Hi everyone, my name is abdelhadi I am a software engineer from Egypt and currently studying computer engineering.
i recently got accepted into google summer of code to work with KDE this summer :)
I will be working specifically with Drawy under Laurent Montel mentorship on Implementing Rich Text project.
Recently a new submodule has landed in Kirigami: “Forms”.
Until this point, Kirigami had only offered the classic “FormLayout” component. which is used for configuration pages throughoug systemsettings, Plasma, and some apps. It’s the classical form used in desktop toolkits for decades:
This is a fairly clean layout which however is starting to slowly become outdated, as modern toolkits are starting to use a different layout nowdays, based on “cards”
Unfortunately FormLayout very bound to the classic layout, and it wasn’t really possible to adapt it to the new look in a compatible way which didn’t break existing applications in unexpected ways.
This is also the reason a new approach was done provided by kirigami addons: “FormCard”, which is used by a lot of applications; for instance here in NeoChat:
We wanted to have this new style of forms in the base Kirigami API, so after a review of the existing FormCard, we decided to make several changes, for two main reasons: First, FormCard is bound to the card style of form as much as FormLayout was bound to the classic flat style. Also, it tried to provide ready-made components for every kind of control; so it had its own TextField, its own RadioButton and so on — effectively becoming its own separate toolkit.
So we opted instead to go down the route of having a more generic API, so the Forms module includes containers that define a semantic structure of a form, which contains all the normal controls — such as textfields, checkboxes and radiobuttons.
This is a code example of the new API:
@abdelhadi-wael:matrix.org
GSoC
notmart import QtQuick.Controls as QQC
import org.kde.kirigami as Kirigami
Kirigami.Form {
Kirigami.FormGroup {
title: "Global Settings"
Kirigami.FormEntry {
contentItem: QQC.CheckBox {
text: "Show Sidebar"
}
}
Kirigami.FormEntry {
contentItem: QQC.CheckBox {
text: "Auto Save"
}
}
}
Kirigami.FormGroup {
title: "Theme Options"
Kirigami.FormEntry {
title: "Colors:"
contentItem: QQC.CheckBox {
text: "Dark Theme"
}
}
Kirigami.FormSeparator {}
Kirigami.FormEntry {
contentItem: QQC.CheckBox {
text: "High Contrast"
}
}
...
}
}
which will look like this:
Or, in mobile mode:
Semantically, a Form will contain one or more FormGroup objects, each of which will contain one or more FormEntry objects. Then a FormEntry will contain the control which represents the configuration of the particular thing. It can be a single control (like a button or a checkbox) or it can be any layout with completely custom contents.
I already ported 4 modules of systemsettings to the new system: the landing page, the “workspace options” kcm, the mouse settings and the touchpad settings.
But wait… this page looks exactly the same as before; why?
A key was to do an API that was as much as separated from any appearance as possible, as we don’t know how UI design trends will evolve in the future. But this also allows us another thing: to have two separate implementations: the new one “card based” and a legacy one which looks exactly like the current FormLayout components. This is used only in systemsettings, so we can port all the kcms without introducing glaring visual inconsistencies, and when we are done, flick the switch and convert the look of everything all in one go.
Since most of KDE’s QML applications already use the existing card-style kirigamiaddons FormCard components, the new look will be used there.
And then in the future, when trends change again, we can re-style all the settings pages in one go.
We ideally want the whole set of systemsettings kcms to be ported as soon as possible to the new system, so we can have soon a nice visual overhaul in the whole systemsettings.
In order for this to happen, we need the help of everyone, so… patches welcome 
As an example, this is the merge request that ported the first four kcms.
When porting, it’s also possible to see how the kcm will look with the new system as well, to make sure it works well for when we flick the switch. If we run in a terminal:
KDE_KIRIGAMI_FORMS_STYLE=cards systemsettingsWe get systemsettings using the new style for pages already ported:
Porting from FormLayout to the new Form/FormGroup/FormEntry system should be really straightforward; it should be possible to make good progress in little time.
With your help, soon KDE’s settings will benefit from greater consistency, a more modern style, and easier adaptation to future designs.
Google Summer of Code (GSoC) is a training/mentorship program that allows new contributors to open source to work on projects for between 175 to 350 hours under the guidance of experienced mentors.
KDE will mentor twelve projects in this year's Google Summer of Code.
Marknote is a rich text notebooks organizer.
Prayag Jain aims to introduce a robust and high performance block editor and a proper markdown parser to Marknote under the guidance of Carl Schwan and Mathis Brüchert.
digiKam is an advanced open-source digital photo management application which provides a comprehensive set of tools for importing, managing, editing, and sharing photos and raw files.
Srirupa Datta, who already successfully completed a GSoC in 2023 working on Krita, will this year work on digiKam to interface the database search engine to an AI-based LLM. This project is mentored by Gilles Caulier, Michael Miller and Maik Qualmann.
Drawy is a recent addition to KDE applications! It is an infinite whiteboard tool useful for brainstorming.
Abdelhadi Wael will improve the text items by implementing rich text handling and other improvements. This project is mentored by Laurent Montel.
Kdenlive is KDE's video editor.
Yash Bavadiya will work on improving the curves, gradients and time remapping effects. This project is mentored by Jean-Baptiste Mardelle.
Shubham Shinde, a successful participant in last year's GSoC working on Merkuro, will work this year again under the guidance of Carl Schwan to improve the Kirigami framework and the developer experience.
Okular is KDE's universal document viewer.
Ojas Maheshwari will work under the guidance of Albert Astals Cid to implement font subsetting when saving PDF files in Poppler (the PDF rendering library used by Okular).
Lokalize is the localization tool for KDE software and other free and open source software.
Navya Sai Sadu will improve Lokalize by redesigning the translation memory tab to help with the translators' experience. This project is mentored by Finley Watson.
KStars is the KDE astronomy software providing an accurate graphical simulation of the night sky, from any location on Earth, at any date and time.
Pavan Kumar S G will add a new AI powered guiding assistant for Ekos under the guidance of Jasem Mutlaq.
KeepSecret is a password manager for viewing, editing, creating, or deleting passwords.
Roshani Kumari will work on improving the user experience and adding new features such as import/export of passwords, adding a built-in password generator, and much more. This work is mentored by Marco Martin.
Ansh Singhal will work on creating a new join.kde.org website which aims to improve the onboarding experience by centralizing the different entry points the KDE community has. This project is mentored by Anish Tak, who was a successful mentee last year on the same topic!
Tushar Gupta will rework some networking modules (KCM) to make them more mobile friendly. This work will be mentored by Carl Schwan and Devin Lin.
Mankala is a two-player board game containing multiple variants.
Sayandeep Dutta will add a new interface to create tournaments for the Mankala game. This project is mentored by Benson Muite and Srisharan V. S. who completed a successful GSoC contributor on the same project last year.
I’m happy to announce that we have sealed bootable container images ready for testing for the Fedora Atomic Desktops!
Note: You can also read this post on the Fedora Magazine.
Sealed bootable container images include all the components needed to create a fully verified boot chain, from the firmware to the operating system composefs image.
This relies on Secure Boot and thus only supports system booting with UEFI on x86_64 & aarch64.
The components are:
Both systemd-boot and the UKI are signed for Secure Boot. The images are test images so the components are not signed with the official keys from Fedora.
The main direct benefit that we will get from this support is that we will be able to enable passwordless disk unlocking using the TPM in a way that will be reasonably secure by default.
See the instructions at github.com/travier/fedora-atomic-desktops-sealed on how to give the pre-built container and disk images a try and how to build your own.
We welcome testing and feedback! Please see the list of known issues and report new issue at github.com/travier/fedora-atomic-desktops-sealed. We’ll redirect them as needed to the right upstream projects.
Beware, those are testing images. The root account does not have a password set and sshd is enabled, by default, to make debugging easier. The UKI and systemd-boot are signed for Secure Boot but, since those are test images, they are not signed with the official keys from Fedora. Don’t use those images in production.
If you want to know more about how sealed images work (i.e. how we make bootable containers, UKI and composefs work together to create a verified boot chain), see the following presentations and documentation:
Thanks to all the contributors that made this possible, notably (but non exhaustively) from the following projects: bootc & bcvk, composefs & composefs-rs, chunkah, podman & buildah and systemd.
Fedora 44 has been released! 🎉 So let’s see what is included in this new release for the Fedora Atomic Desktops variants (Silverblue, Kinoite, Sway Atomic, Budgie Atomic and COSMIC Atomic).
Note: You can also read this post on the Fedora Magazine.
We have moved the cross-variants issue tracker to the new Fedora forge. This is the best place to file issues that impacts all variants or to coordinate work between all of them. If you have issues specific to a given desktop environment then we usually prefer to track them in each respective SIG trackers. They are listed on the README for the atomic-desktops organization.
The unified documentation for all Atomic Desktops is finally live! Unfortunately the translations have not been migrated so we will need help to re-translate everything again, once the translation setup is ready with the new forge. It should be mostly copy/paste from the previous docs and this time we will only have to translate the docs once and not for every (new) variant.
See the tracking issue atomic-desktops#10.
FUSE version 2 has been deprecated and unmaintained for a while so we have removed it from the images. In practice, this means two things:
See the Fedora Change and the tracking issue atomic-desktops#50.
The implications are detailed below.
Some AppImages are still using an old AppImage runtime that relies on FUSE 2 libraries being available on the host. See the discussion thread for examples on how to check the runtime of an AppImage.
If some of your AppImages do not work on Fedora Atomic Desktops 44, we recommend:
KDE upstream no longer recommend using the EncFS nor CryFS backends for Plasma Vaults, notably because they rely on the FUSE 2 libraries.
If you are using one of those backends, you should migrate your data to a new Vault using the only maintained backend (gocryptfs).
Ideally this should occur before the update to Fedora 44.
If you have already updated to Fedora 44 and need access to your data, you can layer the needed packages (cryfs or fuse-encfs) using rpm-ostree install <package>, then migrate your data and finally reset the layers with rpm-ostree reset.
Support for the legacy pkla polkit rules format has been removed. It is unlikely that you were relying on support for those rules as most of the ecosystem has moved on to the new Javascript based format.
See the Fedora Change and the tracking issue atomic-desktops#102.
Fedora Silverblue comes with the latest GNOME 50 release.
For more details about the changes that alongside GNOME 50, see What’s new in Fedora Workstation 44 on the Fedora Magazine.
Fedora Kinoite ships with Plasma 6.6, Frameworks 6.24 and Gear 25.12.
See also What’s new in Fedora KDE Plasma Desktop 44 on the Fedora Magazine.
The brand new Plasma Login Manager replaces SDDM to provide a more integrated experience with systemd and the KDE Plasma session.
See the Fedora Change.
Thanks to the new Plasma Setup, it is now possible to install the system with Anaconda with minimal configuration and then complete the installation on the first boot by creating a new user and selecting the timezone. This is great when you want to install Fedora Kinoite on a computer and don’t want to setup a user in advance.
See the Fedora Change.
Nothing specific for this release.
Fedora Budgie Atomic comes with the latest 10.10.2 Budgie release. This release brings Wayland support to Budgie Atomic. See the 10.10 release announcement for more details.
Fedora COSMIC Atomic comes with the latest 1.0.8 release of the COSMIC desktop. This is now considered stable.
Our friends in the Universal Blue project (Bazzite, Bluefin, Aurora) have prepared the update to Fedora 44. Look for upcoming announcements in their Discourse.
As always, I heavily recommend checking them out, especially if you feel like some things are missing from the Fedora Atomic Desktops and you depend on them (NVIDIA drivers, extra media codec, out of tree kernel drivers, etc.).
If you are interested in contributing to Fedora Atomic Desktops, here are some bugs that we will have to fix in the short term. We would greatly appreciate help with:
Fixing root mount options (atomic-desktops#72): This is a long standing and mostly invisible bug that impacts performance.
Moving away from nss-altfiles (atomic-desktops#108): This is another long standing source of issues that new users regularly face.
Sealed images are now ready for testing! See the other article for all the details.
A lot of work is happening to make the transition to Bootable Containers as smooth as possible for our existing users. You can look at the roadmap for this transition at atomic-desktops#26.
One of the tasks is to move away from our unmaintained installation ISO building scripts to the new image-builder tooling.
This is planned for Fedora 45 for the ostree variants and support for Bootable Container will follow right after.
Another task is to start building the Fedora Atomic Desktops Bootable Container images using the Fedora Konflux instance.
We are looking for contributors to help us make the Fedora Atomic Desktops the best experience for Fedora users.
Some time ago I used a feature in KDE called “Run a command” when an event triggered. It triggered for me when a calendar event fired and used Piper TTS to read the event to me out loud. A small popup and a pling don’t work for me.
I tried to get the feature back into KDE, but since the merge request isn’t going anywhere and people don’t give details how to implement it correctly I wrote Sigrun now. It is named after a Norse Valkyrie and is short for Signal Run.
It is a systemd service running as a user and listening on DBus signals. Once it finds a configured one, it runs its command. The desktop doesn’t matter.
Here is the rule that reads my calendar reminders aloud via kde-tts.py:
[[rule]]
name = "calendar-tts"
[rule.event]
type = "notification"
[rule.filter]
app_name = "kalendarac"
summary = "Meeting.*"
[rule.filter.hints]
"x-kde-eventId" = "reminder"
[rule.action]
command = "/usr/local/bin/kde-tts.py"
args = ["-t", "{summary}", "-d", "{body}"]The Kdenlive team is happy to announce the first major release of 2026. This cycle focuses on stability, interface polish and usability improvements.
For the first time in Kdenlive's history, this version includes features implemented by so many different contributors. Our developer community is growing, don't hesitate to join us building a free and open source video editing program that respects the users privacy and provides a tool to democratize communication.
In case you missed it, check out the State of Kdenlive to learn more about the project's health and the nifty features coming soon.
This feature allows you to mirror any monitor while working in fullscreen mode. It’s especially useful when working with multiple displays or collaborating with others in the editing room, making it easier to share what you're doing without disrupting your workspace.
This release improves effect and transition workflow. We improved the logic for luma files to adapt to different project profiles and automatically reload previews of downloaded lumas. We added a dedicated tab in the transitions list to browse luma files and added the ability to drag-and-drop transitions directly onto timeline clips. This release also comes with a new Euclid Eraser transition and a Heatmap effect. We also fixed issues with audio TAP effects and improved internationalization support, fixed a bug in the Video Noise Generator effect and scaling issues in the Transform effect.
Transitions now include animated previews, making it much easier to visualize how they will look before applying them. Additionally, dropping a transition onto the timeline can now automatically adjust its duration to match the clips above and below, saving time and reducing manual tweaking. The first version of this feature was originally written by Swastik Patel, during KDE's SoK 2025.
Transitions automatically adjust to the length of the clip they are being applied to.
Added basic math expression support in effect spinboxes.
Added a new frei0r heatmap0r effect.
This release comes with many usability and workflow improvements to the timeline, such as a Disable Timeline Effects function to timeline hamburger menu, the ability to import and add clips directly from the timeline context menu with a smart length detection function, and sequences now have audio thumbnails. Other highlights include:
Hold the middle mouse button and drag to continuously pan the timeline even when going outside of the screen edges. Implemented by Abdias J Moya Perez.
Added the option to lock the playhead at the center of the timeline during playback, scrubbing or seeking. Implemented by Abdias J Moya Perez.
Added a button in the Status Bar to toggle between zooming to the mouse cursor position or the playhead position. Implemented by the mrfantastic.
Added the ability to change the playback speed of multiple clips at once either by directly ctrl + dragging in the timeline or by using the Clip Speed tool. Implemented by Vineet Tiwari.
Improved support for external recording devices, now the channel count and sample rate combo boxes only display values supported by the selected hardware. Also added the option to choose the sample format (8bit, 16bit, 32bit, and float) as well as a button to use the hardware's default settings.
This release fixes issues with cutting, moving, and saving subtitles, and solves a crash cutting subtitle clips. It also fixes a problem where searching for multiple words in the Speech Editor did not work correctly. We also improved the installation process for the Whisper and SeamlessM4T plugins and updated their requirements.
Releases are possible thanks to donations by the community. Donate now!
As usual, you will find very useful tips in our documentation website. You can also get help and exchange your ideas in our Kdenlive users Matrix chat room.
Kdenlive relies on its community, your help is always welcome. You can contribute by :
For the full changelog continue reading on kdenlive.org.
Welcome to a new issue of This Week in Plasma!
This week includes an interesting blend of improvements. Lots of visual stuff, so get ready for a ton of screenshots and screen recordings!
The Kicker Application Menu widget can now be configured to show a “Recent Locations” item. (Christoph Wolk, KDE Bugzilla #420951)
Network connections can now be duplicated. (Kartikeya Tyagi, KDE Bugzilla #499188)
There’s now a new “has parent window” window rule you can use to target child dialog windows specifically. (Kai Uwe Broulik, kwin MR #8969)
Dragging a search result for a System Settings page to the desktop now creates a launcher to that page as you would expect. This completes a mini-project to improve dragging-and-dropping things to the desktop that we started a while back. (Antti Savolainen, KDE Bugzilla #500259)
Discover now has fancier application page headers with more obvious install buttons! (Oliver Beard, discover MR #1297)
Discover no longer disables the “More Information” button on list items for in-progress updates. (Tobias Fella, KDE Bugzilla #431719)
Discover now does a better job of handling the rare case where an automatic update to a Flatpak app introduces a compatibility issue you can’t easily recover from. Now it will warn you about this once instead of continuously. (Tobias Fella, KDE Bugzilla #509760)
You can now drag favorite apps out of their areas in Kicker and Dashboard to un-favorite them. Kickoff is coming soon, too! (Christoph Wolk, plasma-desktop MR #3665 and KDE Bugzilla #518749)
When your laptop is plugged in at maximum charge, doing something that changes the power profile to a non-default one now shows only the power profile icon in the System Tray, and omits the fully-charged battery icon because that part is obvious. (Nate Graham, KDE Bugzilla #518802)
Improved some awkwardly-worded labels in System Settings and Plasma. (Philipp Kiemle, plasma-desktop MR #3674 and plasma-workspace MR #6522)
Reduced the amount of blurriness seen in icons throughout QtQuick-based apps using the Kirigami.Icon component when using a low fractional scale factor like 150% or less. (Volodymyr Zolotopupov, kirigami MR #2070)
Before:
Added a search provider for startpage.com, so you can search there from KRunner. (Antti Savolainen, KDE Bugzilla #503976)
Fixed a case where the Plasma Login Manager could crash when connecting and disconnecting multiple monitors while the login screen is visible. (David Edmundson, KDE Bugzilla #519302)
Fixed some cases where Plasma could crash at login. (David Edmundson, plasma-workspace MR #6520)
Fixed multiple accessibility issues: key repeat not working in the Orca screen reader, and various UI elements not being read properly. (Nicolas Fella, KDE Bugzilla #519143, KDE Bugzilla #519333, and KDE Bugzilla #519217)
Fixed a tricky issue in Spectacle that could make large images fail to automatically copy to the clipboard right after the app exits. (David Edmundson, KDE Bugzilla #509065)
Fixed another cause of the issue whereby de-focused full-screen windows could sometimes inappropriately appear at the top of the window stack. (Xaver Hugl, KDE Bugzilla #484155)
Fixed a layout glitch on System Settings’ Colors page that could make UI elements in the color previews overflow when using some non-default fonts and font sizes. (Akseli Lahtinen, KDE Bugzilla #516413)
Changing the brightness or any screen settings no longer terminates Spectacle’s sectangular region recordings. (Xaver Hugl, kwin MR #9127)
Fixed some visual glitches around radio buttons in the Audio Volume widget. (David Edmundson, ksvg MR #103)
Fixed an issue that made System Settings’ Touchscreen page appear while the “highlight changed settings” feature is enabled even if you don’t have a touchscreen. (Jin Liu, KDE Bugzilla #518868)
Turned on the “overlay planes” feature for Intel GPUs, which should improve performance and save some energy when using cooperative games and apps. (Xaver Hugl, kwin MR #8699)
Improved power efficiency for full-screen windows and effects that don’t gain any benefit from using the “direct scan-out” feature; now they’ll only use it if it will save power. (Xaver Hugl, kwin MR #9120)
KDE has become important in the world, and your time and contributions have helped us get there. As we grow, we need your support to keep KDE sustainable.
Would you like to help put together this weekly report? Introduce yourself in the Matrix room and join the team!
Beyond that, you can help KDE by directly getting involved in any other projects. Donating time is actually more impactful than donating money. Each contributor makes a huge difference in KDE — you are not a number or a cog in a machine! You don’t have to be a programmer, either; many other opportunities exist.
You can also help out by making a donation! This helps cover operational costs, salaries, travel expenses for contributors, and in general just keeps KDE bringing Free Software to the world.
Push a commit to the relevant merge request on invent.kde.org.
As an exception to my usual posts, this time I’ll write about a feature that’s already released. Since Plasma 6.6, you can enable automatic brightness in the display settings… let’s take a look at how it works, and why it took so long to make it happen.
This is where the problems start - most laptops unfortunately don’t come with a brightness sensor, and there’s effectively no monitors that have a built-in sensor either (let alone one that can be accessed by the connected PC).
While it’s possible to buy or build a brightness sensor that connects via USB, brightness control for external monitors usually has limitations in how often we can safely adjust the brightness… So for quite some time, there was noone working on Plasma that had the combination of hardware, motivation and knowledge to do something about it.
Luckily, the Framework Laptop 13 comes with a brightness sensor, so on the hardware side I was all set:
Making automatic brightness do something is easy, but making it work well enough that you actually want to use it is a very different story.
My first approach was to assume brightness of the display should scale linearly with environmental brightness. I tried this, and it was sort of usable, but just not good enough. There’s three problems with it:
So a different approach was needed. I looked a bit at other operating systems for inspiration, and from the UX side I definitely wanted to copy Android: You use the brightness slider however you want, and the system should try to replicate what you do on its own. On the implemtation side however, I only saw claims that it uses machine learning, so that wasn’t exactly helpful.
Ultimately, what I settled on is pretty simple: We just store 6 sensor values, one per 20% brightness step. When processing sensor readings, KWin finds the matching brightness setting by linearly interpolating between the two closest sensor values.
When the user touches the brightness slider or uses the brightness shortcuts, KWin adjusts the curve so the current sensor value will result in that desired brightness setting. At first, I only made it adjust the two closest points and enforce the rest of the curve to be monotonic, but it ended up causing problems:
With this configuration, any sensor readings above 100 lux - for example, 101 - resulted in 100% brightness. To fix that, we now enforce a minimum difference of at least 1 lux or 10% between control points, so the curve above would look more like
To ensure the curve would always stay strictly monotonic and so you can have an arbitrary backlight setting at zero lux, values below zero also had to be allowed when updating the curve with those constraints.
However, I was still not done. While it now followed my preferences pretty well, it was still annoying! Especially if you sit in between a light source and the laptop, or if you sit on a train with trees around the tracks (like I recently did, traveling to and from Graz for a Plasma sprint), the brightness constantly fluctuated up and down.
To make it less annoying, I added some more adjustments on top:
This is how it was released in Plasma 6.6, and I’m pretty happy with it on both the Framework 13 and on my OnePlus 6 with Plasma mobile.
Just because I’m happy with it, doesn’t mean it’s done. If you’re using automatic brightness and it’s still annoying you for some reason, please tell me about it! If you’re really happy with it, I won’t complain about being told that either of course ;)
To fully catch up to what phones have done for years though, one feature is still missing: I’d like to adjust not just the brightness, but also the white point of the display to the environment. Unfortunately, none of my devices have a sensor for that… but since the camera module in the Framework 13 is easy to replace, I hope that changes one day!
As many long running projects, Qt too over the years has accumulated some APIs that in hindsight are deemed unsafe or sub-optimal. For example, Qt by default implicitly converts const char* to QString. While that usually only incurs a runtime overhead, maybe encoding problems, but also admittedly less cluttered code, there’s other APIs that can backfire in more subtle ways. One such API is doing a “context-less connect”.
Signals and Slots are a core principle of Qt that make it super easy to connect one object to another and keep a certain separation of concerns. The typical syntax to establish an connection is:
connect(sender, &Foo::somethingHappened, receiver, &Bar::doStuff);This connects the signal somethingHappened in our sender of Type Foo to the member function doStuff in our receiver (context object) of type Bar. Whenever sender “emits” somethingHappened, doStuff on receiver will be called. The neat part about Qt connections is that when receiver gets destroyed, the connection is severed automatically. However, you can not just connect a signal to a member function but also use a lambda:
connect(job, &WallpaperFinder::wallpaperFound, [this](const QString &path) {
m_wallpapers << path;
});In our hypothetical wallpaper selector when the “job” that goes looking for wallpapers found one, it emits a signal and tells us the path of the file, so we can show it to the user. Now what happens when the user closes the dialog (which then gets destroyed) before the job has finished? Well… job still emits the signal which then results in our lambda being called. And then we try to access m_wallpapers on this which is long gone. Boom!
The fix is easy: provide a “context object”, too, just like you would with pointer to member function:
connect(job, &WallpaperFinder::wallpaperFound, this, [this](const QString &path) {
m_wallpapers << path;
});If this gets destroyed, the connection is severed, our lambda will no longer be called and all is well. The receiver object also decides what thread the slot is called, i.e. it will be called in the thread the receivers “lives in”. Context-less connections are always of DirectConnection type. Since Qt 6.7 you can actually enforce the use of a context object by defining QT_NO_CONTEXTLESS_CONNECT.
add_definitions(-DQT_NO_CONTEXTLESS_CONNECT)It requires you think about the lifetime of your objects more and make a conscious decision about what your context object is. It also removes one thing to look out for in code review. I started adding this option to a couple KDE repositories to improve the quality of our code and I encourage you to do that, too! It probably comes to no surprise that in general, the bigger and older a repository, the higher the probability of it using non-ideal code.
Of course, the situation is not always as simple as our example above. Here’s a few tips and tricks:
connect(job, &Job::finished, manager, [job, manager] {
manager->report(job);
});[foo = m_foo] { ... }connect(job, &Job::finished, [timer] {
timer->start();
});connect(job, &Job::finished, timer, &QTimer::start);I’m a huge fan of QT_ENABLE_STRICT_MODE_UP_TO that lets you turn on most strictness features in a single shot. The biggest hurdle of rolling that out more widespread in KDE repositories is actually the Java-style iterators. Qt hates them, many use them, particularly for mutating a container, and imho they’re much more pleasing to look at than STL algorithms. If you start a new project, however, do consider setting your baseline to be as strict as it can be!