GNOME issueshttps://gitlab.gnome.org/groups/GNOME/-/issues2022-09-25T22:58:34Zhttps://gitlab.gnome.org/GNOME/gnumeric/-/issues/671incorrect spacing in html read2022-09-25T22:58:34ZJohnDenkerincorrect spacing in html read**Scenario**
Here is a bit of HTML. It was written by gnumeric, then slightly edited by hand.
[code.html](/uploads/ded51b8081d238f723e6d86ae09c17e3/code.html)
Here's how it looks when viewed in a browser:
![code-browser](/uploads/2e6...**Scenario**
Here is a bit of HTML. It was written by gnumeric, then slightly edited by hand.
[code.html](/uploads/ded51b8081d238f723e6d86ae09c17e3/code.html)
Here's how it looks when viewed in a browser:
![code-browser](/uploads/2e6153ac68cc442526798d89bb8d7251/code-browser.png)
In contrast, here's how it looks when opened by gnumeric:
![code-gnm](/uploads/9b1f2e7e78a5b1971bf7df5deece4926/code-gnm.png)
Here's a patch that makes the bug go away:
[0001-fix-bug-concerning-trimming-spaces-when-appending-ht.patch](/uploads/5847e2b557e59ca4f452eb9ddeabcaa3/0001-fix-bug-concerning-trimming-spaces-when-appending-ht.patch)
**Applicability**
I'm pretty sure this bug affects all versions on all platforms. The relevant code has been unchanged for 15 years.
It mainly shows up when the <td> elements have multiple children. Not everybody does that, but I have very good reasons for doing it.https://gitlab.gnome.org/GNOME/goffice/-/issues/61weak rendering of ( only? ) subnormals in gnumeric, 'shortest' of 1.5E-323 is...2022-09-08T23:50:33Zb. s.weak rendering of ( only? ) subnormals in gnumeric, 'shortest' of 1.5E-323 isn't 1.48219693752374E-323title say's it all?
affected plenty of subnormals, e.g. `4E-312` -> `3.999999999998802E-312` or `2.09999979E-311` -> `2.09999979**00001**E-311`.
filed here as i assume go_dtoa as origin.
strings show up in cell display with format ...title say's it all?
affected plenty of subnormals, e.g. `4E-312` -> `3.999999999998802E-312` or `2.09999979E-311` -> `2.09999979**00001**E-311`.
filed here as i assume go_dtoa as origin.
strings show up in cell display with format general ( shrinked by column width ), formula bar, cell edit ( F2 ) and status bar, some flaws go into file save ( mixed with 'fixed length value save' ) and as a result gnumeric long double produces finalized deviations.
'long' is also affected in it's own subnormal range, e.g. `4E-4950` -> `4.01E-4950`.
tested with plenty versions, e.g. gnumeric 1.12.50 ( kali distro, don't know the corresponding goffice ver. ) and fresh pulled 1.12.53 with fresh pulled goffice 0.10.52,
medium quirk, math value is kept while you stay in gnumeric with double or long datatype, but compatibility gnumeric double -> long and eventually other systems is harmed, and behavior is unexpected for normal users.https://gitlab.gnome.org/GNOME/evince/-/issues/1844Adding annotations and saving makes the app repeatedly crash2023-02-09T22:41:58ZKalytisAdding annotations and saving makes the app repeatedly crashI'm trying to work on a thesis document, which requires annotations. But adding, editing and removing annotations makes the app repeatedly crash. Also, saving the document containing annotations often makes the document crash before anno...I'm trying to work on a thesis document, which requires annotations. But adding, editing and removing annotations makes the app repeatedly crash. Also, saving the document containing annotations often makes the document crash before annotations are saved, hence deleting them.https://gitlab.gnome.org/GNOME/gedit/-/issues/523problem with cursor movement in a text which contains arabic characters2022-09-01T08:43:04ZPiotr S.problem with cursor movement in a text which contains arabic charactersWhen I open this text file in gedit (version 3.36.2):
[problem.txt](/uploads/ce6b6d42da3c6a6472cd6d9873101f42/problem.txt)
and I place a cursor on comma in the second line (between arabic characters), and I press "left arrow" key, to mov...When I open this text file in gedit (version 3.36.2):
[problem.txt](/uploads/ce6b6d42da3c6a6472cd6d9873101f42/problem.txt)
and I place a cursor on comma in the second line (between arabic characters), and I press "left arrow" key, to move left, I can't get to "foo" - when cursor reaches left arabic character, it suddenly jumps to the right arabic character.https://gitlab.gnome.org/GNOME/mutter/-/issues/2411The pointer is async between mouse and touchscreen in Wayland2023-10-09T10:38:25ZJeremy SuThe pointer is async between mouse and touchscreen in Wayland### Affected version
* Your OS and version
Ubuntu 22.04.1 LTS
* Affected Mutter version
42.2-0ubuntu1
* Does this issue appear in XOrg and/or Wayland
Wayland only
### Bug summary
The cursor location from mouse isn't sync with touchscr...### Affected version
* Your OS and version
Ubuntu 22.04.1 LTS
* Affected Mutter version
42.2-0ubuntu1
* Does this issue appear in XOrg and/or Wayland
Wayland only
### Bug summary
The cursor location from mouse isn't sync with touchscreen's point if using them in the same time in Wayland.
### Steps to reproduce
1. Login as Wayland mode.
2. moving the cursor to left side via moving mouse/touchpad
3. press the touchscreen in right side
### What happened
The cursor (from mouse) stay in left side and touchscreen point stay in right side.
If user uses both in the same time, then user can see the cursor blinks quickly.
### What did you expect to happen
Same as in Xorg mode, the cursor location is updated if pressing touchscreen.
<!-- Do not remove the following line. -->![wayland](/uploads/0004ce7f3c42a853b013e630240733fd/wayland.webm)
![Xorg](/uploads/7bd38ffa9e9bbbb88243804bf1c0715e/Xorg.webm)https://gitlab.gnome.org/GNOME/gnome-desktop/-/issues/212default-input-sources.h: Arabic should default to "m17n:ar:kbd"2022-09-15T08:31:12ZJens Petersendefault-input-sources.h: Arabic should default to "m17n:ar:kbd"Based on feedback from Arabic users of Fedora, the default Arabic xkb layout seems insufficient for inputting some Arabic combinations.
See for example https://bugzilla.redhat.com/show_bug.cgi?id=2122899
and https://bugzilla.redhat.com/...Based on feedback from Arabic users of Fedora, the default Arabic xkb layout seems insufficient for inputting some Arabic combinations.
See for example https://bugzilla.redhat.com/show_bug.cgi?id=2122899
and https://bugzilla.redhat.com/show_bug.cgi?id=1076945 for more details.
I would like to propose that ibus "m17n:ar:kbd" be used as the preferred non-Latin Input Source for general Arabic.
I see only ar_Dz (Algeria) is listed in default-input-sources.h, I dunno if it is sufficient to update that entry?https://gitlab.gnome.org/GNOME/gjs/-/issues/504One-time SEGV 10~20 seconds after login in 42.42022-11-07T00:36:56ZThomastw.booker@outlook.comOne-time SEGV 10~20 seconds after login in 42.4<!--
Please read https://wiki.gnome.org/Community/GettingInTouch/BugReportingGuidelines
first to ensure that you create a clear and specific issue.
-->
### Affected version
OS: Arch (1 Sep 2022)
Gnome: 42.4 reported in settings
Disp...<!--
Please read https://wiki.gnome.org/Community/GettingInTouch/BugReportingGuidelines
first to ensure that you create a clear and specific issue.
-->
### Affected version
OS: Arch (1 Sep 2022)
Gnome: 42.4 reported in settings
Display Server: Wayland
<!--
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
I logged into my account where Fragments and Openrgb autostart. I then open the activities view move to the right empty workspace and leave the activities view. Then I moved the mouse around for a bit then the mouse movement froze and a few seconds later I was kicked back to GDM. I then logged in again and looked in coredumpctl where there was a Gnome shell dump.
<!--
Provide a short summary of the bug you encountered.
-->
### Steps to reproduce
As far as I can tell this was a one-time issue that I couldn't reproduce.
### What happened
Gnome shell crashed and returned me to GDM with a core dump in systemd.
### What did you expect to happen
Gnome shell did not crash.
### Relevant logs, screenshots, screencasts etc.
[bt](/uploads/6d43c2049e1f110bc0524d9a59772b4e/bt)
<!-- Do not remove the following line. -->https://gitlab.gnome.org/GNOME/libxslt/-/issues/76compile error for casting between unsigned long and void*2022-10-21T02:57:35ZJoey Arharcompile error for casting between unsigned long and void*Commit 82f6cbf8ca61b1f9e00dc04aa3b15d563e7bbc6d (thanks so much for that patch!) caused a compile error on a peculiar build platform in chromium: https://ci.chromium.org/ui/p/chromium/builders/try/win-libfuzzer-asan-rel/1125277/overview
...Commit 82f6cbf8ca61b1f9e00dc04aa3b15d563e7bbc6d (thanks so much for that patch!) caused a compile error on a peculiar build platform in chromium: https://ci.chromium.org/ui/p/chromium/builders/try/win-libfuzzer-asan-rel/1125277/overview
```
../../third_party/libxslt/src/libxslt/functions.c(741,14): error: cast to smaller integer type 'unsigned long' from 'void *' [-Werror,-Wvoid-pointer-to-int-cast]
id = (unsigned long) *psviPtr;
^~~~~~~~~~~~~~~~~~~~~~~~
../../third_party/libxslt/src/libxslt/functions.c(761,20): error: cast to 'void *' from smaller integer type 'unsigned long' [-Werror,-Wint-to-void-pointer-cast]
*psviPtr = (void *) id;
^~~~~~~~~~~
2 errors generated.
```
I tried making id a intptr_t instead of unsigned long, but that also caused other compile errors when id is used in other spots. I can continue to mess around with different types for id, but I also don't know if it would be better to do something else to convert between unsigned long and void*. Any thoughts?https://gitlab.gnome.org/GNOME/gnome-shell/-/issues/5818One-time SEGV 10~20 seconds arder login in 42.42022-08-31T23:00:42ZThomastw.booker@outlook.comOne-time SEGV 10~20 seconds arder login in 42.4<!--
Please read https://wiki.gnome.org/Community/GettingInTouch/BugReportingGuidelines
first to ensure that you create a clear and specific issue.
-->
### Affected version
OS: Arch (1 Sep 2022)
Gnome: 42.4 reported in settings
Disp...<!--
Please read https://wiki.gnome.org/Community/GettingInTouch/BugReportingGuidelines
first to ensure that you create a clear and specific issue.
-->
### Affected version
OS: Arch (1 Sep 2022)
Gnome: 42.4 reported in settings
Display Server: Wayland
<!--
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
I logged into my account where Fragments and Openrgb autostart. I then open the activities view move to the right empty workspace and leave the activities view. Then I moved the mouse around for a bit then the mouse movement froze and a few seconds later I was kicked back to GDM. I then logged in again and looked in coredumpctl where there was a Gnome shell dump.
<!--
Provide a short summary of the bug you encountered.
-->
### Steps to reproduce
As far as I can tell this was a one-time issue that I couldn't reproduce.
### What happened
Gnome shell crashed and returned me to GDM with a core dump in systemd.
### What did you expect to happen
Gnome shell did not crash.
### Relevant logs, screenshots, screencasts etc.
[bt](/uploads/7f92547ae51f376612d90e0455d8ced2/bt)
<!-- Do not remove the following line. -->https://gitlab.gnome.org/GNOME/simple-scan/-/issues/325canoscan 1220u recognise but not start to working (edited)2023-02-09T22:41:39ZJosé Albertocanoscan 1220u recognise but not start to working (edited)hi, I'd like to tell that my old scanner canoscan 1220u is recognized by the program but it does not start to scan... I know that canon did not made driver for win 7 but maybe in this gorgeous Linux my scan will start again. Edited: I te...hi, I'd like to tell that my old scanner canoscan 1220u is recognized by the program but it does not start to scan... I know that canon did not made driver for win 7 but maybe in this gorgeous Linux my scan will start again. Edited: I tested the scanner in my old PC with Windows Millenium and the scanner did not work. Maybe the problem is in the scanner, and not in the program.
cross my fingers!!!
Regards,
~jazeihttps://gitlab.gnome.org/GNOME/NetworkManager-openvpn/-/issues/106OpenVPN - Show passwords not showing passwords.2022-09-05T10:56:43ZD33M0NOpenVPN - Show passwords not showing passwords.in Manjaro repo:
```
networkmanager 1.38.4-1
networkmanager-openvpn 1.8.18-2
```
![image](/uploads/6451f704a228765f5f0653a16cc55ce8/image.png)in Manjaro repo:
```
networkmanager 1.38.4-1
networkmanager-openvpn 1.8.18-2
```
![image](/uploads/6451f704a228765f5f0653a16cc55ce8/image.png)https://gitlab.gnome.org/GNOME/gnome-maps/-/issues/474Loading spinner are too large2022-08-31T19:40:19ZMarcus LundbladLoading spinner are too largeSome loading spinner are too large (filling all available space) after the GTK4 port.Some loading spinner are too large (filling all available space) after the GTK4 port.GNOME 43https://gitlab.gnome.org/GNOME/meld/-/issues/700How to ignore (/ filter) a linebreak difference in Meld?2022-09-01T08:51:16ZNor WaHow to ignore (/ filter) a linebreak difference in Meld?### Situation
I need to compare 2 text file.
However,
File AA has linebreaks misplaced everywhere.
File BB 's linebreaks are fine.
\---
The linebreak differences are not important to me, so
=> **How to ignore (/ filter) a linebre...### Situation
I need to compare 2 text file.
However,
File AA has linebreaks misplaced everywhere.
File BB 's linebreaks are fine.
\---
The linebreak differences are not important to me, so
=> **How to ignore (/ filter) a linebreak difference in Meld?**
\---
note::
- `Text filter` doesnt work as expected.
#### ex
![image](/uploads/92bf16234e849f2b5390c9591495610f/image.png)https://gitlab.gnome.org/GNOME/recipes/-/issues/156-2022-08-31T19:20:58ZAicelly-==https://gitlab.gnome.org/GNOME/epiphany/-/issues/1848Memory leak and high memory usage2023-07-19T13:48:01ZPeter WeberMemory leak and high memory usageHello!
I don't know if this is related to #1610. I'm running Archlinux and since some days Epiphany uses very high amounts of memory after startup. `Restore Tabs on Startup` is active (currently 12 to 15 tabs) and the last tab which was...Hello!
I don't know if this is related to #1610. I'm running Archlinux and since some days Epiphany uses very high amounts of memory after startup. `Restore Tabs on Startup` is active (currently 12 to 15 tabs) and the last tab which was used is actually loaded. Subsequently memory usage is increasing even further until the system cannot provide more memory.
![Untitled](/uploads/8c23d9ef52822908acf120e915744765/Untitled.png)
The memory consumption in more detail:
![Screenshot_from_2022-08-31_18-49-24](/uploads/fccb7da5d1b44dd6d2f1aaa05a1cc1ca/Screenshot_from_2022-08-31_18-49-24.png)https://gitlab.gnome.org/GNOME/nautilus/-/issues/2468Nautilus thumbnailing (many potentially large) files is blocking GUI, usuabil...2023-07-09T16:28:34ZPercyNautilus thumbnailing (many potentially large) files is blocking GUI, usuability improvement suggetsionsI am watching this now for years and it seams to stay unaddressed.
Thumbnailing is not only slow (OK when many large files) -- but is getting more a usability issue as it more or less blocks the nautilus GUI when busy updating many icon...I am watching this now for years and it seams to stay unaddressed.
Thumbnailing is not only slow (OK when many large files) -- but is getting more a usability issue as it more or less blocks the nautilus GUI when busy updating many icons. Also drag and drop seam to stall a lot when this happens. In some situations deadly locking nautilus and I have to kill it.
This applies to many simple jpeg images and also to other default file formats. And similar applies to custom thumbnailers. I am talking about 1000+ files here in one folder.
I suggest to better manage the thumbnail processes and give them a low (nice) priority if exceeding a larger number to files to be processed to not bog down the machine usability.
Also the gtk level "thumb nail" update of the actual generated icons seam to have a impact. This rendering process also must be managed better to not impact GUI operations.
Another related issue: When updating thumbnail icons in the file view pane -- those often shuffle around crazy (icon size changes oddly temporary, etc.) for no good reason and selecting/working with files is a bit painful. Also when inserting new appearing files -- it should be tried to not shift currently selected/highlighted files and maintain the location/scroll pos and only move icons before/after the active selection if possible.https://gitlab.gnome.org/GNOME/gimp/-/issues/8562.ani cursor generated using 2.99.12 not recognised in Windows 102023-06-30T20:26:26ZDilip.ani cursor generated using 2.99.12 not recognised in Windows 10### Environment/Versions
- GIMP version: 2.99.12
- Package: <!--[flatpak? Installer from gimp.org? If another installer, tell us where from] (write it after the > symbol)--> Windows installer from the official source, downloaded through...### Environment/Versions
- GIMP version: 2.99.12
- Package: <!--[flatpak? Installer from gimp.org? If another installer, tell us where from] (write it after the > symbol)--> Windows installer from the official source, downloaded through official torrent
- Operating System: <!--[Windows? macOS? Linux? All?] (write it after the > symbol) --> Windows 10
<!--Note: bug reporters are expected to have verified the bug still exists
either in the last stable version of GIMP or on updated development code
(master branch).-->
### Description of the bug
<!--Please describe your issue with details.
Add screenshot or other files if needed.(write it after the > symbol)--> I created an 2.ani file, I was able to draw and export without issues. But the .ani file has no thumbnail on desktop. When I tried to replace the busy icon using mouse pointer settings, I was able to browse to 2.ani on the desktop and select it. But nothing changes in the cursor set or cursor preview. It is as if nothing got selected.
Attaching the 2.ani and source 2.xcf
### Reproduction
Is the bug reproducible? <!--[Always / Randomly / Happened only once ] (write it after the > symbol)--> Yes, I created a different other.ani file, but the same issue. I had 3 layers only, while the first one had around 5-6 layers
Reproduction steps:
1. open GIMP
2. Draw in layers
3. Export as .ani
…
Expected result: Busy icon should change when selected through Mouse Pointer settings
Actual result: Busy icon did not change when selected through Mouse Pointer settings
### Additional information
If you have a backtrace for a crash or a warning, paste it here.2.99.14Alx SaAlx Sahttps://gitlab.gnome.org/GNOME/gnome-boxes/-/issues/844Generic Wayland Icon shows in Wayland session instead of Gnome Boxes icon2023-11-26T16:40:35ZEric KirchmanGeneric Wayland Icon shows in Wayland session instead of Gnome Boxes icon![Screenshot_20220831_094858](/uploads/88606561de61a50da200cb851b6084ab/Screenshot_20220831_094858.png)
Gnome Boxes displays the generic Wayland icon, when using KDE Plasma's task manager widget and/or Latte dock. I noticed this issue b...![Screenshot_20220831_094858](/uploads/88606561de61a50da200cb851b6084ab/Screenshot_20220831_094858.png)
Gnome Boxes displays the generic Wayland icon, when using KDE Plasma's task manager widget and/or Latte dock. I noticed this issue both when using the native rpm and the Flatpak version.
```
Gnome Boxes Flatpak version: 42.3
Gnome Boxes RPM version: 42.3
Operating System: Fedora Linux 36
KDE Plasma Version: 5.25.4
KDE Frameworks Version: 5.97.0
Qt Version: 5.15.5
Kernel Version: 5.19.4-200.fc36.x86_64 (64-bit)
Graphics Platform: Wayland
Processors: 8 × Intel® Core™ i5-8250U CPU @ 1.60GHz
Memory: 15.4 GiB of RAM
Graphics Processor: Mesa Intel® UHD Graphics 620
Manufacturer: LENOVO
System Version: ThinkPad T480
```
Terminal output:
```
➜ apps flatpak run org.gnome.Boxes
(gnome-boxes:2): GLib-GIO-WARNING **: 09:52:32.013: Failed to create file monitor for /home/eric/.var/app/org.gnome.Boxes/config/glib-2.0/settings/keyfile: Unable to find default local file monitor type
(gnome-boxes:2): Gtk-WARNING **: 09:52:32.189: GtkFlowBox with a model will ignore sort and filter functions
(gnome-boxes:2): Gtk-WARNING **: 09:52:32.190: GtkListBox with a model will ignore sort and filter functions
(gnome-boxes:2): Handy-WARNING **: 09:52:32.255: Using GtkSettings:gtk-application-prefer-dark-theme together with HdyStyleManager is unsupported. Please use HdyStyleManager:color-scheme instead.
(gnome-boxes:2): GLib-GObject-WARNING **: 09:52:32.871: ../gobject/gsignal.c:2692: handler '2908' of instance '0x563c5d8d0470' is not blocked
```
Other GTK flatpak apps don't have this issue, like [Blanket](https://apps.gnome.org/app/com.rafaelmardojai.Blanket/) or Flatseal.
If you believe this isn't an issue with Gnome Boxes itself, I'll gladly move this issue to where you believe the underlying problem lies. Thankshttps://gitlab.gnome.org/GNOME/gimp/-/issues/8561meson.build: gcc error: unrecognized command-line options '-mmx', '-sse'2022-08-31T14:17:28ZSergey Torokhovmeson.build: gcc error: unrecognized command-line options '-mmx', '-sse'I was confused that gimp-2.99.12 on configuration states that my CPU doesn't support MMX and SSE.
Looking through meson-log.txt I saw related compiler errors:
```
gcc: error: unrecognized command-line options '-mmx'; did you mean '-mmmx...I was confused that gimp-2.99.12 on configuration states that my CPU doesn't support MMX and SSE.
Looking through meson-log.txt I saw related compiler errors:
```
gcc: error: unrecognized command-line options '-mmx'; did you mean '-mmmx'?
gcc: error: unrecognized command-line options '-sse'; did you mean '-msse'?
```
It's seems that meson.build (https://gitlab.gnome.org/GNOME/gimp/-/blob/master/meson.build#L188-189) has typo, isn't it? Is there should be '-mmmx' and '-msse'? I.e.:
```
conf.set('USE_MMX', cc.has_argument('-mmmx'))
conf.set('USE_SSE', cc.has_argument('-msse'))
```
I localy made this changes and meson configuration successfully recognize that CPU and compiler support MMX and SSE.2.99.14JehanJehanhttps://gitlab.gnome.org/GNOME/tracker-miners/-/issues/230Rework tracker-cue-sheet to avoid GStreamer dependency2022-09-26T09:06:16ZSam ThursfieldRework tracker-cue-sheet to avoid GStreamer dependencyCurrently tracker-cue-sheet.c is built around GstTagList. This means that we cannot easily enable cue sheet support with the ffmpeg/libav backend. Also, we cannot use the disc-generic extractor when ffmpeg/libav backend is enabled.
The ...Currently tracker-cue-sheet.c is built around GstTagList. This means that we cannot easily enable cue sheet support with the ffmpeg/libav backend. Also, we cannot use the disc-generic extractor when ffmpeg/libav backend is enabled.
The current API is built around a `TrackerToc` struct, which contains a list of `TrackerTocEntry`. Each entry has a `GstTagList` with metadata about the current entry. The GStreamer extractor calls `tracker_cue_sheet_parse()` or `tracker_cue_sheet_guess_from_uri()` once to get a TrackerToc instance.
I propose modifying the API as follows:
* TrackerToc becomes an opaque pointer, which can refer to a private struct containing the libcue `Cd *` pointer
* new function `tracker_toc_get_total_tracks()` returns the number of available tracks
* new function `tracker_toc_extract_album_disc()` returns a TrackerResource(nmm:AlbumDisc) with the toplevel CUE metadata applied
* new function `tracker_toc_extract_track()` returns a TrackerResource(nmm:MusicPiece) with the track-specific CUE metadata applied
It's then up to the GStreamer extractor to apply any extra metadata from the GstTagList returned by GstDiscoverer.
This code also deals with GstDiscovererToc, so some cleverness (and extra testing) is needed to organise the code neatly.