structs: Bundling Data That Belongs Together

The problem structs solve

A student is not three unrelated variables floating around — it is one thing:

struct Student {
    std::string name;
    int age{};
    double gpa{};
};

Now Student is a type, exactly like int:

Student s{"Linh", 20, 3.8};                    // aggregate initialization
Student t{.name = "Minh", .age = 21, .gpa = 3.5};   // C++20 designated initializers
s.name = "Linh Nguyen";                        // member access with .
Student older = older_of(s, t);                // passes and returns by value like any type

The {} member initializers (int age{}) give every new Student sane defaults — no garbage members, ever.

Structs are values

Assignment copies every member; passing by value copies; == does NOT work automatically (compare members yourself, or write operator== — module 10's neighbor topic). The value semantics you learned in module 2 apply to your own types now.

A taste of member functions

struct Rectangle {
    double width{};
    double height{};

    double area() const { return width * height; }   // const: reads, never writes
};
r.area()

A function inside the struct operates on the members of whichever object it is called on. The const suffix promises it does not modify the object — the same const-correctness as module 2, now per-object. Member functions are the door to module 10.

Design hint: model the domain, not the storage

struct Account { std::string owner; long long cents; }; beats struct AccountData2 { ... }; — name things what they ARE. And when two values always travel together (module 7's hint), that pair deserves a struct with a real name, not std::pair.