Re: Pretest next week

[Jason Rumney]

Adrian Robert wrote:
> > 3) Make sure some kind of interrupt mechanism is in place to pick up 
> > ctrl-g events when emacs core is busy processing and does not itself 
> > make passes through the NS code event loop (colored spinning disk 
> > shown in gui).
>
> Here, the issue is that the SIGIO handler is never called, despite 
> being registered.  It might be overridden by Cocoa in some way, though 
> I haven't been able to fully confirm this.
>
> In any case, I tried to use input polling (Fset_input_interrupt_mode 
> (Qnil)), but poll_for_input() does not get called during, for example, 
> (while t t).  It seems that the QUIT macro does nothing to update 
> timers so the polling timer never fires.  Making the QUIT macro call 
> handle_async_input() every time slows down emacs (esp during startup).
>
> Does anyone have any suggestions?

On w32, we set Vquit_flag directly when Ctrl-G (quit_char) is pressed. 
This is done in the asyncronous key handling where the event gets put on 
the Lisp input queue, so it does not have to wait for the input queue to 
be polled.