An interface in Kotlin can have default implementations for functions:
interface MyInterface {
fun withImplementation() {
print("withImplementation() was called")
}
}
Classes implementing such interfaces will be able to use those functions without reimplementing
class MyClass: MyInterface {
// No need to reimplement here
}
val instance = MyClass()
instance.withImplementation()
Default implementations also work for property getters and setters:
interface MyInterface2 {
val helloWorld
get() = "Hello World!"
}
Interface accessors implementations can't use backing fields
interface MyInterface3 {
// this property won't compile!
var helloWorld: Int
get() = field
set(value) { field = value }
}
When multiple interfaces implement the same function, or all of them define with one or more implementing, the derived class needs to manually resolve proper call
interface A {
fun notImplemented()
fun implementedOnlyInA() { print("only A") }
fun implementedInBoth() { print("both, A") }
fun implementedInOne() { print("implemented in A") }
}
interface B {
fun implementedInBoth() { print("both, B") }
fun implementedInOne() // only defined
}
class MyClass: A, B {
override fun notImplemented() { print("Normal implementation") }
// implementedOnlyInA() can by normally used in instances
// class needs to define how to use interface functions
override fun implementedInBoth() {
super<B>.implementedInBoth()
super<A>.implementedInBoth()
}
// even if there's only one implementation, there multiple definitions
override fun implementedInOne() {
super<A>.implementedInOne()
print("implementedInOne class implementation")
}
}