Bug#968261: sane-airscan: Copyright fails to mention Expat license of http_parser.{c,h} or copyright holders thereof

[Christopher James Halse Rogers]

Package: sane-airscan
Version: 0.99.12-1
Severity: important

Dear Maintainer,

While reviewing an upload of sane-airscan to the Ubuntu NEW queue I
discovered the http_parser.{c,h} files are (C) Joylent, Inc under the
Expat licence, and neither Joylent nor the Expat licence are mentioned
in debian/copyright.

I've verified that this also applies to the current source package in 
Salsa.


-- System Information:
Debian Release: bullseye/sid
 APT prefers focal-updates
 APT policy: (500, 'focal-updates'), (500, 'focal-security'), (500, 
'focal'), (100, 'focal-backports')

Architecture: amd64 (x86_64)
Foreign Architectures: i386, arm-linux-gnueabihf

Kernel: Linux 5.7.0+bcachefs.git20200728.6288f1b6-1-generic (SMP w/8 
CPU cores)
Kernel taint flags: TAINT_PROPRIETARY_MODULE, TAINT_OOT_MODULE, 
TAINT_UNSIGNED_MODULE
Locale: LANG=en_AU.UTF-8, LC_CTYPE=en_AU.UTF-8 (charmap=UTF-8), 
LANGUAGE=en_AU:en (charmap=UTF-8)

Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

Bug#836727: argyll: Please look for libcolordcompat plugin in private library path

[Christopher James Halse Rogers]

Hello Jörg,

On Wed, Sep 7, 2016 at 4:05 PM, Jörg Frings-Fürst 
 wrote:

Hello Chritopher,

thank you for spending your time helping to make Debian better with
this bug report.

Using an absolute path isn't system conform, breaks the library search
and does a lot of work on changes.


I'm sorry, I don't understand your reasoning.

What's not system conformant about searching in /usr/lib/argyll? This 
is a standard pattern used by many other programs with optional 
dynamically-loaded plugins.


It's true that this probably shouldn't go upstream as-is - it should 
probably be reworked so that the plugin path is a configure-time option 
- but that's not particularly relevant to Debian, where we'd always 
define it to be /usr/lib/argyll.


My interpretation of policy is that I currently can't ship 
libcolordcompat.so from colord in a way that can be used by argyll. 
This is a plugin object for argyll, not something that should be in the 
public linker path.

Bug#836727: argyll: Please look for libcolordcompat plugin in private library path

[Christopher James Halse Rogers]

Package: argyll
Version: 1.8.3+repack-2
Severity: wishlist
Tags: upstream patch

dispcal includes a simple plugin-like mechanism to integrate
with the system colord ICC store.

It currently simply calls dlopen("libcolordcompat.so", ...),
requiring the plugin to be in the system library resolution
paths.

The attached patch makes dispcal look for libcolordcompat.so
in /usr/lib/argyll/.



-- System Information:
Debian Release: stretch/sid
  APT prefers yakkety
  APT policy: (500, 'yakkety')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.7+bcachefs-1.git20160828--generic (SMP w/8 CPU cores)
Locale: LANG=en_AU.UTF-8, LC_CTYPE=en_AU.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)

Versions of packages argyll depends on:
ii  argyll-ref1.8.3+repack-2build1
ii  dpkg  1.18.10ubuntu1
ii  libc6 2.24-0ubuntu1
ii  libjpeg8  8c-2ubuntu8
ii  libpng16-16   1.6.24-2
ii  libssl1.0.0   1.0.2g-1ubuntu8
ii  libtiff5  4.0.6-2
ii  libx11-6  2:1.6.3-1ubuntu3
ii  libxext6  2:1.3.3-1
ii  libxinerama1  2:1.1.3-1
ii  libxrandr22:1.5.0-1
ii  libxss1   1:1.2.2-1
ii  libxxf86vm1   1:1.1.4-1

Versions of packages argyll recommends:
ii  libpam-systemd  231-5
ii  udev231-5

Versions of packages argyll suggests:
pn  argyll-doc
ii  colord1.3.2-1
pn  gir1.2-colordgtk-1.0  

-- no debconf information
Index: argyll-1.8.3+repack/spectro/dispwin.c
===
--- argyll-1.8.3+repack.orig/spectro/dispwin.c	2016-09-04 12:17:35.0 +1000
+++ argyll-1.8.3+repack/spectro/dispwin.c	2016-09-04 12:49:15.892851404 +1000
@@ -2341,7 +2341,10 @@
 
 	cd_found = NULL;
 
-	if ((cd_found = dlopen("libcolordcompat.so", RTLD_LAZY)) != NULL) {
+	/* Load from private library path. Load from /usr/lib/argyll under the assumption that
+	 * dispcal isn't going to be foreign-arch.
+	 */
+	if ((cd_found = dlopen("/usr/lib/argyll/libcolordcompat.so", RTLD_LAZY)) != NULL) {
 
 		cd_edid_install_profile = dlsym(cd_found, "cd_edid_install_profile");
 		cd_edid_remove_profile = dlsym(cd_found, "cd_edid_remove_profile");

Bug#815252: Not fixed

[Christopher James Halse Rogers]

reopen 815252
thanks

Bah, closed the wrong bug from the changelog. This is still blocked on 
lcms2 adding support.

Bug#818050: /usr/bin/colormgr: cannot profile monitor with gretag i1-display2

[Christopher James Halse Rogers]

Hm. I see that you don't have colord-sensor-argyll installed. I think 
that might be required to drive your calibration device.


Can you please try installing it and see if calibration works?

It's a Suggests: rather than a Recommends: because it depends on 
argyll, which is almost 100MB on-disc and is not necessary for quite a 
lot of users, but we could certainly fix the error reported.

Bug#815252: colord: please make the build reproducible (timestamps)

[Christopher James Halse Rogers]

Thanks for the patches! I'll add the necessary autofoo checks and 
propose them upstream.


Would it make more sense for cd-it8 to also respect SOURCE_DATE_EPOCH? 
Is there a reason other than ‘it was simpler to just disable 
timestamps’ (before I try duplicating the cd-create-profile logic 
into that patch).

Bug#801401: Workarounds for rootless Xorg

[Christopher James Halse Rogers]

On Fri, Oct 23, 2015 at 5:01 AM, Lingzhu Xiang 
 wrote:

* Privilege to drmSetMaster()

If there is only one drm device no setup is needed.


This is an incorrect understanding of what drmSetMaster() does. It is 
not setting a primary device, it's the process claiming the DRM_MASTER 
capability, which is required for things like modesetting and 
authorising other drm clients' access to the device.


Claiming DRM_MASTER requires root.

Bug#797572: Urgh

[Christopher James Halse Rogers]

Urgh, I thought I'd fixed this *ages* ago.

Indeed, the upstream dependency on gnome-desktop is gone, I've just 
failed to remove the build-dependency.


Sorry.

Bug#751212: colord: obsolete-conffile as found by adequate

[Christopher James Halse Rogers]

Grr. I was *sure* I tested it with a piuparts upgrade. Time to check 
again!

Bug#788297: colord: Need libcolorhug-dev 1.2.9 or later to build fwupd

[Christopher James Halse Rogers]

On Tue, 9 Jun 2015 19:12:35 -0500 Daniel Jared Dominguez 
jared_doming...@dell.com wrote:

 Package: libcolorhug-dev
 Version: 1.2.9
 Severity: wishlist


 Hi there,

 I'm building fwupd for Debian. It requires libcolorhug-dev 1.2.9 or
 later. However, the latest version in unstable is 1.2.1. I notice you
 have 1.2.10 in experimental. I was wondering when you might have 
this in
 unstable. I'm trying to get fwupd into both Debian and Ubuntu, so 
it'd

 really help me if colord were too.


Hi!

It looks like colord is about to be tied up in a systemd transition, so 
now's not an appropriate time to upload to unstable.


There's no good reason for colord to be in experimental anymore though, 
so as soon as the transition is over I'll upload to unstable.



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#773907: libcolord2: libcolordprivate.so.2 - undefined symbol: cmsGetContextUserData

[Christopher James Halse Rogers]

Hm. libcolordprivate.so links to liblcms2.so.2, and that symbol is 
available in liblcms2 2.6, which you appear to have installed.


Do you happen to have a locally-installed copy of lcms2 hanging around 
somewhere? Could you attach the output of “ldd 
/usr/lib/x86_64-linux-gnu/libcolordprivate.so.2”?



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#765120: monogame: support monodevelop 5

[Christopher James Halse Rogers]

Package: monogame
Followup-For: Bug #765120

Alright. So, monogame 2.5.1 (from 2012) doesn't build at all against 
Monodevelop 5. A new upstream version, Monogame 3.2, doesn't build against the 
version of tao-framework in Jessie (also from 2012, project deprecated in 
favour of libopentk, for which Jessie also has a 2012 snapshot). 

It's probably time to either start maintaining these or drop them from the 
archive.


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#768911: [pkg-cli-apps-team] Bug#768911: Bug#768911: Bug#768911: gnome-do: desktop freezes when gnome-do is launched

[Christopher James Halse Rogers]

Ok. I can now reproduce this too, but *only* if Do is set to show on 
startup (which I didn't have).


I wonder if this is related to the GMutex unpleasantness recently. I'll 
investigate.



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#768911: [pkg-cli-apps-team] Bug#768911: gnome-do: desktop freezes when gnome-do is launched

[Christopher James Halse Rogers]

Oooh, wow. That's unexpected!

Can you be more specific than ‘freezes’?
ie:
*) Does the mouse pointer still move?
*) Can you VT switch away from X and back?
*) If you VT switch to a terminal, kill gnome-do, and then VT switch 
back to X is it unfrozen?


If you can VT switch away, kill Do, and VT switch back, could you 
attach the output of running “gnome-do --debug” in a terminal?



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#768911: [pkg-cli-apps-team] Bug#768911: Bug#768911: gnome-do: desktop freezes when gnome-do is launched

[Christopher James Halse Rogers]

Hm. That looks like a pretty normal startup trace.

My guess would be that Do has grabbed input but not managed to show 
itself (does hitting esc unfreeze the desktop?)


I'll give it at try in a GNOME Shell environment to see if I can 
reproduce.



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#763006: Fix known

[Christopher James Halse Rogers]

Ah, yes.

The old “glib changes mutex semantics” gambit.

I'll prepare an upload and run the pkg-cli-apps sponsoring gauntlet :)

Bug#754468: libcolord2: symbol lookup error in libcolordprivate.so.2

[Christopher James Halse Rogers]

Hm. I am confused as to how this could happen.

Checking the liblcms2.so.2 from liblcms2-2 2.6-3 on amd64 I find that 
it exports cmsGetContextUserData, and ldd shows libcolordprivate.so.2 
depending on liblcms2.so.2.


Have you locally installed lcms2 at some point?

Just to check that the write SOs are being loaded, could you attach the 
output of “strace -f -e file /usr/lib/colord/colord”, please?



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#747923: Info received (Bug#747923: transition: colord)

[Christopher James Halse Rogers]

colord 1.2.1-1 uploaded; has built on s390x and (apparently by 
fortunate accident) mipsel.


mips fails with SIGBUS, and will need to be retried once the lcms2 fix 
is uploaded.


kfreebsd seems to be hit by a dependency issue, but that doesn't seem 
to be colord's fault?



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#750533: colord output

[Christopher James Halse Rogers]

Hey there!

What is the actual output colord is spewing to syslog? For me it just 
lists the devices that appear when cups starts (ie: all the printers 
I've got set up), which seems to be reasonable behaviour to me.



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#747923: transition: colord

[Christopher James Halse Rogers]

On Thu, Jun 5, 2014 at 1:38 AM, Emilio Pozuelo Monfort 
po...@debian.org wrote:

On 30/05/14 14:01, Emilio Pozuelo Monfort wrote:

 On 29/05/14 17:09, Emilio Pozuelo Monfort wrote:

 On 29/05/14 17:03, Emilio Pozuelo Monfort wrote:

 Control: tags -1 + pending

 On 25/05/14 12:53, Emilio Pozuelo Monfort wrote:

 Control: tags 747923 confirmed

 On 25/05/14 05:10, Christopher James Halse Rogers wrote:
 On Sun, May 25, 2014 at 12:03 AM, Emilio Pozuelo Monfort 
po...@debian.org wrote:

 On 13/05/14 02:08, Christopher James Halse Rogers wrote:

  Package: release.debian.org
  Severity: normal
  User: release.debian@packages.debian.org
  Usertags: transition
  
  The new release from the new stable branch of colord, 1.2.0, 
contains an

  SONAME bump.


 Can you upload it to experimental now, so that when we ack the 
transition, the

 library doesn't have to go through NEW?


 It has passed NEW and is now in experimental.


 Please go ahead and upload to unstable. Let me know once colord 
has been built

 everywhere in order to schedule the binnmus.


 colord is failing to build on mipsel:

 https://buildd.debian.org/status/logs.php?pkg=colordarch=mipsel


 And mips and s390x failed on 1.2.0-1, and there don't seem to be 
any relevant

 changes in -3:

 
https://buildd.debian.org/status/fetch.php?pkg=colordarch=mipsver=1.2.0-1stamp=1400576819


 
https://buildd.debian.org/status/fetch.php?pkg=colordarch=s390xver=1.2.0-1stamp=1400390062
 
 FTR:
 
 09:28aurel32 for the record, the colord FTBFS on mips/mipsel 
is due to a bug in lcms2
 09:29aurel32 MIPS (and also SPARC if we still care about) 
require that a 64-bit value (double here) is aligned to 64 bits, 
even on a 32-bit machine
 09:30aurel32 and the internal memory allocator only aligned 
entries to void *
 09:30aurel32 i am testing multiple solutions (i have at least 
one working), i will send a patch later today


Aurelien provided a patch for the lcms2 bug, and another one for the 
colord

s390x bug. Can you upload that?


Absolutely. My sponsor is currently running a test-build on the mips 
porter box.



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#750013: colord: FTBS on s390x: wrong cast from uint64_t * to uint32_t *

[Christopher James Halse Rogers]

Thanks for the patch! I'll commit this upstream and do a new upload 
soon.



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#748627: libsnappy-dev: Static archive unsuitable for linking into dynamic object

[Christopher James Halse Rogers]

Package: libsnappy-dev
Version: 1.1.2-1
Severity: wishlist

Dear Maintainer,

I would like to statically link libsnappy into the tracing wrappers for
Apitrace to minimise the chance of symbol conflicts and related shared
library misbehaviour as the tracing wrappers are LD_PRELOADed into arbitrary
processes that we don't control.

The static archive libsnappy.a shipped in libsnappy-dev is not suitable for
this because it's not built with -fPIC (so can't be linked into a dynamic
object on AMD64 at least).

For the moment I'm just using apitrace's bundled copy of snappy, so there's
very little urgency here.

(Reported from an Ubuntu system, but tested with 1.1.2-1 from Sid)


-- System Information:
Debian Release: jessie/sid
  APT prefers utopic-updates
  APT policy: (500, 'utopic-updates'), (500, 'utopic-security'), (500, 
'utopic'), (100, 'utopic-backports')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.13.0-24-generic (SMP w/8 CPU cores)
Locale: LANG=en_AU.UTF-8, LC_CTYPE=en_AU.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages libsnappy-dev depends on:
ii  libsnappy1  1.1.0-1ubuntu1

libsnappy-dev recommends no packages.

libsnappy-dev suggests no packages.

-- no debconf information


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#636679: Sponsoring apitrace package for Debian

[Christopher James Halse Rogers]

On Tue, May 13, 2014 at 10:04 PM, Guus Sliepen g...@debian.org wrote:

On Tue, May 13, 2014 at 01:44:32PM +0200, Guus Sliepen wrote:

 Please tell me where I can download your apitrace packages, and 
unless there is

 something wrong with them I'll upload them to Debian today.


Ok, found something in 
git://git.debian.org/pkg-xorg/app/apitrace.git. Some issues:


- It's an old version, latest is 5.0. Do you intend to update this 
package?


Yes; I'm about halfway through doing so. As you might have noticed, the 
annoying bit is extracting a bunch of embedded code copies - zlib, 
snappy, png.


I hit my buildsystem-madness threshold and then got busy with other 
things. I'll see if I can finish this off tonight.




- You're using pristine-tar, but the upstream version in 
debian/changelog does

  not match any of the tarballs in the repository.

- Why use pristine-tar when upstream does not provide any tarballs at 
all?


So people can easily build packages requiring Debian-only updates given 
only the git packaging branch, rather than the packaging branch plus 
downloading the tarball from the Debian archives.



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#747923: transition: colord

[Christopher James Halse Rogers]

Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: transition

The new release from the new stable branch of colord, 1.2.0, contains an
SONAME bump.

Although it contains some API changes from the existing 1.0 packages, they're
all dropping long-deprecated symbols; none of the packages in archive use
them.

The list of rdepends is:
  weston
  simple-scan
  gtk+3.0
  gnome-settings-daemon
  gnome-control-center
  gnome-color-manager
  darktable
  colorhug-client
  colord-gtk

All of these test-build successfully against libcolord2 and libcolorhug2 on
amd64. The transition should be completable with just binNMUs.

Ben file:

title = colord;
is_affected = .depends ~ libcolord1 | .depends ~ libcolorhug1 | .depends ~ 
libcolord2 | .depends ~ libcolorhug2;
is_good = .depends ~ libcolord2 | .depends ~ libcolorhug2;
is_bad = .depends ~ libcolord1 | .depends ~ libcolorhug1;

-- System Information:
Debian Release: jessie/sid
  APT prefers utopic-updates
  APT policy: (500, 'utopic-updates'), (500, 'utopic-security'), (500, 
'utopic'), (100, 'utopic-backports')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.13.0-24-generic (SMP w/8 CPU cores)
Locale: LANG=en_AU.UTF-8, LC_CTYPE=en_AU.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#636679: ITP apitrace any progress?

[Christopher James Halse Rogers]

Hi!

It's been in an uploadable condition a couple of times over the last 
couple of years, but the DDs on the Debian X team have always been too 
busy to sponsor at those times (and I've never quite got around to 
apply for DD).


I'll give it another update in git; a couple of the other Debian X guys 
are close to DD status, so the bandwidth for uploading should be better 
soon :)



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#705906: Not fixable in fsharp

[Christopher James Halse Rogers]

reassign 705906 mono-gac
forcemerge 570181 705906

thanks

This is no longer an fsharp bug. You've encountered bug #570181, which
you can work around by installing mono-runtime (or mono-runtime-common,
in experimental) first.

Clearly it's time to fix that longstanding problem :)


signature.asc
Description: This is a digitally signed message part

Bug#705906: Pending

[Christopher James Halse Rogers]

I've got the infrastructure required to make this work correctly in
branches of cli-common-dev and mono.

They should be uploadable in the next week or so.


signature.asc
Description: This is a digitally signed message part

Bug#711543: Fixed in 0.1.31-1

[Christopher James Halse Rogers]

fixed 711543 0.1.31-1
thanks

This was fixed in 0.1.31-1. It might be worth doing a stable update for 
those hitting this in wheezy, though.



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/debian-bugs-dist

Bug#668225: colord-sane thread-safety in wheezy

[Christopher James Halse Rogers]

On Fri, 17 May, 2013 at 8:45 AM, Christian Kastner deb...@kvr.at 
wrote:

On 2013-05-16 09:47, Simon McVittie wrote:

 Please use a Subject line that summarizes the subject of the email.


ACK


 On 16/05/13 07:00, Christian Kastner wrote:

 Would it be possible to include [a fix for libdbus-related
 thread-safety problems in colord] in the next Wheezy point
 release (7.0.1)?

 
 I suspect the necessary changes to be rather too large for a stable
 update, given that the changelog describes it as a rewrite. 
However, a
 necessary first step would be for people who reliably get this 
crash to

 confirm that 0.1.31 actually fixes it.


I will try to confirm this over the weekend, but I too doubt that a
rewrite will be accepted into a point release.

 It looks as though a less intrusive fix for wheezy might be to drop 
the
 full SANE plugin and instead backport the udev-based 
cd-plugin-scanner
 module added by commit ebf3e961, which can detect local scanners 
but not

 networked ones:
 
 +# If we should use SANE to add scanner and camera devices.

 +#
 +# If SANE support is installed then this will allow colord to 
manage

 +# all scanners that SANE can detect, including remote scanners.
 +#
 +# If this is disabled then colord will only detect locally 
connected

 +# scanners.

 Another possibility for a lightweight workaround would be to set
 UseSANE to false in the default configuration file, which would 
result

 in colour correction for screens and printers but not scanners.

 I haven't had any response to the upstream libsane bug I opened 
querying

 some code used by colord-sane that uses threads but does not look
 thread-safe
 
(https://alioth.debian.org/tracker/index.php?func=detailaid=313921group_id=30186atid=410366).
 
 Adding a call to dbus_threads_init_default() early in colord-sane's

 main() can't hurt, either.





This was done in colord 0.1.21-2 (and fixed for !Linux in -3); it fixes 
the dbus crash, but this leaves the 
fd-leak-resulting-in-100%-CPU-use-in-select() bug.


Probably the best thing to do is to pull in 0.1.21-4. That has a number 
of miscellaneous fixes, and a rework of colord-sane that avoids all the 
problems. I should probably have pushed this past the freeze.


Although that's a significant patch, it might be reasonable for a point 
release as the existing colord-sane is broken everywhere, so it's 
difficult to make things worse.



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/debian-bugs-dist

Bug#705924: doesn't work

[Christopher James Halse Rogers]

Hm. I wonder why I didn't see this in testing.

I think I see the problem; it looks like I need to fix the packaging
helpers to install some extra F# specific metadata.


signature.asc
Description: This is a digitally signed message part

Bug#705924: Problem isolated, workaround available

[Christopher James Halse Rogers]

Yup, that's it. It's failing to find FSharp.Core.sigdata and
FSharp.Core.optdata; if you copy them
to /usr/lib/mono/gac/FSharp.Core/FSharp.Core/4.3.0.0__b03f5f7f11d50a3a/
you should find that fsharpi works. I clearly had those lying around
from a previous dirty install.

I think we'll need to teach some mono tooling about F#.


signature.asc
Description: This is a digitally signed message part

Bug#702620: ITP: colord-gtk -- GTK+ convenience library for interacting with the colord system daemon

[Christopher James Halse Rogers]

Package: wnpp
Severity: wishlist
Owner: Christopher James Halse Rogers r...@ubuntu.com

* Package name: colord-gtk
  Version : 0.1.24
  Upstream Author : Richard Hughes rich...@hughsie.com
* URL : http://colord.hugsie.com/
* License : LGPL3
  Programming Lang: C
  Description : GTK+ convenience library for interacting with the colord
system daemon

The colord-gtk library was split out of the main colord daemon to avoid a
circular
dependency, so we now need to package it to upgrade colord.


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#699941: Patches

[Christopher James Halse Rogers]

Here are the two diffs, pulled out of my git tree.

The first is a simple update to libgusb 0.1.5.
The second additionally multiarchs the library, and enables the
hardening flags.
diff --git a/debian/changelog b/debian/changelog
index 0bdbcfa..eb92169 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,15 @@
+libgusb (0.1.5-1) UNRELEASED; urgency=low
+
+  * New upstream release
++ Adds gobject-introspection support
+  * Multiarch the library
++ Bump debhelper dependency
++ Bump compat to 9
+  * Enable hardening flags
++ Add lintian-override for no-fortify-functions false-positive
+
+ -- Christopher James Halse Rogers r...@ubuntu.com  Wed, 06 Feb 2013 13:57:28 +0800
+
 libgusb (0.1.3-5) unstable; urgency=low
 
   * Fix upstream homepage link.
diff --git a/debian/compat b/debian/compat
index 45a4fb7..ec63514 100644
--- a/debian/compat
+++ b/debian/compat
@@ -1 +1 @@
-8
+9
diff --git a/debian/control b/debian/control
index 66ced89..5e6d0fb 100644
--- a/debian/control
+++ b/debian/control
@@ -1,11 +1,14 @@
 Source: libgusb
 Priority: extra
 Maintainer: Michal Čihař ni...@debian.org
-Build-Depends: debhelper (= 8.0.0),
+Build-Depends: debhelper (= 9.0.0),
 autotools-dev,
 libgudev-1.0-dev,
 libusb-1.0-0-dev,
-gtk-doc-tools
+gtk-doc-tools,
+gobject-introspection,
+libgirepository1.0-dev,
+valac
 Standards-Version: 3.9.3
 Section: libs
 Homepage: http://www.hughski.com/downloads.html
@@ -18,7 +21,8 @@ Architecture: any
 Depends: libgusb2 (= ${binary:Version}),
 ${misc:Depends},
 libgudev-1.0-dev,
-libusb-1.0-0-dev
+libusb-1.0-0-dev,
+gir1.2-gusb-1.0,
 Description: GLib wrapper around libusb1 - development files
  GUsb is a GObject wrapper for libusb1 that makes it easy to do
  asynchronous control, bulk and interrupt transfers with proper
@@ -42,9 +46,26 @@ Description: GLib wrapper around libusb1 - documentation
 Package: libgusb2
 Architecture: any
 Depends: ${shlibs:Depends}, ${misc:Depends}
+Pre-Depends: ${misc:Pre-Depends}
+Multi-Arch: same
 Description: GLib wrapper around libusb1
  GUsb is a GObject wrapper for libusb1 that makes it easy to do
  asynchronous control, bulk and interrupt transfers with proper
  cancellation and integration into a mainloop.
  .
  This package contains the GUsb shared library.
+
+Package: gir1.2-gusb-1.0
+Section: introspection
+Architecture: any
+Depends: ${gir:Depends},
+ ${shlibs:Depends},
+ ${misc:Depends}
+Description: GObject introspection data for the gusb library
+ This package contains introspection data for libgusb, a GObject
+ wrapper for libusb1 that makes it easy to do asynchronous control,
+ bulk and interrupt transfers with proper cancellation and integration
+ into a mainloop.
+ .
+ It can be used by packages using the GIRepository format to generate
+ dynamic bindings.
diff --git a/debian/gir1.2-gusb-1.0.install b/debian/gir1.2-gusb-1.0.install
new file mode 100644
index 000..d6c0270
--- /dev/null
+++ b/debian/gir1.2-gusb-1.0.install
@@ -0,0 +1 @@
+usr/lib/*/girepository-1.0 usr/lib
diff --git a/debian/libgusb-dev.install b/debian/libgusb-dev.install
index 6cd8ddd..072e031 100644
--- a/debian/libgusb-dev.install
+++ b/debian/libgusb-dev.install
@@ -1,4 +1,6 @@
 usr/include/*
-usr/lib/lib*.a
-usr/lib/lib*.so
-usr/lib/pkgconfig/*
+usr/lib/*/lib*.a
+usr/lib/*/lib*.so
+usr/lib/*/pkgconfig/*
+usr/share/gir-1.0/GUsb-1.0.gir
+usr/share/vala
diff --git a/debian/libgusb2.install b/debian/libgusb2.install
index d0dbfd1..3ddde58 100644
--- a/debian/libgusb2.install
+++ b/debian/libgusb2.install
@@ -1 +1 @@
-usr/lib/lib*.so.*
+usr/lib/*/lib*.so.*
diff --git a/debian/libgusb2.lintian-overrides b/debian/libgusb2.lintian-overrides
new file mode 100644
index 000..45af659
--- /dev/null
+++ b/debian/libgusb2.lintian-overrides
@@ -0,0 +1 @@
+libgusb2 binary: hardening-no-fortify-functions usr/lib/*/libgusb.*
diff --git a/debian/libgusb2.symbols b/debian/libgusb2.symbols
index 0e57d44..4fa661f 100644
--- a/debian/libgusb2.symbols
+++ b/debian/libgusb2.symbols
@@ -35,6 +35,8 @@ libgusb.so.2 libgusb2 #MINVER#
  g_usb_device_list_new@Base 0.1.3
  g_usb_device_open@Base 0.1.3
  g_usb_device_release_interface@Base 0.1.3
+ g_usb_device_reset@Base 0.1.5
  g_usb_device_set_configuration@Base 0.1.3
  g_usb_source_error_quark@Base 0.1.3
  g_usb_source_set_callback@Base 0.1.3
+ g_usb_strerror@Base 0.1.5
diff --git a/debian/rules b/debian/rules
index 45c11b5..5947e5d 100755
--- a/debian/rules
+++ b/debian/rules
@@ -9,13 +9,24 @@
 # Uncomment this to turn on verbose mode.
 #export DH_VERBOSE=1
 
+export DEB_BUILD_MAINT_OPTIONS = hardening=+all
+
 override_dh_install:
-	rm debian/tmp/usr/lib/libgusb.la
+	rm debian/tmp/usr/lib/*/libgusb.la
 	dh_install  --fail-missing
 
 override_dh_installchangelogs:
 	dh_installchangelogs NEWS
 
+override_dh_auto_configure:
+	dh_auto_configure -- --enable-gtk-doc \
+			--enable-introspection \
+			--enable-vala
+
+override_dh_gencontrol:
+	dh_girepository

Bug#699941: Update to 0.1.5 release

[Christopher James Halse Rogers]

Package: libgusb
Severity: wishlist
Tags: patch

It would be nice to get the gusb 0.1.5 release in experimental. It includes
gobject-introspection support which is needed by colord 0.1.29, which I'd like
to get into experimental.

I've got a git tree locally with these changes (and a couple more, like
multiarching and enabling the hardening flags, but they're in separate
commits); tell me where you'd like it and I can push it up for you to
review/merge/cherrypick/ignore.



-- System Information:
Debian Release: wheezy/sid
  APT prefers raring-security
  APT policy: (500, 'raring-security'), (500, 'raring')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.8.0-4-generic (SMP w/4 CPU cores)
Locale: LANG=en_AU.UTF-8, LC_CTYPE=en_AU.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#695232: colord: Please do not install daemon in multiarch path

[Christopher James Halse Rogers]

I've fixed this in git. However, the new version of the colord daemon
loads plugins. Would I be correct in assuming that to consider this bug
fixed these plugins will need to be installed in a non-multiarch path,
too?


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#603306: ITP: fsharp -- Microsoft F# programming language

[Christopher James Halse Rogers]

retitle 603306 ITP: fsharp -- Microsoft F# programming language
owner 603306 !
quit

I plan to package this as a part of the pkg-mono team.


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#668325:

[Christopher James Halse Rogers]

Fun fact: sane_exit(), contrary to its documentation, does not clean up
all the resources used by sane and its backends.

Specifically, a bunch of backends will fail to close the file
descriptors opened in sane_get_devices(). Since colord-sane rescans for
devices on a timer, after some time we run out of fds.

On Ubuntu, this results in a crash in the FORTIFY_SOURCE checks when
sane tries to add an fd  1024 to an FD_SET. It seems we don't build
with that on Debian, so we instead it seems we try to select() on an
invalid fd. With hilarious consequences!

Urgh.

So I'll rework colord-sane to only call sane_get_devices once before
exiting, and just get respawned for each poll.


signature.asc
Description: This is a digitally signed message part

Bug#668325: ditto

[Christopher James Halse Rogers]

Well, that's ruined a perfectly good thread-unsafety hypothesis.

Upstream has entirely removed sane support in git due to bugs, relying
only on udev for scanner enumeration.

I'd like to fix it up instead of remove it, as there are plenty of
network-attached scanners that won't be supported without sane, but
failing that this will be fixed in 0.1.24 when the code is excised.


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#636679: wnpp: 636679: apitrace status?

[Christopher James Halse Rogers]

Paul Wise p...@debian.org wrote:

Hi,

Whats the status of this ITP? It would be great for the games team and
upstream developers if this were added to Debian.

-- 
bye,
pabs

http://wiki.debian.org/PaulWise

It's been mostly done a couple of times in pkg-xorg git.

I've recently updated it to a recent snapshot; it's now waiting for me to 
finish a patch making it work sensibly in the presence of multiarch, then I'll 
hunt for a sponsor. 

I anticipate that this should be done this week. 

Bug#677453: specto: Please upgrade Specto to 0.4.x or later

[Christopher James Halse Rogers]

tags 677453 pending
quit

I've got a 0.4 package locally; it requires a little cleanup and
checking, but it should be ready shortly.



signature.asc
Description: This is a digitally signed message part

Bug#675852: colord: segfault during the boot

[Christopher James Halse Rogers]

On Wed, 2012-06-13 at 13:39 +0200, Paul Menzel wrote:
 ...
  Here is what i did:
  
  1. I rebuilt the colord packages with debuging symbols
 
 It would be great if a package with debugging symbols could be provided.
 James, do you need a separate report for that?

Yeah, I guess I'm resigned to the necessity for a colord-dbg package.
They're a bad workaround for toolchain flaws (and it would be great if
Debian adopted Ubuntu's dbgsym system ☺), but a necessary evil.

The next upload with have a -dbg package.


signature.asc
Description: This is a digitally signed message part

Bug#675852: colord: segfault during the boot

[Christopher James Halse Rogers]

Does colord segfault if run explicitly, either by
running /usr/bin/$(dpkg-architecture -qDEB_HOST_MULTIARCH)/colord/colord
or by starting something which bus-activates colord, such as
gnome-color-manager?

There's not much to debug with just the presence of a segfault ;)


signature.asc
Description: This is a digitally signed message part

Bug#675710: colord-sane write errors in syslog when USB disk is switch on

[Christopher James Halse Rogers]

severity 674710 wishlist
quit

Hello,

 When I switch on an USB disk I obtain the errors hereafter in syslog :
 
 Jun  2 23:34:01 mcfm4 colord-sane: io/hpmud/musb.c 139: unable 
 get_string_descriptor -1: Operation not permitted
 Jun  2 23:34:01 mcfm4 colord-sane: io/hpmud/musb.c 2040: invalid product id 
 string ret=-1
 Jun  2 23:34:01 mcfm4 colord-sane: io/hpmud/musb.c 139: unable 
 get_string_descriptor -1: Operation not permitted
 Jun  2 23:34:02 mcfm4 colord-sane: io/hpmud/musb.c 2045: invalid serial id 
 string ret=-1
 Jun  2 23:34:02 mcfm4 colord-sane: io/hpmud/musb.c 139: unable 
 get_string_descriptor -1: Operation not permitted
 Jun  2 23:34:02 mcfm4 colord-sane: io/hpmud/musb.c 2050: invalid manufacturer 
 string ret=-1
 
 I obtain the same errors when I switch on my HP PSC 1610 printer which have 
 a card reader in it.

This appears to be colord-sane identifying the presence of a new USB
device and trying to probe to see if it's a USB scanner. colord-sane
deliberately doesn't have permission to do this for arbitrary USB
devices, just things that we identify as scanners.

This (should be) is harmless, but it could be silenced. Probably by only
attempting to probe devices udev knows to be scanners.


signature.asc
Description: This is a digitally signed message part

Bug#668325: Please test new version

[Christopher James Halse Rogers]

I can't reproduce this problem on my systems, but it's quite plausible -
the sane backend was split out into a separate binary due to the
difficulties in using sane from a long-running process.

I've just uploaded colord 0.1.21. Although I'm not aware of any specific
colord-sane bugfixes in there, can you please check whether this is
fixed in the new version?

If not, could you please gather some information about what's spinning -
this could be as simple as installing linux-tools, colord debugging
symbols, and running ‘perf top’ when your system is in this state.


signature.asc
Description: This is a digitally signed message part

Bug#668225: Same here

[Christopher James Halse Rogers]

On Tue, 2012-05-15 at 22:58 +0200, Paul Menzel wrote:
...
 Can you reproduce this issue? If yes, could you capture more information
 like documented on the colord site [1]?
 
 I reported that issue upstream now as the Debian maintainer does not
 seem to have time to deal with this.
 

Thanks for that ☺.

I'm most of the way through updating to colord 0.1.20 (which introduces
a some new libraries, and hence needs a bit more attention than normal).
I was hoping to be done by now and ask to retest against the newer
version. Probably by the end of the week.

I think this is likely to end up being a libsane issue; the frequency of
crashes in there was a big part of why upstream split out colord-sane
into a separate process in the first place.



signature.asc
Description: This is a digitally signed message part

Bug#660171: consolekit: Username truncation breaks GDM's frequently-logged-in user check

[Christopher James Halse Rogers]

Package: consolekit
Version: 0.4.5-1build1
Severity: normal
Tags: patch

ConsoleKit's ck-history --frequent output truncates usernames to
8 characters.  This breaks GDM's frequently-logged-in user check,
so remotely authenticated users (ldap, likewise-open) with usernames
greater than 8 characters do not show up. (See also 
https://bugs.launchpad.net/ubuntu/+source/consolekit/+bug/476811 )

This is fixed upstream with
http://cgit.freedesktop.org/ConsoleKit/commit/?id=803cbdfbd78b66b17ead45b1584d65a258e785bf


-- System Information:
Debian Release: wheezy/sid
  APT prefers precise-updates
  APT policy: (500, 'precise-updates'), (500, 'precise-security'), (500, 
'precise'), (100, 'precise-backports')
Architecture: amd64 (x86_64)

Kernel: Linux 3.2.0-16-generic (SMP w/4 CPU cores)
Locale: LANG=en_AU.UTF-8, LC_CTYPE=en_AU.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages consolekit depends on:
ii  dbus   1.4.16-1ubuntu4
ii  libc6  2.15-0ubuntu2
ii  libck-connector0   0.4.5-1build1
ii  libdbus-1-31.4.16-1ubuntu4
ii  libdbus-glib-1-2   0.98-1ubuntu1
ii  libglib2.0-0   2.31.16-0ubuntu2
ii  libpolkit-gobject-1-0  0.104-1
ii  libx11-6   2:1.4.99.1-0ubuntu1
ii  zlib1g 1:1.2.3.4.dfsg-3ubuntu4

Versions of packages consolekit recommends:
ii  libpam-ck-connector  0.4.5-1build1

consolekit suggests no packages.

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#650021: CVE-2011-4349: SQL injection

[Christopher James Halse Rogers]

 tags pending
 thanks

I don't believe this would affect other applications; colord in Debian
is run as the colord system user, rather than as root.

This is fixed in colord 0.1.15, which should be uploaded soon.  Tagging
as such.



signature.asc
Description: This is a digitally signed message part

Bug#648287: postinst install error with 0.1.13.1

[Christopher James Halse Rogers]

Huh.  addgroup is only expected to fail in a couple of ways, none of
which we should be able to hit with that line.

About the only documented way I could see that happening is if you've
already got a non-system group called ‘scanner’.  Does that describe
your system?

Can you remove the --quiet switch from that command and see what gets
printed out?  Thanks.




signature.asc
Description: This is a digitally signed message part

Bug#647481: Extent of testing

[Christopher James Halse Rogers]

You say you've tested this on Ubuntu - to what extent have you tested?
 This looks like it will break everything depending on
libgstreamer0.10-0 because it won't be able to find the plugins until
they're updated to install in the multiarch paths.  If so, then (a)
patches should be posted for the plugin packages (you may have already
done this; I haven't checked), and (b) this package should declare a
Breaks: on the plugin packages with versions less than the version in
which they're multiarched.



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#635251: any news on multi-arch for libpciaccess?

[Christopher James Halse Rogers]

On Wed, 2011-08-24 at 11:54 +0300, Riku Voipio wrote:
 On Wed, Aug 24, 2011 at 09:51:08AM +0200, Julien Cristau wrote:
  On Wed, Aug 24, 2011 at 10:27:56 +0300, Riku Voipio wrote:
   Any updates here? As a udev reverse dependency this is quite important
   for earlt multi-arch conversion.
 
  It'll probably happen on the next upload, but I'm not going to rush
  that.  (What does this have to do with udev?)
 
 Looks like an ubuntuisim. needed inderectly for animated startup graphics,
 sigh. No hurry on debian side then it seems.

libdrm-intel now depends on libpciaccess, plymouth depends on
libdrm-intel, and so on.

The -dev package shouldn't be marked as multiarch; the i386 package is
not parallel installable with the amd64 package as they both
ship /usr/include.  In general, my understanding is that -dev packages
should not be multiarched (yet).

I should apparently also read Debian bug reports more often.  I'd just
pushed a dh7 cleanup branch which also adds multiarch to the
debian-experimental git branch before reading this mail.



signature.asc
Description: This is a digitally signed message part

Bug#635251: any news on multi-arch for libpciaccess?

[Christopher James Halse Rogers]

On Mon, 2011-08-29 at 16:38 +0300, Riku Voipio wrote:
 On Mon, Aug 29, 2011 at 05:48:44PM +1000, Christopher James Halse Rogers 
 wrote:
  The -dev package shouldn't be marked as multiarch; the i386 package is
  not parallel installable with the amd64 package as they both
  ship /usr/include.  In general, my understanding is that -dev packages
  should not be multiarched (yet).
 
 It is ok to incude same files in multiarch coinstallable packages, as long as
 they are identical on all archictectures. Think 
 /usr/share/doc/libpciaccess0/copyright
 for example. Hence all -dev packages where headers are same on all 
 architectures
 are safe for multiarch.

Yeah.  I was confused by the documentation on
wiki.debian.org/MultiArch/Implementation which said policy forbids -dev
packages from being marked as Multi-Arch: same.  As you say, that only
applies where the contents of headers change across architectures.  I've
updated that documentation to hopefully be clearer.



signature.asc
Description: This is a digitally signed message part

Bug#636679: ITP: apitrace -- tool for debugging OpenGL applications and drivers

[Christopher James Halse Rogers]

Package: wnpp
Severity: wishlist
Owner: Christopher James Halse Rogers r...@ubuntu.com


* Package name: apitrace
  Version : 1.0+git
  Upstream Author : José Fonseca jose.r.fons...@gmail.com
* URL : https://github.com/apitrace/apitrace
* License : MIT
  Programming Lang: C++, Python
  Description : tools for debugging OpenGL applications and drivers

apitrace is a suite of tools for debugging OpenGL applications and drivers.
It includes a tool to generate a trace of all the OpenGL calls an applicaton
makes and a tool for replaying these traces and inspecting the rendering and
OpenGL state during the program's execution.
.
This makes it useful for identifying the sources of graphical corruption in
OpenGL applications.



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#633721: ITP: shared-color-profiles -- colour profiles to be used in colour management aware applications

[Christopher James Halse Rogers]

reassign icc-profiles
retitle Provide a package containing only free profiles
severity normal
quit




signature.asc
Description: This is a digitally signed message part

Bug#633721: ITP: shared-color-profiles -- colour profiles to be used in colour management aware applications

[Christopher James Halse Rogers]

On Wed, 2011-07-13 at 13:58 +0200, Jonas Smedegaard wrote:
 On 11-07-13 at 05:22pm, Christopher James Halse Rogers wrote:
  Package: wnpp
  Severity: wishlist
  Owner: Christopher James Halse Rogers r...@ubuntu.com
  
  
  * Package name: shared-color-profiles
Version : 0.1.4
Upstream Author : Richard Hughes rich...@hughsie.com
  * URL : http://www.freedesktop.org/software/colord/
  * License : confused (zlib, CC-BY-SA, MIT, GPL)
Programming Lang: data
Description : colour profiles to be used in colour management aware 
  applications
  
  The shared-color-profiles package contains profiles for various colour 
  spaces
  and standard devices. When combined with profiles for your specific devices,
  these can be used by colour management aware programs to provide a
  colour-managed workflow, ensuring that the colours you see on your camera,
  scanner, display, and printer all match.
  
  These profiles will be used by colord.
  
  The copyright and license for some of the files are a bit confused.  I'm 
  checking with
  upstream to get that sorted out.
 
 I suspect those ICC files not really has freedesktop.org as upstream 
 URL, but are only convenience copies from other canonical sources.
 

This is correct.

 Are you aware of the package icc-profiles in contrib?
 
 Seems more sensible that we join forces on improving that package - 
 including splitting it into binary icc-profiles and 
 icc-profiles-contrib packages.

Yeah, that makes sense.  colord would really like at least a base set of
colour spaces available, so it'd be good to get a set of fully-free
profiles into Debian proper.

There seem to some profiles in s-c-p that aren't in icc-profiles; if
they're useful we should pull them over.

Also, if I understand it correctly, all the profiles from
shared-color-profiles have had cd-fix-profiles (from colord) run over
them to embed a profile ID to make startup faster.

How do you want to do this?

Thanks,
Chris


signature.asc
Description: This is a digitally signed message part

Bug#633721: ITP: shared-color-profiles -- colour profiles to be used in colour management aware applications

[Christopher James Halse Rogers]

Package: wnpp
Severity: wishlist
Owner: Christopher James Halse Rogers r...@ubuntu.com


* Package name: shared-color-profiles
  Version : 0.1.4
  Upstream Author : Richard Hughes rich...@hughsie.com
* URL : http://www.freedesktop.org/software/colord/
* License : confused (zlib, CC-BY-SA, MIT, GPL)
  Programming Lang: data
  Description : colour profiles to be used in colour management aware 
applications

The shared-color-profiles package contains profiles for various colour spaces
and standard devices. When combined with profiles for your specific devices,
these can be used by colour management aware programs to provide a
colour-managed workflow, ensuring that the colours you see on your camera,
scanner, display, and printer all match.

These profiles will be used by colord.

The copyright and license for some of the files are a bit confused.  I'm 
checking with
upstream to get that sorted out.



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#633518: ITP: colord -- system service to manage device colour profiles

[Christopher James Halse Rogers]

On Mon, 2011-07-11 at 12:16 +0200, Michael Biebl wrote:
 Hi,
 
 Am 11.07.2011 08:06, schrieb Christopher James Halse Rogers:
  Package: wnpp
  Severity: wishlist
  Owner: Christopher James Halse Rogersr...@ubuntu.com
 
 
  * Package name: colord
 Version : 0.1.10
 Upstream Author : Richard Hughesrich...@hughsie.com
  * URL : http://colord.hughsie.com/
  * License : GPL2+, LGPL2+
 Programming Lang: C
 Description : system service to manage device colour profiles
 
  colord is a system service that makes it easy to manage, install and 
  generate
  colour profiles to accurately color manage input and output devices.
 
 have you seen [1]? Assuming your email address I guess so.

Yes.  Til cornered me in Dublin and persuaded me to do the packaging for
colord. :)

 What is the state of this effort?

colord is essentially ready to go; I'm just checking the polish at this
point.  After that a number of printingish packages need to be rebuilt
with colord support, and the new version of gnome-color-manager can get
built.

 Has an (alioth) team be formed to 
 maintain colord and related packages?
 

There aren't really many related packages, or at least not related
packages that don't already have a better home - the closest one would
be gnome-color-manager, which is already in pkg-gnome.

I was planning to maintain it with Till Kamppeter and Martin Pitt in the
printing team.  I could be persuaded to make a dedicated team if that's
what's wanted.

Chris


signature.asc
Description: This is a digitally signed message part

Bug#633518: ITP: colord -- system service to manage device colour profiles

[Christopher James Halse Rogers]

Package: wnpp
Severity: wishlist
Owner: Christopher James Halse Rogers r...@ubuntu.com


* Package name: colord
  Version : 0.1.10
  Upstream Author : Richard Hughes rich...@hughsie.com
* URL : http://colord.hughsie.com/
* License : GPL2+, LGPL2+
  Programming Lang: C
  Description : system service to manage device colour profiles

colord is a system service that makes it easy to manage, install and generate
colour profiles to accurately color manage input and output devices.



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#633518: ITP: colord -- system service to manage device colour profiles

[Christopher James Halse Rogers]

On Mon, 2011-07-11 at 09:51 +0300, Timo Juhani Lindfors wrote:
 Christopher James Halse Rogers r...@ubuntu.com writes:
Description : system service to manage device colour profiles
 
  colord is a system service that makes it easy to manage, install and 
  generate
  colour profiles to accurately color manage input and output devices.
 
 You might want to mention freedesktop.org in the description. And
 possibly dbus too?
 

I'm somewhat hesitant to include implementation details in package
descriptions.  That said, this isn't really a user-visible package so
some extra technical detail is probably appropriate.



signature.asc
Description: This is a digitally signed message part

Bug#633518: ITP: colord -- system service to manage device colour profiles

[Christopher James Halse Rogers]

On Mon, 2011-07-11 at 10:23 +0300, Timo Juhani Lindfors wrote:
 Hi,
 
 Christopher James Halse Rogers r...@ubuntu.com writes:
  I'm somewhat hesitant to include implementation details in package
  descriptions.
 
 I generally agree.
 
  That said, this isn't really a user-visible package so some extra
  technical detail is probably appropriate.
 
 Yes, as a system administrator I want to know what sort of a service I
 am installing to the system when I install this package. Since it is a
 dbus service I can't disable it with policy-rc.d or update-rc.d. Is it
 possible to have the service installed but not enabled somehow?
 
 -Timo

Like any autostarted dbus service you could add a servicedir
to /etc/dbus/system.conf and stick an empty
org.freedesktop.ColorManager.service file in that directory to prevent
the service from getting loaded.


signature.asc
Description: This is a digitally signed message part

Bug#612529: Upstream status

[Christopher James Halse Rogers]

The status of the Ubuntu patch upstream, as best I can remember it, is
that they think it's fixing the problem in the wrong place.  The libc
dynamic loader is meant to have some special magic set aside so that at
least some TLS code with the initial-exec class can be loaded, but this
clearly isn't working properly.

I started investigating this, then got busy with other things.


signature.asc
Description: This is a digitally signed message part

Bug#627367: xkb-data: Upgrade from 1.8 to 2.1 adds deadkeys to dvorak-intl

[Christopher James Halse Rogers]

Package: xkb-data
Version: 2.1-2
Severity: normal


In commit b611a52d8 upstream kindly renamed dvorak-intl (which did not have
dead keys) to dvorak-alt-intl.  dvorak-intl now has dead keys, so upgrades from
1.8 or earlier to 2.1 will switch the user's keyboard behaviour.

On upgrade from = 1.8 we should transition users from dvorak-intl to
dvorak-alt-intl.

See also https://bugs.launchpad.net/ubuntu/+source/xkeyboard-config/+bug/742683,
but I've verified this also affects Debian by upgrading an oldstable chroot to
wheezy.


-- System Information:
Debian Release: squeeze/sid
  APT prefers natty-updates
  APT policy: (500, 'natty-updates'), (500, 'natty-security'), (500, 'natty')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.38-8-generic (SMP w/2 CPU cores)
Locale: LANG=en_AU.utf8, LC_CTYPE=en_AU.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#618451: gnome-do: leaks huge amounts of memory

[Christopher James Halse Rogers]

On Mon, 2011-04-25 at 18:29 +0100, Iain Lane wrote:
 reassign 618451 src:gnome-do-plugins
 severity 618451 important
 thanks
 
 Hi,
 
 On Mon, Apr 25, 2011 at 01:47:15PM +0100, Adam D. Barratt wrote:
 On Mon, 2011-03-21 at 17:34 +1100, Christopher James Halse Rogers wrote:
  Hm.  It looks like this is a problem with the mono.addins plugin manager
  rather than specifically a Files  Folders plugin issue.
 
  I'll have a look at that code and see where this is falling over.  I'll
  get back to you if I need more information, or something tested.
 
 Any news on that?  This is now blocking the transition of gdata-sharp.
 
 I just looked into this and I'm going to downgrade to important — the
 plugin isn't enabled by default.
 
 Nevertheless the bug is annoying and ought to be fixed — I hope you
 can find time to look into it soon. :-)

I'm going to be looking at (another!) bug in Files  Folders tonight;
I'll see if I can't fix this while I'm there.


signature.asc
Description: This is a digitally signed message part

Bug#618415: gnome-do: kazach translation in /etc/xdg/ is misformatted

[Christopher James Halse Rogers]

tags 618415 + pending
thanks

This has been fixed upstream in r1334.  I've cherry-picked that commit
into git, and this will be fixed in the next upload.

Thanks

Chris


signature.asc
Description: This is a digitally signed message part

Bug#618451: gnome-do: leaks huge amounts of memory

[Christopher James Halse Rogers]

Hm.  It looks like this is a problem with the mono.addins plugin manager
rather than specifically a Files  Folders plugin issue.

I'll have a look at that code and see where this is falling over.  I'll
get back to you if I need more information, or something tested.

Chris


signature.asc
Description: This is a digitally signed message part

Bug#615667: grass: Use xdg-open for GRASS_HTML_BROWSER when available

[Christopher James Halse Rogers]

Package: grass
Version: 6.4.0-2
Severity: wishlist
Tags: patch

When xdg-open is available it would be good to use it for GRASS_HTML_BROWSER
instead of x-www-browser.  The x-www-browser alternate is system-wide, while
xdg-open will use the browser the user has configured in their DE (eg: in
Preferred Applications in GNOME).

The attached git commit changes the www-browser patch to prefer xdg-open over
x-www-browser when available.


-- System Information:
Debian Release: squeeze/sid
  APT prefers natty-updates
  APT policy: (500, 'natty-updates'), (500, 'natty-security'), (500, 'natty')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.38-5-generic (SMP w/2 CPU cores)
Locale: LANG=en_AU.utf8, LC_CTYPE=en_AU.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
From 596d2d56a70f76d5b5d43d50d722ad16677bee5c Mon Sep 17 00:00:00 2001
From: Thibault Lemaitre thibault.lemai...@laposte.net
Date: Mon, 28 Feb 2011 11:04:59 +1100
Subject: [PATCH] Use xdg-open for GRASS_HTML_BROWSER when available.

---
 debian/patches/www-browser |   15 ---
 1 files changed, 8 insertions(+), 7 deletions(-)

diff --git a/debian/patches/www-browser b/debian/patches/www-browser
index e814734..06d36e6 100644
--- a/debian/patches/www-browser
+++ b/debian/patches/www-browser
@@ -1,8 +1,8 @@
 Index: grass/lib/init/init.sh
 ===
 grass.orig/lib/init/init.sh2011-02-14 17:15:01.0 +0100
-+++ grass/lib/init/init.sh 2011-02-14 17:28:01.0 +0100
-@@ -325,53 +325,11 @@
+--- grass.orig/lib/init/init.sh2011-02-28 10:59:35.514747971 +1100
 grass/lib/init/init.sh 2011-02-28 11:00:17.216340777 +1100
+@@ -325,53 +325,13 @@
  
  # try and find a web browser if one isn't already specified
  if [ ! $GRASS_HTML_BROWSER ] ; then
@@ -18,7 +18,11 @@ Index: grass/lib/init/init.sh
 -  iexplore=$SYSTEMDRIVE/Program Files/Internet Explorer/iexplore.exe
 -  if [ -f $iexplore ] ; then
 -  GRASS_HTML_BROWSER=$iexplore
--  else
++  if [ -x /usr/bin/xdg-open ] ; then
++  GRASS_HTML_BROWSER=xdg-open
++  elif [ -x /usr/bin/x-www-browser ] ; then
++  GRASS_HTML_BROWSER=x-www-browser
+   else
 -  GRASS_HTML_BROWSER=iexplore
 -  fi
 -
@@ -38,9 +42,6 @@ Index: grass/lib/init/init.sh
 -  done
 -  if [ -n $GRASS_HTML_BROWSER ] ; then
 - break
-+  if [ -x /usr/bin/x-www-browser ] ; then
-+  GRASS_HTML_BROWSER=x-www-browser
-+  else 
 +  GRASS_HTML_BROWSER=true
fi
 -  done
-- 
1.7.4.1

Bug#606393: [pkg-cli-apps-team] Bug#606393: gnome-do: crashes on start with Unhandled Exception: System.TypeLoadException

[Christopher James Halse Rogers]

Nothing in that debug log appears to be a problem.

I'm not sure what's causing your error - Do.Interface.ControlOrientation
is an enum defined in the Do.Interface.Linux assembly, so I don't see
how the definition could fail to be found.


signature.asc
Description: This is a digitally signed message part

Bug#606169: tracker: Build fails against evolution-data-server 2.32

[Christopher James Halse Rogers]

Package: tracker
Version: 0.8.17-1
Severity: normal
Tags: patch
User: ubuntu-de...@lists.ubuntu.com
Usertags: origin-ubuntu natty ubuntu-patch



*** /tmp/tmpyMnJSs
In Ubuntu, we've applied the following patch based on BGO: 628672.
It fixes the build against evolution-data-server 2.32.

This should be fixed in the next upstream release.

-- System Information:
Debian Release: squeeze/sid
  APT prefers natty-updates
  APT policy: (500, 'natty-updates'), (500, 'natty-security'), (500, 'natty')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.37-7-generic (SMP w/2 CPU cores)
Locale: LANG=en_AU.utf8, LC_CTYPE=en_AU.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
diff -Nru tracker-0.8.17/debian/patches/02-build-against-eds-2.32.patch tracker-0.8.17/debian/patches/02-build-against-eds-2.32.patch
--- tracker-0.8.17/debian/patches/02-build-against-eds-2.32.patch	1970-01-01 11:00:00.0 +1100
+++ tracker-0.8.17/debian/patches/02-build-against-eds-2.32.patch	2010-11-26 22:18:12.0 +1100
@@ -0,0 +1,14 @@
+Index: tracker-0.8.17/src/plugins/evolution/tracker-evolution-plugin.c
+===
+--- tracker-0.8.17.orig/src/plugins/evolution/tracker-evolution-plugin.c	2010-11-26 22:17:52.616142002 +1100
 tracker-0.8.17/src/plugins/evolution/tracker-evolution-plugin.c	2010-11-26 22:18:08.886142002 +1100
+@@ -41,7 +41,9 @@
+ #include sqlite3.h
+ 
+ #include camel/camel.h
++#ifndef HAVE_EDS_2_31_2
+ #include camel/camel-db.h
++#endif
+ 
+ #include mail/mail-config.h
+ #include mail/mail-session.h
diff -Nru tracker-0.8.17/debian/patches/series tracker-0.8.17/debian/patches/series
--- tracker-0.8.17/debian/patches/series	2010-09-21 13:23:42.0 +1000
+++ tracker-0.8.17/debian/patches/series	2010-11-26 22:17:49.0 +1100
@@ -1,3 +1,4 @@
 # Debian patches for tracker
 01-upower.patch
+02-build-against-eds-2.32.patch
 99-autoreconf.patch

Bug#546741: Kernel patches

[Christopher James Halse Rogers]

The kernel patches on that upstream page were incorporated into the
mainline kernel around 2008.  They're no problem.

So an xf86-video-sis-imedia source package is feasible, although ugly.
Merging the 671/771 support into the freedesktop.org driver would
clearly be better, but… urgh that's a big diff.


signature.asc
Description: This is a digitally signed message part

Bug#586062: mesa-dri-experimental: Missing /usr/lib/dri/nouveau_vieux_dri.so

[Christopher James Halse Rogers]

nouveau_vieux_dri.so is the classic mesa driver for nv04-nv2x nVidia
chips.  It currently doesn't build against the libdrm in experimental.

It appears to be more usable in mesa git master.  I plan to add it when
we start packaging from the mesa 7.9 branch.


signature.asc
Description: This is a digitally signed message part

Bug#585618: mesa: FTBFS on kfreebsd and hurd (gallium/auxiliary: os/os_time.c:51:4: error: #error Unsupported OS

[Christopher James Halse Rogers]

On Sat, 2010-06-12 at 13:59 +0200, Julien Cristau wrote:
 Package: mesa
 Version: 7.8.1-2
 Severity: serious
 
 mesa in experimental ftbfs on kfreebsd and hurd:
 https://buildd.debian.org/pkg.cgi?pkg=mesamaint=dist=experimental
 
 gcc -c -I. -I../../../src/gallium/include
 -I../../../src/gallium/auxiliary -I../../../src/gallium/drivers  -Wall
 -g -O2 -Wall -Wmissing-prototypes -std=c99 -ffast-math
 -fvisibility=hidden -fno-strict-aliasing  -fPIC   -D_GNU_SOURCE
 -DPTHREADS -DHAVE_POSIX_MEMALIGN -DGLX_USE_TLS -DPTHREADS
 -DUSE_EXTERNAL_DXTN_LIB=1 -DIN_DRI_DRIVER -DGLX_INDIRECT_RENDERING
 -DHAVE_ALIAS -D__STDC_CONSTANT_MACROS os/os_time.c -o os/os_time.o
 os/os_time.c:51:4: error: #error Unsupported OS
 os/os_time.c: In function 'os_time_get':
 os/os_time.c:93: warning: control reaches end of non-void function
 make[4]: *** [os/os_time.o] Error 1
 
 Would be nice if porters could take a look.  Otherwise we can disable
 gallium there (I thought we already did, but apparently not).

We only build the DRI drivers on Linux but we do build gallium swrast
everywhere.



signature.asc
Description: This is a digitally signed message part

Bug#585651: nouveau: Fails to use same resolution than console

[Christopher James Halse Rogers]

This probably is due to your high resolution and low video memory - a
1900x1200 truecolour framebuffer takes up slightly more than half your
VRAM.



signature.asc
Description: This is a digitally signed message part

Bug#570183: dh_clideps: Fail with an error when shlib dependencies cannot be resolved

[Christopher James Halse Rogers]

Package: cli-common-dev
Version: 0.7.1
Severity: wishlist
Tags: patch

It would be nice if dh_clideps would fail the build when it can't resolve 
shared library dependencies due to a genuine problem, such as when 
libgnome-desktop-2 bumps its SONAME *again*.

To this end, the first attached patch adds a function to resolve modulerefs 
against private shlibs, by first searching the current binary package, and if 
that fails the other binary packages built from the same source.  If the 
private library is found in a binary package other than the one containing the 
assembly, a strong versioned dependency is added against the binary package 
containing the shlib.

If a moduleref cannot be resolved, either by shlibs or to a private library, 
dh_clideps fails with an error.

The second patch adds an -X option to exclude paths from the search for 
assemblies.  This may not want to be applied, or may want to be re-worked into 
a patch to make unresolved references in the excluded paths not a fatal error.  
This patch was written because the gnome-sharp2-examples package ships some 
assemblies with unresolved references to glib-2.0.

This has been tested against most of the reverse build dependencies of 
cli-common-dev.  With the exception of gnome-sharp2 it doesn't break any 
existing builds.


-- System Information:
Debian Release: squeeze/sid
  APT prefers lucid-security
  APT policy: (500, 'lucid-security'), (500, 'lucid')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.32-12-generic (SMP w/2 CPU cores)
Locale: LANG=en_AU.utf8, LC_CTYPE=en_AU.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages cli-common-dev depends on:
ii  debhelper7.4.11ubuntu1   helper programs for debian/rules
ii  libxml-dom-perl  1.44-1  Perl module for building DOM Level
ii  mono-devel [strong-n 2.4.3+dfsg-1ubuntu1 Mono development tools
ii  mono-utils [cil-disa 2.4.3+dfsg-1ubuntu1 Mono utilities
ii  perl-modules 5.10.1-8ubuntu1 Core Perl modules

cli-common-dev recommends no packages.

cli-common-dev suggests no packages.

-- no debconf information
From 1b56f72feb6dc3b69643aa83921a1935d94f3c7d Mon Sep 17 00:00:00 2001
From: Christopher James Halse Rogers r...@ubuntu.com
Date: Tue, 9 Feb 2010 15:33:38 +1100
Subject: [PATCH 1/2] Add resolvePrivateLibrary function and use it to make 
dh_clideps error rather than warn when shlib refs can't be resolved

---
 dh_clideps |   66 ---
 1 files changed, 62 insertions(+), 4 deletions(-)

diff --git a/dh_clideps b/dh_clideps
index e349f2f..d748f96 100755
--- a/dh_clideps
+++ b/dh_clideps
@@ -576,14 +576,24 @@ sub resolveShlibRefs {
next;
  }
  my $target = $dllmapdata{$name};
+ my $fullTarget = $target;
  
  if (defined($target)) {
$target = basename($target);
verbose_print(Resolved moduleref via DLL map: $name to: $target);
  } elsif (defined($shlibdata{$name})) {
verbose_print(Resolved moduleref via direct match in shlibs);
+ } elsif (resolvePrivateLibrary($package, $name, $package)) {
+   # There is no DllMap, but the package ships the private library 
alongside the assembly
+   verbose_print(Resolved moduleref to private library $name);
+   next;
+ } elsif (resolvePrivateLibrary($package, lib . $name . .so, 
$package)) {
+   # There is no DllMap, the assembly is relying on Mono's foo - 
libfoo.so
+   # translation, and is shipping libfoo.so alongside the assembly
+   verbose_print(Resolved moduleref to private library lib . $name . 
.so);
+   next;
  } else {
-   warning(Warning: Could not resolve moduleref: $name for: 
$assembly_filename!);
+   error(Could not resolve moduleref: $name for: $assembly_filename!);
next;
  }
  
@@ -596,8 +606,24 @@ sub resolveShlibRefs {
# for DLL maps that have an unversioned library as target
$pkgref = $shlibdata{$target..0};
  } else {
-   warning(Warning: Missing shlibs entry: $target or $name for: 
$assembly_filename!);
-   next;
+if(!resolvePrivateLibrary($package, $fullTarget, $package)) {
+# Private library can't be found in the current package.  Try to 
resolve it
+# in the other binary packages, and add a strong dependency if we 
find it.
+foreach my $binary_package (@{$dh{DOPACKAGES}}) {
+if(resolvePrivateLibrary($package, $fullTarget, 
$binary_package)) {
+verbose_print(Found private library in $binary_package);
+$pkgref = $binary_package .  (= \${binary:Version});
+verbose_print(pkgref is $pkgref);
+last;
+}
+}
+if (!defined($pkgref)) {
+error(Missing shlibs entry: $target or $name for: 
$assembly_filename!);
+}
+} else {
+verbose_print(Found private library

Bug#570181: mono-gac: Installing signature-remapped assemblies fails unless mono-runtime is configured first

[Christopher James Halse Rogers]

Package: mono-gac
Version: 2.4.2.3+dfsg-2
Severity: important

Discovered while investigating why rawang's mono-uia packages failed to install 
on the buildds.

09:38 RAOF Ok.  So, I think rawang's mono-uia problem is interesting from a 
policy perspective.  The problem is this: mono-uia is delaysigned with the 
winfx3 public key.  Since mono obviously don't have that private key, the 
assemblies get resigned with mono.snk, which works because of a mapping in 
machine.config.

09:41 RAOF Now, when mono-uia packages are being installed in a chroot, it 
goes like this: everything gets unpacked, then mono-gac tries to get 
configured.  As a part of that, it tries to install libmono-uia3.0 into the 
GAC.  However, mono-runtime _isn't_ configured, and in particular 
machine.config is currently machine.config.dpkg-new, so gacutil can't verify 
the strongname, GAC install fails, mono-gac can't be configured, and so 
mono-runtime can't be configured.

Everything then ends up nicely wedged.

This could be resolved by using a dpkg trigger to run the GAC install, so that 
mono-runtime will be configured before trying to install assemblies into the 
GAC.


-- System Information:
Debian Release: squeeze/sid
  APT prefers lucid-security
  APT policy: (500, 'lucid-security'), (500, 'lucid')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.33-999-generic (SMP w/2 CPU cores)
Locale: LANG=en_AU.UTF-8, LC_CTYPE=en_AU.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages mono-gac depends on:
ii  mono-2.0-gac  2.4.2.3+dfsg-2 Mono GAC tool (for CLI 2.0)

mono-gac recommends no packages.

mono-gac suggests no packages.

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20100105225637.5177.62027.report...@ed.cooperteam.net

Bug#570182: gjs: Enable the test suite, drop some unnecessary dependencies

[Christopher James Halse Rogers]

Package: gjs
Version: 0.4-4
Severity: wishlist
Tags: patch
User: ubuntu-de...@lists.ubuntu.com
Usertags: origin-ubuntu lucid ubuntu-patch



*** /tmp/tmpZdM6qE
In Ubuntu, we've applied the attached patch to achieve the following:

+ debian/control:
  - Add Build-Depends on dbus-x11 and uuid-runtime for the testsuite
+ debian/rules:
  - Apply ltmain-add-as-needed.patch after autoreconf to fix libtool
handling of --as-needed.
  - Add -Wl,--as-needed -Wl,z,defs to LDFLAGS.  This removes unnecessary
linkages from the libraries, fixing some dpkg-shlibdeps warnings.
  - Remove testEverythingBasic.js and testEverythingEncapsulated.js
from the test suite.  These fail because our gobject-introspection
packages don't include Everything-1.0.typelib.  (Launchpad bug
510426 filed for this, Debian bug is #550478).
  - Enable the rest of the testsuite.
+ debian/ltmain-add-as-needed.patch
  - Patch from Debian bug 347650 to fix libtools --as-needed handling.

We thought you might be interested in doing the same. 


-- System Information:
Debian Release: squeeze/sid
  APT prefers lucid-security
  APT policy: (500, 'lucid-security'), (500, 'lucid')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.32-10-generic (SMP w/2 CPU cores)
Locale: LANG=en_AU.utf8, LC_CTYPE=en_AU.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
diff -u gjs-0.4/debian/rules gjs-0.4/debian/rules
--- gjs-0.4/debian/rules
+++ gjs-0.4/debian/rules
@@ -13,15 +13,32 @@
-	dh_auto_configure --sourcedirectory=_build
+	cd _build  patch -p3  ../debian/ltmain-add-as-needed.patch
+	dh_auto_configure --sourcedirectory=_build -- LDFLAGS=$(LDFLAGS) -Wl,-z,defs -Wl,--as-needed
 
 override_dh_auto_build:
 	dh_auto_build --sourcedirectory=_build
 
 override_dh_auto_install:
 	dh_auto_install --sourcedirectory=_build
 
 override_dh_clean:
 	rm -rf _build
 	dh_clean
 
 override_dh_auto_test:
-	true
+	# Our gir-repository packages don't build Everything.typelib, so the
+	# Everything tests will fail.  Just remove them.
+	rm _build/test/js/testEverythingBasic.js
+	rm _build/test/js/testEverythingEncapsulated.js
+	dh_auto_test --sourcedirectory=_build
diff -u gjs-0.4/debian/changelog gjs-0.4/debian/changelog
diff -u gjs-0.4/debian/control gjs-0.4/debian/control
--- gjs-0.4/debian/control
+++ gjs-0.4/debian/control
@@ -1,22 +1,25 @@
 Source: gjs
 Section: interpreters
 Priority: extra
-Maintainer: Gustavo Noronha Silva gustavo.noro...@collabora.co.uk
+Maintainer: Ubuntu Developers ubuntu-devel-disc...@lists.ubuntu.com
+XSBC-Original-Maintainer: Gustavo Noronha Silva gustavo.noro...@collabora.co.uk
 Build-Depends: debhelper (= 7.3.0),
autotools-dev,
autoconf,
automake1.7,
libtool,
pkg-config,
libdbus-glib-1-dev,
libglib2.0-dev,
libgirepository1.0-dev (= 0.6.3),
gobject-introspection (= 0.6.5-3),
gir1.0-glib-2.0,
-   xulrunner-dev (= 1.9)
+   xulrunner-dev (= 1.9),
+   dbus-x11,
+   uuid-runtime
 Standards-Version: 3.8.3
 Homepage: http://live.gnome.org/Gjs
 
 Package: gjs
@@ -32,7 +35,12 @@
 Package: libgjs0
 Architecture: any
 Section: libs
 Depends: ${shlibs:Depends}, ${misc:Depends}, ${gir:Depends}
 Description: Mozilla-based javascript bindings for the GNOME platform
  Makes it possible for applications to use all of GNOME's platform
  libraries using the Javascript language. It's mainly based on the
@@ -45,8 +53,9 @@
 Section: libdevel
 Depends: libgjs0 (= ${binary:Version}),
  ${misc:Depends},
  libgirepository1.0-dev
 Description: Mozilla-based javascript bindings for the GNOME platform
  Makes it possible for applications to use all of GNOME's platform
  libraries using the Javascript language. It's mainly based on the
only in patch2:
unchanged:
--- gjs-0.4.orig/debian/ltmain-add-as-needed.patch
+++ gjs-0.4/debian/ltmain-add-as-needed.patch
@@ -0,0 +1,43 @@
+Description: Patch to add --as-needed magic to ltmain.sh
+ This patch needs to be applied after autoreconf is run, and autoreconf should
+ be copying files, not symlinking (ie: the '-s' option should not be given) so
+ that system-wide files aren't accidentally patched.
+ .
+ This should be more robust than copying a pre-patched version of ltmain.sh
+ from debian/ on build, as ltmain.sh needs to be exactly the same version as 
+ the rest of libtool.
+ .
+ This is somewhat of a hack taken from an attachment to Debian bug 347650
+Author: Josselin Mouette j...@debian.org
+
+
+--- libtool-2.2.6b/libltdl/config/ltmain.sh	2010-01-04 05:33:00.810981679 +
 libtool-as-needed/libltdl/config/ltmain.sh	2010-01-04 00:18:29.71531 +
+@@ -4716,6 +4716,11 @@
+ 	arg=$func_stripname_result
+ 	;;
+ 
++  -Wl,--as-needed)
++	deplibs=$deplibs $arg
++	continue
++	;;
++
+   -Wl,*)
+ 	func_stripname '-Wl,' '' $arg
+ 	

Bug#570180: ITP: docky -- Elegant, powerful, clean dock

[Christopher James Halse Rogers]

Package: wnpp
Severity: wishlist
Owner: Christopher James Halse Rogers r...@ubuntu.com

* Package name: docky
  Version : 2.0.0
  Upstream Author : Jason Smith jassm...@gmail.com
* URL : https://launchpad.net/docky
* License : GPL-3+
  Programming Lang: C#, Python
  Description : Elegant, powerful, clean dock

 Provides an application launcher, running application management, and various
 docklets including a CPU monitor, weather report and clock.  It is similar
 to other docks such as AWN and cairo-dock.
 .
 Applications can integrate with Docky to add extra items to their context
 menus or modify their icons to display more information.  This package 
 includes integration helpers for a number of applications, including Banshee,
 Rhythmbox, Deluge, Tomboy and Zeitgeist.
 .
 Docky is derived from the GNOME Do docky interface.



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20100217011952.27960.55357.report...@ed.cooperteam.net

Bug#568970: dh_clideps should fail when unable to resolve shlib dependencies

[Christopher James Halse Rogers]

Package: cli-common-dev
Severity: wishlist
Tags: patch

It would be nice if dh_clideps would fail the build when it can't
resolve shared library dependencies due to a genuine problem, such as
when libgnome-desktop-2 bumps its SONAME *again*.

To this end, the first attached patch adds a function to resolve
modulerefs against private shlibs, by first searching the current binary
package, and if that fails the other binary packages built from the same
source.  If the private library is found in a binary package other than
the one containing the assembly, a strong versioned dependency is added
against the binary package containing the shlib.

If a moduleref cannot be resolved, either by shlibs or to a private
library, dh_clideps fails with an error.

The second patch adds an -X option to exclude paths from the search for
assemblies.  This may not want to be applied, or may want to be
re-worked into a patch to make unresolved references in the excluded
paths not a fatal error.  This patch was written because the
gnome-sharp2-examples package ships some assemblies with unresolved
references to glib-2.0.

This has been tested against most of the reverse build dependencies of
cli-common-dev.  With the exception of gnome-sharp2 it doesn't break
any existing builds.


From 1b56f72feb6dc3b69643aa83921a1935d94f3c7d Mon Sep 17 00:00:00 2001
From: Christopher James Halse Rogers r...@ubuntu.com
Date: Tue, 9 Feb 2010 15:33:38 +1100
Subject: [PATCH 1/2] Add resolvePrivateLibrary function and use it to make dh_clideps error rather than warn when shlib refs can't be resolved

---
 dh_clideps |   66 ---
 1 files changed, 62 insertions(+), 4 deletions(-)

diff --git a/dh_clideps b/dh_clideps
index e349f2f..d748f96 100755
--- a/dh_clideps
+++ b/dh_clideps
@@ -576,14 +576,24 @@ sub resolveShlibRefs {
next;
  }
  my $target = $dllmapdata{$name};
+ my $fullTarget = $target;
  
  if (defined($target)) {
$target = basename($target);
verbose_print(Resolved moduleref via DLL map: $name to: $target);
  } elsif (defined($shlibdata{$name})) {
verbose_print(Resolved moduleref via direct match in shlibs);
+ } elsif (resolvePrivateLibrary($package, $name, $package)) {
+   # There is no DllMap, but the package ships the private library alongside the assembly
+   verbose_print(Resolved moduleref to private library $name);
+   next;
+ } elsif (resolvePrivateLibrary($package, lib . $name . .so, $package)) {
+   # There is no DllMap, the assembly is relying on Mono's foo - libfoo.so
+   # translation, and is shipping libfoo.so alongside the assembly
+   verbose_print(Resolved moduleref to private library lib . $name . .so);
+   next;
  } else {
-   warning(Warning: Could not resolve moduleref: $name for: $assembly_filename!);
+   error(Could not resolve moduleref: $name for: $assembly_filename!);
next;
  }
  
@@ -596,8 +606,24 @@ sub resolveShlibRefs {
# for DLL maps that have an unversioned library as target
$pkgref = $shlibdata{$target..0};
  } else {
-   warning(Warning: Missing shlibs entry: $target or $name for: $assembly_filename!);
-   next;
+	 if(!resolvePrivateLibrary($package, $fullTarget, $package)) {
+	 # Private library can't be found in the current package.  Try to resolve it
+	 # in the other binary packages, and add a strong dependency if we find it.
+	 foreach my $binary_package (@{$dh{DOPACKAGES}}) {
+		 if(resolvePrivateLibrary($package, $fullTarget, $binary_package)) {
+		 verbose_print(Found private library in $binary_package);
+		 $pkgref = $binary_package .  (= \${binary:Version});
+		 verbose_print(pkgref is $pkgref);
+		 last;
+		 }
+	 }
+	 if (!defined($pkgref)) {
+		 error(Missing shlibs entry: $target or $name for: $assembly_filename!);
+	 }
+	 } else {
+	 verbose_print(Found private library $target for $name);
+	 next;
+	 }
  }
 
  my %overriddenRef = resolveOverride($package, $pkgref);
@@ -606,10 +632,42 @@ sub resolveShlibRefs {
  push(@{$ret{suggests}},   $overriddenRef{suggests});
} 
close(F);
-   
+
return %ret;
 }
 
+sub resolvePrivateLibrary {
+my $package = shift;
+my $target = shift;
+my $resolveIn = shift;
+my $library_file;
+
+use File::Spec;
+if (File::Spec-file_name_is_absolute($target)) {
+	# If the DLLMap target is absolute, we should check that the target
+	# exists in that location.  Since we're currently in the directory
+	# with the assembly, we need to back out first.
+	my @targetDirs = File::Spec-splitdir($target);
+	my @cwdComponents = File::Spec-splitdir(File::Spec-rel2abs(File::Spec-curdir()));
+	my @upDirs = (../, ../);
+
+	# Find where the last occurance of tmpdir is in $curdir, and add enough
+	# updirs to get there from cwd.
+	while(join(/, ($cwdComponents[-2], $cwdComponents[-1])) ne tmpdir

Bug#559009: paprefs: Only show install buttons when PackageKit is available

[Christopher James Halse Rogers]

Package: paprefs
Version: 0.9.9-2
Severity: minor
Tags: patch
User: ubuntu-de...@lists.ubuntu.com
Usertags: origin-ubuntu lucid ubuntu-patch


In Ubuntu, we've applied the attached patch to achieve the following:

  * debian/patches/0004-Show-install-only-with-active-packagekit.patch
+ Show the install buttons only when PackageKit is running.  Otherwise
  these buttons silently fail (LP: #489531).  A functionally-equivalent
  patch is in upstream git; this can be dropped in the next upstream 
  version.

We thought you might be interested in doing the same.


-- System Information:
Debian Release: squeeze/sid
  APT prefers lucid-security
  APT policy: (500, 'lucid-security'), (500, 'lucid')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.32-5-generic (SMP w/2 CPU cores)
Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968)
Shell: /bin/sh linked to /bin/dash
diff -u paprefs-0.9.9/debian/patches/series paprefs-0.9.9/debian/patches/series
--- paprefs-0.9.9/debian/patches/series
+++ paprefs-0.9.9/debian/patches/series
@@ -3,0 +4 @@
+0004-Show-install-only-with-active-packagekit.patch
only in patch2:
unchanged:
--- paprefs-0.9.9.orig/debian/patches/0004-Show-install-only-with-active-packagekit.patch
+++ paprefs-0.9.9/debian/patches/0004-Show-install-only-with-active-packagekit.patch
@@ -0,0 +1,105 @@
+Description: Only show Install buttons when PackageKit is available
+  Check for an owner of the org.freedesktop.PackageKit name on the session
+  bus at startup.  Show the install buttons iff there's an owner of that name.
+Author: Christopher James Halse Rogers r...@ubuntu.com
+Bug-Ubuntu: https://bugs.edge.launchpad.net/ubuntu/+source/paprefs/+bug/489531
+Bug: http://pulseaudio.org/ticket/732
+
+Index: paprefs-0.9.9/src/paprefs.cc
+===
+--- paprefs-0.9.9.orig/src/paprefs.cc	2009-11-30 10:36:32.263971250 +1100
 paprefs-0.9.9/src/paprefs.cc	2009-11-30 10:39:01.932708465 +1100
+@@ -30,6 +30,7 @@
+ #include gconfmm.h
+ #include libintl.h
+ #include dbus/dbus-glib.h
++#include dbus/dbus.h
+ #include gdk/gdkx.h
+ 
+ #include pulse/version.h
+@@ -108,6 +109,9 @@
+ 
+ void onGConfChange(const Glib::ustring key, const Gnome::Conf::Value value);
+ 
++void checkForPackageKit();
++bool packageKitAvailable;
++
+ void installFiles(const char *a, const char *b);
+ void installModules(const char *a, const char *b);
+ 
+@@ -153,6 +157,7 @@
+ x-get_widget(rtpNullSinkRadioButton, rtpNullSinkRadioButton);
+ 
+ checkForModules();
++checkForPackageKit();
+ 
+ gconf = Gnome::Conf::Client::get_default_client();
+ gconf-set_error_handling(Gnome::Conf::CLIENT_HANDLE_ALL);
+@@ -225,37 +230,37 @@
+ upnpMediaServerCheckButton-set_sensitive(upnpAvailable);
+ upnpNullSinkCheckButton-set_sensitive(upnpAvailable  upnpMediaServerCheckButton-get_active());
+ 
+-if (zeroconfDiscoverAvailable)
++if (zeroconfDiscoverAvailable || !packageKitAvailable)
+ zeroconfDiscoverInstallButton-hide();
+ else
+ zeroconfDiscoverInstallButton-show();
+ 
+-if (zeroconfRaopDiscoverAvailable)
++if (zeroconfRaopDiscoverAvailable || !packageKitAvailable)
+ zeroconfRaopDiscoverInstallButton-hide();
+ else
+ zeroconfRaopDiscoverInstallButton-show();
+ 
+-if (remoteAvailable)
++if (remoteAvailable || !packageKitAvailable)
+ remoteInstallButton-hide();
+ else
+ remoteInstallButton-show();
+ 
+-if (zeroconfPublishAvailable)
++if (zeroconfPublishAvailable || !packageKitAvailable)
+ zeroconfPublishInstallButton-hide();
+ else
+ zeroconfPublishInstallButton-show();
+ 
+-if (upnpAvailable)
++if (upnpAvailable || !packageKitAvailable)
+ upnpInstallButton-hide();
+ else
+ upnpInstallButton-show();
+ 
+-if (rtpRecvAvailable)
++if (rtpRecvAvailable || !packageKitAvailable)
+ rtpRecvInstallButton-hide();
+ else
+ rtpRecvInstallButton-show();
+ 
+-if (rtpSendAvailable)
++if (rtpSendAvailable || !packageKitAvailable)
+ rtpSendInstallButton-hide();
+ else
+ rtpSendInstallButton-show();
+@@ -700,6 +705,22 @@
+ g_find_program_in_path(rygel);
+ }
+ 
++void MainWindow::checkForPackageKit() {
++
++DBusError err;
++dbus_error_init(err);
++DBusConnection *sessionBus = dbus_bus_get(DBUS_BUS_SESSION, err);
++
++if(dbus_error_is_set(err)) {
++g_warning(Error connecting to DBus: %s, err.message);
++packageKitAvailable = FALSE;
++} else {
++packageKitAvailable = dbus_bus_name_has_owner(sessionBus, org.freedesktop.PackageKit, NULL);
++dbus_connection_unref(sessionBus);
++}
++dbus_error_free(err);
++}
++
+ int main(int argc, char *argv[]) {
+ 
+ /* Initialize the i18n stuff */

Bug#542351: ITP: hyena -- general purpose CLI library

[Christopher James Halse Rogers]

On Wed, 2009-08-19 at 15:48 +0800, Chow Loong Jin wrote:
 Package: wnpp
 Severity: wishlist
 Owner: Chow Loong Jin hyper...@ubuntu.com
 
 
 * Package name: hyena
   Version : 0.1
   Upstream Author : Gabriel Burt gabriel.b...@gmail.com
 * URL : http://live.gnome.org/Hyena
 * License : MIT/X11
   Programming Lang: C#
   Description : general purpose CLI library
This short description might be a bit broad?  I'd guess that CLI is
also likely to be confused with command-line-interface.  I think
utility library would be the more usual terminology.

Collection of utility libraries for the CLI runtimes?


signature.asc
Description: This is a digitally signed message part

Bug#535143: ITP: gnome-do-docklets -- Dock applets for GNOME Do's Docky interface.

[Christopher James Halse Rogers]

Package: wnpp
Severity: wishlist
Owner: Christopher James Halse Rogers r...@ubuntu.com


* Package name: gnome-do-docklets
  Version : 0.8.2
  Upstream Author : Jason Smith jassm...@gmail.com
* URL : http://do.davebsd.org/
* License : GPLv3+
  Programming Lang: C#
  Description : Dock applets for GNOME Do's Docky interface.

 Extra applets for GNOME Do's Docky interface.  These include a CPU/memory 
 monitor, a weather forcast, a battery monitor, and more.



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#524993: [pkg-cli-apps-team] Bug#524993: gnome-do-plugins: directly only suggests evolution, indirectly pulls it in

[Christopher James Halse Rogers]

As it currently stands, this can't be done in a policy-compliant way -
the Evolution plugin requires libevolution5.0-cil, and
libevolution5.0-cil links against a library only provided in the
evolution package.  We *also* can't split the gnome-do-plugins package
due to the way gnome-do uses mono.addins.

The next gnome-do upstream release will change the mono.addins usage,
making it possible to split the plugins package so plugins with long
dependency chains like evolution will be split out.


signature.asc
Description: This is a digitally signed message part

Bug#418595: Should be fixed

[Christopher James Halse Rogers]

tags 418595 needinfo
quit

This looks like a simple case of a missing dllmap config file.  The
current (0.3.8-1) package contains a config file for the appropriate
assembly, and it looks like the package has done for some time.

Can you reproduce this on a current package?  If not, I'd say this is
fixed.


signature.asc
Description: This is a digitally signed message part

Bug#516194: gnome-do: Installs 2 files in autostart

[Christopher James Halse Rogers]

fixed 516194 0.6.1.0-1
quit

This is fixed in the experimental packages.  These will be uploaded to
unstable once the mono 2.0 transition is ready.


signature.asc
Description: This is a digitally signed message part

Bug#507007: New version uploaded to Ubuntu

[Christopher James Halse Rogers]

 tag patch
kthxbye

I've just updated the version of Homebank in Ubuntu.  It's fairly
straightforward, and I've also cleaned up some lintian warnings.  I'm
attaching the debian/ component of the debdiff on the off-chance it's
helpful.
--- homebank-3.8/debian/changelog	2009-02-05 14:21:24.0 +1100
+++ homebank-4.0.2/debian/changelog	2009-02-05 14:21:24.0 +1100
@@ -1,3 +1,25 @@
+homebank (4.0.2-0ubuntu1) jaunty; urgency=low
+
+  * New upstream release
++ New QIF format import/export support
++ Snazzier charts
++ Autocompletion for Account, Payee, Categories, Description/Memo
++ Tags for transactions
+  * debian/patches/fix_HELP_DIR
++ Update for new upstream version
+  * debian/README.source
++ Add for 3.8.0 policy compliance
+  * debian/control
++ Bump standards version, adding README.source
++ Add ${misc:Depends} to homebank-data dependencies.  Silences lintian,
+  doesn't hurt, might help in future.
+  * debian/copyright:
++ Remove section about repacking the tarball to remove Generator metatag.
+  The original documentation no longer has these tags.
++ (C) - ©.  Silences lintian warning about copyright legalities.
+
+ -- Christopher James Halse Rogers r...@ubuntu.com  Wed, 04 Feb 2009 16:06:23 +1100
+
 homebank (3.8-1) unstable; urgency=low
 
   * New upstream release.
--- homebank-3.8/debian/control	2009-02-05 14:21:24.0 +1100
+++ homebank-4.0.2/debian/control	2009-02-05 14:21:24.0 +1100
@@ -1,9 +1,10 @@
 Source: homebank
 Section: gnome
 Priority: optional
-Maintainer: Francesco Namuri france...@namuri.it
+Maintainer: Ubuntu MOTU Developers ubuntu-m...@lists.ubuntu.com
+XSBC-Original-Maintainer: Francesco Namuri france...@namuri.it
 Build-Depends: debhelper (= 5.0.0), cdbs, autotools-dev, pkg-config, libgtk2.0-dev, libxml-parser-perl, patchutils, libofx-dev
-Standards-Version: 3.7.3
+Standards-Version: 3.8.0
 Vcs-Svn: svn://svn.debian.org/collab-maint/ext-maint/homebank/trunk
 Vcs-Browser: http://svn.debian.org/wsvn/collab-maint/ext-maint/homebank/trunk/
 Homepage: http://homebank.free.fr/
@@ -23,6 +24,7 @@
 
 Package: homebank-data
 Architecture: all
+Depends: ${misc:Depends}
 Recommends: homebank (= ${source:Version})
 Description: Data files for homebank 
  HomeBank is a fast, simple and easy to use program to manage your accounts.
--- homebank-3.8/debian/copyright	2009-02-05 14:21:24.0 +1100
+++ homebank-4.0.2/debian/copyright	2009-02-05 14:21:24.0 +1100
@@ -6,7 +6,7 @@
 Upstream Author: Maxime Doyen homeb...@free.fr
 
 Copyrights:
-   Copyright (C) 1995-2008 Maxime DOYEN
+   Copyright © 1995-2008 Maxime DOYEN
 
 License:
 
@@ -28,29 +28,8 @@
 Public License can be found in `/usr/share/common-licenses/GPL'.
 
 The Debian packaging is 
-(C) 2006, Adrien Cunin adri2...@ubuntu.com
-(C) 2007, Francesco Namuri france...@namuri.it
+© 2006, Adrien Cunin adri2...@ubuntu.com
+© 2007, Francesco Namuri france...@namuri.it
 and is licensed under the GPL, see above.
 
 All files under images/*.svg are under GPLv2 license.
-
-The HTML files under doc/ are no longer generated and the outdated
-metatag Generator has been removed; quoting the author:
-
-
-This is an old metatag outdated from the time I first write the doc in
-Amiga (there were first in amigaguide format and i used a tool
-to convert it to html which added this meta...)
-
-I will remove this.
-
-In waiting you can do it also and re-submit to debian, here is what
-i've done: cd homebank/doc
-find ./ -type f | xargs perl -pi -w -e 's/meta name=Generator
-content=GuideML 2.2//g'
-
-
-Following these instructions, the orig.tar.gz has been repackaged
-removing these metatags. HTML is the preferred format for updating the
-documentation upstream in a normal text editor. Patches against the
-HTML are welcome.
--- homebank-3.8/debian/patches/fix_HELP_DIR.patch	2009-02-05 14:21:24.0 +1100
+++ homebank-4.0.2/debian/patches/fix_HELP_DIR.patch	2009-02-05 14:21:24.0 +1100
@@ -1,22 +1,24 @@
-diff -Nur trunk/src/Makefile.am trunk.new/src/Makefile.am
 trunk/src/Makefile.am	2007-07-21 21:17:54.0 +0200
-+++ trunk.new/src/Makefile.am	2007-08-17 01:52:03.0 +0200
-@@ -68,5 +68,5 @@
- $(DEPS_CFLAGS) \
- -DLOCALE_DIR=\$(datadir)/locale\ \
- -DPIXMAPS_DIR=\$(datadir)/homebank/images\ \
---DHELP_DIR=\$(datadir)/homebank/help\
-+-DHELP_DIR=\/usr/share/doc/homebank-data/help\
+diff -Nur -x '*.orig' -x '*~' homebank-4.0.2/src/Makefile.am homebank-4.0.2.new/src/Makefile.am
+--- homebank-4.0.2/src/Makefile.am	2008-09-03 16:26:18.0 +1000
 homebank-4.0.2.new/src/Makefile.am	2009-02-04 16:33:09.453737686 +1100
+@@ -2,7 +2,7 @@
+ common_defines = \
+ 	-DHOMEBANK_PIXMAPSDIR=\$(datadir)/homebank/images\ \
+ 	-DHOMEBANK_LOCALEDIR=\$(datadir)/locale\ \
+-	-DHOMEBANK_HELPDIR=\$(datadir)/homebank/help\
++	-DHOMEBANK_HELPDIR=\/usr/share/doc/homebank-data/help\
  
-diff -Nur trunk/src/Makefile.in trunk.new/src/Makefile.in
 trunk/src/Makefile.in

Bug#441411: GNOME patch

[Christopher James Halse Rogers]

I've applied the patch[1] attached to comment 21 of the upstream bug to
the Ubuntu packages.  While it only works for gnome-screensaver, it
doesn't break anything else, and should be reasonably easy to add
support for other screensavers there.  I don't have much motivation to
do that work myself though, since gnome-screensaver covers almost all
modern GNOME environments just fine.

[1] http://bugzilla.pculture.org/attachment.cgi?id=1221


signature.asc
Description: This is a digitally signed message part

Bug#436201: Ping!

[Christopher James Halse Rogers]

I know this won't make Lenny, but I just want to get some feedback on
the above patch before it drops off my radar again.  Is this approach
acceptable?  Is the patch as-is likely to be acceptable for Squeeze?



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#485684: New upstream version

[Christopher James Halse Rogers]

The Do 0.5 series of releases is unsuitable for packaging; there's no
way for us to ship the 0.5 plugins, and without the plugins Do is pretty
useless.

Upstream is releasing 0.6, which resolves this issue, this week.  I
should be able to update the Debian package sometime in the next couple
of weeks.


signature.asc
Description: This is a digitally signed message part

Bug#493529: ITP: google-data-cil -- CLI libraries for access to Google services

[Christopher James Halse Rogers]

X-Debbugs-CC: Christopher James Halse Rogers [EMAIL PROTECTED]
Package: wnpp
Owner: Christopher James Halse Rogers [EMAIL PROTECTED]
Severity: wishlist

* Package name: google-data-cil
  Version : 1.2.1.0
  Upstream Author : Google
* URL or Web page : http://code.google.com/p/google-gdata/
* License : Apache 2.0
  Description : CLI libraries for access to Google services

 Provides a set of CLI libraries to access Google services which 
 export a GData interface.  These services include Blogger, Google
 Calendar, Picasa, GMail contacts, and YouTube.


signature.asc
Description: This is a digitally signed message part

Bug#489401:

[Christopher James Halse Rogers]

This is an upstream bug: https://bugs.edge.launchpad.net/do/+bug/207485.

It seems the consensus is that this is a bug in Metacity's compositor -
it doesn't appear to happen with other compositing managers, although
there's talk of a possible work-around being implemented.


signature.asc
Description: This is a digitally signed message part

Bug#418889: Nouveau packaging

[Christopher James Halse Rogers]

I'm RAOF, maintainer of the Ubuntu PPA mentioned earlier.  I've recently
changed the libdrm package in my PPA to produce a module-assistant using
drm-modules-source package, which might make inclusion in experimental a
bit easier.

Speaking with upstream, they very much do not want the 3d portion
shipped at all, anywhere.  It doesn't just need a git snapshot of mesa,
it needs a git snapshot of a branch of the experimental gallium-0.1
branch of mesa.  Accordingly, I don't ship the 3d portion in my PPA, and
it would be courteous to not ship it in experimental.

I'd be happy to help push to experimental, but I don't currently have a
bare-metal Sid install, so driver testing is a little difficult!


signature.asc
Description: This is a digitally signed message part

Bug#436201: Patch

[Christopher James Halse Rogers (RAOF)]

Tags: patch

Here's a patch, tested on i386  AMD64 sid, importing the multi-arch
magic from libasound2.

The BIARCH_TEMPLIBDIR voodoo is there for two reasons:
1) pkg-config doesn't know the difference between archs, so configure
always detects the presence of jack, pulse,  dbus, but then the build
can't find a library of the appropriate bit-ness.

2) ia32-libs doesn't have dev symlinks, so the build fails to link
against the 32bit libpulse for example.

So we detect whether a version of lib{pulse, jack, dbus-1} exists.  If
so, we add a .so symlink to it in BIARCH_TEMPLIBDIR and copy the
pkgconfig file there as well.

Thus the package has the maximum possible functionality where the libs
exist (AMD64), and the build still works where they don't (everything
else, it seems).
Index: debian/control
===
--- debian/control	(revision 2035)
+++ debian/control	(working copy)
@@ -3,7 +3,7 @@
 Priority: optional
 Maintainer: Debian ALSA Maintainers [EMAIL PROTECTED]
 Uploaders: Jordi Mallach [EMAIL PROTECTED], Elimar Riesebieter [EMAIL PROTECTED]
-Build-Depends: debhelper (= 5.0.37), quilt (= 0.40), autotools-dev, libasound2-dev (= 1.0.12), libjack-dev, libpulse-dev, libsamplerate0-dev | libsamplerate-dev, libavcodec-dev
+Build-Depends: debhelper (= 5.0.37), quilt (= 0.40), autotools-dev, libasound2-dev (= 1.0.12), lib32asound2-dev (= 1.0.12) [amd64 ppc64], lib64asound2-dev (= 1.0.12) [sparc s390 i386 powerpc], libpulse-dev, libsamplerate0-dev | libsamplerate-dev, libavcodec-dev, ia32-libs [amd64], ia64-libs [i386 powerpc sparc s390],libc6-dev-powerpc [ppc64], libc6-dev-i386 [amd64], libc6-dev-amd64 [i386], libc6-dev-ppc64 [powerpc], libc6-dev-s390x [s390], libc6-dev-sparc64 [sparc], lib32gcc1 [amd64 ppc64], lib64gcc1 [i386 powerpc sparc s390], gcc-multilib [amd64 i386 powerpc ppc64 s390 sparc], libjack-dev
 Standards-Version: 3.7.2
 Homepage: http://www.alsa-project.org/
 Vcs-Svn: svn://svn.debian.org/pkg-alsa/trunk/alsa-plugins
@@ -26,3 +26,39 @@
  .
  ALSA is the Advanced Linux Sound Architecture.
  JACK is the Jack Audio Connection Kit.
+
+Package: lib32asound2-plugins
+Architecture: amd64 ppc64
+Section: libs
+Depends: ${shlibs:Depends}, ${bilib:depends}
+Enhances: lib32asound2
+Description: ALSA library additional plugins (32 bit)
+ This package contains plugins for the ALSA library that are
+ not included in the main lib32asound2 package.
+ .
+ The ALSA library plugin jack allows the ALSA library to
+ play or capture via JACK. (This should not be confused with
+ the jackd output driver alsa, included in the jackd
+ package, which allows the JACK daemon to play back via the
+ ALSA library.)
+ .
+ ALSA is the Advanced Linux Sound Architecture.
+ JACK is the Jack Audio Connection Kit.
+
+Package: lib64asound2-plugins
+Architecture: sparc s390 i386 powerpc
+Section: libs
+Depends: ${shlibs:Depends}, ${bilib:depends}
+Enhances: lib64asound2
+Description: ALSA library additional plugins (64 bit)
+ This package contains plugins for the ALSA library that are
+ not included in the main lib64asound2 package.
+ .
+ The ALSA library plugin jack allows the ALSA library to
+ play or capture via JACK. (This should not be confused with
+ the jackd output driver alsa, included in the jackd
+ package, which allows the JACK daemon to play back via the
+ ALSA library.)
+ .
+ ALSA is the Advanced Linux Sound Architecture.
+ JACK is the Jack Audio Connection Kit.
Index: debian/changelog
===
--- debian/changelog	(revision 2035)
+++ debian/changelog	(working copy)
@@ -1,9 +1,16 @@
 alsa-plugins (1.0.15-2) UNRELEASED; urgency=low
 
+  [ Jordi Mallach ]
   * Switch to now official Vcs-* control fields.
 
- -- Jordi Mallach [EMAIL PROTECTED]  Mon, 26 Nov 2007 11:24:30 +0100
+  [ Christopher James Halse Rogers ]
+  * debian/control
+  * debian/rules:
++ Import the multi-arch magic from alsa-lib to build a lib32asound2-plugins
+  package wherever lib32asound2 is built, and similarly for lib64asound2.
 
+ -- Christopher James Halse Rogers [EMAIL PROTECTED]  Sun, 10 Feb 2008 18:59:30 +1100
+
 alsa-plugins (1.0.15-1) unstable; urgency=low
 
   * New upstream release.
Index: debian/rules
===
--- debian/rules	(revision 2035)
+++ debian/rules	(working copy)
@@ -7,6 +7,37 @@
 export DEB_HOST_GNU_TYPE  ?= $(shell dpkg-architecture -qDEB_HOST_GNU_TYPE)
 export DEB_BUILD_GNU_TYPE ?= $(shell dpkg-architecture -qDEB_BUILD_GNU_TYPE)
 
+DEB_BUILD_ARCH_CPU ?= $(shell dpkg-architecture -qDEB_BUILD_ARCH_CPU)
+DEB_HOST_ARCH  ?= $(shell dpkg-architecture -qDEB_BUILD_ARCH)
+
+biarch_map := i386=amd64 powerpc=ppc64 sparc=sparc64 s390=s390x \
+		amd64=i386 ppc64=powerpc
+biarch_cpu := $(strip $(patsubst $(DEB_BUILD_ARCH_CPU)=%, %, \
+			   $(filter $(DEB_BUILD_ARCH_CPU)=%, $(biarch_map
+
+ifneq (,$(biarch_cpu))
+  ifneq (,$(findstring /$(DEB_HOST_ARCH

Bug#462016: libxine - pulse plugin

[Christopher James Halse Rogers (RAOF)]

It looks like you're using pulseaudio, or at least have libpulse0
installed.

Does this crash happen reproducibly?  If so, this would seem to be
a(nother) manifestation of bug #427991.


signature.asc
Description: This is a digitally signed message part

Bug#443934: Progress?

[Christopher James Halse Rogers (RAOF)]

What is the progress of this ITP?  Apart from the lack of license
headers in any of the source files, it seems a simple enough packaging
task, although I note that there have been a couple of upstream releases
since you created your packages.

I can help, or take over, if you like.


signature.asc
Description: This is a digitally signed message part

Bug#461353: monodevelop: Fails to start without Gecko

[Christopher James Halse Rogers]

Package: monodevelop
Version: 0.18.1+dfsg-1
Severity: normal

Monodevelop, the program, depends on a gecko renderer being present, but
the package has no dependency on a package which provides one.  Removing
the libxul0d package doesn't remove monodevelop, but trying to run
monodevelop from a terminal results in no output, the shell doesn't
return, and monodevelop doesn't start.


signature.asc
Description: This is a digitally signed message part

Bug#420718: ITP: specto -- Unobtrusive update notification program

[Christopher James Halse Rogers]

Package: wnpp
Owner: Christopher James Halse Rogers [EMAIL PROTECTED]
Severity: wishlist

* Package name: specto
  Version : 0.2.1
  Upstream Author : Jean-Francois 'Kiddo' Fortin Tam [EMAIL PROTECTED]
* URL or Web page : http://code.google.com/p/specto
* License : GPL
  Description : Unobtrusive update notification program

 Specto is a desktop application that will watch any user-specified events
 (web, email, folder, ...). This will allow users, for instance, to specify 
 a website to watch, and Specto will automatically check for updates on the 
 web page. It will then notify the user when there is activity. This allows 
 the user to be informed of new updates/events instead of having to look 
 out for them.

This package will be based upon the specto package already in Ubuntu's Universe
repository


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Bug#415552: ITP: pulseaudio-device-chooser-gtk -- allow quick access to useful features of the PulseAudio sound server

[Christopher James Halse Rogers]

Package: wnpp
Owner: Christopher James Halse Rogers [EMAIL PROTECTED]
Severity: wishlist

* Package name: pulseaudio-device-chooser-gtk
  Version : 0.9.3
  Upstream Author : Lennart Poettering [EMAIL PROTECTED]
* URL or Web page : http://0pointer.de/lennart/projects/padevchooser/
* License : GPL
  Description : allow quick access to useful features of the PulseAudio 
sound server

 PulseAudio Device Chooser is a simple tool that provides a tray icon allowing 
 quick access to some useful features of the PulseAudio sound server.
 Specifically, it:
 .
 Provides notification of new sinks/sources available on the network
 .
 Allows you to easily change the default sink/source/server assigned to the
 current X11 display
 .
 Starts the PulseAudio Volume Control, Volume Meter, Manager and Preferences
 tools (if installed)

I also intend to package these other tools (PulseAudio Volume Control, 
Volume Meter, Manager and Preferences).


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]