[NOTE This question was originally formulated under Swift 2.2. It has been revised for Swift 4, involving two important language changes: the first method parameter external is no longer automatically suppressed, and a selector must be explicitly exposed to Objective-C.]
Let's say I have these two methods in my class:
#objc func test() {}
#objc func test(_ sender:AnyObject?) {}
Now I want to use Swift 2.2's new #selector syntax to make a selector corresponding to the first of these methods, func test(). How do I do it? When I try this:
let selector = #selector(test) // error
... I get an error, "Ambiguous use of test()." But if I say this:
let selector = #selector(test(_:)) // ok, but...
... the error goes away, but I'm now referring to the wrong method, the one with a parameter. I want to refer to the one without any parameter. How do I do it?
[Note: the example is not artificial. NSObject has both Objective-C copy and copy: instance methods, Swift copy() and copy(sender:AnyObject?); so the problem can easily arise in real life.]
[NOTE This answer was originally formulated under Swift 2.2. It has been revised for Swift 4, involving two important language changes: the first method parameter external is no longer automatically suppressed, and a selector must be explicitly exposed to Objective-C.]
You can work around this problem by casting your function reference to the correct method signature:
let selector = #selector(test as () -> Void)
(However, in my opinion, you should not have to do this. I regard this situation as a bug, revealing that Swift's syntax for referring to functions is inadequate. I filed a bug report, but to no avail.)
Just to summarize the new #selector syntax:
The purpose of this syntax is to prevent the all-too-common runtime crashes (typically "unrecognized selector") that can arise when supplying a selector as a literal string. #selector() takes a function reference, and the compiler will check that the function really exists and will resolve the reference to an Objective-C selector for you. Thus, you can't readily make any mistake.
(EDIT: Okay, yes you can. You can be a complete lunkhead and set the target to an instance that doesn't implement the action message specified by the #selector. The compiler won't stop you and you'll crash just like in the good old days. Sigh...)
A function reference can appear in any of three forms:
The bare name of the function. This is sufficient if the function is unambiguous. Thus, for example:
#objc func test(_ sender:AnyObject?) {}
func makeSelector() {
let selector = #selector(test)
}
There is only one test method, so this #selector refers to it even though it takes a parameter and the #selector doesn't mention the parameter. The resolved Objective-C selector, behind the scenes, will still correctly be "test:" (with the colon, indicating a parameter).
The name of the function along with the rest of its signature. For example:
func test() {}
func test(_ sender:AnyObject?) {}
func makeSelector() {
let selector = #selector(test(_:))
}
We have two test methods, so we need to differentiate; the notation test(_:) resolves to the second one, the one with a parameter.
The name of the function with or without the rest of its signature, plus a cast to show the types of the parameters. Thus:
#objc func test(_ integer:Int) {}
#nonobjc func test(_ string:String) {}
func makeSelector() {
let selector1 = #selector(test as (Int) -> Void)
// or:
let selector2 = #selector(test(_:) as (Int) -> Void)
}
Here, we have overloaded test(_:). The overloading cannot be exposed to Objective-C, because Objective-C doesn't permit overloading, so only one of them is exposed, and we can form a selector only for the one that is exposed, because selectors are an Objective-C feature. But we must still disambiguate as far as Swift is concerned, and the cast does that.
(It is this linguistic feature that is used — misused, in my opinion — as the basis of the answer above.)
Also, you might have to help Swift resolve the function reference by telling it what class the function is in:
If the class is the same as this one, or up the superclass chain from this one, no further resolution is usually needed (as shown in the examples above); optionally, you can say self, with dot-notation (e.g. #selector(self.test), and in some situations you might have to do so.
Otherwise, you use either a reference to an instance for which the method is implemented, with dot-notation, as in this real-life example (self.mp is an MPMusicPlayerController):
let pause = UIBarButtonItem(barButtonSystemItem: .pause,
target: self.mp, action: #selector(self.mp.pause))
...or you can use the name of the class, with dot-notation:
class ClassA : NSObject {
#objc func test() {}
}
class ClassB {
func makeSelector() {
let selector = #selector(ClassA.test)
}
}
(This seems a curious notation, because it looks like you're saying test is a class method rather than an instance method, but it will be correctly resolved to a selector nonetheless, which is all that matters.)
I want to add a missing disambiguation: accessing an instance method from outside the class.
class Foo {
#objc func test() {}
#objc func test(_ sender: AnyObject?) {}
}
From the class' perspective the full signature of the test() method is (Foo) -> () -> Void, which you will need to specify in order to get the Selector.
#selector(Foo.test as (Foo) -> () -> Void)
#selector(Foo.test(_:))
Alternatively you can refer to an instance's Selectors as shown in the original answer.
let foo = Foo()
#selector(foo.test as () -> Void)
#selector(foo.test(_:))
In my case (Xcode 11.3.1) the error was only when using lldb while debugging. When running it works properly.
Related
protocol Proto {
associatedtype Entity: NSManagedObject
}
extension Proto {
func subscribe(completionBlock: #escaping (Entity) -> ()) {
var helper = Helper()
helper.subscribe(completionBlock) //error out
}
}
class Helper {
func subscribe(completionBlock: #escaping (NSManagedObject) -> ()){}
}
Even though entity will always be of type NSManagedObject, but it won't let me pass that closure into another function that accepts NSmanagedObject. How to fix?
I can make the code work by doing this:
helper.subscribe(completionBlock as! (NSManagedObject) -> ())
But why do I need to force cast when Entity is of type NSManagedObject? Also are there any risk of it crashing at runtime if I cast like this? Is there ever a scenario if the cast will fail?
Okay, so after some back and forth I believe the question amounts to this. Given this function:
func foo(v:UIView, f:(UIView)->Void) {
f(v)
}
...why is it illegal to pass a (UIButton)->Void into foo as its f parameter?
Well, let's imagine that we could do that. (This is called a reductio ad absurdum proof.) Then we could write this:
let bar : ((UIButton)->Void) = {button in print(button.currentTitle)}
foo(v:UISwitch(), f:bar)
And what would happen? foo would pass a UISwitch into f - legally, because f is typed as taking a UIView, and a UISwitch is indeed a UIView. And we would crash, because a UIView has no currentTitle property.
Do you see the problem? Characterising a function as a (UIButton)->Void gives that function the right, in the eyes of the compiler, to talk to the parameter as if it were a UIButton. But on the outside, f is typed as a function that takes a UIView. So it is legal to pass any UIView into it. Therefore the compiler needs to know that the function being held by f does not think that its parameter is a UIButton, lest it do that very sort of thing.
So the compiler very rightly stops you right at the door and doesn't let you pass bar into foo in the first place. The rule is that function types are contravariant on their parameter types, for passing.
Ready to use example to test in Playgrounds:
import Foundation
class Demo {
let selector = #selector(someFunc(_:))
func someFunc(_ word: String) {
// Do something
}
func someFunc(_ number: Int) {
// Do something
}
}
I'm getting an Ambiguous use of 'someFunc' error.
I have tried this answer and getting something like:
let selector = #selector(someFunc(_:) as (String) -> Void)
But still getting the same result.
Any hints on how to solve this?
Short answer: it can't be done.
Long explanation: when you want to use a #selector, the methods need to be exposed to Objective-C, so your code becomes:
#objc func someFunc(_ word: String) {
// Do something
}
#objc func someFunc(_ number: Int) {
// Do something
}
Thanks to the #objc annotation, they are now exposed to Objective-C and an compatible thunk is generated. Objective-C can use it to call the Swift methods.
In Objective-C we don't call a method directly but instead we try to send a message using objc_msgSend: the compiler is not able to understand that those method are different, since the generated signature is the same, so it won't compile. You will face the error:
Method 'someFunc' with Objective-C selector 'someFunc:' conflicts with previous declaration with the same Objective-C selector.
The only way to fix it is to have different signatures, changing one or both of them.
The selector is obviously ambiguous, neither methods have external parameter names. Remove the empty external parameters to solve this:
#objc func someFunc(word: String) {
// Do something
}
#objc func someFunc(number: Int) {
// Do something
}
After that, you should specify a selector with the parameter name:
#selector(someFunc(word:))
or
#selector(someFunc(number:))
Is there any way to indicate to a "client" of a specific method that a closure parameter is going to be retained?
For instance, having the following code:
import Foundation
typealias MyClosureType = () -> Void
final class MyClass {
private var myClosure: MyClosureType?
func whatever(closure: MyClosureType?) {
myClosure = closure
}
}
Anyone could start using this class and passing closures to the method whatever without any idea about if it is actually being retained or not. Which is error prone and could lead to memory leaks.
For instance, a "client" doing something like this, would be never deallocated
final class MyDummyClient {
let myInstance = MyClass()
func setUp() {
myInstance.whatever {
self.whateverHandler()
}
}
func whateverHandler() {
print("Hey Jude, don't make it bad")
}
}
That is why I would like to know if there is any way to prevent this type of errors. Some type of paramater that I could add to the definition of my method whatever which gives a hint to the client about the need to weakify to avoid leaks
Whether the closure parameter is escaping or non-escaping is some indication to the caller whether it might be retained. In particular, a non-escaping closure param cannot be retained by a function call.
Per SE-0103, non-escaping closures (currently marked #noescape) will become the default in Swift 3, and you'll have to write #escaping if you want to save the closure, so situations like this will become a little more obvious.
Otherwise, there is no language feature to help you here. You'll have to solve this problem with API design and documentation. If it's something like a handler, I'd recommend a property, obj.handler = { ... }, or a method like obj.setHandler({ ... }) or obj.addHandler({ ... }). That way, when reading the code, you can easily tell that the closure is being saved because of the = or set or add.
(In fact, when compiling Obj-C, Clang explicitly looks for methods named set...: or add...: when determining whether to warn the user about retain cycles. It's possible a similar diagnostic could be added to the Swift compiler in the future.)
With the specific case you're presenting the closure itself is the only thing that will be retained, so if you correctly add [weak self] to your closure when you call it there shouldn't be any issues.
I'm not sure what issue you're trying to protect against, but you might also thing about using delegation (protocols) rather than a closure.
I just started to learn Swift (v. 2.x) because I'm curious how the new features play out, especially the protocols with Self-requirements.
The following example is going to compile just fine, but causes arbitrary runtime effects to happen:
// The protocol with Self requirement
protocol Narcissistic {
func getFriend() -> Self
}
// Base class that adopts the protocol
class Mario : Narcissistic {
func getFriend() -> Self {
print("Mario.getFriend()")
return self;
}
}
// Intermediate class that eliminates the
// Self requirement by specifying an explicit type
// (Why does the compiler allow this?)
class SuperMario : Mario {
override func getFriend() -> SuperMario {
print("SuperMario.getFriend()")
return SuperMario();
}
}
// Most specific class that defines a field whose
// (polymorphic) access will cause the world to explode
class FireFlowerMario : SuperMario {
let fireballCount = 42
func throwFireballs() {
print("Throwing " + String(fireballCount) + " fireballs!")
}
}
// Global generic function restricted to the protocol
func queryFriend<T : Narcissistic>(narcissistic: T) -> T {
return narcissistic.getFriend()
}
// Sample client code
// Instantiate the most specific class
let m = FireFlowerMario()
// The call to the generic function is verified to return
// the same type that went in -- 'FireFlowerMario' in this case.
// But in reality, the method returns a 'SuperMario' and the
// call to 'throwFireballs' will cause arbitrary
// things to happen at runtime.
queryFriend(m).throwFireballs()
You can see the example in action on the IBM Swift Sandbox here.
In my browser, the output is as follows:
SuperMario.getFriend()
Throwing 32 fireballs!
(instead of 42! Or rather, 'instead of a runtime exception', as this method is not even defined on the object it is called on.)
Is this a proof that Swift is currently not type-safe?
EDIT #1:
Unpredictable behavior like this has to be unacceptable.
The true question is, what exact meaning the keyword Self (capital first letter) has.
I couldn't find anything online, but there are at least these two possibilities:
Self is simply a syntactic shortcut for the full class name it appears in, and it could be substituted with the latter without any change in meaning. But then, it cannot have the same meaning as when it appears inside a protocol definition.
Self is a sort of generic/associated type (in both protocols and classes) that gets re-instantiated in deriving/adopting classes. If that is the case, the compiler should have refused the override of getFriend in SuperMario.
Maybe the true definition is neither of those. Would be great if someone with more experience with the language could shed some light on the topic.
Yes, there seems to be a contradiction. The Self keyword, when used as a return type, apparently means 'self as an instance of Self'. For example, given this protocol
protocol ReturnsReceived {
/// Returns other.
func doReturn(other: Self) -> Self
}
we can't implement it as follows
class Return: ReturnsReceived {
func doReturn(other: Return) -> Self {
return other // Error
}
}
because we get a compiler error ("Cannot convert return expression of type 'Return' to return type 'Self'"), which disappears if we violate doReturn()'s contract and return self instead of other. And we can't write
class Return: ReturnsReceived {
func doReturn(other: Return) -> Return { // Error
return other
}
}
because this is only allowed in a final class, even if Swift supports covariant return types. (The following actually compiles.)
final class Return: ReturnsReceived {
func doReturn(other: Return) -> Return {
return other
}
}
On the other hand, as you pointed out, a subclass of Return can 'override' the Self requirement and merrily honor the contract of ReturnsReceived, as if Self were a simple placeholder for the conforming class' name.
class SubReturn: Return {
override func doReturn(other: Return) -> SubReturn {
// Of course this crashes if other is not a
// SubReturn instance, but let's ignore this
// problem for now.
return other as! SubReturn
}
}
I could be wrong, but I think that:
if Self as a return type really means 'self as an instance of
Self', the compiler should not accept this kind of Self requirement
overriding, because it makes it possible to return instances which
are not self; otherwise,
if Self as a return type must be simply a placeholder with no further implications, then in our example the compiler should already allow overriding the Self requirement in the Return class.
That said, and here any choice about the precise semantics of Self is not bound to change things, your code illustrates one of those cases where the compiler can easily be fooled, and the best it can do is generate code to defer checks to run-time. In this case, the checks that should be delegated to the runtime have to do with casting, and in my opinion one interesting aspect revealed by your examples is that at a particular spot Swift seems not to delegate anything, hence the inevitable crash is more dramatic than it ought to be.
Swift is able to check casts at run-time. Let's consider the following code.
let sm = SuperMario()
let ffm = sm as! FireFlowerMario
ffm.throwFireballs()
Here we create a SuperMario and downcast it to FireFlowerMario. These two classes are not unrelated, and we are assuring the compiler (as!) that we know what we are doing, so the compiler leaves it as it is and compiles the second and third lines without a hitch. However, the program fails at run-time, complaining that it
Could not cast value of type
'SomeModule.SuperMario' (0x...) to
'SomeModule.FireFlowerMario' (0x...).
when trying the cast in the second line. This is not wrong or surprising behaviour. Java, for example, would do exactly the same: compile the code, and fail at run-time with a ClassCastException. The important thing is that the application reliably crashes at run-time.
Your code is a more elaborate way to fool the compiler, but it boils down to the same problem: there is a SuperMario instead of a FireFlowerMario. The difference is that in your case we don't get a gentle "could not cast" message but, in a real Xcode project, an abrupt and terrific error when calling throwFireballs().
In the same situation, Java fails (at run-time) with the same error we saw above (a ClassCastException), which means it attempts a cast (to FireFlowerMario) before calling throwFireballs() on the object returned by queryFriend(). The presence of an explicit checkcast instruction in the bytecode easily confirms this.
Swift on the contrary, as far as I can see at the moment, does not try any cast before the call (no casting routine is called in the compiled code), so a horrible, uncaught error is the only possible outcome. If, instead, your code produced a run-time "could not cast" error message, or something as gracious as that, I would be completely satisfied with the behaviour of the language.
The issue here seems to be a violation in contract:
You define getFriend() to return an instance of receiver (Self). The problem here is that SuperMario does not return self but it returns a new instance of type SuperMario.
Now, when FireFlowerMario inherits that method the contract says that the method should return a FireFlowerMario but instead, the inherited method returns a SuperMario! This instance is then treated as if it were a FireFlowerMario, specifically: Swift tries to access the instance variable fireballCount which does not exist on SuperMario and you get garbage instead.
You can fix it like this:
class SuperMario : Mario {
required override init() {
super.init()
}
override func getFriend() -> SuperMario {
print("SuperMario.getFriend()")
// Dynamically create new instance of the same type as the receiver.
let myClass = self.dynamicType
return myClass.init()
}
}
Why does the compiler allow it? It has a hard time catching something like this, I guess. For SuperMario, the contract is still valid: the method getFriend does return an instance of the same class. The contract breaks when you create the subclass FireFlowerMario: should the compiler notice that a superclass might violate the contract? This would be an expensive check and probably more suited for a static analyzer, IMHO (Also, what happens if the compiler doesn't have access to SuperMario's source? What happens if that class is from a library?)
So it's actually SuperMario's duty to ensure that the contract is still valid when subclassing.
I have a method that I have implemented, but am having trouble getting the method signature to be elegant.
The method returns all classes which are a subclass of a specified class. My current method signature looks like :
public func allSubclassesOf(baseClass: AnyClass) -> [AnyClass]
which is fine, apart from the return type being AnyClass which means I always end up with a messy cast like this:
allSubClassesOf(UIView).forEach { (subclass:AnyClass) in
UIView *sigh = subclass as! UIView //!< Gross and unnecessary
...
}
This seems like something that generics should be able to solve :)
Things I've tried :
public func allSubclassesOf<T>() -> [T]
Nope, you're not allowed to add generics to a function like that.
extension NSObject {
class func allSubclasses() -> [Self]
}
Nope, Self isn't available here.
Does anyone know how I can pass a type into this method and have the compiler know what type the returning array would hold; removing the need for a cast?
I am not sure of the implementation of your method but you would need to do something like the following.
What I am doing here is applying a generic type to the method function then saying that I would expect the object type as an argument and return instances of the type in an array.
You could add it as an extension, however without more examples for your code I can't help any further
func subclasses<T>(type: T.Type) -> [T] {
....
}
subclasses(UIView).forEach { (view: UIView) in
print(view)
}