Bug#1017098: decopy: may overturn the hierarchy of files w.r.t. the base debian/copyright file

[Maximiliano Curia]

On 2022-09-29 10:22, Jochen Sprickerhof wrote:

But why does decopy strip "et al."?


I looked into the git history but there is no comment why it was added.


It's part of the name normalization logic, it could be extended so if 
the author is set as "$name et all" somewhere to show the entry as such, 
even if it's listed as "$name" everywhere else. But I don't really see 
value in the added information (see bellow), thus it's just dropped.
(in retrospective, it would have been better to call that piece of code 
name normalization instead of simply applying some regexps.)


What's the copyright information that you are adding by that? Who is 
others, who would you contact if you need to relicense?  Why et all in 
french and, not any of the other languages of the world? Any of this 
information is relevant to a Debian package?



And what's the correct way to specify that a copyright is owned
by one main owner plus many other co-owners?



I'm actually not sure of the implication of this. Stating that there
is someone besides the named person holding the copyright does not
really give you any information.


:)


I mean, the correct way for decopy.



If you really want to add it, you could first off all separate it from
the first name. So in your example make Masato Taruishi and et al. two
separate entries:

2002-2020, Masato Taruishi
2002-2020, et alii (or others or..)


Whatever you add in the debian/copyright should be maintained, that's 
the original intention (unless the files mentioned are no longer 
present).


Sadly this part also is missing a normalization logic to merge with the 
found authors. I thought I would do that for the diff mode.


--
Saludos /\/\ /\ >< `/

Bug#952332: kxd: diff for NMU version 0.14-1.1

[Maximiliano Curia]

On 2020-04-05 06:38, Håvard Flaget Aasen wrote:

Control: tags 952332 + patch
Control: tags 952332 + pending

Dear maintainer,

I've prepared an NMU for kxd (versioned as 0.14-1.1) and
uploaded it to mentors. Please feel free to tell me if I
should remove it.


Thanks, could you prepare a merge request against
https://salsa.debian.org/debian/kxd ?

Thanks,
--
Saludos /\/\ /\ >< `/


signature.asc
Description: OpenPGP digital signature

Bug#942009: stgit: please make the build reproducible

[Maximiliano Curia]

On 2019-10-08 23:35, Chris Lamb wrote:

Source: stgit
Version: 0.19-1
Severity: wishlist
Tags: patch
User: reproducible-bui...@lists.alioth.debian.org
Usertags: randomness
X-Debbugs-Cc: reproducible-b...@lists.alioth.debian.org



Whilst working on the Reproducible Builds effort [0] we noticed
that stgit could not be built reproducibly.

This is because the contents of the dynamically-compiled Bash
completion script was not being generated in a deterministic manner,
as well as the cmdlist.py module.



Patch attached.


Thanks. I thought that (python3 > 3.6)'s sets also preserved their 
insertion order (as dictionaries do), apparently not. We could probably 
request that to cpython, that should fix some other set related 
reproducibility issues.


Happy hacking,
--
Saludos /\/\ /\ >< `/


signature.asc
Description: OpenPGP digital signature

Bug#932872: [libkscreenlocker5] Doesn't lock automatically by timeout

[Maximiliano Curia]

Control: severity -1 normal
Control: tags -1 + unreproducible

On 2019-07-24 01:49, Alex Volkov wrote:

Package: libkscreenlocker5
Version: 5.14.5-1
Severity: grave


Doesn't lock automatically by timeout, which presents a security issue. 
Seems
to work in a newly created test user setup, and randomly works 
sometimes in my
user, but I have no idea what makes in work. Probably some app 
"inhibits" the

automatic timeout, but I have no idea how to check it with all that
"brilliant" dbus crap.  Ubuntu has https://wiki.ubuntu.com/
DebuggingScreenLocking, but it's for Gnome, not KDE.


Hi Alex,

I'm really sorry that kscreenlocker is not working right for you, sadly 
I'm not being able to reproduce this issue, and as far as I can see it 
works as expected. But, there are a couple of catches in this, for 
example Xserver won't allow a screen lockeer to work whenever a context 
menu window or a tooltip is being shown, this is a known problem in the 
X11 world [1] [2] [3] that's only solved by migrating to wayland.


From the side of the inhibitors, I'm not sure if there is an api you can 
call to obtain the current inhibitors, but you can always try to lock 
the screen by hand and see if the screen does in fact get locked or not. 
Some known inhibitors are: google-chrome in certain situations, slide 
programs in presentation mode, and most video players.


Hope this help you figuring out where is the problem you are seeing.

[1]: https://bugs.kde.org/show_bug.cgi?id=78871
[2]: 
https://bugs.launchpad.net/ubuntu/+source/gnome-screensaver/+bug/49579
[3]: 
http://blog.martin-graesslin.com/blog/2015/01/why-screen-lockers-on-x11-cannot-be-secure/




--- System information. ---
Architecture:
Kernel:   Linux 4.19.37-bootes0-iommu-p-1000

Debian Release: 10.0
  991 stable  security.debian.org
  991 stable  ftp.fi.debian.org
  990 buster-backports ftp.debian.org
   99 stable  www.deb-multimedia.org
  500 stable  dl.google.com
  500 stable  deb.torproject.org


Happy hacking,
--
Saludos /\/\ /\ >< `/


signature.asc
Description: OpenPGP digital signature

Bug#940861: Fails to install categories files with extra-cmake-modules >= 5.59

[Maximiliano Curia]

Source: kdenlive
Version: 19.08.0-1
Severity: normal
Tags: patch

The new version of extra-cmake-modules introduces a new 
KDE_INSTALL_LOGGINGCATEGORIESDIR variable where the .categories files should 
be now be installed to. This change was adopted by many kde applications in a 
conditional installation path: when the application is built with a previous 
version of extra-cmake-modules the KDE_INSTALL_CONFDIR is used, otherwise the 
new variable is used. This causes kdenlive's debian/rules to not install the 
qtdebug categories file, as it is in a different location, and to 
inadvertently stop installing a conf file.


I've prepared an NMU for fixing this issue, which I'm attaching here and I'll 
upload to the 7 day DELAYED queue.


Happy hacking,

-- System Information:
Debian Release: bullseye/sid
 APT prefers unstable-debug
 APT policy: (500, 'unstable-debug'), (500, 'stable-debug'), (500, 'unstable'), 
(500, 'stable'), (1, 'experimental-debug'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 5.0.0-trunk-amd64 (SMP w/12 CPU cores)
Kernel taint flags: TAINT_PROPRIETARY_MODULE, TAINT_OOT_MODULE, 
TAINT_UNSIGNED_MODULE
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8), 
LANGUAGE=en_GB:en (charmap=UTF-8)
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

--
"Las computadoras son inútiles, solo pueden darte respuestas."
-- Pablo Picasso
Saludos /\/\ /\ >< `/
diff -Nru kdenlive-19.08.0/debian/changelog kdenlive-19.08.0/debian/changelog
--- kdenlive-19.08.0/debian/changelog	2019-08-16 01:48:14.0 -0700
+++ kdenlive-19.08.0/debian/changelog	2019-09-20 17:08:03.0 -0700
@@ -1,3 +1,11 @@
+kdenlive (19.08.0-1.1) unstable; urgency=medium
+
+  * Non-maintainer upload.
+  * Force extra-cmake-modules build dependency version.
+  * Move qdebug files.
+
+ -- Maximiliano Curia   Fri, 20 Sep 2019 17:08:03 -0700
+
 kdenlive (19.08.0-1) unstable; urgency=medium
 
   * New upstream release.
diff -Nru kdenlive-19.08.0/debian/control kdenlive-19.08.0/debian/control
--- kdenlive-19.08.0/debian/control	2019-08-16 01:48:14.0 -0700
+++ kdenlive-19.08.0/debian/control	2019-09-20 17:08:03.0 -0700
@@ -13,7 +13,7 @@
  librttr-dev,
  ffmpeg,
  libqt5opengl5-dev,
- extra-cmake-modules,
+ extra-cmake-modules (>= 5.59~),
  qtscript5-dev,
  qtmultimedia5-dev,
  libqt5svg5-dev,
diff -Nru kdenlive-19.08.0/debian/kdenlive-data.install kdenlive-19.08.0/debian/kdenlive-data.install
--- kdenlive-19.08.0/debian/kdenlive-data.install	2019-08-16 01:48:14.0 -0700
+++ kdenlive-19.08.0/debian/kdenlive-data.install	2019-09-20 17:07:30.0 -0700
@@ -9,3 +9,4 @@
 usr/share/config.kcfg
 usr/share/doc
 usr/share/locale
+usr/share/qlogging-categories5
diff -Nru kdenlive-19.08.0/debian/kdenlive-data.maintscript kdenlive-19.08.0/debian/kdenlive-data.maintscript
--- kdenlive-19.08.0/debian/kdenlive-data.maintscript	1969-12-31 16:00:00.0 -0800
+++ kdenlive-19.08.0/debian/kdenlive-data.maintscript	2019-09-20 17:05:50.0 -0700
@@ -0,0 +1 @@
+rm_conffile /etc/xdg/kdenlive.categories 19.08.0-1.1~


signature.asc
Description: PGP signature

Bug#940794: Please replace transitional dependency kdoctools-dev with libkf5doctools-dev

[Maximiliano Curia]

On 2019-09-19 18:10, Maximiliano Curia wrote:

We are about to drop the transitional package kdoctools-dev from the
kdoctools package, but your package depends on it for building. Please
update the Build-Depends so it uses libkf5doctools-dev instead.


Actually, I've just realized that the kio-dev also needs to be updated 
to libkf5kio-dev.


--
Happy hacking, /\/\ /\ >< `/


signature.asc
Description: OpenPGP digital signature

Bug#940794: Please replace transitional dependency kdoctools-dev with libkf5doctools-dev

[Maximiliano Curia]

Source: kbibtex
Version: 0.8.1-1
Severity: normal

Hi,

We are about to drop the transitional package kdoctools-dev from the kdoctools 
package, but your package depends on it for building. Please update the 
Build-Depends so it uses libkf5doctools-dev instead.


Happy hacking,

-- System Information:
Debian Release: bullseye/sid
 APT prefers unstable-debug
 APT policy: (500, 'unstable-debug'), (500, 'stable-debug'), (500, 'unstable'), 
(500, 'stable'), (1, 'experimental-debug'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 5.0.0-trunk-amd64 (SMP w/12 CPU cores)
Kernel taint flags: TAINT_PROPRIETARY_MODULE, TAINT_OOT_MODULE, 
TAINT_UNSIGNED_MODULE
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8), 
LANGUAGE=en_GB:en (charmap=UTF-8)
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

--
"Any change looks terrible at first." -- Principle of Design Inertia
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#940793: Please replace transitional dependency polkit-kde-1 with polkit-kde-agent-1

[Maximiliano Curia]

Source: apper
Version: 1.0.0-2
Severity: wishlist

Hi,

The next version of the polkit-kde-agent-1 package will drop the transitional 
polkit-kde-1 package. As the transitional package has been around for quite a 
while now. Please update apper to depend on polkit-kde-agent-1.


Happy hacking,

-- System Information:
Debian Release: bullseye/sid
 APT prefers unstable-debug
 APT policy: (500, 'unstable-debug'), (500, 'stable-debug'), (500, 'unstable'), 
(500, 'stable'), (1, 'experimental-debug'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 5.0.0-trunk-amd64 (SMP w/12 CPU cores)
Kernel taint flags: TAINT_PROPRIETARY_MODULE, TAINT_OOT_MODULE, 
TAINT_UNSIGNED_MODULE
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8), 
LANGUAGE=en_GB:en (charmap=UTF-8)
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

--
"La duración de un minuto depende de que lado del baño estés."
-- Ley de la Relatividad (Burke)
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#939400: sddm: Auth: sddm-helper exited with 11

[Maximiliano Curia]

Control: severity -1 normal

¡Hola Konomi!

El 2019-09-05 a las 01:14 +1000, Konomi escribió:

Package: sddm
Version: 0.18.0-1
Severity: grave
Tags: upstream
Justification: renders package unusable



Quoting the upstream bug report:



"sddm-greeter needs execution rights at 
/var/lib/sddm/.cache/sddm-greeter/qmlcache/*.qmlc.
krunner needs execution at /home/user/.cache/krunner/qmlcache/*.qmlc. It breaks
when /var and /home partitions are mounted with noexec flag which is often used 
on
hardened setups or Trusted Path Execution available on hardened linux kernels.



As for now it could be fixed by disabling qml cache by setting 
QML_DISABLE_DISK_CACHE=1
environment variable. Is there a chance that qt will check permission rights 
first then
quietly disable qmlcache per app instead of crashing?



Alternatively it could create cachedir somewhere in /usr/lib/qt same as python3 
have
it's pycache dir."



Upstream bug report: https://bugreports.qt.io/browse/QTBUG-58508



The work around of adding QML_DISABLE_DISK_CACHE=1 to /etc/enviroment got sddm 
working
for me again. Without the work around sddm will start but show just a black 
screen, so
I've marked this bug as grave since it completely breaks the package.


I don't see any reference that a program needs to support a hardened setup in 
Debian, and this is clearly something not supported upstream, as such I'm 
degrading the severity of this issue.


Let's see what do the qt maintainers think about setting a different cache dir 
or patching qt to disable the disk cache if the cache won't be loadable.


Happy hacking,
--
"It's a well-known fact that computing devices such as the abacus were 
invented thousands of years ago. But it's not well known that the first use of 
common computer protocol occurred in the Old Testament. This, of course, was 
when Moses aborted the Egyptians' process with a control-sea."

-- Tom Galloway (recs.arts.comics, February 1992)
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#935477: python-xapp: diff for NMU version 1.2.0-2.1

[Maximiliano Curia]

¡Hola Sandro!

El 2019-08-22 a las 21:21 -0400, Sandro Tosi escribió:

Package: python-xapp
Version: 1.2.0-2
Severity: normal
Tags: patch  pending



I've prepared an NMU for python-xapp (versioned as 1.2.0-2.1) and
uploaded it to DELAYED/10. Please feel free to tell me if I
should delay it longer.



This drop the python 2 package (rdep of python-psutil)


Ok, I merged your changes, go ahead a skip the delayed queue, thanks!

Happy hacking,
--
"Politicians and diapers have one thing in common. They should both be changed 
regularly, and for the same reason."

-- José Maria de Eça de Queiroz
Saludos /\/\ /\ >< `/



signature.asc
Description: PGP signature

Bug#929341: cinnamon-settings-daemon mis-built without wacom support

[Maximiliano Curia]

¡Hola Fabio!

Resending to the bug report.

El 2019-08-22 a las 13:19 +0200, Fabio Fantoni escribió:

@Maxy: why you wrote in description that wacom plugin is unbuildable? archs
where package is missed was already excluded by the build-dep you removed


I don't remember, I assume it wasn't buildable at the time.

Happy hacking,
--
"If the code and the comments disagree, then both are probable wrong."
-- Norm Schryer
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#931483: konsole: Konsole will not launch from mate-panel or from KDE Application Launcher.

[Maximiliano Curia]

Control: severity -1 important
Control: tag -1 moreinfo unreproducible

¡Hola David!

El 2019-07-05 a las 21:19 -0400, David J. Ring escribió:

Package: konsole
Version: 4:18.04.0-1
Severity: grave
Justification: renders package unusable



Dear Maintainer,



konsole will not launch from mate-panel shortcut or from KDE Application menu.



konsole WILL launch from mate-terminal and will subsequently open both new 
windows and new tabs.



konsole is unusable because of the problem, but I do not know if the problem is 
konsole or the KDE launcher or mate-panel.


I'm not able to reproduce the issue as reported, can you check running konsole 
from a different terminal and see what's kind of errors are you getting (it 
might also be useful to check your ~/.xsession-errors)?


Happy hacking,
--
A computer scientist is someone who, when told to "Go to Hell," 
sees the "go to," rather than the destination, as harmful.

Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#904083: Regression run_lintian calls lintian with a non existing filename

[Maximiliano Curia]

¡Hola Johannes!

El 2019-01-04 a las 19:35 +0100, Johannes Schauer escribió:

Quoting Maximiliano Curia (2019-01-04 19:01:52)

On Thu, 19 Jul 2018 13:08:02 +0200 Maximiliano Curia 
wrote:

With the new release of sbuild 0.77 (which moves the run_lintian call outside
of the build function), lintian receives the name of a no longer existing
changes file (as the unlink is part of the copy_changes) and as such it fails
with, for example:
 I: schroot -d /<> -c 
unstable-amd64-sbuild-a8e6ff65-a882-41b6-9ce3-b5e764b4ca8c --run-session -q -u maxy -p -- 
lintian -I --pedantic --show-overrides sddm_0.18.0-1_amd64.changes
 D: Running command: schroot -d /<> -c 
unstable-amd64-sbuild-a8e6ff65-a882-41b6-9ce3-b5e764b4ca8c --run-session -q -u maxy -p -- 
lintian -I --pedantic --show-overrides sddm_0.18.0-1_amd64.changes
 warning: "sddm_0.18.0-1_amd64.changes" cannot be processed.
 warning: It is not a valid lab query and it is not an existing file.



 E: Lintian run failed (runtime error)



The issue is not present using the 0.76 version, and can be workaround
commenting the unlink call from the copy_changes function (as shown in the
attached file).



Thanks for working in sbuild. :)



I really fail to reproduce your problem. Can you give me your exact sbuild
invocation so that I may be able to see the same error that you see? I see it
neither in the version currently in unstable nor in git master.



I'm currently using sbuild 0.77.1-2, calling it with:
 sbuild --dist=sid --arch=amd64 --chroot=sid-amd64-sbuild --arch-all --arch-any --source 
--run-lintian --lintian-opts="-I --pedantic --show-overrides" --no-run-piuparts 
--purge=never /home/maxy/debian/kde/frameworks/build-area/breeze-icons_5.53.0-1~.dsc



An it fails as long as the mentioned unlink is present.



This is really odd.



I tried your command and it works just fine over here.



How did you set up your chroot?



Can you try to set everything up inside a virtual machine or some such? Just to
have a fresh start and a step-by-step instruction of how to get to your error?


Apparently this issue is reproducible only when the host machine (the machine 
running sbuild) has a symlink from /var/lib/sbuild/build to /build.


I've created this symlink long ago in my previous machine, and today, after 
creating the symlink in my new machine the lintian errors started showing up. 
Deleting the symlink makes the builds work as expected.


I guess the unlink is being ignored on machines that don't have a /build, but 
its likely to be bogus non the less.


Happy hacking,
--
"Get your data structures correct first, and the rest of the program will 
write itself"

-- David Jones
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#895901: backup-manager: diff for NMU version 0.7.14-1.1

[Maximiliano Curia]

¡Hola Salvatore!

El 2019-07-29 a las 17:07 +0200, Salvatore Bonaccorso escribió:

El 2019-07-29 a las 08:57 +0200, Salvatore Bonaccorso escribió:

Control: tags 895901 + pending



I've prepared an NMU for backup-manager (versioned as 0.7.14-1.1) and
uploaded it to DELAYED/15. Please feel free to tell me if I should
delay it longer.



I usually wait for upstream to apply the pull request before uploading a
modified version, but I understand that if you are affected by this issue
you would want this fixed. And the patch seems trivial enough, if possible,
please confirm that you tested the fix in the upstream issue/pr.



Yes I can confirm that this fixes the issue on an affected system
running buster. Actually after the fix goes into unstable I would like
to propose this minimal fix as well for the first point release for
buster.


Thanks! Now, I meant the confirmation to be sent to the upstream pr as a 
gentle reminder that the issue is still affecting users, and that the proposed 
fix works.



Happy to hand it back to you if you prefer, this is as well the reason
for the long delayed queue, I do not want to step on others feets :)


No, it's fine, my feets are already covered by debian stuff waiting for me. :)

Feel free to upload this directly, and to upload this fix as an stable 
update.



Btw, there is as well a new usptream version 0.7.15 which incorporates
this dropping of the export of "export BM_BMP_PATH=$bmp" but that
release looks somehow strange. It fixes the above, makes some other
changes but seem to make it on top of 0.7.13?



I guess upstream intended to release 6cf5bb1 as 0.7.15 and somehow released
the 0.7.13. :/



Ah I see, a bit confusing :)


Thanks for your work. I already merge the nmudiff by the way.

Happy hacking,
--
"Seek simplicity, and distrust it." -- Whitehead's Rule
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#895901: backup-manager: diff for NMU version 0.7.14-1.1

[Maximiliano Curia]

¡Hola Salvatore!

El 2019-07-29 a las 08:57 +0200, Salvatore Bonaccorso escribió:

Control: tags 895901 + pending



I've prepared an NMU for backup-manager (versioned as 0.7.14-1.1) and
uploaded it to DELAYED/15. Please feel free to tell me if I should
delay it longer.


I usually wait for upstream to apply the pull request before 
uploading a modified version, but I understand that if you are affected by 
this issue you would want this fixed. And the patch seems trivial enough, 
if possible, please confirm that you tested the fix in the upstream issue/pr.



Btw, there is as well a new usptream version 0.7.15 which incorporates
this dropping of the export of "export BM_BMP_PATH=$bmp" but that
release looks somehow strange. It fixes the above, makes some other
changes but seem to make it on top of 0.7.13?


I guess upstream intended to release 6cf5bb1 as 0.7.15 and somehow released 
the 0.7.13. :/


Happy hacking,
--
"Any change looks terrible at first." -- Principle of Design Inertia
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#924536: unblock: decopy/0.2.4.1-2

[Maximiliano Curia]

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

Please unblock package decopy

decopy pkg uses ronn to generate its manpage, and it seems that ronn changed 
the name of its output file. This change caused an FTBFS in decopy (#924272), 
which is fixed/worked around in 0.2.4.1-2.


unblock decopy/0.2.4.1-2

-- System Information:
Debian Release: buster/sid
 APT prefers testing-debug
 APT policy: (700, 'testing-debug'), (700, 'testing'), (600, 'stable-updates'), 
(600, 'stable-debug'), (600, 'proposed-updates'), (600, 'stable'), (500, 
'buildd-unstable'), (50, 'unstable-debug'), (50, 'unstable'), (1, 
'experimental-debug'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.19.0-2-amd64 (SMP w/4 CPU cores)
Kernel taint flags: TAINT_USER, TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8), 
LANGUAGE=en_GB:en (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

--
"There are two ways of constructing a software design.  One way is to make it 
so simple that there are obviously no deficiencies. And the other way is to 
make it so complicated that there are no obvious deficiencies."

-- C.A.R. Hoare
Saludos /\/\ /\ >< `/
diff -Nru decopy-0.2.4.1/debian/changelog decopy-0.2.4.1/debian/changelog
--- decopy-0.2.4.1/debian/changelog 2018-07-22 11:00:46.0 +0200
+++ decopy-0.2.4.1/debian/changelog 2019-03-13 09:20:54.0 +0100
@@ -1,3 +1,10 @@
+decopy (0.2.4.1-2) unstable; urgency=medium
+
+  [ Vincent Blut ]
+  * Fix FTBFS due to incorrect filename (Closes: 924272)
+
+ -- Maximiliano Curia   Wed, 13 Mar 2019 09:20:54 +0100
+
 decopy (0.2.4.1-1) unstable; urgency=medium
 
   * New upstream release (0.2.4.1)
diff -Nru decopy-0.2.4.1/debian/rules decopy-0.2.4.1/debian/rules
--- decopy-0.2.4.1/debian/rules 2018-07-22 11:00:46.0 +0200
+++ decopy-0.2.4.1/debian/rules 2019-03-13 09:20:54.0 +0100
@@ -8,8 +8,8 @@
 override_dh_auto_build:
dh_auto_build
ronn README.md
-   mv README.1 decopy.1
-   mv README.1.html decopy.1.html
+   mv README*.1 decopy.1
+   mv README*.1.html decopy.1.html
 
 override_dh_auto_test:
LC_ALL=C.UTF-8 dh_auto_test


signature.asc
Description: PGP signature

Bug#924455: unblock: cjs/3.8.0-5

[Maximiliano Curia]

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

Please unblock package cjs

The cjs version currently in testing fails to build from source due to a 
couple of external symbol changes (#923903), this was fixed in unstable with 
the version 3.8.0-5.


unblock cjs/3.8.0-5

-- System Information:
Debian Release: buster/sid
 APT prefers testing-debug
 APT policy: (700, 'testing-debug'), (700, 'testing'), (600, 'stable-updates'), 
(600, 'stable-debug'), (600, 'proposed-updates'), (600, 'stable'), (500, 
'buildd-unstable'), (50, 'unstable-debug'), (50, 'unstable'), (1, 
'experimental-debug'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.19.0-2-amd64 (SMP w/4 CPU cores)
Kernel taint flags: TAINT_USER, TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8), 
LANGUAGE=en_GB:en (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

--
"Nothing ever goes away." -- Commoner's Law of Ecology
Saludos /\/\ /\ >< `/
diff -Nru cjs-3.8.0/debian/changelog cjs-3.8.0/debian/changelog
--- cjs-3.8.0/debian/changelog  2018-08-07 14:51:35.0 +0200
+++ cjs-3.8.0/debian/changelog  2019-03-11 22:35:50.0 +0100
@@ -1,3 +1,14 @@
+cjs (3.8.0-5) unstable; urgency=medium
+
+  [ Maximiliano Curia ]
+  * Update symbols files.
+Thanks to Santiago Vila for the report (Closes: 923903)
+
+  [ Jeremy Bicha ]
+  * debian/libcjs0.symbols: Mark a symbol as optional that disappears on 
ppc64el with -O3
+
+ -- Maximiliano Curia   Mon, 11 Mar 2019 22:35:50 +0100
+
 cjs (3.8.0-4) unstable; urgency=medium
 
   [ Maximiliano Curia ]
diff -Nru cjs-3.8.0/debian/libcjs0.symbols cjs-3.8.0/debian/libcjs0.symbols
--- cjs-3.8.0/debian/libcjs0.symbols2018-08-07 14:51:35.0 +0200
+++ cjs-3.8.0/debian/libcjs0.symbols2019-03-11 22:35:50.0 +0100
@@ -18,7 +18,7 @@
  _Z27gjs_cairo_pattern_get_protoP9JSContext@Base 3.4.4
  
_Z27gjs_cairo_surface_constructP9JSContextN2JS6HandleIP8JSObjectEEP14_cairo_surface@Base
 3.4.4
  _Z27gjs_cairo_surface_get_protoP9JSContext@Base 3.4.4
- 
_Z27gjs_object_require_propertyIPN2JS6RootedIP8JSObjectEEEbP9JSContextNS0_6HandleIS3_EEPKc14GjsConstStringT_@Base
 3.4.4
+ 
(optional=gccO3ppc64el)_Z27gjs_object_require_propertyIPN2JS6RootedIP8JSObjectEEEbP9JSContextNS0_6HandleIS3_EEPKc14GjsConstStringT_@Base
 3.4.4
  
_Z27gjs_object_require_propertyP9JSContextN2JS6HandleIP8JSObjectEEPKcNS2_I4jsidEENS1_13MutableHandleINS1_5ValueEEE@Base
 3.4.4
  
_Z27gjs_object_require_propertyP9JSContextN2JS6HandleIP8JSObjectEEPKcNS2_I4jsidEENS1_13MutableHandleIS4_EE@Base
 3.4.4
  
_Z27gjs_object_require_propertyP9JSContextN2JS6HandleIP8JSObjectEEPKcNS2_I4jsidEEP13GjsAutoJSChar@Base
 3.8.0
@@ -114,7 +114,7 @@
  _ZN9GjsModule8finalizeEP8JSFreeOpP8JSObject@Base 3.8.0
  _ZN9GjsModule9class_opsE@Base 3.8.0
  
(optional=templinst|arch=amd64)_ZNSt10_HashtableImSt4pairIKm11GjsAutoCharESaIS3_ENSt8__detail10_Select1stESt8equal_toImESt4hashImENS5_18_Mod_range_hashingENS5_20_Default_ranged_hashENS5_20_Prime_rehash_policyENS5_17_Hashtable_traitsILb0ELb0ELb121_M_insert_unique_nodeEmmPNS5_10_Hash_nodeIS3_Lb0EEEm@Base
 3.8.0
- 
(optional=templinst|arch=!amd64)_ZNSt10_HashtableImSt4pairIKm11GjsAutoCharESaIS3_ENSt8__detail10_Select1stESt8equal_toImESt4hashImENS5_18_Mod_range_hashingENS5_20_Default_ranged_hashENS5_20_Prime_rehash_policyENS5_17_Hashtable_traitsILb0ELb0ELb19_M_rehashEmRS1_@Base
 3.8.0
+ 
(optional=templinst)_ZNSt10_HashtableImSt4pairIKm11GjsAutoCharESaIS3_ENSt8__detail10_Select1stESt8equal_toImESt4hashImENS5_18_Mod_range_hashingENS5_20_Default_ranged_hashENS5_20_Prime_rehash_policyENS5_17_Hashtable_traitsILb0ELb0ELb19_M_rehashEmRS1_@Base
 3.8.0
  
(optional=templinst|arch=!amd64)_ZNSt10_HashtableIySt4pairIKy11GjsAutoCharESaIS3_ENSt8__detail10_Select1stESt8equal_toIyESt4hashIyENS5_18_Mod_range_hashingENS5_20_Default_ranged_hashENS5_20_Prime_rehash_policyENS5_17_Hashtable_traitsILb0ELb0ELb19_M_rehashEjRKj@Base
 3.8.0
  
(optional=templinst|subst|arch=!amd64)_ZNSt10_HashtableI{uint64_t}St4pairIK{uint64_t}11GjsAutoCharESaIS3_ENSt8__detail10_Select1stESt8equal_toI{uint64_t}ESt4hashI{uint64_t}ENS5_18_Mod_range_hashingENS5_20_Default_ranged_hashENS5_20_Prime_rehash_policyENS5_17_Hashtable_traitsILb0ELb0ELb121_M_insert_unique_nodeE{size_t}{size_t}PNS5_10_Hash_nodeIS3_Lb0EEE@Base
 3.8.0
  
(optional=templinst|subst)_ZNSt10_HashtableI{uint64_t}St4pairIK{uint64_t}11GjsAutoCharESaIS3_ENSt8__detail10_Select1stESt8equal_toI{uint64_t}ESt4hashI{uint64_t}ENS5_18_Mod_range_hashingENS5_20_Default_ranged_hashENS5_20_Prime_rehash_policyENS5_17_Hashtable_traitsILb0ELb0ELb15clearEv@Base
 3.8.0
@@ -124,8 +124,10 @@
  _ZNSt3setImSt4lessImESaImEED1Ev@Base 3.4.4
  _ZNSt3setImSt4lessImESaImEED2Ev@Base 3.4.4
  
_ZNSt5dequeIN11ToggleQueue4ItemESaIS1_EE8_M_eraseESt15_Deque_iteratorIS1_RS1_P

Bug#823037: ITP: flatbuffers -- efficient cross platform serialization library

[Maximiliano Curia]

¡Hola Bálint!

El 2019-03-04 a las 00:23 +0100, Bálint Réczey escribió:

With the following changes IMO the package would be ready for upload:
https://salsa.debian.org/debian/flatbuffers/merge_requests/1


Currently the salsa ci rules test the build against reprotest, and it fails. 
The log mentions that the issue might be due to the use of __DATE__ but that 
gets set by the SOURCE_DATE_EPOCH [1] (using the time and date from the last 
debian/changelog entry).


diffoscope shows the presence of the build paths in the generated binaries, 
I'm not sure how to avoid that.


If you know how to fix this, or you are interested in fixing this, 
please go ahead and make flatbuffers reproducible, otherwise, please disable 
the reprotest in the salsa ci pipeline.


Happy hacking,
[1]: https://wiki.debian.org/ReproducibleBuilds/TimestampsFromCPPMacros
--
"EIEIO Go home and have a glass of warm, dairy-fresh milk"
-- The GNU C Library Reference Manual, Chapter 2.2, Error Codes
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#923056: systemsettings: diff for NMU version 4:5.14.5-1.1

[Maximiliano Curia]

¡Hola Boyuan!

El 2019-02-23 a las 13:23 -0500, Boyuan Yang escribió:

Package: systemsettings
Version: 4:5.14.5-1
Severity: normal
Tags: patch  pending



I've prepared an NMU for systemsettings (versioned as 4:5.14.5-1.1) and
uploaded it to DELAYED/3. Please feel free to tell me if I
should delay it longer.



On behalf of Debian Chinese Team and KDE translation team for zh_CN
(Simplified Chinese), I'm proposing a patch to fix a zh_CN translation error
within systemsettings.desktop. Nothing else is touched.



This issue is fixed upstream with the release of Plasma 5.15. Please drop this
patch after we package Plasma 5.15 into Debian.


Thanks for your work, sadly the attached patch got mangled in the email 
transmission, could please send it as an attachment? Or better yet, could you 
create a new MR in the salsa project [1]?


[1]: https://salsa.debian.org/qt-kde-team/kde/systemsettings

Happy hacking,
--
Porque no respeta el orden natural en el que se leen las cosas

¿Por qué contestar al principio del mensaje es malo?

Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#922502: plasma-desktop: regional settings allow do select system incompatible locales

[Maximiliano Curia]

Control: reassign -1 libqt5core5a/5.11.3+dfsg-2
Control: affects -1 plasma-desktop

Control: severity -1 important

Please, don't abuse the bugs severity just to get more attention.

El 2019-02-17 a las 10:06 +0100, Charlemagne Lasse escribió:

Package: plasma-desktop
Version: 4:5.14.5-1
Severity: critical
Justification: makes unrelated software break on the system



The "regional settings" allow to select various regions which are not
available on the system (even with locales-all). An example here is
en_DE (Germany) for "Time". This is then exported at the next login in
the env variable LC_TIME as "en_DE.UTF-8". This is not supported on
any Debian buster installation and is causing other software to break.


This is a known issue (see https://bugs.debian.org/845788), plasma desktop 
here simply shows the result of calling:

QList allLocales = QLocale::matchingLocales(QLocale::AnyLanguage, 
QLocale::AnyScript, QLocale::AnyCountry);

(See kcms/formats/kcmformats.cpp in plasma-desktop) supplied by qt, which 
contains a rather large set of locales definitions, some of which don't have a 
libc's locales equivalent.


It would be nice to either limit qt's locales to the ones available to libc 
somehow, this might be done patching qtlocale at build time, or patching the 
kcmformats to reduce the list to the ones matching the currently generated 
locales.


Patches to either approach would be welcome.

Happy hacking,
--
"If you are in a hole, stop digging." -- The First Rule of Excavation
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#922055: sddm: SDDM is not conversing properly with pam and poldi

[Maximiliano Curia]

Control: forwarded -1 https://github.com/sddm/sddm/issues/693

¡Hola Eric!

El 2019-02-11 a las 10:55 -0500, Eric Taylor escribió:

Package: sddm
Version: 0.18.0-1
Severity: normal



SDDM doesn't seem to fully support PAM conversations. When attempting to use
the poldi pam module for OpenpPGP cards SDDM seems to deadlock when when poldi
sends/request more information during the login process. I have gotten pam and 
poldi to
successfully work with the gdm and slim desktop managers.



I found a bug for this on github: https://github.com/sddm/sddm/issues/693



and an open pull request that seems to fix this: 
https://github.com/sddm/sddm/pull/776


I'm sorry to hear that sddm doesn't work correctly with your pam setup. Thanks 
for reporting the issue so we are aware of it and making the issue more 
visible to the Debian users. Sadly the set of patches (13 commits and 53 files 
modified) is way beyond the scope of what we can maintain from the Debian 
side, as such this issue needs to be fixed and released upstream in order to 
address it in Debian. Please, consider updating and reviewing the patches 
upstream so the fix gets accepted.


Happy hacking,
--
"There are two ways of constructing a software design.  One way is to make it 
so simple that there are obviously no deficiencies. And the other way is to 
make it so complicated that there are no obvious deficiencies."

-- C.A.R. Hoare
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#823037: ITP: flatbuffers -- efficient cross platform serialization library

[Maximiliano Curia]

¡Hola Bálint!

El 2019-02-10 a las 20:00 +0700, Bálint Réczey escribió:

On Wed, 30 May 2018 10:25:01 +0900 Nobuhiro Iwamatsu <[1]iwama...@debian.org>
wrote:

2018-05-29 17:44 GMT+09:00 Maximiliano Curia <[2]m...@gnuservers.com.ar>:

¡Hola!



El 2018-05-29 a las 09:56 +0200, Maximiliano Curia escribió:



I created a packaging salsa project here:
[3]https://salsa.debian.org/debian/flatbuffers




Ups, that was a lie, the project was created by Nobuhiro Iwamatsu. I was
checking the flatbuffers, sink and kube code base last week and I today,
when I found that an empty project was already present, I tought I created
the repository and forgot about it.



Nobushiro, sorry to meddle with this package. Please let me know if you are
willing to group maintain this package, if not, please take the ownership

of

this bug.



No problem.



I am planning to packaging apache/arrow.
This package depends flatbuffers, if I can, I want to maintain this package.
Coulld you add me to flatbuffers team ?



Kodi 18 needs flatbuffers, too.
I see there is recent activity in the Salsa repo and I'd love to see an upload
to experimental at least and I'm ready to give a helping hand if something is
missing from the packaging.


Any help is welcome, please send merge requests as you see fit. Currently, I 
don't have any real piece of software that uses flatbuffers, so I can't really 
test it. Care to give it a try using the packages generated from the salsa 
repo to build kodi 18?


Happy hacking,
--
"The best way to predict the future is to invent it."
-- Alan Kay
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#895901: backup-manager: fails to purge old remote archives via ftp and ssh

[Maximiliano Curia]

¡Hola Thomas!

El 2018-04-17 a las 14:13 +0200, Thomas Parmelan escribió:

Package: backup-manager
Version: 0.7.14-1
Severity: normal
Tags: patch



Hi there,



I recently noticed that my remote archives were not purged anymore (I am
using FTP but the problem would be the same for SSH). This seems to have
started when I upgraded from 0.7.12-4 to 0.7.14-1.



After investigating, I found the problem lies in the fact that
/usr/sbin/backup-manager exports
BM_BMP_PATH=/usr/bin/backup-manager-purge, whereas backup-manager-upload
tries to run $ENV{BM_BMP_PATH}/backup-manager-purge, thus trying to run
/usr/bin/backup-manager-purge/backup-manager-purge, which does not
exist.



My fix was to simply delete the line "export BM_BMP_PATH=$bmp" (line 37)
in /usr/sbin/backup-manager.



For what it's worth, this line was introduced in a commit related to
upstream issue #91 and I reported it there too:
https://github.com/sukria/Backup-Manager/issues/91#issuecomment-381925277


It seems that the upstream issue report is stale, could you please create a 
pull request to fix the issue upstream?


Happy hacking,
--
"If you have too many special cases, you are doing it wrong." -- Craig Zarouni
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#918510: kwin-wayland incorrectly depends on kwayland-integration

[Maximiliano Curia]

¡Hola Martin!

El 2019-01-06 a las 21:06 +0100, Martin Graesslin escribió:

Package: kwin-wayland
Version: 4:5.13.5-1+b1
Severity: normal


kwin-wayland incorrectly depends on kwayland-integration. KWin neither at 
build nor at run time uses anyting from kwayland-integration.


This dependency was added to have the KF5WindowSystemKWaylandPlugin available 
at runtime. Is this causing a problem?


Happy hacking,
--
"Any change looks terrible at first." -- Principle of Design Inertia
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#907297: breeze-gtk-theme: Breaks xfce4-notes (invalid string constant in "/usr/share/themes/Breeze/gtk-2.0/widgets/styles")

[Maximiliano Curia]

¡Hola Dmitry!

El 2019-01-31 a las 09:50 +1100, Dmitry Smirnov escribió:

On Thursday, 31 January 2019 8:19:38 AM AEDT Maximiliano Curia wrote:

I couldn't reproduce the issue, executing xfce4-notes shows no errors in
the termianl and it seems to be working fine, this of course, while using
the breeze theme.



Interestingly enough this issue is very persistent and it existed for a long
time. Yesterday I've opened #920876 only to realize that I've already
reported that as #907297 earlier...



Is there anything special about your setup that could
somehow caused that the other files in
/usr/share/themes/Breeze/gtk-2.0/widgets/ directory could not be loaded? A
limit on the number of open files, perhaps?



In both cases I had to uninstall "breeze-gtk-theme".


Are you suggesting that even without the breeze theme active you are seeing 
the mentioned errors?


That would suggest that something is forcibly loading the breeze style file 
but not the other files in widgets.


Anyway, please check your ~/.gtkrc-2.0 (and also ~/.config/gtkrc and 
~/.config/gtkrc-2.0). While using cinnamon, these files shouldn't be needed.


Please also test this issue with a new user.


Do you think Breaks/Conflicts could be declared between "breeze-gtk-theme"
and "xfce4-notes"?


In general, I don't think that's a good idea. In particular, the issue is not 
even reproducible by us, so a change in the breeze-gtk package is not 
acceptable.


Happy hacking,
--
"There are only two things wrong with C++: The initial concept and the 
implementation."

-- Bertrand Meyer
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#907297: breeze-gtk-theme: Breaks xfce4-notes (invalid string constant in "/usr/share/themes/Breeze/gtk-2.0/widgets/styles")

[Maximiliano Curia]

Control: severity -1 normal
Control: tag -1 + unreproducible

¡Hola Dmitry!

El 2018-08-26 a las 15:53 +1000, Dmitry Smirnov escribió:

Package: breeze-gtk-theme
Version: 5.13.4-1
Severity: normal
Control: affects -1 xfce4-notes



breeze-gtk-theme breaks xfce4-notes:




$ xfce4-notes
/usr/share/themes/Breeze/gtk-2.0/widgets/styles:7: error: invalid string constant 
"notebook", expected valid string constant
/usr/share/themes/Breeze/gtk-2.0/widgets/styles:2: error: invalid string constant 
"scrollbar", expected valid string constant
/usr/share/themes/Breeze/gtk-2.0/widgets/styles:4: error: invalid string constant 
"entry", expected valid string constant
/usr/share/themes/Breeze/gtk-2.0/widgets/styles:4: error: invalid string constant 
"entry", expected valid string constant
/usr/share/themes/Breeze/gtk-2.0/widgets/styles:1: error: invalid string constant 
"default", expected valid string constant
/usr/share/themes/Breeze/gtk-2.0/widgets/styles:1: error: invalid string constant 
"default", expected valid string constant
/usr/share/themes/Breeze/gtk-2.0/widgets/styles:8: error: invalid string constant 
"range", expected valid string constant
^C




I could only use xfce4-notes after uninstalling "breeze-gtk-theme"...


I couldn't reproduce the issue, executing xfce4-notes shows no errors in the 
termianl and it seems to be working fine, this of course, while using the 
breeze theme.


Thus setting the severity back to normal, and tagging the issue as 
unreproducible. Is there anything special about your setup that could somehow 
caused that the other files in /usr/share/themes/Breeze/gtk-2.0/widgets/ 
directory could not be loaded? A limit on the number of open files, perhaps?


Happy hacking,
--
"Brilliant opportunities are cleverly disguised as insolvable problems."
-- Gardener's Philosophy

"The reverse is also true." -- Corollary
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#914541: [libpam-modules-bin] unix_chkpwd should be SUID instead of SGID, otherwise kscreen_locker does not work

[Maximiliano Curia]

Control: tag -1 + unreproducible

¡Hola Steve!

Please whenever you reassign a bug CC the receiving maintainers so we get an 
useful mail about it, thanks.


El 2019-01-23 a las 08:38 -0800, Steve Langasek escribió:

Control: reassign -1 libkscreenlocker5



That said, the problem could also be in the code of the screen locker.



It could; so let's reassign this bug to the KDE package, whose maintainers
may have more insight into specific configurations that may cause this
screenlocker to fail.


hg42: I haven't heard of a pam issue when using kscreenlocker, thus I'm tagging 
this bug as non reproducible, you might want to test this again with a new 
user or in a new installation without any specific configuration to find the 
real root of the problem you are seeing. To do that I would recommend you 
using a virtual machine with a vanilla Debian Kde installation.


Some other things to consider: are you using selinux rules that might be 
taking effect? any custom apparmor rule? Does your filesystem tweak the file 
permissions? Are you using a remote filesystem?


Happy hacking,
--
"There are two major products that come out of Berkeley: LSD and BSD. 
We don't believe this to be a coincidence."

-- Jeremy S. Anderson
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#904083: Regression run_lintian calls lintian with a non existing filename

[Maximiliano Curia]

¡Hola Johannes!

El 2019-01-04 a las 15:34 +0100, Johannes Schauer escribió:

Control: tag -1 + moreinfo
Control: severity -1 normal



Hi,



On Thu, 19 Jul 2018 13:08:02 +0200 Maximiliano Curia  wrote:

With the new release of sbuild 0.77 (which moves the run_lintian call outside
of the build function), lintian receives the name of a no longer existing
changes file (as the unlink is part of the copy_changes) and as such it fails
with, for example:
 I: schroot -d /<> -c 
unstable-amd64-sbuild-a8e6ff65-a882-41b6-9ce3-b5e764b4ca8c --run-session -q -u maxy -p -- 
lintian -I --pedantic --show-overrides sddm_0.18.0-1_amd64.changes
 D: Running command: schroot -d /<> -c 
unstable-amd64-sbuild-a8e6ff65-a882-41b6-9ce3-b5e764b4ca8c --run-session -q -u maxy -p -- 
lintian -I --pedantic --show-overrides sddm_0.18.0-1_amd64.changes
 warning: "sddm_0.18.0-1_amd64.changes" cannot be processed.
 warning: It is not a valid lab query and it is not an existing file.



 E: Lintian run failed (runtime error)



The issue is not present using the 0.76 version, and can be workaround
commenting the unlink call from the copy_changes function (as shown in the
attached file).



Thanks for working in sbuild. :)



I really fail to reproduce your problem. Can you give me your exact sbuild
invocation so that I may be able to see the same error that you see? I see it
neither in the version currently in unstable nor in git master.


I'm currently using sbuild 0.77.1-2, calling it with:
sbuild --dist=sid --arch=amd64 --chroot=sid-amd64-sbuild --arch-all --arch-any --source 
--run-lintian --lintian-opts="-I --pedantic --show-overrides" --no-run-piuparts 
--purge=never /home/maxy/debian/kde/frameworks/build-area/breeze-icons_5.53.0-1~.dsc

An it fails as long as the mentioned unlink is present.

Happy hacking,
--
"The use of COBOL cripples the mind; its teaching should, therefore, be 
regarded as a criminal offense."

-- Edsger W. Dijkstra
Saludos /\/\ /\ >< `/

Bug#846222: [Pkg-xfce-devel] Bug#846222: x11-common: Please provide a default xsession in /usr/share/xsessions

[Maximiliano Curia]

¡Hola Michael!

El 2017-11-06 a las 13:31 +0100, Michael Biebl escribió:

Another issue I noticed:
gdm3 uses /etc/gdm3/Xsession
sddm uses /etc/sddm/Xsession
lightdm uses /etc/X11/Xsession



to start this special default session.



I wonder if there would be potential for unification (i.e. all affected
display managers using the Xsession script from x11-common) or does
someone know why it could be beneficial to keep separate Xsession
scripts per display manager?


I can only speak for the sddm's Xsession which is a wrapper arround Debian's 
/etc/X11/Xsession, what it does is to try to set the corresponding shell 
environment by loading the user profile files.


For example the bash part reads:
   [ -z "$BASH" ] && exec $SHELL $0 "$@"
   set +o posix
   [ -f /etc/profile ] && . /etc/profile
   if [ -f $HOME/.bash_profile ]; then
 . $HOME/.bash_profile
   elif [ -f $HOME/.bash_login ]; then
 . $HOME/.bash_login
   elif [ -f $HOME/.profile ]; then
 . $HOME/.profile
   fi

The sourced profiles take care of things like adding /usr/games to the PATH, 
which was loudly asked by our users when the wrapper wasn't present.


Happy hacking,
--
"Fighting patents one by one will never eliminate the danger of software 
patents, any more than swatting mosquitoes will eliminate malaria."

-- Richard M. Stallman
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#918080: breeze-gtk: symlink /usr/share/themes/Breeze/{gtk-3.20,gtk-3.0}/

[Maximiliano Curia]

¡Hola Simon!

El 2019-01-02 a las 21:25 -0600, Simon Quigley escribió:

Package: breeze-gtk
Severity: normal
Version: 5.14.3-1



It was raised to my attention from LXQt users that the Breeze GTK theme
cannot be used under LXQt as packaged. This is because the theme exists
in /usr/share/themes/Breeze/gtk-3.18/ and
/usr/share/themes/Breeze/gtk-3.20/ but doesn't exist in
/usr/share/themes/Breeze/gtk-3.0/ which is inconsistent to other GTK
themes in the archive. For example, the arc-theme package just installs
in gtk-3.0.



This raises a few questions for me. Is this just a hidden use of the
standard, or something Plasma-specific? Are the themes meant to only be
used on those minor versions of GTK?



Simply symlinking gtk-3.20 to gtk-3.0 solves the problem under LXQt. Is
it rational to ship this as default?


The theme was adapted to gtk 3.20 (which includes some incompatible changes 
with themes designed for previous gtk versions) and the adapted version is 
shipped in the gtk-3.20 folder, sadly the older version was moved to the 
gtk-3.18 folder. The folder name implies $GTK_VERSION >= 3.20, 
thus it would be wrong to rename it to gtk-3.0 . The gtk-3.18 folder, on the 
other hand, could be renamed, but we don't have a gtk-3 versions older than 
3.18, or even older than 3.20, so that's not really necessary.


I would suspect that lxqt is filtering the themes that that have a gtk-3.0 
folder in them, while a filter for themes that have a gtk-3.* folder would 
yield better results.


Happy hacking,
--
"Politicians and diapers have one thing in common. They should both be changed 
regularly, and for the same reason."

-- José Maria de Eça de Queiroz
Saludos /\/\ /\ >< `/



signature.asc
Description: PGP signature

Bug#916294: statx and statx_timestamp need to be skipped

[Maximiliano Curia]

Package: abi-compliance-checker
Version: 2.3-0.2
Severity: important
Tags: patch

Hi,

with the new glibc version upload sys/stat.h includes bits/statx.h, which in 
turn add two structs (statx and statx_timestamp) which need to be added to the 
C_Structure (TUDump.pm) to avoid a compilation error.


Thanks for maintaining acc in Debian.

Happy hacking,

-- System Information:
Debian Release: buster/sid
 APT prefers testing-debug
 APT policy: (700, 'testing-debug'), (700, 'testing'), (600, 'stable-updates'), 
(600, 'stable-debug'), (600, 'proposed-updates'), (600, 'stable'), (500, 
'buildd-unstable'), (50, 'unstable-debug'), (50, 'unstable'), (1, 
'experimental-debug'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.18.0-2-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8), 
LANGUAGE=en_GB:en (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

Versions of packages abi-compliance-checker depends on:
ii  build-essential  12.5
ii  perl 5.26.2-7+b1

Versions of packages abi-compliance-checker recommends:
ii  exuberant-ctags [ctags]  1:5.9~svn20110310-12

Versions of packages abi-compliance-checker suggests:
pn  doc-base  
pn  icheck

-- no debconf information

--
"La duración de un minuto depende de que lado del baño estés."
-- Ley de la Relatividad (Burke)
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#890503: RFA: oregano -- tool for schematical capture of electronic circuits

[Maximiliano Curia]

¡Hola Ruben!

El 2018-09-29 a las 11:31 +0200, Ruben Undheim escribió:

I would very much like to adopt oregano (within the Electronics team). I am
maintaining a number of other electronics packages (see my status page).


Great, please go ahead with the ITA please.


Thanks you very much in advance.


Thanks to you.

Happy hacking,
--
"If you have too many special cases, you are doing it wrong." -- Craig Zarouni
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#910782: [libkf5wallet5] Support freedesktops Secret Service API

[Maximiliano Curia]

¡Hola Tobias!

El 2018-10-11 a las 09:45 +0200, Tobias Wich escribió:

Package: libkf5wallet5
Version: 5.49.0-1
Severity: wishlist



Hi together,


I'm asking to add support for the Secret Service API [1] to KWallet as I 
would like write a Firefox extension to use KWallet and Gnome Keyring (which 
already provides this API) to manage its passwords. The reason is that the 
old FF extension [2] is not compatible with the Webextension based approach 
and can thus not be used with recent FF releases.


Interesting. I'm sure this would be a great addition to the kde and firefox 
environments, but I don't think that this should be attacked/solved from the 
Debian side. I think that this request should be made upstream, please send 
this request upstream using the kde bugzilla at https://bugs.kde.org, and if 
possible, after reporting the bug, please send the url of the upstream bug to 
this report.


- If I create a ksecrets deb package and provide a patch for the 
libkf5wallet5 package to use it, is there interest to include these changes 
into debian?


In general, we prefer patches that are accepted upstream, and in particular 
for handling user secrets I wouldn't want to diverge upstream at all. So, the 
short answer is not as long as the patches is not accepted upstream.


I would love to hear some feedback from anyone on my plans and if positive I 
would be happy to take care of the mentioned changes. If not I will take 
some other way to make the wallet available in my FF extension.


Oh, if you have a mock up of the changes you are intending to prepare you might 
want to create a WIP review request using https://phabricator.kde.org/, that 
usually gets more eyes that a bug report.


Happy hacking,
--
"A computer program does what you tell it to do, not what you want it to do."
-- Greer's Law
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#908676: [kwin-wayland] Plasma wayland session restarted after pluging or unpluging secondary monitor over thunderbolt

[Maximiliano Curia]

Control: severity -1 normal

¡Hola Jiri!

El 2018-09-12 a las 22:50 +1000, Jiri Kanicky escribió:

Package: kwin-wayland
Version: 4:5.13.5-1
Severity: serious



When I plug or unplug external monitor over thunderbolt the session logs
me out, sometimes the system crashes.


I'm sorry to hear that. Sadly wayland in general is not particularly stable. 
Quoting the package description:


This package provides the wayland version, which is still a work in progress
project, and is available as a PREVIEW release. Don't expect the same
stability as with the x11 version.

On the other hand, I would suspect that the real issue here is not caused by 
kwin but by kscreen or by wayland itself. For the wayland part you could test 
if the issue is reproducible using gnome under wayland. I'm not sure about how 
to test if the issue is in kscreen.


Happy hacking,
--
"I decry the current tendency to seek patents on algorithms. There are better 
ways to earn a living than to prevent other people from making use of one's 
contributions to computer science."

-- Donald Knuth
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#904698: sddm fails to login with user shell fish and bash completion installed

[Maximiliano Curia]

Control: tag -1 + pending

¡Hola Alf!

El 2018-07-26 a las 22:16 +0200, Alf Gaida escribió:

Package: sddm
Version: 0.18.0-1
Severity: normal



fish has some problems with the newly introduced sourcing of /etc/profile - esp
an installed bash-completion and /etc/profile.d/bash_completion.sh


Anyway, I've reverted the source of /etc/profile for fish after checking this:
https://github.com/fish-shell/fish-shell/issues/3665

fish gives syntax errors on things like &&, so I don't think that applying
https://github.com/sddm/sddm/commit/f749f1d65165de7ce7b9ae073b19f057b205ab35 
was a good idea anymore.



A patch to make sourcing of files more fault tolerant for all shells is 
attached.


The "patch" is sourcing all the files in a different subshell, that means, no 
changes in the environment are in effect. This is not an acceptable "fix".


I just commented the /etc/profile line for fish.

Happy hacking,
--
"If a pickpocket meets a saint, he sees only his pockets."
-- Kegley's Law
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#904137: Missing icon for: gtk-find

[Maximiliano Curia]

¡Hola cruncher!

El 2018-07-20 a las 15:03 +0200, cruncher escribió:

Package: oxygen-icon-theme
Version: 5:5.47.0-1
Severity: minor
Tags: upstream



As the title suggests, the icon for "actions/gtk-find" is missing.


As you reported this is an upstream issue, please consider sending the report 
upstream using bugs.kde.org. In the bug report, please try to provide some 
extra information that the one available in this report, such as:


- What kind of issue does the missing icon produces?
- Is this reproducible when using a single program or there are many affected?

Also, consider that it's impossible for an icon theme to provide every icon in 
use, as such the programs using non "standard"/"stock" icons need to provide 
their fallback, and in the same fashion a library providing a "stock" icon, 
well, needs to provide it. :)


If you send this bug upstream please send us a note with the bug url so we can 
keep track of it.


Happy hacking,
--
"C makes it easy to shoot yourself in the foot; C++ makes it harder, 
but when you do it blows your whole leg off."

-- Bjarne Stroustrup
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#884749: plasma-workspace: ksmserver-logout-greeter crashes stopping shutdown

[Maximiliano Curia]

¡Hola Harald!

El 2018-07-23 a las 10:45 +0200, Harald Dunkel escribió:

This package is in the dependency chain of KDE, i.e. the crash affects
all the KDE users. Would you mind to adjust the severity of this bug
report accordingly?


Were you able to reproduce this issue?

I don't think this bug is currently affecting any users. In fact, as reported 
it seems to be a one time crash, probably caused by a plasma upgrade while 
running plasma (which is explicitly not supported upstream).


Happy hacking,
--
"The most important thing in the programming language is the name. A language 
will not succeed without a good name. I have recently invented a very good 
name and now I am looking for a suitable language."

-- Donald Knuth
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#903713: plasma-browser-integration: "This_file_is_part_of_KDE" in debian/copyright?

[Maximiliano Curia]

¡Hola Albert!

El 2018-07-19 a las 18:54 +0200, Albert Astals Cid escribió:

Let's see a sample header like ar/messages/kdegraphics/okular_mobi.po



# Copyright (C) YEAR This_file_is_part_of_KDE
# This file is distributed under the same license as the PACKAGE package.
# Zayed Al-Saidi , 2009.
# Abdalrahim G. Fakhouri , 2014.



The first one is the one you mentioned, personally i think we can just
delete the first line (or change them to "For Copyright see the individual
names below"), they are "worthless/wrong" and if people use the "right"
tools for translation their copyright is added after those lines, i.e. lines
3 and 4.



The problem with this kind of copyright assignment is that it's not machine
readable (a line with a name is very hard to distinguish from a piece of
text), and tools like decopy, licensecheck, scancode, etc will simply ignore
the list of authors. So, if you are going to change this format, please
consider adding a "Copyright: " to the list of authors, then this would
become:



 # Copyright: Zayed Al-Saidi , 2009.
 # Copyright: Abdalrahim G. Fakhouri , 2014.


That's not possible, the Copyright line gets added by third party tools we 
do not control, and even if we updated Lokalize (the tool "most" of our 
translators use) *today*, it wouldn't get to them until years later thanks 
to how distribution of applications in Linux works at this time.


Alright, what about:

# This file is copyright:
# Zayed Al-Saidi , 2009.
# Abdalrahim G. Fakhouri , 2014.

this would only mean changing the proposed "For Copyright see the individual 
names below" to a string that is already supported by automatic copyright 
parsers.


Happy hacking,
--
"There are two ways of constructing a software design.  One way is to make it 
so simple that there are obviously no deficiencies. And the other way is to 
make it so complicated that there are no obvious deficiencies."

-- C.A.R. Hoare
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#904083: Regression run_lintian calls lintian with a non existing filename

[Maximiliano Curia]

Package: sbuild
Version: 0.77.0-3
Severity: important

Hi,

With the new release of sbuild 0.77 (which moves the run_lintian call outside 
of the build function), lintian receives the name of a no longer existing 
changes file (as the unlink is part of the copy_changes) and as such it fails 
with, for example:

I: schroot -d /<> -c 
unstable-amd64-sbuild-a8e6ff65-a882-41b6-9ce3-b5e764b4ca8c --run-session -q -u maxy -p -- 
lintian -I --pedantic --show-overrides sddm_0.18.0-1_amd64.changes
D: Running command: schroot -d /<> -c 
unstable-amd64-sbuild-a8e6ff65-a882-41b6-9ce3-b5e764b4ca8c --run-session -q -u maxy -p -- 
lintian -I --pedantic --show-overrides sddm_0.18.0-1_amd64.changes
warning: "sddm_0.18.0-1_amd64.changes" cannot be processed.
warning: It is not a valid lab query and it is not an existing file.

E: Lintian run failed (runtime error)

The issue is not present using the 0.76 version, and can be workaround 
commenting the unlink call from the copy_changes function (as shown in the 
attached file).


Thanks for working in sbuild. :)

Happy hacking,

-- System Information:
Debian Release: buster/sid
 APT prefers testing-debug
 APT policy: (700, 'testing-debug'), (700, 'testing'), (600, 'stable-updates'), 
(600, 'stable-debug'), (600, 'proposed-updates'), (600, 'stable'), (500, 
'oldstable-updates'), (500, 'oldstable-proposed-updates'), (500, 
'buildd-unstable'), (500, 'oldstable'), (50, 'unstable-debug'), (50, 
'unstable'), (1, 'experimental-debug'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.16.0-2-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8), 
LANGUAGE=en_GB:en (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

Versions of packages sbuild depends on:
ii  adduser 3.117
ii  libsbuild-perl  0.77.0-3
ii  perl5.26.2-6

Versions of packages sbuild recommends:
ii  autopkgtest  5.4.1
ii  debootstrap  1.0.106
ii  schroot  1.6.10-5

Versions of packages sbuild suggests:
pn  deborphan  
ii  kmod   25-1
ii  wget   1.19.5-1

-- Configuration Files:
/etc/sbuild/sbuild.conf changed:
1;


-- no debconf information

--
"It is not the task of the University to offer what society asks for, but to 
give what society needs."

-- Edsger W. Dijkstra
Saludos /\/\ /\ >< `/
diff --git a/lib/Sbuild/Build.pm b/lib/Sbuild/Build.pm
index 18c8656c..cf039e91 100644
--- a/lib/Sbuild/Build.pm
+++ b/lib/Sbuild/Build.pm
@@ -2598,8 +2598,8 @@ sub build {
 		rename("$sys_build_dir/$changes.new", "$sys_build_dir/$changes")
 		or $self->log("$sys_build_dir/$changes.new could not be " .
 		"renamed to $sys_build_dir/$changes: $!\n");
-		unlink("$build_dir/$changes")
-		if $build_dir;
+		# unlink("$build_dir/$changes")
+		# if $build_dir;
 	}
 
 	return $pchanges;


signature.asc
Description: PGP signature

Bug#903713: plasma-browser-integration: "This_file_is_part_of_KDE" in debian/copyright?

[Maximiliano Curia]

¡Hola Albert!

Thanks for the follow up.

El 2018-07-19 a las 00:32 +0200, Albert Astals Cid escribió:
Yes, there is a long standing issue with the translations of the .po files 
that carry not really good copyright information most of the times.



Let's see a sample header like ar/messages/kdegraphics/okular_mobi.po



# Copyright (C) YEAR This_file_is_part_of_KDE
# This file is distributed under the same license as the PACKAGE package.
# Zayed Al-Saidi , 2009.
# Abdalrahim G. Fakhouri , 2014.


The first one is the one you mentioned, personally i think we can just 
delete the first line (or change them to "For Copyright see the individual 
names below"), they are "worthless/wrong" and if people use the "right" 
tools for translation their copyright is added after those lines, i.e. lines 
3 and 4.


The problem with this kind of copyright assignment is that it's not machine 
readable (a line with a name is very hard to distinguish from a piece of 
text), and tools like decopy, licensecheck, scancode, etc will simply ignore 
the list of authors. So, if you are going to change this format, please 
consider adding a "Copyright: " to the list of authors, then this would 
become:


# Copyright: Zayed Al-Saidi , 2009.
# Copyright: Abdalrahim G. Fakhouri , 2014.

Which is easily detectable by any of the mentioned tools.

The second line is also wrong, what is "PACKAGE"? In my opinion now that we 
ship the translations as part of the application tarballs themselves it's 
clear-ish that unless otherwise stated in the file, the files are under the 
copyright stated in the COPYING file so we may as well delete those lines 
too.



Opinions?


I agree that deleting the second line and "inheriting" the license from a 
LICENSE or COPYING file is clearer than what's currently being shipped. And 
from my part I would welcome such a change.


Happy hackings,
--
"Any sufficiently advanced technology is indistinguishable from magic."
-- Clarke's Third Law
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#903223: libkf5kiocore5: Symlink behavior over SFTP connection broken

[Maximiliano Curia]

¡Hola Alex!

El 2018-07-07 a las 18:23 -0400, Alex Krusemark escribió:

Package: libkf5kiocore5
Version: 5.47.0-1
Severity: normal



Since a recent upgrade (I think it was the transition from KDE Frameworks 5.46
to 5.47), on filesystems accessed over an SFTP connection symbolic/soft links
whose target is a directory have broken behavior. Symlinks whose targets are
files still work properly.



Links to directories are shown as a link to a file with an unknown mimetype,
rather than that of a directory, and can neither be double-clicked to follow
nor expanded in tree view. The only way I see to follow them is by right-
clicking and choosing "Show Target" in Dolphin, with no obvious workaround in
other applications. I confirmed that this problem was present in Dolphin,
Krusader, Konqueror, Kate, and Calligra.


Does kio-extras 4:18.04.3-1 (currently available in experimental) fixes this 
issue for you?


Happy hacking,
--
: You are in a dark room with a compiler, emacs, an internet connection,
: and a thermos of coffee.
: Your move ?
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#891068: im-config: Failed to setup fcitx when zsh used as login shell

[Maximiliano Curia]

Control: tag -1 - upstream

¡Hola!

El 2018-03-21 a las 19:04 +0800, 崔灏 escribió:

Hi,



Hmmm SDDM change shell parsing Xsession, then things may break.



Are you sure im-config is the problem?



I agree it's likely SDDM is to blame, not im-config.


I think that what is to blame here is the zsh's nomatch option that's being set 
to be used even in a non interactive script. Please move this option to your 
zshrc file.


As some background information: Running the XSession without the Xsession 
wrapper shipped by sddm ignores the /etc/profile and $HOME/.profile files (and 
in particular for zsh, also the zlogin and zprofile files), this causes 
that the shells run under the graphical user session to start an interactive 
session without loading the commands set to be run in a login session, which 
is an undesirable state. In particular for zsh this also causes an incomplete 
PATH.


The workaround that's in place in sddm has been in use for many years now (it 
was previously part of kdm), and fcitx has been successfully used with it.


Happy hacking,
--
"The cheapest, fastest and most reliable components of a computer system are 
those that aren't there."

-- Gordon Bell
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#903713: plasma-browser-integration: "This_file_is_part_of_KDE" in debian/copyright?

[Maximiliano Curia]

¡Hola Luigi!

El 2018-07-14 a las 10:37 +0100, Chris Lamb escribió:

My interpretation of this is that the intention is to assign the copyright
to the kde project, although it's not a hundred percent clear.



I should have been clearer, sorry — I understand you are going with
whatever the file says but I am requesting that you make this clearer,
perhaps by getting a statement from upstream or similar.



"This_file_is_part_of_KDE" is really not suitable as an author,
whatever the file says, after all.


Chris raised the issue of the po files distributed by kde containing some (not 
very clear) template parts, in particular the copyright assignments to 
This_file_is_part_of_KDE.


With your kde i18n team hat on, would you consider it feasible to replace 
these strings with something clearer?


If the intention is for the translators to assign the copyright to kde it 
should be assigned to KDE.e.V, if the intention is for each translator to keep 
the copyright assignment the This_file_is_part_of_KDE part of the template 
needs to be updated to say AUTHOR .


The first case should be "scriptable" the second case, would need to manually 
modifying each po file that contains the "This_file_is_part_of_KDE" text.


Happy hacking,
--
"Brilliant opportunities are cleverly disguised as insolvable problems."
-- Gardener's Philosophy

"The reverse is also true." -- Corollary
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#903713: plasma-browser-integration: "This_file_is_part_of_KDE" in debian/copyright?

[Maximiliano Curia]

¡Hola Chris!

El 2018-07-13 a las 16:39 +0100, Chris Lamb escribió:

Source: plasma-browser-integration
Version: 5.13.1-1
Severity: serious
X-Debbugs-CC: Maximiliano Curia 



I just ACCEPTed plasma-browser-integration from NEW but noticed it
declares "This_file_is_part_of_KDE" as an copyright holder.



This seems very... odd.


Agreed, this is caused by the templates used in po files. For example in 
po/ca/plasma_runner_browsertabs.po it says:

# Translation of plasma_runner_browsertabs.po to Catalan
# Copyright (C) 2017 This_file_is_part_of_KDE
# This file is distributed under the license LGPL version 2.1 or
# version 3 or later versions approved by the membership of KDE e.V.
#
# Josep Ma. Ferrer , 2017.

Josep here is clearly the author, but there is no copyright assigment 
statement with his name. My interpretation of this is that the intention 
is to assign the copyright to the kde project, although it's not a hundred 
percent clear. So I go with whatever the file says.


Happy hacking,
--
"The best way to predict the future is to invent it."
-- Alan Kay
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#897221: Session logout doesn't TERMinate all processes

[Maximiliano Curia]

¡Hola Didier!

El 2018-06-06 a las 19:38 +0200, Didier 'OdyX' Raboud escribió:

Please check loginctl session-status  (The id can be obtained with
loginctl list-sessions)



I noticed it also appears when disconnecting, and reconnecting back. Akonadi
is then broken, etc etc.



Hereattached is a log of the processes from a stopped session. It really feels
like plasma doesn't stop its child processes from the systray either.



Unit: session-3.scope
  ├─ 2690 /usr/lib/postgresql/9.6/bin/postgres -D 
/home/didier/.local/share/akonadi/db_data -k/tmp/akonadi-didier.bR0FSO -h
  ├─ 2695 postgres: checkpointer process
  ├─ 2696 postgres: writer process
  ├─ 2697 postgres: wal writer process
  ├─ 2698 postgres: autovacuum launcher process
  ├─ 2699 postgres: stats collector process
  ├─ 2955 /usr/lib/at-spi2-core/at-spi-bus-launcher 
--launch-immediately


I've noticed that at-spi-bus-launcher won't close when the session finishes, 
I've seen this behaviour in cinnamon also. You might want to try killing it 
and checking if that reduces the amount of waiting processes.



  ├─ 2976 /usr/bin/dbus-daemon 
--config-file=/usr/share/defaults/at-spi2/accessibility.conf --nofork 
--print-address 3
  ├─ 2986 /usr/bin/perl /usr/bin/parcimonie-applet
  ├─ 3083 /usr/bin/python3 /usr/bin/hp-systray -x


I would expect these two to die "gracefully" when there is no X connection. 
Are these two zombies? Is plasmashell still running? Is X still running?



  ├─ 3086 /usr/bin/python3 /usr/bin/hp-systray -x
  ├─ 3920 
/usr/lib/x86_64-linux-gnu/libexec/kf5/kio_http_cache_cleaner


I'm sure this one will go away on its own.


  ├─20252 postgres: didier akonadi [local] idle
  ├─20277 postgres: didier akonadi [local] idle
  ├─20278 postgres: didier akonadi [local] idle
  ├─20507 postgres: didier akonadi [local] idle
  ├─20509 postgres: didier akonadi [local] idle
  ├─20792 /usr/lib/at-spi2-core/at-spi2-registryd 
--use-gnome-session


The --use-gnome-session here seems wrong.


  ├─21250 postgres: didier akonadi [local] idle
  ├─21252 postgres: didier akonadi [local] idle
  ├─21318 postgres: didier akonadi [local] idle
  └─21377 postgres: didier akonadi [local] idle


There is no akonadi process in this list, but a postgres running with your 
user (for akonadi purposes), I don't think they are even started by akonadi. 
You might want to migrate this db to a pgcluster configuration, so it's 
independent of your login session.


In any case you might want to keep checking the output of loginctl 
session-status  and killing the "offending" processes till you have a 
minimal set of processes that are not being correctly stopped.


I wonder if at-spi-bus-launcher and postgres would behave "better" if set as a 
systemd user service.



Mmh, this is a complex problem. Debian systemd reverts the default logind
configuration KillUserProcesses, that effectively terminates the session
when the user logs out, and most desktop environments assume this is set to
it's default value, so they don't have to deal with the processes started
by systemd --user, which, by the way, the desktop environment had nothing
to do with.



Also, kde does nothing to remember which processes were started as autostart
desktop files.



Gnome "fixes" this by terminating the logind session, as if
KillUserProcesses was set, at the cost of breaking tmux and screen (which is
the reason why Debian reverts the KillUserProcesses setting on the first
place).



So, either we need to fix tmux and screen to they can be detached from the
logind session, or fix the kde concept of session so it matches better what
logind expects.



The first alternative looks like the correct thing to do to me. So a route to
fix this is for systemd to revert the KillUserProcesses back to upstream's
default, right? This was apparently last reinforced in systemd 234-3 and was
discussed in #825394.



But it doesn't seem like this would gather consensus currently. How hard is it
to get KDE to grow better login session processes handling?


Well, at least the the ksmserver/startup.cpp launches the autostart desktop 
files with a QProcess::startDetached call, this would need to be refactor so 
the processes are waited and notified on shutdown.


There are probably other places where a process ends up being part of the 
logind session, but I guess this should be a good start.


Happy hacking,
--
"Fighting patents one by one will never eliminate the danger of software 
patents, any more than swatting mosquitoes will eliminate malaria."

-- Richard M. Stallman
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#902211: mark breeze-icon-theme Multi-Arch: foreign

[Maximiliano Curia]

¡Hola Helmut!

El 2018-06-23 a las 14:33 +0200, Helmut Grohne escribió:

Package: breeze-icon-theme
Version: 4:5.47.0-1
Tags: patch
User: helm...@debian.org
Usertags: rebootstrap
Control: affects -1 + src:kpimtextedit



kpimtextedit cannot satisfy its cross Build-Depends, because its
dependency on breeze-icon-theme is unsatisfiable. In general,
Architecture: all packages can never satisfy cross Build-Depends unless
marked Multi-Arch: foreign. In this case such a marking is correct,
because the only dependency - hicolor-icon-theme - is already thus
marked and its maintainer scripst only update icon caches, which is an
architecture independent operation. breeze-icon-theme-rcc can also be
thus marked, because it lacks dependencies and maintainer scripts.
Please consider applying the attached patch.


I might have already asked this, but if so, I really don't remember the 
answer. Is there any case where an Arch: all package shouldn't/couldn't be 
correctly taken as Multi-Arch: foreign?


Happy hacking,
--
"If you think your users are idiots, only idiots will use it."
-- Linus Torvalds
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#901655: plasma-workspace: Keyboard input and clipboard freezes when virt-manager is used

[Maximiliano Curia]

¡Hola Phil!

El 2018-06-16 a las 06:02 -0400, Phil Wolf escribió:

Package: plasma-workspace
Version: 4:5.12.5-1
Severity: normal



after update to Buster I found that I cannot use virtualization (KVM/QEMU, 
virt-manager) on KDE Plasma 5.12
since clipboard is constantly freezing. I tested this behaviour both with 
Debian VM with Plasma and Windows VM.



To reproduce this issue simply install and upgrade host to Buster and run one 
or more VMs with
"spice-vdagent" that handles clipboard. Simply copy text (Ctrl+C/V) from host 
to host or from host to VM or
from VM to host or within. In random intervals paste fails and time to time 
keyboard input hangs so you
cannot even type.



The solution is to close virt-manager. Then everything is OK again and you can 
copy and paste.


As a workaround, you might want to disable the kde clipboard manager, you can 
disable it in the system tray settings (right click on top of the system tray 
applet), and deselect the clipboard checkbox.


As for the bug, this is an upstream bug, either in spice-vdagent or in klipper 
or in spice-vdagent, could you report the issue to klipper in bugs.kde.org? If 
you do this, please add a note in this bug with the link of the reported bug.


Happy hacking,
--
"It is practically impossible to teach good programming to students that have 
had a prior exposure to BASIC: as potential programmers they are mentally 
mutilated beyond hope of regeneration."

-- Edsger W. Dijkstra
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#901734: /usr/bin/plasmashell: recently taskbar becames unreadable after some time of use

[Maximiliano Curia]

¡Hola Paul!

El 2018-06-17 a las 10:39 +0200, Paul Gevers escribió:

Recently I had multiple occurances where the taskbar of my KDE session suddenly
became unreadable. As an example, I have attached a screen shot of the task bar
in such a case.



I haven't figured out what triggers the task bar to behave like that.
Nevertheless I already wanted to report the issue. If you have debugging
tips, feel free to contact me for more info.


I would suspect that this is related to a change in xorg 1.20, maybe the 
change to DRI3 by default in some video drivers (including modesetting).


I haven't seen this behaviour, though (using modesetting + intel).

What video card are you using?


Thanks for maintaining KDE in Debian.


You are welcome.

Happy hacking,
--
"People get lost in thought because it is unfamiliar territory."
-- Fix's Principle
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#900373: Update - plasmashell too

[Maximiliano Curia]

¡Hola Alex!

El 2018-06-07 a las 22:29 +0300, Alex Dănilă escribió:
The same started to happen for plasmashell too. It happens from time 
to time when it shows a notification or a pop-up (for example the 
window preview when hovering the task bar). Here's the stack:


This is probably #900145 [1]

Which versions of mesa (libgl1-mesa-dri) and xorg-server (xserver-xorg-core) 
are you running?


[1]: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=900145

Happy hacking,
--
"A computer program does what you tell it to do, not what you want it to do."
-- Greer's Law
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#852680: 答复: Bug#852680: plasma-workspace: Please add dbus-user-session as alternative dependency for dbus-x11

[Maximiliano Curia]

¡Hola 张!

El 2018-06-07 a las 10:35 +, 张 敬强 escribió:


Hi,

 Does it make any difference? What's the rationale behind this request?



According to dbus NEWS file, dbus-user-session provides systemd units for
a per-user bus listening (in effect, treating all concurrent graphical and
non-graphical login sessions as one large session)



dbus-session-bus is a virtual package, which is also provided by dbus, so
that's most likely going be fulfilled by dbus if used as a dependency, also
dependencies against virtual packages need to be given a non virtual
preference. Probably depending on: dbus | dbus-session-bus would work.



The recommended way seems to be:
default-dbus-session-bus | dbus-session-bus
, which is already used by cinnamon-session, but not by gnome-session-bin.



And the binary package dbus itself doesn't provide dbus-session-bus.



But breaking the session dbus for our users wouldn't be nice so this needs
a somewhat thorough testing, in an enviroment with only dbus installed, with
only dbus-user-session installed, with only dbus-x11 installed, etc, checking
for differences in each session.



So now we only have to test dbus-x11 and dbus-user-session.
There is a check for dbus-user-session in dbus-x11 file
/etc/X11/Xsession.d/75dbus_dbus-launch, so in theory dbus-user-session
replace dbus-x11 if they both get installed.
I haven't encountered any difference when using only one graphical session.


Ok, you convinced me. Thanks for the report and follow up. For future reports, 
whenever possible, please add this level of information.


Happy hacking,
--
"Politicians and diapers have one thing in common. They should both be changed 
regularly, and for the same reason."

-- José Maria de Eça de Queiroz
Saludos /\/\ /\ >< `/



signature.asc
Description: PGP signature

Bug#852680: plasma-workspace: Please add dbus-user-session as alternative dependency for dbus-x11

[Maximiliano Curia]

¡Hola 张!

El 2018-06-07 a las 03:15 +, 张 敬强 escribió:

Please consider add dbus-user-session as alternative for dbus-x11.
Or just use dbus-session-bus as dependency.

Still no pending action?


Does it make any difference? What's the rationale behind this request?

dbus-x11's dbus-launch is explicitly used by upstream's unittests, iirc, 
that's the main reason why the dbus-x11 dependency is present. It could 
probably be dropped (and added to the testsuite dependencies).


dbus-session-bus is a virtual package, which is also provided by dbus, so 
that's most likely going be fulfilled by dbus if used as a dependency, also 
dependencies against virtual packages need to be given a non virtual 
preference. Probably depending on: dbus | dbus-session-bus would work.


But breaking the session dbus for our users wouldn't be nice so this needs 
a somewhat thorough testing, in an enviroment with only dbus installed, with 
only dbus-user-session installed, with only dbus-x11 installed, etc, checking 
for differences in each session.


If you do this, please open a merge request in salsa.

Happy hacking,
--
"Las computadoras son inútiles, solo pueden darte respuestas."
-- Pablo Picasso
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#900523: plasma-widgets-addons: System load viewer depends on ksysguard

[Maximiliano Curia]

¡Hola Benoit!

El 2018-06-01 a las 11:18 +0200, Benoit Friry escribió:

Bonjour Maximilianon,



Le 01/06/2018 à 10:18, Maximiliano Curia a écrit :

Without KSysGuard, the system load widget does not display any value.



The package plasma-widgets-addons should recommand or suggest ksysguard
package.



Thanks for the report.



Looking at the code the ksysguard is only used when you click on the
applet, otherwise it seems to be using the systemmonitor engine from
plasma-workspace, which in turn uses ksysguardd (?), mmh.



Without ksysguard installed:
- the applet bars remain empty
- when I click on the applet, I get a processus table, without any
  access to the system load tab (graphs)


Have you tried having ksysguardd installed and not ksysguard?

Here I get the contents of the applet having ksysguardd installed, without 
ksysguard. Which also makes sense with the code I've read.


Happy hacking,
--
"Any change looks terrible at first." -- Principle of Design Inertia
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#900523: plasma-widgets-addons: System load viewer depends on ksysguard

[Maximiliano Curia]

¡Hola Benoit!

El 2018-05-31 a las 21:29 +0200, Benoit Friry escribió:

Package: plasma-widgets-addons
Version: 4:5.12.5-1
Severity: minor



Without KSysGuard, the system load widget does not display any value.



The package plasma-widgets-addons should recommand or suggest ksysguard
package.


Thanks for the report.

Looking at the code the ksysguard is only used when 
you click on the applet, otherwise it seems to be using the systemmonitor 
engine from plasma-workspace, which in turn uses ksysguardd (?), mmh.


Adding both recommends.

Happy hacking,
--
"Backtracking algorithms are nondeterministic, not in the sense of being 
random, but in the sense of having free will."

-- Robert W. Floyd
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#900145: plasmashell freezes with xserver 1.20

[Maximiliano Curia]

Control: block -1 by 900352

¡Hola 张!

El 2018-05-29 a las 10:43 +, 张 敬强 escribió:

It seems that the freeze is due to xserver-xorg 1.20.
Downgrade xserver-xorg to 1.19 solve the problem.
I have tried use 'root' to `startx`, which has a default plasma config, and 
still get the freeze.
So there may be some incompatible between plasma 5.12 and xorg 1.20.


Ok, I could reproduce the issue with the new xorg. Thanks for reporting. I 
reported a bug to the xorg-server package.


Happy hacking,
--
"La duración de un minuto depende de que lado del baño estés."
-- Ley de la Relatividad (Burke)
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#900352: new xorg-server version causes a random freezes in plasmashell

[Maximiliano Curia]

Package: src:xorg-server
Version: 2:1.20.0-2
Severity: critical
Tags: upstream

Hi,

The severity is set as it breaks "unrelated programs" although I'm not sure a 
desktop environment can be called "unrelated" to x, but in any case, it would 
be better if this version of xorg does not migrate to testing till this is 
fixed.


The new xorg-server version seems to be causing plasmashell to freeze. This 
was first reported in #900145, and it's also seen in other distros: 
https://bugs.archlinux.org/task/58549


Upstream seems to have a patch for this (actually two patches that fix this 
with two different aproaches), that I havent tested: 
https://lists.x.org/archives/xorg-devel/2018-May/056829.html


Happy hacking,

-- System Information:
Debian Release: buster/sid
 APT prefers testing-debug
 APT policy: (700, 'testing-debug'), (700, 'testing'), (600, 'stable-updates'), 
(600, 'stable-debug'), (600, 'proposed-updates'), (600, 'stable'), (500, 
'buildd-unstable'), (50, 'unstable-debug'), (50, 'unstable'), (1, 
'experimental-debug'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.16.0-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8), 
LANGUAGE=en_GB:en (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

--
"Always code as if the person who ends up maintaining your code is a violent 
psychopath who knows where you live."

-- John Woods
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#823037: ITP: flatbuffers -- efficient cross platform serialization library

[Maximiliano Curia]

¡Hola!

El 2018-05-29 a las 09:56 +0200, Maximiliano Curia escribió:

I created a packaging salsa project here:
https://salsa.debian.org/debian/flatbuffers


Ups, that was a lie, the project was created by Nobuhiro Iwamatsu. I was 
checking the flatbuffers, sink and kube code base last week and I today, 
when I found that an empty project was already present, I tought I created 
the repository and forgot about it.


Nobushiro, sorry to meddle with this package. Please let me know if you are 
willing to group maintain this package, if not, please take the ownership of 
this bug.


Happy hacking,
--
"By definition, when you are investigating the unknown, you do not know what 
you will find"

-- The Ultimate Principle
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#823037: ITP: flatbuffers -- efficient cross platform serialization library

[Maximiliano Curia]

Control: retitle -1 ITP: flatbuffers -- efficient cross platform serialization 
library
Control: owner -1 !

¡Hola!

El 2016-04-30 a las 18:37 +1000, Jonathon Love escribió:

Package: wnpp
Severity: wishlist
Owner: Jonathon Love 



* Package name: flatbuffers
 Version : 1.3.0
 Upstream Author : Wouter van Oortmerssen
* URL : http://google.github.io/flatbuffers/
* License : Apache-V2
 Programming Lang: C++, Java, Javascript
 Description : efficient cross platform serialization library



FlatBuffers is an efficient cross platform serialization library for
games and other memory constrained apps. It allows you to directly
access serialized data without unpacking/parsing it first, while still
having great forwards/backwards compatibility.


I'll give flatbuffers a try, I'm interested in flatbuffers as a dependency 
needed for kde's kube (https://kube.kde.org/).


I created a packaging salsa project here:
https://salsa.debian.org/debian/flatbuffers

Merge requests are welcome. :)

Happy hacking,
--
"If it ain't broke, don't fix it" -- Bert Lance

"If we can't fix it, it ain't broke" -- Lieutenant Colonel Walt Weir
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#900145: plasma-workspace: plasmashell freezes every few times it appears

[Maximiliano Curia]

Control: tag -1 + moreinfo

¡Hola Jason!

El 2018-05-26 a las 15:17 -0400, Jason Riedy escribió:

Package: plasma-workspace
Version: 4:5.12.5-1
Severity: important


The plasmashell component freezes every few (5-15?) times the system-key 
menu appears.  I have to kill and restart plasmashell.  There has been 
nothing obvious in the output (but I have not yet captured the output to 
send).


What do you mean by the "system-key menu"? The context menu that appears when 
pressing the menu key? The System activity window that appaear when pressing 
"Control + Esc"? The application launcher that appears when pressing the 
"Super" or "Windows" key? The screenshot dialog that appears when pressing 
"Print screen/System request"?


For most of this combinations I use them quite frequently and I'm pretty sure 
they work as expected, can you please describe exactly how to reproduce the 
issue you are seeing.


Can you reproduce the issue if you start a new plasma session with a newly 
created user? (this should rule out some specific configuration in your user)


Happy hacking,
--
"Good judgement comes from experience, and experience comes from bad 
judgement."

-- Fred Brooks
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#899984: sddm: Password automatically shown in clear at every user login

[Maximiliano Curia]

Control: tag -1 + unreproducible

¡Hola r.ductor!

El 2018-05-24 a las 11:42 +0200, r.ductor escribió:

Package: sddm
Version: 0.17.0-1
Severity: normal


The greeter ui is part of the theme, the themes shipped with sddm don't have a 
"show password" button (afaik), the sddm-theme-breeze has this feature, 
provided by plasma-frameworks' plasmacomponentstandard TextField widget, which 
seems to work fine. In any case, this is part of the theme, not part of sddm 
itself. Which theme are you using?



At every login autentifications passwords are automatically shown in clear when 
entered.


None of the themes I've seen show the passwords in plain text by default, and 
only breeze has the option to show it by pressing the "eye" button.



Maybe this is not a bug and you think this is a very useful feature,
and maybe I've inadvertently clicked on the eye icon that I see on other 
machines but not on this one.
If this is the case there should be a simple way out from this unsafe behavior,
so take this as a wish of better documentation (I find no mention of this in 
the man pages)
or better hide/unhide icons.


I'm not sure what you are trying to say here. Is your bug report about the 
existance of the reveal password option (the "eye" icon)? As mentioned, this 
is part of the theme, if you don't like it, please use a different one.


Happy hacking,
--
"The cheapest, fastest and most reliable components of a computer system are 
those that aren't there."

-- Gordon Bell
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#899378: plasma-desktop: Focus stealing prevention breaks when set to "high"

[Maximiliano Curia]

¡Hola Emmanuel!

El 2018-05-23 a las 15:59 +0200, Emmanuel Revah escribió:

Package: plasma-desktop
Version: 4:5.8.6-1
Severity: normal



  * What led up to the situation?



I set focus stealing prevention to "high"



  * What was the outcome of this action?


Focus stealing prevention blocks krunner and kmenu (as well as the 
clock), these elements very briefly appear before being "blocked".


Interesting, but I'm not sure if this can be considered a bug or a feature. In 
any case this is an upstream issue, can you please report this issue in the 
kde bug tracker (https://bugs.kde.org, you'll need a bugzilla account for 
reporting this), if you do this, please leave a note in this bug with the url 
of the upstream bug so we can track it.



  * What outcome did you expect instead?


I was expecting the focus stealing prevention to help avoid having new 
windows steal the focus while I'm focused and active on a program.


For example, I open Firefox, but it takes a while, so I start typing 
in a terminal, and then, Firefox opens and steals the focus, and bam, 
I'm searching the Internet for my root password.


I don't see this behaviour, testing it with xterm, firefox and konsole. Was 
this just an hypothetical example or is this reproducible in your setup? What 
other rules do you have in place? What are you using for activating windows?


Also, please check your windows rules, you can set specific values to stealing 
prevention, and stealing protection using them.


With the "smart" focus stealing prevention that was available before 
(I'm not sure when), I could open new windows, they would take the 
focus only if I wasn't actively using another program.


There is no smart focus stealing prevention, at least in the plasma 5 
versions, was that ever part of kwin?


Regardless, of the "smart" aspect, setting FSP to "high" probably 
shouldn't block krunner and kmenu.


I'm not sure if a keybinding should follow a different rule, please, discuss 
this upstream. Saying that, even pressing on the application launcher doesn't 
work with FSP set to "high", that can't be right.


Happy hacking,
--
"Some people worry that artificial intelligence will make us feel inferior, 
but then, anybody in his right mind should have an inferiority complex every 
time he looks at a flower."

-- Alan Kay
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#899144: [oxygen-icon-theme] Please include scalable version

[Maximiliano Curia]

¡Hola Bastien!

El 2018-05-19 a las 21:42 +0200, Bastien ROUCARIÈS escribió:

Package: oxygen-icon-theme
Version: 5:5.45.0-1
Severity: important
control: tags -1 + patch



Hi,



Some package packages scalable version (svg) of this theme. Could be possible
to get scalable version here in order to create symbolic link and reduce code
duplication



Here patch untested by could help


Can you please send the patch upstream? Currently upstream processes the 
patches through the phabricator.kde.org (the documentation on how to use it 
can be found: https://community.kde.org/Infrastructure/Phabricator)


Please consider adding a rationale on why are the svg and svgz useful for you 
in the installed paths (as requested by Pino).


Happy hacking,
--
"Nothing ever goes away." -- Commoner's Law of Ecology
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#797331: dracut: doesn't show the name of the LUKS volume to unlock in the graphical plymouth interface

[Maximiliano Curia]

¡Hola intrigeri!

El 2015-08-29 a las 17:36 +0200, intrig...@debian.org escribió:

Package: dracut
Version: 043-2
Severity: minor



I'm using dracut + plymouth (0.9.0-9) on sid, and booting my system
requires unlocking a LUKS volume. This works fine, except in the
graphical plymouth screen that prompts me for my passphrase, I'm not
told which volume this prompt is about (it's not even made clear that
the prompt is about an encrypted storage volume). With initramfs-tools
I was told why I had to type a passphrase.



Now, if I switch to the text interface with ESC, then I see a prompt
that displays this missing information.



This is no big deal for me (hence severity << normal), but this might
be a more severe UX problem for users who need to unlock multiple LUKS
volumes at boot time... especially if the order of the prompts is
even-driven and not deterministic.



FWIW, I've reproduced this problem on a clean Fedora 22 installation.



Any idea if that's a bug in dracut, in plymouth, some integration
issue between them, or a feature that I'm not (yet) able to make
sense of?


This is still reproducible.

I think this is related to the missing fonts. As 
/usr/share/initramfs-tools/hooks/plymouth (from the plymouth package) compared 
to /usr/lib/dracut/modules.d/50plymouth/plymouth-populate-initrd.sh 
installs the dejavu fonts and the fontconfig configuration.


Happy hacking,
--
"Some people worry that artificial intelligence will make us feel inferior, 
but then, anybody in his right mind should have an inferiority complex every 
time he looks at a flower."

-- Alan Kay
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#898556: [kwin-wayland] Closing windows in Plasma Wayland crashes the entire session

[Maximiliano Curia]

Control: severity -1 important

¡Hola Alexander!

El 2018-05-13 a las 17:09 +0300, Alexander Kernozhitsky escribió:

Package: kwin-wayland
Version: 4:5.12.5-1
Severity: grave



I tried Plasma session on Wayland today. Closing a window in it crashes the
session and returns me back to SDDM.


I'm sorry to hear that, but most of the wayland support in Debian is in 
preview release mode, in fact, I think that that problem that you are 
reporting might be in xwayland or in a lower layer of the wayland stack (see 
#897390[1], for example).


I couldn't reproduce the issue in my machine. Is there any other wayland 
desktop session that does work for you?


Given this, I'm lowering the severity.

Happy hacking,

[1]: https://bugs.debian.org/897390
--
"It's a well-known fact that computing devices such as the abacus were 
invented thousands of years ago. But it's not well known that the first use of 
common computer protocol occurred in the Old Testament. This, of course, was 
when Moses aborted the Egyptians' process with a control-sea."

-- Tom Galloway (recs.arts.comics, February 1992)
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#877871: RFP: pyside2 -- Python bindings for Qt5

[Maximiliano Curia]

¡Hola Raphael!

El 2018-05-11 a las 15:33 +0200, Raphael Hertzog escribió:

Thank you. Note that you granted me "developer" access which doesn't let
me create new repositories, so I will need one of you to create the
repository and also to add sophie (sbrun-guest) as member of the
pyside2 repository.


I've created: https://salsa.debian.org/qt-kde-team/qt/pyside2

I've pushed an empty commit, as otherwise, some configurations don't seem to 
stick.


I've also added you as master for the pyside2 repository.

We have not yet decided when to give team wide Master or Admin levels, sorry 
about that, it's in the agenda for our next meeting, but the meeting has no 
date set.



Right, but please take into account that the packages maintained under the
qt tree use a debian directory only packaging branch, and in general, they
use all the same packaging structure (no dpm, no upstream branches or tags
in the public repos, etc). This set of rules is more relaxed in qt-extras
and kde-extras.



Are the rules documented somewhere?


They are somewhat documented in http://pkg-kde.alioth.debian.org/, I would 
say: http://pkg-kde.alioth.debian.org/gitguidelines.html -> but the tagging 
string here is wrong (we used to enforce that tags needed the "$version 
$distribution; urgency=$urgency" format as done by pkgkde-git tag -s), and I 
wouldn't blindly follow the instructions under "Integrating original source 
into local repository".


I'm not particularly fond to this rules as to work on updating them, please 
ask further advice to the qt maintainers (Lisandro and Dmitry).


Happy hacking,
--
"The day Microsoft makes something that doesn't suck, 
is probably the day Microsoft starts making vacuum cleaners."

-- Ernst Jan Plugge
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#877871: RFP: pyside2 -- Python bindings for Qt5

[Maximiliano Curia]

¡Hola Raphael!

El 2018-05-11 a las 09:46 +0200, Raphael Hertzog escribió:

Control: retitle -1 ITP: pyside2 -- Qt for Python
Control: owner -1 sop...@freexian.com



On Sat, 14 Apr 2018 10:09:38 +0200 Francesco Poli  
wrote:

Moreover, if I read the [announcement] correctly, it seems that PySide2
is going to be renamed as "Qt for Python" and adopted as official Qt
bindings for Python...



[announcement]: 




I really hope someone with good Python Debian packaging skills will
soon package pyside2 for inclusion in Debian!



Sophie and me are working to bring pyside2 into Debian. We would like to
put it in pkg-kde-extras. I requested access to the salsa group for this.


Welcome to the team!


Should it go in "Qt" or "Qt extras"? It's going to be something official by
the upstream Qt project so I think that "Qt" might be good.


Right, but please take into account that the packages maintained under the qt 
tree use a debian directory only packaging branch, and in general, they use 
all the same packaging structure (no dpm, no upstream branches or tags in the 
public repos, etc). This set of rules is more relaxed in qt-extras 
and kde-extras.


Also, I don't see in the bug log any comment about the current pyside 
maintenance, are the pyside maintainers ok with moving the new version of the 
project to a different repository in a different team?


Happy hacking,
--
"If a million people believe a foolish thing, it is still a foolish thing."
-- France's Rule of Folly
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#898338: plasma-workspace: Different components hang, high CPU load

[Maximiliano Curia]

¡Hola Dominik!

El 2018-05-10 a las 22:34 +0200, Dominik George escribió:

OK, I found something:



When I kill plasmashell and start it again, in a terminal, I get the
following several times per second:



KActivities: Database can not be opened in WAL mode. Check the SQLite version 
(required >3.7.0). And whether your filesystem supports shared memory
Closing SQL connection:  "kactivities_db_resources_139838218086080_readonly"
KActivities ERROR: There is no database. This probably means that you do not 
have the Activity Manager running, or that something else is broken on your 
system. Recent documents and alike will not work!
KActivities: FATAL ERROR: Failed to contact the activity manager daemon



The last message of these is also filling up mi .xsession-errors, so I
figure it's the same error that is causing the hangs.



-nik


I think that means that the kactivities database is broken, you might want to 
try moving ~/.local/share/kactivitymanagerd.


Happy hacking,
--
"When explaining a command, or language feature, or hardware widget, first 
describe the problem it is designed to solve."

-- David Martin
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#898338: plasma-workspace: Different components hang, high CPU load

[Maximiliano Curia]

¡Hola Dominik!

El 2018-05-10 a las 14:05 +0200, Dominik George escribió:

Package: plasma-workspace
Version: 4:5.12.5-1
Severity: important



-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256



Since one of the three recent updates, the Plasma workspace is close to 
unusable.



* Panel hangs completely, windows in the panel do not change, menu is
  unresponsive, clock is stuck
* Window switcher reacts minutes after pressing Alt-Tab
* Sometimes, some component triggered minutes before start popping up, drawing 
only their
  borders but no content



The plasmashell process is constantly consuming between 100% and 150% of CPU
time, and the Xorg process also uses around 100% when doing nothing special
UI-wise, but only sometimes.



- -- System Information:
Debian Release: buster/sid
 APT prefers unstable-debug
 APT policy: (500, 'unstable-debug'), (500, 'unstable'), (1, 
'experimental-debug')
Architecture: amd64 (x86_64)
Foreign Architectures: i386



Kernel: Linux 4.16.0-1-amd64 (SMP w/4 CPU cores)


You might want to try using a 4.15 kernel till:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=898021
gets fixed. If the issue solves itself using a 4.15 kernel please send a mail 
to this bug with line saying Control: block 898338 with 898021


Happy hacking,
--
"If the code and the comments disagree, then both are probable wrong."
-- Norm Schryer
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#894815: plasma-workspace: Lock screen media controls cannot be disabled

[Maximiliano Curia]

¡Hola Gregor!

El 2018-04-30 a las 12:52 +0200, Gregor Riepl escribió:

In Plasma 5.12.2, a system setting was introduced that allows enabling and
disabling the feature.
In the Debian version of lockscreen however, these settings aren't shown, for
an unknown reason.



I can't reproduce the issue here. I see the Show media controls option in the
Screen Locking / Appearance kcm interface.



That is very strange indeed.
Perhaps something went wrong during the upgrade or wasn't replace properly?



Just to be certain, have you tried restarting the session after upgrading
plasma-workspace?



I have actuall rebooted the machine several times in the meantime, and the UI
is still the same.



Somehow breeze was not included in the generated list, could you please check
and include the output of:
apt policy breeze breeze-cursor-theme breeze-dev kde-style-breeze
kwin-style-breeze qml-module-qtquick-controls-styles-breeze



Here you go:


Well, following the code, it seems that the option is only shown if you have 
org.kde.breeze.desktop or org.kde.breezedark.desktop set as your 
lookandfeelpackage configuration (~/.config/kdeglobals [KDE] LookAndFeelPackage)


Can you please check if your configuration is says so?

Happy hacking,
--
"The day Microsoft makes something that doesn't suck, 
is probably the day Microsoft starts making vacuum cleaners."

-- Ernst Jan Plugge
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#898191: sddm: Does not honor PAM's ambient supplemental groups (via e.g. pam_group.so)

[Maximiliano Curia]

¡Hola Salvatore!

El 2018-05-08 a las 15:14 +0200, Salvatore Bonaccorso escribió:

Package: sddm
Version: 0.14.0-4
Severity: normal
Tags: upstream
Forwarded: https://github.com/sddm/sddm/pull/834



In sddm setups where the sddm pam configuration is configured to use
pam_group to add additional groups on login depending on the
/etc/security/group.conf configuration does not work. This was as well
reported upstream as https://github.com/sddm/sddm/pull/834 .



Upstream fix
https://github.com/sddm/sddm/commit/1bc813d08b8130e458a6550ec47fb2bfbe6de080


It seems to me that the the commit is missing free(buffer) in the first if.


Note, similar configuration worked with kdm under Jessie, so such
setups are suffering from a slight regression when updating from
Jessie to Stretch with kdm (respective later sddm) using pam_group
setup.



Could you consider to update it in unstable and possibly via point
release in stretch? I would be happy to test any backported fix.


I have looked at this commit and after giving it some thought, I don't think 
the issue it's fixing is worth accelerating. I would rather wait for this to be 
released.


I recommend that you talk to upstream in order to get it into the 0.18.0 
release.

If you are planning to use sddm in a large enough environment you might also 
be interested in the patches from Dmitry Torokhov:

- d598c324b589b84e631bd23108f2709a7a8d9634 UserModel: optimize setting of 
default user icon (#999)
- 011020cbd14e302a0b80e9b208379cd1dca0f2b8 UserModel: fix filtering out 
duplicate users (#998)
- 0edc3ffe15f2cdd89da40f4c830da8860bd9a13e UserModel: optimize filtering out 
duplicate users (#995)

Happy hacking,
--
"If it ain't broke, don't fix it" -- Bert Lance

"If we can't fix it, it ain't broke" -- Lieutenant Colonel Walt Weir
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#898092: sddm: takes extremely long time to start

[Maximiliano Curia]

Control: block -1 with 898021

¡Hola Leandro!

El 2018-05-07 a las 03:12 +, Leandro Doctors escribió:

Package: sddm
Version: 0.17.0-1
Severity: important



   * What led up to the situation?



Intalling Plasma Desktop after having problems with several GTK-based
Desktop Environments (may fill bug on that once this one is solved).



   * What exactly did you do (or not do) that was effective (or
 ineffective)?



For now, just started to do some debugging... (I copy the output of
relevant systemd-related commands below.)



It looks as if the program was spending those three minutes trying to find
something else, and it starts only after that time.




   * What was the outcome of this action?



SSDM takes three (3!) minutes to initialize.




   * What outcome did you expect instead?



SSDM to take a couple of seconds to initialize.





-- System Information:
Debian Release: buster/sid
  APT prefers testing
  APT policy: (900, 'testing'), (500, 'unstable')
Architecture: amd64 (x86_64)



Kernel: Linux 4.16.0-1-amd64 (SMP w/8 CPU cores)


It seems that there is a reported bug on 4.16.0 about this:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=898021

You might want to downgrade to 4.15.0 just to be sure that you a are affected 
by the same issue.


Happy hacking,
--
"Programming today is a race between software engineers striving to build 
bigger and better idiot-proof programs, and the Universe trying to produce 
bigger and better idiots. So far, the Universe is winning."

-- Rich Cook
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#897687: libpam-kwallet5: kwallet stopped working

[Maximiliano Curia]

Hi!

As far as I could test the issue it should be solved in the new version, could 
you please check if the issue is also solved for you?


Happy hacking,
--
"Las computadoras son inútiles, solo pueden darte respuestas."
-- Pablo Picasso
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#897668: [breeze-cursor] Severe Visibility Issue

[Maximiliano Curia]

Control: tag -1 + upstream

¡Hola Lumin!

El 2018-05-04 a las 01:35 +, Lumin escribió:

Package: breeze-cursor-theme
Version: 4:5.12.4-1
Severity: important



I'm setting severity to important because the visibility of a cursor
theme is really important.



The problematic cursor is the cross cursor of Breeze and Breeze_Snow,
which are located at the svg files in source tree:



 ./cursors/Breeze_Snow/src/cursors.svg
 ./cursors/Breeze/src/cursors.svg



They look good, but when we actually use the cross cursor in a
low-contrast background it is definitely a pain to find the location of
cursor on the screen.



For example, my cursor theme is Breeze. When I'm taking a screenshot
with flameshot, the whole screen is shadowed, according to source code,
the cursor will be set to the cross. However, unlike other breeze cursors
which are padded with a white margin, the cross cursor is padded with
nothing. In this case the cursor has a very low contrast to the
background hence very hard to caught by eyes.



Similarly, I checked Breeze_Snow. And I guess the cross cursor will also
be hard to find when the background is basically grey or white.



I suggest a fix with high-contrast margin added to the cross cursor.


This seems to be a feature request to be discussed upstream, at least from 
my part, I'm in no position to make statement about a cursor design, and you 
definitely don't want to see my drawing skills. :)


So, please, send this feature request upstream directly using the 
https://bugs.kde.org bugtracker (you'll need to create an account there).


After submitting the request upstream, please also add the link here.

Happy hacking,
--
"Always code as if the person who ends up maintaining your code is a violent 
psychopath who knows where you live."

-- John Woods
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#897221: Session logout doesn't TERMinate all processes

[Maximiliano Curia]

¡Hola Didier!

El 2018-04-30 a las 12:21 +0200, Didier 'OdyX' Raboud escribió:

Package: sddm
Version: 0.17.0-1
Severity: important
Tags: upstream



I routinely run two sessions in parallel (pro & private); and I noticed that
sddm (which I assume is handling the logind session) doesn't coherently
TERMinate (or KILL after some delay) processes from a logged-out session.



Some of these consume ressources (akonadiserver is the usual culprit), but more
importantly; keeping these alive will  break subsequent sessions for the same
user (plasma, akonadi, etc).


I guess akonadiserver is started as a detached process, so ksmserver doesn't 
know about it.


Anyway the problem are the processes that are in the systemd concept of 
session but not in the kde concept of session, those are the ones that systemd 
is waiting for, and ksmserver has long forgotten about them.


Please check loginctl session-status  (The id can be obtained with 
loginctl list-sessions)



Running:
# loginctl terminate-user 
circumvents the issue, but that should really be done by the login manager.



Please tell me if there's something I can do to help here.


Mmh, this is a complex problem. Debian systemd reverts the default logind 
configuration KillUserProcesses, that effectively terminates the session when 
the user logs out, and most desktop environments assume this is set to it's 
default value, so they don't have to deal with the processes started by 
systemd --user, which, by the way, the desktop environment had nothing to do 
with.


Also, kde does nothing to remember which processes were started as autostart 
desktop files.


Gnome "fixes" this by terminating the logind session, as if 
KillUserProcesses was set, at the cost of breaking tmux and screen (which is 
the reason why Debian reverts the KillUserProcesses setting on the first 
place).


So, either we need to fix tmux and screen to they can be detached from the 
logind session, or fix the kde concept of session so it matches better what 
logind expects.


Happy hacking,
--
"I decry the current tendency to seek patents on algorithms. There are better 
ways to earn a living than to prevent other people from making use of one's 
contributions to computer science."

-- Donald Knuth
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#897577: cinnamon-settings-daemon: stop using deprecated libdbus-glib-1-dev

[Maximiliano Curia]

¡Hola Helmut!

El 2018-05-03 a las 07:12 +0200, Helmut Grohne escribió:

Source: cinnamon-settings-daemon
Version: 3.6.2-1
Tags: upstream
User: helm...@debian.org
Usertags: rebootstrap
Control: block 895291 by -1



libdbus-glib-1-dev: is deprecated (#895291). We want to remove it in the
long run, so cinnamon-settings-daemon should stop using it. In
particular, it is presently using dbus-binding-tool and that happens to
not work well for cross compilation (see #873617). Rather than fixing
dbus-glib, the work should be invested in reducing reverse dependencies.
cinnamon-settings-daemon is one of those. Refer to #895291 for
suggestions on how to move forward.


What do you suggests as a replacement of dbus-binding-tool?

Happy hacking,
--
"If you can't write it down in English, you can't code it." -- Peter Halpern
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#897338: Tested on hardware

[Maximiliano Curia]

¡Hola Александр!

El 2018-05-03 a las 07:54 +0300, Александр Керножицкий escribió:

I tried Debian Testing on real hardware and SDDM seems to work there. So I can 
reproduce the bug only on VirtualBox.


It seems that the new sddm requires virtualbox-guest-dkms and 
virtualbox-guest-x11 to run inside a virtualbox, I'm not sure why.


Happy hacking,
--
"We must be very careful when we give advice to younger people: sometimes they 
follow it!"

-- Edsger W. Dijkstra
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#892594: plasma-discover: Invalid Kaddressbook_themes backend

[Maximiliano Curia]

¡Hola Matthias!

El 2018-03-11 a las 10:01 +0100, luca escribió:

Package: plasma-discover
Version: 5.12.2-1
Severity: normal



on start discover return this error:



org.kde.knewstuff.core: Could not find category "KAddressbook Theme"
invalid kns backend! "/etc/xdg/kaddressbook_themes.knsrc" because: "All
categories are missing"
Discarding invalid backend "kaddressbook_themes.knsrc"
qrc:/qml/DiscoverDrawer.qml:145: TypeError: Cannot read property 'name' of null
kns error "/etc/xdg/kaddressbook_themes.knsrc" "Invalid Kaddressbook_themes
backend, contact your distributor."
qml: message: kaddressbook_themes.knsrc: Invalid Kaddressbook_themes backend,
contact your distributor.


Matthias is the expert regarding plasma-discover, but it seems to me that it 
could skip the backend if it's invalid. Something like this:


diff --git a/libdiscover/Category/CategoryModel.cpp 
b/libdiscover/Category/CategoryModel.cpp
index 2e6835b5..b9876938 100644
--- a/libdiscover/Category/CategoryModel.cpp
+++ b/libdiscover/Category/CategoryModel.cpp
@@ -49,6 +49,9 @@ void CategoryModel::populateCategories()
QVector ret;
CategoriesReader cr;
Q_FOREACH (const auto backend, backends) {
+if (!backend->isValid())  {
+continue;
+}
const QVector cats = cr.loadCategoriesFile(backend);

if(ret.isEmpty()) {

I guess this kind of errors could happend by a network hiccup.

Happy hacking,
--
A computer scientist is someone who, when told to "Go to Hell," 
sees the "go to," rather than the destination, as harmful.

Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#897388: Logs accessed files, etc. to syslog

[Maximiliano Curia]

Control: forwarded -1 https://phabricator.kde.org/D12656

I'm not sure if bts link supports phabricator, let's see how this goes.

¡Hola Anthony!

El 2018-05-01 a las 18:01 -0400, Anthony DeRobertis escribió:

Package: kactivitymanagerd
Version: 5.12.1-1
Severity: important



Similar (but nowhere near as bad as) bug #805399, ActivityManager is
logging files I access to the systemd journal & syslog. Some examples:



while hopefully the database itself is in my home director and
mode go-rw, the same can't be said for syslog and journal. This violates
user privacy on a multi-user system as the sysadmin is expected to read
syslog, but respect the privacy of $HOME. In addition, syslog and
journal are available to members of group adm, who may not have root.



From the journal, it appears that kactivymanagerd may be speweing this
to stdout, which is ultimately being picked up by systemd (I think
that's what _TRANSPORT of stdout means):


Upstream already accepted a patch for this, so it would be solved in the next 
release.


Happy hacking,
--
"Seek simplicity, and distrust it." -- Whitehead's Rule
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#897338: sddm: Cannot launch SDDM at boot time and from console

[Maximiliano Curia]

¡Hola Alexander!

El 2018-05-01 a las 17:12 +0300, Alexander Kernozhitsky escribió:

Package: sddm
Version: 0.17.0-1
Severity: grave
Justification: renders package unusable


I tried to do a fresh install of Debian Testing with KDE. After installing 
it, SDDM did not launch. Launching it manually with "sudo systemctl start 
sddm" did not have effect also.



"systemctl status sddm" shows the following:



● sddm.service - Simple Desktop Display Manager
  Loaded: loaded (/lib/systemd/system/sddm.service; indirect; vendor preset: 
enabled)
  Active: active (running) since Tue 2018-05-01 16:56:33 +03; 12min ago
Docs: man:sddm(1)
  man:sddm.conf(5)
 Process: 412 ExecStartPre=/bin/sh -c [ "$(cat /etc/X11/default-display-manager 
2>/dev/
Main PID: 421 (sddm)
   Tasks: 2 (limit: 1156)
  Memory: 11.7M
  CGroup: /system.slice/sddm.service
  └─421 /usr/bin/sddm



Though I don't see the login screen, tty7 is blank.


I tried to rebuild and install sddm from Ubuntu Bionic (where it works), but 
met the same problem. Installing sddm 0.14.0-4 from stable did solve the 
problem.


Is X started at all?

What's the output of `` update-alternatives --query sddm-debian-theme `` ?

It seems that there is a known issue upstream with some nvidia drivers:
https://bugs.launchpad.net/ubuntu/+source/sddm/+bug/1762885
https://github.com/sddm/sddm/issues/1019

Are you using a nvidia card? Can you please check if the problem goes away 
with an older version of the nvidia drivers?


Happy hacking,
--
"Anytime you have a fifty-fifty chance of getting something right, there's a
90 percent probability you'll get it wrong."
-- The 50-50-90 rule
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#873402: Radio: frequency can get reset to preset #1 on startup

[Maximiliano Curia]

¡Hola Ben!

El 2017-08-27 a las 06:59 -0700, Ben Wong escribió:

Package: radio
Version: 3.103-3+b1
Severity: normal
Tags: patch



Dear Maintainer,



The radio program incorrectly changes the frequency on startup under
certain circumstances because it does not correctly calculate the band.



To recreate:



1. Create a file called "~/.radio" with contents like this:



   [Buttons]
   1= 8955
   2= 9035



2. Tune in a frequency that is not one of the presets.
  For example, 'radio -f 94.9'  (then 'q' to quit)



3. Run 'radio' again and note that the frequency gets changed to the
  first preset.



(Note: This bug doesn't occur if the radio is set to a frequency that
is one of the presets).



The problem is that the radio program converts the current frequency
to a band by looking at frequencies that are listed in the presets. It
ought to look at the frequency ranges of the available bands.



I have included a patch which does that.


Sorry that it took me so long to reply your email.

You are reporting an upstream issue, and it seems to me that it would be 
better if you send this kind of patches upstream directly. You can contact 
upstream through the linux-media mailing list: linux-me...@vger.kernel.org


Although the project is no longer in active development the still accept 
patches/fixes.


Happy hacking,
--
"Seek simplicity, and distrust it." -- Whitehead's Rule
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#873397: radio: input frequency uses floating point, causing rounding errors

[Maximiliano Curia]

¡Hola Ben!

El 2017-08-27 a las 06:21 -0700, Ben Wong escribió:

Package: radio
Version: 3.103-3+b1
Severity: normal
Tags: patch



Dear Maintainer,



  To replicate: Run 'radio -f 107.7'
Expected result: Frequency should be 107.7.
 Actual result: Frequency is 107.69.



The 'radio' program uses floating point when reading in the frequency
from the user. (E.g. "107.7"). However, the frequency actually set are
usually wrong due to floating point rounding.



The fix is to read the input as a string and manipulate it as
characters. I've included a patch which does exactly that.


Sorry that it took me so long to reply your email.

You are reporting an upstream issue, and it seems to me that it would be 
better if you send this kind of patches upstream directly. You can contact 
upstream through the linux-media mailing list: linux-me...@vger.kernel.org


Although the project is no longer in active development the still accept 
patches/fixes.


Happy hacking,
--
"People get lost in thought because it is unfamiliar territory."
-- Fix's Principle
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#895747: cinnamon - Missing close button in some themes in some dialog boxes

[Maximiliano Curia]

On 2018-04-15 16:42, Peter Mueller wrote:

Package: cinnamon
Version: 3.6.7-8
Severity: normal



   * What led up to the situation?
Installed Debian Buster/Sid with Cinnamon 3.6.7-8



   * What exactly did you do (or not do) that was effective
 (or ineffective)?
Switching themes.



   * What was the outcome of this action?
If you switch between different "window borders" in the theme
settings, you will find that in some themes (for example in Default,
Greybird, Menta, BlueMenta ...) in some dialog boxes the close button
at the top right of the dialog is missing!
The theme Bluebird contains the Close button.
See also the two attached screenshots to illustrate the problem.


Modal windows don't show a close button, that's just the way things are. 
Modal windows should show a Close or YES/NO button at the bottom.


Also, gtk apps could use client side decorations which override the 
window manager settings, so you might be seeing some windows that are 
designed no to show a titlebar with buttons in it.


If you are really annoyed by this you might want to try giving 
gtk3-nocsd, but personally I consider gtk3-nocsd a hack.


In any case this doesn't look like cinnamon bug but a gtk3 behavior.

Happy hacking,
--
Saludos,
/\/\ /\ >< `/

Bug#895617: Cinnamon (Nemo): Missing extensions/plugins

[Maximiliano Curia]

Control: severity -1 wishlist

On 2018-04-13 16:53, Peter Mueller wrote:

Package: cinnamon
Version: 3.6.7-8
Severity: normal



Dear Maintainer,


*** Reporter, please consider answering these questions, where 
appropriate ***



   * What led up to the situation?
Installed Debian Buster with Cinnamon.



   * What exactly did you do (or not do) that was effective (or
 ineffective)?
Tried to install further Plugins/Extensions for Nemo.



   * What was the outcome of this action?
Unfortunately, only 2 plugins for Nemo where available in Debian, but
Nemo comes with more plugins, e.g.:



nemo-compare: nice support for a compare tool, for example Meld (which
is available in Debian).



 * What outcome did you expect instead?
Nemo Plugins/Extension available in Debian Cinnamon.


It seems to me that you are not really reporting a bug, but requesting 
that we package more nemo extensions. In Debian the procedure for that 
is to send a bug to the wnpp [1][2][3] meta package with the RFP: prefix 
in the subject for each specific package, you can also cc us in the bug 
report if you consider that appropriate. The cinnamon team has no 
intention to package every extension, we don't have the manpower to do 
so, and many extensions are not up to the Debian standards. As rule of 
thumb, if it embeds/forks other well known application/libs, we would 
prefer to do without. On the other hand, we  welcome collaborators 
willing to maintain acceptable extensions under the team umbrella.


Happy hacking,

[1] https://www.debian.org/devel/wnpp/
[2] https://wiki.debian.org/WNPP
[3] https://wnpp.debian.net/
--
Saludos,
/\/\ /\ >< `/

Bug#852170: [pkg-cinnamon] Bug#852170: cinnamon-control-center: Inconsistency in settings for Trackpad (edge-scrolling vs two-finger-scrolling)

[Maximiliano Curia]

¡Hola lopiuh!

El 2017-01-22 a las 11:31 +0100, lopiuh escribió:

Package: cinnamon-control-center
Version: 3.2.1-3
Severity: minor



setting  to true internaly turns  off. But 
 is being able to be true/enabled in the user interface.
Setting  off enables functionality of  
again.



I would expect, that  is disabled when  
is true. Otherwise UI tells something which is not done by the software.
On the other hand: If both is possible independently again in future we would 
have loss of functionality. Best would be, old behaviour would be restored,
that both features could exist next to each other independently.



https://ask.fedoraproject.org/en/question/89443/how-to-enable-touchpad-edge-scrolling/


With the release of cinnamon 3.4 this functionality was rewritten and it's now 
part of cinnamon-common (see cs_mouse.py). The scrolling methods are defined 
as:

clickpad_list = [[0, _("No scrolling")], [3, _("Automatic")], [1, _("Two-finger 
scrolling")], [2, _("Edge scrolling")]]
so you can choose either one or the other. I hope this fixes the issue you 
reported. As far as I understand this, libinput only allows either "Two-finger 
scrolling" or "Edge scrolling", while synaptics accepted switching them 
independently.


Happy hacking,
--
"Programs must be written for people to read, and only incidentally for 
machines to execute."

-- Hal Abelson, "Structure and Interpretation of Computer Programs"
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#894815: plasma-workspace: Lock screen media controls cannot be disabled

[Maximiliano Curia]

¡Hola Gregor!

El 2018-04-04 a las 15:34 +0200, Gregor Riepl escribió:

Package: plasma-workspace
Version: 4:5.12.4-1
Severity: normal



The Plasma lockscreen displays start/stop/forward/backward controls when an
media player application like VLC, Amarok or SMPlayer is active.
It was previously impossible to disable these controls.



In Plasma 5.12.2, a system setting was introduced that allows enabling and
disabling the feature.
In the Debian version of lockscreen however, these settings aren't shown, for
an unknown reason.



The original KDE bug report is here:
https://bugs.kde.org/show_bug.cgi?id=389483
A user confirmed that the setting indeed doesn't appear in Debian and Ubuntu,
but it works fine in Neon and Arch Linux.



Please fix this bug as soon as possible.


I can't reproduce the issue here. I see the Show media controls option in the 
Screen Locking / Appearance kcm interface.


Just to be certain, have you tried restarting the session after upgrading 
plasma-workspace?



Versions of packages plasma-workspace depends on:


Given that the files mentioned in the bug are under a breeze related path, it 
might be possible that breeze is implied in this bug.


Somehow breeze was not included in the generated list, could you please check 
and include the output of:

apt policy breeze breeze-cursor-theme breeze-dev kde-style-breeze 
kwin-style-breeze qml-module-qtquick-controls-styles-breeze

Happy hacking,
--
"If you have too many special cases, you are doing it wrong." -- Craig Zarouni
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#895260: kde-cli-tools: Hard dependency on sudo, which weakens system security

[Maximiliano Curia]

Control: reassign -1 src:kdesu
Control: severity -1 wishlist
Control: retitle -1 Please use the su backend (instead of sudo) by default

On Monday, 9 April 2018 01:58:15 CEST Alex Hvostov wrote:
> Package: kde-cli-tools
> Version: 4:5.10.5-2

> kde-cli-tools 4:5.12.4-1 has a hard dependency on kdesu, which
> indirectly depends on sudo, making it impossible to upgrade KDE without
> creating a serious, unnecessary security risk.

We clearly disagree on considering sudo a security concern. At least, not from 
the kde packaging point of view. I'm downgrading the severity value to 
wishlist.

>From the packaging point of view, the kdesu links against libkf5su5, thus the 
hard dependency, that's not a bug in kde-cli-tools. And in turn libkf5su5 uses 
sudo by default [1]. So, I'm reassigning this bug to src:kdesu.

> Frankly, I consider it a bug that sudo is available in Debian at all.
> Others obviously disagree, but that's no reason to tie unrelated
> packages to it like this.

> Please move kdesu into its own package, and make it optional again.
 
The kdesu tool isn't optional, it's even used by kio to handle certain 
desktop files.

> In the mean time, others with my concern can mitigate this risk by
> neutralizing sudo before installing it. To do that, run the following
> command (as root) before installing sudo:
 
> # dpkg-statoverride --add root root 644 /usr/bin/sudo

Or replacing sudo with a locally equivs generated package, or rebuilding 
libkf5su5 without the sudo dependency and defaulting back to su.

Happy hacking,
[1]: https://salsa.debian.org/qt-kde-team/kde/kdesu/blob/master/debian/
rules#L10
-- 
"Brilliant opportunities are cleverly disguised as insolvable problems."
-- Gardener's Philosophy

"The reverse is also true." -- Corollary
 Saludos /\/\ /\ >< `/


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

Bug#894807: qtbase-opensource-src: Consider recommending qt5ct perhaps in widgets or gui

[Maximiliano Curia]

¡Hola Dmitry!

El 2018-04-04 a las 21:53 +0300, Dmitry Shachnev escribió:
However there is still one major problem with it: it does not support native 
GTK+ 3 dialogs. Test case: run Qt Designer under GNOME with qt5ct and without 
it. If you click Open button, the native GTK+ dialog will be used only when 
qt5ct is not used. This is a significant regression for users of GNOME, Xfce, 
Cinnamon, etc (compared to normal Qt setup).


Interesting, but at the same time you wouldn't have a well integrated gnome 
dialog if the gtk and qt5ct themes don't match. I'm not sure if there is a 
proper way to fix that.


Also, kde applications look "sane" when using qt5ct, and using 
the gtk2 or gtk3 platformtheme some ui annoyances are shown, specially if your 
kde theme doesn't match your gtk theme.


Given this issue, and the fact that it is difficult for an average user to 
disable qt5ct, I think we should not recommend it in Qt. Suggests is fine 
though.


The user should be able to set QT_QPA_PLATFORMTHEME in ~/.xsessionrc.

As Maximiliano mentioned some other things in his IRC message, I will reply 
to them too:



currently nothing depends on qt5-gtk-platformtheme qt5-style-plugins


libqt5gui5 currently recommends qt5-gtk-platformtheme. We cannot depend on 
it because we do not want to depend on GTK+.


qt5-style-plugins is a collection of obsolete styles that are no longer 
maintained or supported by The Qt Company. It can be sometimes safe to use, 
but we do not want to suggest using it.


Ok, then from the cinnamon side of things, I'm fine using the gtk3 
platformtheme by default, and then the user might prefer to use qt5ct. I guess 
we could add a suggests for that.



At least on the cinnamon side of things we are setting QT_PLATFORMTHEME=gtk2


s/_/_QPA_/

There is no sense in doing that (and this is the wrong variable name anyway). 
Qt automatically enables GTK+ integration support when it is run on Cinnamon 
or on other GTK+ based desktop:



https://code.qt.io/cgit/qt/qtbase.git/tree/src/platformsupport/themes/genericunix/qgenericunixthemes.cpp#n863


cinnamon-session upstream already tries to tweak QT_QCA_PLATFORMTHEME, we 
patched that as the expected platformtheme (qt5ct) was not available in 
Debian. I don't remember why we used the gtk2 platformtheme but I'm sure 
that it was the best option at that time.


From my side, this bug could be closed.

Happy hacking,
--
"Don't let what you cannot do interfere with what you can do."
-- Wooden's Rule
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#894725: baloo-kf5: auto-starts on unrelated DE

[Maximiliano Curia]

¡Hola Norbert!

El 2018-04-04 a las 01:01 +0900, Norbert Preining escribió:

(you are everywhere ... ;-)


:)

In Debian, baloo_file is disabled by default (https://salsa.debian.org/qt-kde-team/kde/baloo/blob/master/debian/patches/Disable-the-file-indexer-by-default.patch) 
in order to start it you need to have enabled it. If you do, it seems 
reasonable to have it running.


H I am not aware that I changed anything, but baloo was 
definitely running on my cinnamon DE.



How do I *disable* this setting again?


The configuration file should be in ~/.config/baloofilerc, in Debian, the 
default value is:


[Basic Settings]
Indexing-Enabled=false

This can also be configured using KDE system settings (sytemsettings5) -> 
Workspace -> File Search (or kcmshell5 kcm_baloofile)


I'm not completely sure if a running indexer would notice a change in the 
configuration file, so you might need to signal the baloo_file processes.


Happy hacking,
--
Se necesitan voluntarios para dominar el mundo.
Saludos /\/\ /\ >< `


signature.asc
Description: PGP signature

Bug#894300: Missing dependency: libmpfr-dev (included by libqalculate/Number.h)

[Maximiliano Curia]

Package: libqalculate-dev
Version: 2.2.1-1
Severity: important

Building plasma-workspace against the experimental version of qalculate causes 
plasma-workspace to FTBFS, due to the a missing dependency of libmpfr-dev in 
qalculate's -dev package.


Happy hacking,
--
"A computer program does what you tell it to do, not what you want it to do."
-- Greer's Law
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#892596: [systemsettings] Comes up with black windows

[Maximiliano Curia]

Control: severity -1 important
Control: tag -1 unreproducible

¡Hola David!

El 2018-03-11 a las 11:29 +0200, David Baron escribió:
Package: systemsettings 
Version: 4:5.12.3-1 
Severity: grave


--- Please enter the report below this line. --- 
Previous version and current version:
Application comes up with black windows. The left pane will show its MouseOver 
toolboxes.


I failed to reproduce this issue. Can you still reproduce it? Do you get any 
output in the terminale if you launch systemsettings from it?


Debian Release: buster/sid 
 500 yakkety ppa.launchpad.net 
 500 sid linux.dropbox.com 
 500 lucid   ppa.launchpad.net 


reportbug-ng is not very helpful here.

--- Package information. --- 
Depends(Version) | Installed 
-+- 
qml-module-org-kde-kcm   | 5.42.0-2 
qml-module-org-kde-kirigami2 | 5.42.0-3 
qml-module-qtquick-controls  | 5.9.2-2 
qml-module-qtquick-layouts   | 5.9.2-3 
qml-module-qtquick2  | 5.9.2-3 
kio  | 5.42.0-3 
kpackagetool5| 5.42.0-2 
libc6  (>= 2.14) | 
libkf5activities5(>= 4.96.0) | 
libkf5activitiesstats1 (>= 5.20) | 
libkf5auth5  (>= 4.96.0) | 
libkf5completion5(>= 4.97.0) | 
libkf5configcore5(>= 4.98.0) | 
libkf5configgui5 (>= 4.97.0) | 
libkf5configwidgets5 (>= 4.96.0) | 
libkf5coreaddons5(>= 5.16.0) | 
libkf5crash5 (>= 5.15.0) | 
libkf5dbusaddons5(>= 4.99.0) | 
libkf5declarative5   (>= 4.96.0) | 
libkf5i18n5  (>= 5.17.0) | 
libkf5iconthemes5(>= 4.96.0) | 
libkf5itemviews5 (>= 4.96.0) | 
libkf5kcmutils5  (>= 4.96.0) | 
libkf5khtml5 (>= 4.96.0) | 
libkf5kiowidgets5(>= 4.96.0) | 
libkf5package5(>= 5.5.0+git) | 
libkf5service-bin| 
libkf5service5   (>= 4.99.0) | 
libkf5widgetsaddons5 (>= 5.35.0) | 
libkf5windowsystem5  (>= 4.96.0) | 
libkf5xmlgui5(>= 4.98.0) | 
libqt5core5a (>= 5.9.0~beta) | 
libqt5dbus5  (>= 5.7.0~) | 
libqt5gui5(>= 5.7.0) | 
libqt5qml5(>= 5.0.2) | 
libqt5quick5  (>= 5.0.2) | 
libqt5quickwidgets5   (>= 5.5.0) | 
libqt5widgets5   (>= 5.7.0~) | 
libstdc++6(>= 4.1.1) |


And definitelly not helpful in not showing the installed versions, for most of 
the dependencies. Could you please use reportbug to generate the package 
information again?


Happy hacking,
--
"Fighting patents one by one will never eliminate the danger of software
patents, any more than swatting mosquitoes will eliminate malaria."
-- Richard M. Stallman
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#894218: cinnamon-control-center: really necessary to depend on GOA?

[Maximiliano Curia]

¡Hola Fabio!

El 2018-03-27 a las 17:36 +0200, Fabio Fantoni escribió:

El 2018-03-27 a las 15:19 +0200, Christoph Anton Mitterer escribió:

Package: cinnamon-control-center Version: 3.6.5-1 Severity: wishlist


Is it really necessary to depend on GOA and force people it's 
installation?


Mmh, the panel that requires this is online-accounts, we could 
probably move that panel to a different (optional) package. If you 
prepare a patch for this I could review it.


Additional package for this don't seems a good idea for me, is possible 
instead other way like "load" it only if GOA is installed (if I'm not 
wrong some cinnamon control panel entry already do something similar 
checking the exist of the binary they need) and make it recommends 
instead dep.?


That would need some heavy hacking, in order to load the library that's 
currently being compiled against it would need to load the library as a 
plugin, which is not an elegant solution for something that it's not intended 
to be used that way.


In the Python components this could be done, but again this means a complete 
rewrite of the panel in a different language.


I wouldn't consider that feasible.

Moving the file that links agains goa to a different package, is a packaging 
solution, my offer stands for reviewing a packaging solution. For doing this 
from a different aproach, please contact upstream.


Happy hacking,
--
"First, solve the problem. Then, write the code." -- John Johnson
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#894218: cinnamon-control-center: really necessary to depend on GOA?

[Maximiliano Curia]

¡Hola Christoph!

El 2018-03-27 a las 15:19 +0200, Christoph Anton Mitterer escribió:
Package: cinnamon-control-center 
Version: 3.6.5-1 
Severity: wishlist



Is it really necessary to depend on GOA and force people it's installation?


Mmh, the panel that requires this is online-accounts, we could probably move 
that panel to a different (optional) package. If you prepare a patch for this 
I could review it.


Happy hacking,
--
"Programming today is a race between software engineers striving to build
bigger and better idiot-proof programs, and the Universe trying to produce
bigger and better idiots. So far, the Universe is winning."
-- Rich Cook
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#891499: New mailing lists: debian-cinnamon

[Maximiliano Curia]

Package: lists.debian.org
Severity: wishlist

Dear listmaster,

I would like to migrate pkg-cinnamon-team@l.a.d.o to lists.debian.org, the former 
list is used as a communication point for the Debian Cinnamon team, and as the 
main maintainer address for the packages.


List name: debian-cinnamon
Rationale: The list pkg-cinnamon-team@l.a.d.o is used by the Debian Cinnamon
team as the point of contact, being the main maintainer address, a way to
coordinate with contributions, etc.
Short description: Debian Cinnamon team
Long description: Discussion and coordination among maintainers of Debian's 
Cinnamon.
Category: Developers
Subscription Policy: Open
Post Policy: Open
Web Archive: yes

Happy hacking,
--
"If I ask another professor what he teaches in the introductory programming
course, whether he answers proudly "Pascal" or diffidently "FORTRAN," I know
that he is teaching a grammar, a set of semantic rules, and some finished
algorithms, leaving the students to discover, on their own, some process of
design."
-- Robert W. Floyd
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#890503: RFA: oregano -- tool for schematical capture of electronic circuits

[Maximiliano Curia]

Package: wnpp
Severity: normal

I request an adopter for the oregano package, as I haven't been working on if 
for quite some time now, due to, among other factors, the fact that I haven't 
needed oregano, and that the changes on the upstream side make it diffucult to 
mantain it, the "newer" versions live in https://github.com/drahnr/oregano, 
but it uses waf (https://wiki.debian.org/UpstreamGuide#waf), and it includes 
some non free spice models.


The package description is:
Oregano is a GNOME application for schematic capture and printing of
electronic circuits. It can simulate the circuits using Gnucap,
ng-spice or Berkeley spice.

Happy hacking,
--
"I decry the current tendency to seek patents on algorithms. There are better
ways to earn a living than to prevent other people from making use of one's
contributions to computer science."
-- Donald Knuth
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#887703: kinit: Kinit cannot open/locate libkdeinit5 specific shared object files

[Maximiliano Curia]

¡Hola Andrew!

El 2018-01-19 a las 03:51 -0500, Andrew Lawrence DeMarsh escribió:
Package: kinit 
Version: 5.37.0-2 
Severity: important



Dear Maintainer,



*** Reporter, please consider answering these questions, where appropriate ***

  * What led up to the situation?



a kinit update


The kinit version that you are reporting against was upload in september last 
year, so I guess you have just migrated your system to testing or unstable, if 
that's the case, KDE in general doesn't support upgrades while the session is 
running, you might need to restart your kde session and most likely the 
problem will go away.


If that doesn't fix it:

Re: Bug#887703: kinit: Kinit cannot open/locate libkdeinit5 specific shared object files 
Cannot load library libkdeinit5_minetest: (libkdeinit5_minetest: cannot open shared object file: No such file or directory)


Unless you have created it, there is no libkdeinit5_minetest, so the message 
is correct. minetest (if you have it installed) is in /usr/games/minetest, it 
could be that /usr/games is not in your PATH.


Happy hacking,
--
"If programmers deserve to be rewarded for creating innovative programs, by
the same token they deserve to be punished if they restrict the use of these
programs."
-- Richard M. Stallman
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#873370: kmenuedit: All versions in repositories are extremely out of date.

[Maximiliano Curia]

¡Hola Brice!

El 2017-08-26 a las 21:51 -0600, Brice Hunt escribió:
Package: kmenuedit 
Version: 4:5.8.4-1 
Severity: grave 


Don't abuse the bug severities.


Justification: renders package unusable


The out of date kmenuedit saves configuration files in the wrong directories, causing most menu edits to be lost. 
The newest version in repositories is 4.8.5. The newest version available from upstream is 5.10.5. The version of kmenuedit 
in the repositories does not work with the version of plasma that is available in the repositories.



The recommended solution is to package a more up-to-date version for Debian and 
submit it.


Closing the issue as, it's not really an issue for 4:5.8.4-1 (when using with 
plasma 5.8) nor a current issue with plasma 5.10.


Happy hacking,
--
"By definition, when you are investigating the unknown, you do not know what
you will find"
-- The Ultimate Principle
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#881755: baloo: Baloo can take up all system resources when indexing

[Maximiliano Curia]

¡Hola Leos!

El 2017-11-14 a las 14:54 -0500, Leos Pohl escribió:
Package: baloo 
Version: 4:5.28.0-2 
Severity: important


When baloo indexes files, it takes up all system resources (CPU, memory, 
disk) and the computer becomes unusable until it finishes which can take 
from several minutes to almost half an hour. 
This is erroneous behavious as it equals crashing the system. There need to 
be implemented limitations that baloo has to respect to provide a usable 
indexing. Also the indexing runs not only when the computer 
is idle but also under load. When I get the console running and ask balooctl 
status, I after almost a minute:
Baloo File Indexer is running 
Indexer state: Idle 
Indexed 34283 / 34283 files 
Current size of index is 15.80 GiB



How can it be idle when even the terminal is unusable...


Please maintainer, is there any fix that could be applied to stable debian 
that resolves this issue?


In Debian, the baloo file indexer is disabled by default. As the current baloo 
implementation doesn't scale nicely, (or at all in 32bit systems).


Please consider disabling it or following the issues with the baloo file 
indexer with the upstream developers, as it's infeasible to fix it from the 
Debian side.


Happy hacking,
--
"Don't let what you cannot do interfere with what you can do."
-- Wooden's Rule
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#785274: [git-buildpackage/master] pq: Parse DEP3 headers

[Maximiliano Curia]

tag 785274 pending
thanks

Date:   Sun Sep 11 16:23:11 2016 +0200
Author: Maximiliano Curia <m...@debian.org>
Commit ID: 17a471d1fc07935dd85c31d3a7c4ae3ea5c39208
Commit URL: 
https://git.sigxcpu.org/cgit/git-buildpackage//commit/?id=17a471d1fc07935dd85c31d3a7c4ae3ea5c39208
Patch URL: 
https://git.sigxcpu.org/cgit/git-buildpackage//patch/?id=17a471d1fc07935dd85c31d3a7c4ae3ea5c39208

pq: Parse DEP3 headers

Currently the patch headers in DEP3 format are partially
supported, as git's mailinfo only reads the From and Subject fields from
the first paragraph. But the default in dep3 patches is Description and
Author, that are ignored by git. Even worse, when this fields are in the
first paragraph (again the default) git mailinfo drops all the contained
information.

This patch parses the dep3 headers if git's mailinfo couldn't obtain any
useful information, any header other than Subject|Description and
Author|From is appended to the patch message.

The description field is splitted in first line for the short
description and the rest is prepended to the patch message.

Closes: #785274

  

Bug#879901: kded5 mem ballooning consumes over 6520.81MB of RAM!

[Maximiliano Curia]

Control: severity -1 important

kded acts as a proxy to the background kde services, most likely the leak is 
in one of them, not in kded per se. I'm degrading the severity as I can't 
reproduce it.


¡Hola Nicholas!

El 2017-11-29 a las 05:14 -0500, Nicholas D Steeves escribió:
Please let me know what I can do to help debug this further, because 
I'd like to help solve this asap.


It would be helpful to know which threads kded is executing, in order to do 
that please run the following command: pstree -tp | less, and look for the kded 
processes.


Also it would be interesting to know which services are being managed by kded, 
to find that out you can use the system settings' "Startup and Shutdown" 
"Background services", you can see which services are currently running in you 
session, you can attach your ~/.config/kded5rc that stores this information.


A known culprit on wasting resources is the baloo file indexer (which should 
be disabled by default), I don't remember seeing a leak in kded related to 
baloo, but I haven't paid much attention to it since we disabled it. You can 
check if your file indexer is disabled in: "System Settings" "Search" "File 
Search" "Enable File Search"


Happy hacking,
--
"Any change looks terrible at first." -- Principle of Design Inertia
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#874330: cinnamon-doc: Please make the package multi-arch aware

[Maximiliano Curia]

¡Hola Hugh!

El 2017-09-05 a las 06:27 +, Hugh McMaster escribió:
Package: cinnamon-doc 
Version: 3.2.7-4 
Severity: wishlist


Please mark cinnamon-doc as Multi-Arch: foreign to satisfy the Debian 
duplication check.


cinnamon-doc is an arch:all package, I'm not sure what kind of duplication 
check you are using, but what's the gain on tagging m-a: foreign to arch: all 
packages? Wouldn't it be better to consider arch:all packages as non 
duplicates?


Happy hacking,
--
"The most important thing in the programming language is the name. A language
will not succeed without a good name. I have recently invented a very good
name and now I am looking for a suitable language."
-- Donald Knuth
Saludos /\/\ /\ >< `/


signature.asc
Description: PGP signature

Bug#875001: marked as done ([ktp-call-ui] Future Qt4 removal from Buster)

[Maximiliano Curia]

On Sunday, 10 September 2017 21:55:14 CEST Lisandro Damián Nicanor Pérez Meyer 
wrote:
> On 10 September 2017 at 16:51, Debian Bug Tracking System
>  wrote:
> > -- Forwarded message --
> > From: Pino Toscano 

> > Theoretically this was fixed in experimental already, although it would
> > be nice to know why the telepathy-kde stack was not updated as a whole
> > in unstable, time ago...

> IIRC freeze and then busy maintainers. It would be good to have it in
> unstable.

And the fact that the ktp in experimental is not functional. I'm currently 
updating the packages to 17.08, from what I gathered from the kde ml's it 
seems that it should be functional but in an almost unmaintainable state... In 
the current state, it would make sense to request the removal from unstable, 
and keep it only in experimental till it works again (or is dropped from 
applications).

Happy hacking,
-- 
"The first 90% of the code accounts for the first 90% of the development time.
The remaining 10% of the code accounts for the other 90% of the development
time."
-- Tom Cargill
 Saludos /\/\ /\ >< `/


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