After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 310306 - Crashes with failed assertion: file ev-page-cache.c: line 187 (_ev_page_cache_new): assertion failed
Crashes with failed assertion: file ev-page-cache.c: line 187 (_ev_page_cache...
Status: RESOLVED FIXED
Product: evince
Classification: Core
Component: general
0.3.x
Other other
: High critical
: ---
Assigned To: Evince Maintainers
Evince Maintainers
: 310948 313698 (view as bug list)
Depends on:
Blocks: 312019
 
 
Reported: 2005-07-14 02:23 UTC by Bradley Baetz
Modified: 2005-08-17 14:32 UTC
See Also:
GNOME target: ---
GNOME version: 2.9/2.10


Attachments
crashing pdf (395 bytes, application/pdf)
2005-07-14 02:23 UTC, Bradley Baetz
  Details
Proposed patch (11.71 KB, patch)
2005-07-15 00:29 UTC, Nickolay V. Shmyrev
rejected Details | Review

Description Bradley Baetz 2005-07-14 02:23:04 UTC
Distribution: Fedora Core release 4 (Stentz)
Package: evince
Severity: critical
Version: GNOME2.10.0 0.3.x
Gnome-Distributor: Red Hat, Inc
Synopsis: Crashes with failed assertion: file ev-page-cache.c: line 187 (_ev_page_cache_new): assertion failed
Bugzilla-Product: evince
Bugzilla-Component: general
Bugzilla-Version: 0.3.x
BugBuddy-GnomeVersion: 2.0 (2.10.0)
Description:
Description of the crash:
Evince crashes when opening a particualar pdf file:

** ERROR **: file ev-page-cache.c: line 187 (_ev_page_cache_new):
assertion failed: (page_cache->uniform_width > 0 &&
page_cache->uniform_height > 0)

Steps to reproduce the crash:
1. evince /tmp/saffron_output_29BD.pdf

Expected Results:
Not crash

How often does this happen?
Always

Additional Information:
PDF was obtained from:

http://gecko.dynalivery.com/geckoresults.asp?fname=http%3A%2F%2Fwww.bbc.com&outputType=1&footer=Powered+By+Mozilla+Gecko+%26+Dynalivery+Saffron

but thats dynamic, so I'll attach the file. I suspect that the PDF is
invalid since its too small to really be the BBC home page...


Debugging Information:

Backtrace was generated from '/usr/bin/evince'

Using host libthread_db library "/lib/libthread_db.so.1".
[Thread debugging using libthread_db enabled]
[New Thread -1209088320 (LWP 11045)]
[New Thread -1211216976 (LWP 11046)]
0x002ae402 in ?? ()

Thread 1 (Thread -1209088320 (LWP 11045))

  • #0 ??
  • #1 __waitpid_nocancel
    from /lib/libpthread.so.0
  • #2 libgnomeui_module_info_get
    from /usr/lib/libgnomeui-2.so.0
  • #3 <signal handler called>
  • #4 ??
  • #5 raise
    from /lib/libc.so.6
  • #6 abort
    from /lib/libc.so.6
  • #7 g_logv
    from /usr/lib/libglib-2.0.so.0
  • #8 g_log
    from /usr/lib/libglib-2.0.so.0
  • #9 g_assert_warning
    from /usr/lib/libglib-2.0.so.0
  • #10 _ev_page_cache_new
    at ev-page-cache.c line 187




------- Bug moved to this database by unknown@gnome.bugs 2005-07-14 02:23 UTC -------

Comment 1 Bradley Baetz 2005-07-14 02:23:54 UTC
Created attachment 49137 [details]
crashing pdf
Comment 2 Teppo Turtiainen 2005-07-14 07:11:28 UTC
According to simple-dup-finder this is a unique stack trace.
Comment 3 Bradley Baetz 2005-07-14 08:18:51 UTC
FWIW, acrobat complains that the document can't be opened because it has no pages.
Comment 4 Nickolay V. Shmyrev 2005-07-15 00:29:36 UTC
Created attachment 49198 [details] [review]
Proposed patch

I think we might have errors in page cache creation function, not the only
errors handled by this assert. And the right way to find this errors - pass
GError to page cache creation. So I propose the following patch.
Comment 5 Marco Pesenti Gritti 2005-07-20 09:03:51 UTC
If I remember correctly Bryan didnt want to handle this case (no pages) with an
error dialog but by simply not showing any page.

I think part of the code is handling that correctly, because it used to work...
Comment 6 Bryan W Clark 2005-07-20 15:17:54 UTC
I think we decided on showing no pages instead of an error dialog.
Comment 7 Nickolay V. Shmyrev 2005-07-20 20:04:08 UTC
*** Bug 310948 has been marked as a duplicate of this bug. ***
Comment 8 Marco Pesenti Gritti 2005-07-26 11:35:46 UTC
Rejecting patch since an error dialog is not what we want.
Comment 9 Jonathan Blandford 2005-07-30 03:16:01 UTC
Fixed, though it looks mildly weird.  I wonder if the dialog makes more sense. 
You can bring up the properties, though, so that's cool.
Comment 10 Nickolay V. Shmyrev 2005-08-17 14:32:33 UTC
*** Bug 313698 has been marked as a duplicate of this bug. ***