Bruno Haible writes:
> Simon Josefsson wrote:
>> Please install as far as I'm concerned.
>
> Done. Thanks also for your fnmatch.m4 fix.
I also noticed there were no self-test for fnmatch. Pushed.
/Simon
>From 19b2a30859ba963627fb42987944efd35599dc75 Mon Sep 17 00:00:00 2001
From: Simon Josefs
On Sun, May 3, 2009 at 6:33 PM, Bruno Haible wrote:
>
> Can you please also provide an augmented unit test that exhibits the bug?
> (Additional code in tests/test-file-has-acl* ?)
>
I added a test case and made a few fixes for FreeBSD. I successfully
ran the acl tests on Linux and FreeBSD (I don'
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Jim Meyering on 5/3/2009 7:40 AM:
>> Can we please make this VPATH safe? $(srcdir)/.git
>>
>>> + diff=$$(git submodule -q foreach git diff-index --name-only HEAD) \
>> diff=$$(cd $(srcdir) && git submodule...
>
> Of course.
> Thanks!
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 5/3/2009 4:39 PM:
> We have the backported AC_DEFUN_ONCE also in 00gnulib.m4.
>
> It looks like it would be possible to simplify your commit from 2009-01-21
> now, like this. What do think?
>
>
> -dnl This macro must pa
David Bartley wrote:
> The acl_get_entry related ones fix the loop logic;
> before the loop would infinitely loop (observable on freebsd via 'ls
> -ldv foo' where foo has a default acl set).
Thanks for reporting this.
Can you please also provide an augmented unit test that exhibits the bug?
(Addi
David Bartley wrote:
> The lib/set-mode-acl.c fix is benign, as sizeof(ace_t) == sizeof(aclent_t).
A bad bug nevertheless. Thank you. I applied this fix.
Bruno
Ralf Wildenhues wrote:
> > No, that would require the VPATH build to write those files into the
> > source directory, which defeats the purpose of a VPATH build.
>
> It is no problem if a git tree updates files in the source tree.
> ...
> The GNU Coding Standards are pretty clear about this too.
Hi,
The following patch fixes a few bugs in the acl module. The
lib/set-mode-acl.c fix is benign, as sizeof(ace_t) ==
sizeof(aclent_t). The acl_get_entry related ones fix the loop logic;
before the loop would infinitely loop (observable on freebsd via 'ls
-ldv foo' where foo has a default acl set)
Stefan Bienert wrote:
> For testing I fetched the rand48 module into my project and stdlib.h
> contains a line
>
> # if addr...@hidden@
Oh, you took the patch off the mailing list archive :-(
I send you a copy of Richard's post in private email.
Bruno
Ludovic Courtès wrote:
> on alpha*-linux-gnu, `-mieee' is apparently not needed:
>
> --8<---cut here---start->8---
> #include
>
> int
> main (int argc, char *argv[])
> {
> return isnan (NAN);
> }
> --8<---cut here---end---
Hi,
Bruno Haible writes:
>> So people compiling with GCC on this platform should pass `-mieee'
>> (`configure' could add it automatically.)
>
> The gnulib module 'fpieee' adds it automatically.
This module appears to be overzealous: `gl_FP_IEEE' just checks for the
host CPU, but on alpha*-linux
Hi Eric,
After your introduction of AC_DEFUN_ONCE in autoconf, the documentation says:
either the macro must be declared with AC_DEFUN_ONCE (although
this only helps in Autoconf 2.64 or newer), or all
uses of that macro should be through AC_REQUIRE
We have the backported AC_DEFUN_ONCE also
Simon Josefsson wrote:
> Please install as far as I'm concerned.
Done. Thanks also for your fnmatch.m4 fix.
Bruno
Hello Ralf,
> when I try to
> gnulib-tool --with-tests --test
>
> with git Autoconf, I get this set of warnings:
>
> configure.ac:2159: warning: AC_REQUIRE: `gl_PREREQ_VASNPRINTF_LONG_DOUBLE'
> was expanded before it was required
> glm4/vasnprintf.m4:101: gl_PREREQ_VASNPRINTF_INFINITE_LONG_DO
Bruno Haible writes:
> We had a long-standing incompatibility between the modules fnmatch-gnu and
> fnmatch-posix: It was not possible to both in the same configure.ac. This
> fixes it.
It fails to build because config.h contains:
#define fnmatch ${gl_fnmatch_required_lowercase}_fnmatch
I've p
Bruno Haible writes:
> And here's a proposed simplification: Deprecate the 'fnmatch-posix'
> module, and let the 'fnmatch' module take over its role. (Like we do
> for the 'getpass' vs. 'getpass-gnu' modules.)
>
> Any objections?
Please install as far as I'm concerned. Fixing the glob module pr
Hi again,
OK, so everything in it's own file...
Question:
For testing I fetched the rand48 module into my project and stdlib.h
contains a line
# if addr...@hidden@
and gcc complains about it:
./stdlib.h:273:6: error: "address" is not defined
./stdlib.h:273:13: error: token "@" is not valid i
David Bartley wrote:
...
> Updated the gnulib patch accordingly.
>
>> Jim, I leave it to you to commit this for David, since I don't want to
>> interfere with your coreutils release.
Thanks to both of you.
I've just pushed that to gnulib, and the following to coreutils:
>From 6249eb0596dfe50e1f95
Eric Blake wrote:
> According to Jim Meyering on 5/2/2009 2:34 PM:
>> * top/maint.mk (no-submodule-changes): New rule.
>> (alpha beta major): Depend on it.
>> +.PHONY: no-submodule-changes
>> +no-submodule-changes:
>> +if test -d .git; then \
>
> Can we
Bruno Haible wrote:
>> * lib/propername.c (proper_name_utf8): Test not pointer equality,
>> but rather string equality.
>
> Thanks, applied with a comment, as below.
>
>> Without this, coreutils' "make check-AUTHORS" test fails.
>
> The check-AUTHORS rule in coreutils/src/Makefile.am also has a sec
According to Jim Meyering on 5/2/2009 2:34 PM:
> * top/maint.mk (no-submodule-changes): New rule.
> (alpha beta major): Depend on it.
> +.PHONY: no-submodule-changes
> +no-submodule-changes:
> + if test -d .git; then \
Can we please make this VPATH saf
Hi Jim,
> * lib/propername.c (proper_name_utf8): Test not pointer equality,
> but rather string equality.
Thanks, applied with a comment, as below.
> Without this, coreutils' "make check-AUTHORS" test fails.
The check-AUTHORS rule in coreutils/src/Makefile.am also has a second
problem: it expec
Jim Meyering wrote:
> How could it happen?
A possible way to avoid such mistakes is to "diff -r" subsequent snapshots.
I do this for the 2 or 3 last snapshots before a release, to avoid all kinds
of unintended changes: debugging statements, changes in automake version due
to a different PATH, zero
David Bartley wrote:
> Updated the gnulib patch accordingly.
Thanks. From my point of view, it's perfect now.
I leave it to Jim to decide when to commit it.
Bruno
Giuseppe Scrivano wrote:
> I fixed the regression you found and added some new tests.
...
>>From 0e1af7775a82aed00331a535ddadee2753d12e5e Mon Sep 17 00:00:00 2001
> From: Giuseppe Scrivano
> Date: Fri, 1 May 2009 15:44:30 +0200
> Subject: [PATCH]
> =?utf-8?q?*=20lib/getdate.y=20(get=5Fdate):=20Co
25 matches
Mail list logo
