Re: Glib runloop

[Adrian Robert]

On 2005-04-17 10:54:13 -0400 Helge Hess <address@hidden> 
wrote:

> On 16. Apr 2005, at 14:31 Uhr, Banlu Kemiyatorn wrote:
> > Has anyone tried to unify Glib runloop and NSRunLoop?
>
> Well, yes, I did for GTKKit:
>
>   
> http://mail.gnome.org/archives/gtk-devel-list/1999-March/msg00158.html
>
> but 1999 is quite a while ago ;-)
>
> > Does this mean to reimplement NSRunLoop entirely?
>
> I think not necessarily, but it might be the easiest approach. I 
> suppose a 
> lot of things have changed in glib since 1999, so it might be rather 
> easy to 
> wrap the glib runloop as an NSRunLoop object.
>
> Personally I would be quite interested in such a work.


As an additional source of ideas, Emacs.app runs the NSRunLoop 
inferior to the main event loop run by the lisp engine.  When the lisp 
engine calls a select() (with nonzero timeout), [NSApp run] is called 
to process pending NS events.  A timer is set up to periodically call 
select() (with zero timeout).  When an NS event is received, it is 
converted to be handed over to the lisp engine, then an 
NSApplicationDefined event is appended to the queue.  NSApplication 
-sendEvent is overridden to intercept these and call [NSApp stop].  (I 
tried using NSFileHandle instead of timer-based polling to handle the 
select(), but this caused a problem with NSAutoreleasePool management 
on OS X.)

You could probably do something similar in a plain NSRunloop (no 
NSApp) environment, but I haven't tried, and I'm not sure whether it 
would generalize well to multithreading.  (And note the NSApp approach 
on GNUstep requires a patch to NSApplication which hasn't been checked 
in yet.. I'm hoping to do it soon.)