Swift vs ObjC initialisation process?

Question

In ObjectiveC we create objects like

 -(instancetype)init()
{
  return [super init]; // Here it returns initialised value
}  

Class *obj =   [[Class alloc]init] 

But swift initialiser wont return any value.

From Swift docs

Unlike Objective-C initializers, Swift initializers do not return a value. Their primary role is to ensure that new instances of a type are correctly initialized before they are used for the first time.

init()
{
  super.init()
}  

let obj = Class()

Now how swift initialiser returns the instance to variable obj?.

How the allocation and initialisation occurs in swift?

Answer 1

As @NikolayKasyanov says, with the init family of initialisers, the return (of self) is implicit, and you can't return nil. However, if you want to initialise an optional that could return nil, use a class function. EG:

class NumberLessThan5: Int {

    var myNumber: Int
    init (i: Int) {
        self.myNumber = i
    }
    class func createWithInt(i: Int) -> NumberLessThan5? {
        if i < 5 {
            return NumberLessThan5(i)
        } else {
            return nil
        }    
    }
}

Answer 2

It's just a convention. Swift initialiser sets up a valid instance and could not theoretically return anything other that a valid instance, so there's no point in explicit return.

So (from my point of view) allocation & initialisation sequence looks like this:

1. Runtime allocates instance of requested class
2. Initializer is called with self set to allocated instance
3. Initializer performs setup
4. Runtime returns initialised instance to client code

Although this approach breaks some useful Objective-C patterns like initialisers returning nil on error, the guarantee that instantiation always succeeds allows compiler to perform some optimisations. Also without dropping initialisers returning nil it would be impossible to actually remove nil from language, it would seem weird if initialisers were returning optionals.