TL;DR: Long chains of collaborations generate coupling and ripple effect, where any change in the chain breaks the code.
Problems
Coupling
Break encapsulation
Solutions
Create intermediate methods.
Consider the Law of Demeter.
Create higher-level messages.
Sample Code
Wrong
class Dog(val feet: Array<Foot>) {
}
class Foot {
fun move(): Unit = TODO()
}
fun main() {
val feet = arrayOf(Foot(), Foot(), Foot(), Foot())
val dog = Dog(feet)
for(foot in dog.feet) {
foot.move()
}
}
Right
// We're copying the reference of the array, so in theory, it can still
// be changed from outside, which is yet another code smell. A better
// approach would be to create a copy of the array inside the class
// instead of holding the reference.
class Dog(private val feet: Array<Foot>) {
fun walk() {
for(foot in feet) {
foot.move()
}
}
}
class Foot {
fun move(): Unit = TODO()
}
fun main() {
val feet = arrayOf(Foot(), Foot(), Foot(), Foot())
val dog = Dog(feet)
dog.walk()
}
Conclusion
Avoid successive message calls. Try to hide the intermediate collaborations and create new protocols. This way, not only will you protect your code from breaking in the future, but you will also maintain good encapsulation of your class.
Stay updated with my latest thoughts and ideas by registering for my newsletter. Connect with me on LinkedIn or Twitter. Let's stay connected and keep the conversation going!
Top comments (0)