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 353526 - crash in Deskbar:
crash in Deskbar:
Status: RESOLVED FIXED
Product: deskbar-applet
Classification: Deprecated
Component: general
Git master
Other All
: High critical
: 2.16
Assigned To: Deskbar Applet Maintainer(s)
Deskbar Applet Maintainer(s)
: 353555 354164 354546 (view as bug list)
Depends on:
Blocks:
 
 
Reported: 2006-08-30 10:14 UTC by myhansheng
Modified: 2006-09-06 05:25 UTC
See Also:
GNOME target: 2.16.x
GNOME version: 2.15/2.16


Attachments
Check that last_visible_path is not null (1.51 KB, patch)
2006-08-30 21:07 UTC, Mikkel Kamstrup Erlandsen
none Details | Review
same patch as above, but it removes a misleading FIXME too (1.52 KB, patch)
2006-08-30 21:12 UTC, Mikkel Kamstrup Erlandsen
none Details | Review

Description myhansheng 2006-08-30 10:14:20 UTC
What were you doing when the application crashed?



Distribution: Ubuntu 6.10 (edgy)
Gnome Release: 2.15.92 2006-08-22 (Ubuntu)
BugBuddy Version: 2.15.92

Memory status: size: 0 vsize: 0 resident: 0 share: 0 rss: 0 rss_rlim: 0
CPU usage: start_time: 0 rtime: 0 utime: 0 stime: 0 cutime:0 cstime: 0 timeout: 0 it_real_value: 0 frequency: 0

Traceback (most recent call last):
  • File "/usr/lib/python2.4/site-packages/deskbar/ui/cuemiac/CuemiacUIManager.py", line 333 in _on_entry_activate
    iter = self.model.get_iter (self.cview.last_visible_path())
TypeError: GtkTreeModel.get_iter requires a tree path as its argument

Comment 1 Mikkel Kamstrup Erlandsen 2006-08-30 18:07:13 UTC
How did you encounter this bug? Is there any way we can reproduce it?
Comment 2 Karsten Bräckelmann 2006-08-30 18:19:02 UTC
*** Bug 353555 has been marked as a duplicate of this bug. ***
Comment 3 myhansheng 2006-08-30 18:24:35 UTC
this is auto reported bug when i first time using the Deskbar search..  TQ for reply.
Comment 4 Mikkel Kamstrup Erlandsen 2006-08-30 21:07:37 UTC
Created attachment 71926 [details] [review]
Check that last_visible_path is not null

*Gulp*, I can reproduce this by simply hitting enter in an empty entry... Button or entry view - it doesn't matter.

Luckily I have a patch. We need release team clearance though, since we are in hard freeze.

This bug is a bit related to bug #345806 - "Search string truncated if user is a fast typist". I think this bug might be triggered when you type something very fast and then hit enter before any keystrokes are registered at all.
Comment 5 Mikkel Kamstrup Erlandsen 2006-08-30 21:12:06 UTC
Created attachment 71928 [details] [review]
same patch as above, but it removes a misleading FIXME too
Comment 6 Nigel Tao 2006-08-31 04:59:27 UTC
Hmm... just to be safe, should we also have
----------------------
else:
  iter = None
----------------------
to cover the case where (iter != None and self.model.iter_has_child(iter))??

And, yes we will need freeze-break clearance.  The process is at http://live.gnome.org/ReleasePlanning/Freezes
Comment 7 Mikkel Kamstrup Erlandsen 2006-08-31 05:11:06 UTC
not that i can tell... We set iter = None at the top of that block anyway...
Comment 8 Nigel Tao 2006-08-31 05:28:47 UTC
We do set iter = None, but soon after, we (conditionally) call iter = self.model.get_iter (path).

You are probably right, but at this stage, I would rather be safe than sorry.  :-)
Comment 9 Mikkel Kamstrup Erlandsen 2006-09-03 09:53:29 UTC
Ok. I understand your confusion Nigel. The code is a bit tricky, but correct nonetheless (to my understanding atleast :-D).

Consider the case where nothing is selected in the tree, and you hit enter in the entry to activate the top hit. The topmost element of the tree will be a category header, so you have to select the child of the first element.

This is exactly the case you describe where (iter != None and self.model.iter_has_child(iter)).

Maybe this ought to be documented in the code.
Comment 10 Raphael Slinckx 2006-09-03 15:54:55 UTC
Committed in HEAD since this fixes a crasher. I didn't use nigel's modificaiton. Mikkel i hope you're right ;)
Comment 11 Sergej Kotliar 2006-09-03 20:16:48 UTC
*** Bug 354164 has been marked as a duplicate of this bug. ***
Comment 12 Mikkel Kamstrup Erlandsen 2006-09-06 05:25:56 UTC
*** Bug 354546 has been marked as a duplicate of this bug. ***