Package: kjc
Version: 2:1.1.4.PRECVS6-1
Severity: serious
The kaffe source package does not include the source code for kjc.
Instead, there is just a kjc.jar full of compiled class files.
/usr/share/doc/kjc/copyright describes:
> * libraries/javalib/kjc.jar
>
> Kopi Java Compiler 2.1B with kaff
Kalle Olavi Niemitalo <[EMAIL PROTECTED]> writes:
> Frank Lichtenheld <[EMAIL PROTECTED]> writes:
>
>> emacs20 was removed before sarge release. Build errors of it on
>> unstable certainly will not be fixed... Closing the bug.
>
> I don't think the build
Frank Küster <[EMAIL PROTECTED]> writes:
> Kalle Olavi Niemitalo <[EMAIL PROTECTED]> wrote:
>
>> Could you test the following patch, and report your Emacs and SSH versions?
>
> Yes, the patch fixed it: Now the graphical ssh-askpass dialog is shown.
Thanks.
Package: pcregrep
Version: 6.4-1.1
Severity: normal
When pcregrep is outputting the input lines that matched the regexp,
it truncates each output line at the first null character. AFAICT,
the bug affects only the output: the regexp engine correctly finds
characters that follow a null character in
Florian Weimer <[EMAIL PROTECTED]> writes:
> IIRC, the correct idiom is to use fboundp, as in:
>
> (when (fboundp 'set-buffer-file-coding-system)
>(set-buffer-file-coding-system 'undecided-unix nil))
Stefan Reichör made this change in revision 12943 of psvn.el.
The fix is included in Subver
Frank Küster <[EMAIL PROTECTED]> writes:
> When using psvn.el for a working copy that uses svn+ssh, I get a prompt
> to type in the password (actually, the passphrase for my RSA key). The
> prompt is written in the *svn-process* buffer (I think), but it does not
> accept input. Therefore, commit
Pierre THIERRY <[EMAIL PROTECTED]> writes:
> When having filenames containing characters outside ASCII, it is shown
> escaped in the $Id$ keyword, thus being not very easy to read...
That was fixed in upstream revision 15073.
The forthcoming Subversion 1.3.0 will include the fix.
However, this c
LI Daobing <[EMAIL PROTECTED]> writes:
> Kalle Olavi Niemitalo wrote:
>> Note that the text here was "0 file(s) marked", whereas you had
>> "0 files marked" without the parentheses.
The parentheses of "file(s) marked" were added in revision 1
LI Daobing <[EMAIL PROTECTED]> writes:
> I delete psvn.elc and then everything seems right.
> How about delete this file in preinst or postrm?
I don't decide how Subversion is packaged for Debian.
I only hack psvn.el in the upstream repository.
In my opinion, the best course of action would be t
Frank Küster <[EMAIL PROTECTED]> writes:
> In pcvs.el, x11-ssh-askpass is used.
Not in Emacs 21.4, nor in emacs--multi-tty--0--patch-426.
In pcvs-parse.el, cvs-parse-ignored-messages holds a regexp
with which pcvs skips messages where ssh reports it is starting
ssh-askpass. But pcvs itself never
Could you test the following patch, and report your Emacs and SSH versions?
--- psvn.el (revision 13845)
+++ psvn.el (local)
@@ -368,7 +368,7 @@ (defun svn-run-svn (run-asynchron clear-process-buffer cmdtype &rest arglist)
(svn-status-update-mode-line)
(sit-for 0.1)
main): Don't recognize -i and -p at all. (See: #193170)
-- Kalle Olavi Niemitalo <[EMAIL PROTECTED]> Sat, 30 Apr 2005 12:38:13 +0300
--- coreutils_5.2.1-2/debian/patches/21_uname-no-unknowns.diff 2005-04-30 12:21:04.0 +0300
+++ coreutils_5.2.1-2.kon.1/debian/patches/21_uname-
ive up if there are not enough
good spots, instead of falling in an infinite loop.
-- Kalle Olavi Niemitalo <[EMAIL PROTECTED]> Sat, 18 Jun 2005 10:07:51 +0300
--- crawl-4.0.0beta26.orig/source/dungeon.cc
+++ crawl-4.0.0beta26/source/dungeon.cc
@@ -5,7 +5,8 @@
*
* Change H
tags 318508 + patch
quit
Daniel Schepler <[EMAIL PROTECTED]> writes:
>From my build log (reproduced using pbuilder in an i386 chroot):
Also reproduced on AMD Duron running Debian i386.
> Unfortunately, since this is in build-tree/db2 instead of
> build-tree/db-2.7.7, I can't figure out how to g
Package: libevent1
Version: 1.0b-1.1
Severity: normal
I updated libevent1 from 0.8-2 to 1.0b-1.1 at 2005-04-12T23:15,
and now I see my tinysnmp-agent 0.8.1 installation has not been
working since. I get lines like this in syslog:
Apr 14 12:55:02 [tinysnmpd] recvfrom failed: Bad address
Downgrad
Guus Sliepen <[EMAIL PROTECTED]> writes:
> I cannot reproduce this problem. Do you have a savegame of this
> situation?
I don't, but I'll try to construct one.
pgppsuuS82DDX.pgp
Description: PGP signature
tag 323079 - unreproducible
quit
Kalle Olavi Niemitalo <[EMAIL PROTECTED]> writes:
> Guus Sliepen <[EMAIL PROTECTED]> writes:
>
>> I cannot reproduce this problem. Do you have a savegame of this
>> situation?
>
> I don't, but I'll try to construct
Frank Lichtenheld <[EMAIL PROTECTED]> writes:
> emacs20 was removed before sarge release. Build errors of it on
> unstable certainly will not be fixed... Closing the bug.
I don't think the build errors are specific to unstable. The bug
that causes them is in debian/rules and does not depend on t
Package: linux-2.6
Version: 2.6.18-8
Severity: normal
Date: Sat, 30 Dec 2006 16:50:41 +0200
...
find
'debian/linux-patch-debian-2.6.18/usr/src/kernel-patches/all/2.6.18/debian' !
-path '*/series/*' -type f -execdir bzip2 '{}
' ';' -execdir chmod 644 '{}.bz2' ';'
find: invalid predicate `-execdir
Gabor Gombas <[EMAIL PROTECTED]> writes:
> It seems that elinks cannot handle UTF-8 input/output properly. Symptoms
> include links containing accented characters jumping some positions to
> the left or becoming garbled when selected, or wrong cursor positioning
> in input fields when I enter acce
These changes are now in the upstream source trees from which
ELinks 0.11.3 and 0.12.0 will be released.
pgpUdmJj87562.pgp
Description: PGP signature
package elinks
forwarded 399188 http://bugzilla.elinks.cz/show_bug.cgi?id=841
quit
Stefan Fritsch <[EMAIL PROTECTED]> writes:
> A vulnerability has been found in elinks:
> Links web browser 1.00pre12 and Elinks 0.9.2 with smbclient installed
> allows remote attackers to execute arbitrary code via
Robert Millan <[EMAIL PROTECTED]> writes:
> $ echo "à" | elinks -dump
>`a
>
> It should print "à" instead (at least on utf-8).
ELinks 0.11.* do not support UTF-8 as the charset for dumps or
terminal output. As a workaround, you can select some other
charset with the -dump-charset option and
Package: tcc
Version: 0.9.23-4.kon.1
Severity: normal
gcc-4.0 4.0.3-3 compiles this program all right, but tcc fails:
sizeof.c:5: pointer expected
int
main (void)
{
char x[1];
return (sizeof (x) [0]) - 1;
}
tcc 0.9.23-4.kon.1 is 0.9.23-4 + the patch from bug 418360.
/* C99 6.5.3p1 */
unary-
Package: jadetex
Version: 3.13-6
Severity: important
Building jadetex, I get an error:
> rm -rf debian/jadetex
> install -d -o root -g root -m 755 debian/jadetex/usr/bin
> debian/jadetex/usr/share/doc/jadetex debian/jadetex/usr/share/man/man1
>\
> debian/jadetex/u
It is even possible to contrive programs where this bug does not
cause compilation to fail but rather results in incorrect output.
I think such constructs are very rare outside IOCCC, fortunately.
#include
#include
int
main (void)
{
static char x [2];
if (sizeof (*x) [x])
{
puts (
In ELinks 0.10.6, the toggle-numbered-links action (listed
in the View menu and bound to "." by default) did not mark
the document.browse.links.numbering option as modified, so
it did not get saved with the default config.saving_style =
3 setting. This has been fixed in ELinks 0.11.0 and later:
th
Miciah Dashiel Butler Masters <[EMAIL PROTECTED]> writes:
> ELinks will crash if one selects a bookmark, selects Delete, allows
> the session snapshotting (or anything else) to delete that bookmark,
> and then confirms the deletion. This is due to poor reference counting
> that I intend to fix whe
I get no such error with /usr/bin/elinks and
/usr/share/doc/elinks/examples/contrib/lua/hooks.lua.gz from
the Debian elinks 0.11.1-1 package, so I suppose this bug
could be closed.
pgpoWvO6VO88y.pgp
Description: PGP signature
Ralf Hildebrandt <[EMAIL PROTECTED]> writes:
> After the proxy "trickled" the first by back to elinks,
> it crashed with a "floating point exception".
There was a change made on 2005-04-20 that may have fixed this.
http://pasky.or.cz/gitweb.cgi?p=elinks-history.git;a=commit;h=621ccbab1a61370ac74
Package: tcc
Version: 0.9.23-2
Severity: normal
Tags: patch
I have reflowed some of the long lines below.
$ cat dummy.c
int
main ()
{
return 0;
}
$ tcc -c dummy.c
$ ls -l dummy.o
-rw-rw-r-- 1 Kalle Kalle 484 2007-04-09 13:31 dummy.o
$ tcc dummy.o -lgnutls -lguile
Segmentation fault (core dumped
Josselin Mouette writes:
> I think wheezy would be a good time to finally ditch non-UTF8 locales.
> IIRC, we made the switch to UTF8 by default in etch (and we were already
> way too late in doing that), and supporting non-UTF8 stuff becomes
> harder and harder, at least for desktop software.
Jakub Wilk writes:
> elinks follows HTTP 302 redirects to file:// URLs. This can cause
> information disclosure or, if protocol.file.allow_special_files is
> enabled, denial of service:
If a local user is running ELinks and getting the output to a
terminal, then ELinks will display the contents
package xserver-xorg-video-intel
tags 651741 - fixed-upstream
found 651741 xserver-xorg-video-intel/2:2.19.0-6
quit
The upstream developers of xf86-video-intel denied my request to
blacklist the faulty VAIO ACPI implementation. They said Linux
should be patched instead. I have not yet reported t
package gnome-power-manager
reassign 651741 xserver-xorg-video-intel
quit
Kalle Olavi Niemitalo writes:
> So, the next thing I'll try will be moving it back to the top.
That works, and I no longer think gnome-power-manager should
be changed, so I'm reassigning the bug.
package xserver-xorg-video-intel
tags 651741 + patch
quit
Kalle Olavi Niemitalo writes:
> (b) PCI subsystem. Here, lspci -vn shows:
>
> 00:02.0 0300: 8086:0046 (rev 02) (prog-if 00 [VGA controller])
> Subsystem: 104d:907c
>
> intel_driver.h defines th
Package: stow
Version: 2.2.0-1
Severity: normal
I have a script that configures and builds a package and stows
the resulting binaries. As a precaution against hypothetical
package names that start with a dash, it uses "--" like so:
( cd -- "${stowdir}" && stow --restow -- "${PKG_STOWNAME}" )
Th
Package: stow
Version: 2.2.0-1
Severity: minor
Kalle@Niukka:~/prefix/stow 1% stow --restow ./emacs
Undefined subroutine &main::error called at /usr/bin/stow line 568, line
18.
It tries to call error("Slashes are not permitted in package names")
but never defines that subroutine. This is the on
tags 681752 + patch
quit
This is how I think it should be fixed.
Both "stow --restow emacs" and "stow --restow -- emacs" work now.
This patch is rather simple, perhaps not copyrightable.
Either way, I am not assigning copyright to the FSF. According
to /usr/share/doc/stow/copyright, the FSF has
Package: iceweasel-l10n-fi
Version: 1:10.0.6esr-1
File: /usr/lib/iceweasel/extensions/langpack...@iceweasel.mozilla.org.xpi
Severity: minor
Tags: upstream, l10n
In the Finnish language pack, in the dialog box that asks whether
to accept a cookie, the check box and the Deny ("Estä") button
both hav
Dominique Dumont writes:
> Le jeudi 21 février 2013 23:12:28, vous avez écrit :
>> It looks like the license of GNUTLS 3.1.x may eventually be
>> changed from LGPLv3+ back to LGPLv2.1+. If that succeeds,
>> I believe it would solve the incompatibility with Pan.
>> http://lists.gnutls.org/piperma
Dominique Dumont writes:
> I'll put back SSL support for Pan in Debian unstable once the problematic
> code
> is relicensed or re-written.
It looks like the license of GNUTLS 3.1.x may eventually be
changed from LGPLv3+ back to LGPLv2.1+. If that succeeds,
I believe it would solve the incompa
Source: xulrunner
Version: 1.9.1.8-3
Mike Hommey writes:
>* debian/control:
> + Add missing ${misc:Depends}
> + Build-depend on binutils with readelf -p support. Closes: #56.
The build-dependency and the runtime dependency on binutils are now:
binutils (>= 2.18
Toni Mueller writes:
> I need an option in elinks to select the address to bind to,
> and to force an address family (eg. one of IPv4 or IPv6 only).
For the latter, please see the connection.try_ipv4 and
connection.try_ipv6 options.
http://elinks.cz/documentation/manpages/elinks.conf.5.html#con
Toni Mueller writes:
> thanks for the hint, but: There appears to be no command line switch to
> do it, right?
elinks -eval "set connection.try_ipv4 = 0"
but if that connects to an existing elinks process, then the
setting will be lost, I think. So if that is a problem, you may
have to combine
Brice Goglin writes:
> So the crash is reproducible?
So far, it has only happened once here. I used "info symbol" and
"print/x" in gdb to add the symbols to the original backtrace.
I now tried replugging the tablet a dozen times without
hibernation and once with hibernation but the X server di
Package: libgtk-3-0
Version: 3.0.11-1
Severity: normal
Start cheese 3.0.1-2. It is in photo mode by default. Click on
the empty pane where photos would appear if you took any. Press
the Right arrow key. The process dies with SIGSEGV.
This crash does not occur if you do have any photos in the
Package: id-utils
Version: 4.5-3
Severity: normal
/usr/share/doc/id-utils/copyright claims:
> Copyright:
>
> 1986-2006 Free Software Foundation, Inc.
>
> License:
>
>This package is free software; you can redistribute it and/or modify
>it under the terms of the GNU General Public
Package: ardentryst
Version: 1.71-4
Severity: minor
If the inventory menu has scrolled so that the first few entries
are no longer visible, and you then press Left to return to the
main menu and Right to reenter the inventory menu, the game
selects the first inventory entry but does not scroll it
Package: ardentryst
Version: 1.71-4
Severity: minor
In a shop, if you first choose Sell and press C to sell an item,
but then press X to cancel the transaction, then the shopkeeper
asks what you want to buy, even though you're still selling
rather than buying.
For example, enter Elchim's Shop in
Package: ardentryst
Version: 1.71-4
Severity: normal
I was playing as Pyralis in Worm Woods (Sempridge5). Normally,
if you try to jump above the top of the level, the hero hits a
ceiling. However, at the left side of the NPC who asks for
Anneludine Shells, there is a hole in the ceiling. By wea
package abiword
found 623136 abiword/2.9.1-0.2
quit
Here's a patch that I think should fix the bug.
However, I could not test it, because abiword build-depends on
libpsiconv-dev, which is not currently available in wheezy.
Thus not adding the patch tag.
diff -ru abiword_2.9.1-0.2/debian/abiword.mi
Package: gnome-power-manager
Version: 3.0.2-3
Severity: normal
On a Sony Vaio VPCYA1V9E laptop, if I press Fn+F5 or Fn+F6, then
gnome-power-manager attempts to change the brightness of the
display backlight, and shows a window as if it had worked.
However, the brightness does not change. Likewise
Package: manpages-dev
Version: 3.32-0.2
Severity: wishlist
File: /usr/share/man/man2/ioprio_get.2.gz
The ioprio_get(2) manual page describes the meanings of the which
and who parameters:
> IOPRIO_WHO_PROCESS
>who is a process ID identifying a single process.
>
> IOPRIO_WHO_PGRP
>w
Package: util-linux
Version: 2.19.1-5
Severity: wishlist
File: /usr/bin/ionice
The ionice(2) manual page describes the two syntaxes:
> ionice [[-c class] [-n classdata] [-t]] -p PID [PID]...
> ionice [-c class] [-n classdata] [-t] COMMAND [ARG]...
However, when I use the latter syntax with neith
Enrico Tassi writes:
> On Wed, Dec 09, 2009 at 09:42:18AM +0100, Erich Schubert wrote:
>> As far as I know (from discussion with upstream and on #debian-devel) it
>> is currently impossible for me to re-use the Debian lua packages for building
>> Enigma. The main reason is that C++ exceptions and
Enrico Tassi writes:
> I've prepared that patch a while ago (so it may not apply 100% clean).
> I'm attaching it, it is for the lua5.1 package of course.
> Could you please test it?
Thank you. The patch conflicted in debian/changelog only.
I had some difficulty getting the Makefile changes app
I investigated the possibility of just putting the C++ code in
liblua5.1.so: keeping the ABI compatible but adding exception
support. I think this is feasible for liblua5.1.so but
unfortunately not for liblua5.1.a, because the C++ support needs
several symbols from libsupc++.a and we cannot expect
Enrico Tassi writes:
> Please try the attached patch, I'm planning to upload this to unstable
> as soon as possible.
In the liblua5.1-0 package built from your 5.1.4-12 patch, the
shlibs file contains:
liblua5.1-c++ 0 liblua5.1-0
liblua5.1 0 liblua5.1-0
Because liblua5.1-c++.so.0 did not exist
package egoboo-data
found 163629 egoboo/1:2.8.1-1
quit
Mark Purcell writes:
>* Upstream fixes:
> - Closes: #163629: egoboo-data: rogue can gain full speed even with ball
Nope, it has not been fixed.
When you pick up the iron ball, it lowers your maximum acceleration:
| Hardware watchp
Package: inkscape
Version: 0.48.3.1-1+b1
To clarify: the large circle near the top left corner of
http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=5;filename=cssbug.svg;att=1;bug=588912
should have a dark grey stroke and a light grey fill,
but Inkscape instead shows a red stroke.
pgpcYgD6Frdeh.p
Martin Erik Werner writes:
> Here is a patch which fixes this issue for me, it is based on a
> suggestion at
> http://egoboo.sourceforge.net/phpBB3/viewtopic.php?f=3&t=1177&p=61333#p61333,
> see the DEP-3 header for more information.
That patch adds casts to two control_is_pressed subtractions
Jonathan Nieder writes:
> Closing optimistically. As usual, confirmation either way about the
> fix would be welcome.
I installed linux-image-3.2.0-2-amd64 3.2.15-1, commented out
"blacklist acer-wmi", and rebooted to that kernel. The acer-wmi
module was loaded and WLAN still works.
pgpHyw6o
package gcc-4.3
found 517653 gcc-4.4/4.4.7-1
found 517653 gcc-4.6/4.6.3-1
found 517653 gcc-4.7/4.7.0-3
quit
Kalle Olavi Niemitalo writes:
> Compiling this with gcc-4.3 -Wall -c
>
> int
> main (void)
> {
> return "hello" == "there";
> }
>
> res
Package: inkscape
Version: 0.48.3.1-1+b1
Severity: normal
Tags: l10n
When I run Inkscape with LANG=fi_FI.utf8, the menu bar has the
following mnemonics for menus (English equivalents in parentheses):
_Tiedosto (_File)
_Muokkaa (_Edit)
_Näytä(_View)
_Taso (_Layer)
Jonathan Nieder writes:
> Please test the attached patch, following instructions from [1] or the
> following instructions:
Thank you; I didn't realize this had already been fixed upstream.
stable/linux-3.2.y was v3.2.13 = 02905906ddc049720ded93ded03bbc2652f4646a.
That version has the bug; acer-
I upgraded my Sony Vaio VPCYA1V9E from Wheezy to Jessie.
It now has these versions of packages:
xserver-xorg-video-intel2:2.21.15-2+b2 amd64
linux-image-3.16.0-4-amd64 3.16.7-ckt25-1 amd64
As a test, I commented out all of /etc/X11/xorg.conf.
/var/log/Xorg.0.log then showed:
[166061.586]
Package: rsyslog
Version: 8.16.0-1+b4
Severity: normal
User: debian-h...@lists.debian.org
Usertags: hurd
I get this to /var/log/syslog on each boot:
Jul 2 16:19:36 luuska rsyslogd: [origin software="rsyslogd" swVersion="8.16.0"
x-pid="1207" x-info="http://www.rsyslog.com";] start
Jul 2 16:19:3
althaser writes:
> Could you please still reproduce this issue with newer
> gnome-settings-daemon version like 3.4.2+git20121218.7c1322-3+deb7u3 or
> 3.8.5-2 ?
I upgraded the laptop to Wheezy and the bug no longer occurs.
Test A:
1. Disable the touchpad by pressing the touchpad toggle button.
Kalle Olavi Niemitalo writes:
> Support for XF86TouchpadOff or XF86TouchpadOn was first included
> in GNOME_SETTINGS_DAEMON_2_91_4. The experimental Debian package
> of gnome-settings-daemon 2.91.5.1-1 includes that change.
> However, I'm not sure it'd be right to mark th
Kalle Olavi Niemitalo writes:
> Normally, if you try to jump above the top of the level, the
> hero hits a ceiling. However, at the left side of the NPC who
> asks for Anneludine Shells, there is a hole in the ceiling. By
> wearing Winged Boots, Pyralis was able to jump through tha
Package: xul-ext-noscript
Version: 2.6.8.19-1~deb7u1
Severity: normal
Originally with these packages:
ii iceweasel 31.8.0esr-1~deb7u1 amd64 Web browser based on Firefox
ii xul-ext-noscript 2.6.8.19-1~deb7u1 allpermissions manager for Icew
I reset the noscript settings, restart
David Prévot writes:
> + [ Kalle Olavi Niemitalo ]
> + * Temporarily allow scripts with recent iceweasel
> +(Closes: #797043)
Those functions in noscriptService.js are used not only for
setting up the menu (from which the user could temporarily
whitelist sites), but also for
Possibly related upstream bugs:
Bug 331257 - data: images show up when 'load images'
(Tools->Options->content->checkbox) is disabled
Bug 449273 - SVG:display desc (and title?) as text if images are blocked
Bug 1196784 - img tag with srcset attribute bypasses permissions.default.image=2
https://b
David Prévot writes in Bug#797079:
> Uploaded (with the improved changelog and metadata suggested by Kalle),
> thanks.
I installed xul-ext-noscript 2.6.8.19-1~deb7u2 from
wheezy-proposed-updates, and it works OK.
However, I see the patch now has the following line:
Origin: backport,
http://an
Markus Koschany writes:
> I now assume that those bugs are either fixed in the latest
> version of Powermanga, 0.92, or that more information are
> needed to fix them.
Bug #561670 "D-pad won't move the ship straight left" doesn't seem
seem fixed in the upstream Powermanga 0.92. If you search fo
# Bcc: control@ so that nobody accidentally sends replies there.
package powermanga
found 561670 0.92-1
quit
Markus Koschany writes:
> In this case more information means that it would be interesting to know
> what would happen if you changed the variables to a different value but
> I understand
Package: iceweasel-l10n-fi
Version: 1:31.2.0esr-2~deb7u1
File:
/usr/lib/iceweasel/browser/extensions/langpack...@iceweasel.mozilla.org.xpi
Severity: minor
Tags: upstream, l10n
Click the menu button at the right end of the tool bar, and then
click the "Muokkaa" button at the bottom of the menu to
Philipp Kewisch writes:
> I do make the official Lightning packages, but if Icedove is using
> something different then I don't create those packages. For any language
> specific issues, I'd suggest filing a bug in the respective language
> component on bugzilla.mozilla.org.
Before I refile the
This Bug #801348 looks pretty similar to Bug #651741, which I
reported originally in 2011. The hardware is different though.
You can probably make it work by putting this in /etc/X11/xorg.conf:
Section "Device"
Option "Backlight" "intel_backlight"
Identifier "Card0"
Source: openmpi
Version: 2.0.1-7
Severity: normal
User: debian-h...@lists.debian.org
Usertags: hurd
Samuel Thibault writes:
> openmpi 2.0.1-7 currently FTBFS, which makes a lot of package
> unbuildable because openmpi is the default mpi implementation. Version
> 2.0.1-4 did build, so there's pro
Kalle Olavi Niemitalo writes:
> Please disable the Java bindings of openmpi on hurd-i386, like
> they are already disabled on hppa.
I verified that disabling Java bindings with the following patch
lets openmpi 2.0.1-7 build on hurd-i386. However, I'm afraid
this patch would also af
Matthias Klose writes:
> It would be good to have a self-contained example to show the
> exact issue.
Enable the "de_AT.UTF-8" locale in "dpkg-reconfigure locales",
copy the attached files to a directory, and run "make check"
there. The C version outputs "test" with glibc 2.19, but the
Python v
Kalle Olavi Niemitalo writes:
> Content-Type: text/plain; charset=iso-8859-15
> Content-Disposition: attachment; filename=de.po
> Content-Transfer-Encoding: quoted-printable
Oops, that file was UTF-8 originally, but my MUA recoded it to
iso-8859-15, which is inconsistent with the Con
Robert Luberda writes:
> According to GNU gettext documentation[1]: "The variable LANGUAGE is
> ignored if the locale is set to ‘C’."
That exception was added on 2001-01-03, for glibc 2.2.1.
In glibc 2.2, LANGUAGE used to override LC_ALL=C.
In Python 2.0 (released on 2000-10-16), 2.7, and 3.5.0
Package: hurd
Version: 1:0.8.git20160826-1
Richard Braun writes:
> This was famously shown with the example of the
> firmlink translator used in /tmp, which would cause the removal of
> any file targeted by the firmlink on /tmp cleanup during system
> startup.
That was already fixed in daemons/
Richard Braun writes:
> A true fix would mean we provide the same assumptions the init
> scripts on Unix.
Programs deleting or modifying files in /tmp already have to be
careful not to follow other users' symlinks. Perhaps you could
reuse that somehow. __hurd_file_name_lookup_retry in glibc ha
Package: hurd
Version: 1:0.8.git20160826-1
Severity: normal
The hurd package makes /etc/hurd/runsystem a symlink that is
managed with the alternatives system. The default is
runsystem.sysv, and runsystem.gnu is also available.
However, if I use 'update-alternatives --config runsystem' to
select
Samuel Thibault writes:
> It is really non-technical work, a matter of using the check-copyright
> script to check that the various licences are referenced in
> debian/copyright (there is no hard need to reference files exactly,
> the only minimal need is knowing which licences end up in the gnum
This message applies to gnumach 2:1.7+git20160809-2.
The following files are not used by "dpkg-buildpackage -uc -b
-nc", i.e. their atimes do not change during this binary-arch
build, and the build succeeds even if they are removed.
./ChangeLog.0
./ChangeLog.00
./DEVELOPMENT
./ddb/db_mp.h
./ddb/t
sam/_libc.h has:
OREAD = 0,
OWRITE = 1,
ORDWR = 2,
OCEXEC = 4,
but it looks like nothing #includes that file, so perhaps that
doesn't have to be patched. Have you tested sam on the Hurd?
If these flags are sent in the Topen and Tcreate requests of the
9P protocol
My inventory of the licenses in gnumach 2:1.7+git20160809-2 is not
yet complete. I'm seeing two kinds of license violations so far.
University of Utah advertising clause vs. GNU GPL
-
Several files have a license notice like this:
> Copyright (c)
Package: hurd
Version: 1:0.8.git20160826-1
Severity: serious
File: /hurd/ext2fs.static
The ext2fs translator contains GPLv2-only code copied from Linux,
but it is linked (through libstore) with libparted, which is
GPLv3-or-later since 2007. This combination violates at least
one of the licenses.
Samuel Thibault writes:
> But storeio can be used as an intermediate between the two.
"storeio --store-type=part 1:device:hd0" apparently supports
file_get_storage_info and reports the partition boundaries there,
so the I/O would not have to go through the storeio translator.
libstore/encode.c
Package: hurd
Version: 1:0.8.git20160809-1
Severity: normal
File: /bin/rpctrace
If the COMMAND argument of rpctrace does not contain slash, then
rpctrace searches for it in PATH. If the file found is a script
file, then the interpreter of the script needs to know the file
name so that it can open
This fixes the following test case:
mkdir testy
echo '#! /bin/bash' > testy/prog
echo 'printf "%s\n" "$0"' >> testy/prog
chmod +x testy/prog
PATH=$(pwd)/testy /bin/rpctrace -E PATH=/usr/bin:/bin -o /dev/null prog
Before this patch, the output is:
/bin/bash: prog: No such file or dire
Package: hurd
Version: 1:0.8.git20160522-4
Severity: normal
Connect to the hurd machine via ssh. Type the password when
prompted. Run "stty -a" via the ssh connection. It outputs:
speed 38400 baud; rows 36; columns 149;
intr = ^C; quit = ^\; erase = ^?; kill = ^U; eof = ^D; eol = ; eol2 =
; s
Control: retitle -1 hurd: should clear discard-output flag when pseudoterminal
is destroyed
Samuel Thibault noted that pressing ^O a second time makes the
terminal work again. It turns out this is a documented feature.
https://www.gnu.org/software/libc/manual/html_node/Other-Special.html#index-V
I patched term/users.c (open_hook) so it clears the FLUSH_OUTPUT
flag if the tty was not already open.
I built hurd from commit bc170131016969f1d79409337833046ae1f4501b
(2016-06-14 "pfinet: fix memory leak") + the following changes:
* debian/patches/exec_filename_*.patch from the Debian hurd
1:
Control: retitle -1 term should clear discard-output flag in more situations
Although the FreeBSD manual page for termios(4)
https://www.freebsd.org/cgi/man.cgi?query=termios&apropos=0&sektion=4&manpath=FreeBSD+10.3-RELEASE+and+Ports&arch=default&format=html
documents the DISCARD key, the kernel d
1 - 100 of 254 matches
Mail list logo
