Classes, their instances (objects), and access to objects using reference variables form the basics of ABAP Objects. These means already allow you to model typical business applications, such as customers, orders, order items, invoices, and so on, using objects, and to implement solutions using ABAP Objects.
However, it is often necessary for similar classes to provide similar functions that are coded differently in each class but which should provide a uniform point of contact for the user. For example, you might have two similar classes, savings account and check account, both of which have a method for calculating end of year charges. The interfaces and names of the methods are the same, but the actual implementation is different. The user of the classes and their instances must also be able to run the end of year method for all accounts, without having to worry about the actual type of each individual account.
ABAP Objects makes this possible by using interfaces. Interfaces are independent structures that you can implement in a class to extend the scope of that class. The class-specific scope of a class is defined by its components and visibility sections. For example, the public components of a class define its public scope, since all of its attributes and method parameters can be addressed by all users. The protected components of a class define its scope with regard to its subclasses. (However, inheritance is not supported in Release 4.5B).
Interfaces extend the scope of a class by adding their own components to its public section. This allows users to address different classes via a universal point of contact. Interfaces, along with inheritance, provide one of the pillars of polymorphism, since they allow a single method within an interface to behave differently in different classes.
Like classes, you can define interfaces either globally in the R/3 Repository or locally in an ABAP program. For information about how to define local interfaces, refer to theClass Builder section of the ABAP Workbench Tools documentation. The definition of a local interface <intf> is enclosed in the statements:
The definition contains the declaration for all components (attributes, methods, events) of the interface. You can define the same components in an interface as in a class. The components of interfaces do not have to be assigned individually to a visibility section, since they automatically belong to the public section of the class in which the interface is implemented. Interfaces do not have an implementation part, since their methods are implemented in the class that implements the interface.
Unlike classes, interfaces do not have instances. Instead, interfaces are implemented by classes. To implement an interface in a class, use the statement
in the declaration part of the class. This statement may only appear in the public section of the class.
When you implement an interface in a class, the components of the interface are added to the other components in the public section. A component <icomp> of an interface <intf> can be addressed as though it were a member of the class under the name <intf~icomp>.
The class must implement the methods of all interfaces implemented in it. The implementation part of the class must contain a method implementation for each interface method <imeth>:
Interfaces can be implemented by different classes. Each of these classes is extended by the same set of components. However, the methods of the interface can be implemented differently in each class.
Interfaces allow you to use different classes in a uniform way using interface references (polymorphism). For example, interfaces that are implemented in different classes extend the public scope of each class by the same set of components. If a class does not have any class-specific public components, the interfaces define the entire public face of the class.
Reference variables allow you to access objects (refer toWorking with Objects). Instead of creating reference variables with reference to a class, you can also define them with reference to an interface. This kind of reference variable can contain references to objects of classes that implement the corresponding interface.
To define an interface reference, use the addition TYPE REF TO <intf> in the TYPES or DATA statement. <intf> must be an interface that has been declared to the program before the actual reference declaration occurs. A reference variable with the type interface reference is called a interface reference variable, or interface reference for short.
An interface reference <iref> allows a user to use the form <iref>-><icomp> to address all visible interface components <icomp> of the object to which the object reference is pointing. It allows the user to access all of the components of the object that were added to its definition by the implementation of the interface.
Addressing Objects Using Interface References
To create an object of the class <class>, you must first have declared a reference variable <cref> with reference to the class. If the class <class> implements an interface <intf>, you can use the following assignment between the class reference variable <cref> and an interface reference <iref> to make the interface reference in <iref> point to the same object as the class reference in <cref>:
<iref> = <cref>
If the interface <intf> contains an instance attribute <attr> and an instance method <meth>, you can address the interface components as follows:
Using the class reference variable <cref>:
Using the interface reference variable <iref>:
As far as the static components of interfaces are concerned, you can only use the interface name to access constants:
Addressing a constant <const>: <intf>=><const>
For all other static components of an interface, you can only use object references or the class <class> that implements the interface:
Addressing a static attribute <attr>: <class>=><intf~attr>
Calling a static method <meth>:CALL METHOD <class>=><intf~meth>
Assignment Using Interface References - Casting
Like class references, you can assign interface references to different reference variables. You can also make assignments between class reference variables and interface reference variables. When you use the MOVE statement or the assignment operator (=) to assign reference variables, the system must be able to recognize in the syntax check whether an assignment is possible.
Suppose we have a class reference <cref> and interface references <iref>, <iref1>, and <iref2>. The following assignments with interface references can be checked statically:
Both interface references must refer to the same interface, or the interface of <iref1> must contain the interface <iref2> as a component.
The class of the class reference <cref> must implement the interface of the interface reference <iref>.
The class of <cref> must be the predefined empty class OBJECT.
In all other cases, you would have to work with the statement MOVE ...? TO or the casting operator (?=). The casting operator replaces the assignment operator (=). In the MOVE... ? TO statement, or when you use the casting operator, there is no static type check. Instead, the system checks at runtime whether the object reference in the source variable points to an object to which the object reference in the target variable can also point. If the assignment is possible, the system makes it, otherwise, the catchable runtime error MOVE_CAST_ERROR occurs.
You must always use casting for assigning an interface reference to a class reference if <cref> does not refer to the predefined empty class OBJECT:
<cref> ?= <iref>
For the casting to be successful, the object to which <iref> points must be an object of the same class as the type of the class variable <cref>.
See also:Overview Graphics