Overview

Intensive dealing with pointers

Now that we have introduced static data structures as well as dynamic data structures (also called pointers; see Section 3.7), this chapter deals more intensively with that subject. Dynamic data structures are important, because the only way to create objects in object-oriented programming is to create them dynamically.

This chapter begins with a discussion of data abstraction, which will help you understand the concept of an object-oriented class easily, then discusses algorithms that use dynamic data structures. Although such algorithms can also be used in COBOL, experience has shown that they rarely are, especially by programmers who made their way into the COBOL world through a crash course from a different discipline, where they did not get a chance to work with any programming language intensively. This chapter is particularly important for those who feel addressed by this statement.

4.1 Abstract Data Types

The tutorial shows a few preparatory steps for how programmers can create their own data types. Our discussion concentrates less on syntactic details (which have been dealt with in Section 3.4.1) than on how this operation works efficiently from the contents point of view. We know that data types are normally made available to various other programs. Therefore, the client needs to know only the essential things about the data type. In most cases, no detailed knowledge is required. Behind this concept is the term abstraction. The abstract data structure and abstract data type build on this abstraction.