std::pair vs Custom Struct

The choice between using std::pair and defining your own struct or class depends on the specific requirements and semantics of your code. Here are some guidelines:

Use std::pair when:

• You need to group two values together temporarily or pass them as a single unit.
• The two values do not have a strong logical connection or a specific meaning.
• You don't need to define any custom behavior or member functions for the grouped values.

Use a custom struct or class when:

• The grouped values have a strong logical connection and represent a meaningful entity in your program.
• You need to define custom behavior, member functions, or operators for the grouped values.
• You want to provide a more expressive and self-explanatory name for the type.

Example of using std::pair:

1#include <iostream>
2#include <utility>
3
4std::pair<int, int> getMinMax(int a, int b) {
5  return (a < b) ?
6    std::make_pair(a, b) : std::make_pair(b, a);  
7}
8
9int main() {
10  auto [min, max] = getMinMax(5, 3);
11  std::cout << "Min: " << min
12    << ", Max: " << max;
13}

1Min: 3, Max: 5

In this example, std::pair is used to group and return the minimum and maximum values from the getMinMax function. The pair doesn't have a specific meaning beyond holding the min and max values temporarily.

Example of using a custom struct:

1#include <iostream>
2
3struct Point {
4  int x;
5  int y;
6
7  void print() const {
8    std::cout << "(" << x << ", " << y << ")";
9  }
10};
11
12int main() {
13  Point p{3, 5};
14  p.print();
15}

1(3, 5)

In this case, we define a custom Point struct to represent a 2D point. The x and y values have a strong logical connection, and we provide a custom print member function to display the point in a specific format.

Consider the semantics and requirements of your code when deciding between std::pair and a custom struct or class.