GNOME Bugzilla – Bug 437528
StateSet.equals fails
Last modified: 2019-03-27 20:11:57 UTC
Please describe the problem: Working on collection I have found some problems using StateSet both on the "C" libspi implementatin and on th Pyorbit one. I have found these two things: 1- Steps to reproduce: 1. If i do s=obj.getState() ; s.equals(s) ==> I get False This happens both on C and Python 2. If I create a State in python then when I try to get it from C( get_atk_state_from_servant()) the program crashes. Actual results: Expected results: Does this happen every time? Other information:
Created attachment 87979 [details] Test case 1. This script shows the use of stateset using calls from both python and C and the results. You need to have gcalctool running.
The equals bug is my fault in Python. The crash could be how we're implementing the state set object. I'm grasping at straws about how to do it in Python.
Possibly a problem with calling methods on the Python implementation. Look at http://bugzilla.gnome.org/attachment.cgi?id=85914&action=view where the methods are invoked on the _this() object instead.
Doing: ss = StateSetImpl() poa = orb.resolve_initial_references("RootPOA") ss._this().add(a.STATE_VISIBLE) seems to solve the crash problem. I will test on Collection. Still pending the equals... thans ariel
Created attachment 87990 [details] [review] fixes equals
Ariel, I'm closing this bug as the problem with equals and opening a new one about the crashing problems.