On 3/13/08, Jo?o Mesquita <jmesquita at contactnet.com.br> wrote: > Yes, I do have a worker that polls handle_events every 50ms. Why 50ms? > Cos I saw on the example ... :D 50ms seems to give good signaling responsiveness without hogging the CPU with busy polling. > The worker is not really a different thread since I am attaching that to > the GTK main loop using gobject.timeout_add function. > > Do I have to remove the source before cleaning up? I have no idea about GTK, but yes you need to somehow stop it before you call destroy(). > The __del__ function is the destroyer, but my guess is that py_pjsua is > not blocking, so gtk.main_quit() simply kills the loop b4 anything else > gets done. How has blocking or non-blocking got anything to do with it? Like I said, I have no idea about using GTK with Python, but I would expect that if you call destroy(), then Python will allow destroy() function to complete and not just kill it. cheers, -benny > Thanks, > > > Mesquita > > > On Thu, 2008-03-13 at 21:18 +0000, Benny Prijono wrote: > > On 3/13/08, Jo?o Mesquita <jmesquita at contactnet.com.br> wrote: > > > I am doing > > > py_pjsua.destroy() > > > gtk.main_quit() > > > > That's not exactly what I'm asking. In your script, do you have a > > worker thread? Or something that calls py_pjsua.handle_events()? Is > > so, then this worker thread needs to be stopped before you call > > py_pjsua.destroy(). > > > > > If I wrap py_pjsua on another module/class and call py_pjsua.destroy() > > > on the __del__ method, it simply quits and does no cleanup... > > > > I'm not sure about this. Maybe py_pjsua.destroy() is not called at all? > > > > > What is the cons of doing the latter? > > > > > > > You should make sure that py_pjsua.destroy() is called or otherwise > > things don't get cleaned up properly (dangling calls, dangling > > registration in the server, etc.). > > > > cheers, > > -benny > > > > > > > Mesquita > > > > > _______________________________________________ > > Visit our blog: http://blog.pjsip.org > > > > pjsip mailing list > > pjsip at lists.pjsip.org > > http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org > > > _______________________________________________ > Visit our blog: http://blog.pjsip.org > > pjsip mailing list > pjsip at lists.pjsip.org > http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org >
