A list of rules, together with a context for evaluating them and interpreting results, for use in constructing data-driven logic or fuzzy logic systems.
- iOS 9.0+
- macOS 10.11+
- Mac Catalyst 13.0+
- tvOS 9.0+
GKRule object manages a list of rules (
GKRule objects). A rule system also offers methods for evaluating its list of rules in a context defined by two features: a
state dictionary containing information to be tested by rules, and a set of
facts representing the conclusions drawn as a result of rule evaluation. You can evaluate facts based on a binary truth state—that is, a fact either is or is not in the set—or on a continuously variable membership grade, representing different levels of veracity, confidence, or strength for use in fuzzy logic.
You construct a rule system by creating
GKRule objects and adding them to the system’s list of rules. There are multiple ways to construct rules: for greater reusability, use the methods listed in Creating Data-Driven Rules; or for greater flexibility, use the
rule method or create a custom subclass of
GKNSPredicate. Then, add rules to the system with the methods listed in Managing a System’s List of Rules.
To evaluate a system, call the
evaluate method. This method processes each rule in the system in the order it appears in the system’s
agenda list. You set this order with the
salience property of each rule, or with the order in which you add rules to the system. As the system processes each rule, it tests the rule’s
evaluate method to determine whether the rule is satisfied in the context of the system. If the rule’s predicate is satisfied, the system executes the rule’s
perform method and moves the rule to the
executed list (so the further evaluation of the agenda doesn’t repeatedly trigger the rule’s action).
Rules typically use the system’s
state dictionary as input and its set of
facts as output. (However, more complex systems can include sets of rules whose predicates test facts or whose actions mutate the system’s state.) After evaluating a rule system, you can examine the set of facts it has produced using the methods listed in Drawing Conclusions from Facts. You can then use the presence of a fact in the set, the value of its membership grade, or the combined membership grades of a group of facts to influence the behaviors in your game.