user d...@packages.debian.org
usertag 523980 + install-info
usertag 518737 + install-info
thanks
On Tue, 14 Apr 2009, jida...@jidanni.org wrote:
Rather than relying on each package to clean up after itself, perhaps
just rebuild the whole thing each time anything changes?
This is already
On Tue, 07 Apr 2009, Jayen Ashar wrote:
I've created a package for a proprietary software that bundles with
java. We've decided since then to replace the bundled java with a
symlink to the system java, but when our systems update the package,
they are left with an empty directory. I have
tag 392570 + unreproducible
thanks
On Mon, 23 Oct 2006, Bruce Sass wrote:
Note: the problem occurs with 2 versions of the available file one
generated directly from the DVD's and on from converting the the DVD's
to a hardidsk based store.
Can you provide those files so that we can reproduce
On Wed, 19 Apr 2006, Justin Pryzby wrote:
Hello Eric,
Could you provide more information about this bug you submitted?
http://bugs.debian.org/242795
[SEGFAULT]dpkg segfaults when running apt-get --fix-missing upgrade
No new information provided, so we can't do anything with this bug
report.
On Mon, 14 Oct 2002, Thomas Hood wrote:
Can you reproduce the bug with a later version of dpkg?
No answer from bug submitter. Furthermore the message DPKG ERROR is
not a message from dpkg (at least not in 1.7.0) so it's not dpkg that
restarted itself but some other frontend (unspecified).
On Mon, 30 Mar 2009, Modestas Vainius wrote:
I understand the need to mark symbols in
multiple ways so that specific treatment/conversions are made but I'm not
convinced that reusing the comment-based approach used for missing symbols
is the best choice.
That's the way it was designed, I
Author: Raphael Hertzog hert...@debian.org
Date: Fri Apr 17 22:24:19 2009 +0200
dpkg-source: accept dashes in component name of additional tarballs
The regular expression used to identify additional tarballs (used in
formats 2.0 and 3.0 (quilt)) was too strict and refused dashes
On Sat, 25 Apr 2009, Josselin Mouette wrote:
Hi,
older versions of the totem-gstreamer and totem-xine added an
alternative in the postinst. Somewhere around 2.22, it was changed to
add two new slave alternatives.
However, it looks like the slave alternatives are completely ignored,
and
On Mon, 27 Apr 2009, Colin Watson wrote:
Here's a patch to add a keyrings vendor hook. This allows Ubuntu
systems to supply an additional keyring against which dpkg-source will
verify source package signatures, and in general for vendors to supply
any number of additional keyrings.
Thanks, I
tag 525834 pending
thanks
Hello,
Bug #525834 reported by you has been fixed in the Git repository. You can
see the changelog below, and you can check the diff of the fix at:
http://git.debian.org/?p=dpkg/dpkg.git;a=commitdiff;h=cbde6f5
---
commit cbde6f587130724749c23908cce129380dc7652d
Author: Raphael Hertzog hert...@debian.org
Date: Mon May 4 20:49:42 2009 +0200
update-alternatives: don't remove real files conflicting with link
When an alternative is installed and a real file is in conflict, the real
file is kept. When the same alternative is fully removed
Author: Raphael Hertzog hert...@debian.org
Date: Thu Apr 9 18:07:09 2009 +0200
dpkg-vendor: new script to query various vendor information
This tool is meant to be used in debian/rules files to have common source
packages across multiple distributions and yet still have slightly
Author: Raphael Hertzog hert...@debian.org
Date: Sun May 10 17:30:02 2009 +0200
dpkg-source: fix automatic patch registration in format 3.0 (quilt)
When the automatic patch is updated due to a second call to
dpkg-source, it's not always properly registered with quilt.
It must
Author: Raphael Hertzog hert...@debian.org
Date: Sun May 10 21:31:40 2009 +0200
dpkg-buildpackage: add execute right on debian/rules when missing
That right should normally not be missing as dpkg-source -x automatically
sets it. However manually applying a Debian diff doesn't
Hi,
On Thu, 14 May 2009, Modestas Vainius wrote:
what's the status of this bug? Can I rely on the patch being applied some day
(it's quite important for the future of my tool)?
I still have your last mail in my debian inbox (i.e. I know that I have to
reply and decide about it/apply it). Most
On Sat, 18 Apr 2009, Modestas Vainius wrote:
wildcard=c++:Foo::Bar::Private::foobar()@Base 1.0
optional:wildcard=c++:Foo::Bar::Private::foobar()@Base 1.0
Hmm, I fail to understand what this = means here.
?? It's just a separator between the tag and its value.
From
On Fri, 15 May 2009, Modestas Vainius wrote:
+sub get_name {
+my $self = shift;
+my $symbol = $self-{symbol};
+$symbol =~ s/['']//g;
+return $symbol;
+}
The quotes should be stripped in parse() and not be stored ?
(And then restored in dump() only if
Author: Raphael Hertzog hert...@debian.org
Date: Fri May 15 16:28:45 2009 +0200
dpkg-buildpackage: allow options -nc and -S together
This combination is generally not recommended as you want to ensure
that a source package contains only source and not any intermediary files
://www.ouaza.com/wp/2009/03/02/contribuer-a-debian-gagner-un-livre/
From bc62fd60bb2347e4550bb3ccba701561cab5e1fe Mon Sep 17 00:00:00 2001
From: Raphael Hertzog hert...@debian.org
Date: Fri, 15 May 2009 18:09:25 +0200
Subject: [PATCH] dpkg-source/dpkg-genchanges: split long Binary: field values
---
scripts
On Thu, 30 Aug 2007, Guillem Jover wrote:
If multiple backports (for different target distributions) can be built
from the same source version, it would be preferrable to only keep one
copy of the sources, for that to work, the binary packages need to point
to an explicit source version
Author: Raphael Hertzog hert...@debian.org
Date: Thu May 28 15:31:32 2009 +0200
Unset TAR_OPTIONS when handling tar archives for source packages
The command line interface is the only way that should control tar's
behaviour in its usage by dpkg as we want the same behaviour
On Thu, 28 May 2009, Stefano Costa wrote:
Opening synaptic, I get the following error message:
E: dpkg was interrupted, you must manually run 'dpkg --configure -a' to
correct the problem.
E: _cache-open() failed, please report.
Report against synaptic I guess... and not dpkg.
dpkg
reassign 529393 popularity-contest 1.47
thanks
Please don't hijack bugs and don't falsely reassign it once
you have identified that your bug is about something else...
I reported the bug and I am a dpkg maintainer. If you have another problem
that differs from the one that I reported (undesired
On Sun, 07 Jun 2009, Neil Williams wrote:
Is there some hidden assumption that the filename must match the Vendor
because the existing default file does not follow that convention, at
least as far as case-sensitivity is concerned. (debian vs Debian).
Yes. Case-sensitivity is taken care of (we
On Sun, 07 Jun 2009, Neil Williams wrote:
The behaviour with DEB_VENDOR should match the behaviour with --vendor
and there should be some way of handling vendor names that include
spaces without needing to use quote marks when using the environment
variables or --vendor options.
That's plain
Author: Raphael Hertzog hert...@debian.org
Date: Sun Jun 7 21:05:41 2009 +0200
update-alternatives: handle dangling alternative symlink
update-alternatives did not properly handle when the
alternative symlink in /etc/alternatives was dangling
and pointing to a choice that does
On Sun, 07 Jun 2009, Neil Williams wrote:
Those options expect a vendor name and not a
filename.
There's no documentation, so how am I to know what those options expect?
--vendor vendor
Assumes the current vendor is vendor instead of discovering it with
the
On Thu, 11 Jun 2009, Joachim Breitner wrote:
Therefore, dpkg-generated should warn about unused substvars. This could
be implemented in Dpkg::Substvars by remembering which vars have been
get()ted after being set()ted, and returning this list upon request.
Please indicate if you agree and
Author: Raphael Hertzog hert...@debian.org
Date: Sat Jun 13 21:50:46 2009 +0200
dpkg-buildpackage(1): add missing paragraph separator
Reported-by: Per Andersson avtob...@gmail.com
diff --git a/debian/changelog b/debian/changelog
index f377ec0..6ca21c8 100644
--- a/debian/changelog
On Mon, 15 Jun 2009, Lionel Elie Mamane wrote:
Possibly that is your plan, but you are waiting for a few hardcoded
calls to /usr/sbin/install-info to disappear, but I didn't see this
in the various bug reports.
There's a plan indeed, see
http://wiki.debian.org/Transitions/DpkgToGnuInstallInfo
On Wed, 17 Jun 2009, David Stansby wrote:
I've downloaded the latest git version, using git clone
git://git.debian.org/git/dpkg/dpkg.git and have changed all the .po
files and the files in /src to the correct grammar. How do I upload
the changes to git now?
You can't upload to git directly.
On Fri, 19 Jun 2009, Christoph Egger wrote:
Building the NEW package I'm working on (irrlicht [3]) on my
ARM(el)[4] box (up-to-date sid pbuilder) causes dpkg-shlibdeps to
complain about missing symbols [0]. These symbols seem to be some gcc
internals that should be covered by libgcc_s.so
Author: Raphael Hertzog hert...@debian.org
Date: Sat Jun 20 20:55:46 2009 +0200
Dpkg::Shlibs::SymbolFile: assume an implicit version Base for unversioned
symbols
diff --git a/debian/changelog b/debian/changelog
index c2cfe9d..56d93a5 100644
--- a/debian/changelog
+++ b/debian/changelog
Author: Raphael Hertzog hert...@debian.org
Date: Sat Jun 20 20:30:59 2009 +0200
dpkg-gensymbols: do not output deprecated symbols
The deprecated symbols appear commented with “#MISSING:”, they should
appear in the explanatory diff but not in the file generated.
diff --git a/debian
Author: Raphael Hertzog hert...@debian.org
Date: Sat Jun 20 19:28:29 2009 +0200
Add changelog entry for new dpkg-gensymbols feature
diff --git a/debian/changelog b/debian/changelog
index 94e451b..8a34e44 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -17,6 +17,12 @@ dpkg (1.15.3
Author: Raphael Hertzog hert...@debian.org
Date: Sat Jun 20 21:45:48 2009 +0200
dpkg-gensymbols: add new “ignore-blacklist” tag
Add new tag ignore-blacklist to force-include symbols which are
normally blacklisted. This can be useful for libgcc to include symbols
Author: Raphael Hertzog hert...@debian.org
Date: Sun Jun 21 00:42:37 2009 +0200
dpkg-genchanges: skip warning about lower version for backports
Backports have by definition a version number lower than the original
version (to enable upgrades from backports to official packages). So
On Fri, 15 May 2009, Raphael Hertzog wrote:
On Mon, 11 Aug 2008, Bastian Blank wrote:
dpkg-genchanges should fold lines in the output to a sane length. There
is a package in the archive (linux-modules-extra-2.6) which produces a
25k long Binary line, which is cropped by gpg during signing
Hi,
On Sun, 31 May 2009, Julien BLACHE wrote:
It may be necessary for one reason or another to determine in debian/rules
whether an architecture is 64bit or not.
There isn't really a good way to do that today, and dpkg-architecture seems
like the natural place to add this.
So, I think a
Hello,
it is well known that C++ symbol mangling result in different symbol
names from one architecture to the other. It means that libraries that
want to provide symbol files have to maintain one symbol file for each
architecture. To avoid this problem Modestas Vainius has written a patch
that
On Sun, 28 Jun 2009, Guillem Jover wrote:
Actually I had some discussion about exporting the wordsize and
endianness few months ago on #debian-dpkg with Hector Oron.
So I've done some preliminary coding, something like this could do,
after some cleanup:
On Tue, 30 Jun 2009, Craig Sanders wrote:
upgrading dpkg fails with /var/lib/dpkg/tmp.ci/preinst: line 38: read:
read error: 0: Is a directory
What do you have in /var/lib/dpkg/alternatives/ ?
Please give the output of ls -alR /var/lib/dpkg/alternatives/.
I suspect that you have an
Author: Raphael Hertzog hert...@debian.org
Date: Fri Jul 3 15:35:50 2009 +0200
Add changelog entry for the install-info transition
diff --git a/debian/changelog b/debian/changelog
index a45f189..5722977 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -11,6 +11,13 @@ dpkg (1.15.4
Author: Raphael Hertzog hert...@debian.org
Date: Fri Jul 3 15:43:29 2009 +0200
Fix dpkg's preinst when alternatives dir contains unexpected subdirs
Adjust the lintian overrides accordingly.
diff --git a/debian/changelog b/debian/changelog
index 5722977..2dd90db 100644
--- a/debian
: Raphael Hertzog hert...@debian.org
Date: Fri Jul 3 15:35:50 2009 +0200
Add changelog entry for the install-info transition
diff --git a/debian/changelog b/debian/changelog
index a45f189..5722977 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -11,6 +11,13 @@ dpkg (1.15.4
reassign 536482 libc6 2.9-19
severity 536482 serious
retitle 536482 symbols file is broken in 2.9-19
clone 536482 -1
reassign -1 libvirt0 0.6.5-1
retitle -1 symbols file is broken in 0.6.5-1
thanks
On Fri, 10 Jul 2009, Sven Joachim wrote:
I suspect this is a consequence of bug #536034¹ which
. :-)
commit a27d14c4423c44b400764c99b669824f4ba6e9a1
Author: Raphael Hertzog hert...@debian.org
Date: Wed Jul 15 15:59:03 2009 +0200
Add new test case for symbols file parsing
Ensure that an #include can change the name of the currently parsed
library (ELF object).
Feel free to try
On Wed, 01 Jul 2009, Emilio Pozuelo Monfort wrote:
I'm implementing automatic creation of debug packages, and to generate
their binary control file I'm using dpkg-gencontrol. The debug package
are not listed in debian/control, and thus when calling dpkg-gencontrol
I get something like:
$
Author: Raphael Hertzog hert...@debian.org
Date: Wed Jul 15 16:44:21 2009 +0200
deb-substvars(5): list fields where substvars do not work
diff --git a/debian/changelog b/debian/changelog
index 23e0f5c..b429933 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -33,6 +33,8 @@ dpkg (1.15.4
On Tue, 07 Jul 2009, Colin Watson wrote:
dpkg-parsechangelog doesn't emit the Launchpad-Bugs-Fixed field when the
vendor is Ubuntu.
Why? What's wrong in the current implementation?
In general I think handling this in dpkg-genchanges is
just the wrong place for it; other scripts mention
retitle 485330 Allow context diff in debian/patches/ in 3.0 (quilt) format
thanks
On Thu, 06 Aug 2009, Pierre Habouzit wrote:
That said, yes, using non-unified diff is as laughable as using RCS or
SCCS nowadays. Though I consider it a bug if dpkg refuses to apply a
patch that patch(1) (that it
On Tue, 11 Aug 2009, Peter Fritzsche wrote:
Ok, tried again and it seems to be that the first time it gets installed it
exists. The next time it is removed and doesnt appear again.
What removes it? I suspect it's not dpkg... (or we would have noticed it
already)
So
$
On Tue, 11 Aug 2009, Peter Fritzsche wrote:
What removes it? I suspect it's not dpkg... (or we would have noticed it
already)
Hm, only started dpkg. So I suspected it.
Did your machine crash/lose power during an upgrade?
work. The only thing I did in between was to reboot. Maybe it is/was
On Thu, 13 Aug 2009, Guillem Jover wrote:
On Wed, 2009-08-12 at 23:27:05 +0200, Raphael Hertzog wrote:
On Tue, 11 Aug 2009, Peter Fritzsche wrote:
What removes it? I suspect it's not dpkg... (or we would have noticed it
already)
Hm, only started dpkg. So I suspected it.
Did your
On Fri, 14 Aug 2009, Kurt Roeckx wrote:
The documentation says that -c2 should fail if new symbols got
added, but it doesn't for me.
Wildcard symbols do not allow this currently since you don't have any
comprehensive list of symbols in the reference symbol file. Any symbol
in the final file is
On Fri, 14 Aug 2009, Leo 'costela' Antunes wrote:
Currently the code responsible for creating the files in
/var/backups/dpkg.status.* is in /etc/cron.daily/standard, which belongs
to the cron package.
I'd suggest this functionality be brought into the dpkg package and
slightly better
tag 533916 + wontfix
summary 533916 0
thanks
with the feedback received and from what I understand, it's not a
feature that I would like to support on the long term since there is no
finite set of substitutions and those can change depending on external
include files.
Externalizing the
Hi,
On Sun, 16 Aug 2009, Joachim Breitner wrote:
dpkg-buildpackage(1) claims that -C, -m and -e options are passed to
dpkg-source and documented in it’s manpage, but they are not mentioned
in either dpkg-source(1) nor dpkg-source --help.
Quoting dpkg-buildpackage(1) I see:
On Tue, 18 Aug 2009, Jonathan Nieder wrote:
I have been working on packaging XZ Utils, which is supposed to
eventually supersede LZMA Utils. To that end, it provides “lzma”,
“unlzma”, etc commands for compatibility with lzma, but scripts and
other programs are encouraged to use “xz” directly
Hi,
On Thu, 20 Aug 2009, Zoran Dzelajlija wrote:
A related bug, choking on leftover zero-length files in
/var/lib/dpkg/alternatives, might or might not warrant a separate report.
The report already exists, it's 537558 and it's fixed in 1.15.4 (in git
only for now).
Cheers,
--
Raphaël Hertzog
On Thu, 20 Aug 2009, Modestas Vainius wrote:
Unfortunately, I have not got around to write tests / update man page for
these changes yet hence I never submitted the patch for review :/ I may post
it in the current state if you want to see it.
Yeah, feel free to push it in a private git repo
Package: dpkg-dev
Severity: wishlist
User: d...@packages.debian.org
Usertags: dpkg-source
Automatic patches (debian/patches/debian-changes-version) generated when
using 3.0 (quilt) source packages should contain an autogenerated DEP-3
compliant header. This would facilitate usage of this format
forcemerge 537800 543603
thanks
On Tue, 25 Aug 2009, Raphael Geissert wrote:
It is currently impossible to include the first changelog entry by the means
of dpkg-genchange's -v option, since specifying the first version will only
include the change sets greater than that version and
clone 539692 -1
reassign -1 gnupg 1.4.9-4
retitle -1 gnupg should respect default-key when --local-user matches several
private keys
retitle 539692 dpkg-buildpackage should have a config file
severity 539692 wishlist
user d...@packages.debian.org
usertags 539692 dpkg-buildpackage
thanks
On Mon,
reassing 544481 apt 0.7.23.1
severity 544481 normal
retitle 544481 apt mixes essential flag from all sources
On Tue, 01 Sep 2009, Vincent Lefevre wrote:
I could help you to determine what's wrong in your system, but *not*
on the basis that I have to do something that I've *already* done,
forcemerge 537558 544732
thanks
On Wed, 02 Sep 2009, Colin Tuckley wrote:
Package: dpkg
Version: 1.15.3.1
Severity: normal
dpkg preinst fails if there is a zero length file in
/var/lib/dpkg/alternatives
The bug is already reported, did you check before filing it? Anyway, it's
fixed in
On Thu, 03 Sep 2009, Vincent Lefevre wrote:
Hi,
What's the status of this bug?
On http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=139569, it is said:
Blocking fix for 411722: coreutils: info printf just gets the man
page, 483554: coreutils: info chcon gives the man page instead of
On Thu, 03 Sep 2009, Vincent Lefevre wrote:
On 2009-09-03 11:34:30 +0200, Raphael Hertzog wrote:
What about reading the last message in the history? dpkg's
install-info is going away in 1.15.4 and all install-info bugs will
be closed/reassigned to the new install-info package
forcemerge 229357 545081
thanks
On Fri, 04 Sep 2009, Zack Weinberg wrote:
As is well known, dpkg-buildpackage -B does not know whether it can
If you knew well, you would have known that this bug already exists...
it's #229357 and you're not helping us by opening a new bug report.
dpkg-dev has
On Fri, 04 Sep 2009, Zack Weinberg wrote:
I disagree that this is 229357. That bug is about Build-Options.
This is a new proposal to *unconditionally* use build-arch in
dpkg-buildpackage -B. I'm not going to play control ping-pong, but
please reconsider.
It's all the same, we can't fix the
Hello,
On Sun, 06 Sep 2009, Ian Bruce wrote:
$ dpkg-source --require-valid-signature -x psutils_1.17-27.dsc
gpgv: keyblock resource `/home/ian/.gnupg/trustedkeys.gpg': general error
gpgv: Signature made Wed 19 Aug 2009 04:21:54 PM PDT using DSA key ID
D688E0A7
gpgv: Can't
On Sun, 06 Sep 2009, ian_br...@fastmail.net wrote:
On Sun, 6 Sep 2009 10:38:37 +0200
Raphael Hertzog hert...@debian.org wrote:
What's up with /home/ian/.gnupg/trustedkeys.gpg?
Is it a good keyring file? Does it have correct permissions?
Try again after moving it out of the way
On Sun, 06 Sep 2009, Petr Vandrovec wrote:
Since today's update run dselect is completely confused and does not
remember which packages are new and which are old - it lists all
available and not installed packages as new, again and again. And
not only that, it also marks some of them for
On Mon, 07 Sep 2009, Carsten Grohmann wrote:
New version number isn't formatted properly. Especially the colon (:,
hex 3a) won't be displayed correctly. It's properly an i18n/l10n problem
because I use German language setting.
What led you to believe that it's
severity 545366 serious
thanks
On Tue, 08 Sep 2009, Petr Vandrovec wrote:
IMHO change should be reverted until old functionality can be
achieved through some other way - unless you can achieve such
functionality today.
I'd like to remind everybody that we're looking for someone to step up and
Package: dpkg
Version: 1.15.4
Severity: wishlist
Please export DPKG_MAINTSCRIPT_NAME in the environment of maintainer
scripts, it should contain the name of the maintainer script (prerm,
postrm, preinst, postinst, config) that is currently running.
See #546165 for an example of where it would
On Tue, 15 Sep 2009, Aleksey Midenkov wrote:
Guys! Placing some hints on dpkg about packages that may be broken is
wrong. Imagine that count of such packages is infinity, so you will
be keep putting all them inside dpkg control info? This is an
abstraction and the next is reality. You fixed
Hi,
On Tue, 22 Sep 2009, Javier Serrano Polo wrote:
Could someone explain (again?) why dpkg-buildpackage -B should not add
build-arch to DEB_BUILD_OPTIONS? It doesn't require any control fields,
any make fiddling, it's compatible with old rules files... it just
works.
What are you trying to
forcemerge 32877 548663
thanks
On Sun, 27 Sep 2009, Stefan Monnier wrote:
I think dpkg should systematically save the default content of each and
every file that might reasonably be expected to be modified by the user.
This request is not new... please try to check open bug reports on the
reassign 548661 general
thanks
On Mon, 28 Sep 2009, Stefan Monnier wrote:
reopen 548661
thanks
Please don't play ping-pong with bugs. Start a discussion on -devel if you
want to get input on how to properly solve your problem, but it's far from
being a reasonable wishlist request for dpkg at
reassign 549011 safe-rm
retitle 549011 safe-rm provides a conflicting and less reliable rm command
severity 549011 normal
thanks
On Wed, 30 Sep 2009, Didier Raboud wrote:
I had safe-rm installed and perl broke (#548848). dpkg went unuseable because
it was using /usr/bin/rm from safe-rm, broken
Author: Raphael Hertzog hert...@debian.org
Date: Tue Jul 21 10:14:00 2009 +0200
Dpkg::Changelog: fix heuristic to identify correct start entry
The heuristic to identify the version number that is most approaching
was not working properly for the case where the version number
Author: Raphael Hertzog hert...@debian.org
Date: Mon Jul 20 08:47:09 2009 +0200
Fix preinst to not fail on read on broken alternatives status files
The internal structure of alternatives status file is line based.
Empty lines end each section. read call can fail if the file
: Raphael Hertzog hert...@debian.org
Date: Fri Jul 3 15:35:50 2009 +0200
Add changelog entry for the install-info transition
diff --git a/debian/changelog b/debian/changelog
index a45f189..5722977 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -11,6 +11,13 @@ dpkg (1.15.4
Author: Raphael Hertzog hert...@debian.org
Date: Fri Jul 3 15:35:50 2009 +0200
Add changelog entry for the install-info transition
diff --git a/debian/changelog b/debian/changelog
index a45f189..5722977 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -11,6 +11,13 @@ dpkg (1.15.4
Author: Raphael Hertzog hert...@debian.org
Date: Fri Jul 3 15:43:29 2009 +0200
Fix dpkg's preinst when alternatives dir contains unexpected subdirs
Adjust the lintian overrides accordingly.
diff --git a/debian/changelog b/debian/changelog
index 5722977..2dd90db 100644
--- a/debian
On Wed, 21 Oct 2009, Sandro Weiser wrote:
Today I installed dpkg, dpkg-dev and dselect 1.15.4.1.
I hoped that the error is fixed, but he is still there.
I use _only_ dselect, because I love it!
Can you please fix it?
What is precisely the error that you're seeing?
Cheers,
--
Raphaël Hertzog
On Sat, 24 Oct 2009, Takehiko Abe wrote:
The fix does not work for me. dselect behaves the same way as it did
with 1.15.4.1.
What dselect method are you all using?
Cheers,
--
Raphaël Hertzog
--
To UNSUBSCRIBE, email to debian-dpkg-bugs-requ...@lists.debian.org
with a subject of
Version: 1.15.4.1
On Sun, 25 Oct 2009, Piotr Engelking wrote:
The only difference between 1.15.4 and 1.15.4.1 is that former marks
new and old packages as new, and the latter marks new and old packages
as old.
That's what we wanted. Closing this bug again. The fact that it doesn't
keep track
On Tue, 27 Oct 2009, Takehiko Abe wrote:
That's what we wanted. Closing this bug again. The fact that it
doesn't keep track of new packages is tracked in #551638 and is not
release critical for us.
What about the it also marks some of them for installation although I
do not want them part
On Wed, 28 Oct 2009, Christian Perrier wrote:
Frankly speaking, for other packages where there is no real
maintainer, where the userbase is small and that have many bugs, we
are not that sentimental and the packages are removed from the
archive. Period, basta and voilà.
Who is really
Package: dpkg-dev
Severity: wishlist
It's been my plan since some time to modify dpkg-source to produce new
source formats by default, now that the archive accepts the new formats,
it's about time to look closer on how to make this happen.
For Debian as a whole, we need to fix the bugs listed
On Sun, 01 Nov 2009, Markus Fischer wrote:
When running inside a vserver, the capability to alter the IO priority may not
be available.
This error has already been demoted to a simple warning in the upcoming
dpkg 1.15.5.
Cheers,
--
Raphaël Hertzog
--
To UNSUBSCRIBE, email to
On Thu, 05 Nov 2009, Mike Hommey wrote:
A bit more information. It *hum*hum* works, actually, and the error
message above was due to remaining stuff in
debian/source/include-binaries.
No, it really doesn't work as expected. It works in lenny but doesn't work
properly in sid. I think it might
On Thu, 05 Nov 2009, Mike Hommey wrote:
No, it really doesn't work as expected. It works in lenny but doesn't work
properly in sid. I think it might be due to a behavioural difference
in perl's File::Find...
In fact no, it's a regression that I introduced in
On Fri, 06 Nov 2009, Mike Hommey wrote:
On Thu, Nov 05, 2009 at 09:56:03AM +0100, Raphael Hertzog wrote:
What I can do however is add a supplementary option like
--create-empty-orig that would help bootstrap the process for your
specific case (that would not encourage repacking an upstream
On Fri, 06 Nov 2009, Jaldhar H. Vyas wrote:
On Thu, 5 Nov 2009, Raphael Hertzog wrote:
On Wed, 04 Nov 2009, Jaldhar H. Vyas wrote:
What I would like to do instead is to use the multiple .orig.tar.gz
feature of the 3.0 (quilt) source format. Each module would become a
l-a-b-p-b-p_$version
On Fri, 06 Nov 2009, Mike Hommey wrote:
That pretty much depends on what dak is expecting to be the right files.
If it can handle the additional tarballs, doesn't it mean it properly
parses the .dsc and/or .changes files ?
It does and verify that we have all the required files for the
On Sat, 07 Nov 2009, Mike Hommey wrote:
We can go even further, with another optimization.
As appeared in the profiles, abs2rel and canonpath take a lot of time.
It look like abs2rel is bloated for what it is used for.
I applied the patch as well. Thanks!
Cheers,
--
Raphaël Hertzog
--
On Tue, 17 Nov 2009, Michael Biebl wrote:
since the update of dpkg-dev from 1.15.4 to 1.15.5, my packages fail to
build with the following message:
Can't call method epoch on an undefined value at
/usr/share/perl5/Dpkg/Version.pm line 132.
dh_shlibdeps: dpkg-shlibdeps returned exit code 2
301 - 400 of 801 matches
Mail list logo
