Swift 2.0 introduced .init, which offers functional access to initialization. You can construct an array of strings from integers by mapping String.init:
[1, 2, 3, 4, 5, 6].map(String.init)
Contrast with the old construction, which isn’t nearly as nice:
[1, 2, 3, 4, 5, 6].map{String($0)}
You can also populate enumerations with associated values. This example uses the Container enumeration I wrote about a few weeks ago.
What you don’t want to do is use this as a long-hand for standard initialization. I think it’s a good rule to prefer String(5) over String.init(5).
Apple writes, “.init is still implicit when constructing using a static type, as in String(5). .init is required when using dynamic type objects, or when referring to the initializer as a function value.”
Update 1:
Trying to decide whether this is worth a radar:
Update 2:
I oopsed on the initial post, using ()’s instead of {}’s for the 2nd map. But why does the first one work? (p.s. You want to debugPrint to see the quotation marks to confirm as String, not print)
5 Comments
> But why does the first one work?
In first case: because .init you can use as closure, and you set it in map as callback.
In second: I’m sure work case like .map { String.init($0) }
I agree the last line is equivalent to String.init($0). Didn’t know you could use init as a closure. Thanks.
to be clear, I’m providing example:
class A { var x: Int!, y: Int! init(completion: A -> Void) { completion(self) } } let a = A { $0.x = 5 $0.y = 25 }Where in the Apple docs did you find that “Swift 2.0 introduced .init, which offers functional access to initialization.”?
Release notes