gcompris crashes when 'Rebuild the Mosaic' is the first board used

trying to help

Hi, I had some difficulty finding that application. For me it was 1st item on the left-side menu Discovery Activities), 4th item on the first row (Color Activities), 1st item on the submenu (Rebuild the mosaic).
Is that the correct application ?

What do you mean by 'draw' the piece ? I can click on any of the colors on the row of colors at the bottom and then a green 'check-sign' will be in that color. Then I can click on any of the blocks to the top-right side and it will become that color.

Please supply more information (maybe a screenshot ?)

Attached is the screenshot. When I select a color down the bottom, then then fill one of the empty items on the right top, it crashes. Both in full screen as in windowed.

ii gcompris 8.3.2-1ubuntu2 Educational games for small children

===============
$ gcompris
exec_prefix NULL
XF86VidMode: Compiled with XF86VidMode.
If you have problems starting GCompris in fullscreen, try the -x option to disable XF86VidMode.

** (process:6587): WARNING **: Binary relocation disabled
package_data_dir = /usr/share/gcompris/boards
package_locale_dir = /usr/share/locale
package_plugin_dir = /usr/lib/gcompris
package_python_plugin_dir= /usr/share/gcompris/python
Infos:
   Config dir '/home/egonw/.config/gcompris'
   Users dir '/home/egonw/My GCompris'
   Database '/home/egonw/.config/gcompris/gcompris_sqlite.db'
===============

It does not dump any error messages... :(

Full screen or windowed both crash. That '-x' option in the above output mentioned does not help either.

I can fully reproduce this now on Gutsy and with the same version of gcompris.

After some more tests, I found out that the application 'Rebuild the Mosaic' will NOT crash when first playing another game.

Could you try that and confirm that? Thanks!

PS I ran gcompris -D but it didnt' give any information about the crash.
The last line before it crashed was:
gcompris: sdlplayer_fx(/usr/share/gcompris/boards/sounds/bleep.wav) ended.

Confirmed: it does not crash when not started as first module.

as per guidelines in https://wiki.ubuntu.com/Bugs/HowToTriage I assigned the big to nobidy now it's confirmed.

I confirm the issue but I cannot explain it. Here is the backtrace:
[Switching to Thread -1224730944 (LWP 14744)]
_wrap_gnome_canvas_item_set (self=0x84a04b4, args=0x8652404, kwargs=0x84a446c) at canvas.override:172
172 canvas.override: No such file or directory.
        in canvas.override
(gdb) bt
#0 _wrap_gnome_canvas_item_set (self=0x84a04b4, args=0x8652404, kwargs=0x84a446c) at canvas.override:172
#1 0xb513c85d in PyCFunction_Call (func=0x8493b4c, arg=0x8652404, kw=0x0) at ../Objects/methodobject.c:77
#2 0xb5105c17 in PyObject_Call (func=0x84a04b4, arg=0x8652404, kw=0x84a446c) at ../Objects/abstract.c:1860
#3 0xb51858cc in PyEval_EvalFrameEx (f=0x8334714, throwflag=0) at ../Python/ceval.c:3775
#4 0xb518b734 in PyEval_EvalCodeEx (co=0x87f95c0, globals=0x849cd74, locals=0x0, args=0x84a81a0,
    argcount=4, kws=0x0, kwcount=0, defs=0x0, defcount=0, closure=0x0) at ../Python/ceval.c:2831
#5 0xb5128a4a in function_call (func=0xb50b0d84, arg=0x84a8194, kw=0x0) at ../Objects/funcobject.c:517
#6 0xb5105c17 in PyObject_Call (func=0x84a04b4, arg=0x84a8194, kw=0x0) at ../Objects/abstract.c:1860
#7 0xb510d438 in instancemethod_call (func=0x84a0504, arg=0x84a8194, kw=0x0)
    at ../Objects/classobject.c:2509
#8 0xb5105c17 in PyObject_Call (func=0x84a04b4, arg=0x84a0644, kw=0x0) at ../Objects/abstract.c:1860
#9 0xb518380c in PyEval_CallObjectWithKeywords (func=0x84a0504, arg=0x84a0644, kw=0x0)
    at ../Python/ceval.c:3433
#10 0xb5105f2c in PyObject_CallObject (o=0x84a0504, a=0x84a0644) at ../Objects/abstract.c:1851
#11 0xb5055607 in ?? () from /var/lib/python-support/python2.5/gtk-2.0/gobject/_gobject.so

I have absolutely no idea what may cause this crash. The workaround is to play at another python based activity first.
I tested on gutsy GCompris 8.3.4, the latest stable in the 8.3 branch and it does work fine.
Sadly, there are no changes I can see concerning this code. There maybe a side effect in the build process or from another
fix.

Can someone build a 8.3.4 on gutsy following the official process and check if the problem is still there.

I can confirm that this still happens in 8.4.2-1ubuntu3 but when I install a newly built 8.4.4-1.1 it works fine.

This bug was fixed in the package gcompris - 8.4.4-1.1ubuntu1

---------------
gcompris (8.4.4-1.1ubuntu1) hardy; urgency=low

  * Merge from Debian unstable (Closes LP: #156679), remaining changes:
    - debian/control:
       + use gnome-games instead of gnuchess
       + add intltool as a build dependency
       + depend on python (>= 2.5) | python-pysqlite2
    - debian/rules: add intltool-update to generate .pot
    - update all .py files importing pysqlite2 to try importing
      from sqlite3 first before pysqlite2
  * Modify Maintainer value to match the DebianMaintainerField
    specification.

gcompris (8.4.4-1.1) unstable; urgency=low

  * Non-maintainer upload.
  * Fix the docdir screwup. Closes: #469430, #469433
  * Drop python-xml dependency. Closes: #468589

gcompris (8.4.4-1) unstable; urgency=low

  * New upstream release.
  * New voice packages for Bulgarian and Urdu.
  * Setup gcompris-dbg docdir as a symlink as well.
  * Added the LAL text to debian/copyright for the LuneRouge sounds, do
    not install the license file.
  * Bumped Standards-Version to 3.7.3, no additional change required.

 -- Jordan Mantha <email address hidden> Wed, 26 Mar 2008 11:07:46 -0700