Bug 49506 - cannot display file content and crash on exit
Summary: cannot display file content and crash on exit
Status: RESOLVED DUPLICATE of bug 56521
Alias: None
Product: kghostview
Classification: Applications
Component: general (show other bugs)
Version: unspecified
Platform: RedHat Enterprise Linux Linux
: NOR crash
Target Milestone: ---
Assignee: Wilco Greven
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2002-10-21 21:12 UTC by Alias
Modified: 2003-04-09 00:21 UTC (History)
0 users

See Also:
Latest Commit:
Version Fixed In:


Attachments
dummy postscript file created by SO 5.2 (5.53 KB, application/postscript)
2003-04-08 21:47 UTC, Alias
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Alias 2002-10-21 21:12:54 UTC
Version:           kghostview 0.13 (using KDE KDE 3.0)
Installed from:    RedHat RPMs
OS:          Linux

kghostview cannot display content of a PostScript file. It doesn't show anything but doesn't crash. Cursor shows working
On exit, it crashes.
the PS-file is from staroffice. it can be seen thru gv (w/o anti-aliasing) and ggv (with anti-aliasing) though.
anti-alias on with gv brings an error message.
Anyway, kghostview shouldn't crash and do at least the same as gv.
anti-aliasing is on as a preference for kghostview.

I don't think, it's a staroffice pb as i could display PS files before upgrading to KDE3.
GNU gs version: 6.52

traceback from kcrash
(no debugging symbols found)...(no debugging symbols found)...
(no debugging symbols found)...(no debugging symbols found)...
(no debugging symbols found)...(no debugging symbols found)...
(no debugging symbols found)...(no debugging symbols found)...
(no debugging symbols found)...(no debugging symbols found)...
(no debugging symbols found)...[New Thread 1024 (LWP 16347)]
0x420b4769 in wait4 () from /lib/i686/libc.so.6
#0  0x420b4769 in wait4 () from /lib/i686/libc.so.6
#1  0x4213030c in __DTOR_END__ () from /lib/i686/libc.so.6
#2  0x40fe8ca3 in waitpid () from /lib/i686/libpthread.so.0
#3  0x407688d2 in KCrash::defaultCrashHandler ()
   from /usr/lib/libkdecore-gcc2.96.so.4
#4  0x40fe6f75 in pthread_sighandler () from /lib/i686/libpthread.so.0
#5  <signal handler called>
#6  0x4004dcc8 in KPSWidget::stopInterpreter ()
   from /usr/lib/libkghostview-gcc2.96.so
#7  0x4004c25f in KPSWidget::~KPSWidget ()
   from /usr/lib/libkghostview-gcc2.96.so
#8  0x40aa013a in QWidget::~QWidget ()
   from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#9  0x4006133b in KGVPageDecorator::~KGVPageDecorator ()
   from /usr/lib/libkghostview-gcc2.96.so
#10 0x40aa013a in QWidget::~QWidget ()
   from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#11 0x40b79d44 in QViewportWidget::~QViewportWidget ()
   from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#12 0x40b73d0a in QScrollView::~QScrollView ()
   from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#13 0x40060fe4 in KGVPageView::~KGVPageView ()
   from /usr/lib/libkghostview-gcc2.96.so
#14 0x40aa013a in QWidget::~QWidget ()
   from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#15 0x400613b4 in KGVMainWidget::~KGVMainWidget ()
   from /usr/lib/libkghostview-gcc2.96.so
#16 0x40130270 in KParts::Part::~Part () from /usr/lib/libkparts-gcc2.96.so.2
#17 0x401307c4 in KParts::ReadOnlyPart::~ReadOnlyPart ()
   from /usr/lib/libkparts-gcc2.96.so.2
#18 0x40053279 in KGVPart::~KGVPart () from /usr/lib/libkghostview-gcc2.96.so
#19 0x40aa013a in QWidget::~QWidget ()
   from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#20 0x40b579e4 in QMainWindow::~QMainWindow ()
   from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#21 0x40547449 in KMainWindow::~KMainWindow ()
   from /usr/lib/libkdeui-gcc2.96.so.4
#22 0x40137a43 in KParts::MainWindow::~MainWindow ()
   from /usr/lib/libkparts-gcc2.96.so.2
#23 0x4004af4c in KGVShell::~KGVShell () from /usr/lib/libkghostview-gcc2.96.so
#24 0x40a73350 in QObject::event () from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#25 0x40aa4e31 in QWidget::event () from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#26 0x40b5be5f in QMainWindow::event ()
   from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#27 0x40a14ff3 in QApplication::internalNotify ()
   from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#28 0x40a14e8c in QApplication::notify ()
   from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#29 0x406cdf1b in KApplication::notify () from /usr/lib/libkdecore-gcc2.96.so.4
#30 0x40a1590f in QApplication::sendPostedEvents ()
   from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#31 0x40a1578b in QApplication::sendPostedEvents ()
   from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#32 0x409b96f3 in QApplication::processNextEvent ()
   from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#33 0x40a164bc in QApplication::enter_loop ()
   from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#34 0x409b9637 in QApplication::exec ()
   from /usr/lib/qt3-gcc2.96/lib/libqt-mt.so.3
#35 0x0804912e in main ()
#36 0x42017499 in __libc_start_main () from /lib/i686/libc.so.6
Comment 1 Moritz Moeller-Herrmann 2002-10-22 01:50:05 UTC
Please send in the file if the bug persists, closing for now... Can't reproduce 
behaviour with SO6 generated PS files. 
Comment 2 Alias 2002-10-22 20:36:26 UTC
i don't think, the problem is due only to SO5.2 output file but instead, that it
is a GNU gs pb. I could display SO5.2 output file before upgrading to KDE3.0 (gs
was upgraded too).
As I wrote, with anti-aliasing on, gv cannot display the file, it brings a error
message and stop. The pb of kghostview is, that it doesn't display anything and
doesn't show any error message. As you can't stop it, you have to exit. When
exiting, kghostview crashes. So, although, you cannot probably solve entirely
the pb (if it's a GNU gs one), you could prevent kghostview not to display an
error message for instance and also that it crashes on exit.
If you need more info, please ask.
Comment 3 Luís Pedro Coelho 2003-04-08 21:20:02 UTC
Can anyone post a testfile from SO 5.2 ? 
Comment 4 Alias 2003-04-08 21:47:39 UTC
Created attachment 1326 [details]
dummy postscript file created by SO 5.2

I am sending a file created by SO5.2 It can't be displayed by kghostview when
antialiasing is on. No crash on exit. (It happened when I reported the bug,
can't reproduce it now)
It crashes for me if you do the following: option antialiasing is off
call kghostview test-so5.2.ps
content is displayed
go in option and select antialiasing on
content vanishes
and kghostview crashes on exit
message on console: KCrash: crashing.... crashRecursionCounter = 2
KCrash: Application Name = kghostview path = <unknown> pid = 27590

kbug pops up
---
as I previously wrote, gv can't display the file  when antialiasing is on
so it could be that this version of gs is buggy (6.52) At least gv doesn't
crash.

hope it helps
Comment 5 Luís Pedro Coelho 2003-04-08 23:00:34 UTC
I had to reboot twice because of that file, once for gs 7.05 another for gs 6.52, but I 
was trying pure gs ("gs file.ps"). gs allocates too much memory and everything 
becomes unusable. kghostview uses another method (dsc structure) to guess the 
size of the pixmap it will allocate and so it doesn't crash. 
 
So, the display problems come from gs (or from StarOffice, which is generating 
broken postscript). I have already realised that kghostview doesn't tell you anything 
when gs fails and I will solve it really soon (I have a patch in my computer, but my 
copy if half-broken and I wanted to talk to Wilco who is the maintainer first .... but it 
will be in CVS in a few days). 
 
I also think that this issue has come up in another report where we established that 
gs was failing and kghostview wasn't saying anything and this needed to be fixed. If 
I find that bug, I will mark this as a duplicate. 
 
Interestingly, if you do ps2ps brokenFile.ps brokenFileFixed.ps you will get a file 
which will work perfectly with gs and be viewable with kghostview although 
kghostview will guess the paper-size wrong (kghostview uses the DSC structure 
which gs ignores when parsing. ps2ps which hangs on gs generates DSC structure 
but apparently here it is wrong).  
If you like this sort of thing, iterating through ps2ps will get you an alternating 
sequence after n=2 (consider n=0 to be the original broken file). 
 
Comment 6 Luís Pedro Coelho 2003-04-09 00:21:31 UTC
Here is my reasoning: 
 
- That rendering fails is INVALID bc it is gs's fault. 
- That we don't tell the user (and sometimes crash) is a DUPLICATE 
 
So, I mark it as a DUPLICATE. 
 
Thank you for reporting. Do so again if you find other bugs. 
luis pedro coelho 

*** This bug has been marked as a duplicate of 56521 ***