Why is this property's didSet being called - swift

protocol FooType {
var num:Int { get set }
}
class Foo: FooType {
var num: Int = 0 {
didSet {
print("Did set num")
}
}
}
class Bar {
var foo: FooType = Foo() {
didSet {
print("Did set Foo")
}
}
func changeNum(num:Int) {
foo.num = num
}
}
let bar = Bar()
bar.changeNum(5)
Did set num
Did set Foo
In this example, setting a property on foo causes Bar's didSet for foo to get called.
I would expect only Foo's num didSet to get called.
If I remove the protocol constraint of Bar's foo property to FooType, it behaves as I expected.
protocol FooType {
var num:Int { get set }
}
class Foo: FooType {
var num: Int = 0 {
didSet {
print("Did set num")
}
}
}
class Bar {
var foo = Foo() {
didSet {
print("Did set Foo")
}
}
func changeNum(num:Int) {
foo.num = num
}
}
let bar = Bar()
bar.changeNum(5)
Did set num
If I keep the conformance to FooType, but add a class constraint (FooType: class), it also behaves as expected.
protocol FooType: class {
var num:Int { get set }
}
class Foo: FooType {
var num: Int = 0 {
didSet {
print("Did set num")
}
}
}
class Bar {
var foo: FooType = Foo() {
didSet {
print("Did set Foo")
}
}
func changeNum(num:Int) {
foo.num = num
}
}
let bar = Bar()
bar.changeNum(5)
Did set num
If I remove the protocol completely and make Foo a struct rather than a class, we're back to both setters being called.
struct Foo {
var num: Int = 0 {
didSet {
print("Did set num")
}
}
}
class Bar {
var foo = Foo() {
didSet {
print("Did set Foo")
}
}
func changeNum(num:Int) {
foo.num = num
}
}
let bar = Bar()
bar.changeNum(5)
Did set num
Did set Foo
In the case changing Foo to a struct, I can see why it's happening... it mentions it in the swift documentation because a struct is a value type, it makes a copy, etc. (although I didn't expect it at first).
But the other cases I don't get...

you can solve this by conformance FooType to AnyObject
protocol FooType: AnyObject {
var num:Int { get set }
}

Related

How can I get a specific struct type returned?

I want to initialize the struct by receiving the "struct type"(?) by a specific logic as below.
As I abstracted the return value of the struct into MyProtocol had a declaration of init(), which seems a little awkward.
I'm not sure I can do this.
I'd like to get an undecided struct type returned, what should I do?
Is this for the best?
For your information, Opaque Type is not available because it needs to support iOS 13 or earlier.
protocol MyProtocol {
init() // Is this for the best?
}
struct AAA: MyProtocol {
var numberAAA: Int // Sample variable.
init() {
print("init AAA")
numberAAA = 100
}
}
struct BBB: MyProtocol {
var numberBBB: Int // Sample variable.
init() {
print("init BBB")
numberBBB = 200
}
}
class MyClass {
func mainLogic() {
let myStruct = randomStruct()
myStruct.init() // This is the reason init () declared in the MyProtocol.
}
func randomStruct() -> MyProtocol.Type {
if Bool.random() {
return AAA.self
} else {
return BBB.self
}
}
}
init() as a protocol requirement seems odd. No one is stopping you to do this and compiler should allow this, however I would consider making protocol based on some other requirement rather than just init().
Here's an attempt to do so -
protocol NumberOperation {
var number: Int { get set }
mutating func perform()
}
struct Incrementer: NumberOperation {
var number: Int
mutating func perform() {
number += 1
}
}
struct Decrementer: NumberOperation {
var number: Int
mutating func perform() {
number -= 1
}
}
struct Record<O: NumberOperation> {
public var operation: O
mutating func perform() {
operation.perform()
}
}
class MyClass {
func mainLogic() {
var record = getRecord(type: Incrementer.self)
record.perform()
}
func getRecord<O: NumberOperation>(type: O.Type) -> Record<O> {
if type == Incrementer.self {
return Record(operation: Incrementer(number: 1) as! O)
}
return Record(operation: Decrementer(number: 10) as! O)
}
}
This introduces a container type Record that holds/wraps our type based on the same protocol conformation. This does the same what you were doing, probably is easier to read/understand.

Swift: Default value for properties in Protocol

I am trying to provide a default value for a variable in a Protocol. I am getting an error:
Type ViewController does not conform to protocol Test
Code:
protocol Test {
var aValue: CGFloat { get set }
}
extension Test {
var aValue: CGFloat {
return 0.3
}
}
class ViewController: UIViewController, Test {
override func viewDidLoad() {
super.viewDidLoad()
print("value \(aValue)")
}
}
How can I provide a default value so the ViewController can use the default value (in the protocol extension) without to declaring it?
protocol Test {
var aValue: CGFloat { get set }
}
extension Test {
var aValue: CGFloat {
get {
return 0.3
}
set {
print("the new value is \(newValue)")
}
}
}
class Default: Test {
init() {
print("value \(aValue)")
}
}
class ViewController: Test {
var aValue: CGFloat {
get {
return 0.4
}
set {
print("i am overriding the setter")
}
}
init() {
print("value \(aValue)")
}
}
var d = Default() // value 0.3
d.aValue = 1 // the new value is 1.0
var vc = ViewController() // value 0.4
vc.aValue = 1 // i am overriding the setter
Since you have a protocol extension, you don't have to implement neither the getter nor the setter if you don't want to.
https://docs.swift.org/swift-book/LanguageGuide/Properties.html#//apple_ref/doc/uid/TP40014097-CH14-ID259
In addition to stored properties, classes, structures, and
enumerations can define computed properties, which do not actually
store a value. Instead, they provide a getter and an optional setter
to retrieve and set other properties and values indirectly.
You can't set the value of the same variable in the setter itself.
protocol Test {
var aValue: CGFloat {get set}
}
extension Test {
// var aValue: CGFloat {
// return 0.3
// }
var aValue: CGFloat {
get {
return 0.3
}
set {
debugPrint("new value is \(aValue)")
}
}
}
struct TestClass: Test {
func printData() {
debugPrint(aValue)
}
}
let aTestClass = TestClass()
aTestClass.printData()

Generic class with generic delegate. How can I distinguish which object's delegate is executing the callback?

MyClass is a generic class which has a generic delegate.
AClass contains two iVar instances of MyClass and also implements MyClassDelegate.
Where AClass implements MyClassDelegate, how can I distinguish which object is calling into the interface? With non-generic classes, == is acceptable.
Please see the comments and error messages at the bottom of this code snippet.
protocol MyClassDelegate: class {
func myClass<T>(_ myClass: MyClass<T>, valueDidChange value: T)
}
class MyClass<T: Comparable> {
private var _value: T
var value: T {
set {
delegate?.myClass(self, valueDidChange: newValue)
}
get {
return _value
}
}
var delegate: MyClassDelegate?
init(value: T) {
_value = value
}
}
class AClass {
private var thing1 = MyClass(value: Int(10))
private var thing2 = MyClass(value: Int(100))
private var thing3 = MyClass(value: TimeInterval(10))
private var thing4 = MyClass(value: TimeInterval(100))
init() {
thing1.delegate = self
thing2.delegate = self
thing3.delegate = self
thing4.delegate = self
}
}
extension AClass: MyClassDelegate {
func myClass<T>(_ myClass: MyClass<T>, valueDidChange value: T) {
// This fails to complile
// Binary operator '==' cannot be applied to operands of type 'MyClass<T>' and 'MyClass<Int>'
if myClass == thing1 {
}
// Binary operator '==' cannot be applied to operands of type 'MyClass<T>' and 'MyClass<TimeInterval>' (aka 'MyClass<Double>')
else if myClass == thing3 {
}
}
}
Restrict the T type in your protocol method signature func myClass<T>(...) to be Comparable. Since you restricted it only in one particular extension of that protocol, the method in AClass can take any kind of T, not just Comparable ones.
I tried and tried but couldn't configure the generic delegate in a manner where I could compare callers.
I read the answer from NRitH, but couldn't get the compiler to accept the syntax. Perhaps there is a way to do this, but I went a different route with success.
I gave up on the protocol/delegate technique and ended up using a callback closure. Instead of defining a delegate in MyClass and implementing that delegate in AClass, I defined a callback closure in MyClass in implemented that in AClass like so:
class MyClass<T: Comparable>: Equatable {
var valueChanged: ((_ clazz: MyClass) -> Void)?
private var _value: T
var value: T {
set {
_value = newValue
valueChanged?(self)
}
get {
return _value
}
}
init(value: T) {
_value = value
}
static public func ==(lhs: MyClass<T>, rhs: MyClass<T>) -> Bool {
return lhs.value == rhs.value
}
}
class AClass {
fileprivate var thing1: MyClass<Int> = MyClass(value: 10)
fileprivate var thing2: MyClass<Int> = MyClass(value: 100)
fileprivate var thing3: MyClass<TimeInterval> = MyClass(value: 10)
fileprivate var thing4: MyClass<TimeInterval> = MyClass(value: 100)
init() {
thing1.valueChanged = { (thing) in
// You can compare here, but you already know which object is callling back
if thing == self.thing1 {
}
}
thing2.valueChanged = { (thing) in
print("Thing2 changed: \(thing.value)")
}
}
}

swift 3 downcast to dynamic class

I am trying to create a couple of objects which are dependent one to each other and they mush have a method to downcast directly the concrete class of the other object. Something like this:
protocol aProt
{
var bVar:bProt! { get set }
}
protocol bProt
{
var aVar:aProt! { get set }
}
class a: aProt
{
var bVar: bProt!
func bConcrete() -> b {
return bVar as! b
}
}
class b: bProt
{
var aVar: aProt!
func aConcrete() -> a {
return aVar as! a
}
Now, the problem is that I want this behavior (func aConcrete(),func bConcrete()) to be inherited by the subclasses of a and b. Then I thought the perfect way of doing this was using generics, but... There's no way of doing this.
class a: aProt
{
var bVar: bProt!
func bConcrete() -> T {
return bVar as! T
}
}
class b: bProt
{
var aVar: aProt!
func aConcrete<T>() -> T {
return aVar as! T
}
You can do it but when you have to use it you must downcast the variable anyway, so there is no way of doing it in a clean manner:
let aObject = a()
let bSubclassObject = a.bConcrete() // The compiler complains it cannot infer the class of T
let bSubclassObject = a.bConcrete() as! bSubclass // this works, but this is exactly which I wanted to avoid... :(
Define the generic function and add where to T:
protocol aProt {
var bVar: bProt! { get set }
}
protocol bProt {
var aVar:aProt! { get set }
}
class a: aProt {
var bVar: bProt!
func bConcrete<T: b>(_ type: T.Type) -> T? {
return bVar as? T
}
}
class b: bProt {
var aVar: aProt!
func aConcrete<T: a>(_ type: T.Type) -> T? {
return aVar as? T
}
}
class a1: a { }
class b1: b {
var fullName: String = "new object"
}
let aObj = a()
aObj.bVar = b1()
let bObj = aObj.bConcrete(b1.self)
bObj?.fullName
According to your requirement, calls bConcrete(b1.self) might still not good enough, but at least you need to know what type of data you are expecting to return.

How to initialize variables inside subclasses' initializer that will be used in the super class's initializer?

In the following code, my intention is to have repeatNum declared in the base class because it is used in the base class (inside init and other functions). And each subclass should set repeatNum because only the subclass knows its own value.
class Base {
var repeatNum: Int
init() {
for var i=0; i<repeatNum; ++i {
print("*")
}
}
}
class SubClass1 : Base {
override init() {
repeatNum = 10
super.init()
}
}
class SubClass2 : Base {
override init() {
repeatNum = 5
super.init()
}
}
Of course, it prompts some errors:
For the base class:
'self.repeatNum' not initialized
Return from initializer without initiating all stored properties.
For the subclasses:
use of 'self' in property access 'repeatNum' before super.init initializes self
I know I can simply call super.init() before setting repeatNum in the subclasses, but I really need to use repeatNum inside the base's init() in the real case. It becomes a paradox for me here somehow.
I also understand I can change the design to make the errors go away, but I am wondering if it's possible to achieve my original intention with some tweaks? I probably miss something here.
Your current code doesn't even compile, does it? It should complain about trying to use repeatNum before being initialized...
Maybe something like this could be an option for you?
class Base {
var repeatNum: Int
init(repeatNum: Int) {
self.repeatNum = repeatNum
for _ in 0..<repeatNum {
print("*")
}
}
}
class SubClass1 : Base {
init() {
super.init(repeatNum: 10)
}
}
class SubClass2 : Base {
init() {
super.init(repeatNum: 5)
}
}
i don't know what is you want,the following code is that ok
class Base {
var repeatNum: Int?
init() {
}
func printyourneed(){
for var i=0; i<repeatNum; ++i {
print("*")
}
}
}
class SubClass1 : Base {
override init() {
super.init()
self.repeatNum = 10
self.printyourneed()
}
}
class SubClass2 : Base {
override init() {
super.init()
self.repeatNum = 5
self.printyourneed()
}
}
or
class Base {
var _repeatNum:Int?;
var repeatNum: Int?{
get{
return _repeatNum
}
set{
_repeatNum = newValue
printyourneed()
}
}
init() {
}
func printyourneed(){
for var i=0; i<repeatNum; ++i {
print("*")
}
}
}
class SubClass1 : Base {
override init() {
super.init()
self.repeatNum = 10
}
}
class SubClass2 : Base {
override init() {
super.init()
self.repeatNum = 5
}
}