GNOME issueshttps://gitlab.gnome.org/groups/GNOME/-/issues2023-11-08T11:42:33Zhttps://gitlab.gnome.org/GNOME/console/-/issues/126Allow customizing keyboard shortcuts2023-11-08T11:42:33ZJeremy BichaAllow customizing keyboard shortcutsWhen you add some basic preferences, please also add some keyboard shortcuts.
Usecase 1
---------
The Alt+number shortcuts to switch between tabs may conflict with keyboard shortcuts used by a CLI app.
Usecase 2
---------
I set a keybo...When you add some basic preferences, please also add some keyboard shortcuts.
Usecase 1
---------
The Alt+number shortcuts to switch between tabs may conflict with keyboard shortcuts used by a CLI app.
Usecase 2
---------
I set a keyboard shortcut for `Reset and clear` in GNOME Terminal. Since I also have infinite scrollback, this allows me to clean up my scrollback while I'm running a long command (like a project build). Otherwise, I guess I'd need to background the current process, then run a reset, then resume the process to get the same effect.https://gitlab.gnome.org/GNOME/file-roller/-/issues/153Broken rar file extraction2022-03-30T22:35:52ZlinsuiBroken rar file extractionI created a file with winrar. File roller can unzip it but the extracted files are broken. However, bsdtar does works well. Since file roller uses libarchive, this is weird.
This file is created with winrar with default options.
[This_...I created a file with winrar. File roller can unzip it but the extracted files are broken. However, bsdtar does works well. Since file roller uses libarchive, this is weird.
This file is created with winrar with default options.
[This_is_a_test_file.rar](/uploads/ccdb57598eb1765ec63c310249fc402b/This_is_a_test_file.rar)https://gitlab.gnome.org/GNOME/libxml2/-/issues/361Incorrect line number reported if higher than 65535 in some cases2022-09-02T17:52:28ZJosé M. SánchezIncorrect line number reported if higher than 65535 in some caseslibxml2 internally stores the line number of the nodes as a two bytes unsigned integer, but has the possibility to set in parser context options XML_PARSE_BIG_LINES and in that case the line number is stored in text PSVI field. This happ...libxml2 internally stores the line number of the nodes as a two bytes unsigned integer, but has the possibility to set in parser context options XML_PARSE_BIG_LINES and in that case the line number is stored in text PSVI field. This happens only for text nodes, but not in other places.
I have created this patch which enhances tree.c:xmlGetLineNoInternal and SAX2.c:xmlSAX2StartElementNs so that if XML_PARSE_BIG_LINES is set then the line number is provided just like for text nodes.
If this patch is correct, could it be integrated into future libxml2 releases?
[Incorrect-line-number-reported-if-higher-than-65535.patch](/uploads/197f156f725d6e1a48795260c83dde8a/Incorrect-line-number-reported-if-higher-than-65535.patch)https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/5270Shell keyboard navigation (after pressing Ctrl+Alt+Tab) doesn't work in Xorg ...2023-06-09T08:52:06ZJeremy LincicomeShell keyboard navigation (after pressing Ctrl+Alt+Tab) doesn't work in Xorg sessionsOriginal bug filed at <https://bugs.launchpad.net/ubuntu/+source/gnome-shell/+bug/1965696>:
>>>
As of [2022-03-18], I can no longer quickly get to system controls such as the Top Bar using the keyboard. According to the keyboard shortcu...Original bug filed at <https://bugs.launchpad.net/ubuntu/+source/gnome-shell/+bug/1965696>:
>>>
As of [2022-03-18], I can no longer quickly get to system controls such as the Top Bar using the keyboard. According to the keyboard shortcuts settings window, I should be able to press "CTRL+Alt+Tab" to access these controls. When I press this key combination, I'm told by the Orca screen reader that I'm on the Top Bar. When I release the keys, focus jumps back to the window I was in before. I do not know if this is the case without Orca running. I'm unable to test without Orca running. I've reset Gnome settings, with no change.
>>>
[edited to copy the summary from the Launchpad bug -smcv]https://gitlab.gnome.org/GNOME/gtk/-/issues/4800infinite loop with label in list box containing æſð2023-05-03T00:20:18ZR. N.infinite loop with label in list box containing æſð**Version:** `42.0-1-g985c8dd` (installed via Flatpak on Debian 11)
**Steps to reproduce:**
1. Start GNOME Text Editor
2. Paste these three characters: `æſð`
3. Close the editor
The editor then hangs until GNOME offers to force close it.**Version:** `42.0-1-g985c8dd` (installed via Flatpak on Debian 11)
**Steps to reproduce:**
1. Start GNOME Text Editor
2. Paste these three characters: `æſð`
3. Close the editor
The editor then hangs until GNOME offers to force close it.https://gitlab.gnome.org/GNOME/gnome-text-editor/-/issues/349Zoom buttons should not use `list-add`/`list-remove` icons2022-03-29T10:51:41ZJohn FactotumZoom buttons should not use `list-add`/`list-remove` iconsCurrently, `list-add-symbolic` and `list-remove-symbolic` are used for the icons for zoom in and out buttons.
Semantically, they should be `zoom-in-symbolic`/`zoom-out-symbolic`, or `value-increase-symbolic`/`value-decrease-symbolic`.Currently, `list-add-symbolic` and `list-remove-symbolic` are used for the icons for zoom in and out buttons.
Semantically, they should be `zoom-in-symbolic`/`zoom-out-symbolic`, or `value-increase-symbolic`/`value-decrease-symbolic`.https://gitlab.gnome.org/GNOME/console/-/issues/125Button arrangement2023-08-05T06:22:54ZdikaspButton arrangementShould console follow common button layout found in gnome apps ?
I mean in appplication window, new (+) button should be placed in leftmost and search button usually placed in rightmost near menu and close button
apps that uses this la...Should console follow common button layout found in gnome apps ?
I mean in appplication window, new (+) button should be placed in leftmost and search button usually placed in rightmost near menu and close button
apps that uses this layout: previous gnome terminal, gedit, evince, nautilus, gnome text editor, etc ...https://gitlab.gnome.org/GNOME/connections/-/issues/96"Windows" keyword is translated inconsistently2022-03-29T12:45:32ZWill Thompson"Windows" keyword is translated inconsistentlyHere is an excerpt from Connections (42.0 from Flathub)'s `.desktop` file:
```ini
Keywords=vnc;rdp;remote;desktop;windows;support;access;view
# ^^^^^^^
Keywords[fr]=vnc;rdp;distant;distance;bureau;windows;p...Here is an excerpt from Connections (42.0 from Flathub)'s `.desktop` file:
```ini
Keywords=vnc;rdp;remote;desktop;windows;support;access;view
# ^^^^^^^
Keywords[fr]=vnc;rdp;distant;distance;bureau;windows;prise en main;prise en charge;support;accès;visualisation
# ^^^^^^^
Keywords[es]=vnc;rdp;remoto;escritorio;ventanas;soporte;acceso;ver
# ^^^^^^^^
Keywords[pt]=vnc;rdp;remoto;desktop;janelas;suporte;acesso;visualização
# ^^^^^^^
```
I believe the keyword "windows" is supposed to refer to the Microsoft operating system brand, rather than the metaphor used to represent open applications by all major graphical desktops. I believe based on looking at Wikipedia in each language that the OS is called "Windows" in French, Spanish & Portuguese.
The French translation agrees with my interpretation and leaves the brand untranslated. (I believe the metaphor, like the real thing on the side of a house, is called « une fenêtre ».)
The Spanish & Portuguese translations use the other interpretation, and translate the keyword as the metaphor/hole in a house.https://gitlab.gnome.org/GNOME/gnome-control-center/-/issues/1730Accessing colors settings crashes the gnome-control-center2022-06-06T07:53:46ZAfonso BarrachaAccessing colors settings crashes the gnome-control-center<!--
Not following the communication guidelines [1] will mean your issue or comment
will be removed. Read it carefully before submitting this issue.
[1] https://gitlab.gnome.org/GNOME/gnome-control-center/blob/master/docs/CONTRIBUTING...<!--
Not following the communication guidelines [1] will mean your issue or comment
will be removed. Read it carefully before submitting this issue.
[1] https://gitlab.gnome.org/GNOME/gnome-control-center/blob/master/docs/CONTRIBUTING.md#communication-guideline
-->
Detailed description of the issue. Put as much information as you can, potentially
with images showing the issue.
Steps to reproduce:
1. Open GNOME Settings on Gnome 42 on Fedora 36.
2. Go to Colors.
3. Settings Crashes.
I know this error is caused by a service of gds-color but I can't find which. To get to color settings without the settings crashing I have to do the following commands to access it:
```bash
~ killall gsd-color
~ nohup /usr/libexec/gsd-color &
```https://gitlab.gnome.org/GNOME/libxslt/-/issues/65Please support symbol versioning with CMake2022-03-29T17:30:39ZYasuhiro KimuraPlease support symbol versioning with CMakelibxslt supports symbol versioning. But it is only with GNU Autotools and not with CMake. So please support it with the latter.libxslt supports symbol versioning. But it is only with GNU Autotools and not with CMake. So please support it with the latter.https://gitlab.gnome.org/GNOME/gnome-contacts/-/issues/237GTK4 regression on narrow screens2022-07-24T17:28:00ZPablo Correa GomezGTK4 regression on narrow screensI have been trying GNOME contacts for update to 42 in alpine/postmarketOS, and I have found that the contact view does no longer fit in very narrow screens. Probably screenshots explain it better than me:
<details><summary>GNOME 42.0 b...I have been trying GNOME contacts for update to 42 in alpine/postmarketOS, and I have found that the contact view does no longer fit in very narrow screens. Probably screenshots explain it better than me:
<details><summary>GNOME 42.0 build</summary>
![contacts-42](/uploads/485b7dfbc4ee921143bcb594cbd9c9bd/contacts-42.png)
</details>
<details><summary>GNOME 41.0 build with !140 applied</summary>
![contacts-41](/uploads/f4dae2dab094ac68f3ec37b9384c3c9f/contacts-41.png)
</details>
For GNOME 41.0, I could also send a screenshot without !140 if that were needed, but that doesn't change the problem. Also, the same problem where some parts of the app on the right do not fit happens when editing the contact.
@bertob I have seen that you have been the one doing UI papercuts and reviews here. I could likely send a patch to fix it, but would like to know where to trim space. If it should be done elipsizing strings, removing some padding on the sides or with some other UX approach.
Just for completion, that is a pinephone with Phosh at 200% scale (360px screen width) and postmarketos edge.
And BTW, the GTK4 port looks amazing, I love it!
<!-- Ignore the text under this line. -->https://gitlab.gnome.org/GNOME/libxml2/-/issues/360Please support symbol versioning with CMake2022-09-04T00:46:23ZYasuhiro KimuraPlease support symbol versioning with CMakelibxml2 supports symbol versioning. But it is only with GNU Autotools and not with CMake. So please support it with the latter.libxml2 supports symbol versioning. But it is only with GNU Autotools and not with CMake. So please support it with the latter.https://gitlab.gnome.org/GNOME/gnome-text-editor/-/issues/348Extensions2022-03-29T10:57:20ZJohn Erling BladExtensionsWhen (if at all) would you guys provide an extension API if not the same extensions? For me lack of extensions are a showstopper, and I would have to use [gedit](https://flathub.org/apps/details/org.gnome.gedit) anyhow.
I would expect a...When (if at all) would you guys provide an extension API if not the same extensions? For me lack of extensions are a showstopper, and I would have to use [gedit](https://flathub.org/apps/details/org.gnome.gedit) anyhow.
I would expect an editor to replace gedit would have the same core features, but it seems like this editor simply tries to be a visually pleasing replacement without providing the same feature sethttps://gitlab.gnome.org/GNOME/gnome-boxes-logos/-/issues/3Rename development branch to "main"2022-03-29T09:01:30ZFelipe Borgesfelipeborges@gnome.orgRename development branch to "main"This repository is used for things that GNOME Boxes hotlinks to (logos and screenshots). Therefore these permalinks shouldn't change otherwise we would be breaking old stable releases.
What can be done is to create a new branch and lock...This repository is used for things that GNOME Boxes hotlinks to (logos and screenshots). Therefore these permalinks shouldn't change otherwise we would be breaking old stable releases.
What can be done is to create a new branch and lock the "master" one for merges and pushes until all stable releases after this date are EOL.https://gitlab.gnome.org/GNOME/totem/-/issues/513Fix "Totem could not" error messages2022-03-31T12:51:01ZBastien NoceraFix "Totem could not" error messagesThey shouldn't mention "Totem", "Videos" is the name of the app.They shouldn't mention "Totem", "Videos" is the name of the app.https://gitlab.gnome.org/GNOME/gnome-software/-/issues/1700Occasional 'Gs failed to load metadata: cancelled by user action' runtime wa...2022-03-29T11:48:56ZMilan CrhaOccasional 'Gs failed to load metadata: cancelled by user action' runtime warning on consoleThe warning itself is related to the PackageKit plugin, printed from:
```
by g_log() at gmessages.c:1494
by gs_loading_page_refresh_cb() at gs-loading-page.c:89
by g_task_return_now() at gtask.c:1224
by g_task_return() at...The warning itself is related to the PackageKit plugin, printed from:
```
by g_log() at gmessages.c:1494
by gs_loading_page_refresh_cb() at gs-loading-page.c:89
by g_task_return_now() at gtask.c:1224
by g_task_return() at gtask.c:1290
by run_job_cb() at gs-plugin-loader.c:3656
by g_task_return_now() at gtask.c:1224
by g_task_return() at gtask.c:1290
by finish_op() at gs-plugin-job-refresh-metadata.c:418
by plugin_refresh_metadata_cb() at gs-plugin-job-refresh-metadata.c:373
by g_task_return_now() at gtask.c:1224
by g_task_return() at gtask.c:1290
by refresh_metadata_cb() at gs-plugin-packagekit.c:3764
by g_simple_async_result_complete() at gsimpleasyncresult.c:805
by complete_in_idle_cb() at gsimpleasyncresult.c:817
by g_idle_dispatch() at gmain.c:5912
by g_main_context_dispatch() at gmain.c:3394
by g_main_context_iterate.constprop.0() at gmain.c:4188
by g_main_context_iteration() at gmain.c:4254
by g_application_run() at gapplication.c:2564
by main() at gs-main.c:49
```
The `cancelled by user action` comes from the libdnf. Tracking this even closer, the PackageKit cancels the operation on its own:
```
09:26:54 PackageKit adding transaction 0x1a63190
09:26:54 PackageKit sending object path: '/1_cecacdbd'
09:26:54 PackageKit notify::connected
09:26:54 PackageKit SetHints method called: locale=en_US.UTF-8, background=true, interactive=false, cache-age=4294967295
09:26:54 PackageKit locale changed to en_US.UTF-8
09:26:54 PackageKit cache-age changed to 4294967295
09:26:54 PackageKit RefreshCache method called: 0
09:26:54 PackageKit transaction now waiting-for-auth
09:26:54 PackageKit authorizing action org.freedesktop.packagekit.system-sources-refresh
09:26:54 PackageKit transaction now ready
09:26:54 PackageKit changing transaction to exclusive mode
09:26:54 PackageKit 1 transactions in list, 1 committed but not finished
09:26:54 PackageKit transaction now running
09:26:54 PackageKit install uid now 1000
09:26:54 PackageKit install cmdline now PackageKit: refresh-cache
09:26:54 PackageKit setting role for /1_cecacdbd to refresh-cache
09:26:54 PackageKit-DNF got state loading-cache with hint (null)
09:26:54 PackageKit CreateTransaction method called
09:26:54 PackageKit job count now 2
09:26:54 PackageKit trying to open database '/var/lib/PackageKit/transactions.db'
09:26:54 PackageKit job count is now at 0
09:26:54 PackageKit transaction now new
09:26:54 PackageKit setting sender to :1.247
09:26:54 PackageKit adding transaction 0x1a63310
09:26:54 PackageKit sending object path: '/2_eaccbabc'
09:26:54 PackageKit SetHints method called: locale=en_US.UTF-8, background=false, interactive=false, cache-age=4294967295
09:26:54 PackageKit locale changed to en_US.UTF-8
09:26:54 PackageKit cache-age changed to 4294967295
09:26:54 PackageKit Resolve method called: 327680, glibc-all-langpacks&glibc-common&glibc-gconv-extra&....
09:26:54 PackageKit transaction now ready
09:26:54 PackageKit changing transaction to exclusive mode
09:26:54 PackageKit 2 transactions in list, 2 committed but not finished
09:26:54 PackageKit cancelling running background transactions and instead running /2_eaccbabc
09:26:54 PackageKit cancelling running background transaction /1_cecacdbd
09:26:54 PackageKit CancelBg method called on /1_cecacdbd
...
09:26:54 PackageKit emitting allow-cancel 0
09:26:54 PackageKit 2 transactions in list, 2 committed but not finished
09:26:54 PackageKit emitting error-code transaction-cancelled, 'cancelled by user action'
09:26:54 PackageKit emitting allow-cancel 1
09:26:54 PackageKit 2 transactions in list, 2 committed but not finished
09:26:54 PackageKit backend was running for 307 ms
09:26:54 PackageKit emitting finished 'cancelled-priority', 307
09:26:54 PackageKit transaction now finished
09:26:54 PackageKit running /2_eaccbabc as previous one finished
09:26:54 PackageKit transaction now running
09:26:54 PackageKit 2 transactions in list, 1 committed but not finished
09:26:54 PackageKit install uid now 1000
09:26:54 PackageKit install cmdline now PackageKit: resolve
09:26:54 PackageKit setting role for /2_eaccbabc to resolve
09:26:54 PackageKit-DNF got state query with hint (null)
```
The `/1_cecacdbd` is a background operation to refresh cache. The following `/2_eaccbabc` is a foreground operation, which cancels the running background operation - it's cancelled by the PackageKit itself.
I do not know what to do with it. Either the cache refresh cannot be set as a background operation, or the next op can run only after the first is finished, but that won't fix a case where another application runs a transaction while the background transaction is running.https://gitlab.gnome.org/GNOME/gnumeric/-/issues/625go_dtoa imprecise for last digit of 'long' values?2023-01-19T12:05:33Zb. s.go_dtoa imprecise for last digit of 'long' values?( disclaimer as a 'newbie': it's at least as likely that I have built in a bug somewhere, as that gnumeric or go_dtoa have one, but I have to ask to move things forward. )
in a bigger project I - mean to - have spotted a little inaccur...( disclaimer as a 'newbie': it's at least as likely that I have built in a bug somewhere, as that gnumeric or go_dtoa have one, but I have to ask to move things forward. )
in a bigger project I - mean to - have spotted a little inaccuracy of go_dtoa for long doubles when using 'shortest'.
( "!" GNM_FORMAT_g, IMHO that's what is used for cell edit display and write to files with 'conventions '-1''? )
can boil it down to a simple reproducer:
```
16470255126.0330839185 and
16470255126.0330839194 are distinct FP-values with 80-bit IEEE 754 'long doubles', they have a difference of
~9.313225746154785156E-10 gnumeric can calculate that but assigns the string
16470255126.033083919 to both for 'cell edit', file save with option 'shortest', or when I implement a
function using go_dtoa with option 'shortest'.
```
Ulf Adam's 'ryu' stays distinctive and produces
```
1.64702551260330839185E10 and
1.6470255126033083919E10 I think MSVC and gcc have adopted ryu for 'to_chars' ( or similar ) shortly,
but evtl. only for C++ usage?.
```
( it is not! a matter if I really need that accuracy, it is! a question of whether or not the building blocks of a system meet the specifications they state. the more we let go with errors the better we fulfill the IT mantra 'floating point math is inaccurate', but that's not what gnumeric set out for? For the UI it's a question to 'get what you see', editing two cells to the same string and then having a difference between them is 'irritating'. )https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/5269Can't close multiple notifications2022-07-26T12:03:21ZKolja LampeCan't close multiple notifications<!--
Please read https://wiki.gnome.org/Community/GettingInTouch/BugReportingGuidelines
first to ensure that you create a clear and specific issue.
-->
### Affected version
Gnome 42
Fedora 36 Beta
<!--
Provide at least the following i...<!--
Please read https://wiki.gnome.org/Community/GettingInTouch/BugReportingGuidelines
first to ensure that you create a clear and specific issue.
-->
### Affected version
Gnome 42
Fedora 36 Beta
<!--
Provide at least the following information:
* Your OS and version
* Affected GNOME Shell version (see https://wiki.gnome.org/Schedule for currently supported versions)
* Does this issue appear in XOrg and/or Wayland
-->
### Bug summary
If you have multiple notifications and want to close them manually, the second click will just close the notification center.
### Steps to reproduce
1. Have multiple closeable notifications
2. Open the notification center
3. Close the first notification by clicking `X`
4. The next notification will move right under your mouse cursor, so just click the `X` again
### What happened
The notification center closes
### What did you expect to happen
The notification center should not close, but the notification should be dismissed.
### Relevant logs, screenshots, screencasts etc.
I can add a screengrab later, if needed.
<!-- Do not remove the following line. -->https://gitlab.gnome.org/GNOME/gjs/-/issues/476Improve pretty-print output for GObject-introspected objects2022-06-22T18:06:56ZPhilip ChimentoImprove pretty-print output for GObject-introspected objectsThe old logging of objects with a toString() method did give some more information that the new pretty-printer didn't:
Old:
```
gjs> imports.gi.versions.Gdk = '3.0'
"3.0"
gjs> Gdk = imports.gi.Gdk
[object GIRepositoryNamespace]
gjs> new...The old logging of objects with a toString() method did give some more information that the new pretty-printer didn't:
Old:
```
gjs> imports.gi.versions.Gdk = '3.0'
"3.0"
gjs> Gdk = imports.gi.Gdk
[object GIRepositoryNamespace]
gjs> new Gdk.Rectangle()
[boxed instance wrapper GIName:Gdk.Rectangle jsobj@0xeadfc7890c0 native@0x55610677abf0]
gjs> GObject = imports.gi.GObject
[object GIRepositoryNamespace]
gjs> new GObject.Object()
[object instance wrapper GIName:GObject.Object jsobj@0xeadfc789100 native@0x55610690eea0]
```
New:
```
gjs> imports.gi.versions.Gdk = '3.0'
3.0
gjs> Gdk = imports.gi.Gdk
{ }
gjs> new Gdk.Rectangle()
{ }
gjs> GObject = imports.gi.GObject
(gjs-console:164032): Gjs-WARNING **: 21:13:39.102: JS ERROR: TypeError: can't convert symbol to string
formatObject@resource:///org/gnome/gjs/modules/script/_bootstrap/default.js:56:49
prettyPrint@resource:///org/gnome/gjs/modules/script/_bootstrap/default.js:29:20
@<stdin>:1:42
[error printing value]
gjs> new GObject.Object()
{ }
```
We should consider special pretty printers for introspected objects (boxed objects such as Gdk.Rectangle, objects that inherit from GObject.Object, GI namespace objects...)
One simple way to do this might be to check if the object's `toString` property equals `Object.prototype.toString` or `Array.prototype.toString`, and only do the pretty-printing if the `toString` is one of these standard functions; otherwise call the custom one and print that. Or we could do something more specific for each of these types of objects. Or a combination of the custom `toString` and the pretty-printer.
We should also figure out why printing `imports.gi.GObject` throws an exception and fix that.GNOME 43Nasah KumaNasah Kumahttps://gitlab.gnome.org/GNOME/gnome-text-editor/-/issues/347Crash when reopening Find overlay after performing another search2022-03-29T10:43:08ZDaniel AleksandersenCrash when reopening Find overlay after performing another search1. Type <kbd>example</kbd> into a new document.
2. Press <kbd>Ctrl+F</kbd>.
3. Type <kbd>example</kbd> into the find dialog.
2. Press <kbd>Escape</kbd>.
2. Press <kbd>Ctrl+F</kbd>.
I’m not able to generate a stacktrace from inside the F...1. Type <kbd>example</kbd> into a new document.
2. Press <kbd>Ctrl+F</kbd>.
3. Type <kbd>example</kbd> into the find dialog.
2. Press <kbd>Escape</kbd>.
2. Press <kbd>Ctrl+F</kbd>.
I’m not able to generate a stacktrace from inside the Flathub container. (I’m unfamiliar with `flatpak-coredumpctl`.) However, TextEditor always ouputs this to the console when the crash happens:
```
Highlighting a single line took too much time, syntax highlighting will be disabled
**
Gtk:ERROR:../gtk/gtktext.c:6574:blink_cb: assertion failed: (priv->selection_bound == priv->current_pos)
Bail out! Gtk:ERROR:../gtk/gtktext.c:6574:blink_cb: assertion failed: (priv->selection_bound == priv->current_pos)
```
Version 42 via Flathub on Fedora 35.