Em 31/07/2009, às 10:50, Dario Freddi escreveu: > In data venerdì 31 luglio 2009 15:46:10, Gustavo Boiko ha scritto: >>> The interval is already the ID. Actually, since the idle time is >>> always the >>> same, setting 2 intervals in the very same moment will end up in >>> having a >>> single 500ms interval, since both would fire at the very same time. >> >> Yes, I know they will fire at the same time. But my concern is if one >> of the timers added ask for removing the interval, will the other >> still fire? >> A refcount would probably do the job, but since there is a similar >> API >> for other timeout event, I thought it would be a good thing to keep >> it >> consistent. > > You're right on a second thought, so I think it's sensible to > implement such a > thing. so setIdleTimeout should return an int, carrying the token, > but what > about the signal? should it emit the interval or the token number?
The token number, probably (or at least it is how QObject::timerEvent() works). The one who adds the timeout watching should be responsible to know what the timeout is for that id. > By the way, I'm about to leave for some holidays, so I'd probably > have to > implement this when I'll come back in september. But if you are > willing to do > so (and it would be better, since I didn't know about this kind of > API) you > could try and commit the change (please ping toma if you do so) or > if you > don't have time simply drop me a mail with the API changes needed :) Well, I'm not that experienced with the API either. I just happened to have used QObject::timerEvent some times now :) If I happen to have time I will ping toma and discuss about, but I can't promise I'll be able to do so though. Cheers Boiko _______________________________________________ kopete-devel mailing list kopete-devel@kde.org https://mail.kde.org/mailman/listinfo/kopete-devel