Inheritance allows you to derive a new class from an existing class. You do this using the INHERITING FROM addition in the

CLASS <subclass> DEFINITION INHERITING FROM <superclass>.

statement. The new class <subclass> inherits all of the components of the existing class <superclass>. The new class is called the subclass of the class from which it is derived. The original class is called the superclass of the new class.

If you do not add any new declarations to the subclass, it contains the same components as the superclass. However, only the public and protected components of the superclass are visible in the subclass. Although the private components of the superclass exist in the subclass, they are not visible. You can declare private components in a subclass that have the same names as private components of the superclass. Each class works with its own private components. Methods that a subclass inherits from a superclass use the private attributes of the superclass, and not any private components of the subclass with the same names.

If the superclass does not have a private visibility section, the subclass is an exact replica of the superclass. However, you can add new components to the subclass. This allows you to turn the subclass into a specialized version of the superclass. If a subclass is itself the superclass of further classes, you introduce a new level of specialization.

A class can have more than one direct subclass, but it may only have one direct superclass. This is called single inheritance. When subclasses inherit from superclasses and the superclass is itself the subclass of another class, all of the classes involved form an inheritance tree, whose degree of specialization increases with each new hierarchical level you add. Conversely, the classes become more generalized until you reach the root node of the inheritance tree. The root node of all inheritance trees in ABAP Objects is the predefined empty class OBJECT. This is the most generalized class possible, since it contains neither attributes nor methods. When you define a new class, you do not have to specify it explicitly as the superclass - the relationship is always implicitly defined. Within an inheritance tree, two adjacent nodes are the direct superclass or direct subclass of one another. Other related nodes are referred to as superclasses and subclasses. The component declarations in a subclass are distributed across all levels of the inheritance tree.

All subclasses contain the components of all classes between themselves and the root node in an inheritance tree. The visibility of a component cannot be changed. However, you can use the REDEFINITION addition in the METHODS statement to redefine an inherited public or protected instance method in a subclass and make its function more specialized. When you redefine a method, you cannot change its interface. The method retains the same name and interface, but has a new implementation.

The method declaration and implementation in the superclass is not affected when you redefine the method in a subclass. The implementation of the redefinition in the subclass obscures the original implementation in the superclass.

Any reference that points to an object of the subclass uses the redefined method, even if the reference was defined with reference to the superclass. This particularly applies to the self-reference ME->. If, for example, a superclass method M1 contains a call CALL METHOD [ME->]M2, and M2 is redefined in a subclass, calling M1 from an instance of the subclass will cause the original method M2 to be called, and calling M1 from an instance of the subclass will cause the redefined method M2 to be called.

Within a redefine method, you can use the pseudoreference SUPER-> to access the obscured method. This enables you to use the existing function of the method in the superclass without having to recode it in the subclass.

Reference variables defined with reference to a superclass or an interface defined with reference to it can also contain references to any of its subclasses. Since subclasses contain all of the components of all of their superclasses, and given that the interfaces of methods cannot be changed, a reference variable defined with reference to a superclass or an interface implemented by a superclass can contain references to instances of any of its subclasses. In particular, you can define the target variable with reference to the generic class OBJECT.

When you create an object using the CREATE OBJECT statement and a reference variable typed with reference to a subclass, you can use the TYPE addition to create an instance of a subclass, to which the reference in the reference variable will then point.

A static user can use a reference variable to address the components visible to it in the superclass to which the reference variable refers. However, it cannot address any specialization implemented in the subclass. If you use a dynamic method call, you can address all components of the class.

If you redefine an instance method in one or more subclasses, you can use a single reference variable to call different implementations of the method, depending on the position in the inheritance tree at which the referenced object occurs. This concept that different classes can have the same interface and therefore be addressed using reference variables with a single type is called polymorphism.

Subclasses contain all of the components of all of their superclasses within the inheritance tree. Of these components, only the public and protected ones are visible. All public and protected components within an inheritance tree belong to the same namespace, and consequently must have unique names. The names of private components, on the other hand, must only be unique within their class.

When you redefine methods, the new implementation of the method obscures the method of the superclass with the same name. However, the new definition replaces the previous method implementation, so the name is still unique. You can use the pseudoreference SUPER-> to access a method definition in a superclass that has been obscured by a redefinition in a subclass.

Like all components, static attributes only exist once in each inheritance tree. A subclass can access the public and protected static attributes of all of its superclasses. Conversely, a superclass shares its public and protected static attributes with all of its subclasses. In terms of inheritance, static attributes are not assigned to a single class, but to a part of the inheritance tree. You can change them from outside the class using the class component selector with any class name, or within any class in which they are shared. They are visible in all classes in the inheritance tree.

When you address a static attribute that belongs to part of an inheritance tree, you always address the class in which the attribute is declared, irrespective of the class you specify in the class selector. This is particularly important when you call the static constructors of classes in inheritance. Static constructors are executed the first time you address a class. If you address a static attribute declared in a superclass using the class name of a subclass, only the static constructor of the superclass is executed.

There are special rules governing constructors in inheritance.

Every class has an instance constructor called CONSTRUCTOR. This is an exception to the rule that states that component names within an inheritance tree must be unique. However, the instance constructors of the various classes in an inheritance tree are fully independent of one another. You cannot redefine the instance constructor of a superclass in a subclass, neither can you call one specifically using the statement CALL METHOD CONSTRUCTOR. Consequently, no naming conflicts can occur.

The instance constructor of a class is called by the system when you instantiate the class using CREATE OBJECT. Since a subclass contains all of the visible attributes of its superclasses, which can also be set by instance constructors, the instance constructor of a subclass has to ensure that the instance constructors of all of its superclasses are also called. To do this, the instance constructor of each subclass must contain a CALL METHOD SUPER->CONSTRUCTOR statement. The only exception to this rule are direct subclasses of the root node OBJECT.

In superclasses without an explicitly-defined instance constructor, the implicit instance constructor is called. This automatically ensures that the instance constructor of the immediate superclass is called.

When you call an instance constructor, you must supply values for all of its non-optional interface parameters. There are various ways of doing this:

• Using CREATE OBJECT

• Using CALL METHOD SUPER->CONSTRUCTOR

In both CREATE OBJECT and CALL METHOD SUPER->CONSTRUCTOR, you must look at the next-available explicit instance constructor and, if it has an interface, pass values to it. The same applies to exception handling for instance constructors. When you work with inheritance, you need an precise knowledge of the entire inheritance tree. When you instantiate a class at the bottom of the inheritance tree, you may need to pass parameters to the constructor of a class that is much nearer the root node.

The instance constructor of a subclass is divided into two parts by the CALL METHOD SUPER->CONSTRUCTOR statement. In the statements before the call, the constructor behaves like a static method, that is, it cannot access the instance attributes of its class. You cannot address instance attributes until after the call. Use the statements before the call to determine the actual parameters for the interface of the instance constructor of the superclass. You can only use static attributes or local data to do this.

When you instantiate a subclass, the instance constructors are called hierarchically. The first nesting level in which you can address instance attributes is the highest-level superclass. When you return to the constructors of the lower-level classes, you can also successively address their instance attributes.

In a constructor method, the methods of the subclasses of the class are not visible. If an instance constructor calls an instance method of the same class using the implicit self-reference ME->, the method is called as it is implemented in the class of the instance constructor, and not in any redefined form that may occur in the subclass you want to instantiate. This is an exception to the rule that states that when you call instance methods, the system always calls the method as it is implemented in the class to whose instance the reference is pointing.

Every class has a static constructor called CLASS_CONSTRUCTOR. As far as the namespace within an inheritance tree, the same applies to static constructors as to instance constructors.

The first time you address a subclass in a program, its static constructor is executed. However, before it can be executed, the static constructors of all of its superclasses must already have been executed. A static constructor may only be called once per program. Therefore, when you first address a subclass, the system looks for the next-highest superclass whose static constructor has not yet been executed. It executes the static constructor of that class, followed by those of all classes between that class and the subclass you addressed.