Closed albertZhangTJ closed 11 months ago
I thought about having two separate functions for defining an ID and for accessing an ID. The reason behind is if, for example, a quantifier node instantiated an RP_ ID to a value that another node cannot follow, then we have a problem.
But then I realized that this kind of declare-access relationship might not be very clear in some cases and is certainly something the tester might not get.
So I am going to resort to the solution of throwing an error when this happens and regenerate the entire test case for now.
Decide to resort to throwing an error when no identifier is available and regenerate (after all generating a test case is rather cheap and the computational overhead for a full solution might not be worth it). Will close this issue for now.
Wont fix
Quantity limit must be set correctly when, for example, the number of available identifiers does not match the originally generated repetition number.