SUMMARY I don't do anything specific. I just translate my .po files from a project of posummit. The crash happens quite fast from the start of the application STEPS TO REPRODUCE 1. OBSERVED RESULT Crash. EXPECTED RESULT No crash. SOFTWARE/OS VERSIONS Linux/KDE Plasma: KDE Plasma Version: KDE Frameworks Version: Qt Version: 6.9.1 ADDITIONAL INFORMATION #0 Catalog::numberOfEntries (this=0x556e71a95c30) at /usr/src/debug/lokalize-25.04.2-1.fc42.x86_64/src/catalog/catalog.cpp:159 #1 CatalogTreeModel::rowCount (this=<optimized out>, parent=...) at /usr/src/debug/lokalize-25.04.2-1.fc42.x86_64/src/cataloglistview/catalogmodel.cpp:81 #2 0x00007f5ca8384422 in QAbstractItemModel::hasChildren(QModelIndex const&) const () at /lib64/libQt6Core.so.6 #3 0x00007f5ca83e2fa4 in QSortFilterProxyModel::hasChildren(QModelIndex const&) const () at /lib64/libQt6Core.so.6 #4 0x00007f5ca9805265 in QTreeView::doItemsLayout() () at /lib64/libQt6Widgets.so.6 #5 0x00007f5ca9770646 in QAbstractItemView::event(QEvent*) () at /lib64/libQt6Widgets.so.6 #6 0x00007f5ca943d97f in QApplicationPrivate::notify_helper(QObject*, QEvent*) () at /lib64/libQt6Widgets.so.6 #7 0x00007f5ca80f9f38 in QCoreApplication::notifyInternal2(QObject*, QEvent*) () at /lib64/libQt6Core.so.6 #8 0x00007f5ca9494f33 in QWidgetPrivate::setStyle_helper(QStyle*, bool) () at /lib64/libQt6Widgets.so.6 #9 0x00007f5ca9494eb8 in QWidgetPrivate::setStyle_helper(QStyle*, bool) () at /lib64/libQt6Widgets.so.6 #10 0x00007f5ca9494eb8 in QWidgetPrivate::setStyle_helper(QStyle*, bool) () at /lib64/libQt6Widgets.so.6 #11 0x00007f5ca94a817a in QWidget::setParent(QWidget*, QFlags<Qt::WindowType>) () at /lib64/libQt6Widgets.so.6 #12 0x00007f5ca95d830a in QDockWidget::~QDockWidget() () at /lib64/libQt6Widgets.so.6 #13 0x0000556e4f0acb3f in CatalogView::~CatalogView (this=0x556e7220c1d0, this=<optimized out>) at /usr/src/debug/lokalize-25.04.2-1.fc42.x86_64/src/cataloglistview/cataloglistview.cpp:135 #14 CatalogView::~CatalogView (this=0x556e7220c1d0, this=<optimized out>) at /usr/src/debug/lokalize-25.04.2-1.fc42.x86_64/src/cataloglistview/cataloglistview.cpp:135 #15 0x00007f5ca815cda2 in QObjectPrivate::deleteChildren() () at /lib64/libQt6Core.so.6 #16 0x00007f5ca94a1248 in QWidget::~QWidget() () at /lib64/libQt6Widgets.so.6 #17 0x0000556e4f0693b5 in EditorTab::~EditorTab (this=0x556e7156f870, this=<optimized out>) at /usr/src/debug/lokalize-25.04.2-1.fc42.x86_64/src/editortab.cpp:110 #18 0x00007f5ca815652c in QObject::event(QEvent*) () at /lib64/libQt6Core.so.6 #19 0x00007f5ca943d97f in QApplicationPrivate::notify_helper(QObject*, QEvent*) () at /lib64/libQt6Widgets.so.6 #20 0x00007f5ca80f9f38 in QCoreApplication::notifyInternal2(QObject*, QEvent*) () at /lib64/libQt6Core.so.6 #21 0x00007f5ca80fdd26 in QCoreApplicationPrivate::sendPostedEvents(QObject*, int, QThreadData*) () at /lib64/libQt6Core.so.6 #22 0x00007f5ca841016f in postEventSourceDispatch(_GSource*, int (*)(void*), void*) () at /lib64/libQt6Core.so.6 #23 0x00007f5ca6ceb880 in g_main_context_dispatch_unlocked.lto_priv () at /lib64/libglib-2.0.so.0 #24 0x00007f5ca6cf47a8 in g_main_context_iterate_unlocked.isra () at /lib64/libglib-2.0.so.0 #25 0x00007f5ca6cf4953 in g_main_context_iteration () at /lib64/libglib-2.0.so.0 #26 0x00007f5ca840f9ad in QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () at /lib64/libQt6Core.so.6 #27 0x00007f5ca8107b03 in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () at /lib64/libQt6Core.so.6 #28 0x00007f5ca8103419 in QCoreApplication::exec() () at /lib64/libQt6Core.so.6 #29 0x0000556e4f0565ed in main (argc=<optimized out>, argv=<optimized out>) at /usr/src/debug/lokalize-25.04.2-1.fc42.x86_64/src/main.cpp:147
Sorry about this. I don't see these crashes for Arch. I will make this a priority though, if we can figure out what's wrong. Are you able to try opening Lokalize with new state data? Rename these files (so that you can restore them later, if you want to restore your state data) and open Lokalize, and see what happens. ~/.local/state/lokalizestaterc ~/.config/lokalizerc
What does "quite fast" mean here? Do you open a specific file, is the crash seen in specific projects only, does it crash as Lokalize opens and loads the files...?
(In reply to Finley Watson from comment #1) > Sorry about this. I don't see these crashes for Arch. I will make this a > priority though, if we can figure out what's wrong. Thanks. > Are you able to try opening Lokalize with new state data? > Rename these files (so that you can restore them later, if you want to > restore your state data) and open Lokalize, and see what happens. > > ~/.local/state/lokalizestaterc > ~/.config/lokalizerc It doesn't help. (In reply to Finley Watson from comment #2) > What does "quite fast" mean here? Do you open a specific file, is the crash > seen in specific projects only, does it crash as Lokalize opens and loads > the files...? I narrowed it down more. I: 1. open a file in my project 2. translate one unit 3. press ctrl+s to save the file 4. press ctrl+w to close its tab and return to my project's tab Here occurs the crash. No specific file. If a file is small, then I translate it fast, so I close it fast, so it happens quite fast. BTW, the status bar is messed up. It switches between showing untranslated/not ready/total from a project and from a file, and it should show a specific file data when a specific file is opened.
This is useful, thanks. If you open a partially translated PO file, translate one translation pair, and save then close the file, does it happen? I still can't replicate yet.
A possibly relevant merge request was started @ https://invent.kde.org/sdk/lokalize/-/merge_requests/244
What file types cause the crash? Can you trigger the crash by opening a file, saving, closing? I.e. without actually changing the file?
(In reply to Finley Watson from comment #4) > This is useful, thanks. > If you open a partially translated PO file, translate one translation pair, > and save then close the file, does it happen? I still can't replicate yet. Yes, but actually only the following is needed to reproduce the crash: 1. open a file in my project 4. press ctrl+w to close its tab and return to my project's tab (In reply to Finley Watson from comment #6) > What file types cause the crash? .po > Can you trigger the crash by opening a file, saving, closing? I.e. without > actually changing the file? Yes. See above.
Does this only happen with a specific project? I still can't replicate. Try closing the project, opening only a file, then closing the file. Looking at the code I think the problem is coming from the process of closing an editor tab, so I expect you would still see the crash even without a project open.
I had nearly the same problem. After deleting and creating the translation memory new, it got better. Could it be, that the numbers in the statusbar are not always correct and then is something happening. Stack trace of thread 71323: #0 0x00007faca8faf74c n/a (libc.so.6 + 0x9774c) #1 0x00007faca8f55dc0 raise (libc.so.6 + 0x3ddc0) #2 0x00007facab643e97 _ZN6KCrash19defaultCrashHandlerEi (libKF6Crash.so.6 + 0x7e97) #3 0x00007faca8f55ef0 n/a (libc.so.6 + 0x3def0) #4 0x00007faca9a0d922 _ZNK21QSortFilterProxyModel6parentERK11QModelIndex (libQt6Core.so.6 + 0x40d922) #5 0x0000562388dddf0c n/a (/usr/bin/lokalize + 0x12bf0c) #6 0x00007faca97c0e34 _ZN7QObject5eventEP6QEvent (libQt6Core.so.6 + 0x1c0e34) #7 0x00007facaa901c70 _ZN19QApplicationPrivate13notify_helperEP7QObjectP6QEvent (libQt6Widgets.so.6 + 0x101c70) #8 0x00007faca9768118 _ZN16QCoreApplication15notifyInternal2EP7QObjectP6QEvent (libQt6Core.so.6 + 0x168118) #9 0x00007faca97684fb _ZN23QCoreApplicationPrivate16sendPostedEventsEP7QObjectiP11QThreadData (libQt6Core.so.6 + 0x1684fb) #10 0x00007faca9a3fcf8 n/a (libQt6Core.so.6 + 0x43fcf8) #11 0x00007faca810887d n/a (libglib-2.0.so.0 + 0x5e87d) #12 0x00007faca8109cd7 n/a (libglib-2.0.so.0 + 0x5fcd7) #13 0x00007faca8109ee5 g_main_context_iteration (libglib-2.0.so.0 + 0x5fee5) #14 0x00007faca9a3c5e2 _ZN20QEventDispatcherGlib13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (libQt6Core.so.6 + 0x43c5e2) #15 0x00007faca97744b6 _ZN10QEventLoop4execE6QFlagsINS_17ProcessEventsFlagEE (libQt6Core.so.6 + 0x1744b6) #16 0x00007faca976c7c1 _ZN16QCoreApplication4execEv (libQt6Core.so.6 + 0x16c7c1) #17 0x0000562388d01975 n/a (/usr/bin/lokalize + 0x4f975) #18 0x00007faca8f3f6b5 n/a (libc.so.6 + 0x276b5) #19 0x00007faca8f3f769 __libc_start_main (libc.so.6 + 0x27769) #20 0x0000562388d035c5 n/a (/usr/bin/lokalize + 0x515c5) Does this point to https://invent.kde.org/sdk/lokalize/-/blob/master/src/tm/dbfilesmodel.cpp?ref_type=heads#L154 ?
(In reply to Finley Watson from comment #8) > Does this only happen with a specific project? I still can't replicate. > > Try closing the project, opening only a file, then closing the file. Looking > at the code I think the problem is coming from the process of closing an > editor tab, so I expect you would still see the crash even without a project > open. No project and still crashing.
(In reply to Alois Spitzbart from comment #9) > I had nearly the same problem. After deleting and creating the translation > memory new, it got better. Could it be, that the numbers in the statusbar > are not always correct and then is something happening. You have no debug symbols installed. It's hard to say anything then.
STEPS TO REPRODUCE being empty is an insult. Next time please give us EXACT STEPS TO REPRODUCE from the start
You all should install debug symbols https://community.kde.org/Guidelines_and_HOWTOs/Debugging/How_to_create_useful_crash_reports#Backtraces Once done that gives us a backtrace with symbols. Also run valgrind lokalize and make what you usually do to make it crash and attach the valgrind log (have patience valgrind is super slow)
Follow Albert's advice with valgrind and maybe we can figure out what is causing this problem. You might alternatively want to try moving the translation memory databases into a different folder / deleting them, and seeing if they are causing the crash.
(In reply to Albert Astals Cid from comment #12) > STEPS TO REPRODUCE being empty is an insult. > > Next time please give us EXACT STEPS TO REPRODUCE from the start I don't want to collaborate in an atmosphere, where I'm being accused of something I didn't mean. I reported your comment in hope for this to stop early.
I put hundreds of hours into making Lokalize work, for free, for people like you that want to translate. All Albert is asking for is clear steps that will allow me or others to fix your issue instead of wasting our time going back and forth trying to understand if this is an issue with your set up or something else. I can close this issue if you would prefer, but let's try to reach a positive conclusion? If you're able to run valgrind and give me the output then it's likely we can figure what is going wrong.
Created attachment 182957 [details] backtrace of lokalize Does a backtrace still help? I followed the arch site and I got this file. Maybe you can get a step further.
...what on earth is this? looks like there might be something going wrong with the paths to files or something /home/alois/kde/src/powerplant/tmp/var/run/host/var/run/host/var/run/host/var/run/host/var/run/host/var/run/host/var/run/host/var/run/host/var/run/host/var/run/host/var/run/host/var/run/host/var/r
Created attachment 182958 [details] backtrace of lokalize with juk.po opened Strange... one thing comes in my mind. Could that be connected to the search to the sources of the string to be translated. Probably I have clicked on the link to the source-file. The second time it crashed, the file is smaller. I hope this one helps
Reading over this thread, I am seeing traces of impatience and demands. I want to advise that we change the tone of this conversation. We must recognize that not everything we type is understood and members of the KDE team should demonstrate helpful traits. Should the tone not change, please email the Community Working Group at community-wg@kde.org
๐๐งน โ ๏ธ This bug has been in NEEDSINFO status with no change for at least 15 days. Please provide the requested information, then set the bug status to REPORTED. If there is no change for at least 30 days, it will be automatically closed as RESOLVED WORKSFORME. For more information about our bug triaging procedures, please read https://community.kde.org/Guidelines_and_HOWTOs/Bug_triaging. Thank you for helping us make KDE software even better for everyone!
๐๐งน This bug has been in NEEDSINFO status with no change for at least 30 days. Closing as RESOLVED WORKSFORME.