The new patch in comment #24 seems to further support the idea that bug
2098390 is the same issue.
** Changed in: gnome-shell-extension-desktop-icons-ng (Ubuntu)
Status: Fix Committed => In Progress
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is s
That is a good fix; you can also keep it after gnome-shell is fixed.
LGTM
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2098102
Title:
Desktop switching shows windows from previous desktop
To manag
This is a quick fix, until it is fixed upstream in Gnome Shell.
** Patch added: "fix-double-creation-bug.patch"
https://bugs.launchpad.net/ubuntu/+source/gnome-shell/+bug/2098102/+attachment/5859002/+files/fix-double-creation-bug.patch
--
You received this bug notification because you are a
The problem is even worse: the extension itself is created twice.
** Attachment added: "Captura desde 2025-02-20 12-28-57.png"
https://bugs.launchpad.net/ubuntu/+source/gnome-shell/+bug/2098102/+attachment/5859000/+files/Captura%20desde%202025-02-20%2012-28-57.png
--
You received this bug no
** Changed in: gnome-shell (Ubuntu)
Status: Won't Fix => In Progress
** Changed in: gnome-shell (Ubuntu)
Assignee: (unassigned) => Sergio Costas (rastersoft-gmail)
** Changed in: gnome-shell (Ubuntu)
Importance: Undecided => Medium
--
You received this bug notification because yo
This is my idea for an asyncMutex fix, but I don't know why the test is
failing... Running it manually from a command line (and some little
changes to allow it to run from command line, of course) does work as
expected.
https://gitlab.gnome.org/GNOME/gnome-shell/-/merge_requests/3640
--
You rece
Fixed upstream, we hope:
https://gitlab.com/rastersoft/desktop-icons-ng/-/merge_requests/439
** Changed in: gnome-shell-extension-desktop-icons-ng (Ubuntu)
Status: In Progress => Fix Committed
** Tags added: fixed-in-ding-48 fixed-upstream
--
You received this bug notification because yo
Yes, that patch would work, but I still think that the right-way-to-do-
it(tm) would be to use an async mutex, mainly because that can ensure
that no other yet-to-find race conditions can happen. Of course, I'm
talking about a patch for upstream.
--
You received this bug notification because you
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2098102
Title:
Desktop switching shows windows from previous desktop
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+so
An async function behaves sequentially until it executes an `await`,
where it breaks the flow and returns a Promise.
If we update the list of enabled extensions before yielding there will
be no race. Here's an untested patch.
** Patch added: "extensionSystem.diff"
https://bugs.launchpad.net/u
I think that a single mutex should protect the status of the extensions.
It should do a lock at the beginning of any async function calling any
of the `_callExtensionInit()`, `_callExtensionEnable()` or
`_callExtensionDisable()` async functions.
--
You received this bug notification because you a
Unfortunately, I think that GLib.Mutex won't work here, because they
would block the GJS thread itself... an async Javascript mutex is what I
think that we need: something where an async function can do an `await
mutex.lock();`
--
You received this bug notification because you are a member of Ubu
Ok, I think that the previous solution won't work... I think that the
solution is some kind of async mutex in `_onEnabledExtensionsChanged()`.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2098102
Tit
Seeing the code, I think that the problem is that
`_onEnabledExtensionsChanged()` is called both when the list of enabled
extensions change, and when the list of disabled extensions change. The
problem is that it is an async function, and the call in the callback
doesn't wait it to finish, so it ca
Oh! Good catch! Anyway, you are right: moving there.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2098102
Title:
Desktop switching shows windows from previous desktop
To manage notifications about
Ooops, the patch in comment #9 may be unreliable because you could have
two racing enable() calls and the first one hasn't flagged isEnabled yet
because innerEnable was deferred:
if (Main.layoutManager._startingUp) {
this.data.startupPreparedId =
Main.layoutManager.connect('st
The patch in comment #9 looks good but I won't push it to Plucky yet if
it's going to get in the way of releasing DING version 48.
That wouldn't be such a problem if the new version was in Salsa already,
but it's not: https://salsa.debian.org/gnome-team/shell-
extensions/gnome-shell-extension-desk
** Changed in: gnome-shell-extension-desktop-icons-ng (Ubuntu)
Assignee: (unassigned) => Sergio Costas (rastersoft-gmail)
** Changed in: gnome-shell-extension-desktop-icons-ng (Ubuntu)
Status: Confirmed => In Progress
** Changed in: gnome-shell-extension-desktop-icons-ng (Ubuntu)
I
Simpler patch
** Patch added: "fix-double-enable.diff"
https://bugs.launchpad.net/ubuntu/+source/gnome-shell/+bug/2098102/+attachment/5857233/+files/fix-double-enable.diff
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bu
The attachment "fix-double-enable.diff" seems to be a patch. If it
isn't, please remove the "patch" flag from the attachment, remove the
"patch" tag, and if you are a member of the ~ubuntu-reviewers,
unsubscribe the team.
[This is an automated message performed by a Launchpad user owned by
~brian
This is a workaround.
** Patch added: "fix-double-enable.diff"
https://bugs.launchpad.net/ubuntu/+source/gnome-shell/+bug/2098102/+attachment/5857209/+files/fix-double-enable.diff
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
ht
As I suspected... when calling enable() twice, the old pointers to the
previous methods are overwritten with the new function pointer; and
that's why disabling the extension won't restore them.
A quick workaround is to keep track of the current status, and just
ignore any enable() called when the
For the record, I've tested disabling the extension. It doesn't solve
the issue:
1. Start the system. The issue is there.
2. Disable the extension on the command line using `gnome-extensions`.
3. Check `gnome-extensions info d...@rastersoft.com`. It shows the
status of the extension as inactive.
It's a bug in the gnome-shell extension manager.
I made DING print traces of when it receives enable(). Notice we get two
calls, first from extensionSystem.js:739:18, then from
extensionSystem.js:736:18
Feb 13 09:48:24 ubuntu gnome-shell[2776]: Trace
enable@file:///usr/share/gnome-shell/extens
This sounds like stale window contents on the background "window", which
usually can only happen in X11 sessions.
Please try:
gnome-extensions disable d...@rastersoft.com
Does the problem stop happening?
** Changed in: gnome-shell (Ubuntu)
Status: New => Incomplete
** Changed in: gno
Marek mentioned elsewhere that this happened on the live-cd session, and
indeed I can reproduce by booting the same 20250211.1 daily live-cd.
Instead, I cannot reproduce on the 24.04.1 live-cd.
To reproduce, simply boot the daily live-cd and close the installer to
reach the desktop.
The logs poin
This bug has been reported on the Ubuntu ISO testing tracker.
A list of all reports related to this bug can be found here:
http://iso.qa.ubuntu.com/qatracker/reports/bugs/2098102
** Tags added: iso-testing
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is s
27 matches
Mail list logo
