Modular Programming

Modules in modular programming enforce logical boundaries between components and improve maintainability. They are incorporated through interfaces. They are designed in such a way as to minimize dependencies between different modules. Teams can develop modules separately and do not require knowledge of all modules in the system.

Each and every modular application has a version number associated with it. This provides developers flexibility in module maintenance. If any changes have to be applied to a module, only the affected subroutines have to be changed. This makes the program easier to read and understand.

Modular programming has a main module and many auxiliary modules. The main module is compiled as an executable (EXE), which calls the auxiliary module functions. Auxiliary modules exist as separate executable files, which load when the main EXE runs. Each module has a unique name assigned in the PROGRAM statement. Function names across modules should be unique for easy access if functions used by the main module must be exported.

Languages that support the module concept are IBM Assembler, COBOL, RPG, FORTRAN, Morpho, Zonnon and Erlang, among others.

The benefits of using modular programming include:

• Less code has to be written.
• A single procedure can be developed for reuse, eliminating the need to retype the code many times.
• Programs can be designed more easily because a small team deals with only a small part of the entire code.
• Modular programming allows many programmers to collaborate on the same application.
• The code is stored across multiple files.
• Code is short, simple and easy to understand.
• Errors can easily be identified, as they are localized to a subroutine or function.
• The same code can be used in many applications.
• The scoping of variables can easily be controlled.