This week we did a lot of work to fix bugs and make UI improvements, but we still need help to knock out the remaining issues found during the Plasma 5.25 beta. If you are an experienced developer, please consider fixing one or two of these bugs within the week! Plasma 5.25 is a big release, but let’s not let it get a reputation for big bugs too. 🙂
15-Minute Bugs Resolved
Current number of bugs: 65, up from 64. 1 re-opened and 0 resolved.
This is the second week in a row that the number has risen. The reason is because Plasma’s core developers are all very busy with other projects. So if you want this number to go down, it’s going to take more contributions from volunteers. What are you waiting for!? 🙂
You can now remove individual items from the “Recent Files” and “Recent Locations” lists in Dolphin, the file dialogs, and other places (Méven Car, Dolphin 22.08):
It’s now easier to preview wallpapers: just click on them and the desktop will change to show you what the wallpaper would look like. The preview will only be applied if you click on either the “OK” or “Apply” button, of course (Fushan Wen, Plasma 5.26)
File open/save dialogs now let you sort hidden files last, just like you can do in Dolphin. And when you do show hidden files, they’re lightened–again, just like in Dolphin (Eugene Popov, Frameworks 5.95)
Bugfixes & Performance Improvements
Sorting by access time in Dolphin now works properly (Méven Car, Dolphin 22.04.2)
Spectacle’s global shortcut for “take screenshot of window under cursor” (
Meta+Ctrl+Print screen) now works properly and doesn’t cause the app to launch incorrectly and get stuck in memory when you close it (Paul Worral, Spectacle 22.04.2)
Konsole is now more reliable about parsing URLs that include things like port numbers or are IPV6 addresses (Ahmad Samir, Konsole 22.08)
Elisa’s “Files” view is now rooted at
/ rather than your home folder, so you can now use it to access music that isn’t in your home folder (Roman Lebedev, Elisa 22.08)
kded daemon no longer leaks XCB client connections when the screen configuration is changed, so it no longer eventually causes you to be unable to open new apps (Stefan Becker, Plasma 5.24.6)
3rd-party cursor themes can once again be applied and deleted (Alexander Lohnau, Plasma 5.24.6)
KRunner no longer freezes when it tries to display a search result with text that would consume more than three lines (Ismael Asensio, Plasma 5.24.6)
KWin’s lowest latency setting now actually works (Malte Dronskowski, Plasma 5.24.6)
When you sync your Plasma settings to the SDDM login screen while using a color scheme other than Breeze Light, the user interface elements in SDDM now respect the new color scheme without you having to manually clear the Plasma cache first (me: Nate Graham, Plasma 5.24.6)
Changing the delimiter character from a space to a colon (or vice versa) in KRunner’s web shortcuts now works without you needing to restart KRunner first (Alexander Lohnau, Plasma 5.24.6)
In the wallpaper chooser window, wallpapers now appear in the aspect ratio of the screen they will be applied to, not the aspect ratio of the screen that the window is currently located on (Fushan Wen, Plasma 5.24.6)
Discover now finds apps from their AppStream URLs when the trailing
.desktop suffix is omitted, which in particular makes it able to handle all the links at https://apps.kde.org (Antonio Rojas, Plasma 5.25)
Widget resize handles on the desktop now uses the correct cursor shapes when Plasma is running in reversed/RTL language mode (Ivan Tkachenko, Plasma 5.25)
Sliders are now drawn correctly with when Plasma is running in reversed/RTL language mode (Jan Blackquill, Plasma 5.25)
The “Astronomical Events” calendar plugin no longer shows an event for the intermediate moon phases (e.g. “waxing gibbous”) every single day (Volker Krause, Plasma 5.25)
It’s now possible to use wallpapers with ampersands in their filenames (Fushan Wen, Plasma 5.26)
Previews for various types of RAW image files are once again generated as expected (Alexander Lohnau, Frameworks 5.95)
Fixed a significant memory leak in the Plasma Wayland session (Méven Car, Frameworks 5.95)
Dolphin’s “All Tags” view now shows the correct name for all tags (Méven Car, Frameworks 5.95)
Fixed an issue in the common Kirigami Scrollview that could cause Kirigami-based apps–especially Discover–to freeze (Marco Martin, Frameworks 5.95)
Progress bars and sliders in QtQuick-based apps now have smoother animations (Ivan Tkachenko, Frameworks 5.95):
User Interface Improvements
In Elisa, you can now sort the Tracks view by “Date Modified”, which can be useful to find things you’ve recently added or changed (Shantanu Tushar, Elisa 22.08)
Tapping a song in Elisa’s playlist using a touchscreen now plays it immediately rather than simply selecting it. In addition, playlist items become taller and more finger-friendly when the app is interacted with using a touchscreen (me: Nate Graham, Elisa 22.08):
When you stretch Partition Manager’s window vertically, text in the information panel no longer awkwardly stretches as well (Ivan Tkachenko, Partition Manager 22.08)
Partition Manager now shows human-readable text for the amount of time that a disk has been powered on (Ivan Tkachenko, Partition Manager 22.08)
Global keyboard shortcuts in Plasma that currently don’t use the Meta key now do; here are the new keyboard shortcuts:
- Switch keyboard layout:
- Activate window that wants attention:
- Kill Window: Ctrl+Alt+Esc -> Meta+Ctrl+Esc
- Automatic Action Popup Menu: Ctrl+Alt+X -> Meta+Ctrl+X
- Manually Invoke Action on Current Clipboard:
Note that this change will only take effect for new installations; no shortcuts will be changed for existing users (me: Nate Graham, Plasma 5.25)
You can now click on a letter header in Kickoff’s “All Applications” view to be taken to a view where you can choose a letter and be zoomed right to the apps that start with that letter (Fushan Wen, Plasma 5.26):
Clicking the “Cancel” button in the desktop configuration dialog now shows a warning about unsaved changes, if you have unsaved changes (Fushan Wen, Plasma 5.26)
…And everything else
This blog only covers the tip of the iceberg! Tons of KDE apps whose development I don’t have time to follow aren’t represented here, and I also don’t mention backend refactoring, improved test coverage, and other changes that are generally not user-facing. If you’re hungry for more, check out https://planet.kde.org, where you can find more news from other KDE contributors.
How You Can Help
If you’re a developer, check out our 15-Minute Bug Initiative. Working on these issues makes a big difference quickly!
Otherwise, have a look at https://community.kde.org/Get_Involved to discover ways to be part of a project that really matters. Each contributor makes a huge difference in KDE; you are not a number or a cog in a machine! You don’t have to already be a programmer, either. I wasn’t when I got started. Try it, you’ll like it! We don’t bite!
Finally, consider making a tax-deductible donation to the KDE e.V. foundation.
20 thoughts on “This week in KDE: Fixing bugs, and let’s fix more”
Looking forwrards to 5.25 its going to be an awesome release great work everyone.
I do have a suggestion if its possible, some of us with 2-in-1 devices are having issues with tablet mode etc due to driver issues For example AMD still has issues with their Sensor Fusion Hub. I have screen rotation working fine but have no tablet mode due to issues with the driver for the lid switch. I was wondering if there could be an option to force tablet mode when the screen is rotated to a position opther than ‘Normal’. This would then disable the keyboard, trackpad and enable the larger icons for touch making it usable as tablet. Anyway just a thought.
Anyway good job everyone.
LikeLiked by 1 person
Keep up the good work!
LikeLiked by 2 people
You have not published how many new bugs you have introduced by fixing bugs. You keep fixing bugs yet KDE remains buggy as hell. You fix tons of bugs every month for years already instead of not introducing any new bugs by just writing f*ing unit tests. Unit tests distinguish run-of-the-mill developers from professional engineers.
Would you like to help us by contributing some more unit tests?
LikeLiked by 2 people
Hei, is there a dashboard to see for which code coverage would need improvement?
Not to my knowledge. But pretty much everything in Plasma needs more tests. There’s quite a bit of low-hanging fruit, for someone so inclined.
Dude… Entitlement much? Wth?
> You have not published how many new bugs you have introduced by fixing bugs. You keep fixing bugs yet KDE remains buggy as hell.
In my experience the remaining bugs are there because they just haven’t been fixed yet — a bugfix introducing new ones is not really as common you think, as in, it alone does not account for most let alone all of the bugs. The backlog is huge.
Also, please fix your attitude. Putting your points across politely and respectfully makes it much more likely that they will be taken seriously and constructively. You’re doing yourself a disservice by being this rude.
That xcb client leak might be describing what I’ve been encountering lately.
Runs out of window handles somehow, whatever they are.
.. What bugs me is this is on kubuntu 22.04 lts.
But hey you know, at least it’s fixed. (I hope.. lol)
One thing I guess~ I should mention is that 10-bit isn’t 100% yet, it might seem that way in some ways.
Not nagging or anything, just tired and bored.
Thanks for the updates.
Thanks for the blog thing as a whole.
It helps to follow parts of it.
To learn were parts are and crap like that.
For ex. kwin and plasma-desktop tweaks. (hoping the rest of the kickoff widget is in there somewhere lol.., trying to get rid of the default shutdown sub-menu)
Having plenty of C/C++ experience, and being on the MX Linux dev team, I have some ideas. Please don’t take this the wrong way, but I feel the barrier to entry for KDE development is unnecessarily high.
1. Fix your build system. It’s extremely frustrating that the build has to be started all over again just because a dependency problem is found half way.
2. Fix the documentation (pertinent to the point above). When the instructions on the wiki pages on how to build KDE don’t work (often bailing out when a dependency issue is found) it’s very discouraging when I have to attend to the many-hour build process the entire time.
There are a few things I can think of that will help utilise existing manpower/resources in a more effective way:
3. Remove dead or rarely used features. Do a survey on what KDE apps and features are or are not in use. For instance, who here uses rotating widgets? With the availability of SQLite, perhaps akonadi could go on a diet? It’s great to have apps and features, but I feel like the lack of manpower will get to you when you have to maintain all that. The sheer number of components in the Bugzilla I’ve never heard of is scary.
The “scream test” is a good tool (when used in moderation): disable the feature by default, then if nobody screams, disable it at compile time. If nobody screams still, then get rid of it altogether.
4. Exercise some discipline when accepting features or components. There isn’t anything wrong with saying no to a feature or app when it won’t help the project. It’s easy to say things like “if you don’t let volunteers add whatever they want they’ll leave” but ultimately it’s important to have a project that’s maintainable in the first place. What if the person who added the feature leaves, then you now have a feature that others need to maintain.
5. Remove unmaintained features and apps. In my experience, getting rid of unmaintained code can also fix bugs in other code, since it allows you to clean the code up.
And finally, from a user’s perspective:
6. Expose good features. I see, for example, activities mentioned a lot, yet I’m not sure how it can improve my work flow. Perhaps it can, but it isn’t exposed as well as it could be (eg. by means of documentation). I’m sure there are a number of other features out there, but unless I know about them somehow, I won’t use them.
Now, the stuff KDE does right:
1. Not being GNOME
2. Using modern UI elements the right way instead of just jumping on the smartphone bandwagon.
3. This blog.
4. Sensible defaults (to a degree).
5. Incremental easy-to-incorporate UI improvements instead of just changing the UI altogether (eg. Microsoft with the ribbon, GNOME, Windows 3.1->95 and Windows 7->8).
6. Not being GNOME.
I am in no way suggesting turning KDE into IceWM, but it’s important to understand how the users use your software. I use KDE every day, and I’d love to be able to patch something up instead of just reporting to Bugzilla and hope it gets fixed. Anyway, keep up the good work!
LikeLiked by 2 people
Thanks for your writings. You’re more than welcome to improve it if you think you can do (or delegate) it.
If you are such an expert that you claim you are. Then do it yourself.
Our build system already lets you resume from where you left off; see the `–resume-from` argument for `kdesrc-build`. This is documented at https://community.kde.org/Get_Involved/development#How_to_solve_build_problems. If that’s not obvious for you, you’re welcome to edit the page. it’s just a wiki, after all.
3, 4, and 5 are culturally very difficult for us to do, and also contradict “not being GNOME”. To avoid becoming like GNOME, we need to preserve our distinctiveness: broad scope, UX flexibility to meet the needs of beginners as well as experts. To preserve that, we need to expand our resources to match our scope. GNOME went in the opposite direction and the result was a much smaller project with a scope that matches its resources. If we did that too, what would be the difference between us and GNOME? We would be leaving our niche where we’re strong and playing in GNOME’s niche where they’re strong. And we would lose users to them, because they would be better at being them than we are.
LikeLiked by 1 person
I might have to try that again, last time I tried it last year it didn’t work and although it didn’t rebuild everything, it spent a lot of time skipping through components that it already built, then eventually gets to one where it cant find a whole bunch of dependencies.
GNOME’s problem isn’t about scope, their problem is they love dictating to users things they don’t want (hence my point 5), changing the UX in the most insane ways possible (eg. CSD, smartphone-ish interfaces, just to name a few).
Again, I am in no way suggesting turning KDE into IceWM or a “lightweight” desktop, but feature creep is a common theme when bugs are on the loose: “The reason is because Plasma’s core developers are all very busy with other projects.”
> 4. Exercise some discipline when accepting features or components. There isn’t anything wrong with saying no to a feature or app when it won’t help the project. It’s easy to say things like “if you don’t let volunteers add whatever they want they’ll leave” but ultimately it’s important to have a project that’s maintainable in the first place. What if the person who added the feature leaves, then you now have a feature that others need to maintain.
I see this being done already. Look at discussions in MRs, there is a lot of thinking before adding features or especially new off-by-default options.
3. Akonadi: Akonadi supports multiple backends over QSQL, Why should go Akonadi on a diet when the backend ist your choice? Akonadi could make switching backends easier but having a choice is a good thing.
I like to spend more memory (actually not that much more) by using the PSQL backend.
Seeing the subtle difference in progress bar animation. This tells me KDE is a mature project. Ironing out such minor details. Another things i really like about KDE is in for example GNOME there is one person who says this is important to the end users and this is not. And the ideas behind that are very subjective. In KDE if you feel something can be improved then you can do it. And as you won’t be spending your time on implementing something useless that comes down to KDE being a mature project and still evolving and improving itself. Not to be a victim of some niche group of people or end users.
LikeLiked by 1 person
Great as always.
> Kill Window: Ctrl+Alt+Esc -> Meta+Ctrl+Esc
Not sure about that one tbh… a lot of Windows users are used to Ctrl+Alt+Esc to get to the task manager so they can kill frozen apps. I always thought Kill Window being mapped to that was a nice touch.
well nvm on Windows it’s actually Ctrl+Shift+Esc