GNOME Bugzilla – Bug 587717
Remove stray error message about missing symbol.
Last modified: 2011-11-10 04:56:01 UTC
Please describe the problem: To make gst-python-0.10.15, download, make, and install the following packages: glib-2.21.2, gst-python-0.10.15, pygobject-2.18.0, gst-plugins-base-0.10.23, gstreamer-0.10.23, and pygtk-2.14.1. The make of gst-python-0.10.15 produces many warnings along the lines of Could not write method GstStructure.get_uint: No ArgType for 'guint*' Could not write virtual accessor method GstIndex.get_assoc_entry: No ArgType for 'GCompareDataFunc' Could not write virtual accessor method GstClock.wait: No ArgType for 'GstClockEntry*' ... The coverage values are all less than 100% except for interface proxies. The make will complete. Run python and import gst. You will get the message AttributeError: 'module' object has no attribute 'Element' As Element is the base class for all elements, its absence is a serious problem. Steps to reproduce: 1. 2. 3. See above. Actual results: See above. Expected results: import gst should not produce an error message. Does this happen every time? Yes. Other information:
I built GStreamer on Kubuntu to have a base unadulterated by anything to do with GTK. I downloaded tar packages gst-python-0.10.15, gstreamer-0.10.23, gst-plugins-base-0.10.23, pygtk-2.14.1, and pygobject-2.18.0. I installed the deb packages pkg-config, gettext, liboil0.3, and liboil0.3-dev. I set LD_LIBRARY_PATH=/usr/local/lib. I ran ./configure, make, sudo make install on the 6 directories created by the tar packages. All makes completed successfully (though not without comments). I ran python. I ran import gst. I got the message "'module' object has no attribute 'Element'".
* You don't need pygtk * You can make gst-python * The coverage warnings are not fatal (they're just an indication for the maintainers) * That error message is non-fatal So basically the only issue is to remove that error message.
I thought that the error message might provide an explanation for the actual problem that I am observing: The volume parameter has no effect with playbin2. If the error message is itself erroneous, then it would appear that there is a separate bug. Should I file a separate report?
Jeffrey: is this still an issue with current versions of gst-python?
Please feel free to reopen this bug if the problem still occurs with a newer version of gst-python.