Re: Update progress bar without returning to main loop?

[Toralf Lund <toralf procaptura com>]

Paul Davis wrote:

> Threads are easy enough.
>
> Oh, and I forgot to attach the tar file.
>
> Its not hard at all.

Threads are not that hard to program, but they still represent a major 
complexity "behind the scenes". One that shouldn't be introduced just 
like that, IMO, and, if the toolkit requires me to do it just to achieve 
a *simple* GUI update, then I really think there is something 
fundamentally wrong with its design.  But I'm thinking there must be 
another way...

> Toralf Lund wrote:
>
> > Paul Davis wrote:
> >
> > > No time to explain. I'm running late.
> > >
> > > Here it is.  Use threads.  Email questions.
> >
> >
> >
> > It's a very simple app, the complexity of threads has no place in it. 
> > Especially if it's just to make the GUI behave. In fact, introducing 
> > threads just for this seems very wrong for any program...
> >
> > > Good luck.
> > >
> > > Toralf Lund wrote:
> > >
> > > > I have what I believe is a classic issue with progress bars, but I 
> > > > have never found a good solution in the docs or via a web search:
> > > >
> > > > I'm trying to indicate progress of a signal handler that takes a 
> > > > *long* time via a Gtk::ProgressBar and multiple calls (as the work 
> > > > progresses) using the set_fraction() method. The problem is that 
> > > > these calls have no real effect, since the actual GUI update isn't 
> > > > done until after the application returns to the main loop (i.e. 
> > > > when the handler is done and the progress has reached 100%). The 
> > > > only way I've found to ensure it's redrawn as the work progresses, 
> > > > is to flush events using something like:
> > > >
> > > > while(Gtk::Main::events_pending())
> > > >   Gtk::Main::iteration(false);
> > > >
> > > > But this has some nasty side effects. In particular, it means that 
> > > > user input like button clicks, text field updates also handled, 
> > > > which is not what I want; I need the window to be unresponsive 
> > > > while the above mentioned callback is being executed.
> > > >
> > > > Is there an alternative?
> > > >
> > > > - Toralf
> > > >
> > > >
> > > >
> > > > _______________________________________________
> > > > gtkmm-list mailing list
> > > > gtkmm-list gnome org
> > > > http://mail.gnome.org/mailman/listinfo/gtkmm-list