A pointer type can also have methods. These are defined like normal methods, but taking care that the receiver (go-types-and-methods) is a pointer and thus needs to be dereferenced.

E.g.:

1type MyInt int
2
3func (input *MyInt) Double() {
4  *input *= 2
5}

Methods with pointer receiver can modify the value to which the receiver points.

N.B.: while functions with a pointer argument must take a pointer, methods with pointer receivers take either a value or a pointer as the receiver when they are called. When they are passed the value, the method with the pointer receiver is called automatically.

1x = MyInt(3)
2// you can do this:
3(&x).Double()
4// you can also do this:
5x.Double()

In the example above, when you pass x by value, Go will interpret the statement as (&x).Double() as a convenience.

Moreover, Go supports automatic de-referencing. We know that when a function (or a method) wants to do something with a pointer, it needs to dereference it first, using the * operator, to get the value that it points to.

E.g.:

1func (b *Book) SetPriceCents(price int) error {
2    (*b).PriceCents = price
3    return nil
4}

where Book is a struct with PriceCents field.

In structs, Go knows that pointers don’t have fields, only structs do. So if you write something like b.PriceCents that looks like a struct field, Go is smart enough to assume you meant “dereference b to get the struct it points to, then modify its PriceCents field”.

1func (b *Book) SetPriceCents(price int) error {
2    b.PriceCents = price
3    return nil
4}

This is called automatic dereferencing, and since most methods on structs are pointer methods, it’s very helpful.

References

• John Arundel, For the Love of Go
• https://go.dev/tour/methods/6

Next -> choosing-a-value-or-pointer-receiver-in-go