Why is a UIAlertAction's handler required to have a parameter that references the UIAlertAction?

Question

I'm working with a simple button that, when pressed, runs a function I created.

Originally, I wrote the code like this:

func askQuestion() {
     // runs code to ask a question
}

@IBAction func buttonTapped(_ sender: UIButton) {
     let ac = UIAlertController(title: title, message: "You tapped the button.", preferredStyle: .alert)
     ac.addAction(UIAlertAction(title: "Continue", style: .default, handler: askQuestion))
    present(ac, animated: true)

But that returns an error:

Cannot convert value of type '() -> ()' to expected argument type '((UIAlertAction) -> Void)?'

Which is fixed when you add the following parameter to askQuestion():

askQuestion(action: UIAlertAction! = nil)

Why does a handler method passed into the UIAlertAction require that it accept a UIAlertAction parameter? It seems like an unnecessary step, but I was thinking it might be a way to scan your code for hints that this function is triggered by a button.

Answer 1

You could have a single handler responsible for handling multiple actions

var continueAction: UIAlertAction!
var cancelAction: UIAlertAction!

override func viewDidLoad() {
    super.viewDidLoad()

    continueAction = UIAlertAction(title: "Continue", style: .default, handler: masterHandler)
    cancelAction = UIAlertAction(title: "Cancel", style: .default, handler: masterHandler)
}

//...

 let ac = UIAlertController(title: title, message: "You tapped the button.", preferredStyle: .alert)
 ac.addAction(continueAction)
 ac.addAction(cancelAction)

Personally, I don't know why you might do this, but the API designers felt it was a good idea to provide you with the flexibility to design your solution which best meet your needs.

So, in most cases, while I appreciate it seems weird (especially when you can use closures), I do appreciate having the information available to make my own choices