On 19/04/2017 17:50, Daniel P. Berrange wrote:
>> But perhaps it's simpler to add "#undef recv" and "#undef select" near
>> the top of the file, in addition to this change.
>
> If we add the undef's then we don't need to change anything related to
> windows_compute_revents_socket() as it'd be
On 18/04/2017 17:07, Daniel P. Berrange wrote:
> We're seeing a failure in libvirt on Win32 platforms whereby poll() often
> returns POLLERR. I traced this down to the rpl_recv() function calling
> FD_TO_SOCKET() and getting INVALID_SOCKET back. This is propagated back
> to
On 19/11/2016 11:26, Bruno Haible wrote:
> Hi Paolo,
>
> For the 'frexpl-nolibm' module I also need your approval, since you
> contributed lib/frexpl.c on 2003-02-18.
>
> Would you agree to relicense your initial lib/frexpl.c
> under 'LGPLv3+ or GPLv2'?
> Or possibly even under LGPLv2+?
>
>
Il 18/07/2014 19:52, Oliver Schneider ha scritto:
Hi,
I happened to be working on an ancient Redhat system with GCC 2.96 and
got the following error:
gcc -DHAVE_CONFIG_H -I. -I.. -I../lib -I.. -I../lib -g -O2 -MT
regex.o -MD -MP -MF .deps/regex.Tpo -c -o regex.o regex.c
In file
Il 30/05/2013 02:59, Paul Eggert ha scritto:
On 05/26/2013 03:11 PM, Paolo Bonzini wrote:
Use the lock module instead.
Adding the lock module should work. But this will require
some reengineering of Guile, so that Guile uses the lock module
rather than its own thread packaging. Another
Il 20/05/2013 16:20, Ludovic Courtès ha scritto:
Paul Eggert egg...@cs.ucla.edu skribis:
On 05/14/2013 02:21 PM, Ludovic Courtès wrote:
How should that be fixed? Shouldn’t __libc_lock_unlock co. be rebased
on top of pthread_mutex_t?
Yes, thanks, that sounds right. I pushed the
Il 24/04/2013 14:18, Pádraig Brady ha scritto:
On 04/23/2013 11:16 AM, Dmitry V. Levin wrote:
On Tue, Apr 23, 2013 at 10:54:03AM +0100, Pádraig Brady wrote:
I'm still thinking the gnulib patch should be applied though,
Since the regression in glibc was detected by this gnulib test,
I'm not
Il 20/02/2013 20:58, Andy Wingo ha scritto:
if (full_read (fd, cookie, sizeof cookie) != sizeof cookie
|| full_read (fd, SCM_BYTEVECTOR_CONTENTS (bv),
SCM_BYTEVECTOR_LENGTH (bv)) != SCM_BYTEVECTOR_LENGTH
(bv))
{
int errno_save = errno;
Il 21/02/2013 17:47, Andy Wingo ha scritto:
Doesn't it need to set errno to zero before calling full_read?
Not sure! If full_read requires that errno be 0 going in, then yes, we
should. Is this a requirement?
Hmm, I think usually errno is undefined for return values other than -1,
so
atanl:Paolo Bonzini
bind:Paolo Bonzini, Simon Josefsson
close:Paolo Bonzini, Simon Josefsson, Bruno Haible
connect:Paolo Bonzini, Simon Josefsson
cosl:Paolo Bonzini
frexp:Bruno Haible, Paolo Bonzini
frexp-nolibm:Bruno Haible, Paolo Bonzini
frexpl:Bruno Haible, Paolo Bonzini
frexpl-nolibm:Bruno
Il 21/12/2012 01:02, Michael Goffioul ha scritto:
Hi,
We've got a bug report in octave [1] that seems to indicate that
gnulib's replacement of isatty is incorrect on Windows 8 (it's fine up
to Windows 7). Looking at the implementation, it first calls the Windows
_isatty version, then checks
Il 17/10/2012 17:46, Eric Blake ha scritto:
On 10/17/2012 09:35 AM, Richard W.M. Jones wrote:
libguestfs (an LGPLv2+ library) would like to use the useful quotearg
functions to ensure we can run system(3)-style external commands while
safely quoting shell arguments.
However this would
Il 12/09/2012 19:02, Joachim Schmitz ha scritto:
Which Monday? ;-)
September 31st. Doing it now, sorry.
Paolo
Il 07/09/2012 09:39, Joachim Schmitz ha scritto:
I suppose it works to always handle ENOTSOCK that way, even on
non-__TANDEM systems.
Will you be fixing this in gnulib? How?
I don't have access to the system, so it's best if you post the patches
yourself to bug-gnulib and git mailing lists
Il 07/09/2012 12:58, Joachim Schmitz ha scritto:
From: Paolo Bonzini [mailto:paolo.bonz...@gmail.com] On Behalf Of Paolo
Bonzini
Sent: Friday, September 07, 2012 11:41 AM
To: Joachim Schmitz
Cc: g...@vger.kernel.org; 'Junio C Hamano'; 'Erik Faye-Lund';
bug-gnulib@gnu.org; rsbec
Il 07/09/2012 17:01, Joachim Schmitz ha scritto:
From: Paolo Bonzini [mailto:paolo.bonz...@gmail.com] On Behalf Of Paolo
Bonzini
Sent: Friday, September 07, 2012 4:47 PM
To: Joachim Schmitz
Cc: bug-gnulib@gnu.org; rsbec...@nexbridge.com
Subject: Re: poll() emulation in git/gnulib
Il 07/09
Il 06/09/2012 16:02, Joachim Schmitz ha scritto:
But is there something that could be done to make git work even without
poll()?
It is used in 5 places:
$ grep -n poll\( *.c */*.c
credential-cache--daemon.c:175: if (poll(pfd, 1, 1000 * wakeup) 0) {
daemon.c:1018: if (poll(pfd,
Il 06/09/2012 16:44, Joachim Schmitz ha scritto:
Yes, it's an usleep(autocorrect * 10) basically (poll takes
milliseconds, not micro).
OK, it is _supposed_ to do this usleep(), but is does not, as poll() returns
early with EFAULT in this case:
/* EFAULT is not necessary to implement,
Il 05/09/2012 13:24, Joachim Schmitz ha scritto:
However: this poll implementation, while compiling OK, doesn't work properly.
Because it uses recv(...,MSG_PEEK), it works on sockets only (returns
ENOTSOCK on anything else), while the real poll() works on all
kind if file descriptors, at
Il 05/09/2012 15:36, Joachim Schmitz ha scritto:
Does your system have a working FIONREAD ioctl for pipes?
It does have FIONREAD ioctl. Whether it works properly is to be
determined...
I'll test if you could show me how?
Oh, now I see what you aimed at, but no, that Mac OS X method
On the other hand, so far I saw no reply to my attempts to refute
counterarguments
against these patches. So, should I just submit git patches for one (or both)
of them,
for inclusion? Or does anybody still have reservations about this?
I'm waiting for feedback from the Gnulib guys. This
On Sun, Jun 17, 2012 at 11:40 PM, Bruno Haible br...@clisp.org wrote:
Isaac Dunham wrote:
The test as it stands is error out on unsupported platforms unless
user specifies to use slow method.
My proposal is On unsupported platforms, use the slow method instead
of erroring out.
If we did
On Sun, Jun 17, 2012 at 6:15 PM, Bruno Haible br...@clisp.org wrote:
Paul Eggert wrote:
On 06/12/2012 04:21 AM, Paolo Bonzini wrote:
perhaps we can follow the suggestion and
replace if (freadahead (f)) with if (freading(f) !feof(f)) in
closein.c.
Yes, thanks, I like this idea the best
Il 08/06/2012 12:19, Pedro Alves ha scritto:
Have you any plans to address these problems? In particular, it does
seem odd to place a burden on libc authors of porting gnulib to it,
rather than just not supporting those functions which require
non-standard APIs on such libc's.
I've heard
From e2aa7434ad06a0ec4e2c47b57564313d16561c14 Mon Sep 17 00:00:00 2001
From: Paolo Bonzini pbonz...@redhat.com
Date: Tue, 12 Jun 2012 13:26:57 +0200
Subject: [PATCH 1/1] freadahead, freadptr, freadseek: Never fail compilation
2012-06-12 Paolo Bonzini bonz...@gnu.org
* lib
Il 12/06/2012 14:14, Eric Blake ha scritto:
While I agree with this, perhaps we can follow the suggestion and
replace if (freadahead (f)) with if (freading(f) !feof(f)) in
closein.c.
freading() is just as much an extension as freadahead(), but it might be
an easier extension to implement.
Il 07/06/2012 14:50, Eric Blake ha scritto:
The fix could be to have two different locale_charset() functions,
one that returns US-ASCII and another one that returns UTF-8.
The first one to be used when MB_CUR_MAX and mbrtowc() are used as
well, the second one to be used by gettext(). But
Il 07/06/2012 16:51, Eric Blake ha scritto:
On 06/07/2012 08:13 AM, Paolo Bonzini wrote:
Il 07/06/2012 14:50, Eric Blake ha scritto:
The fix could be to have two different locale_charset() functions,
one that returns US-ASCII and another one that returns UTF-8.
The first one to be used when
On Sat, Jun 2, 2012 at 6:23 AM, Paul Eggert egg...@cs.ucla.edu wrote:
now UTF-8 is set but MB_CUR_MAX is 1
That sounds bad.
Yeah, that's the real bug perhaps, but can we force an override of MB_CUR_MAX?
Paolo
Here is a report from a GNU sed user.
Paolo
SETUP:
$ sw_vers
ProductName: Mac OS X
ProductVersion: 10.7.4
BuildVersion: 11E53
$ ~/gnu/bin/sed --version
GNU sed version 4.2.1
PROBLEM: With UTF-8 input, but LANG and LC_ALL set to C, sed regular
expressions break on multibyte
Here is what I applied. Thanks!
2012-05-21 Paolo Bonzini bonz...@gnu.org
poll/select: prevent busy-waiting. SwitchToThread() only gives away
the rest of the current time slice to another thread in the current
process. So if the thread that feeds the file decscriptor
I noticed that errno doesn't work on Fedora 17's MinGW64 because it
defines some of MSVC10's new error codes, but not all. This patch
fixes it, while avoiding that glibc gets the replacement only because
it lacks EOTHER (which is not even in POSIX).
Ok?
2012-05-21 Paolo Bonzini bonz
Il 16/05/2012 21:16, Erik Faye-Lund ha scritto:
From: theoleblond theodore.lebl...@gmail.com
SwitchToThread() only gives away the rest of the current time
slice to another thread in the current process. So if the
thread that feeds the file decscriptor we're polling is not
in the current
Il 26/04/2012 22:40, Aharon Robbins ha scritto:
Hi Jim, Paul, Paolo, anyone else I missed,
I sent in paperwork for copyright assignments in grep and gnulib to the
FSF; they were mailed last Friday so they should be there by now.
Can we move ahead getting the Rational Range Interpretation
Il 27/04/2012 11:07, Aharon Robbins ha scritto:
From 5c7665f2ced46d2e830958bce1bf46469995d3de Mon Sep 17 00:00:00 2001
From: Arnold D. Robbins arn...@skeeve.com
Date: Fri, 27 Apr 2012 12:04:22 +0300
Subject: [PATCH] Implement Rational Range Interpretation in Gnulib.
---
lib/regcomp.c |
Il 01/03/2012 04:56, Bruno Haible ha scritto:
The IEEE tests show that on IRIX 6.5
cbrtf(-0.0) = -Inf(ouch!!)
cbrtl(-0.0) = +0.0
The patches are similar; here's the second one.
I suppose the one for cbrtf does not add an -ieee variant?
Paolo
Il 29/02/2012 12:29, Bruno Haible ha scritto:
When writing new math functions code, it is useful to know where the
existing code in gnulib comes from. Assuming that no one here has the
expertise to compute Chebyshev polynomials for transcendental functions,
it is clear from which files Paolo
On 01/06/2012 09:16 PM, Paul Eggert wrote:
The new quoting policy is inconsistent, since many diagnostics now use
quotes like this 'token', while quote (foo) still produces `foo' in
the C locale.
The idea is that Paolo's quotearg changes (see the thread at
On Sat, Jan 7, 2012 at 16:11, Bruno Haible br...@clisp.org wrote:
Paolo Bonzini wrote:
The idea is that Paolo's quotearg changes (see the thread at
http://lists.gnu.org/archive/html/bug-gnulib/2011-12/msg00145.html)
will be pushed to fix this inconsistency.
Paolo, would you like to do
On 01/03/2012 09:36 AM, Eli Zaretskii wrote:
From: Bruno Haiblebr...@clisp.org
Cc: bastien ROUCARIESroucaries.bast...@gmail.com, Eli Zaretskiie...@gnu.org, Eric
Blakeebl...@redhat.com, bonz...@gnu.org, bug-g...@gnu.org
Date: Tue, 03 Jan 2012 03:56:56 +0100
I'm adding this new module. Feel free
On 01/02/2012 11:57 PM, Paul Eggert wrote:
Anyway, impressing people is not the main goal here.
The main goal is to come up with a useful name that
doesn't praise the Windows API or operating environment.
win32 is not the right choice for that, we must use
something better, and there is a
On 01/03/2012 12:59 AM, Bruno Haible wrote:
1) It contrasts with the NtDll API.[2] For example, file names passed
to NtDll functions can use a different syntax than file names
passed to Win32 functions.[3] This distinction matters because
programs build with gnulib should
On 01/03/2012 12:59 AM, Bruno Haible wrote:
Hi Paul,
Attached is a proposed Gnulib patch to fix some occurrences of
the win terminology problem.
It mostly just substitutes Woe32 for Win32, except that for
Win32 API it substitutes Windows API.
It's good to clean up some of these terms once
On 12/24/2011 01:59 PM, Eli Zaretskii wrote:
+ return
+isatty (fd)
+#ifdef __MINGW32__
+/* Without the lseek call, Windows isatty returns non-zero for the
+ null device as well. */
+ lseek (fd, SEEK_CUR, 0) == -1
+#endif
This fix to isatty should be done in gnulib.
Otherwise
On 12/28/2011 01:58 PM, Eric Blake wrote:
For glibc, this is a very unlikely assumption even considering patch 1
only.
Not entirely unlikely - after all, glibc commit 2127a186 changed
getopt.c output from `' to '' style, back in 2008.
Yes, but patch 1 more or less covers only test code
On 12/27/2011 02:39 AM, Paul Eggert wrote:
As before, the assumption is that the changes affecting files
imported from upstream will be sent upstream; we shouldn't stop
importing merely because of quoting issues or minor commentary
changes, even if the changes are rejected upstream.
I'm also
On 12/23/2011 10:07 AM, Paul Eggert wrote:
[Adding bug-gnulib since I'm attaching a gnulib patch.]
On 12/23/11 00:22, Paolo Bonzini wrote:
This permits GNU gettext to replace the string @code{Processing file
! \%s\...} with a translated version.
This is ugly...
How about if we change
From: Paolo Bonzini pbonz...@redhat.com
The first two patches are improvement in documentation. The first
fixes a broken link, the second avoids the distinction between
British and American locales: most people are unaware of the
different typographic conventions and also there is usually
From: Paolo Bonzini pbonz...@redhat.com
* lib/quotearg.c (quotearg_buffer_restyled): Fix link to Wikipedia.
---
lib/quotearg.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/lib/quotearg.c b/lib/quotearg.c
index 03fbfe7..fdcb8da 100644
--- a/lib/quotearg.c
+++ b/lib
From: Paolo Bonzini pbonz...@redhat.com
Most programs are using single quotation marks even though this
conflicts with the American typographic convention. Since most
of the time there will actually be no translation catalogs for
English, suggest a convention that will clash little
From: Paolo Bonzini pbonz...@redhat.com
* lib/quotearg.c (gettext_quote): Map ` to ' for locale_quoting_style.
(quotearg_buffer_restyled): Fix example.
---
lib/quotearg.c |7 ++-
1 files changed, 2 insertions(+), 5 deletions(-)
diff --git a/lib/quotearg.c b/lib/quotearg.c
index a60baba
From: Paolo Bonzini pbonz...@redhat.com
* lib/quotearg.c (gettext_quote): Hard-code U+2018 and U+2019 when using
an UTF-8 locale.
* lib/quotearg.h (locale_quoting_style): Do not put an example in text,
since this would be wrong when using Unicode.
* modules/quotearg: Depend on c-strcaseeq
On 12/19/2011 09:11 AM, Paolo Bonzini wrote:
For more information, seehttp://www.cl.cam.ac.uk/~mgk25/ucs/apostrophe.html
Oops, see http://www.cl.cam.ac.uk/~mgk25/ucs/quotes.html :)
Paolo
On 12/19/2011 04:40 PM, Paul Eggert wrote:
'. If the catalog has no translation,
- locale_quoting_style quotes `like this', and
+ locale_quoting_style quotes 'like this', and
clocale_quoting_style quotes like this.
This
On Tue, Dec 20, 2011 at 01:22, Bruno Haible br...@clisp.org wrote:
I am very fond of my ` characters and hate the fact that stupid
standards broke what was the obviously useful thing to do ...
I don't know what you consider to be the obviously useful thing.
I find Markus Kuhn's write-ups at
On 12/13/2011 11:32 AM, Bruno Haible wrote:
I'm attaching the benchmark program I'm experimenting with. So far, it seems
that locale_charset() is really slow, whereas the is_cjk stuff is not a big
speed problem.
I would love to have locale_charset be either faster or use some thread-safe
On 08/31/2011 05:48 PM, Jim Meyering wrote:
The test-float test is failing on ppc64 with:
gcc version 4.4.4 20100630 (Red Hat 4.4.4-10) (GCC)
(albeit an aging Fedora 12 system)
due to the failure of this assertion:
ASSERT (LDBL_MIN_EXP= DBL_MIN_EXP);
It fails because of these
On 11/26/2011 02:14 PM, Bastien ROUCARIES wrote:
Malloc page than realloc to smaller does not work ?
There is no constraint that realloc(X, N) returns X, even if N is
smaller than the previous allocation size.
Paolo
On 11/26/2011 03:27 PM, Bruno Haible wrote:
- gnulib needs to add support for en@quot and en@boldquot in gnulib's
bootstrap script.
If you want so, why not. It's not that complicated. Take the rules from
gettext's Rules-quot, quot.sed, boldquot.sed files.
Yes, I'll look into it.
- perhaps
[Cc bug-gnulib/gettext/standards/texinfo, Bcc gnu-prog-discuss]
On 11/25/2011 02:51 PM, Thien-Thi Nguyen wrote:
GNU programs are urged to use `symbol' (grave, symbol, apostrophe)
in README, Texinfo uses those to implement @code in Info files, etc.
I propose GNU adopt U+2018 symbol U+2019
On 10/06/2011 10:43 AM, Bruno Haible wrote:
Bernhard Voelker wrote:
Alternatively, can't we use _exit() here?
This would not change the problem: _exit() is, like exit(), declared
as nonreturning on glibc systems and not declared this way on some
other platforms.
Is this true? exit is
On 09/25/2011 11:46 PM, Bastien ROUCARIES wrote:
Did you seehttp://bugs.python.org/issue4804 ?
Particularlyhttp://bugs.python.org/file12953/__pioinfo.patch
A little bit hackhish but seems to work
That would provide a lot more flexibility, including:
* the ability to implement O_INHERIT
On 09/26/2011 10:37 AM, Bastien ROUCARIES wrote:
python seehttp://svn.python.org/projects/python/trunk/Modules/posixmodule.c
seems to have solved a lot the maintenabilty stuff by determining the
actual size of the structure is determined at runtime.
Still hackish, though slightly better. What
On 09/26/2011 11:01 AM, Bruno Haible wrote:
Additionally, there's a legal problem here: Code like
Stopped reading here. Do not include snippets, we _can_ do clean-room
reverse engineering if necessary.
We can also take code from Wine's MSVCRT implementation if need be, so
that at least it
On 09/24/2011 06:11 PM, Bruno Haible wrote:
* modules/fclose-tests (Depends-on): Add fdopen.
* modules/fflush-tests (Depends-on): Likewise.
* modules/fgetc-tests (Depends-on): Likewise.
* modules/fputc-tests (Depends-on): Likewise.
* modules/fread-tests
On 09/16/2011 11:42 PM, Reuben Thomas wrote:
1. Drop the AC_DEFINE item, as it's a dead end (and, being non-GCS,
unusable) not a logical refinement of a developing technique (I do
agree that gradually building up complex techniques, with motivation
and explanation, is a good feature of GNU
On 07/27/2011 05:11 PM, Sam Steingold wrote:
Hi,
When the included regex is used I must prepend -I$(srcdir)/gllib to
CPPFLAGS to ensure that the gnulib regex.h is included in the sources.
(when this is not the case, I_must_ not do that!)
^^
On 07/28/2011 04:16 PM, Sam Steingold wrote:
If course, you can claim that, if gnulib regex.h were incompatible with
the system one, then configure would have detected that and included
gnulib regex in libgnu.a.
The gnulib regex.h is always a superset of the system one. ABI-wise,
the only
On 07/28/2011 06:43 PM, Sam Steingold wrote:
* Paolo Bonziniobam...@tah.bet [2011-07-28 16:54:12 +0200]:
The current situation is ugly---apart from the inconsistency in syntax
definitions, there is no reason indeed why regex shouldn't use
regex.in.h too, and no reason why it should require
On 07/26/2011 10:26 PM, Jim Meyering wrote:
I don't know about typical mingw file system usage.
Do mingw users use NTFS often?
Yes, all of them. MinGW is just the name of the free compiler
toolchain for Windows. Unlike cygwin, it uses the MS C library, so
anything MS documents
On 07/26/2011 12:18 PM, Jim Meyering wrote:
Regarding NTFS, can you point to a real gnulib-using application that is
misbehaving because of this? I've seen that some NTFS implementations
*do* have usable inode support. Both cygwin and fuse-based ones do,
so you must mean mingw.
Yes, I mean
On 07/26/2011 03:42 PM, Jim Meyering wrote:
I don't know about typical mingw file system usage.
Do mingw users use NTFS often?
Yes, all of them. MinGW is just the name of the free compiler toolchain
for Windows. Unlike cygwin, it uses the MS C library, so anything MS
documents applies to
On 07/24/2011 09:06 PM, Jim Meyering wrote:
What about a few POSIX-violating fringe operating systems (Windows and
DJGPP come to mind)?:) For Windows we can write our own stat
function in cygwin, but for DJGPP I think we're in a bad situation...
AFAIK, DJGPP is not relevant these days.
On 07/23/2011 11:19 PM, Bruno Haible wrote:
The only inelegance is having
to code up a struct type to pass around the privdata argument even for
something as simple as passing data in and reading it out again.
That's life, in C. A callback consists of a function pointer plus a privdata
On 07/23/2011 09:05 PM, Reuben Thomas wrote:
I just used pipe-filter-ii to reimplement GNU Zile's shell-command (as
in Emacs). It's a bit more complicated in structure than the original
with all the callbacks
Perhaps pipe-filter-gi can be used instead if your use case allowed you
to use
On 07/23/2011 11:12 PM, Jim Meyering wrote:
In my experience, dev/ino is sufficient, as long as you're not using one of
a few POSIX-violating fringe file systems (clearcase's MVS comes to mind).
What about a few POSIX-violating fringe operating systems (Windows and
DJGPP come to mind)? :)
On 07/06/2011 10:23 AM, Erik Faye-Lund wrote:
Without proper kernel-level support for poll, we probably can't have a
poll on Windows that suspend the thread until an event happens; we'll
have to accept a bit of polling. But I believe that looking and
quacking like POSIX is better than
On 07/06/2011 12:10 PM, Erik Faye-Lund wrote:
* lib/poll.c: Loop with yield if no events occured
---
OK, this should be a slightly more complete version. Added
SwitchToThread and mailing as a proper git-patch (including
a ChangeLog-entry) that can be applied with git-am.
Looks good to me.
On 07/03/2011 06:28 PM, Bastien ROUCARIES wrote:
On Sun, Jul 3, 2011 at 5:36 PM, Bruno Haiblebr...@clisp.org wrote:
Hi all,
I wrote:
... in programs that don't install signal handlers, EINTR ... also occurs in
MacOS X!
It is worse than that:
1) Even on Linux, even when the signal
On 06/30/2011 07:56 PM, Eric Blake wrote:
Yes, Windows pipes are that broken.:(
Using socketpair is a possibly good idea, but I would do it on
libvirtd only. I don't know exactly how libvirtd uses this pipe, but
perhaps it can be changed to an eventfd-like abstraction that can be
the string being sought.
Ok?
Paolo
2011-06-27 Paolo Bonzini bonz...@gnu.org
* m4/mmap-anon.m4: Do not use special characters in the
regular expression.
diff --git a/m4/mmap-anon.m4 b/m4/mmap-anon.m4
index 7ba7fd2..952536f 100644
--- a/m4/mmap-anon.m4
+++ b/m4/mmap-anon.m4
@@ -27,18
* A couple of tests fail:
FAIL: test-mbrtowc4.sh
FAIL: test-mbrtowc3.sh
FAIL: test-mbsrtowcs4.sh
FAIL: test-setlocale2.sh
FAIL: test-mbrtowc3.sh (exit: 262)
FAIL: test-mbrtowc4.sh (exit: 262)
FAIL: test-mbsrtowcs4.sh (exit: 262)
FAIL: test-setlocale2.sh (exit: 1)
This is for gnulib, I'll
* A couple of tests fail:
FAIL: test-mbrtowc4.sh
FAIL: test-mbrtowc3.sh
FAIL: test-mbsrtowcs4.sh
FAIL: test-setlocale2.sh
FAIL: test-mbrtowc3.sh (exit: 262)
FAIL: test-mbrtowc4.sh (exit: 262)
FAIL: test-mbsrtowcs4.sh (exit: 262)
FAIL: test-setlocale2.sh (exit: 1)
This is
On 06/16/2011 10:44 AM, Jim Meyering wrote:
To make this proposed change go through, that configure-time option would
have to be eliminated, so that we always build with the gnulib-provided
regex code. Of course, if glibc ever changes, we can detect that and
automatically prefer it when
On 06/15/2011 09:12 PM, Jim Meyering wrote:
However, backreferences force these tools to skip the DFA-based
optimization and resort to running the regexp code. In that case,
there is a dichotomy. Adding a backreference to a range-including
regexp would have the surprising consequence of
On 06/16/2011 09:06 PM, Aharon Robbins wrote:
I have already corresponded with Chet. He plans to add a shell option
to enable RRI, and we can hope that at some point, it might become the
default. So that has already been started.
I'd do the other way round---make it the default, and add a
In principle, I'm all for this, but in practice, I'm going to leave gawk's
code alone for now (there's always 4.1 :-).
As long as --posix is not affecting the choice, that's fine. However,
please make sure that compiling gawk --without-included-regex works
(it should go without saying)!
[making this public, there should be no reason not to]
On 06/08/2011 10:14 PM, Aharon Robbins wrote:
Hi. As we've discussed a little previously, I finally got tired of
trying to explain to users why the character range [a-z] was matching
most uppercase letters also. (I've found a bug in gawk!
On 06/09/2011 11:33 AM, Jim Meyering wrote:
I like the idea.
However a potential sticking point is the equivalence class (e.g., using
[=e=] to match e as well as accented versions like é, è and ê).
That is the one feature that you get with glibc, and that you would
sacrifice when building
On 06/09/2011 11:58 AM, Bruno Haible wrote:
Paolo,
[=e=] to match e as well as accented versions like é, è and ê).
That is the one feature that you get with glibc, and that you would
sacrifice when building --with-included-regex.
I agree. It's up to distros to choose, of course.
If you
On 06/09/2011 01:12 PM, Bruno Haible wrote:
What would it take to let distros/people use --with-included-regex and
get understandable semantics for ranges + working equivalence classes?
I would prefer that to your proposal, because it cannot be seen as a
regression by people who care about
On 06/09/2011 01:53 PM, Bruno Haible wrote:
Paolo,
My proposal wouldn't change defaults, which is why I believe that this
is a separate topic.
But at the same time you are pushing for the use of --with-included-regex.
We found out that by doing this, the equivalence classes feature gets
On Thu, Jun 9, 2011 at 19:14, Paul Eggert egg...@cs.ucla.edu wrote:
On 06/08/2011 10:14 PM, Aharon Robbins wrote:
So, for the upcoming gawk 4.0, I decided (as Karl put it) to cut the
Gordian knot and make ranges behave like the C locale, the way it's long
been documented, and as most people
On 06/07/2011 12:38 AM, Sam Steingold wrote:
note that the windows version would have to operate on the values of
GetLastError() instead of errno.
At least this part does not belong in gnulib, I think, because gnulib
tries to map Windows errors to errno values whenever necessary. You
would
On Sat, Jun 4, 2011 at 00:37, Matthias Bolte
matthias.bo...@googlemail.com wrote:
After testing a while and reading MSDN docs the problem seems to be
that MsgWaitForMultipleObjects doesn't work on pipes. It doesn't
actually wait but just returns immediately. Digging MSDN and googling
about
On 05/05/2011 07:23 AM, Gary V. Vaughan wrote:
This makes it a lot easier to distribute software through Apple's
various AppStore channels under the more onerous terms they impose,
while still having the freedom to share the actual code under the
GPL.
I'm afraid this would only be a Pyrrhic
On 05/05/2011 10:48 AM, Jim Meyering wrote:
Bruce Korb wrote:
...
Is it a small thing if a half dozen engineers spend days
exchanging emails trying to figure it out? Not so small to me.
Fortunately, this doesn't happen often, and once it does for a given
module, it's not likely to happen
On 05/07/2011 05:32 PM, Bruno Haible wrote:
So, my vote for the license of libposix would be LGPLv3+ | GPLv2+.
For that I agree.
However, relicensing-wise, the best path to achieving this, is to first
ensure that all gnulib modules meant for libposix are LGPLv2+. Then
libposix can be
On 05/04/2011 02:55 AM, Bruno Haible wrote:
libposix is intended to be under LGPL, but we have some way to go
(agreements to get from the authors) for some modules:
http://lists.gnu.org/archive/html/bug-gnulib/2010-12/msg00184.html
FWIW, I'm okay with LGPL for the mathl modules. They
On 05/04/2011 08:04 PM, Bruce Korb wrote:
So, if I've not mis-stated nor missed anything, we only wait for Derek,
Paolo and Ralf.
Ok by me.
Paolo
1 - 100 of 618 matches
Mail list logo
