gnome-shell issueshttps://gitlab.gnome.org/GNOME/gnome-shell/-/issues2023-10-11T02:49:19Zhttps://gitlab.gnome.org/GNOME/gnome-shell/-/issues/165osk: Pressing 'ë' inputs 'd'2023-10-11T02:49:19ZAdrien Plazasosk: Pressing 'ë' inputs 'd'Secondary keys seem to be broken as they input completely different keys, for example when pressing 'ë' in QWERTY I have 'd' instead.Secondary keys seem to be broken as they input completely different keys, for example when pressing 'ë' in QWERTY I have 'd' instead.GNOME 3.28https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/164osk: Can't input tons of 'e's2023-10-11T02:49:19ZAdrien Plazasosk: Can't input tons of 'e'sWhen holding down a key on the OSK, either this key is inputed many times or a popover with more keys appear. The behavior is inconsistent:
- If I want to check whether a key hides more keys and it doesn't, I end up with tons of times th...When holding down a key on the OSK, either this key is inputed many times or a popover with more keys appear. The behavior is inconsistent:
- If I want to check whether a key hides more keys and it doesn't, I end up with tons of times the same character, which is not good.
- If I want to press many times a key hiding more keys (like 'e'), I can't just hold as it will just make the popover appear, and I have to press it many times.
A way to solve that is by always having a popover appearing when holding a key, this popover would have the secondary keys but also the main one, and by having only keys from a popover to be holdable. This way I know what holding a key will do, I can still input characters many times if needed and I can do it for all character.
If I want to maintain a key pressed, I can afford the extra time of first holding a primary key and then a secondary one for consistency and predictability even if the secondary one is exactly the same as the primary one.GNOME 3.28https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/45Use Cantarell Light prominently2023-12-12T14:08:43ZJakub SteinerUse Cantarell Light prominentlyOne of the great new hotness from Cantarell is new weights, particularly the thin weight useful for large titles. It would be desirable to use it for:
- time for the lock screen
- time and notification popover datesOne of the great new hotness from Cantarell is new weights, particularly the thin weight useful for large titles. It would be desirable to use it for:
- time for the lock screen
- time and notification popover datesGNOME 3.28Jakub SteinerJakub Steinerhttps://gitlab.gnome.org/GNOME/gnome-shell/-/issues/539Crash in 3.30 when switching workspaces2021-02-12T19:04:08ZGeorges Basile Stavracas NetoCrash in 3.30 when switching workspacesI'm using GNOME Shell + Mutter 3.30, and I see a crash that happens mainly when switching workspaces. I do not know if it happens under other circumstances.
Here's the backtrace:
```
#0 0x00007f287578a310 in meta_window_actor_is_destr...I'm using GNOME Shell + Mutter 3.30, and I see a crash that happens mainly when switching workspaces. I do not know if it happens under other circumstances.
Here's the backtrace:
```
#0 0x00007f287578a310 in meta_window_actor_is_destroyed () from /usr/lib/libmutter-3.so.0
[Current thread is 1 (Thread 0x7f2867e4a100 (LWP 893))]
(gdb) bt
#0 0x00007f287578a310 in meta_window_actor_is_destroyed () at /usr/lib/libmutter-3.so.0
#1 0x00007f287333b1c8 in ffi_call_unix64 () at /usr/lib/libffi.so.6
#2 0x00007f287333ac2a in ffi_call () at /usr/lib/libffi.so.6
#3 0x00007f2875a7a7b9 in gjs_invoke_c_function(JSContext*, Function*, JS::HandleObject, JS::HandleValueArray const&, mozilla::Maybe<JS::MutableHandle<JS::Value> >, GIArgument*)
(context=0x559f43d6e530, function=0x559f461ad300, obj=..., args=..., js_rval=..., r_value=0x0) at ../gi/function.cpp:1096
#4 0x00007f2875a7bf36 in function_call(JSContext*, unsigned int, JS::Value*) (context=0x559f43d6e530, js_argc=0, vp=0x559f46157050) at /usr/include/mozjs-60/js/RootingAPI.h:1128
#5 0x00007f2870b4e36d in () at /usr/lib/libmozjs-60.so.0
#6 0x00007f2870b402b0 in () at /usr/lib/libmozjs-60.so.0
#7 0x00007f2870b4dde5 in () at /usr/lib/libmozjs-60.so.0
#8 0x00007f2870b4e31a in () at /usr/lib/libmozjs-60.so.0
#9 0x00007f2870b4ec5f in () at /usr/lib/libmozjs-60.so.0
#10 0x00007f2870f65370 in () at /usr/lib/libmozjs-60.so.0
#11 0x00007f2870b4e1b3 in () at /usr/lib/libmozjs-60.so.0
#12 0x00007f2870b402b0 in () at /usr/lib/libmozjs-60.so.0
#13 0x00007f2870b4dde5 in () at /usr/lib/libmozjs-60.so.0
#14 0x00007f2870b4e31a in () at /usr/lib/libmozjs-60.so.0
#15 0x00007f2870b4ec5f in () at /usr/lib/libmozjs-60.so.0
#16 0x00007f2870f65370 in () at /usr/lib/libmozjs-60.so.0
#17 0x00007f2870b4e1b3 in () at /usr/lib/libmozjs-60.so.0
#18 0x00007f2870b402b0 in () at /usr/lib/libmozjs-60.so.0
#19 0x00007f2870b4dde5 in () at /usr/lib/libmozjs-60.so.0
#20 0x00007f2870b4e31a in () at /usr/lib/libmozjs-60.so.0
#21 0x00007f2870b4ec5f in () at /usr/lib/libmozjs-60.so.0
#22 0x00007f2870f65370 in () at /usr/lib/libmozjs-60.so.0
#23 0x00007f2870b4e1b3 in () at /usr/lib/libmozjs-60.so.0
#24 0x00007f2870b402b0 in () at /usr/lib/libmozjs-60.so.0
#25 0x00007f2870b4dde5 in () at /usr/lib/libmozjs-60.so.0
#26 0x00007f2870b4e31a in () at /usr/lib/libmozjs-60.so.0
#27 0x00007f2870b4ec5f in () at /usr/lib/libmozjs-60.so.0
#28 0x00007f2870f65370 in () at /usr/lib/libmozjs-60.so.0
#29 0x00007f2870b4e1b3 in () at /usr/lib/libmozjs-60.so.0
#30 0x00007f2870b402b0 in () at /usr/lib/libmozjs-60.so.0
#31 0x00007f2870b4dde5 in () at /usr/lib/libmozjs-60.so.0
#32 0x00007f2870b4e31a in () at /usr/lib/libmozjs-60.so.0
#33 0x00007f2870b4ec5f in () at /usr/lib/libmozjs-60.so.0
#34 0x00007f2870f65370 in () at /usr/lib/libmozjs-60.so.0
#35 0x00007f2870b4e1b3 in () at /usr/lib/libmozjs-60.so.0
#36 0x00007f2870b402b0 in () at /usr/lib/libmozjs-60.so.0
#37 0x00007f2870b4dde5 in () at /usr/lib/libmozjs-60.so.0
#38 0x00007f2870b4e31a in () at /usr/lib/libmozjs-60.so.0
#39 0x00007f2870b4ec5f in () at /usr/lib/libmozjs-60.so.0
#40 0x00007f2870f65370 in () at /usr/lib/libmozjs-60.so.0
#41 0x00007f2870b4e1b3 in () at /usr/lib/libmozjs-60.so.0
#42 0x00007f2870b402b0 in () at /usr/lib/libmozjs-60.so.0
#43 0x00007f2870b4dde5 in () at /usr/lib/libmozjs-60.so.0
#44 0x00007f2870b4e31a in () at /usr/lib/libmozjs-60.so.0
#45 0x00007f2870b4ec5f in () at /usr/lib/libmozjs-60.so.0
#46 0x00007f2870e8d818 in JS_CallFunctionValue(JSContext*, JS::Handle<JSObject*>, JS::Handle<JS::Value>, JS::HandleValueArray const&, JS::MutableHandle<JS::Value>) () at /usr/lib/libmozjs-60.so.0
#47 0x00007f2875aa2760 in gjs_call_function_value(JSContext*, JS::HandleObject, JS::HandleValue, JS::HandleValueArray const&, JS::MutableHandleValue) (context=context@entry=0x559f43d6e530, obj=obj@entry=..., fval=fval@entry=..., args=..., rval=rval@entry=...)
at ../gjs/jsapi-util.cpp:633
#48 0x00007f2875a75c55 in gjs_closure_invoke(GClosure*, JS::HandleObject, JS::HandleValueArray const&, JS::MutableHandleValue, bool)
(closure=closure@entry=0x559f44182c30, this_obj=this_obj@entry=..., args=..., retval=retval@entry=..., return_exception=return_exception@entry=false) at /usr/include/mozjs-60/js/RootingAPI.h:1128
#49 0x00007f2875a96c13 in closure_marshal(GClosure*, GValue*, guint, GValue const*, gpointer, gpointer) (closure=0x559f44182c30, return_value=0x0, n_param_values=<optimized out>, param_values=0x7ffd9df9c290, invocation_hint=<optimized out>, marshal_data=<optimized out>)
at /usr/include/mozjs-60/js/RootingAPI.h:493
#50 0x00007f28764023d5 in g_closure_invoke () at /usr/lib/libgobject-2.0.so.0
#51 0x00007f28763ef195 in () at /usr/lib/libgobject-2.0.so.0
#52 0x00007f28763f301e in g_signal_emit_valist () at /usr/lib/libgobject-2.0.so.0
#53 0x00007f28763f3a80 in g_signal_emit () at /usr/lib/libgobject-2.0.so.0
#54 0x00007f2875992de4 in () at /usr/lib/mutter/libmutter-clutter-3.so
#55 0x00007f28759692b1 in () at /usr/lib/mutter/libmutter-clutter-3.so
#56 0x00007f287631b3cf in g_main_context_dispatch () at /usr/lib/libglib-2.0.so.0
#57 0x00007f287631cf89 in () at /usr/lib/libglib-2.0.so.0
#58 0x00007f287631df62 in g_main_loop_run () at /usr/lib/libglib-2.0.so.0
#59 0x00007f287579fc90 in meta_run () at /usr/lib/libmutter-3.so.0
--Type <RET> for more, q to quit, c to continue without paging--
#60 0x0000559f42a313ab in ()
#61 0x00007f2876839223 in __libc_start_main () at /usr/lib/libc.so.6
#62 0x0000559f42a3150e in ()
```GNOME 3.30https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/408Address legibility issues with Top Bar transparency2023-09-17T19:02:04ZJakub SteinerAddress legibility issues with Top Bar transparencyThe [eye-candy drive](https://bugzilla.gnome.org/show_bug.cgi?id=747163) to put transparency into the top bar has introduced legibility issues derived from the fact we don't control the background. A high density detail or very bright wa...The [eye-candy drive](https://bugzilla.gnome.org/show_bug.cgi?id=747163) to put transparency into the top bar has introduced legibility issues derived from the fact we don't control the background. A high density detail or very bright wallpapers will make text and icons difficult to read. There are various approaches to fix this:
1) Revert. Go back to using the top as a negative space outside of the working space. Use black background. Top notch legibility and model.
2) More elaborate metods.
Use Blur to address high frequency detail. @3v1n0 has been working on getting the search results more legible by blurring the high frequency wallpaper detail below. This could work for the top bar context.
Use average lightness sampling to pick foreground color rather than using dropshadow. FIXME: There is prior art.
3) Use a darker overlay for the top bar, but keep the top bar "airy" feeling by using a gradient rather than clip borders of the top bar. FIXME: Find mockups and code.GNOME 3.30https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/1220Wacom Intuos 5 Touch lacks button mapping UI2019-04-30T09:43:40ZJakub SteinerWacom Intuos 5 Touch lacks button mapping UIThe on screen UI is blank for this particular model. You only get the dark overlay, Table name label at the top and the edit button at the bottom. No button map is drawn. Tested in Fedora 30 beta.
![tst](/uploads/6f012a7ee4d5e8091c69aca...The on screen UI is blank for this particular model. You only get the dark overlay, Table name label at the top and the edit button at the bottom. No button map is drawn. Tested in Fedora 30 beta.
![tst](/uploads/6f012a7ee4d5e8091c69acad9db02962/tst.jpg)GNOME 3.32https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/1143Weather: Too little padding2020-01-09T17:28:12ZMattias BengtssonWeather: Too little paddingI think there's too little padding in the weather section of the notification bar.![Screenshot_from_2019-04-05_17-59-05](/uploads/982a866dc4806f75cf639c5c998ea74f/Screenshot_from_2019-04-05_17-59-05.png)
Mostly it's a case of "Göteborg-...I think there's too little padding in the weather section of the notification bar.![Screenshot_from_2019-04-05_17-59-05](/uploads/982a866dc4806f75cf639c5c998ea74f/Screenshot_from_2019-04-05_17-59-05.png)
Mostly it's a case of "Göteborg-Landvetter Airport" being a rather long label, but even the low amount of padding around the specific weather day entries make the whole section feel a bit crowded. At least to me.GNOME 3.32https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/1123gnome-shell fails to resume after suspend2019-03-30T09:30:05ZMichael Hillgnome-shell fails to resume after suspendWith a vm running in Boxes on Rawhide, closing the laptop lid to suspend kills gnome-shell and on resume only GDM is left. Suspend works as expected if Boxes is not running.
```
```Thread 1 "gnome-shell" received signal SIGABRT, Abor...With a vm running in Boxes on Rawhide, closing the laptop lid to suspend kills gnome-shell and on resume only GDM is left. Suspend works as expected if Boxes is not running.
```
```Thread 1 "gnome-shell" received signal SIGABRT, Aborted.
0x00007f30387b0ef5 in raise () from /lib64/libc.so.6
#0 0x00007f30387b0ef5 in raise () from /lib64/libc.so.6
No symbol table info available.
#1 0x00007f303879b8d5 in abort () from /lib64/libc.so.6
No symbol table info available.
#2 0x00007f303955eb53 in g_assertion_message (domain=<optimized out>, file=<optimized out>, line=<optimized out>, func=0x7f3038a8a050 "meta_gpu_kms_flip_crtc", message=<optimized out>) at ../glib/gtestutils.c:2878
lstr = "276\000\374\177\000\000\000D\nC\222\006Z\370\320\316'\335\374\177\000\000`\237\250\070\060\177\000"
s = 0x55e9434a0e40 "0S9C\351U"
#3 0x00007f30395ba90f in g_assertion_message_expr (domain=0x7f3038a72bbf "mutter", file=0x7f3038a89f18 "../src/backends/native/meta-gpu-kms.c", line=276, func=0x7f3038a8a050 "meta_gpu_kms_flip_crtc", expr=<optimized out>) at ../glib/gtestutils.c:2904
s = 0x55e9449e1dd0 "assertion failed: (monitor_manager->power_save_mode == META_POWER_SAVE_ON)"
#4 0x00007f3038a5cd36 in ?? () from /lib64/libmutter-4.so.0
No symbol table info available.
#5 0x00007f3038a61dc1 in ?? () from /lib64/libmutter-4.so.0
No symbol table info available.
#6 0x00007f3039591fa0 in g_main_dispatch (context=0x55e942753f30) at ../glib/gmain.c:3189
dispatch = <optimized out>
prev_source = 0x0
was_in_call = 0
user_data = 0x55e9431aff00
callback = 0x7f3038a61c80
cb_funcs = 0x7f3039664280 <g_source_callback_funcs>
cb_data = 0x7f301c0abe70
need_destroy = <optimized out>
source = 0x55e94393e7a0
current = 0x55e94276c950
i = 1
__FUNCTION__ = "g_main_dispatch"
#7 g_main_context_dispatch (context=context@entry=0x55e942753f30) at ../glib/gmain.c:3854
No locals.
#8 0x00007f3039592338 in g_main_context_iterate (context=0x55e942753f30, block=block@entry=1, dispatch=dispatch@entry=1, self=<optimized out>) at ../glib/gmain.c:3927
max_priority = 0
timeout = 0
some_ready = 1
nfds = <optimized out>
allocated_nfds = 19
fds = 0x55e944b9d860
#9 0x00007f3039592683 in g_main_loop_run (loop=0x55e942a87ee0) at ../glib/gmain.c:4123
self = <optimized out>
__FUNCTION__ = "g_main_loop_run"
#10 0x00007f30389ee040 in meta_run () from /lib64/libmutter-4.so.0
No symbol table info available.
#11 0x000055e940c86b94 in main (argc=<optimized out>, argv=<optimized out>) at ../src/main.c:499
ctx = <optimized out>
error = 0x0
ecode = <optimized out>
'gjs_dumpstack' has unknown return type; cast the call to its declared return type
GNOME 3.32https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/1064alt+esc doesn't work anymore in 3.32: TypeError: this._switcherList is null2019-03-18T09:54:07ZAhmed Khaledalt+esc doesn't work anymore in 3.32: TypeError: this._switcherList is nullPressing *Alt+Esc* doesn't do anything after upgrading to Gnome 3.32 and it prints this error message in the systemd journal when I press it:
```
JS ERROR: TypeError: this._switcherList is null
vfunc_allocate@resource:///org/gnome/shell/...Pressing *Alt+Esc* doesn't do anything after upgrading to Gnome 3.32 and it prints this error message in the systemd journal when I press it:
```
JS ERROR: TypeError: this._switcherList is null
vfunc_allocate@resource:///org/gnome/shell/ui/switcherPopup.js:81:13
```GNOME 3.32https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/1051Big gaps (empty space) in app menus2023-11-21T20:34:03ZAllan DayBig gaps (empty space) in app menusTesting 3.31.92 in Fedora 30, a lot of the apps have a big gap in their app menus:
![Screenshot_from_2019-03-14_11-27-13](/uploads/574a669e80b0d1e998d936d35d820fb6/Screenshot_from_2019-03-14_11-27-13.png)
Examples include Characters, S...Testing 3.31.92 in Fedora 30, a lot of the apps have a big gap in their app menus:
![Screenshot_from_2019-03-14_11-27-13](/uploads/574a669e80b0d1e998d936d35d820fb6/Screenshot_from_2019-03-14_11-27-13.png)
Examples include Characters, Software, Calendar, Contacts. They're all single instance - maybe that has something to do with it.GNOME 3.32https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/968App menu organisation improvements2019-04-17T19:29:09ZAllan DayApp menu organisation improvementsIn 3.31.x the app menu has been replaced with generic items similar to those found in the dash menu. I think we can consider a few improvements to these:
1. Show Details is in the same menu group as "Quit". I think these are different ...In 3.31.x the app menu has been replaced with generic items similar to those found in the dash menu. I think we can consider a few improvements to these:
1. Show Details is in the same menu group as "Quit". I think these are different enough to warrant their own groups.
2. At first I was a bit confused by the first menu item (it was an open window). I think this can be a particular issue if a) there's only one window and b) it has a name that could be read as an action like "Downloads".
For 2, one possibility might be ta add a heading, like "Open Windows". I'm not sure if this is feasible or not, though.GNOME 3.32https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/841sync with gtk:adwaita for 3.322019-06-24T16:24:14ZJakub Steinersync with gtk:adwaita for 3.32There's been a number of changes in adwaita. Once the dust settles we need to fold those changes back to the shell.
- [x] Color updates (primary color scheme is Adwaita:light, overview is toned down (osd style). Prepare for easy Adwait...There's been a number of changes in adwaita. Once the dust settles we need to fold those changes back to the shell.
- [x] Color updates (primary color scheme is Adwaita:light, overview is toned down (osd style). Prepare for easy Adwaita:dark opt in.)
- [x] New switch design
- [x] New rounded menubars, popovers (5x border-raidus as opposed to 3)GNOME 3.32Jakub SteinerJakub Steinerhttps://gitlab.gnome.org/GNOME/gnome-shell/-/issues/811Login screen: remove border around user images2019-03-05T05:04:38ZAllan DayLogin screen: remove border around user imagesThere's [an ongoing initiative to make all the user images round](https://gitlab.gnome.org/GNOME/Initiatives/issues/6), as is the modern convention. The login screen currently places a square border around user images, which doesn't matc...There's [an ongoing initiative to make all the user images round](https://gitlab.gnome.org/GNOME/Initiatives/issues/6), as is the modern convention. The login screen currently places a square border around user images, which doesn't match a round image:
![Screenshot_from_2018-11-20_12-00-29](/uploads/8731071aeebad86f47d4a57bcaf6ecd8/Screenshot_from_2018-11-20_12-00-29.png)
It would probably be best just to remove the border altogether.GNOME 3.32https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/647Alt+tab thumbnails do not show up on primary external monitor2018-10-25T16:21:10ZGhost UserAlt+tab thumbnails do not show up on primary external monitorSwitching between apps functionality does work with alt+tab, however thumbnails do not appear.
When the built-in (laptop) monitor is set as primary, the thumbnails appear normally.
* Version 3.30.1
* Arch linux, 4.14.74-1-lts kernelSwitching between apps functionality does work with alt+tab, however thumbnails do not appear.
When the built-in (laptop) monitor is set as primary, the thumbnails appear normally.
* Version 3.30.1
* Arch linux, 4.14.74-1-lts kernelGNOME 3.32Georges Basile Stavracas NetoGeorges Basile Stavracas Netohttps://gitlab.gnome.org/GNOME/gnome-shell/-/issues/646[regression] Desktop zoom is vertically distorted2019-03-12T00:46:12ZDaniel van Vugtdaniel.van.vugt@canonical.com[regression] Desktop zoom is vertically distortedDesktop zoom is vertically distorted, sometimes.
![Screenshot_from_2018-10-10_18-30-27](/uploads/bb70d643f369dbff30b0619f5b261c1a/Screenshot_from_2018-10-10_18-30-27.png)
Test case:
1. Super+A to to show the app grid.
2. Accessibil...Desktop zoom is vertically distorted, sometimes.
![Screenshot_from_2018-10-10_18-30-27](/uploads/bb70d643f369dbff30b0619f5b261c1a/Screenshot_from_2018-10-10_18-30-27.png)
Test case:
1. Super+A to to show the app grid.
2. Accessibility menu > Zoom = ON
Bisected:
```
f4682748faf466498d3c73961a60099d48dc2254 is the first bad commit
commit f4682748faf466498d3c73961a60099d48dc2254
Author: Georges Basile Stavracas Neto <georges.stavracas@gmail.com>
Date: Sat Jul 7 13:12:33 2018 +0200
layoutManager: Stop using Shell.GenericContainer
This one was remarkably easy to port. In order to make it,
replace the Shell.GenericContainer handlers by a constraint
and simply replace it by a St.Widget.
https://gitlab.gnome.org/GNOME/gnome-shell/merge_requests/153
:040000 040000 b2331095f7c9ae82f4ecc101f3cd1b2e80606714 3eb59c857b343f959acfa8bba39297a69df0e411 M js
```
Note that it's also clipped wrong, but that's a much older bug (mutter#282).GNOME 3.32Georges Basile Stavracas NetoGeorges Basile Stavracas Netohttps://gitlab.gnome.org/GNOME/gnome-shell/-/issues/436Shell Search Provider v32023-11-21T18:53:36ZMarco Trevisanmail@3v1n0.netShell Search Provider v3As per some discussion, one problem that suffers the actual protocol is that, once the overview is closed or the search is stopped by the user, the providers won't be canceled, and they will continue going without being stopped.
Provider...As per some discussion, one problem that suffers the actual protocol is that, once the overview is closed or the search is stopped by the user, the providers won't be canceled, and they will continue going without being stopped.
Providers like nautilus could be affected more by this, or the ones which perform network searches.
A part from various documentation that needs to be added, the basic of the protocol could be something like this ([IRC logs for first discussion](https://paste.ubuntu.com/p/YTyXN7YXvH/)):
```xml
<node name="/" xmlns:doc="http://www.freedesktop.org/dbus/1.0/doc.dtd">
<interface name='org.gnome.Shell.SearchProvider3'>
<method name='Search'>
<arg type='as' name='Terms' direction='in' />
<arg type='o' name='ResultsObject' direction='out' />
</method>
<method name='LaunchSearch'>
<arg type='as' name='Terms' direction='in' />
<arg type='u' name='Timestamp' direction='in' />
</method>
</interface>
<interface name='org.gnome.Shell.SearchProvider3.Result'>
<method name='GetIDs'>
<arg type='as' name='Results' direction='out' />
</method>
<method name='GetMetas'>
<arg type='as' name='ResultsFilter' direction='in' />
<arg type='aa{sv}' name='Metas' direction='out' />
</method>
<method name='Activate'>
<arg type='s' name='Result ID' direction='in' />
<arg type='u' name='Timestamp' direction='in' />
</method>
<method name='Cancel' />
<!-- bool cancelled property? -->
<signal name='ResultsAdded'>
<arg type='as' name='Results' />
</signal>
<signal name='ResultsReady' />
</interface>
</node>
```
So basically, when calling `org.gnome.Shell.SearchProvider3.Search` an object implementing `org.gnome.Shell.SearchProvider3.Result` interface path will be returned and from that we can fetch Result IDs, Meta's or activate it. If the sarch hs not been completed yet, calling `Cancel` should make the provider to stop its action.
`GetSubsearchResultSet` has also been removed.
This is indeed open for discussion, and based on a quick draft, so happy to get inputs on this.GNOME 3.32Marco Trevisanmail@3v1n0.netMarco Trevisanmail@3v1n0.nethttps://gitlab.gnome.org/GNOME/gnome-shell/-/issues/7531Distrobox exported running app icon not showing correctly2024-03-29T05:38:05ZConor DockryDistrobox exported running app icon not showing correctly### Affected version
* OS `openSUSE Aeon/MicroOS 20240327-2826.1`
* GNOME Shell `46.0.1`
* Wayland
### Bug summary
On Aeon after the upgrade to Gnome 46, distrobox exported apps when running do not show the proper icon, rather they sh...### Affected version
* OS `openSUSE Aeon/MicroOS 20240327-2826.1`
* GNOME Shell `46.0.1`
* Wayland
### Bug summary
On Aeon after the upgrade to Gnome 46, distrobox exported apps when running do not show the proper icon, rather they show as the generic placeholder. In the app grid or pinned to dash the correct icon is displayed. But when the app is running it does not mark the pinned icon but displays a separate generic icon in the tray.
Cross posted to [89luca89/distrobox](https://github.com/89luca89/distrobox/issues/1306) and [forums.opensuse.org](https://forums.opensuse.org/t/distrobox-exported-running-app-icon-not-showing/173647)
### Steps to reproduce
Install openSUSE Aeon running Gnome 46
`distrobox enter`\
`sudo zypper in opi`\
`opi vscode`\
`distrobox-export --app code`
Log out, log in.
### What happened
See correct icon displayed in app grid and after pinning to dash.\
Run app, see wrong generic icon in dash tray.
Same with qtcreator and any other GUI app I tried.
### What did you expect to happen
As with Gnome 45, the correct app icon shows when running the app. If the app is pinned to dash, it does not create an extra icon in tray.
### Relevant logs, screenshots, screencasts etc.
Running distrobox exported VSCode
![Screenshot_from_2024-03-23_16-51-22](/uploads/7517e58b3b55bfb6974182c55d29b6eb/Screenshot_from_2024-03-23_16-51-22.png)https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/7530Unprivileged application is able to bypass GNOME Lockscreen2024-03-29T02:03:34ZKyle GospodnetichUnprivileged application is able to bypass GNOME Lockscreen<!--
Please read https://gitlab.gnome.org/GNOME/gnome-shell/-/tree/main#reporting-bugs
first to ensure that you create a clear and specific issue.
-->
Known affected versions:
- OS: Fedora Silverblue 39, Fedora Workstation 39, Bazzite-G...<!--
Please read https://gitlab.gnome.org/GNOME/gnome-shell/-/tree/main#reporting-bugs
first to ensure that you create a clear and specific issue.
-->
Known affected versions:
- OS: Fedora Silverblue 39, Fedora Workstation 39, Bazzite-GNOME 39
- GNOME Shell Version: 45.4
- Appears in: XORG & Wayland
- Happens without extensions: Yes
### Bug summary
This issue was reported by a Bazzite user earlier today. An unprivileged application (joystickwake) that Bazzite includes to prevent the screen from blanking on HTPCs, is able to skip the lock screen entirely. This is also reproducible on stock Fedora installs simply by installing joystickwake.
### Steps to reproduce
1. Install joystickwake (https://github.com/foresto/joystickwake)
2. Connect a controller
3. Lock the screen
4. Press any button or move any stick to skip the lock screen and enter the desktop
#### Alternative
1. Open a Terminal
2. Run the command:
```sleep 10; dbus-send --type=method_call --print-reply --dest=org.gnome.ScreenSaver /org/gnome/ScreenSaver org.gnome.ScreenSaver.SetActive boolean:false```
3. Press <kbd>Super</kbd> + <kbd>L</kbd> to lock the screen
4. Wait up to 10 seconds and the screen unlocks on its own
### More details
Video of the problem can be found at:
https://www.reddit.com/r/gnome/comments/1bq2zl1/bluetooth_controller_can_bypass_gnome/
The source code for this application can be found at:
https://github.com/foresto/joystickwake
This is a security risk since other applications could replicate what this is doing and render the screen lock pointless.
<!-- Do not remove the following line. -->https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/7529GNOME extensions crashing desktop2024-03-28T13:50:17ZMichael BaileyGNOME extensions crashing desktopEverytime I enable extensions my whole GNOME session goes black then signs out, when I log back in I get pushed out again, login again to the extensions being turned off to
OS: Arch Linux x86_64
Host: ThinkPad T560
Kernel: 6.8.2-arch1-1...Everytime I enable extensions my whole GNOME session goes black then signs out, when I log back in I get pushed out again, login again to the extensions being turned off to
OS: Arch Linux x86_64
Host: ThinkPad T560
Kernel: 6.8.2-arch1-1
Uptime: 6 hours, 4 mins
Packages: 909 (pacman), 19 (flatpak)
Shell: bash 5.2.26
Resolution: 1920x1080
DE: GNOME 46.0
WM: Mutter
WM Theme: Adwaita
Theme: Adwaita-dark [GTK2/3]
Icons: Adwaita [GTK2/3]
Terminal: kgx
CPU: Intel i7-6600U (4) @ 3.400GHz
GPU: NVIDIA GeForce 940MX
GPU: Intel Skylake GT2 [HD Graphics 520]
Mmory: 3125MiB / 15853MiB
I was able to switch over to GNOME 46 without any issues but the following update this started happening. there were some extensions that were unsupported after the update so I uninstalled them but still having the same issue every time turn on the extensions
I'm sorry if this is a duplicate issue post I am fairly new to reporting issues on these forums...https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/7524popupMenu.js: submenu inside a submenu won't work2024-03-27T21:52:38ZentepopupMenu.js: submenu inside a submenu won't work### Affected version
Gnome 46 on Arch Linux (up to date)
Wayland
The issue is inside extensions using [popupMenu.js](https://gitlab.gnome.org/GNOME/gnome-shell/-/blob/main/js/ui/popupMenu.js)
### Bug summary
Using popupMenu.js create ...### Affected version
Gnome 46 on Arch Linux (up to date)
Wayland
The issue is inside extensions using [popupMenu.js](https://gitlab.gnome.org/GNOME/gnome-shell/-/blob/main/js/ui/popupMenu.js)
### Bug summary
Using popupMenu.js create a PopupSubMenuMenuItem "inner" inside a PopupSubMenuMenuItem "outer". If you try to open the "inner" PopupSubMenuMenuItem, the "outer" PopupSubMenuMenuItem will close and thus you won't be able to use any menu item inside the "inner" PopupSubMenuMenuItem. Once you again open the "outer" PopupSubMenuMenuItem this will close the "inner" PopupSubMenuMenuItem.
I believe the issue is located in line 902 (`_setOpenedSubMenu`) in combination with line 1272 `_subMenuOpenStateChanged(menu, open)`: if a submenu opens (e.g. "inner") it goes up to the top hierarchy menu (line 1272: `this._getTopMenu()`) and announces itself as "open". The top menu unfortunately will see if there is already an open menu (e.g. "outer") and close it (line 903/904).
Keeping a single submenu open is probably because you only want a single menu in a single hierarchy open. unfortunately you prevent nested submenus. To fix it, you need to implement a more complex logic. It may be easier to just put nested submenu out of scope.
<!-- Do not remove the following line. -->