Package: ttf-bitstream-vera
Version: 1.10-8.2
The character U+00A0 NO-BREAK SPACE (unbreakable space) is wider than
U+0020 SPACE (normal space) in fonts Bitstream Vera Sans and Bitstream
Vera Serif.
This is evidenced by running, say:
LC_ALL=C.UTF-8 pango-view --font="Bitstream Vera Serif 80" -q
> Index: sispmctl-3.1/src/sispm_ctl.c
> ===
> --- sispmctl-3.1.orig/src/sispm_ctl.c
> +++ sispmctl-3.1/src/sispm_ctl.c
Apologies for this part of the bug-report, which was included in
error, and which has nothing to do with mdadm (it
Package: mdadm
Version: 3.4-4+b1
The mdadm tool uses a config file, generally called mdadm.conf, to
configure how arrays are to be created. Many of the options to mdadm
--assemble can be set via this config file, but, inexplicably, not
all, and in particular, there is no config file equivalent to
Package: sispmctl
Version: 3.1-1
The source file src/sispm_ctl.c of sispmctl contains the following
line (where buffer is an array of unsigned char):
wmemset((wchar_t*)(buffer+5), (wchar_t)0x3FFF3FFF, (0x27-5)/sizeof(wchar_t));
The author's intent was to fill bytes 0x05 inclusive through 0x25
uot; by Chris Liddell in the above-mentioned bug report
page, but I think it is perfectly sensible. Otherwise, backport the
patch made to the ghostscript upstream itself.
--
David A. Madore
( http://www.madore.org/~david/ )
Description: fix ghostscript bug #695862 by using error-checking
On Sat, Sep 17, 2016 at 11:18:20PM +0300, Paul Fertser wrote:
> Hm, can't see anything good about it...
Ah, sorry, as I mentioned, behaviour is somewhat erratic even with
openocd-0.5.0. This log is definitely good (I was able to upload
U-Boot). Maybe the comparison between the two can be instruc
On Fri, Sep 16, 2016 at 06:25:27PM +0300, Paul Fertser wrote:
> In this case it would help to see -d3 output.
Attached are the output of
/tmp/openocd-0.5.0/src/openocd -s /tmp/openocd-0.5.0/tcl -f
/tmp/openocd-0.5.0/tcl/board/sheevaplug.cfg -d3
("openocd-good.log"), which is version 0.5.0-1 fro
On Fri, Sep 16, 2016 at 11:13:53AM +0300, Paul Fertser wrote:
> On Fri, Sep 16, 2016 at 09:55:21AM +0200, David Madore wrote:
> > A SheevaPlug JTAGKey FT2232D (USB identifiers 9e88:9e8f) being
>
> Please also provide lsusb -vvv data for this device.
Bus 003 Device 002: ID 9e8
Package: openocd
Version: 0.8.0-4
(I am reporting this against openocd version 0.8.0-4 because that is
what is found in Debian jessie, currently stable, but I also tried
0.9.0-1+b1, the version currently in sid.)
A SheevaPlug JTAGKey FT2232D (USB identifiers 9e88:9e8f) being
connected, version 0.
Package: emacs24
Version: 24.4+1-5
To reproduce: create an empty .html file (touch /tmp/blah.html) then
open it with Emacs in a terminal (emacs -q -nw /tmp/blah.html). Now
type C-c C-c h to insert a link and try entering a URL with a '?'
(question mark), say "https://www.google.com/search?q=emacs
Package: strace
Version: 4.5.20-2.3
On 32-bit systems, strace is not compiled with -D_FILE_OFFSET_BITS=64
(and the essential message of this bug-report is that ALL programs
that use stat() in any way whatsoever should always be compiled with
this define).
As a result, stat() can fail with EOVERFL
reopen 664854
thanks
Sorry for being slow at replying. I tried versions 13.003 and 13.008
of kernel-package, and the armel architecture is still broken:
* in 13.003, the arches/armel.mk rule file was re-added, but
architecture.mk does not include it, so it is inoperative;
* and in 13.008, it
Package: isc-dhcp-client
Version: 4.2.2.dfsg.1-5+deb70u6
Severity: major
My server's dhclient regularly dies (every few days or weeks) with the
fatal error message "Impossible condition at dhc6.c:4111."
It is started with dhclient -cf /etc/dhcp/dhclient.conf -6 -P -v eth0
The config file is as f
Attached is a(n absolutely trivial) patch that adds the IUTF8 terminal
mode (with code 42) to OpenSSH. Prebuilt Debian packages for those
architectures I was able to build for, for testing and stable
distributions, are available in ftp://quatramaran.ens.fr/pub/madore/misc/openssh-with-iutf8/
>.
Package: evince
Version: 3.4.0-3.1
Severity: minor
Open the file http://www.madore.org/~david/.misc/astral.pdf >
using evince: this contains a table of contents with a single entry
whose name is the same as on the document itself, i.e., "The name "
followed by the four characters U+10909 U+10904 U
Package: socat
Version: 1.7.1.3-1.4
Severity: wishlist
Hi,
Linux has an IP-level (level=SOL_IP) socket option called
IP_TRANSPARENT, which allows binding on a non-local address (with
superuser privileges) to implement transparent proxy support. It
would be useful if socat could have an option to
Just to confirm that Wolfram's patch works beautifully for me, and
fixes the utter brokenness of xpdf on Ubuntu Precise as well as its
not-so-obvious brokenness on Debian Wheezy. So, many thanks to him
for figuring this one out and fixing it!
Is there any hope of this fix entering Debian Wheezy b
On Sun, May 20, 2012 at 10:41:52AM -0700, Andrew Pollock wrote:
> On Sun, May 20, 2012 at 06:39:41PM +0200, David Madore wrote:
> > The best I
> > found was adding "exit 0" inside /etc/default/isc-dhcp-server but
&g
Package: isc-dhcp-server
Version: 4.2.2.dfsg.1-5
Severity: minor
Most Debian packages containing a daemon provide a way to not start
the daemon (for those people who want the daemon binary to be
available for whatever reason, and yet not run it, or at least not run
it with the default init script)
Package: ifupdown
Version: 0.7~alpha5+really0.6.16
Running ifup $DEV for a network device that does not yet exist in the
kernel, will be created by autoloading the netdev-$DEV module and will
be renamed by udev persistent-net.rules, incurs a race-condition: udev
does not instantaneously rename the
Package: kernel-package
Version: 12.036+nmu2
Severity: important
make-kpkg no longer recognizes "armel" as a valid parameter to the
--arch switch: passing it causes the following error when trying to do
anything:
debian/ruleset/misc/checks.mk:36: *** Error. I do not know where the kernel
image g
Package: wnpp
Severity: wishlist
* Package name: gcc-4.3
Version : 4.3.6
Upstream Author : many
* URL : http://gcc.gnu.org/gcc-4.3/
* License : GPLv3
Description : The GNU C Compiler version 4.3
Debian wheezy currently has packages for gcc versions (at le
Package: radvd
Version: 1:1.8-1.2
When radvd has been configured to listen on an interface but ignore if
missing, and when that interface is later brought up, a router
advertisement is sent for the prefix of the newly activated interface,
but it is sent to the wrong interface (i.e., some already e
Or, more generally, if someone can explain how to revert to pre-1.16
behavior, I'd be very grateful, because I have dozens of locally
compiled kernel packages installed on my system, causing every
invocation of dpkg to spew hundreds of lines of useless warnings,
making it nearly unusable (any valid
On Sat, Apr 09, 2011 at 10:32:14AM +0200, Guillem Jover wrote:
> What I've locally now is the following (might get slightly refined
> before the push though):
>
> --compare-versions is lax by default now, will still warn though.
> One of the reasons I ended up with the strict parser for
> --compar
Package: librsvg2-2
Version: 2.26.3-1
Consider the file: http://www.madore.org/~david/.tmp/svgbug.svg
> (which is rendered correctly, e.g., by Firefox-4.0): the upper and
lower parts of the image should be identical. However, RSVG renders
it as: http://www.madore.org/~david/.tmp/svgbug.png
>, t
Package: ntp
Version: 1:4.2.6.p2+dfsg-1+b1
The init scripts for ntp depend on the lockfile utilities to function
correctly. If they are not present, ntp will try to start but will
probably die because the socket is in use by ntpdate. So the ntp
package should at the very least recommend the lock
Package: gnucash
Version: 2.2.9-10
Transaction cut & paste, which used to work fine in previous versions
of GnuCash, now seems to fail systematically: cutting the transaction
works, but any attempt to paste it now produces the meaningless error
message "This transaction is already being edited in
Package: gap-core
Version: 4r4p12-1
Severity: minor
When run on a terminal with a large number of columns, gap is liable
to segfault when displaying a character table. I don't know the exact
circumstances, but the following seems to reproduce the problem: open
a terminal with 272 columns, open ga
Package: e2fsprogs
Version: 1.41.12-1
Severity: grave
Debian testing currently has e2fsprogs 1.41.12-1 and libblkid1
2.16.2-0 but the latter does not satisfy the dependencies of the
former. This makes mke2fs non-functional.
pleiades david ~ $ sudo mke2fs
mke2fs: /lib/libblkid.so.1: version `BLKI
Package: libreadline6
Version: 6.1-1
I believe the following behavior is libreadline6's fault because I
noticed it in every program linked against libreadline6 that I could
find, but I'm completely sure (maybe gdb, sqlite3 and the like are
using it incorrectly):
When starting up in an xterm, libr
On Thu, Mar 25, 2010 at 06:37:03PM +1100, Aníbal Monsalve Salazar wrote:
> On Wed, Mar 24, 2010 at 09:04:19PM +0100, David Madore wrote:
> >Whether this is a bug is perhaps debatable (because of the vagueness
> >of specifications in the matter), and maybe needs to be discussed
>
Hi,
On Mon, Mar 08, 2010 at 03:17:55PM +0100, Daniel wrote:
> Starting advanced IEEE 802.11 management: hostapdioctl[SIOCGIWRANGE]:
> Invalid argument
> ioctl[SIOCSIWRTS]: Invalid argument
> rmdir[ctrl_interface]: No such file or directory
> (warning).
I've encountered the exact same errors wit
Package: grep
Version: 2.5.4-4
Whether this is a bug is perhaps debatable (because of the vagueness
of specifications in the matter), and maybe needs to be discussed
upstream, but the following behavior is extremely confusing at the
very least:
vega david ~ $ locale
LANG=
LC_CTYPE=en_US.UTF-8
LC_
Package: vsftpd
Version: 2.3.0~pre1-1
There seems to be a trend in Debian and/or Linux to activate the
sysctl net.ipv6.bindv6only=1 by default, meaning that binding on the
IPv6 unspecified address (::) will not also bind IPv4 (contrarily to
the former Linux tradition). This means that running vsf
Package: libsocket6-perl
Version: 0.23-1
The SOL_IPV6(=IPPROTO_IPV6)-level constants for setsockopt which are
documented in ipv6(7) under Linux, many of which are common to other
operating systems, are not defined by the Socket6 module, as they
probably should. Of particular importance is IPV6_V6
Package: openssh-client
Version: 1:5.1p1-5
Severity: minor
ssh-keygen -r wants its input_keyfile in "public key file" format,
which differs from "known hosts" format by the fact that there is no
hostname at the beginning of the line. Unlike ssh-keygen -l, which is
lenient about the input file for
Package: hal-info
Version: 20091130-1
My keyboard and mice devices aren't properly configured by HAL: they
don't receive the info.category = 'input' key or the right
info.capabilities (and persumably many other keys are missing as a
consequence):
vega david ~ $ hal-find-by-capability --capability
Package: ttf-linux-libertine
Version: 4.4.1-1
The ttf-linux-libertine package was recently upgraded from 4.1.8-1 to
4.4.1-1; something in the changes brought by this upgrade has broken
XeTeX's ability to use "fi" ligature in the bold version of the font.
A minimal example showing the problem is t
Package: kernel-package
Version: 12.010
Option --overlay-dir appears to be broken (I try to use it to provide
a useful changelog; it used to be the case that the changelog file
could be modified between the configure and build stages, but now it
seems to be overwritten again, so I tried to fix thi
Package: xtrans-dev
Version: 1.2.3-3
The following code (lines 2058-2066 of Xtranssock.c) is wrong:
else if (olderrno == ENOENT || olderrno == ECONNREFUSED) {
/* If opening as abstract socket failed, try again normally */
if (abstract) {
Package: gnupg2
Version: 2.0.9-3.1
Severity: wishlist
The OpenPGP message format provides a way (RFC 4880 section 5.2.3.22)
to specify, when a signature is produced, which of the various uids of
a key is doing the signing. GnuPG does not seem to allow this, nor
does it display it when it is used.
Package: eog
Version: 2.22.3-2
Severity: minor
(I don't know whether this should be considered as a bug or whether
the developers consciously chose this, so please update severity
accordingly.)
When eog is invoked to open a single named file (eog /tmp/foo.jpg), it
nevertheless opens and reads eve
Package: coreutils
Version: 6.10-6
Severity: wishlist
dd is typically the sort of program where one would like to enter
numbers (for bs, count, skip and seek) in hex (or possibly octal)
format. Unfortunately, dd won't parse them:
pleiades david ~ $ dd if=/dev/zero bs=0x1000 count=1
dd: invalid n
On Fri, Jan 23, 2009 at 05:38:51PM +0100, Josselin Mouette wrote:
> Le mercredi 10 décembre 2008 à 17:01 +0100, David Madore a écrit :
> > the statement: "The gdk-pixbuf library does not need to be
> > initialized". This is false, as the following trivial program sh
Package: metacity
Version: 1:2.22.0-2
Severity: wishlist
I know metacity isn't meant to be configurable, but as the gconf
variable /apps/metacity/general/action_middle_click_titlebar exists,
it would be nice if this could take the value "move", to allow a
window to be moved with the middle mouse b
Package: libgtk2.0-dev
Version: 2.12.11-4
The page , or, alternatively, http://library.gnome.org/devel/gdk-pixbuf/2.12/apas02.html > contains
the statement: "The gdk-pixbuf library does not need to be
initialized". This is false, as the following trivial program shows:
/* --- cut after --- */
#i
Package: hal
Version: 0.5.11-6
The file /etc/udev/rules.d/90-hal.rules contains the line
RUN+="socket:/org/freedesktop/hal/udev_event"
Obviously the /org/freedesktop/hal/udev_event Unix domain socket does
not exist nor is it supposed to. However, the socket of the same name
exists in the abstra
Package: apt
Version: 0.7.14+b1
Severity: wishlist
It would be nice to have an option --ignore-nonexistent to instruct
apt-get to emit a non-fatal warning if any nonexistent packages are
requested (on install, remove, etc.) rather than terminating with an
error. E.g., "apt-get install --ignore-no
Package: ttf-sil-gentium
Version: 1.02+dfsg1-5
Severity: wishlist
It would be very helpful to add the characters U+0401 CYRILLIC CAPITAL
LETTER IO and U+0451 CYRILLIC SMALL LETTER IO to Gentium Regular,
because they are all that is missing to support Russian. Luckily,
these characters are also ve
Package: kvm-source
Version: 66+dfsg-1
I compiled kvm-source on a 2.6.24.4 kernel (with module-assistant) and
am using it on an Intel KVM (a Core 2 Quad Q6600 CPU) with the kvm and
kvm-intel modules loaded. Attempting to boot the EeeXUbuntu 7.10
release 3 boot CD (found at http://wiki.eeeuser.com
Package: seahorse
Version: 2.22.0-1
I discovered of seahorse's existence because it is started from
/etc/X11/Xsession.d/60seahorse (thus without much of my consent) and
then prevents ssh-agent from being started (because it sets
SSH_AUTH_SOCK). This would be fine if it behaved as a drop-in
replac
Package: net-tools
Version: 1.60-19
When more than 256 network interfaces exist or have existed on a given
system, /proc/net/if_inet6 contains an interface number (the second
column, immediately after the IPv6 address) which is no longer
formated as a two-digit hexadecimal number and this confuses
Package: manpages
Version: 2.78-1
The ipv6(7) manual page does not document the IPV6_V6ONLY socket
option, nor does it mention SOL_IPV6 as an alternative to IPPROTO_IPV6
for socket level. It would be nice if it could.
--
David A. Madore
([EMAIL PROTECTED],
http://www.madore.org/~d
Package: xinetd
Version: 1:2.3.14-1.1
xinetd does not bind to IPv6 addresses (and does not seem to have an
option to do so) when used in -inetd_compat mode. As current inetd's
are IPv6-aware, this is a problem: this means xinetd cannot be used as
a drop-in inetd replacement.
The attached patch i
Package: libc6
Version: 2.7-6
Hi,
I have a host with a (global, routable) IPv4 address and a 6to4
(2002::/16) IPv6 address that is derived from it (this is a quite
common setup), and I am trying to connect to another host with the
same setup in the DNS (that is, one A record for an IPv4 address a
Package: xclip
Version: 0.08-7
Hi,
If I select some Unicode characters that are not latin-1 in, say,
Firefox, and try to copy them to a file using "xclip -o > file", only
the latin-1 characters come through nicely, and they are encoded as
iso-8859-1, even if LC_CTYPE is set to en_US.UTF-8 (the ot
Package: socat
Version: 1.6.0.0-1
Severity: wishlist
Hi,
Is there some chance of having Stream Control Transmission Protocol
(aka SCTP) support in socat? (See, e.g., http://en.wikipedia.org/wiki/SCTP > for background on this.) At least
basic SCTP support should be no more difficult to write tha
On Fri, Dec 28, 2007 at 09:50:03PM +0100, David Madore wrote:
> The wireless configuration window for network-manager only offers the
> choice between "WEP key (hexadecimal)" and "WEP key (ascii)": no WPA
> here. Yet wpa_supplicant, when invoked manually (whether wi
Package: network-manager
Version: 0.6.5-3
The wireless configuration window for network-manager only offers the
choice between "WEP key (hexadecimal)" and "WEP key (ascii)": no WPA
here. Yet wpa_supplicant, when invoked manually (whether with -Dwext
or -Dmadwifi, see below for card model) works p
Package: linux-uvc-source
Version: 0.1.0.svn120-2
Why is the 04_makefile_no_depmod patch commented out? I really
believe it should be reinstated: calling depmod causes many problems,
notably if one is using the -l option of module-assistant (compiling
for a kernel version other than the one curre
Package: live-helper
Version: 1.0~a37-2
I'm trying to build a live USB stick that uses resolvconf whereas I
don't use it on the host system. So chroot/etc/resolv.conf is a
symlink to the (nonexistent) /etc/resolvconf/run/resolv.conf (this is
how resolvconf functions). The lh_chroot_resolv stage
Package: xbase-clients
Version: 1:7.2.ds2-2
Severity: normal
The /etc/X11/xinit/xserverrc file shipped by xbase-clients contains an
explicit dpi option to the X server: this means that starting X by
running "startx" will typically give fonts with wrong sizes.
--
To UNSUBSCRIBE, email to [EMAI
On Fri, Nov 16, 2007 at 05:35:36PM +0900, Asheesh Laroia wrote:
> Have you checked on the current alpine? I vaguely remember thinking
> about fixing it.
I checked 0.9+dfsg-1 from unstable.
> If it still doesn't work, a build-depend on sendmail should do the job,
> and I'll add that and upl
On Fri, Jun 01, 2007 at 11:52:03PM +0200, Reuben Thomas wrote:
> There seems to be no default posting method (surely should be
> sendmail?): if I try sending mail with no smtp server or posting command
> set, which should just use the system default, presumably sendmail, I
> get the error: "Error s
Package: gs-esp
Version: 8.15.3.dfsg.1-1
I do not know whether this bug should be filed against gs-esp or
ttf-kochi-gothic or something else: I hope someone can redirect it if
necessary.
Assuming ttf-kochi-gothic version 1.0.20030809-4 is installed,
ghostscript is unable to load the TTF font (Koc
Package: lilypond
Version: 2.10.29-1
lilypond 2.10.29-1 in Debian lenny (testing), with guile-1.8-libs
1.8.1+1-5 (the current version found in testing), segfaults at
startup.
Upgrading to guile-1.8-libs version 1.8.2+1-2 solves the problem (but
causes other problems in other packages, unfortunate
Package: linux-image-2.6.18-4-686
Version: 2.6.18.dfsg.1-12etch1
Hi,
My on-motherboard Ethernet adapter,
01:05.0 Ethernet controller: Marvell Technology Group Ltd. 88E8001 Gigabit
Ethernet Controller (rev 13)
is fully functional using the sk98lin driver but only partially
functional using skge
On Wed, Sep 07, 2005 at 12:38:07AM +0200, Kilian Krause wrote:
> Upon installation xinetd replaces the /etc/init.d/inetd, which may or
> may not be desired. Nevertheless the openbsd-inetd, which is needed for
> ipv6-enabled inetd environments uses /etc/init.d/openbsd-inetd which is
> left alone. Th
On Sat, Oct 28, 2006 at 06:23:50PM +0200, Marco d'Itri wrote:
> On Oct 28, David Madore <[EMAIL PROTECTED]> wrote:
> > If inetd can't yet be removed as a dependency of netbase, is it at
> > least possible to include xinetd as an alternative in this list?
> There
# Version: 4.27
# Depends: ifupdown (>= 0.6.4-4.9), openbsd-inetd | inet-superserver,
iputils-ping | ping, lsb-base (>= 3.0-6), debconf (>= 0.5) | debconf-2.0
If inetd can't yet be removed as a dependency of netbase, is it at
least possible to include xinetd as an alternative in this list?
--
Package: wodim
Version: 5:1.0~pre4-1.1
Severity: minor
growisofs has a -dvd-compat option, documented as follows:
-dvd-compat
Provide maximum media compatibility with DVD-ROM/-Video. In
write-once context (DVD+R and DVD-R) this results in unappend-
Package: xutils
Version: 7.1.ds-1
The "showrgb" program, which used to be in this package (at least
until version 4.3.0.dfsg.1-14sarge1, found in the Stable distribution)
seems to have vanished.
I'm not sure whether this should be filed as a bug-report against
xutils or whether it should be a new
On Fri, Aug 25, 2006 at 12:04:10PM +0200, Samuel Thibault wrote:
> There are some concerns with the window operations that XTerm
> emulates. CSI 21t (report window title) in particular, because since OSC
> 0/1/2 ST let you decide of the window title, one can decide what CSI 21t
> returns, which mig
I filed the bug upstream: http://bugzilla.elinks.or.cz/show_bug.cgi?id=811
>.
I also have a trivial patch (attached) that probably fixes this bug
(but perhaps not in The Right Way).
Cheers,
--
David A. Madore
([EMAIL PROTECTED],
http://www.madore.org/~david/ )
--- elinks-0.11.1.o
Package: zsh
Version: 4.3.2-17
Severity: minor
zsh seems to autocomplete non-printable characters by representing
them as something like $'\377', but then it fails to understand them
when autocompleting one further level of directories.
For example, suppose we start in an empty directory (/tmp/te
Package: qpsmtpd
Version: 0.32-3
If I try to add any more paths in /etc/qpsmtpd/plugin_dirs beyond the
default /usr/share/qpsmtpd/plugins, only the first listed path has any
effect. Trying to load a plugin from one of the other directories
listed in the file has the same effect as though the plug
Package: vlc
Version: 0.8.6-svn20060823.debian-3
Severity: minor
I believe the following packages are missing from vlc's
build-dependencies: freeglut3-dev and libncurses5w-dev (the latter
owing to the libcaca plugin; not sure about the former).
--
David A. Madore
([EMAIL PROTECTED],
Package: vlc
Version: 0.8.6-svn20060823.debian-3
Severity: minor
The (apparently handwritten) configure script in extras/x264/ uses
uname in various ways to detect host+build architecture (it apparently
won't allow for cross-compilation, but that's not Debian's problem).
This is a problem in at le
On Sun, Aug 27, 2006 at 06:46:06PM +0200, Mattia Dongili wrote:
> except I don't have any amd64 box and can't test it, nor create a safe
> config.amd64 :)
Ah, sorry. (I had thought Debian provided its maintainers with access
to test boxen of all supported architectures for such cases.)
> I'll be
Package: user-mode-linux
Version: 2.6.17-1um-2
Please provide an amd64 build of user-mode-linux. AFAICS, there is no
difficulty involved, the package builds correctly from the Debian
source with essentially no changes (just provide a config.adm64 file).
As a proof of concept, see the package file
On Wed, Jul 26, 2006 at 07:30:12AM -0400, Michael Stone wrote:
> I'll admit the "FUCK debian" commentary makes me less interested in
> working with you.
I'll offer apologies (in private) if you took this personally - it was
not meant as such. But nobody says you have to work with me: the
patch
Just a reminder for Debian users watching this bug: while waiting for
this to be integrated upstream (should be in forthcoming 6.x release
of coreutils), I maintain Debian packages which are identical to the
Sarge packages except they include the sha2 utilities. The packages
in question, along wit
On Tue, May 16, 2006 at 08:59:55PM -0500, Enrico Zini wrote:
> Thanks for packaging xlaby! It's *such a great toy*! I gave a talk
> about "Advanced Ways of Wasting Time" at Debconf6 and people got crazy
> about it.
Thanks. :-) I just read this report of yours (I was kiboifying on
Google groups s
On Fri, Oct 21, 2005 at 05:25:25PM -0700, Steve Langasek wrote:
>The obvious
> workaround here would be to not list apt sources in your configuration that
> you don't actually want apt to use...
So I guess my bug report wasn't clean e
Package: apt
Version: 0.6.41
Severity: serious
The --allow-unauthenticated option does not revert to the exact
pre-0.6.27 behavior: even when this flag is passed on the command
line, apt still favors an authenticated package on the network to an
unauthenticated package on the local host (say with
On Tue, Sep 13, 2005 at 08:11:49AM -0600, Barak Pearlmutter wrote:
> Sure, but there is an automated system that does the
> unstable-to-testing migration.
Ah, I see. Maybe it has something to do with the incompatibility of
libdjvulibre1 and libdjvulibre15 (bug #327190)? The apt-get
dist-upgrade
On Tue, Sep 13, 2005 at 07:38:56AM -0600, Barak Pearlmutter wrote:
> This isn't djview's fault, is it?
Shouldn't djview have remained in unstable as long as some of its
dependencies were still there? Maybe I'm misunderstanding the working
of Debian distributions, but surely migration must proceed
Package: djview
Version: 3.5.15-1
Severity: serious
djview version 3.5.15-1 found in "testing" depends on libqt3-mt which
is only available in "unstable". This is broken:
vega david ~ $ sudo apt-get -s install djview
Reading Package Lists... Done
Building Dependency Tree... Done
Some packages co
Package: bind9
Version: 9.3.1-2
Severity: minor
The README.Debian file for bind9 states that:
# Note that when running named as a user other than root, it will not be able
# to find new interfaces that appear dynamically, such as during a PCMCIA card
# insertion, or if you're running some flavors
On Tue, Aug 30, 2005 at 01:20:37PM +0900, Horms wrote:
> Appart from my general feeling that no one should use Reiser FS,
Why is that? I mean, certainly every filesystem has its problems, but
I don't think there's a consensus that ReiserFS is much worse than the
others? I personally find it usef
On Sat, Aug 27, 2005 at 04:30:51PM +, Eric Blake wrote:
> Would it make more sense to have a single sha utility that takes
> an argument of which algorithm to use, defaulting to the most
> secure, then the user can define wrappers to invoke a non-default
> algorithm? For example, 'sha --algori
Package: coreutils
Version: 5.2.1-2
Severity: wishlist
md5 has been broken for some time now; sha1 also has been, more
recently (an actual collision is not known, so far, but a way to
produce one in a "reasonable" amount of computer time has been
provided). Consequently, the md5sum and sha1sum ut
Package: elinks
Version: 0.10.4-7
Even when elinks is configured to use Setup -> Character set -> ISO
8859-1, it still does not send the non-breaking space (U+00A0)
character correctly in forms: for some reason, it sends ^A (0x01)
instead.
How to reproduce: go to http://www.madore.org/~david/.tes
Package: linux-image-2.6.12-1-386
Version: 2.6.12-5
Severity: minor
Tag: patch sid
Please include the following config snippet on all archs:
CONFIG_REISERFS_PROC_INFO=y
CONFIG_REISERFS_FS_XATTR=y
CONFIG_REISERFS_FS_POSIX_ACL=y
CONFIG_REISERFS_FS_SECURITY=y
(at least the last three lines). It ma
Package: libgimp-perl
Version: 2.0.dfsg-4
It seems that libgimp-perl refuses to start even with the most trivial
program:
machine madore ~ $ cat /tmp/test.pl
#!/usr/bin/perl
use strict;
use Gimp;
Gimp::init;
machine madore ~ $ /tmp/test.pl
test.pl: protocol error (1) at /usr/lib/perl5/Gimp/Net.pm
Package: kernel-image-2.6.8-2-686
Version: 2.6.8-13
The configuration of the stock Debian kernels does not include
CONFIG_SOUND_AWE32_SYNTH. It should. Not all SB AWE32 sound cards
work with ALSA (cf. a post which I recently made to the alsa-user
mailing-list, which will be available as soon as
97 matches
Mail list logo
