What is the difference between a value type and a reference type in Go?

In Go, data types can generally be categorized into two types: Value Types and Reference Types. Understanding the distinction between these types is essential for efficiently utilizing Go.

Value Types

Value Types encompass basic data types such as int, float, bool, and string, as well as composite structures built from them, like array and struct. A key characteristic of Value Types is that they always perform a value copy when assigned or passed as parameters.

Example:

Consider the following code:

go
type Point struct {
    X int
    Y int
}

func main() {
    p1 := Point{1, 2}
    p2 := p1 // Assign the value of p1 to p2
    p2.X = 3

    fmt.Println(p1.X) // Output: 1, as p2 is a copy of p1 and modifications to p2 do not affect p1
}

In this example, p2 is a copy of p1, and changes to p2 do not alter the value of p1.

Reference Types

Reference Types include slice, map, channel, interface, and pointers. When assigned or passed as parameters, these types do not copy the value itself but instead copy references or pointers to the underlying data structures.

Example:

Consider the following code:

go
func main() {
    slice1 := []int{1, 2, 3}
    slice2 := slice1 // Assign the reference to slice1
    slice2[0] = 9

    fmt.Println(slice1[0]) // Output: 9, as slice1 and slice2 share the same underlying array
}

In this example, slice2 is a copy of the reference to slice1, not a deep copy of the data. Thus, modifications to slice2 also affect slice1.

Summary

Understanding the difference between Value Types and Reference Types primarily lies in comprehending how data flows through a program. Value Types are ideal for scenarios requiring full data copies, while Reference Types are suited for situations where multiple functions or program sections need to share or modify the same data. This understanding facilitates writing more efficient, readable, and maintainable Go code.