Summary: | Dolphin crashes when accessing/manipulating a PPS/PPT presentation file (Strigi) [readLittleEndianInt32, Private, OleInputStream, OleEndAnalyzer::analyze] | ||
---|---|---|---|
Product: | [Applications] dolphin | Reporter: | Arnaud BERTHIER <aberthier> |
Component: | general | Assignee: | Peter Penz <peter.penz19> |
Status: | RESOLVED FIXED | ||
Severity: | crash | CC: | andresbajotierra, bittajam, dgonza.us, diego.ml, frank78ac, greg.freemyer, hhablaty, intek, johnsc301, kotlyar.dd, markg85, martinknocik, moabi2000, n.behrendorff, n.underwood78, p92, pctriplett, timid3000, tjm1983, vorsma |
Priority: | NOR | ||
Version: | 16.12.2 | ||
Target Milestone: | --- | ||
Platform: | Ubuntu | ||
OS: | Linux | ||
Latest Commit: | Version Fixed In: | 4.7.0 | |
Sentry Crash Report: | |||
Attachments: |
New crash information added by DrKonqi
New crash information added by DrKonqi New crash information added by DrKonqi 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
Arnaud BERTHIER
2010-09-02 12:34:01 UTC
Thanks for the report. Would it be possible that you attach the presentation to this bug-report? This looks like an issue in the corresponding metadata-analyzer and it would be helpful for the corresponding developer to be able to reproduce this issue. > Thanks for the report. Would it be possible that you attach the presentation to > this bug-report? This looks like an issue in the corresponding > metadata-analyzer and it would be helpful for the corresponding developer to be > able to reproduce this issue. Ok, Thank you Peter. This is this file (123M! ;) : http://www.partage-fichiers.com/dl.php?f=Poster_Judit_Salamanca_09.ppt User: 5vlg54jv Password: nxs6ci9w ) . Be note that it's a scientific presentation from my pH. D. student. Thanks. Oh wow! never saw a presentation file that big before. Anyway, while trying to reproduce this crash i got a few crashes and never got to the point to rename it.. it had already crashed dolphin by that time. For me it does not crash in trunk when renaming. However if you have the tooltips turned on or the information panel (which is on by default) then dolphin crashes as soon as you click on the file. The crash is again in the strigi analyse which is odd considering my dolphin preview setting are set to 5MB and imho it shouldn't even try to analyse files that are bigger.. So that might be a bug as well This is my crash log: Application: Dolphin (kdeinit4), signal: Segmentation fault [Current thread is 1 (Thread 0x7f2f539c8760 (LWP 4209))] Thread 4 (Thread 0x7f2f3c717710 (LWP 4210)): #0 0xffffffffff60017b in ?? () #1 0x00007f2f3c716ae0 in ?? () #2 0x00007fff011ff6e2 in ?? () Backtrace stopped: previous frame identical to this frame (corrupt stack?) Thread 3 (Thread 0x7f2f3ba84710 (LWP 4215)): #0 0x00007f2f511040b3 in select () from /lib/libc.so.6 #1 0x00007f2f52653e3d in ?? () from /usr/lib/libQtCore.so.4 #2 0x00007f2f5258c485 in ?? () from /usr/lib/libQtCore.so.4 #3 0x00007f2f52304cb0 in start_thread () from /lib/libpthread.so.0 #4 0x00007f2f5110a7ad in clone () from /lib/libc.so.6 #5 0x0000000000000000 in ?? () Thread 2 (Thread 0x7f2f3b283710 (LWP 4225)): [KCrash Handler] #6 0x00007f2f4a872d9a in Strigi::readLittleEndianInt32(char const*) () from /home/kde-devel/kde/lib/libstreams.so.0 #7 0x00007f2f4a871fd4 in Strigi::OleInputStream::Private::Private(Strigi::OleInputStream*, Strigi::StreamBase<char>*) () from /home/kde-devel/kde/lib/libstreams.so.0 #8 0x00007f2f4a871a3f in Strigi::OleInputStream::OleInputStream(Strigi::StreamBase<char>*) () from /home/kde-devel/kde/lib/libstreams.so.0 #9 0x00007f2f4ab8d5a8 in OleEndAnalyzer::analyze(Strigi::AnalysisResult&, Strigi::StreamBase<char>*) () from /home/kde-devel/kde/lib/libstreamanalyzer.so.0.7 #10 0x00007f2f4ab6980a in Strigi::StreamAnalyzerPrivate::analyze(Strigi::AnalysisResult&, Strigi::StreamBase<char>*) () from /home/kde-devel/kde/lib/libstreamanalyzer.so.0.7 #11 0x00007f2f4ab692f2 in Strigi::StreamAnalyzer::analyze(Strigi::AnalysisResult&, Strigi::StreamBase<char>*) () from /home/kde-devel/kde/lib/libstreamanalyzer.so.0.7 #12 0x00007f2f4b62ee65 in KFileMetaInfoPrivate::init (this=<value optimized out>, stream=..., url=<value optimized out>, mtime=1284592051, w=<value optimized out>) at /home/kde-devel/kde/src/trunk/KDE/kdelibs/kio/kio/kfilemetainfo.cpp:256 #13 0x00007f2f4b62f728 in KFileMetaInfo::KFileMetaInfo (this=0x7f2f3b282d00, path=<value optimized out>, w=<value optimized out>) at /home/kde-devel/kde/src/trunk/KDE/kdelibs/kio/kio/kfilemetainfo.cpp:285 #14 0x00007f2f4b6f949e in KLoadFileMetaDataThread::run (this=0x183dbc0) at /home/kde-devel/kde/src/trunk/KDE/kdelibs/kio/kfile/kloadfilemetadatathread.cpp:142 #15 0x00007f2f5258c485 in ?? () from /usr/lib/libQtCore.so.4 #16 0x00007f2f52304cb0 in start_thread () from /lib/libpthread.so.0 #17 0x00007f2f5110a7ad in clone () from /lib/libc.so.6 #18 0x0000000000000000 in ?? () Thread 1 (Thread 0x7f2f539c8760 (LWP 4209)): #0 0x00007f2f5230940c in pthread_cond_wait@@GLIBC_2.3.2 () from /lib/libpthread.so.0 #1 0x00007f2f5258cb4b in QWaitCondition::wait(QMutex*, unsigned long) () from /usr/lib/libQtCore.so.4 #2 0x00007f2f5258bbf1 in QThread::wait(unsigned long) () from /usr/lib/libQtCore.so.4 #3 0x00007f2f52647f30 in QFileSystemWatcher::~QFileSystemWatcher() () from /usr/lib/libQtCore.so.4 #4 0x00007f2f52648009 in QFileSystemWatcher::~QFileSystemWatcher() () from /usr/lib/libQtCore.so.4 #5 0x00007f2f52683a74 in QObjectPrivate::deleteChildren() () from /usr/lib/libQtCore.so.4 #6 0x00007f2f52688725 in QObject::~QObject() () from /usr/lib/libQtCore.so.4 #7 0x00007f2f4a5d2f49 in Solid::Backends::Fstab::FstabWatcher::~FstabWatcher (this=0x1416780, __in_chrg=<value optimized out>) at /home/kde-devel/kde/src/trunk/KDE/kdelibs/solid/solid/backends/fstab/fstabwatcher.cpp:53 #8 0x00007f2f51071151 in ?? () from /lib/libc.so.6 #9 0x00007f2f510711a5 in exit () from /lib/libc.so.6 #10 0x00007f2f518b70e8 in ?? () from /usr/lib/libQtGui.so.4 #11 0x00007f2f533ce518 in KApplication::xioErrhandler (this=0x7fff01168320, dpy=0x1256600) at /home/kde-devel/kde/src/trunk/KDE/kdelibs/kdeui/kernel/kapplication.cpp:416 #12 0x00007f2f52e98f8e in _XIOError () from /usr/lib/libX11.so.6 #13 0x00007f2f52e9ff3d in _XEventsQueued () from /usr/lib/libX11.so.6 #14 0x00007f2f52e88eaf in XEventsQueued () from /usr/lib/libX11.so.6 #15 0x00007f2f518ee1cc in ?? () from /usr/lib/libQtGui.so.4 #16 0x00007f2f4d3967b8 in g_main_context_check () from /usr/lib/libglib-2.0.so.0 #17 0x00007f2f4d397092 in g_main_context_iterate () from /usr/lib/libglib-2.0.so.0 #18 0x00007f2f4d3975ad in g_main_context_iteration () from /usr/lib/libglib-2.0.so.0 #19 0x00007f2f5269b886 in QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () from /usr/lib/libQtCore.so.4 #20 0x00007f2f518ee37e in ?? () from /usr/lib/libQtGui.so.4 #21 0x00007f2f52672122 in QEventLoop::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () from /usr/lib/libQtCore.so.4 #22 0x00007f2f5267235c in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () from /usr/lib/libQtCore.so.4 #23 0x00007f2f5267676b in QCoreApplication::exec() () from /usr/lib/libQtCore.so.4 #24 0x00007f2f437b0a05 in kdemain (argc=5, argv=0x1212740) at /home/kde-devel/kde/src/trunk/KDE/kdebase/apps/dolphin/src/main.cpp:98 #25 0x0000000000406a83 in _start () *** Bug 251575 has been marked as a duplicate of this bug. *** Reopening because the file has been provided. *** Bug 253134 has been marked as a duplicate of this bug. *** Looks like it could be related to bug 251701. Created attachment 52276 [details]
New crash information added by DrKonqi
dolphin (1.5) on KDE Platform 4.5.1 (KDE 4.5.1) "release 3" using Qt 4.6.3
- What I was doing when the application crashed:
Trying to individually select a .ppt file in Dolphin by left-clicking, or including it in a group select by dragging the cursor. In both cases Dolphin crashes immediately. Crash could be reproduced several times, with three different .ppt files. Dolphin does not crash when right-clicking on any of the same files. File sizes respectively: 18, 22, 72 Mb
-- Backtrace (Reduced):
#6 readLittleEndianInt32 (this=0x7ffbdc0dff00, s=<value optimized out>, input=<value optimized out>) at /usr/src/debug/strigi-0.7.2/src/streams/textutils.h:78
#7 Strigi::OleInputStream::Private::Private (this=0x7ffbdc0dff00, s=<value optimized out>, input=<value optimized out>) at /usr/src/debug/strigi-0.7.2/src/streams/oleinputstream.cpp:208
#8 0x00007ffbef44da82 in Strigi::OleInputStream::OleInputStream (this=0x7ffbe21ed6a0, input=<value optimized out>) at /usr/src/debug/strigi-0.7.2/src/streams/oleinputstream.cpp:151
#9 0x00007ffbf0505036 in OleEndAnalyzer::analyze (this=0x7ffbdc0556f0, ar=..., in=<value optimized out>) at /usr/src/debug/strigi-0.7.2/src/streamanalyzer/endanalyzers/oleendanalyzer.cpp:377
#10 0x00007ffbf04ee332 in Strigi::StreamAnalyzerPrivate::analyze (this=0x19, idx=..., input=0x7ffbdc03a7f0) at /usr/src/debug/strigi-0.7.2/src/streamanalyzer/streamanalyzer.cpp:421
*** Bug 255519 has been marked as a duplicate of this bug. *** *** Bug 255615 has been marked as a duplicate of this bug. *** *** Bug 255618 has been marked as a duplicate of this bug. *** Same problem here while trying to select a ppt file, even with the keyboard. If I disable the "preview" panel, dolphin doesn't crash anymore. I've submitted a related bug-report at https://sourceforge.net/tracker/?func=detail&aid=2929294&group_id=171000&atid=856302 and have linked to this bug-report. *** Bug 257789 has been marked as a duplicate of this bug. *** Created attachment 53883 [details]
New crash information added by DrKonqi
dolphin (1.5.9) on KDE Platform 4.5.80 (4.6 Beta1) using Qt 4.7.1
- What I was doing when the application crashed:
The same littleEndian32 crash - this time on a .doc file.
I appreciate that you have closed this upstream, but this crash still occurs (4.6 beta 1) even when I have disabled strigi indexing in system settings - that is a KDE bug for still trying to use it when I have it disabled, isn't it?
-- Backtrace (Reduced):
#6 readLittleEndianInt32 (this=0x270e7b0, s=<value optimized out>, input=0x7f9446445534) at /usr/src/debug/strigi-0.7.2/src/streams/textutils.h:78
#7 Strigi::OleInputStream::Private::Private (this=0x270e7b0, s=<value optimized out>, input=0x7f9446445534) at /usr/src/debug/strigi-0.7.2/src/streams/oleinputstream.cpp:208
#8 0x00000038e0828fc2 in Strigi::OleInputStream::OleInputStream (this=<value optimized out>, input=0x2480610) at /usr/src/debug/strigi-0.7.2/src/streams/oleinputstream.cpp:151
#9 0x00000038dc453c05 in OleEndAnalyzer::analyze (this=0x22bfa90, ar=..., in=<value optimized out>) at /usr/src/debug/strigi-0.7.2/src/streamanalyzer/endanalyzers/oleendanalyzer.cpp:377
#10 0x00000038dc43e24a in Strigi::StreamAnalyzerPrivate::analyze (this=0x1a, idx=..., input=0x2480610) at /usr/src/debug/strigi-0.7.2/src/streamanalyzer/streamanalyzer.cpp:421
*** Bug 259073 has been marked as a duplicate of this bug. *** *** Bug 259829 has been marked as a duplicate of this bug. *** Created attachment 56671 [details]
New crash information added by DrKonqi
dolphin (1.5) on KDE Platform 4.5.5 (KDE 4.5.5) using Qt 4.7.1
- What I was doing when the application crashed:
Open Dolphin, hover a PPT presentation file and crash!
-- Backtrace (Reduced):
#6 readLittleEndianInt32 (this=0x7f4c2812a350, s=<value optimized out>, input=0x7f4c2fb50584) at /usr/src/debug/strigi-0.7.2/src/streams/textutils.h:78
#7 Strigi::OleInputStream::Private::Private (this=0x7f4c2812a350, s=<value optimized out>, input=0x7f4c2fb50584) at /usr/src/debug/strigi-0.7.2/src/streams/oleinputstream.cpp:208
#8 0x0000003f43628fc2 in Strigi::OleInputStream::OleInputStream (this=<value optimized out>, input=0x7f4c2805dce0) at /usr/src/debug/strigi-0.7.2/src/streams/oleinputstream.cpp:151
#9 0x0000003751e53c05 in OleEndAnalyzer::analyze (this=0x7f4c2803c3e0, ar=..., in=<value optimized out>) at /usr/src/debug/strigi-0.7.2/src/streamanalyzer/endanalyzers/oleendanalyzer.cpp:377
#10 0x0000003751e3e24a in Strigi::StreamAnalyzerPrivate::analyze (this=0x1a, idx=..., input=0x7f4c2805dce0) at /usr/src/debug/strigi-0.7.2/src/streamanalyzer/streamanalyzer.cpp:421
seems fixed in 4.6.0 Just for record - this is dupe of #251701 Kubuntu 10.10 64bit kde 4.6 I have the same issue. Created attachment 57479 [details]
New crash information added by DrKonqi
dolphin (1.6) on KDE Platform 4.6.00 (4.6.0) using Qt 4.7.0
- What I was doing when the application crashed:
I opened dolphin from a desktop shortcut and once my mouse hovered the .ppt file saved in Documents, Dolphin crashed. it crashes everytime doing the same thing.
-- Backtrace (Reduced):
#6 readLittleEndianInt32 (this=0x7f85a4169210, s=<value optimized out>, input=<value optimized out>) at /build/buildd/strigi-0.7.2/src/streams/textutils.h:78
#7 Strigi::OleInputStream::Private::Private (this=0x7f85a4169210, s=<value optimized out>, input=<value optimized out>) at /build/buildd/strigi-0.7.2/src/streams/oleinputstream.cpp:208
#8 0x00007f85b3f482c2 in Strigi::OleInputStream::OleInputStream (this=0x7f85a24a8640, input=0x7f85a406f7d0) at /build/buildd/strigi-0.7.2/src/streams/oleinputstream.cpp:151
#9 0x00007f85b79c3a7d in OleEndAnalyzer::analyze (this=<value optimized out>, ar=..., in=0x7f859cdaa010) at /build/buildd/strigi-0.7.2/src/streamanalyzer/endanalyzers/oleendanalyzer.cpp:377
#10 0x00007f85b79aff7a in Strigi::StreamAnalyzerPrivate::analyze (this=0x7f85a4054d80, idx=..., input=0x7f85a406f7d0) at /build/buildd/strigi-0.7.2/src/streamanalyzer/streamanalyzer.cpp:421
Created attachment 57480 [details]
New crash information added by DrKonqi
dolphin (1.6) on KDE Platform 4.6.00 (4.6.0) using Qt 4.7.0
- What I was doing when the application crashed:
I moved my ppt file from Downloads to Documents and now when my mouse hovers the file in Documents, Dolphin crashes. It only crashes from hovering ppt documents in the Documents folder, none other.
-- Backtrace (Reduced):
#6 readLittleEndianInt32 (this=0x7fdd2c04fc10, s=<value optimized out>, input=<value optimized out>) at /build/buildd/strigi-0.7.2/src/streams/textutils.h:78
#7 Strigi::OleInputStream::Private::Private (this=0x7fdd2c04fc10, s=<value optimized out>, input=<value optimized out>) at /build/buildd/strigi-0.7.2/src/streams/oleinputstream.cpp:208
#8 0x00007fdd496232c2 in Strigi::OleInputStream::OleInputStream (this=0x7fdd37d1f640, input=0x7fdd2c02e250) at /build/buildd/strigi-0.7.2/src/streams/oleinputstream.cpp:151
#9 0x00007fdd4d09ea7d in OleEndAnalyzer::analyze (this=<value optimized out>, ar=..., in=0x7fdd3241d010) at /build/buildd/strigi-0.7.2/src/streamanalyzer/endanalyzers/oleendanalyzer.cpp:377
#10 0x00007fdd4d08af7a in Strigi::StreamAnalyzerPrivate::analyze (this=0x7fdd2c002b30, idx=..., input=0x7fdd2c02e250) at /build/buildd/strigi-0.7.2/src/streamanalyzer/streamanalyzer.cpp:421
Dear KDE Developers As should be evident from the number of duplicates here, this is a frequently encountered bug. It is also a MAJOR bug - a file manager is a crucial, frequently-used part of a GUI desktop environment. Having it crash randomly is totally unacceptable behaviour in the 21st century! (FWIW I now use Nautilus instead). No matter how many Fantastic new features dolphin has (the filtering is great as is faceted browsing) this application cannot be used for any serious work if it keeps crashing. I DON'T CARE that this is a strigi bug. The symptoms are in dolphin and that is what needs fixing. How hard would it be to work around this until it is fixed? if ((strigiVersion==TheBuggyOne) AND (FileType==PowerPoint)) then DisableMetaDataParsing ?? If you leave it like this it is basically unusable for any serious work. These kinds of highly obvious, workflow-destroying bugs in critical parts of the KDE infrastructure are what makes KDE an amateur hobbyists tool rather than a serious desktop environment. Sorry for the rant, I love dolphin, it is really useful and I can find and act on files way faster than with nautilus and friends, but random crashes if I open a folder with ppt files in makes it unusable. bon courage M. @mps: I understand your frustration and agree that from a users point of view it does not matter which subsystem is guilty for the crash. Users blame Dolphin for being unstable and as maintainer of course I'm not happy with this. However fixing it in the "if ((strigiVersion==TheBuggyOne) AND (FileType==PowerPoint)) then DisableMetaDataParsing" in Dolphin would just be egoistic as not only Dolphin is affected by this crash. If you want to help getting this bug fixed, please add a comment to https://sourceforge.net/tracker/?func=detail&aid=2929294&group_id=171000&atid=856302. Most probably the fix is just a few lines of code and probably less effort than implementing the "if ((strigiVersion..."-hack. Thanks. Oke, a few things to note here. 1. In the first place i find it bad that strigi manages to keep this broken for so long.. 2. When diving in the KDE code i find this: http://lxr.kde.org/source/kde/kdelibs-git/kio/kio/kfilemetainfo.cpp#246 which has a "// TODO: get data from Nepomuk" .. Why not implement the TODO.. 3. Since this strigi can, obviously, crash it seems logical to me to call it through IPC (like DBus) so that if it crashes you as a Dolphin user won't notice it and dolphin stays running. I don't know if nepomuk is currently running strigi that way or if nepomuk simply included the strigi libs and calls it directly. I don't even know how hard it is to make strigi available trhough DBus if possible at all. Imho it should use IPC to prevent crashes or strigi should get some heavy error checking. @Mark:
> // TODO: get data from Nepomuk" .. Why not implement the TODO..
Because it won't help to solve this issue at all: Most people out there don't have Nepomuk enabled but still want to be able to get the meta-data of files (like it worked in KDE3). Beside this Nepomuk uses Strigi to get the meta-data and hence the Nepomuk-Indexer will crash when trying to parse this file.
As said above: We can add a lot of workarounds in Dolphin and other apps, but it is absolute overkill to use an IPC or DBus just to get some meta-data out of a 64KB file-snipplet. KFileMetaInfo worked nice in KDE3 with the same approach for years. The problems here are really some Strigi-analyzers that must be fixed. Parsing meta-data from files might not be always trivial, but at least not crashing when doing some parsing really is no vodoo-magic... My personal point-of-view is that the Strigi-API is too low-level for file-parsing and makes it too easy to write analyzers that crash.
Now please don't waste further energy with comments on this bug-report, please use it to push the strigi-maintainers ;-)
(In reply to comment #27) > Now please don't waste further energy with comments on this bug-report, please > use it to push the strigi-maintainers ;-) Hehe, true, but the strigi site is "down/unusable" or something.. http://strigi.sourceforge.net/ But the bug is reported here (http://sourceforge.net/tracker/index.php?func=detail&aid=2567315&group_id=171000&atid=856302 over 2 freaking years ago!!) And their bugzilla is here: http://sourceforge.net/tracker/?group_id=171000&atid=856302 which is full with crash reports! Maybe it might make sense to ask on the strigi-users mailinglist about the maintenance-state of the analyzers... I already tried to contact the creators of some analyzers directly during the last years, but it did not help much. (http://sourceforge.net/mail/?group_id=171000) Git commit d1254968eac95f4d9d8c47552e27f0cc40e6aed9 by Peter Penz. Committed on 27/02/2011 at 20:01. Pushed by ppenz into branch 'master'. KFileMetaDataProvider: Don't crash if a Strigi analyzer crashes Some Strigi analyzers are still unstable and crash under certain circumstances. This patch moves the reading of the meta-data into a custom process instead of doing it in a thread, which assures that a crashing Strigi analyzer does not result in crashing Dolphin or Konqueror when hovering some items. FIXED-IN: 4.7.0 BUG: 257964 BUG: 258715 BUG: 262299 BUG: 263468 BUG: 263502 BUG: 264254 BUG: 234799 BUG: 251462 BUG: 251701 BUG: 258918 BUG: 192377 BUG: 195564 BUG: 199368 BUG: 210841 BUG: 245376 BUG: 246461 BUG: 246901 BUG: 249150 BUG: 261952 BUG: 249876 BUG: 179376 BUG: 179417 BUG: 179420 BUG: 181591 BUG: 183269 BUG: 183722 BUG: 185667 BUG: 188596 BUG: 191864 BUG: 193112 BUG: 205813 BUG: 244621 BUG: 245451 BUG: 248214 BUG: 249655 BUG: 249876 BUG: 265549 BUG: 267079 M +11 -1 kio/CMakeLists.txt M +68 -38 kio/kfile/kfilemetadataprovider.cpp M +2 -1 kio/kfile/kfilemetadataprovider_p.h A +174 -0 kio/kfile/kfilemetadatareader.cpp [License: LGPL (v2+)] M +0 -2 kio/kfile/kfilemetadatawidget.cpp D +0 -178 kio/kfile/kloadfilemetadatathread.cpp D +0 -96 kio/kfile/kloadfilemetadatathread_p.h http://commits.kde.org/kdelibs/d1254968eac95f4d9d8c47552e27f0cc40e6aed9 Hey, you "somewhat" took my advice to run strigi in a way that strigi crashes don't crash dolphin :) Perhaps using dbus for that is overkill. I like your solution! Nice job! @Mark: Yes, I followed your advice :-) The trigger was that I started a discussion on strigi-devel and for me the outcome was that I cannot influence the quality of strigi-analyzers and that probably the quality of some strigi-analyzers will remain as unstable also in far future. The current approach needs some finetuning but basically it works quite well and fast. Awesome! I guess that will save a lot of bug reports ;) Excellent, well done. Look forward to trying this out! M. *** Bug 269275 has been marked as a duplicate of this bug. *** *** Bug 269828 has been marked as a duplicate of this bug. *** The only "problem" with this patch is that we no longer get bug reports (and associated test files), which means we shifted the Strigi bug fixing from "not fixed for a very long time" to "never" :/ (In reply to comment #37) > The only "problem" with this patch is that we no longer get bug reports (and > associated test files), which means we shifted the Strigi bug fixing from "not > fixed for a very long time" to "never" :/ I remember a discussion on #fedora-kde mentioning that: a) tools like abrt[1] would be able to detect the crash, and then it could be reported upstream; b) (not sure about this) moving drkonqui to kdecore would permit somehow to detect the crash. So getting reports from users will probably still be possible; just the route will be abrt -> fedora bugzilla -> strigi bug tracker instead of dr.konqui -> kde bugzilla -> strigi bug tracker. So, let's face it: the real problem is that Strigi is short in manpower. As comment #28 stated Strigi bug tracker is already full of crash reports, but this doesn't fix them magically, nor keeping Dolphin crashing will. Just developers fix bugs. [1]: https://fedorahosted.org/abrt/ *** Bug 270271 has been marked as a duplicate of this bug. *** Created attachment 59420 [details]
New crash information added by DrKonqi
dolphin (1.6.1) on KDE Platform 4.6.2 (4.6.2) using Qt 4.7.2
- What I was doing when the application crashed:
howering over a larger ppt file dolphin crashes
-- Backtrace (Reduced):
#6 readLittleEndianInt32 (this=0x7f2c70124dd0, s=<value optimized out>, input=<value optimized out>) at /build/buildd/strigi-0.7.2/src/streams/textutils.h:78
#7 Strigi::OleInputStream::Private::Private (this=0x7f2c70124dd0, s=<value optimized out>, input=<value optimized out>) at /build/buildd/strigi-0.7.2/src/streams/oleinputstream.cpp:208
#8 0x00007f2c7e0db162 in Strigi::OleInputStream::OleInputStream (this=<value optimized out>, input=<value optimized out>) at /build/buildd/strigi-0.7.2/src/streams/oleinputstream.cpp:151
#9 0x00007f2c823fcdfd in OleEndAnalyzer::analyze (this=0x7f2c70095950, ar=..., in=<value optimized out>) at /build/buildd/strigi-0.7.2/src/streamanalyzer/endanalyzers/oleendanalyzer.cpp:377
#10 0x00007f2c823e6eca in Strigi::StreamAnalyzerPrivate::analyze (this=0x7f2c70002580, idx=..., input=0x7f2c7007f9d0) at /build/buildd/strigi-0.7.2/src/streamanalyzer/streamanalyzer.cpp:421
Created attachment 61976 [details]
New crash information added by DrKonqi
dolphin (1.6.1) on KDE Platform 4.6.2 (4.6.2) using Qt 4.7.2
- What I was doing when the application crashed:
I had just opened a folder containing large ppt files. Upon hovering the mouse over one of them, dolphin crashes. This happens every time in this folder. It normally does not happen with smaller powerpoints.
-- Backtrace (Reduced):
#6 readLittleEndianInt32 (this=0x7faac41302b0, s=<value optimized out>, input=<value optimized out>) at /build/buildd/strigi-0.7.2/src/streams/textutils.h:78
#7 Strigi::OleInputStream::Private::Private (this=0x7faac41302b0, s=<value optimized out>, input=<value optimized out>) at /build/buildd/strigi-0.7.2/src/streams/oleinputstream.cpp:208
#8 0x00007faad4a71162 in Strigi::OleInputStream::OleInputStream (this=<value optimized out>, input=<value optimized out>) at /build/buildd/strigi-0.7.2/src/streams/oleinputstream.cpp:151
#9 0x00007faad8d92dfd in OleEndAnalyzer::analyze (this=0x7faac4021330, ar=..., in=<value optimized out>) at /build/buildd/strigi-0.7.2/src/streamanalyzer/endanalyzers/oleendanalyzer.cpp:377
#10 0x00007faad8d7ceca in Strigi::StreamAnalyzerPrivate::analyze (this=0x7faac40a9e50, idx=..., input=0x7faac402ec70) at /build/buildd/strigi-0.7.2/src/streamanalyzer/streamanalyzer.cpp:421
*** Bug 279548 has been marked as a duplicate of this bug. *** *** Bug 297102 has been marked as a duplicate of this bug. *** |