Yes. I agree. It would "prevent" the "no case" by returning an error on parse/evaluation. However the solution described here can do better. It can correctly pa

Yes. I agree. It would "prevent" the "no case" by returning an error on parse/evaluation. However the solution described here can do better. It can correctly parse the no case. Basically by knowing it is parsing a string the grammar can be simpler, it doesn't have to decide if it is a int/bool/string anymore.