| Summary: | When accepting a change of a shared event (received via e-mail), Kontact crashes | ||
|---|---|---|---|
| Product: | [Applications] kontact | Reporter: | Matija Šuklje <matija> |
| Component: | general | Assignee: | Sergio Martins <smartins> |
| Status: | RESOLVED WORKSFORME | ||
| Severity: | crash | CC: | ismail, jem, klakin, oneofthegods, paul.f.fee, smartins |
| Priority: | NOR | ||
| Version: | unspecified | ||
| Target Milestone: | --- | ||
| Platform: | Gentoo Packages | ||
| OS: | Linux | ||
| Latest Commit: | Version Fixed In: | ||
| Sentry Crash Report: | |||
| Attachments: |
The attached iCal, anonymised
New crash information added by DrKonqi New crash information added by DrKonqi New crash information added by DrKonqi New crash information added by DrKonqi |
||
|
Description
Matija Šuklje
2011-08-15 19:25:13 UTC
I tried to reproduce the bug 5 times so far ant it always crashed. Also it crashes *before* changing the event, so the event changes don't even get applied. Played around a bit and basically clicking *anything* from this list (basically everything that the system lets you do with the event) makes it crash: * [Record invitation in my calendar] * Accept * Accept cond. * Counter proposal * Decline Delegate * Forward * Check my calendar If I click the attached iCal file though, I can manually import it into my calendar (but not as a change, only as a new event). Can you attach the invitation ? Created attachment 62853 [details]
The attached iCal, anonymised
This is the iCal I got attached in the e-mail, cleaned up of identifiable information.
Git commit 3a531e958c5dc585a1258f0d50964f8831625836 by Sergio Martins. Committed on 15/08/2011 at 21:49. Pushed by smartins into branch 'master'. Add assert while debugging this. CCBUG: 280144 M +9 -0 calendarsupport/groupware.cpp http://commits.kde.org/kdepim/3a531e958c5dc585a1258f0d50964f8831625836 I got the same crash today when accepting an invitation. Good to see the fix has been made, just need to wait for it to trickled down via the distros. By the way, having looked at the diff, if the call to calendar->incidence() returns NULL, then a message is displayed and CalendarSupport::Groupware::finishHandlingInvitation() returns. Will the invitation still make it into the user's calendar? An existing entry won't be updated (since the lookup failed to find it). However a new entry should be added. Just wanted to check if the check will have the desired effect. Thanks. (In reply to comment #6) > I got the same crash today when accepting an invitation. Good to see the fix > has been made, just need to wait for it to trickled down via the distros. > The issue is not fixed yet but it's been examined *** Bug 280171 has been marked as a duplicate of this bug. *** Created attachment 65461 [details]
New crash information added by DrKonqi
kmail (4.7.2) on KDE Platform 4.7.2 (4.7.2) using Qt 4.7.4
- What I was doing when the application crashed:
Accepted an update to an event in my personal calendar. Crash!
Have not tried to reproduce. I hope that it's not reproducable! (Does the KDE PIM team need more hands to do bug squashing? I can code, and have poked around in KDE internals in the past.)
-- Backtrace (Reduced):
#6 QHash (other=) at /usr/include/qt4/QtCore/qhash.h:283
#7 QSet (other=) at /usr/include/qt4/QtCore/qset.h:60
#8 KCalCore::IncidenceBase::dirtyFields (this=0x0) at ../../kcalcore/incidencebase.cpp:622
#9 0x00007f0f7022c25c in CalendarSupport::Groupware::finishHandlingInvitation (this=0x542e4c0) at ../../calendarsupport/groupware.cpp:217
#10 0x00007f0f7022db6c in CalendarSupport::Groupware::qt_metacall (this=0x542e4c0, _c=QMetaObject::InvokeMetaMethod, _id=<optimized out>, _a=0x7fff16b24700) at ./groupware.moc:82
Created attachment 65462 [details]
New crash information added by DrKonqi
kmail (4.7.2) on KDE Platform 4.7.2 (4.7.2) using Qt 4.7.4
Yep. This is 100% reproducable. This sucks. :(
-- Backtrace (Reduced):
#6 QHash (other=) at /usr/include/qt4/QtCore/qhash.h:283
#7 QSet (other=) at /usr/include/qt4/QtCore/qset.h:60
#8 KCalCore::IncidenceBase::dirtyFields (this=0x0) at ../../kcalcore/incidencebase.cpp:622
#9 0x00007fd6117ac25c in CalendarSupport::Groupware::finishHandlingInvitation (this=0x3b50690) at ../../calendarsupport/groupware.cpp:217
#10 0x00007fd6117adb6c in CalendarSupport::Groupware::qt_metacall (this=0x3b50690, _c=QMetaObject::InvokeMetaMethod, _id=<optimized out>, _a=0x7fff463c2cf0) at ./groupware.moc:82
(In reply to comment #9) > (Does the KDE > PIM team need more hands to do bug squashing? I can code, and have poked around > in KDE internals in the past.) > Yes, your help is welcome. Do you build from source ? If yes, please update master. Make it crash and post debug output. I'm interested in the output of this debug statement: kDebug() << "cloning. SchedulingId=" << incidence->uid() << "; SchedulingId2=" << existingIncidence->schedulingID() << "; new uid=" << existingIncidence->uid() << "; willCrash=" << willCrash << "; action=" << action; which in kdepim/calendarsupport/groupware.cpp Git commit dc76c02f29c730b94ffc575d3166c03305687e8f by Sergio Martins. Committed on 10/11/2011 at 00:41. Pushed by smartins into branch 'KDE/4.7'. Add debug info so reporter can give me more info. CCBUG: 280144 M +7 -1 calendarsupport/groupware.cpp http://commits.kde.org/kdepim/dc76c02f29c730b94ffc575d3166c03305687e8f Git commit 97ff696e0a23deee0cf49a7fd71a9b6aae658cde by Sergio Martins. Committed on 10/11/2011 at 00:41. Pushed by smartins into branch 'master'. Add debug info so reporter can give me more info. CCBUG: 280144 (cherry picked from commit dc76c02f29c730b94ffc575d3166c03305687e8f) Conflicts: calendarsupport/groupware.cpp M +7 -1 calendarsupport/groupware.cpp http://commits.kde.org/kdepim/97ff696e0a23deee0cf49a7fd71a9b6aae658cde (In reply to comment #11) > Yes, your help is welcome. Do you build from source ? If yes, please update > master. I do not build from source at work. I'll see if I can build from source at home and reproduce the bug. I might not get around to this until this weekend. :( Kmail2 lives in the KDEPIM module? (In reply to comment #14) > (In reply to comment #11) > > Yes, your help is welcome. Do you build from source ? If yes, please update > > master. > > I do not build from source at work. I'll see if I can build from source at home > and reproduce the bug. I might not get around to this until this weekend. :( > > Kmail2 lives in the KDEPIM module? Yes, you'll need kdelibs, kdepimlibs and kdepim-runtime, kdepim Created attachment 69128 [details]
New crash information added by DrKonqi
kmail (4.7.2) on KDE Platform 4.7.2 (4.7.2) "release 5" using Qt 4.7.4
- What I was doing when the application crashed:
Kmail immediately crashes when I accept a meeting invitation.
-- Backtrace (Reduced):
#6 QHash (other=) at /usr/include/QtCore/qhash.h:283
#7 QSet (other=) at /usr/include/QtCore/qset.h:60
#8 KCalCore::IncidenceBase::dirtyFields (this=0x0) at /usr/src/debug/kdepimlibs-4.7.2/kcalcore/incidencebase.cpp:622
#9 0x00007fcbcb25b55c in CalendarSupport::Groupware::finishHandlingInvitation (this=0x1f6ebf0) at /usr/src/debug/kdepim-4.7.2/calendarsupport/groupware.cpp:217
#10 0x00007fcbcb25ce6c in CalendarSupport::Groupware::qt_metacall (this=0x1f6ebf0, _c=QMetaObject::InvokeMetaMethod, _id=<optimized out>, _a=0x7fffba425c60) at /usr/src/debug/kdepim-4.7.2/build/calendarsupport/groupware.moc:82
Created attachment 70513 [details]
New crash information added by DrKonqi
kmail (4.7.3) on KDE Platform 4.7.4 (4.7.4) using Qt 4.7.4
- What I was doing when the application crashed:
I clicked the "Remove this meeting from my calendar" link in the (autogenerated?) email that accompanied a meeting cancellation. BOOM! This crash happened. This meeting is a recurring meeting that is on my local calender.
-- Backtrace (Reduced):
#6 QHash (other=) at /usr/include/qt4/QtCore/qhash.h:283
#7 QSet (other=) at /usr/include/qt4/QtCore/qset.h:60
#8 KCalCore::IncidenceBase::dirtyFields (this=0x0) at ../../kcalcore/incidencebase.cpp:622
#9 0x00007fb4f78ce324 in CalendarSupport::Groupware::finishHandlingInvitation (this=0x2ec3090) at ../../calendarsupport/groupware.cpp:224
#10 0x00007fb4f78cfe9c in CalendarSupport::Groupware::qt_metacall (this=0x2ec3090, _c=QMetaObject::InvokeMetaMethod, _id=<optimized out>, _a=0x7fff7a7d21b0) at ./groupware.moc:82
I can't reproduce for the moment. (In reply to comment #17) > - What I was doing when the application crashed: > > I clicked the "Remove this meeting from my calendar" link in the > (autogenerated?) email that accompanied a meeting cancellation. BOOM! This > crash happened. This meeting is a recurring meeting that is on my local > calender. > Tried creating an event, send an invitation to another account, accept it, return to the calendar & delete the event, send the email, hit the remove from calendar link. No crash, just this warning: kontact(9693)/kdepimlibs (kcalcore) KCalCore::ICalFormat::parseScheduleMessage: kcalcore library reported a problem while parsing: kontact(9693)/kdepimlibs (kcalcore) KCalCore::ICalFormat::parseScheduleMessage: "Cancel" : "Failed iTIP restrictions for SEQUENCE property. Expected 1 instances of the property and got 0 I just accepted an invitation in 4.8.5 and nothing crashed. good. anyway, stuff has changed a lot for 4.11, so we won't be seeing this again |