After reading my initial question, I think I should've worded it differently.
> On Tue, 2007-10-16 at 16:12 -0600, Jeff Eberl wrote: > [...] > > First is that the display doesn't do anything when my program is doing > > it's thing. So it looks like it freezes. I'm okay with that. I've > > never used threads before, and I'm not sure this is the best project to > > start with. > > ofcourse this can be fixed by splitting up whatever your app does into > smaller less cpu/load intensive itterations thus updating the gui more > often, but that involves work (just like integration your execution > into timeout/idle callbacks would involve work). That's true, except when the parts that are taking away from the gtk loop are user prompts. This has really become a minor issue now though. > > Second is the most annoying part. Since there is no loop, > > gtk_main_quit() doesn't work, and when I click the little [X] in the > > corner, the destruction begins, calls my WindowDestroyEvent that I set > > up in the signals, but my program has no idea that the window has been > > destroyed. > > How can your program have no idea that the window is being closed > if you set the signal handler for "delete-event" ? > I should have mentioned that there is a class that I use that is a dort of gtk wrapper that connects this signal to it's function. I don't really want to change it though, because a lot of other code depends on that working properly. Making it virtual and inheriting it wouldn't cause too many problems. That's then next thing I'll do. > > Then the program keeps making the same calls to gtk. So gtk > > spits out a bunch of messages about how everything is failing. > > > > Is there a way to check to see if the window has been destroyed? I > > could put a member boolean that changes when WindowDestroyEvent gets > > called but that seems like a suspicious hack. > > However you do it, its definitely up to you to bookkeep that information > (some might keep a pointer to the active display window and set it to > NULL when the window is destroyed, I dont see how thats better than > a boolean variable). > > For elegancy, you might want to add it to a structure that gets passed > around to all callbacks instead of crudely packing it into global scope. > > Cheers, > -Tristan > That's the info I needed. If anyone wants to disagree with Tristan and tell me that there actually _is_ a way to check that, then please do. Otherwise, I think my question is answered. Thanks. _______________________________________________ gtk-app-devel-list mailing list gtk-app-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/gtk-app-devel-list
