Comment 1 Benedikt Meurer 2007-08-11 12:33:48 CEST

The assertion in thunar-vfs-path.c:1207 verifies that all path objects were freed prior to calling thunar_vfs_shutdown(). The assertion is disabled if thunar is build w/o debug support.

This basically means that the places plugin leaks ThunarVfsPaths objects somewhere. So this is not a bug in thunar-vfs, but in the places plugin.

Comment 2 Jérôme Guelfucci 2007-08-11 12:49:32 CEST

Ok thank you very much.

Comment 3 Jérôme Guelfucci 2007-08-11 12:51:55 CEST

Sorry this was reported here : https://bugs.launchpad.net/ubuntu/+source/xfce4-places-plugin/+bug/123804

Filled the bug report too fast...

Benny,
Would the following leak a thunar-vfs-path object? It's the only mention of thunar_vfs_path in v0.3 of the places plugin (for which this bug was reported).
thunar_vfs_path_dup_uri( thunar_vfs_volume_get_mount_point(volume) );

Jérôme,
Has this been reproduced with v0.9.991?

Thanks,
Diego

Comment 5 Jérôme Guelfucci 2007-08-23 18:22:14 CEST

Diego :

This happened with xfce4-places-plugins 1.0~beta1-0ubuntu2, I didn't try to reproduce it as this package is only available under the development release of Xubuntu which I don't run.

"This happened with xfce4-places-plugins 1.0~beta1-0ubuntu2"
Actually, launchpad and the traces show that this was v0.3.

I just tried reproducing the bug again with v0.3 and could not. I'm on debian lenny (testing) and have added --enable-debug=full (along with a couple extra g_printf's) to the thunar packages.

The following things would help:
1) Whether or not this bug affects v0.9.991
2) More specific instructions on reproducing this bug (including exactly how the language was changed and when the crash occurred)

Comment 7 Jérôme Guelfucci 2007-09-08 11:10:50 CEST

Asked to original reporter.

Comment 8 Jérôme Guelfucci 2007-09-15 15:56:26 CEST

Original reporter is suffering of an xserver bug so he can't try to reproduce this   now, I will report you any progress.

Comment 9 Jérôme Guelfucci 2007-09-15 15:57:02 CEST

Forget to copy that :

"The language was changed via SCIM using a font called bolnagri which didn't need SCIM. http://www.indlinux.org/blog/?cat=7"

I just tried reproducing this again, this time on a Xubuntu Gutsy VirtualBox image, with no luck.

If anyone can reproduce this, I would like to know. On the other hand, this is currently the only outstanding issue that I know of, and I've been promising the big 1.0 for a while now. Let's consider this a final call for anyone that can reproduce the issue to step up/back up.

Comment 11 Jérôme Guelfucci 2007-10-18 15:37:44 CEST

I asked this to the original reporter but he is kind of reluctant to answer...

Please reopen this bug if you receive any more information.

Comment 13 Jérôme Guelfucci 2007-10-26 15:53:14 CEST

Ok, sorry about that...