The biggest things are:
- getting tactics to not take affect until their duration To do this, I can add behavior to where events are processed i.e., when a start event happens, instead of having the system state accept it, just put a complete event into the timeline, which will trigger the actual change or not
I must also update how the probabilities are handled to be compatible with this, i.e., avoid double counting them (first event always gets 1.0)
-
continuing along these lines, I need to fix handing probabilities for tactics that fail for reasons other than simulation (intentional fail)
-
building back in support for try-catch and for loop tactics
This can be done by having those start events put the other events on the list
- Then I can put the parallel tactics back in This will work like try-catch, simply putting the needed events on the timeline