Robin Dunn wrote:
Petar Dimitrijevic wrote:
It's not a problem to send you the entire code but I don't think you would be able to run it without a controller :(.
I wouldn't want to take the time to grok the whole app anyway...
Can you suggest something that I could do that could shed some light on this ?
Incrementally disable parts of the app until the problem goes away, then that will tell you what part you need to focus on. Perhaps then you can think of a way to simulate just that part in a sample app.
Also would it make a difference if I wrote the screen update routine in the wx.IDLE event instead of using wx.CallAfter ?
Perhaps. You can put the info you need to do the update on a Queue from the worker thread, and then call wx.WakeUpIdle(). Then the EVT_IDLE handler can just read the info from the queue and do the updates as needed. This might result in the same problem, but since the wxEvent processing part of wx.CallAfter is eliminated it might be enough to avoid the bug.
Ok. I'll dissect it and if something new comes up I'll notify the list.
Thanks for the help.