Boris Zbarsky <[EMAIL PROTECTED]> wrote: > Ben Hutchings wrote: >> As it is, the loop never exits. My actual code for >> on_net_state_change includes some output statments which report its >> arguments. They show that it is never called with the REQUEST and >> STOP flags. If I remove the loop, it is called with these flags >> during the next call to DispatchEvent. > > Oh, hmm.... Does spinning the GTK event loop end up spinning the mozilla > event > loop too, here? That is, is PL_ProcessEvents getting called while you spin > the > event loop?
I assume you mean PL_ProcessPendingEvents? That is not being called. I wonder whether there's something wrong with the way I'm dispatching the event, that can cause further events to be blocked? I tried releasing the event (by setting the nsCOMPtr objects to 0) before running the loop, but this had no effect. I also changed the relatedTarget argument to DispatchEvent to be 0 since the previous value was clearly wrong. This had no effect either. Perhaps I've run into an obscure bug? I can't think that it would impact an ordinary browser at all, so it could have gone entirely unnoticed. I should be going to bed now, but tomorrow I shall try adding some code to fake additional dummy events since that seems to flush the network event through. -- Ben Hutchings Never attribute to conspiracy what can adequately be explained by stupidity. _______________________________________________ mozilla-embedding mailing list mozilla-embedding@mozilla.org http://mail.mozilla.org/listinfo/mozilla-embedding
