Go stores interfaces as a pair of values: a type, and the actual value, and will only be equal to nil if both are.

As a consequence, a nil value stored in an interface variable, but as a pointer to another type, will not pass the == nil test. To illustrate:

type myInterface interface {
    sayHi()
}

type myName string

func (n myName) sayHi() {
    fmt.Println("Hi there! My name is " + n)
}

func main() {
    test1 := myName("Camilo")
    run("test 1", test1)

    var test2 myInterface
    run("test 2", test2)

    var test3 *myName
    run("test 3", test3)
}

func run(label string, a myInterface) {
    fmt.Println("======== " + label + " ========")
    fmt.Println("value:", reflect.ValueOf(a))
    fmt.Println("type: ", reflect.TypeOf(a))
    if a == nil {
        fmt.Println("I'm nil")
    } else if reflect.ValueOf(a).Kind() == reflect.Ptr && reflect.ValueOf(a).IsNil() {
        fmt.Println("I'm nil (weird case!)")
    } else {
        a.sayHi()
    }
}
Previous on Go