Bug#1032293: [PATCH] Restore HTML DOCTYPE declarations

[Grzegorz Szymaszek]

This reverts commit 504b8f4a8bb8268d0dd94e45fc4936a63e8c4c0c
("Remove DOCTYPE declaration, that makes the rendering not show the events"),
which has removed the DOCTYPE declaration from index.html to enable web
browsers' quirks mode. This way Simile timelines, which have their
height set to 75%, were visible; otherwise, modern browsers would
interpret the 75% as zero.

Restore the DOCTYPE declarations, with a workaround for Simile, which
sets the timeline height to 75vw, which is 75% of the user agent
viewport. This syntax, according to caniuse.com[1], is supported by
major web browser since around 2013, most likely including Internet
Explorer 9.

 [1]: https://caniuse.com/viewport-units
---
 __history__.html  | 1 +
 index.html| 1 +
 media/debian-timeline.css | 5 +
 3 files changed, 7 insertions(+)

diff --git a/__history__.html b/__history__.html
index 0dc101b..8ba582f 100644
--- a/__history__.html
+++ b/__history__.html
@@ -1 +1,2 @@
+
 
diff --git a/index.html b/index.html
index 5fc85a2..6e01a90 100644
--- a/index.html
+++ b/index.html
@@ -1,3 +1,4 @@
+
 
 
 
diff --git a/media/debian-timeline.css b/media/debian-timeline.css
index 718e3b6..849dd72 100644
--- a/media/debian-timeline.css
+++ b/media/debian-timeline.css
@@ -102,6 +102,11 @@ table
 margin: 0.5em 0 0.5em 0;
 }
 
+#the-timeline {
+/* Simile generates height:75%, which seems unsupported in HTML5 */
+height: 75vh !important;
+}
+
 .timeline-default {
 font-family: Trebuchet MS, Helvetica, Arial, sans serif;
 font-size: 1em;
-- 
2.40.1

Bug#884744: gpsd grabs cp2102 usb-serial converter

[Grzegorz Szymaszek]

Hi,

FWIW, here are two upstream issues on this topic, where it appears
gpsd.rules is considered a source of problems:
- https://gitlab.com/gpsd/gpsd/-/issues/187
- https://gitlab.com/gpsd/gpsd/-/issues/235

Here is a recent Ubuntu bug report with some more opinions:
https://bugs.launchpad.net/ubuntu/+source/gpsd/+bug/2012207

And here's my patch that should fix this specific bug by disabling
cp210x hotplugging rules:
https://salsa.debian.org/debian-gps-team/pkg-gpsd/-/merge_requests/14

-- 
Grzegorz Szymaszek


signature.asc
Description: PGP signature

Bug#1032989: marked as done (Liferea 1.14.1-1 segfaults on startup when trying to read gsettings)

[Grzegorz Szymaszek]

Hi,

On Sat, Apr 01, 2023 at 07:57:23PM +0200, Paul Gevers wrote:
> I would appreciate it a lot if you could install the version from
> unstable once [...]

1.14.4-1 from unstable (upgraded from the 1.14.1-2 test build) works
well for me.

Thanks!

-- 
Grzegorz


signature.asc
Description: PGP signature

Bug#1032989: Liferea 1.14.1-1 segfaults on startup when trying to read gsettings

[Grzegorz Szymaszek]

On Thu, Mar 23, 2023 at 07:34:49PM +0100, Paul Gevers wrote:
> On 23-03-2023 18:28, Grzegorz Szymaszek wrote:
> > Would it be possible for you to run another build?
> http://debomatic-amd64.debian.net/distribution#testing/liferea/1.14.1-2/buildlog

Thanks! Well, this one does not crash.


signature.asc
Description: PGP signature

Bug#1032989: Liferea 1.14.1-1 segfaults on startup when trying to read gsettings

[Grzegorz Szymaszek]

Hi,

On Thu, Mar 23, 2023 at 12:41:52PM +0100, Paul Gevers wrote:
> I've went ahead and the build is running now here: 
> http://debomatic-amd64.debian.net/distribution#unstable/liferea/1.14.1-2/buildlog
> 
> Once done, you can pull the liferea-data and liferea deb's from that page
> and install them with $(dpkg -i ).

Unfortunately these new packages depend on
libjavascriptcoregtk-4.0-18 >= 2.40.0, while bookworm only has 2.38.5-1.
Would it be possible for you to run another build? I can do it myself,
but my old SSD is not very happy with this kind of work; it’s not a
problem to wait those 15–20 minutes, I’m just worried about having to
find a new disk relatively soon.

-- 
Grzegorz


signature.asc
Description: PGP signature

Bug#1032989: Liferea 1.14.1-1 segfaults on startup when trying to read gsettings

[Grzegorz Szymaszek]

Hi Paul,

On Fri, Mar 17, 2023 at 09:49:31PM +0100, Paul Gevers wrote:
> It seems upstream is on to something:
> https://github.com/lwindolf/liferea/issues/1212#issuecomment-1472904260

Upstream issue 1212 is now closed and version 1.14.2 is released. I've
built and tested two Liferea Debian packages:
 (1) the same as git tag debian/1.14.1-1,
 (2) the same as above, but with all upstream changes between 1.14.1 and
 1.14.2[1] applied (except commit
 f04dae4ee0f54ac7ec630f35548aed4f31e2c5dc[2] which would cause the
 build to fail) as package patches, so technically version 1.14.2.

Both segfault in the same place as I have originally reported. I think
we can set "forwarded" to issue 1214[3] which seems to more clearly
describe this specific crash.

[1]: https://github.com/lwindolf/liferea/compare/v1.14.1...v1.14.2
[2]: 
https://github.com/lwindolf/liferea/commit/f04dae4ee0f54ac7ec630f35548aed4f31e2c5dc
[3]: https://github.com/lwindolf/liferea/issues/1214

Thanks!

-- 
Grzegorz


signature.asc
Description: PGP signature

Bug#1032989: Liferea 1.14.1-1 segfaults on startup when trying to read gsettings

[Grzegorz Szymaszek]

I'm still trying to find the culprit in gdb. Here's what I've found so
far:
 - feedlist_selection_changed() calls itemlist_unload(), which calls
   itemview_clear(),
   - itemview_clear() sets itemview->needsHTMLViewUpdate = true,
 - feedlist_selection_changed() then calls itemlist_load() which causes
   itemview_update() to be called, and since needsHTMLViewUpdate is
   still true, it calls
   liferea_browser_update(itemview->htmlview, itemview->mode), but
   itemview->htmlview is NULL at this point,
   - liferea_browser_update() calls the conf_get_bool_value macro with
 &(browser->readerMode) as the second argument, but since browser is
 NULL, the browser->readerMode reference causes a segfault.

-- 
Grzegorz


signature.asc
Description: PGP signature

Bug#1032989: Liferea 1.14.1-1 segfaults on startup when trying to read gsettings

[Grzegorz Szymaszek]

Thanks for the response.

On Thu, Mar 16, 2023 at 03:12:24PM +0100, Paul Gevers wrote:
> 1) Could you backup ~/.config/liferea and ~/.local/share/liferea  and
> then test what happens if you remove either or both of these locations?
> [...]

It crashes even with all of ~/.{cache,config,local/share}/liferea
deleted. I did not try resetting gsettings yet, I hope I will be able to
test it tomorrow.

> 2) What desktop engine are you using? I'm on KDE (with X).

Sway 1.7 (Wayland).

-- 
Grzegorz


signature.asc
Description: PGP signature

Bug#1032989: Liferea 1.14.1-1 segfaults on startup when trying to read gsettings

[Grzegorz Szymaszek]

Package: liferea
Version: 1.14.1-1

Dear Maintainer,

After I have upgraded liferea from 1.14.0-1 to 1.14.1-1 today, it no
longer starts, segfaulting when trying to read the "enable-reader-mode"
setting. I have attached (liferea.crash) the result of the following
command:

gdb -batch -n -ex 'set pagination off' -ex run -ex bt -ex 'bt full' -ex 
'thread apply all bt full' --args liferea

from which the short summary is:

Thread 1 "liferea" received signal SIGSEGV, Segmentation fault.
0x55577a6e in conf_get_bool_value_from_schema (gsettings=, gsettings@entry=0x0, key=key@entry=0x555cb33a "enable-reader-mode", 
value=value@entry=0x54) at ../conf.c:258
258 *value = g_settings_get_boolean (gsettings,key);

Let me know if there is more info I can provide to help.

-- 
Grzegorz Szymaszek
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[New Thread 0x7fffebdff6c0 (LWP 202012)]
[New Thread 0x7fffeb5fe6c0 (LWP 202013)]
[New Thread 0x7fffeadfd6c0 (LWP 202014)]
[New Thread 0x7fffea5fc6c0 (LWP 202015)]
[Detaching after fork from child process 202016]
[New Thread 0x7fffdbfff6c0 (LWP 202017)]
[Thread 0x7fffdbfff6c0 (LWP 202017) exited]
[New Thread 0x7fffdbfff6c0 (LWP 202018)]
[New Thread 0x7fffdb7fe6c0 (LWP 202019)]
[Thread 0x7fffdbfff6c0 (LWP 202018) exited]
[Thread 0x7fffdb7fe6c0 (LWP 202019) exited]
[New Thread 0x7fffdb7fe6c0 (LWP 202020)]
[New Thread 0x7fffdbfff6c0 (LWP 202021)]
[Thread 0x7fffdb7fe6c0 (LWP 202020) exited]
[Thread 0x7fffdbfff6c0 (LWP 202021) exited]
[New Thread 0x7fffdbfff6c0 (LWP 202022)]
[New Thread 0x7fffdb7fe6c0 (LWP 202023)]
[Thread 0x7fffdbfff6c0 (LWP 202022) exited]
[Thread 0x7fffdb7fe6c0 (LWP 202023) exited]
[New Thread 0x7fffdb7fe6c0 (LWP 202024)]
[New Thread 0x7fffdbfff6c0 (LWP 202025)]
[Thread 0x7fffdb7fe6c0 (LWP 202024) exited]
[Thread 0x7fffdbfff6c0 (LWP 202025) exited]
[New Thread 0x7fffdbfff6c0 (LWP 202026)]
[New Thread 0x7fffdb7fe6c0 (LWP 202027)]
[Thread 0x7fffdbfff6c0 (LWP 202026) exited]
[Thread 0x7fffdb7fe6c0 (LWP 202027) exited]

Thread 1 "liferea" received signal SIGSEGV, Segmentation fault.
0x55577a6e in conf_get_bool_value_from_schema (gsettings=, gsettings@entry=0x0, key=key@entry=0x555cb33a "enable-reader-mode", 
value=value@entry=0x54) at ../conf.c:258
258 *value = g_settings_get_boolean (gsettings,key);
#0  0x55577a6e in conf_get_bool_value_from_schema (gsettings=, gsettings@entry=0x0, key=key@entry=0x555cb33a "enable-reader-mode", 
value=value@entry=0x54) at ../conf.c:258
#1  0x555a74fd in liferea_browser_update (browser=0x0, mode=1) at 
./src/ui/liferea_browser.c:727
#2  0x555a54e7 in itemview_update () at ./src/ui/itemview.c:297
#3  0x555851cd in itemlist_merge_itemset 
(itemSet=itemSet@entry=0x564b7740) at ../itemlist.c:299
#4  0x55585405 in itemlist_load (node=node@entry=0x55eb2cc0) at 
../itemlist.c:354
#5  0x5558170d in feedlist_selection_changed (obj=, 
nodeId=0x56438d20 "jnwlxys", data=) at ../feedlist.c:585
#9  0x73c8ffd8 in  (instance=0x55b45400, 
detailed_signal=detailed_signal@entry=0x555c04ce "selection-changed") at 
../../../gobject/gsignal.c:3648
#6  0x73c763b0 in g_closure_invoke (closure=0x55fdc420, 
return_value=return_value@entry=0x0, n_param_values=2, 
param_values=param_values@entry=0x7fffb770, 
invocation_hint=invocation_hint@entry=0x7fffb6f0) at 
../../../gobject/gclosure.c:832
#7  0x73c89076 in signal_emit_unlocked_R 
(node=node@entry=0x55840800, detail=detail@entry=0, 
instance=instance@entry=0x55b45400, 
emission_return=emission_return@entry=0x0, 
instance_and_params=instance_and_params@entry=0x7fffb770) at 
../../../gobject/gsignal.c:3796
#8  0x73c8fbf5 in g_signal_emit_valist 
(instance=instance@entry=0x55b45400, signal_id=signal_id@entry=282, 
detail=, var_args=var_args@entry=0x7fffb928) at 
../../../gobject/gsignal.c:3549
#10 0x555a2fdd in feed_list_view_selection_changed_cb 
(selection=, data=) at 
./src/ui/feed_list_view.c:134
#11 0x73c765a9 in _g_closure_invoke_va 
(closure=closure@entry=0x55b726b0, return_value=return_value@entry=0x0, 
instance=instance@entry=0x55719030, args=args@entry=0x7fffbcd0, 
n_params=0, param_types=0x0) at ../../../gobject/gclosure.c:895
#12 0x73c8fbbf in g_signal_emit_valist (instance=0x55719030, 
signal_id=257, detail=, var_args=var_args@entry=0x7fffbcd0) 
at ../../../gobject/gsignal.c:3456
#13 0x73c8fdbf in g_signal_emit (instance=, 
signal_id=, detail=detail@entry=0) at 
../../../gobject/gsignal.c:3606
#14 0x71d3a9ed in _gtk_tree_selection_internal_select_node 
(selection=, node=, tree=, 
path=path@entry=0x563e7850, mode=, 
override_browse_mode=override_browse_mode@entry=0) at 
..

Bug#1032281: nc(1) references nonexistent documentation files

[Grzegorz Szymaszek]

Package: netcat-traditional
Version: 1.10-47

Dear Maintainer,

the netcat-traditional's Debian manual page, nc(1), references two
README files that do not exist: /usr/share/doc/netcat/README.gz and
/usr/share/doc/netcat/README.Debian.gz. The "netcat" part of the path
should be replaced with "netcat-traditional" in order to find the files
actually shipped by the Debian package.


All the best

-- 
Grzegorz


signature.asc
Description: PGP signature

Bug#1023301: Krita 4.4.2 reports support for an invalid MIME type: jpeg/jfif

[Grzegorz Szymaszek]

Package: krita
Version: 1:4.4.2+dfsg-1
Severity: minor

Dear Maintainer,

The Krita 4.4.2 package contains a Desktop Entry file krita_jpeg.desktop
that has "jpeg/jfif" set as one of the supported MIME types. Since it is
not a valid type, it causes update-desktop-database(1) (and most likely
other tools) to report an error.

This is an upstream bug[1], introduced in March 2020[2], that is already
fixed upstream[3] in Krita 4.4.7[4]. Please consider backporting the fix
to the Debian bullseye krita package.

[1]: https://bugs.kde.org/show_bug.cgi?id=424888
[2]: 
https://invent.kde.org/graphics/krita/-/commit/e1735d0654a0fbd761658172dc2d79758d01c2a3
[3]: https://invent.kde.org/graphics/krita/-/merge_requests/663
[4]: 
https://invent.kde.org/graphics/krita/-/commit/5f84290b7ce98c37de589aa7925f27ebfc4353a0


All the best

-- 
Grzegorz


signature.asc
Description: PGP signature

Bug#1012390: Gajim 1.4.0 crashes at startup if gir1.2-gst-plugins-base-1.0 is not installed

[Grzegorz Szymaszek]

On Mon, Jun 06, 2022 at 10:11:37AM +, Martin wrote:
> On 2022-06-06 11:00, Grzegorz Szymaszek wrote:
> > Installing gir1.2-gst-plugins-base-1.0 makes it work again. gajim
> > currently recommends this package, perhaps it should depend on it?
> 
> Yes and it's already done in 1.4.3-1, now in Debian unstable.

Great, thank you!

-- 
Grzegorz


signature.asc
Description: PGP signature

Bug#1012390: Gajim 1.4.0 crashes at startup if gir1.2-gst-plugins-base-1.0 is not installed

[Grzegorz Szymaszek]

Package: gajim
Version: 1.4.0-1

Dear Maintainer,

after upgrading to Gajim 1.4.0, the program no longer starts correctly
and shows the crash reporting dialog instead. Here are the software
component versions listed in the dialog:
- OS: Debian GNU/Linux bookworm/sid testing (bookworm)
- GTK Version: 3.24.34
- PyGObject Version: 3.42.1
- GLib Version : 2.72.0
- python-nbxmpp Version: 3.0.1
- Gajim Version: 1.4.0
And here is the traceback:
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/gajim/gtk/application.py", line 233, 
in _startup
from gajim.gui_interface import Interface
  File "/usr/lib/python3/dist-packages/gajim/gui_interface.py", line 74, in 

from gajim.common.preview import PreviewManager
  File "/usr/lib/python3/dist-packages/gajim/common/preview.py", line 38, 
in 
from gajim.common.preview_helpers import aes_decrypt
  File "/usr/lib/python3/dist-packages/gajim/common/preview_helpers.py", 
line 34, in 
from gi.repository import GstPbutils
  File "/usr/lib/python3/dist-packages/gi/importer.py", line 136, in 
load_module
raise ImportError('cannot import name %s, '
ImportError: cannot import name GstPbutils, introspection typelib not found

Installing gir1.2-gst-plugins-base-1.0 makes it work again. gajim
currently recommends this package, perhaps it should depend on it?


All the best

-- 
Grzegorz


signature.asc
Description: PGP signature

Bug#956440: freecad: Should suggest python3-phply for SCAD import.

[Grzegorz Szymaszek]

On Thu, Oct 21, 2021 at 06:10:38PM +0200, Tobias Frost wrote:
> Well, it Depends: on it already, so I guess this is not sufficient...

It does depend currently, but it did not at the time the bug was
reported:
https://salsa.debian.org/science-team/freecad/-/commit/5b061b9bd107d757eab1c45455365790d2a33715#58ef006ab62b83b4bec5d81fe5b32c3b4c2d1cc2

-- 
Grzegorz


signature.asc
Description: PGP signature

Bug#910504: Long delay until kernel logs “random: crng init done” and allows eg. gdm3 to start

[Grzegorz Szymaszek]

Dear Roger,

I’ve tested this again on the same machine that caused this bug report.
Debian bullseye, upgraded from buster just today; neither rng-tools nor
haveged are installed. GDM (or, in general, any text console or
graphical display manager) starts immediately when broadcom-sta-dkms is
installed, as well as when it is not. From this point of view, the bug
is fixed.

Thanks!

-- 
Grzegorz Szymaszek


signature.asc
Description: PGP signature

Bug#910504: Long delay until kernel logs “random: crng init done” and allows eg. gdm3 to start

[Grzegorz Szymaszek]

Dear Roger,

I am going to upgrade the affected systems to bullseye and will check
this issue while doing so. But looks like it should be solved indeed.

Thanks for reminding me about this bug report.

All the best

-- 
Grzegorz Szymaszek


signature.asc
Description: PGP signature

Bug#989962: shim-signed 1.36~1+deb10u1 fails to boot some systems

[Grzegorz Szymaszek]

On Tue, Jun 22, 2021 at 10:36:48PM +0100, Steve McIntyre wrote:
> Could you please verify if this new build fixes the problem you're
> seeing on your hardware? […] It may still complain about resource
> failures and "import_mok_state() failed", but should then boot anyway in
> non-secure mode.

It works as you’ve described on an E6410.


signature.asc
Description: PGP signature

Bug#989962: shim-signed 1.36~1+deb10u1 fails to boot some systems

[Grzegorz Szymaszek]

On Tue, Jun 22, 2021 at 01:35:51PM +0200, Grzegorz Szymaszek wrote:
> I have recently upgraded several buster amd64 machines; shim-signed went
> up from 1.33 to 1.36~1+deb10u1. […]

FWIW, upgrading to 1.36~1+deb10u2 brings the problem back.

Looks like the same as #990158.


signature.asc
Description: PGP signature

Bug#989962: shim-signed 1.36~1+deb10u1 fails to boot some systems

[Grzegorz Szymaszek]

Hi,

I have recently upgraded several buster amd64 machines; shim-signed went
up from 1.33 to 1.36~1+deb10u1. This has caused some machines to fail to
boot with the mentioned error (“MoklistRT out of resources”).

The affected systems include a Dell Latitude E6410 as well as a E6420,
while a Dell Optiplex 3010 was not affected. All systems are installed
in the UEFI mode and have secure boot disabled/unavailable.

As a workaround I’ve set grubx64.efi as the first boot entry.

-- 
Grzegorz Szymaszek


signature.asc
Description: PGP signature

Bug#989417: mc could recognize foot’s default TERM

[Grzegorz Szymaszek]

Package: mc
Version: 3:4.8.26-1

Dear Maintainer,

foot is a “fast, lightweight and minimalistic Wayland terminal
emulator”[1]. Its Debian package is recommended by the Debian sway (a
Wayland compositor) package[2]. Its default TERM value is “foot”, which
is unrecognized by the mc version packaged in bullseye, and prevents mc
from updating the terminal window title, handling mouse events and key
shortcuts like Shift+Tab (in mc dialogs).

Recently a patch was merged upstream[3] that adds foot to the recognized
TERMs. Please consider backporting it in bullseye.

[1]: https://codeberg.org/dnkl/foot
[2]: https://packages.debian.org/bullseye/sway
[3]: https://midnight-commander.org/ticket/4251

-- 
Grzegorz


signature.asc
Description: PGP signature

Bug#989139: gnome-flashback segfaults on login

[Grzegorz Szymaszek]

Hi Dmitry!

On Sun, May 30, 2021 at 05:14:26PM +0300, Dmitry Shachnev wrote:
> On Fri, May 28, 2021 at 12:12:19AM +0200, Grzegorz Szymaszek wrote:
> > […] the patch fixes it. Is it going to land in bullseye?
> 
> Yes. As it makes gnome-flashback unusable on 32-bit systems, it is a
> release-critical bug, and I will request an unblock in a few days.

Great, thank you very much!

> If you could test the version from unstable and confirm that it works
> fine, it would be nice.

I’ve just tested this version and it works well.

-- 
Grzegorz


signature.asc
Description: PGP signature

Bug#989139: gnome-flashback segfaults on login

[Grzegorz Szymaszek]

Thank you for quick response.

On Wed, May 26, 2021 at 06:52:58PM +0300, Alberts Muktupāvels wrote:
> Maybe this helps?:
> https://gitlab.gnome.org/GNOME/gnome-flashback/-/commit/7676c3a243aad62150b1108dd126132b5f91f8d7

Yep, its the same issue as
 and the
patch fixes it. Is it going to land in bullseye?

-- 
Grzegorz


signature.asc
Description: PGP signature

Bug#989139: gnome-flashback segfaults on login

[Grzegorz Szymaszek]

Package: gnome-session-flashback
Version: 3.38.0-1

When I login to a GNOME Flashback session from GDM3, it fails with a
“something has gone wrong” screen. When I click “Log Out”, I am able to
see the GNOME Flashback desktop, but I get logged out anyway. Relevant
dmesg fragment:

gnome-flashback[…]: segfault at fff0 ip 00486734 sp bf9fd99c error 5 in 
gnome-flashback[47e000+9]
Code: 00 5b 5e c3 66 90 83 c4 04 31 c0 5b 5e c3 8d b4 26 00 00 00 00 90 e8 
84 fb ff ff 05 6f 93 0f 00 8b 54 24 04 8b 80 38 18 00 00 <8b> 04 02 c3 8d b4 26 
00 00 00 00 90 e8 64 fb ff ff 05 4f 93 0f 00

This is an old Intel Atom‐based netbook with Debian bullseye. All other
GNOME sessions—the default Wayland session, the alternative X11 session,
as well as the X11 “Classic” session—work correctly.


signature.asc
Description: PGP signature

Bug#956440: freecad: Should suggest python3-phply for SCAD import.

[Grzegorz Szymaszek]

Hi,

actually, I think it should should suggest python3-ply[1][2].
python3-phply is a “PHP parser written in Python 3 using PLY”[3], not
something FreeCAD would need to import OpenSCAD files.

[1]: https://packages.debian.org/bullseye/python3-ply
[2]: https://wiki.freecadweb.org/OpenSCAD_Module#Dependencies
[3]: https://packages.debian.org/bullseye/python3-phply

-- 
Grzegorz


signature.asc
Description: PGP signature

Bug#950866: Update pyserial homepage

[Grzegorz Szymaszek]

Package: pyserial
Version: 3.4-5.1

Hello,

the homepage of the pyserial project should be updated. The old one[1]
points to the new URL: [2].

Thanks!

[1]: http://pyserial.sourceforge.net/
[2]: https://github.com/pyserial/pyserial

-- 
Grzegorz Szymaszek


signature.asc
Description: PGP signature

Bug#933367: Kdenlive should depend on more qml-module packages

[Grzegorz Szymaszek]

Package: kdenlive
Version: 19.04.3-2

Hello,

Kdenlive should probably depend on the following qml-module packages:
— qml-module-qtgraphicaleffects,
— qml-module-qtqml-models2,
— qml-module-qtquick-controls2,
— qml-module-qtquick-dialogs,
— qml-module-qtquick-layouts,
— qml-module-qtquick-window2.

You can find them by grepping the source code with something like:

grep -Fhr 'import Qt' src | cut -d ' ' -f 2-3 | sort | uniq

Although it may be a rare use case, Kdenlive is unusable due to missing
modules if was installed:
1. on buster, and
2. from Deb Multimedia or sid, and
3. with the “--no-install-recommends” option.

A short explanation of the above requirements:
1. I tested it only on buster (reproducible on the Gnome live CD);
2. Kdenlive 18.12 which ships with buster works correctly;
3. if “install-recommends” is enabled, libkf5newstuffcore5 causes
   qml-module-org-kde-newstuff installation, which in turn pulls the
   dependencies that I’ve listed above (libkf5newstuffcore5 _recommends_
   qml-module-org-kde-newstuff) and Kdenlive works correctly.

See also https://bugs.launchpad.net/ubuntu/+source/kdenlive/+bug/1634478
and Debian bug #822279.

Thanks!

-- 
Grzegorz Szymaszek


signature.asc
Description: PGP signature

Bug#850778: Plasma reports failure when mounting encrypted disks

[Grzegorz Szymaszek]

Perhaps the problem is related to disk encryption? Plasma mounts both 
unencrypted and encrypted drives just fine, but reports failure only in the 
latter case.


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

Bug#910504: Long delay until kernel logs “random: crng init done” and allows eg. gdm3 to start

[Grzegorz Szymaszek]

Package: broadcom-sta-dkms
Version: 6.30.223.271-8~bpo9+2

Dell Latitude E6420 with a Dell Wireless 1530 (Broadcom BCM43228) card.

When this package is installed, there’s a long delay between systemd’s
“Startup finished” message and kernel’s “crng init done” message, around
3–4 minutes. While the SSH server is available immediately (once
connected to WiFi), GDM does not start until the latter message gets
logged.

The problem does not occur if the rng-tools package is installed.

See also .



signature.asc
Description: PGP signature

Bug#876063: fixed in ktexteditor 5.41.0-1

[Grzegorz Szymaszek]

On Wed, Feb 28, 2018 at 11:38:47PM +0100, Daniele Scasciafratte wrote:
> I am checking if kate support them.

Implemented options: charset, end_of_line, indent_size, indent_style,
insert_final_newline, max_line_length, tab_width,
trim_trailing_whitespace.

> looking on the internet seems that the behaviour used to locate and
> load .editorconfig is the same of kateconfig files.

If .kateconfig is found, KTextEditor doesn’t try to find .editorconfig.

> So I don't find a way right now to specify where is the editorconfig or
> what to do specifically.

KTextEditor just calls `editorconfig_parse()`, so its behavior should be
similar to that of other EditorConfig-enabled editors, you may want to
read “Where are these files stored?” (http://editorconfig.org/).

HTH



signature.asc
Description: PGP signature

Bug#890274: Invalid jquery-minicolors installation path

[Grzegorz Szymaszek]

Package: gitea
Version: 1.3.2+dfsg-3

Hello,

In [1], the jquery-minicolors library gets installed into the
`/usr/share/gitea/public/vendor` directory, while it is expected to land
into the `plugins` subdirectory (that’s where the web browser expects it).

Thanks!

[1]: 
https://anonscm.debian.org/cgit/pkg-go/packages/golang-code.gitea-gitea.git/tree/debian/gitea-common.links#n71



signature.asc
Description: PGP signature

Bug#823599: task-polish-desktop: Please add firefox-l10n-pl to package recommendations

[Grzegorz Szymaszek]

It is, but I don't know how to fix it. And I'm not sure if
recommending Firefox-based browsers is enough, maybe it should be
x-www-browser or similar package? task-gnome-desktop recommends
iceweasel - why not epiphany-browser?

2016-05-07 7:11 GMT+02:00 Christian PERRIER :
> Isn't this a more general issue with iceweasel|firefox in desktop
> tasks ?
>
> To Mozilla package maintainers, since the change from iceweasel to
> firefox, isn't there something we should change in tasksel tasks?
>
> (please keep all addresses CC'ed when answering, except /me as I'm
> subscribed to tasksel bugs)
>