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 585458 - crash in gtk.RecentInfo.get_application_info()
crash in gtk.RecentInfo.get_application_info()
Status: RESOLVED FIXED
Product: pygtk
Classification: Bindings
Component: general
2.14.x
Other Linux
: Normal normal
: ---
Assigned To: Nobody's working on this now (help wanted and appreciated)
Python bindings maintainers
Depends on:
Blocks:
 
 
Reported: 2009-06-11 18:02 UTC by Markus Korn
Modified: 2009-08-11 13:26 UTC
See Also:
GNOME target: ---
GNOME version: 2.25/2.26


Attachments
terminal output (8.19 KB, text/plain)
2009-06-11 18:05 UTC, Markus Korn
Details

Description Markus Korn 2009-06-11 18:02:27 UTC
The following code crashes for me reproducible:

###
import gtk

recent = gtk.recent_manager_get_default()
for info in recent.get_items():
    last_application = info.last_application()
    application_info = info.get_application_info(last_application)
###

With something like
*** glibc detected *** python: double free or corruption (!prev): 0x00000000024c4e20 ***

please see the attachment for the complete backtrace.

I'm not sure which further information are needed, but I'm happy to provide any information to debug this further.

Markus
Comment 1 Markus Korn 2009-06-11 18:05:34 UTC
Created attachment 136344 [details]
terminal output
Comment 2 Paul Pogonyshev 2009-06-11 18:46:49 UTC
I cannot reproduce this.  Can you try if the bug happens with the latest (2.15.1) version for you?  Are all statements above necessary?  E.g. can you remove get_application_info call and still get a crash?
Comment 3 Paul Pogonyshev 2009-06-11 18:51:39 UTC
Nevermind, actually I can reproduce.  Tried this in interactive mode and error hapenned only after I quit the interpreter.
Comment 4 Paul Pogonyshev 2009-06-11 19:02:19 UTC
Fixed in development branch.
Comment 5 Markus Korn 2009-06-12 09:18:26 UTC
hi Paul,
wow, you are fast, thanks for the fix.
Is the development branch available somewhere to the public? - I would like to check if this fixes my initial problems.
When can I expect the release of this fix, is there a pygtk roadmap somewhere?

Thanks again,
Markus
Comment 6 Paul Pogonyshev 2009-06-12 21:31:16 UTC
(In reply to comment #5)
> Is the development branch available somewhere to the public? - I would like to
> check if this fixes my initial problems.

Sure, see http://git.gnome.org/cgit/.  You want 'master' branch.

> When can I expect the release of this fix, is there a pygtk roadmap somewhere?

We are way too low on human resources to make any promises.  Hopefully soon, but no guarantee.
Comment 7 Siegfried Gevatter (RainCT) 2009-08-11 13:16:43 UTC
Thanks for fixing this.

Can you say in which version this will be fixed? (2, 15, 2), (2, 16, 1), something else?

We want to use this function in Zeitgeist and will have to use a workaround for now (parsing the XML files instead of using PyGTK for it). If you can guarantee in which version number this will be fixed I can add a conditional check to only use the workaround in older PyGTK versions.
Comment 8 Gian Mario Tagliaretti 2009-08-11 13:26:32 UTC
(In reply to comment #7)
> Thanks for fixing this.
> 
> Can you say in which version this will be fixed? (2, 15, 2), (2, 16, 1),
> something else?

It will be really soon into 2.16.0.