Re: revised mainEventLoop

Le 8 mar 2013 à 17:32, chris nuernberger a écrit :

> Furthermore you could have conditions in the datamodel which have changed thus enabling eventless transitions even though to an outside observer you are testing the same eventless transition set over and over again.

The SC itself can't do anything while you're just dequeuing events from the internal queue. So any change to the datamodel (except for _event) will have to be done by the platform, which should know better than to allow them in the middle of an event loop. Or by side-effects in event conditions. But please let's assume the platform is safe enough and the authors aren't hacking with side-effects. That means the only thing changing in the whole SC, while dequeueing events as I suggested, is _event.

			David

Received on Friday, 8 March 2013 17:10:38 UTC