Package: module-init-tools
Version: 3.2-pre9-1
Severity: important
Module-init-tools diverts /usr/share/man/man8/ksyms.8.gz and kallsyms.8.gz
to ksyms.modutils.8.gz and kallsyms.modutils.8.gz. This is unneccessary, and
so is a violation of Debian policy on diversions, since module-init-tools
On 04/10/05, Karl Chen [EMAIL PROTECTED] wrote:
gecko mis-parses the following:
!DOCTYPE html PUBLIC -//W3C//DTD XHTML 1.0 Transitional//EN
http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd;
htmlfoo
!-- a -- b --
bar/html
firefox, galeon, etc. display:
foo -- a -- b -- bar
Please also include upstream changelog (this is, actually, a policy
requirement).
Package: libbz2-1.0
Version: 1.0.2-10
Severity: minor
libbz2-1.0 suggests libbz2-dev, but libbz2-dev doesn't enhance the
functionality of libbz2-1.0.
-- System Information:
Debian Release: testing/unstable
APT prefers testing
APT policy: (500, 'testing')
Architecture: i386 (i686)
Shell:
Package: xpdf-reader
Version: 3.01-2
Severity: normal
'xpdf foo bar' opens foo and silently ignores the second argument. Expected
behaviour: xpdf should complain about bar not being a number.
(Fixing #160258 instead would be, of course, even better.)
-- System Information:
Debian Release:
Package: xpdf-reader
Version: 3.01-2
Severity: normal
$ ls -l
total 332
-rw--- 1 hippy hippy 195872 Oct 25 00:33 FOO
-rw--- 1 stoner stoner 141902 Oct 25 00:33 foo
$ who am i
hippytty10Oct 24 19:43
In this situation, 'xpdf foo' opens FOO!
-- System Information:
Debian
I believe that xarchon really should recommend the font package it
uses. A fat collection of xfonts-* packages isn't something that any
sane system should have anymore, since nowadays programs use freetype
instead.
By the way, how did this bug end being tagged 'patch'?
Package: help2man
Version: 1.36.1
Severity: normal
Tags: l10n
In 1.36.1, the following snippet was added to help2man.PL:
# Add default territory to locale.
$locale .= _\U$locale if $locale =~ /^[a-z]{2}$/;
This is hopelessly broken for many languages. Uppercasing an ISO 639-1 code
often
On 12/10/05, Alexander Gattin [EMAIL PROTECTED] wrote:
On Wed, Oct 12, 2005 at 09:39:59AM +0200, Piotr Engelking wrote:
Please also include upstream changelog (this is, actually, a policy
Please, specify which chapter of debian policy are you
referring to.
requirement).
BTW
reopen 402558
severity 402557 serious
severity 402558 serious
thanks
Etch RC policy § 2:
Packages must not install programs in the default PATH with
different functionality with the same file name, even if they
Conflict:.
reopen 382923
thanks
The file has been uploaded under an incorrect name, 'debian/po/cz.po'.
The correct ISO 639-1 code for the Czech language is 'cs'.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
severity 383467 grave
tags 383467 + security lfs upstream
There are two problems here:
* Upstream bug: The 'generic' target silently miscalculates large file sizes.
* Packaging bug: On Linux, debian/rules runs builds the 'generic'
target instead of the 'linux' one.
The attached patch fixes the
Package: evince
Version: 0.4.0-2+b1
Severity: normal
Attempt to open a gzipped file in evince results in following error:
Unable to open document
Unhandled MIME type: 'application/x-gzip'
This is quite inconvenient, considering that documentation shipped in Debian
in formats supported by
reopen 396949
thanks
Version 2.8.7dev2 attempts to fix the bug by checking if the user owns
the .mime.types and .mailcap files before opening them. The assumption
that user trusts the files he owns, is, however, flawed. Consider, for
an example:
* files downloaded by the user
* files unpacked
Package: lynx-cur
Version: 2.8.7dev1-1
Severity: grave
Tags: security
Justification: user security hole
Lynx attempts to use the .mime.types and .mailcap files located in the
current directory:
$ strace lynx -dump 21 | grep '^open([^/]'
open(.mailcap, O_RDONLY) = -1 ENOENT (No such
Package: linux-source-2.6.18
Version: 2.6.18-4
Severity: serious
Justification: no longer builds from source
debian/patches/bugfix/copy-user-highpage.patch adds fourth argument to
copy_user_highpage(), but fails to update function call in mm/hugetlb.c
accordingly:
CC mm/hugetlb.o
maximilian attems [EMAIL PROTECTED] wrote:
forcemerge 397139 397207
stop and over
can you please look for duplicates, before reporting!
Hmm... For some strange reason http://bugs.debian.org/src:linux-2.6
redirects to
http://bugs.debian.org/cgi-bin/pkgreport.cgi?src=linux-2.6;dist=unstable,
Package: zsnes
Version: 1.420-2.1
Severity: normal
Some useful documentation, such as docs/Linux/readme.htm, is not included in
the binary package. Please consider including it.
-- System Information:
Debian Release: testing/unstable
APT prefers testing
APT policy: (500, 'testing')
Package: nvi
Version: 1.79-24
Severity: minor
nvi ships an useless empty directory: /usr/share/man/cat1
-- System Information:
Debian Release: testing/unstable
APT prefers testing
APT policy: (500, 'testing')
Architecture: i386 (x86_64)
Shell: /bin/sh linked to /bin/dash
Kernel: Linux
Package: libnids-dev
Version: 1.20-3
Severity: minor
Tags: patch
The libnids(3) manpage lists an incorrect return type for the
nids_register_chksum_ctl() function. The attached patch fixes this problem.
-- System Information:
Debian Release: testing/unstable
APT prefers testing
APT policy:
tags 372536 + patch
thanks
The problem comes from the fact, that modern Linux kernels often don't
calculate the checksum for outgoing packets, and that libnids by
default filters packets with incorrect checksum. The attached patch
disables this filtering.
A more fine-grained solution might be
Package: mimms
Version: 2.0.1-1
Severity: normal
Please include the upstream changelog in the package, as Debian policy 12.7
requests.
-- System Information:
Debian Release: testing/unstable
APT prefers testing
APT policy: (500, 'testing')
Architecture: i386 (x86_64)
Shell: /bin/sh linked to
tags 365626 + patch
tags 365632 + patch
thanks
The attached patch fixes both problems. If you would prefer it
splitted, just say so.
diff -ur netcat-1.10-32/netcat.c netcat/netcat.c
--- netcat-1.10-32/netcat.c 2006-05-01 18:46:54.0 +0200
+++ netcat/netcat.c 2006-05-01 19:36:11.0
Package: openswan
Version: 1:2.4.6+dfsg-1
Severity: minor
$ ls -ld /etc/ipsec.d/examples/
drwxrwxr-x 2 root root 232 Sep 5 05:34 /etc/ipsec.d/examples/
$
It seems that, on i386, openswan has been built with weird umask.
While we are at it, lintian reports some errors in your package, perhaps
Package: cpuid
Version: 3.3-4
Severity: minor
The man page for cpuid says:
This manual page was written for the Debian distribution because the
original program does not have a manual page. Instead, it has
documentation in the GNU Info format; see below.
The info documentation you
Package: java-common
Version: 0.23
Severity: wishlist
Currently, java-common suggests using equivs to create dummy java packages.
It should probably suggest using java-package instead. As far as I
understand, it is the preferred way of installing non-free java on Debian
nowadays.
-- System
Package: firefox
Version: 1.0.7-1
Severity: normal
When Firefox executes external programs, it passes to them all file
descriptors it has open. This is undesirable, especially in the case of
/dev/dsp, since two programs cannot use it at the same time. This bug is
present in version 1.5, as well.
Package: rrdcollect
Version: 0.2.3-2
Severity: minor
Your package depends on libpcap on i386 (the version built by you) but not
on other architectures (the auto-built ones). If you consider packet capture
support useful, please explicitly enable it, and update build dependencies
accordingly. If
On 29/10/05, Brendan O'Dea [EMAIL PROTECTED] wrote:
On Sat, Oct 29, 2005 at 12:21:29PM +0200, Piotr Engelking wrote:
In 1.36.1, the following snippet was added to help2man.PL:
# Add default territory to locale.
$locale .= _\U$locale if $locale =~ /^[a-z]{2}$/;
This is hopelessly
On 20/01/06, Eric Dorland [EMAIL PROTECTED] wrote:
Yes, this would be cleaner, but what exactly is the problem? just
passing the open file descriptor doesn't mean the execed process has
access to it, unless it went through it's descriptor list looking for
it.
The problem is that now xpdf
Package: joe
Version: 3.5-1.1
Severity: normal
When scrolling a text one line at a time, joe pads some lines with trailing
spaces to the length of the lines replaced during scrolling. These extra
spaces are visible to the X text selection.
The problem does not seem to appear in nvi, so I guess
Package: keytouch
Version: 2.3.2-2
Severity: serious
Justification: Policy 12.5, 12.7
The keytouch binary package doesn't ship debian/changelog nor
debian/copyright, which is a violation of section 1 of the RC policy.
-- System Information:
Debian Release: lenny/sid
APT prefers testing
APT
Package: gnome-panel
Version: 2.20.2-2
Severity: normal
Tags: l10n
The clock applet uses the locale setting of LC_MESSAGES to decide which date
format to use. It should use LC_TIME, instead.
Steps to reproduce:
With the following locale settings:
$ locale | egrep 'TIME|MESSAGES'
Package: gnome-panel
Version: 2.20.2-2
Severity: wishlist
The clock applet has an option to display the date in a custom format. This
option is, however, not visible to the user. It can be exposed by setting
the /apps/panel/applets/clock/prefs/format gconf key to 'custom', but
neither this
tag 436008 patch
thanks
The bug is caused by hex-a-hop unnecessarily initializing the SDL
cdrom subsystem. The attached patch fixes this problem.
diff -ur hex-a-hop-0.0.20070315/gfx.cpp hex-a-hop/gfx.cpp
--- hex-a-hop-0.0.20070315/gfx.cpp 2007-07-06 16:18:54.0 +0200
+++ hex-a-hop/gfx.cpp
As this bug is actually a more general privacy issue caused by the
fact that oprofiled makes the collected samples world-readable, I'd
suggest to make them owned and readable only by root:adm, similarly to
other log files containing sensitive information.
--
To UNSUBSCRIBE, email to [EMAIL
Package: oprofile
Version: 0.9.2-4
Severity: normal
The upstream changelog seems to be incomplete - 2004 and 2005 changes are
undocumented. ChangeLog refers to ChangeLog-2005, which doesn't exist.
-- System Information:
Debian Release: lenny/sid
APT prefers testing
APT policy: (500,
Package: oprofile
Version: 0.9.2-4
Severity: minor
Much of the Debian-specific documentation, including almost the entire
contents of the HOWTO-* files is kernel-2.4-specific and as such no longer
applicable to lenny.
Timestamps in the HOWTO-* files don't include years, by the way.
-- System
Package: oprofile
Version: 0.9.2-4
Severity: normal
At the moment, oprofile includes three binary packages:
* oprofile, containing the documentation
* oprofile-common, containing the binaries
* oprofile-gui, containing the x11 gui
with oprofile depending on the other two. As a result, the user
John Levon [EMAIL PROTECTED] wrote:
Please explain the privacy issue. Are you trying to suggest that the
profile of an application might reveal something private to a user? Note
there is no way to correlate such a profile back to an individual user.
It can reveal, e.g., which library and
Package: hex-a-hop
Version: 0.0.20070315-5
Severity: wishlist
Tags: patch
The attached patch adds mouse wheel support for the help system.
-- System Information:
Debian Release: lenny/sid
APT prefers testing
APT policy: (500, 'testing')
Architecture: i386 (x86_64)
Kernel: Linux 2.6.22
Package: hex-a-hop
Version: 0.0.20070315-5
Severity: minor
Tags: patch l10n
The Info: string in menus.h, used as a default hint title the first time a
hint is displayed, is not gettextized. The attached patch fixes this
problem.
-- System Information:
Debian Release: lenny/sid
APT prefers
reassign 437709 libdb4.5
found 437709 4.5.20-3
severity 437709 serious
thanks
Version 4.5.20-3 of libdb4.5 has introduced symbol versioning, but
hasn't bumped the shlibs accordingly. This is a violation of section 2
of lenny RC policy, and has a potential for serious breakage when a
package
Package: libc6
Version: 2.6.1-1
Severity: normal
The i386 dhelp 0.5.24-0.1 package in lenny has been built against libdb4.5
4.5.20-5, which uses symbol versioning, but libdb4.5 4.5.20-1 in lenny
doesn't use it (see bug 437709). Such a combination works, but when LD_DEBUG
is used, ld.so segfaults:
Clint Adams [EMAIL PROTECTED] wrote:
Googling for lenny rc policy, I only find mails from you. Where can I
find this document?
http://release.debian.org/etch_rc_policy.txt
(Despite the name, it applies to lenny as well, for now. Sorry for confusion.)
--
To UNSUBSCRIBE, email to [EMAIL
Package: hex-a-hop
Version: 0.0.20070315-3
Severity: minor
Tags: patch
One of the level names is displayed as Toughened\n\nTiles, with
superfluous end of line. The attached patch fixes this problem.
-- System Information:
Debian Release: lenny/sid
APT prefers testing
APT policy: (500,
Package: hex-a-hop
Version: 0.0.20070315-3
Severity: normal
The font introduced in 0.0.20070315-3 is somewhat less readable compared to
earlier versions. I'd suggest using the same typeface that upstream uses. If
there is some particular reason not to do so, bold variant of the current
font would
Package: hex-a-hop
Version: 0.0.20070315-3
Severity: normal
Tags: patch
debian/patches/compiler_warnings.patch, introduced in 0.0.20070315-3, has
changed an assignment of global variable into an assignment of local
variable, and, as a result, broken credits displayed when you complete all
levels.
Jens Seidel [EMAIL PROTECTED] wrote:
To be honest I do not know how to change the font. It may indeed depend
on your local font settings.
I fear I have no idea how sdl-pango handles fonts, either.
But do you really consider the current font so bad? It is a little bit
different from the
Jens Seidel [EMAIL PROTECTED] wrote:
PS: Please note that the files are in DOS mode. That's why you have to
sedn patches in binary mode (e.g. base64 encoding). Otherwise they
cannot be applied.
Hmm... I have just downloaded the attachments from the BTS and they
seem to contain dos line
Jens Seidel [EMAIL PROTECTED] wrote:
PS: Please note that the files are in DOS mode. That's why you have to
sedn patches in binary mode (e.g. base64 encoding). Otherwise they
cannot be applied.
Hmm... I have just downloaded the attachments from the BTS and they
seem to contain dos
reopen 441030
reassign 441030 hobbit-client
retitle 441030 hobbit-client: bb(1) conflicts with bb(6) in package bb
severity 441030 serious
thanks
RC policy § 2:
Packages must not install programs in the default PATH with
different functionality with the same file name, even if they
Package: bb
Version: 1.3rc1-8
Severity: important
The linux driver of bb doesn't work correctly if the console is in utf-8
mode. Texts (such as 'Music?[Y/n]') are displayed as garbage.
-- System Information:
Debian Release: lenny/sid
APT prefers testing
APT policy: (500, 'testing')
I have fixed two typos in the Polish localization and made a few
punctuation and whitespace corrections. Please apply the attached
patch.
l10n-pl-fixes.patch.gz
Description: GNU Zip compressed data
Package: hex-a-hop
Version: 0.0.20070315-3
Severity: wishlist
Tags: patch l10n
A Polish localization of hex-a-hop is included. Comments:
* In one place it was necessary to replace gettext() with ngettext(). Patch
attached, with all the other translations fixed, as well. The Russian
Package: groff-base
Version: 1.18.1.1-12
Severity: wishlist
Please consider changing the default justification mode of man pages
rendered for a device using monospaced fonts (such as a terminal) from both
margins to left margin only. The current default causes spacing between
words to vary
Package: apt
Version: 0.7.3
Severity: minor
Almost the entire content of the COPYING file, copied verbatim into
debian/copyright, is devoted to custom licensing terms that are no longer in
force since 2000, and, as such, are no longer useful. For the sake of
simplicity, I'd suggest removing them.
Package: bugs.debian.org
Severity: normal
http://bugs.debian.org/ eats first character of every line of reported bugs,
if it is a space. Compare, for an example:
* http://bugs.debian.org/cgi-bin/bugreport.cgi?msg=5;att=0;bug=415057
(rendered correctly)
and:
*
Package: fetchmail
Version: 6.3.6~rc3-1
Severity: minor
/etc/init.d/fetchmail creates the idfile, with comment create $UIDL if it
doesn't exist, because the daemon won't have the permission. This is no
longer useful, as since version 6.3 fetchmail doesn't write directly to the
idfile, but
Eric Dorland [EMAIL PROTECTED] wrote:
* Piotr Engelking ([EMAIL PROTECTED]) wrote:
When asked to set the default browser in Gnome
(Edit-Preferences-Main-Check Now-Yes), Iceweasel sets it to
'/usr/lib/iceweasel/firefox %s'. Since /usr/lib/iceweasel/firefox doesn't
exist, this command always
Package: iceweasel
Version: 2.0.0.1+dfsg-1
Severity: important
I noticed that one of the children of hanged Iceweasel was netstat (in the
zombie state). It seems that security/nss/lib/freebl/unix_rand.c runs
netstat -ni, despite debian/changelog claiming otherwise:
$ fgrep netstat
Package: fetchmail
Version: 6.3.6~rc3-1
Severity: normal
Please include the upstream changelog in the package, as Debian policy 12.7
requests.
-- System Information:
Debian Release: 4.0
APT prefers testing
APT policy: (500, 'testing')
Architecture: i386 (x86_64)
Shell: /bin/sh linked to
Package: openoffice.org-core
Version: 2.0.4.dfsg.2-5
Severity: normal
Open Office attempts to use libgcj.so.6, which doesn't exist in etch.
Exporting a file as XHTML (File - Export) results in the following error:
JRE is Defective
OpenOffice.org requires a Java runtime environment (JRE)
Rene Engelhard [EMAIL PROTECTED] wrote:
Which JRE do you try to use? Which JREs are installed?.
Which libgcjs are there?. Do you have java-gcj-compat installed?
Is your gcj default 4.x (which it should be)?
The only JRE I have installed is java-gcj-compat (1.0.65-10) + gij-4.1
(4.1.1-20).
I
Jim Watson [EMAIL PROTECTED] wrote:
On 26/03/2007, at 8:48 PM, Piotr Engelking wrote:
I have tried running ooo with an empty profile and now it writes
'javaldx: Could not find a Java Runtime Environment!' to stderr on
startup.
After that, did you try Tools-Options-OOo-Java ?
Yes
Package: libc6
Version: 2.3.6.ds1-13
Severity: normal
The result of strxfrm() varies depending on the length of the destination
buffer. In particular, in the pl_PL.UTF8 locale, strxfrm(buf, a, i)
returns 6 for i = 6, and 5 for i 6 (and the string placed in the
destination buffer does, indeed,
Package: python2.4
Version: 2.4.4-2
Severity: important
Tags: security patch
In Modules/_localemodule.c, PyLocale_strxfrm() miscalculates the length of
the strxfrm() destination buffer, which causes the function to return a
wrong string, and to read past the destination buffer, which may (and
Package: alpine
Version: 0.83+dfsg-1
Severity: wishlist
As alpine is a drop-in replacement for pine, please make 'pine' an alias for
alpine (through the alternatives system, preferably).
-- System Information:
Debian Release: 4.0
APT prefers testing
APT policy: (500, 'testing')
Architecture:
Package: alpine
Version: 0.83+dfsg-1
Severity: normal
At startup, after processing INBOX, alpine displays the following warning:
Folder vulnerable - directory /var/spool/mail must have 1777 protection
However, according to the Debian policy, /var/mail should have access
permissions set to
Package: alpine
Version: 0.83+dfsg-1
Severity: important
Attempt to send mail from alpine with default configuration results in the
following error:
Error sending: No default posting command.
-- System Information:
Debian Release: 4.0
APT prefers testing
APT policy: (500, 'testing')
Package: alpine
Version: 0.83+dfsg-1
Severity: normal
$ alpine -d 1
Argument Error: unknown flag d, debugging not compiled in
[...]
Alpine seems to have been built with --disable-debug. Disabling this
feature doesn't have any benefit other than making the binary slightly
smaller, and makes it
Santiago Vila [EMAIL PROTECTED] wrote:
On Sat, 10 Mar 2007, Piotr Engelking wrote:
As alpine is a drop-in replacement for pine, please make 'pine' an alias for
alpine (through the alternatives system, preferably).
alpine is not a drop-in replacement for pine yet. It does not have
maildir
I didn't notice this bug is related to #405689. (I should have read
the changelog before submitting, sorry.)
Anyway, to fix bug #405689 it should be enough to build with
--with-debug-level=0, instead.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact
Package: alpine
Version: 0.83+dfsg-1
Severity: wishlist
Tags: l10n
Please include iso-8859-2 in the default posting-character-set. It is the
recommended character encoding in the pl.* hierarchy of Usenet.
-- System Information:
Debian Release: 4.0
APT prefers testing
APT policy: (500,
Santiago Vila [EMAIL PROTECTED] wrote:
I don't understand this request. Why iso-8859-2 and not iso-8859-3,
iso-8859-4, or iso-8859-5?
[ pine uses iso-8859-1, but only as a better default than us-ascii ]
The Posting-Character-Set is used when sending messages. The default
behavior obtained
Asheesh Laroia [EMAIL PROTECTED] wrote:
Piotr, I'll close this bug, since there's nothing Alpine can do to
ensure it sends email in the correct way per site. I will however make a
note of this in README.Debian in the next release. If you think I should
do something else/different, do reply and
Package: bsdmainutils
Version: 6.1.6
Severity: minor
Tags: patch l10n
At the moment, cal pads the weekday abbreviations on the right:
marzec 2007
Pn Wt Śr Cz Pt So N
1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31
Please pad them on the
Package: iceweasel
Version: 2.0.0.2+dfsg-3
Severity: minor
Opening the help menu results in the following error:
Error: checkForUpdates has no properties
Source File: chrome://browser/content/utilityOverlay.js
Line: 401
(To make it appear in the error console, you need to use the
Package: gnome-applets
Version: 2.14.3-4
Severity: serious
Justification: Policy 2.2.1
The gweather/Locations.xml file seems to be autogenerated from
gweather/Locations.xml.in, which isn't included in the Debian source
package. Please include it.
(While this is a violation of section 1 of the
Steve Langasek [EMAIL PROTECTED] wrote:
On Sun, Mar 18, 2007 at 10:58:20AM +0100, Piotr Engelking wrote:
The gweather/Locations.xml file seems to be autogenerated from
gweather/Locations.xml.in, which isn't included in the Debian source
package. Please include it.
Based on what? The only
close 348420 9.0.21.55.1
thanks
Since Flash 9 on Linux has switched to ALSA, holding the sound device
open is no longer a problem.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Package: iceweasel
Version: 2.0+dfsg-1
Severity: normal
When asked to set the default browser in Gnome
(Edit-Preferences-Main-Check Now-Yes), Iceweasel sets it to
'/usr/lib/iceweasel/firefox %s'. Since /usr/lib/iceweasel/firefox doesn't
exist, this command always fails.
Moreover, on subsequent
Package: qemu
Version: 0.8.2-5lenny1
Severity: normal
Qemu uses ctrl+alt+f to switch between windowed and fullscreen mode. The
emulated system, however, receives the keypress as well. This is highly
inconvenient, since many programs interpret such a key sequence as a
command.
Steps to
Package: localechooser
Version: 1.37
Severity: normal
Tags: patch
ME (Montenegro) and RS (Serbia) are missing from the regionmap. As a result,
localechooser shows them in the 'other' category, instead of 'Europe'.
The attached patch fixes this problem and removes an obsolete CS (Serbia and
Package: localechooser
Version: 1.37
Severity: normal
Localechooser doesn't display AX (Åland Islands) nor CI (Côte d'Ivoire)
entries, presumably because their names contain non-US-ASCII characters.
-- System Information:
Debian Release: lenny/sid
APT prefers testing
APT policy: (500,
Package: localechooser
Version: 1.37
Severity: minor
In the graphical installer, native names of languages are misaligned.
Localechooser attempts to align them by displaying a fixed number of
characters before the dash, which doesn't work, since the graphical
installer doesn't use monospace
Package: localechooser
Version: 1.37
Severity: normal
Tags: l10n
In non-English locales, countries are sorted by their English name. Aside
from being ugly, it makes the country chooser hard to use.
-- System Information:
Debian Release: lenny/sid
APT prefers testing
APT policy: (500,
Package: localechooser
Version: 1.37
Severity: normal
Tags: patch
Several countries in the regionmap are misplaced:
[territory: incorrect location - correct location]
* CC (Cocos Islands): Oceania - Indian Ocean
* CX (Christmas Island): Oceania - Indian Ocean
*
The missing patch. :P
diff -ur localechooser-1.37/regionmap localechooser/regionmap
--- localechooser-1.37/regionmap 2006-07-26 00:49:25.0 +0200
+++ localechooser/regionmap 2007-05-31 16:54:51.0 +0200
@@ -1,4 +1,3 @@
-ML -- Africa --
AO -- Africa --
BF -- Africa --
BI -- Africa
On 31/05/07, Christian Perrier [EMAIL PROTECTED] wrote:
* GL (Greenland): Europe - North America
GL is an associated terortiry of Denmark which makes the probability
for people to seek for it in Europe pretty high.
This is, in my opinion, too much an overstretch. Under
On 02/06/07, Christian Perrier [EMAIL PROTECTED] wrote:
Wile we're at it, I think that, following the same definition of the
Antarctic Ocean, the Bouvet Island (54°S) should be considered as part of
Atlantic Ocean, not Antarctica.
The same goes for South Georgia (54°S) and the South Sandwich
Package: debsecan
Version: 0.4.7
Severity: normal
The debconf template asks whether to track sarge, etch, or sid, but doesn't
give an option to track lenny.
By the way, perhaps it might be a better idea to ask to choose between
oldstable, stable, testing, and unstable instead of (or in addition
reopen 418247
thanks
The bug doesn't seem to be fixed:
$ fgrep sid /var/lib/dpkg/info/debsecan.*
/var/lib/dpkg/info/debsecan.config: sarge|etch|sid)
/var/lib/dpkg/info/debsecan.templates:Choices: GENERIC, sarge, etch, sid
$ dpkg-query -W debsecan
debsecan0.4.8
$
--
To UNSUBSCRIBE,
Package: amd64-libs
Version: 1.2
Severity: important
Your version of ldd doesn't print anything on setgid programs.
$ ldd /usr/bin/crontab
$ echo $?
1
$ ldd.i386 /usr/bin/crontab
linux-gate.so.1 = (0xe000)
libpam.so.0 = /lib/libpam.so.0 (0xb7fa3000)
libselinux.so.1 =
On 9/12/05, Daniel Jacobowitz [EMAIL PROTECTED] wrote:
On Mon, Sep 12, 2005 at 02:47:26AM +0200, Piotr Engelking wrote:
Your version of ldd doesn't print anything on setgid programs.
It does for me:
[EMAIL PROTECTED]:~% ls -l =crontab
-rwxr-sr-x 1 root crontab 26872 Feb 16 2005 /usr
tags 208532 patch
thanks
diff -ur dpkg-1.13.11/lib/database.c dpkg/lib/database.c
--- dpkg-1.13.11/lib/database.c 2005-06-06 06:07:12.0 +0200
+++ dpkg/lib/database.c 2005-09-16 19:58:28.0 +0200
@@ -100,9 +100,11 @@
pkg-eflag != eflagv_ok ||
pkg-status !=
Package: libpcap0.8
Version: 0.9.8-3
Severity: wishlist
Please introduce a flag to mark captured outgoing packets whose checksum is
not correctly calculated yet (because it will be done in hardware), both in
the live packet capture interface and in the packet dump file. Such a
feature would be
block 420129 by 481240
severity 420129 important
thanks
The behaviour you observe is caused by the fact that libnids by default
filters packets with incorrect checksum, and that, if checksum calculation
is done in hardware, libpcap is only able to capture outgoing packets before
the checksum is
Lubomir Kundrak [EMAIL PROTECTED] wrote:
Piotr: Could you please provide a reproducer, or a string/locale couple
that triggered th bug for you?
Ok, sorry for being so terse in the original report:
$ cat foo
#!/usr/bin/python
import locale
print locale.setlocale(locale.LC_COLLATE,
I believe that this bug has been resolved incorrectly. Section 10.4 of the
Debian Policy says:
All command scripts, including the package maintainer scripts inside the
package and used by dpkg, should have a #! line naming the shell to be
used to interpret them.
In the case of
1 - 100 of 349 matches
Mail list logo
