[kwin] [Bug 487098] KWin on X11: No compositing on startup, cannot be re-enabled
https://bugs.kde.org/show_bug.cgi?id=487098 Andrew Gunnerson changed: What|Removed |Added CC||accounts+...@chiller3.com --- Comment #1 from Andrew Gunnerson --- This was a Fedora-specific issue and should be fixed in the `kwin-x11-6.0.4.1-3.fc40` update: https://bugzilla.redhat.com/show_bug.cgi?id=2280899 -- You are receiving this mail because: You are watching all bug changes.
[dolphin] [Bug 485404] New: Terminal captures Tab key, but not Shift+Tab
https://bugs.kde.org/show_bug.cgi?id=485404 Bug ID: 485404 Summary: Terminal captures Tab key, but not Shift+Tab Classification: Applications Product: dolphin Version: 24.02.1 Platform: Other OS: Linux Status: REPORTED Severity: normal Priority: NOR Component: panels: terminal Assignee: dolphin-bugs-n...@kde.org Reporter: accounts+...@chiller3.com CC: kfm-de...@kde.org Target Milestone: --- SUMMARY When the terminal inside Dolphin is focused, it successfully captures the Tab key, but Shift+Tab is not captured. Pressing Shift+Tab causes the focus to move away from the terminal component and into the free disk space widget in the bottom right corner. This prevents certain functionality in CLI applications from working. For example, zsh allows using Shift+Tab to move backwards in a tab completion list. STEPS TO REPRODUCE 1. Open Dolphin 2. Open the terminal with F4 3. Press Tab (it is sent to the shell) 4. Press Shift+Tab OBSERVED RESULT The focus shifts away from the terminal component. EXPECTED RESULT Shift+Tab is sent to the command running in the terminal. SOFTWARE/OS VERSIONS Linux/KDE Plasma: Fedora 40 KDE Plasma Version: 6.0.3-2.fc40.x86_64 KDE Frameworks Version: 6.0.0-1.fc40.x86_64 Qt Version: 6.6.2-1.fc40.x86_64 Dolphin Version: 24.02.1-1.fc40.x86_64 -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kded] [Bug 466387] New: Intermittent segfault when plugging in PS5 or PS4 controller
https://bugs.kde.org/show_bug.cgi?id=466387 Bug ID: 466387 Summary: Intermittent segfault when plugging in PS5 or PS4 controller Classification: Frameworks and Libraries Product: frameworks-kded Version: 5.103.0 Platform: Other OS: Linux Status: REPORTED Severity: normal Priority: NOR Component: general Assignee: fa...@kde.org Reporter: accounts+...@chiller3.com CC: kdelibs-b...@kde.org Target Milestone: --- SUMMARY When I plug in my PS5 or PS4 controllers via USB, kded will sometimes segfault. I have not changed any settings other than disabling the controller's touchpad from KDE's system settings. It happens pretty frequently, but not always. There seems to be a higher chance of a crash after resuming from suspend, but I don't have any hard data on that. STEPS TO REPRODUCE 1. Plug in PS5 or PS4 controller via USB OBSERVED RESULT kded segfaults (which triggers side effects related to that, like night color breaking) EXPECTED RESULT kded shouldn't crash SOFTWARE/OS VERSIONS Linux/KDE Plasma: Fedora 37 (x11 session) KDE Plasma Version: 5.27.0 KDE Frameworks Version: 5.103.0 Qt Version: 5.15.8 ADDITIONAL INFORMATION journalctl messages starting from plugging in the controller to the crash: Feb 24 21:38:02 cxl-desktop-1 kernel: usb 1-9: new high-speed USB device number 22 using xhci_hcd Feb 24 21:38:02 cxl-desktop-1 kernel: usb 1-9: New USB device found, idVendor=054c, idProduct=0ce6, bcdDevice= 1.00 Feb 24 21:38:02 cxl-desktop-1 kernel: usb 1-9: New USB device strings: Mfr=1, Product=2, SerialNumber=0 Feb 24 21:38:02 cxl-desktop-1 kernel: usb 1-9: Product: Wireless Controller Feb 24 21:38:02 cxl-desktop-1 kernel: usb 1-9: Manufacturer: Sony Interactive Entertainment Feb 24 21:38:02 cxl-desktop-1 kernel: input: Sony Interactive Entertainment Wireless Controller as /devices/pci:00/:00:14.0/usb1/1-9/1-9:1.3/0003:054C:0CE6.0022/input/input47 Feb 24 21:38:02 cxl-desktop-1 kernel: hid-generic 0003:054C:0CE6.0022: input,hidraw15: USB HID v1.11 Gamepad [Sony Interactive Entertainment Wireless Controller] on usb-:00:14.0-9/input3 Feb 24 21:38:02 cxl-desktop-1 mtp-probe[275156]: checking bus 1, device 22: "/sys/devices/pci:00/:00:14.0/usb1/1-9" Feb 24 21:38:02 cxl-desktop-1 mtp-probe[275156]: bus: 1, device: 22 was not an MTP device Feb 24 21:38:02 cxl-desktop-1 kernel: playstation 0003:054C:0CE6.0022: hidraw15: USB HID v1.11 Gamepad [Sony Interactive Entertainment Wireless Controller] on usb-:00:14.0-9/input3 Feb 24 21:38:02 cxl-desktop-1 kernel: input: Sony Interactive Entertainment Wireless Controller as /devices/pci:00/:00:14.0/usb1/1-9/1-9:1.3/0003:054C:0CE6.0022/input/input48 Feb 24 21:38:02 cxl-desktop-1 kernel: input: Sony Interactive Entertainment Wireless Controller Motion Sensors as /devices/pci:00/:00:14.0/usb1/1-9/1-9:1.3/0003:054C:0CE6.0022/input/input49 Feb 24 21:38:02 cxl-desktop-1 kernel: input: Sony Interactive Entertainment Wireless Controller Touchpad as /devices/pci:00/:00:14.0/usb1/1-9/1-9:1.3/0003:054C:0CE6.0022/input/input50 Feb 24 21:38:02 cxl-desktop-1 kernel: playstation 0003:054C:0CE6.0022: Registered DualSense controller hw_version=0x0313 fw_version=0x0104000c Feb 24 21:38:02 cxl-desktop-1 mtp-probe[275167]: checking bus 1, device 22: "/sys/devices/pci:00/:00:14.0/usb1/1-9" Feb 24 21:38:02 cxl-desktop-1 mtp-probe[275167]: bus: 1, device: 22 was not an MTP device Feb 24 21:38:02 cxl-desktop-1 kded5[91269]: Touchpad reset Feb 24 21:38:03 cxl-desktop-1 kded5[91269]: KCrash: Attempting to start /usr/bin/kded5 Feb 24 21:38:03 cxl-desktop-1 kded5[91269]: KCrash: Application 'kded5' crashing... Feb 24 21:38:03 cxl-desktop-1 kded5[91269]: KCrash: Attempting to start /usr/libexec/drkonqi Backtrace from gdb (with debuginfo): Core was generated by `/usr/bin/kded5'. Program terminated with signal SIGSEGV, Segmentation fault. #0 __pthread_kill_implementation (threadid=, signo=signo@entry=11, no_tid=no_tid@entry=0) at pthread_kill.c:44 Downloading 0.00 MB source file /usr/src/debug/glibc-2.36-9.fc37.x86_64/nptl/pthread_kill.c 44return INTERNAL_SYSCALL_ERROR_P (ret) ? INTERNAL_SYSCALL_ERRNO (ret) : 0; [Current thread is 1 (Thread 0x7f5ac256c840 (LWP 275201))] (gdb) bt #0 __pthread_kill_implementation (threadid=, signo=signo@entry=11, no_tid=no_tid@entry=0) at pthread_kill.c:44 #1 0x7f5ac1aafec3 in __pthread_kill_internal (signo=11, threadid=) at pthread_kill.c:78 #2 0x7f5ac1a5fa76 in __GI_raise (sig=11) at ../sysdeps/posix/raise.c:26 #3 0x7f5ac374b7b1 in KCrash::defaultCrashHandler(int) (sig=11) at /usr/src/debug/kf5-kcrash-5.103.0-1.fc37.x86_64/src/kcrash.cpp:611 #4 0x7f5ac1a5fb20 in () at /lib64/libc.so.6 #5 0x7f5ac1ef14b9 in XPending (dpy=0x7f5a9a94e4a0 ) at
[KScreen] [Bug 464590] kded5 crashes in KScreen::Config::setOutputPriority() after monitor wakes from sleep
https://bugs.kde.org/show_bug.cgi?id=464590 Andrew Gunnerson changed: What|Removed |Added CC||accounts+...@chiller3.com -- You are receiving this mail because: You are watching all bug changes.
[KScreen] [Bug 466036] kded5 crash after resume from blank
https://bugs.kde.org/show_bug.cgi?id=466036 Andrew Gunnerson changed: What|Removed |Added CC||accounts+...@chiller3.com --- Comment #3 from Andrew Gunnerson --- I've having the same issue as well after the Plasma 5.27.0 update. When this occurs, the screen also stays blank until I switch to another VT and then back. Software info: * Distro: Fedora 37 * Session type: X11 * kernel: 6.1.11-200.fc37.x86_64 * kf5: 5.103.0-1.fc37.x86_64 * plasma: 5.27.0-2.fc37.x86_64 * qt5: 5.15.8-5.fc37.x86_64 * Nvidia proprietary drivers: 525.89.02-1.fc37.x86_64 Backtrace with debuginfo: Core was generated by `/usr/bin/kded5'. Program terminated with signal SIGSEGV, Segmentation fault. #0 KScreen::Output::id (this=0x0) at /usr/src/debug/libkscreen-qt5-5.27.0-1.fc37.x86_64/src/output.cpp:189 Downloading 0.00 MB source file /usr/src/debug/libkscreen-qt5-5.27.0-1.fc37.x86_64/src/output.cpp 189 return d->id; [Current thread is 1 (Thread 0x7f3ae576c840 (LWP 593267))] (gdb) bt #0 KScreen::Output::id() const (this=0x0) at /usr/src/debug/libkscreen-qt5-5.27.0-1.fc37.x86_64/src/output.cpp:189 #1 0x7f3abc09110c in KScreen::Config::setOutputPriority(QSharedPointer const&, unsigned int) (this=0x558343e4d3f0, output=..., priority=1) at /usr/include/qt5/QtCore/qsharedpointer_impl.h:301 #2 0x7f3abc09aa5e in KScreen::SetConfigOperationPrivate::fixPrimaryOutput() (this=) at /usr/include/qt5/QtCore/qsharedpointer_impl.h:301 #3 KScreen::SetConfigOperation::start() (this=0x558343ae4550) at /usr/src/debug/libkscreen-qt5-5.27.0-1.fc37.x86_64/src/setconfigoperation.cpp:114 #4 0x7f3ae54c8134 in QObject::event(QEvent*) (this=0x558343ae4550, e=0x558343e21070) at kernel/qobject.cpp:1347 #5 0x7f3ae63aed62 in QApplicationPrivate::notify_helper(QObject*, QEvent*) (this=, receiver=0x558343ae4550, e=0x558343e21070) at kernel/qapplication.cpp:3640 #6 0x7f3ae549d4e8 in QCoreApplication::notifyInternal2(QObject*, QEvent*) (receiver=0x558343ae4550, event=0x558343e21070) at kernel/qcoreapplication.cpp:1064 #7 0x7f3ae54a0854 in QCoreApplicationPrivate::sendPostedEvents(QObject*, int, QThreadData*) (receiver=0x0, event_type=0, data=0x5583433de340) at kernel/qcoreapplication.cpp:1821 #8 0x7f3ae54eeb07 in postEventSourceDispatch(GSource*, GSourceFunc, gpointer) (s=0x5583434bfe20) at kernel/qeventdispatcher_glib.cpp:277 #9 0x7f3ae4264cbf in g_main_dispatch (context=0x7f3acc005010) at ../glib/gmain.c:3444 #10 g_main_context_dispatch (context=0x7f3acc005010) at ../glib/gmain.c:4162 #11 0x7f3ae42ba598 in g_main_context_iterate.constprop.0 (context=0x7f3acc005010, block=1, dispatch=1, self=) at ../glib/gmain.c:4238 #12 0x7f3ae4261f40 in g_main_context_iteration (context=0x7f3acc005010, may_block=1) at ../glib/gmain.c:4303 #13 0x7f3ae54ee5fa in QEventDispatcherGlib::processEvents(QFlags) (this=0x5583434bf6e0, flags=...) at kernel/qeventdispatcher_glib.cpp:423 #14 0x7f3ae549bf3a in QEventLoop::exec(QFlags) (this=this@entry=0x7fff5b65c820, flags=..., flags@entry=...) at ../../include/QtCore/../../src/corelib/global/qflags.h:69 #15 0x7f3ae54a4002 in QCoreApplication::exec() () at ../../include/QtCore/../../src/corelib/global/qflags.h:121 #16 0x7f3ae595fad0 in QGuiApplication::exec() () at kernel/qguiapplication.cpp:1863 #17 0x7f3ae63aecd9 in QApplication::exec() () at kernel/qapplication.cpp:2832 #18 0x55834213d5ab in main(int, char**) (argc=, argv=) at /usr/src/debug/kf5-kded-5.103.0-1.fc37.x86_64/src/kded.cpp:785 -- You are receiving this mail because: You are watching all bug changes.
[dolphin] [Bug 465357] Dolphin's UI freezes when stat() is slow
https://bugs.kde.org/show_bug.cgi?id=465357 --- Comment #10 from Andrew Gunnerson --- Woohoo! Any chance we could leave this issue open? The patch is good enough for my use case, but the general problem still exists. -- You are receiving this mail because: You are watching all bug changes.
[dolphin] [Bug 465357] Dolphin's UI freezes when stat() is slow
https://bugs.kde.org/show_bug.cgi?id=465357 --- Comment #8 from Andrew Gunnerson --- Thanks for the info, Dāvis! Looking forward to the upcoming changes :) I'm quite new to the KIO and Dolphin code bases, but happy to help wherever I can if needed. -- You are receiving this mail because: You are watching all bug changes.
[dolphin] [Bug 465357] Dolphin's UI freezes when stat() is slow
https://bugs.kde.org/show_bug.cgi?id=465357 --- Comment #5 from Andrew Gunnerson --- I submitted an MR for skipping directory counts when KFileItem::isSlow() here: https://invent.kde.org/system/dolphin/-/merge_requests/507 --- Regarding the slow stat() calls blocking the UI, I did some further investigation. The stat() calls are coming from KIO::WorkerThread threads, not the UI thread. The stack trace looks something like this: Thread 14 (Thread 0x7fffbd7fa6c0 (LWP 255122) "KIO::WorkerThre"): #0 stat (pathname=0x7fff9c00abe8 "/home/chenxiaolong", statbuf=0x7fffbd7f9780) at slow_stat.cpp:35 #1 0x76f16646 in acl_get_file (path_p=0x7fff9c00abe8 "/home/chenxiaolong", type=16384) at libacl/acl_get_file.c:73 #2 0x7fffd80be93e in createUDSEntry(QString const&, QByteArray const&, KIO::UDSEntry&, QFlags, QString const&) [clone .lto_priv.0] () from /usr/lib64/qt5/plugins/kf5/kio/kio_file.so #3 0x7fffd80c64c6 in FileProtocol::stat(QUrl const&) () from /usr/lib64/qt5/plugins/kf5/kio/kio_file.so #4 0x7790ff15 in KIO::SlaveBase::dispatch(int, QByteArray const&) () from /lib64/libKF5KIOCore.so.5 #5 0x7790869e in KIO::SlaveBase::dispatchLoop() () from /lib64/libKF5KIOCore.so.5 #6 0x77983553 in KIO::WorkerThread::run() () from /lib64/libKF5KIOCore.so.5 #7 0x75ae8507 in QThreadPrivate::start (arg=0x55683ff0) at thread/qthread_unix.cpp:330 #8 0x754ae12d in start_thread (arg=) at pthread_create.c:442 #9 0x7552fbc0 in clone3 () at ../sysdeps/unix/sysv/linux/x86_64/clone3.S:81 Since the UI is freezing despite the slow stat calls being on another thread, I'm assuming there's a StatJob::exec()/KJob::exec() call somewhere forcing the UI to wait for its completion. Unfortunately, I've had no luck at all tracking down where the job came from. -- You are receiving this mail because: You are watching all bug changes.
[dolphin] [Bug 465357] Dolphin's UI freezes when stat() is slow
https://bugs.kde.org/show_bug.cgi?id=465357 --- Comment #3 from Andrew Gunnerson --- (In reply to Nate Graham from comment #2) > > Specifically, when I try to access a `cifs` mount point > For just this reason, it's not recommended to manually mount network shares. > Is there a reason why you have to do this, and accessing the share from its > network URL when needed doesn't work for you? Yeah, unfortunately, kio smb:// isn't quite fast enough for my (admittedly unusual) use case. I'm connecting to a very high bandwidth server and frequently access files both from GUI apps and the CLI. With kernel cifs mounts, I get around ~3 GiB/s sequential reads/writes, and Dolphin + smb:// only hits around ~500 MiB/s and kio-fuse around ~80 MiB/s. > Regardless, skipping directory counts for known slow paths would probably > help indeed. Can you please submit it at > https://invent.kde.org/system/dolphin/-/merge_requests? Sure thing! -- You are receiving this mail because: You are watching all bug changes.
[dolphin] [Bug 465357] Dolphin's UI freezes when stat() is slow
https://bugs.kde.org/show_bug.cgi?id=465357 --- Comment #1 from Andrew Gunnerson --- Created attachment 155993 --> https://bugs.kde.org/attachment.cgi?id=155993=edit A tiny library to intentionally slow down stat() for testing -- You are receiving this mail because: You are watching all bug changes.
[dolphin] [Bug 465357] New: Dolphin's UI freezes when stat() is slow
https://bugs.kde.org/show_bug.cgi?id=465357 Bug ID: 465357 Summary: Dolphin's UI freezes when stat() is slow Classification: Applications Product: dolphin Version: 22.12.2 Platform: Other OS: Linux Status: REPORTED Severity: normal Priority: NOR Component: general Assignee: dolphin-bugs-n...@kde.org Reporter: accounts+...@chiller3.com CC: kfm-de...@kde.org Target Milestone: --- Created attachment 155992 --> https://bugs.kde.org/attachment.cgi?id=155992=edit A patch to skip directory counts when KFileItem::isSlow SUMMARY When stat() does not complete quickly, Dolphin's UI freezes until it returns. This seems to be the root cause for the freezes I'm encountering. Specifically, when I try to access a `cifs` mount point, Dolphin kicks off `KDirectoryContentsCounter.scanDirectory()` for all the subdirectories visible on screen. The opendir/readdir calls are enough to slow down the server to the point where each stat takes a couple of seconds. Once stat starts slowing down, the Dolphin's UI stops responding. I've been working around the problem by making Dolphin skip `KDirectoryContentsCounter.scanDirectory()` if `KFileItem::isSlow()` is true (patch attached). Even though this hides the issue, I wonder if it may be good to have anyway since it matches the behavior of kio smb://, where directory counts don't happen. STEPS TO REPRODUCE To avoid the need to connect to a slow server, I've attached an LD_PRELOAD'able library to help reproduce this. It slows down all stat calls to paths under /home/, except for paths containing "/." so that Dolphin's startup isn't slowed down when reading configs. 1. $ g++ -shared -fPIC slow_stat.cpp -o slow_stat.so 2. $ LD_PRELOAD=$(pwd)/slow_stat.so dolphin 3. Navigate to somewhere under /home/ 4. Try to perform UI actions, like hovering over items, selecting, right clicking, opening menus, etc. OBSERVED RESULT Dolphin's UI freezes until stat returns EXPECTED RESULT Dolphin should ideally remain responsive while stat occurs SOFTWARE/OS VERSIONS Windows: n/a macOS: n/a Linux/KDE Plasma: Fedora 37 KDE Plasma Version: 5.26.5 KDE Frameworks Version: 5.102.0 Qt Version: 5.15.8 ADDITIONAL INFORMATION (Patch for my workaround is attached) I've also tried compiling the latest commit in dolphin (580f3bf0d5c8735cf230208fd6e4b288a3e55722) and there's no change in behavior. -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 423499] KFileItem::iconName ignorant of isSlow() and can block gui thread
https://bugs.kde.org/show_bug.cgi?id=423499 Andrew Gunnerson changed: What|Removed |Added CC||accounts+...@chiller3.com -- You are receiving this mail because: You are watching all bug changes.
[kwin] [Bug 452119] Poor performance on intel igp on wayland versus Xorg
https://bugs.kde.org/show_bug.cgi?id=452119 Andrew Gunnerson changed: What|Removed |Added CC||accounts+...@chiller3.com -- You are receiving this mail because: You are watching all bug changes.
[kiofuse] [Bug 432627] drag and drop from remote locations (dolphin/kio-fuse) to apps not working
https://bugs.kde.org/show_bug.cgi?id=432627 Andrew Gunnerson changed: What|Removed |Added CC||accounts+...@chiller3.com -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 463989] Remote RAW image previews don't work because of missing file extension in thumbnail temp file
https://bugs.kde.org/show_bug.cgi?id=463989 --- Comment #2 from Andrew Gunnerson --- In case approach #2 is acceptable, I've created an MR to implement that here: https://invent.kde.org/frameworks/kio/-/merge_requests/1103 -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 463989] New: Remote RAW image previews don't work because of missing file extension in thumbnail temp file
https://bugs.kde.org/show_bug.cgi?id=463989 Bug ID: 463989 Summary: Remote RAW image previews don't work because of missing file extension in thumbnail temp file Classification: Frameworks and Libraries Product: frameworks-kio Version: 5.101.0 Platform: Other OS: Linux Status: REPORTED Severity: normal Priority: NOR Component: general Assignee: kio-bugs-n...@kde.org Reporter: accounts+...@chiller3.com CC: kdelibs-b...@kde.org Target Milestone: --- SUMMARY In Dolphin, when browsing a directory containing RAW image files (eg. Canon's .cr2), previews work for local files, but not remote files. I believe I've tracked down the reason: On the Dolphin side, when it runs the PreviewJob, it takes this code path for remote files: https://invent.kde.org/frameworks/kio/-/blob/v5.101.0/src/widgets/previewjob.cpp#L742. This creates a file called /tmp/dolphin.XX, which is passed to the thumbnailer. On the kioslave/thumbnail.so side, RAWCreator::create() is called as expected (https://invent.kde.org/graphics/kdegraphics-thumbnailers/-/blob/v22.12.0/raw/rawcreator.cpp#L47), which then calls https://invent.kde.org/graphics/libkdcraw/-/blob/v22.12.0/src/kdcraw.cpp#L87. KDcraw::loadEmbeddedPreview() returns false because of the `!rawFilesExt.toUpper().contains(ext)` condition. libkdcraw requires the filename to have a known file extension. Off the top of my head, it seems like there are a couple ways to solve this: 1. Change libkdcraw to not depend on the file extension (not sure how feasible this is). 2. Change kio so that the temporary file it creates has the original file extension. STEPS TO REPRODUCE 1. In Dolphin, change the settings for the max file limit for remote thumbnails. 2. Navigate to a remote folder containing RAW images. OBSERVED RESULT There are no thumbnails. EXPECTED RESULT Thumbnails should be shown. SOFTWARE/OS VERSIONS Linux/KDE Plasma: Fedora 37 KDE Plasma Version: 5.26.4 KDE Frameworks Version: 5.101.0 Qt Version: 5.15.7 ADDITIONAL INFORMATION (Included above) -- You are receiving this mail because: You are watching all bug changes.
[dolphin] [Bug 463952] Moving a file within the same remote share results in a copy (via kio-fuse) + delete
https://bugs.kde.org/show_bug.cgi?id=463952 --- Comment #1 from Andrew Gunnerson --- I did a bit more testing to try and track this down. The latest commit in `master` (https://invent.kde.org/system/dolphin/-/commit/1c4baff760c7975d17381f72029a145943b32ff6) still has this problem. Doing a git bisect shows that https://invent.kde.org/system/dolphin/-/commit/8d7e600f63a1961294dfe2c278a710b4ce0716e9 is the commit that first introduced the problem. Reverting that commit on top of `master` fixes the problem. The `release/22.12` branch also works because of the revert at https://invent.kde.org/system/dolphin/-/merge_requests/479. -- You are receiving this mail because: You are watching all bug changes.
[dolphin] [Bug 463952] New: Moving a file within the same remote share results in a copy (via kio-fuse) + delete
https://bugs.kde.org/show_bug.cgi?id=463952 Bug ID: 463952 Summary: Moving a file within the same remote share results in a copy (via kio-fuse) + delete Classification: Applications Product: dolphin Version: 22.12.0 Platform: Other OS: Linux Status: REPORTED Severity: normal Priority: NOR Component: general Assignee: dolphin-bugs-n...@kde.org Reporter: accounts+...@chiller3.com CC: kfm-de...@kde.org Target Milestone: --- SUMMARY When trying to move a large file on a remote share into a different directory within the same share, dolphin doesn't perform a move operation. Instead, it's doing a copy via kio-fuse and then a delete. This seems to only happen in dolphin. Doing the move operation via kioclient works fine: kioclient --interactive move smb://hostname/home/large_file.iso smb://hostname/home/new_folder/ STEPS TO REPRODUCE 1. Navigate to a remote share in Dolphin (eg. smb://hostname) 2. Cut a large file 3. Navigate to another directory in the same share 4. Paste the file OBSERVED RESULT Dolphin pops up a notification showing that it's moving the file from /run/user/1000/kio-fuse-XX/smb/hostname/home/large_file.iso to smb://hostname/home/new_folder/large_file.iso. From the speed and network bandwidth, it's apparent that the data is being downloaded and reuploaded. EXPECTED RESULT kio-fuse shouldn't be involved in a move operation across the same remote share. SOFTWARE/OS VERSIONS Linux/KDE Plasma: Fedora 37 KDE KDE Plasma Version: 5.26.4 KDE Frameworks Version: 5.101.0 Qt Version: 5.15.7 ADDITIONAL INFORMATION kio-fuse Version: 5.0.1-4.fc37.x86_64 dolphin Version: 22.12.0-1.fc37.x86_64 -- You are receiving this mail because: You are watching all bug changes.
[konsole] [Bug 372116] Feature Request: Support OSC 52 (copy to clipboard)
https://bugs.kde.org/show_bug.cgi?id=372116 Andrew Gunnerson changed: What|Removed |Added CC||accounts+...@chiller3.com -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 459616] DesktopExecParser: Some terminals require "-e " while others just take the command directly
https://bugs.kde.org/show_bug.cgi?id=459616 --- Comment #3 from Andrew Gunnerson --- I went through Arch Wiki's list of terminal emulators and tested most of them: https://wiki.archlinux.org/title/List_of_applications#Terminal_emulators Looks like there are many different ways with how they accept the commands :( Requires `-e` where `-e` signals the end of argument parsing and all further arguments specify the command to run: - alacritty - konsole - lxterminal - mlterm - putty (documented in the help output, but the argument parser fails to handle -e) - qterminal (fails to run command if it contains arguments starting with dash. Eg. `-c` in `qterminal -e bash -c 'sleep 5'`) - roxterm - tilix - urxvt - xterm Requires `-e`, but requires the command to be specified as a single string: - guake - mate-terminal - pantheon-terminal (does not exit after command completion) - terminator - xfce4-terminal Requires no flag after the Exec= command in the .desktop file: - gnome-terminal - kitty - wezterm Requires no flag, but also supports `-e` optionally: - st Don't support running a command at all: - liri-terminal - yakuake -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 459616] DesktopExecParser: Some terminals require "-e " while others just take the command directly
https://bugs.kde.org/show_bug.cgi?id=459616 --- Comment #2 from Andrew Gunnerson --- Unfortunately, yes. With xterm, it fails when parsing the arguments without `-e`. With Konsole, konsole -e bash -c 'echo abc' runs `bash -c 'echo abc'` as expected, but: konsole -- bash -c 'echo abc' seems to just ignore the extra arguments. (And arguments cannot be parsed without the `--`.) -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 459616] New: DesktopExecParser: Some terminals require "-e " while others just take the command directly
https://bugs.kde.org/show_bug.cgi?id=459616 Bug ID: 459616 Summary: DesktopExecParser: Some terminals require "-e " while others just take the command directly Classification: Frameworks and Libraries Product: frameworks-kio Version: 5.98.0 Platform: Other OS: Linux Status: REPORTED Severity: normal Priority: NOR Component: general Assignee: kio-bugs-n...@kde.org Reporter: accounts+...@chiller3.com CC: kdelibs-b...@kde.org Target Milestone: --- SUMMARY It seems that terminal emulators generally allow spawning an application inside a new window with either " -e " or just " ". Currently, only the former (with "-e") is supported in DesktopExecParser. This causes "Terminal=true" desktop files to fail to launch when the default terminal application is set to one that doesn't use/support "-e". STEPS TO REPRODUCE 1. Set the default terminal to an application that doesn't use "-e" (eg. wezterm or gnome-terminal) 2. Create a desktop file with "Terminal=true" 3. Launch the desktop file OBSERVED RESULT The default terminal fails to launch with some error message about "-e" being an unknown argument (or similar). EXPECTED RESULT The default terminal should launch successfully with the specified command. SOFTWARE/OS VERSIONS Windows: n/a macOS: n/a Linux/KDE Plasma: Fedora 36 KDE Plasma Version: 5.25.5 KDE Frameworks Version: 5.98.0 Qt Version: 5.15.5 ADDITIONAL INFORMATION I'm not sure there's a clean/foolproof way to make this work. I don't think there's any sort of metadata (in the .desktop or similar) that describes how a terminal expects a command to be passed to it. -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 459408] DesktopExecParser fails to find terminal if TerminalApplication has multiple arguments
https://bugs.kde.org/show_bug.cgi?id=459408 Andrew Gunnerson changed: What|Removed |Added Version|5.97.0 |5.98.0 -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 459408] DesktopExecParser fails to find terminal if TerminalApplication has multiple arguments
https://bugs.kde.org/show_bug.cgi?id=459408 Andrew Gunnerson changed: What|Removed |Added Resolution|FIXED |--- Status|RESOLVED|REOPENED --- Comment #3 from Andrew Gunnerson --- Thanks for the reply, Nate! Fedora 36 just got the 5.98 update today so I gave it a try again and it seems like it's still an issue (I'll update the version field in this bug report). It doesn't look like DesktopExecParser was changed between 5.97 and 5.98. -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 459408] DesktopExecParser fails to find terminal if TerminalApplication has multiple arguments
https://bugs.kde.org/show_bug.cgi?id=459408 Andrew Gunnerson changed: What|Removed |Added Summary|DesktopExecParser fails to |DesktopExecParser fails to |launch terminal if |find terminal if |TerminalApplication has |TerminalApplication has |multiple arguments |multiple arguments -- You are receiving this mail because: You are watching all bug changes.
[frameworks-kio] [Bug 459408] New: DesktopExecParser fails to launch terminal if TerminalApplication has multiple arguments
https://bugs.kde.org/show_bug.cgi?id=459408 Bug ID: 459408 Summary: DesktopExecParser fails to launch terminal if TerminalApplication has multiple arguments Classification: Unclassified Product: frameworks-kio Version: 5.97.0 Platform: Fedora RPMs OS: Linux Status: REPORTED Severity: normal Priority: NOR Component: general Assignee: kio-bugs-n...@kde.org Reporter: accounts+...@chiller3.com CC: kdelibs-b...@kde.org Target Milestone: --- SUMMARY When the default terminal application is set to a terminal whose .desktop file has multiple arguments in Exec=, launching the default terminal (eg. via KTerminalLauncherJob) will fail. STEPS TO REPRODUCE 1. Install a terminal that has multiple arguments in its .desktop entry (eg. Wezterm's "Exec=wezterm start --cwd . --") 2. Set that terminal (eg. Wezterm) as the default terminal in system settings 3. Create a new .desktop file containing "Terminal=true" (to easily test starting a terminal) 4. Launch the new .desktop file OBSERVED RESULT plasmashell fails to spawn a terminal with the following errors showing up in journald: kf.kio.core: Terminal "wezterm start --cwd . --" not found, service "TestDesktopFile" kf.kio.gui: "Terminal wezterm start --cwd . -- not found while trying to run /home/chenxiaolong/.local/share/applications/testdesktopfile.desktop" EXPECTED RESULT The terminal should launch successfully. SOFTWARE/OS VERSIONS Windows: n/a macOS: n/a Linux/KDE Plasma: Fedora 36 KDE Plasma Version: 5.25.5 KDE Frameworks Version: 5.97.0 Qt Version: 5.15.5 ADDITIONAL INFORMATION I believe the issue is how the default terminal config option is parsed in KIO. When I set Wezterm as the default terminal, the following is written to ~/.config/kdeglobals: [General] TerminalApplication=wezterm start --cwd . -- TerminalService=org.wezfurlong.wezterm.desktop The TerminalApplication value is queried here: https://invent.kde.org/frameworks/kio/-/blob/v5.97.0/src/core/desktopexecparser.cpp#L488 and passed to QStandardPaths::findExecutable() verbatim. I believe it's searching PATH for the literal string "wezterm start --cwd . --" instead of "wezterm". -- You are receiving this mail because: You are watching all bug changes.
[kio-extras] [Bug 325574] Unable to modifying file on Android device using kio-mtp
https://bugs.kde.org/show_bug.cgi?id=325574 Andrew Gunnerson changed: What|Removed |Added CC||accounts+...@chiller3.com -- You are receiving this mail because: You are watching all bug changes.
[kdeconnect] [Bug 447636] Android 11: scoped storage does not allow to share the root of the filesystem
https://bugs.kde.org/show_bug.cgi?id=447636 Andrew Gunnerson changed: What|Removed |Added CC||accounts+...@chiller3.com --- Comment #6 from Andrew Gunnerson --- I was able to get it working with MANAGE_EXTERNAL_STORAGE, but it's not quite as simple as adding the permission to the manifest. Once the "All files" permission is granted by the user via the Android Settings [1], KDE Connect is immediately allowed access to the root of the storage volume. However, it must do so via standard filesystem APIs, not via SAF's ACTION_OPEN_DOCUMENT_TREE. Even with the permission, accessing the storage root via SAF is not allowed. I made a very hacky hardcoded proof of concept here: https://invent.kde.org/chenxiaolong/kdeconnect-android/-/commit/53370709d607055a0593a1dd743733858977cf05 It's working fine on a Google Pixel 6 Pro running Android 13 Beta 2.1. I can read and write files from the primary storage root. For a proper implementation, I believe these are the necessary changes: * A UI option for selecting the primary storage volume. If the "All files" permission hasn't been granted, the ACTION_MANAGE_APP_ALL_FILES_ACCESS_PERMISSION intent can take the user straight to the permissions settings. The file:// URI for the primary storage volume can be retrieved from StorageManager.getPrimaryStorageVolume().getDirectory(). * Adjust some Build.VERSION.SDK_INT checks that enable file:// support to be uri.getScheme().equals(ContentResolver.SCHEME_FILE) instead. * Maybe refactor the FileSystemView abstraction to allow a mixture of SAF and non-SAF roots? Eg. the user might select the primary storage (non-SAF with MANAGE_EXTERNAL_STORAGE) as well as an SD card (SAF). I unfortunately don't have time to implement this myself, but hope this information might be useful to others. [1] Either by sending the user there via the ACTION_MANAGE_APP_ALL_FILES_ACCESS_PERMISSION intent or manually going to Settings -> Apps -> Special app access -> All files access -> KDE Connect -> Allow -- You are receiving this mail because: You are watching all bug changes.
[kwin] [Bug 446674] In Wayland with fractional scaling fonts and images are blurry from downscaling
https://bugs.kde.org/show_bug.cgi?id=446674 Andrew Gunnerson changed: What|Removed |Added CC||accounts+...@chiller3.com -- You are receiving this mail because: You are watching all bug changes.
[plasmashell] [Bug 448479] kwallet-pam no longer works on Wayland
https://bugs.kde.org/show_bug.cgi?id=448479 Andrew Gunnerson changed: What|Removed |Added CC||accounts+...@chiller3.com -- You are receiving this mail because: You are watching all bug changes.
[plasmashell] [Bug 418411] [Wayland] Cursor appears as a black box.
https://bugs.kde.org/show_bug.cgi?id=418411 Andrew Gunnerson changed: What|Removed |Added CC||accounts+...@chiller3.com -- You are receiving this mail because: You are watching all bug changes.
[plasma-nm] [Bug 409392] NetworkManager widget freezes on connection issues blocking all Plasma
https://bugs.kde.org/show_bug.cgi?id=409392 Andrew Gunnerson changed: What|Removed |Added CC||accounts+...@chiller3.com -- You are receiving this mail because: You are watching all bug changes.
[kwallet-pam] [Bug 436442] kwallet fails to open at login most of the time.
https://bugs.kde.org/show_bug.cgi?id=436442 Andrew Gunnerson changed: What|Removed |Added CC||accounts+...@chiller3.com -- You are receiving this mail because: You are watching all bug changes.
