PGP signature
Hi. This doc change encourages consistent use of FDL in manuals;
several packages I looked at (e.g., coreutils) already follow this
pattern.
/Simon
From 6b8032a320a75053c6607fcd37d2d67ffe61fd77 Mon Sep 17 00:00:00 2001
From: Simon Josefsson
Date: Sat, 4 Sep 2021 09:19:37 +0200
Subject: [PATCH]
Hi. It seems a lot of projects that use gnulib is using our bootstrap
script, but documentation is fairly hidden in the manual. This is just
a small improvement to the current section.
/Simon
* doc/gnulib-tool.texi (VCS Issues): Make title more positive,
s/Issues/Integration/. Some more
Bruno Haible writes:
> What exactly is the problem that you were seeing? (Commands to reproduce,
> and compiler output, please.)
I haven't had time to debug it, but since a few days libidn fails to
build with a reference to rpl_free, that looks related to this thread
and the recent gnulib
Hi,
Some syntax-check rules have a really poor failure mode (it prints all
source code files to stdout) when the 'gnulib_dir' points to an empty
git submodule checkout, see for example:
Hi. I'm trying to debug a regexp segfault [1] in latest inetutils
release, but the following fails:
jas@latte:~/src/gnulib$ ./gnulib-tool --create-testdir --dir m regex
...
executing autoheader
executing touch config.h.in
executing automake --add-missing --copy
parallel-tests: installing
Jim Meyering writes:
> On Wed, Jul 28, 2021 at 1:08 AM Simon Josefsson via Gnulib discussion
> list wrote:
>> Hi. I replaced GNU InetUtils' old custom fts implementation with the
>> one from gnulib, but self-tests started failing. Looking at the code,
>> it seems gnu
Jim Meyering writes:
> Feel free to make the script generate a full fingerprint and even
> (though it feels a little like giving up) add a checksum or two.
I think checksums still serve a purpose.
Many announcement e-mails are OpenPGP signed (and sometimes with a
different key than the release
sön 2021-08-01 klockan 17:47 +0200 skrev Bernhard Voelker:
> On 7/27/21 11:38 AM, Simon Josefsson via Gnulib discussion list
> wrote:
> > Let's discuss and see what we can do.
> Isn't this what the "release GPG keys" on Savannah are for?
>
> Each project maintainer
tis 2021-08-03 klockan 16:51 -0700 skrev Jim Meyering:
> On Tue, Aug 3, 2021 at 12:25 PM Paul Eggert
> wrote:
> > On 8/3/21 12:20 PM, Simon Josefsson via Gnulib discussion list
> > wrote:
> > > + print "\nThe SHA256 checksum is base64 encoded and not
> > &g
Jim Meyering writes:
> Thanks, Simon! I too am all for B64-formatted checksums.
Good, it is a trade-off between output readability and code complexity.
Aligning 'sha*sum' with OpenBSD's 'sha*' tools would be nice, and base64
support is one missing piece.
> You may want to coordinate with
mån 2021-09-20 klockan 07:36 -0700 skrev Jim Meyering:
> On Mon, Sep 20, 2021 at 12:49 AM Simon Josefsson via Gnulib
> discussion
> list wrote:
> > Hi,
> >
> > Some syntax-check rules have a really poor failure mode (it prints
> > all
> > source code files
Bruno Haible writes:
> Thanks for the heads-up, Simon. I debugged it, and indeed the cause is
> in gnulib, not in libidn. The patch below fixes it.
Thanks for debugging this Bruno!
/Simon
signature.asc
Description: PGP signature
Paul Eggert writes:
> Good question. Addressed in the attached patch, which I pushed.
Thanks for improving this -- it addresses my concerns.
/Simon
signature.asc
Description: PGP signature
Hi. This may mostly be for Bruno, but I believe it is more relevent to
gnulib than gettext, even though it is gettext-related, and maybe others
on this list can provide feedback too.
I got a bug report that suggested using AM_GNU_GETTEXT_REQUIRE_VERSION:
Bruno Haible writes:
>> > Also, I would expect that the GNULIB_REVISION environment variable is not
>> > so frequently used, and that the more frequent use is with git submodules.
>> > Is it possible to apply the same trick to the git submodules case? Or does
>> > the use of submodules always
Robbie Harwood writes:
> The gnulib module makes use of booleans via the header. As
> GRUB does not provide any POSIX wrapper header for this, but instead
> implements support for bool in , we need to patch
> base64.h to not use anymore. We unfortunately cannot include
> instead, as it would
Robbie Harwood writes:
> So I think our way forward is to move where we nerf _GL_ATTRIBUTE_CONST
> in grub2. I've tested that this works and will submit to grub2.
>
> Longer-term, this problem could be avoided by dropping the const
> attribute from isbase64(). Since uchar_in_range is a macro,
Hi! I noticed that the HTML output of gendocs.sh did not look the same
locally as it does on www.gnu.org, and it was because gendocs.sh
hard-coded a relative URL. I have pushed this patch to use an absolute
URL instead.
/Simon
From c567dcac24dd90a2be051772d9a8c8bbf869221a Mon Sep 17 00:00:00
Hi. Our announce-gen contains:
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 $gpg_key_id
Given recent OpenPGP key server issues, that doesn't work reliably any
more, and behave different
Hi. I noticed some typos causing the Cc line in mail announcements to
be empy where that (most likely) wasn't the intention.
/Simon
* top/maint.mk (announcement_Cc_stable): Rename to
announcement_mail_Cc_stable.
(announcement_Cc_alpha): Rename to announcement_mail_Cc_alpha.
Hi. I replaced GNU InetUtils' old custom fts implementation with the
one from gnulib, but self-tests started failing. Looking at the code,
it seems gnulib's fts implementation has diverged compared to glibc, and
has some optimizations that (I think) change the API (wrt stat and
chdir). Also,
Simon Josefsson via Gnulib discussion list writes:
> Hi. I noticed some typos causing the Cc line in mail announcements to
> be empy where that (most likely) wasn't the intention.
It seems some packages (coreutils, grep, ...) set 'announcement_Cc_' in
cfg.mk manually, thus hiding the pro
"Darshit Shah" writes:
> I don't have push access to gnulib, so could you please push it for me?
I ended up pushing the patch below. There are several considerations
here:
* We want the OpenPGP Key identifier to be mentioned in the e-mail, it
provides a strong hash-based coupling between
Thanks for quick debugging and conclusion!
Reading the ideas in your responses, I think gnulib could help
developers to use ASan/UBSan in their project by assisting with these
choices. I'll see if I can come up with anything that is generally
useful, once I get a couple of projects to build and
Darshit Shah writes:
> + --gpg-keyring-url=URLURL pointing to the GnuPG Keyring containing
> +the key used to sign the tarballs
...
> If that command fails because you don't have the required public key,
> then run this command to import it:
>
> -
sön 2022-02-20 klockan 21:08 +0100 skrev Bruno Haible:
> Mike Frysinger wrote:
> > your automake version is too old. %reldir% was introduced in 1.14.
> > from your log:
> > > ---> Package automake.noarch 0:1.13.4-3.el7 will be installed
>
> Thanks for the analysis. Let me document it (patch
Bruno Haible writes:
> Simon Josefsson wrote:
>> diff --git a/build-aux/bootstrap b/build-aux/bootstrap
>> index dc2238f4ad..771ef65d22 100755
>> --- a/build-aux/bootstrap
>> +++ b/build-aux/bootstrap
>> @@ -675,7 +675,7 @@ fi
>> # If either is not listed, add it (with minimum version) as a
>>
Bruno Haible writes:
> Hi Simon,
>
>> > Can 'bootstrap' be used without gnulib-tool? I would guess yes.
>>
>> Good point. How about this instead?
>
> This is good. Please push it.
Done.
We have a lot of bug workaround code for automake < 1.14 behaviour that
could be dropped. But let's wait
Hi. I encountered the following runtime error when running libidn2
self-tests (log in [1]) with ASAN/UBSAN checks [2] and recent gnulib.
uninorm/composition.c:75:22: runtime error: implicit conversion from type
'unsigned int' of value 218 (32-bit, unsigned) to type 'char' changed the value
to
Bruno Haible writes:
> I could rewrite this line as
>
> codes[2] = (char) (unsigned char) (uc1 & 0xff);
Here is the slightly smaller local patch I am using:
https://gitlab.com/libidn/libidn2/-/commit/d792d91eb7197691cd3d2878b110c4d5ce58b2a9#7257b9bace195314d2201030e03b7ee2b83e4c83
Hi. I updated gnulib in libidn2, and it fails on CentOS7 like this:
mv string.h-t string.h
/usr/bin/mkdir -p '%reldir%/sys'
sed -e 1h -e '1s,.*,/* DO NOT EDIT! GENERATED AUTOMATICALLY! */,' -e 1G \
-e 's|@''GUARD_PREFIX''@|GL|g' \
-e 's|@''INCLUDE_NEXT''@|include_next|g' \
...
Bruno Haible writes:
> * Commit b93de66735cd6f935ee0970f8cb26908d113e09d introduced mcel.h, but
> it has tabs. Can we untabify
> mcel.h
> mountlist.c
> verify.h
> (as we do with all source files that are not shared with glibc)?
We may have discussed this before, but what do you
Bruno Haible writes:
> -#define LIBFOO_DLL_EXPORTED __attribute__((__visibility__("default")))
> -#elif (defined _WIN32 && !defined __CYGWIN__) && BUILDING_SHARED &&
> BUILDING_LIBFOO
> -#define LIBFOO_DLL_EXPORTED __declspec(dllexport)
> -#elif (defined _WIN32 && !defined __CYGWIN__) &&
Pádraig Brady writes:
> However if there are good use-cases for bad inputs
> we may need to adjust this patch,
> rather than failing unconditionally.
>
> For example we could just flag non canonical input in the context,
> and leave it up to the caller how to deal with that.
That adds
Pádraig Brady writes:
> To give a little more context, this will avoid
> round trip issues like the following, by failing early:
>
> $ echo "HelloWorld==" | base64 -d | base64
> HelloWorlQ==
Thanks for background and patches! There are use-cases for bad inputs
(both for good and malicious
Bruno Haible writes:
> Vivien Kraus wrote:
>> I don’t know how to do it at the moment; the
>> configure script automatically gets a --enable-shared=… flag but I
>> can’t find a trace of it in config.log, and I don’t know how to use it.
>> Could you elaborate on how to get a value for
Tim Rühsen writes:
> a) The maintainer/contributor/hacker setup
> This is when you re-run configure relatively often for the same project(s).
> I do this normally and and came up with
> https://gitlab.com/gnuwget/wget2/-/wikis/Developer-hints:-Increasing-speed-of-GNU-toolchain.
>
> It may be a
Hi. I got complaints that the announcement of a recent release did not
include my full GnuPG key fingerprint.
While the instructions on GnuPG verification is getting longer and
longer, having the full fingerprint in the email seems like the right
thing. The OpenPGP infrastructure is under
Bruno Haible writes:
> Hi Simon,
>
> Two remarks regarding 'announce-gen':
>
> 1) I used the command
>$ $GNULIB_SRCDIR/build-aux/announce-gen --release-type stable \
>--package-name libunistring --previous-version 0.9.10 \
>--current-version 1.0 --gpg-key-id F5BE8B267C6A406D
Reuben Thomas writes:
> Just noticed that it says: "Running without arguments will suffice in most
> cases." However, there is a mandatory argument!
Thank you! I pushed this fix.
/Simon
From be1b6d4bfaf1ad73780f30d6cd12cae3633b816a Mon Sep 17 00:00:00 2001
From: Simon Josefsson
Date: Sat, 9
Hi
In libidn and libidn2 we (used to) print this during ./configure:
Libiconv: $am_cv_func_iconv $LTLIBICONV
It prints the right thing in most cases, that is, one of these:
Libiconv: yes -liconv ...
Libiconv: no, consider installing GNU libiconv
However on
Bruno Haible writes:
> Simon Josefsson wrote:
>> + my $gpg_fingerprint = `LANG=C gpg --fingerprint $gpg_key_id | grep -v
>> ^sub`;
>
> LANG=C has no effect if LC_ALL is set, since LC_ALL has a higher precedence
> than
> LANG [1].
>
> [1]
>
Bruno Haible writes:
> The last line gives an overfull line in the PDF output, so let me add a line
> break.
>
> Also, there is some collision risk for a shell variable named 'shared'.
>
> I'm therefore applying this:
Thank you Bruno!
/Simon
signature.asc
Description: PGP signature
Bruno Haible writes:
>> I can't find any uses of am_func_iconv in any M4 files in gnulib, so I'm
>> hoping that nobody is testing it for == 'no' strings but only uses !=
>> 'yes' comparisons.
>
> It's too dangerous to change the value of this variable. This macro is
> in the wild for more than
Hi
I'm reading
https://www.gnu.org/software/gnulib/manual/html_node/gettextize-and-autopoint.html
that recommends using the 'gettext' module but it is marked obsolete:
This module is obsolete. Use the module 'gettext-h' instead to make
your program capable of internationalization, when
Bruno Haible writes:
> Hi Simon,
>
>> I'm reading
>>
>>
>> https://www.gnu.org/software/gnulib/manual/html_node/gettextize-and-autopoint.html
>>
>> that recommends using the 'gettext' module but it is marked obsolete:
>>
>> This module is obsolete. Use the module 'gettext-h' instead to
Bruno Haible writes:
> Therefore I would now like to actually do it.
Thanks for working on this, it is important!
Maybe I am getting old, but one year seems like a fairly short period of
time. The list would be shortened with the following names if we used
two years:
Daiki Ueno
Bruno Haible writes:
> Recent discussion in gnu-prog-discuss has shown that making the separation
> into two phases (1) and (2) explicit will have several benefits:
Wonderful, thank you! I have yet to digest everything, so I'll fall
back on stylistic comments:
> The first phase is a script
Bruno Haible writes:
> The handling of the GNULIB_SRCDIR environment variable in 'bootstrap'
> is hard to understand, despite the long documentation in `bootstrap --help`.
Thank you! I have encountered this a couple of times, and always had to
read the source code to learn how it worked, and
Bruno Haible writes:
> Since we are already stating in the generic INSTALL file, since 2008:
>
> On Solaris, don't put '/usr/ucb' early in your 'PATH'. This
> directory contains several dysfunctional programs; working variants of
> these programs are available in '/usr/bin'. So, if
I think some of these patches introduced a build failure of GNU
InetUtils on Debian 6, see build error below. The following looks
strange:
-if (INT_MULTIPLY_WRAPV (plen, sizeof (CHAR), ) \
-|| INT_ADD_WRAPV (new_used, plensize, _used)) \
+
Bruno Haible writes:
> On MSVC, with libunistring installed as a shared library, I get this link
> error:
>
> /home/bruno/msvc/compile cl -nologo -MD -L/usr/local/msvc64/lib -o
> test-categ_none.exe unictype/test-categ_none.obj libtests.a ../gllib/libgnu.a
> libtests.a ../gllib/libgnu.a
Hi. I have committed this.
/Simon
From 4b17a1ae49e69df1ac5dc35a4f60b20ab958baf2 Mon Sep 17 00:00:00 2001
From: Simon Josefsson
Date: Tue, 6 Sep 2022 14:32:05 +0200
Subject: [PATCH] gnumakefile: Improve tarball reproducibility.
* top/GNUmakefile (TAR_OPTIONS): Add --sort=name. Suggested by
Paul Eggert writes:
>> +if (ckd_mul (, plen, sizeof (CHAR))
>> \
>
> Indeed it should. Thanks for reporting that. I installed the attached.
Thank you!
>> I wonder why no self-check has caught this?
>
> Apparently I tested it only on platforms with working
Bruno Haible writes:
> In this case, you'll better modify the unit test to pipe the result
> through "tr -d '\r'".
This is unrelated, but alas I've not found a more portable way to trim
CR than this since some tr do not support \r:
if echo solaris | tr -d '\r' | grep solais > /dev/null; then
tor 2022-08-25 klockan 19:21 +0200 skrev Bruno Haible:
> Simon Josefsson wrote:
> > > In GNU gettext, many tests use "tr -d '\r'" since 2007 already,
> > > and no one
> > > ever has reported a problem with it.
> >
> > But does it fail fatally when it doesn't work? tests/parser.sh in
> > libtasn1
1 - 100 of 178 matches
Mail list logo
