When should we encapsulate codes to be a "Class"? [closed]

Question

Closed. This question is opinion-based. It is not currently accepting answers.

---

Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.

Closed 8 years ago.

Improve this question

I looked back on my code so far and found that I rarely wrote class. Or, in other words, I don't see the value of using them. I mean, almost everything in coding, I can work them out without using classes by just writing functions and group them into different .h/.cpp files based on their functionalities.

Even after I've participated in several projects, I still do not feel the urgency to do this as progress goes well without big challenges. I am working mostly with C++ (MATLAB sometimes).

Note that, I did use struct a lot and I know that struct and class have no big difference for C++. But I only use struct for data-storage (group data together). The data requires no flow control (i.e. getters/setters that maintain or protect an internal state) nor starts acquiring any major functionality.

Maybe I am missing somewhere about basic things about OOP. So when should we encapsulate our codes using classes? Can anyone share your opinions?

Answer 1

In C++, a struct and a class are just the same except with a different default access modifier. struct members are by default public, whereas class members are by default private. If we think of classes in C++ as being a class with encapsulated data members and public member functions that expose functionality related to that data, the following are some good suggestions that you want to use a class:

1. You need to store some state consisting of related items of data, and...
2. You need to impose some invariants on this state or provide some alternative interface to this state.

If you have only suggestion 1, then you probably just want a struct. See Classes Should Enforce Invariants.

Answer 2

You can write classless code in most languages. In Java you would merely use static methods and nothing else.

Using C/C++ structs binds groups of values together into a unit to form a single component that can be created and manipulated separately.

The moment you discover you need classes and instances of classes (generally objects) is the moment you grow up in your coding. With a class you bind a number of fields together (as with structs) but here you attach methods to them too. You are now working with objects that have structure and function.

Once that step is complete you discover a whole new world of programming where you put these objects inside other objects or into queues for transport or in maps and sets for storage and then you discover that a map and a set and a queue is in itself an object with its own class.

In summary - working with just process is like building your house with sand. Working with structs is building your house with sticks. Working with classes allows you to instantiate objects and manipulate them as entities - build your house with stone - you can then step away from the process you have been so far toiling with and begin to consider the architecture of your system.