iOS – 5 Popular Ways Of Making Network Requests In Swift

Hello Readers, CoolMonkTechie heartily welcomes you in this article.

In this article, We will learn about five major ways of making network request in Swift. Many iOS application are clients of a REST API. This article gives an overview of common ways to accomplish tasks associated with making HTTP requests and handling responses.

A famous quote about learning is :

Change is the end result of all true learning.

So Let’s begin.

First we need to understand that how to threading relates from network requests.


How does threading relate from network requests?

In iOS much of the code that runs in your application is triggered by an event on the main event loop. The main event loop is responsible for executing code to respond to things like user interaction (e.g triggering an @IBAction) or events in a view controller’s lifecycle (e.g. viewDidLoad). Code executed from the main event loop is run on the main thread. This is convenient for us because any updates to an application’s UI elements must happen on the main thread. We’ll want to keep this rule in mind when working with network requests.

iOS provides a couple of higher level libraries for concurrent programming: Grand Central Dispatch and NSOperationQueue. We’ll be able use either to ensure us that a piece of code does or does not run on the main thread.

We should never make a synchronous network request on the main thread since this will block thread and UI will appear frozen while our request is pending. We’ll rarely run into instances where we’ll need to make synchronous requests.

When we make an asynchronous request, any of the above libraries will execute the request on a background (i.e. not the main) thread. Some methods will allow us to specify the dispatch queue on which we want the response handler to run, others will provide no guarantees. If we need to update the UI in our response handler we must ensure that the code that manipulates the UI is run on the main thread. This can be tricky because we may call into a method that calls into another method that after a long stack of calls eventually updates a UI element.

One simple way to ensure a block of code is run on the main thread using Grand Central Dispatch is as follows:

DispatchQueue.main.async {  
  // This code will be executed on the main thread
}  


Overview of popular network programming methods

There is a wide variety of ways to make HTTP requests in iOS with which we might at least want to be familiar. We will discuss the below 4 major ways of making network requests in network programming :


1. NSURLConnection

NSURLConnection is a lower level mechanism to make URL requests and is part of Apple’s Foundation Framework.

  • simplest way to make URL request.
  • provides synchronous and asynchronous requests via with completion handler blocks and delegates.
  • does not have much support for authentication or a session concept.
  • does not have an “operation” or “task” concept associated with requests so there’s no convenient way to handle queue of requests or to pause/resume.
  • does not handle parsing of common content types.
  • not much built in support for HTTP error codes / request parameters.
  • deprecated in iOS 9.


Example

Notice that we are forced to specify a operation queue on which the completion handler will run.

import UIKit

private let apiKey = "53eb9541b4374660d6f3c0001d6249ca:19:70900879"
private let resourceUrl = NSURL(string: "http://api.nytimes.com/svc/topstories/v1/home.json?api-key=\(apiKey)")!

class Story {
    var headline: String?
    var thumbnailUrl: String?

    init(jsonResult: NSDictionary) {
        ...
    }

    class func fetchStories(successCallback: ([Story]) -> Void, error: ((NSError?) -> Void)?) {
        let request = NSURLRequest(URL: resourceUrl)
        NSURLConnection.sendAsynchronousRequest(request, queue: NSOperationQueue.mainQueue()) { (response, data, requestError) -> Void in
            if let requestError = requestError? {
                error?(requestError)
            } else {
                if let data = data? {
                    let json = NSJSONSerialization.JSONObjectWithData(data, options: nil, error: nil) as NSDictionary
                    if let results = json["results"] as? NSArray {
                        var stories: [Story] = []
                        for result in results as [NSDictionary] {
                            stories.append(Story(jsonResult: result))
                        }
                        successCallback(stories)
                    }
                } else {
                    // unexpected error happened
                    error?(nil)
                }
            }
        }
    }
}


2. URLSession

NSURLSession a higher level library that is part of Apple’s Foundation Framework.

  • built on top of NSURLConnection
  • better support for authentication and has a session concept
  • concept of “task” enables pausing/resuming requests
  • can perform requests while your app is in the background
  • does not handle parsing of common content types
  • not much built in support for HTTP error codes / request parameters

URLSession is now the preferred built-in method of performing network requests on iOS.


Example

class Movie {

    // ... 

    class func fetchMovies(successCallBack: @escaping (NSDictionary) -> (), errorCallBack: ((Error?) -> ())?) {
        let apiKey = "Put_Your_Client_Id_Here"
        let url = URL(string: "https://api.themoviedb.org/3/movie/now_playing?api_key=\(apiKey)")!
        let request = URLRequest(url: url, cachePolicy: .reloadIgnoringLocalCacheData, timeoutInterval: 10)
        let session = URLSession(configuration: .default, delegate: nil, delegateQueue: OperationQueue.main)
        let task: URLSessionDataTask = session.dataTask(with: request) { (data: Data?, response: URLResponse?, error: Error?) in
            if let error = error {
                errorCallBack?(error)
            } else if let data = data,
                let dataDictionary = try! JSONSerialization.jsonObject(with: data, options: []) as? NSDictionary {
                //print(dataDictionary)
                successCallBack(dataDictionary)
            }
        }
        task.resume()
    }
    // ...
}


3. Codable

Codable is Apple’s latest powerful contribution to efforts to better improve the built-in networking libraries available to iOS and Mac OS engineers. Codable is actually a typealias for Encodable and Decodable protocols that allows us to quickly decode and encode external representations (such as JSON strings) as native Structs in Swift.

Before Codable was introduced to the Swift language in Swift 4, many developers had to rely on third party frameworks or building their own JSON decoding code which required a lot of boilerplate code. However, with the introduction of Codable, it’s actually really easy to write 100% Swift networking code! Let’s give it a try!


Example

Let’s assume we are implementing a movie list viewing application that retrieves a list of movies from a server to show the user, with the following JSON response.

{
    "totalFilms": 100,
    "films": [{
            "id": 11219,
            "image_url": "https://movieguru.org/sample1.png",
            "title": "The Hunt for Red October",
            "score": 4.5
        },
        {
            "id": 11169,
            "image_url": "https://movieguru.org/sample1.png",
            "title": "Wolf of Wall Street",
            "score": 5.0
        },
        {
            "id": 13671,
            "image_url": "https://movieguru.org/sample1.png",
            "title": "Superbad",
            "score": 4.2
        }
    ]
}

With the power of Codable, you can implement a native Struct object with the following code.

struct MovieResponse: Codable {
    let totalFilms: Int
    let films: [Film]
}

struct Film: Codable {
    let id: Int
    let imageURL, title: String
    let score: Double

    enum CodingKeys: String, CodingKey {
        case id
        case imageURL = "image_url"
        case title, score
    }
}

As we can see, the Struct looks almost exactly like we would want it to look if we were using it to drive a UITableViewDataSource or a custom film details View Controller.

The CodingKey is simply an enum that allows the JSONDecoder to perform an internal switch on each key of the JSON response in order to match each key to the property name of the Struct.

Can we guess why imageURL is the only case in the enum that has a declared raw value? If we’re thinking it’s related to Snake Case, we’re right! While Snake Case works in Swift, it’s not best practice, and Apple cleverly considered that an application might need properties to have different names compared to their network properties so an engineer can overwrite the property name by mapping a different variable name to each JSON parameter. If we wanted, we could make imageURL read “thumbnailURL” instead, as long as the encoding key is equal to “image_url”, the JSONDecoder will know that the JSON value for key “image_url” is set to thumbnailURL.

So how do we use it? Easy!!

Let’s go back to the example from URLSession, and instead of a general NSDictionary (which would require a lot more code on the consumption side like a MovieObject class with init(fromDict dict: NSDictionary) in order to be usable in your code base), let’s substitute it with our Codable compatible struct.

class Movie {

    // ... 

    class func fetchMovies(successCallBack: @escaping ([Film]?) -> (), errorCallBack: ((Error?) -> ())?) {
        let apiKey = "Put_Your_Client_Id_Here"
        let url = URL(string: "https://api.themoviedb.org/3/movie/now_playing?api_key=\(apiKey)")!
        let request = URLRequest(url: url, cachePolicy: .reloadIgnoringLocalCacheData, timeoutInterval: 10)
        let session = URLSession(configuration: .default, delegate: nil, delegateQueue: OperationQueue.main)
        let task: URLSessionDataTask = session.dataTask(with: request) { (data: Data?, response: URLResponse?, error: Error?) in
            if let error = error {
                errorCallBack?(error)
            } else if let data = data,
                let filmResponse = try! JSONDecoder().decode(MovieResponse.self, from: data) {
                //print(filmResponse.films)
                successCallBack(filmResponse.films)
            }
        }
        task.resume()
    }
    // ...
}

It might not look like much, but a significant amount of code is saved from the Movie class object, and when we consume this API call, on the other side we’ll get the Film objects we need to drive our UI, rather than a Dictionary you’d have to iterate over, verifying each value.

Codable can save us a significant amount of time in writing networking code, and it’s growing in popularity, so we highly recommend picking it up!!


4. AFNetworking

AFNetworking is the most popular library for and is the de facto gold standard for networking tasks in the iOS world. Chances are we will want to use this library if accessing an API and making network requests is a key part of your application.

  • built on top of NSURLSession.
  • built-in support for parsing common content-types.
  • great support for common HTTP operations including handling of request params, headers, error codes.
  • great integration with UIKit components making complex behavior like loading remote images asynchronously very easy.


Example

This code starts to look a little cleaner with AFNetworking. AFNetworking does some error handling for us and gives us a way to provide a failure handler. Also note that we no longer have to parse the JSON result ourselves since AFNetworking does this automatically based on the content type. Finally note that we were able to supply our GET parameters as a Swift dictionary. This is not particularly useful here, but becomes very nice to have if there is a large number of parameters.

private let params = ["api-key": "53eb9541b4374660d6f3c0001d6249ca:19:70900879"]
private let resourceUrl = "http://api.nytimes.com/svc/topstories/v1/home.json"

class Story {
    var headline: String?
    var thumbnailUrl: String?

    init(jsonResult: NSDictionary) {
       ...
    }

    class func fetchStories(successCallback: ([Story]) -> Void, error: ((NSError?) -> Void)?) {
        let manager = AFHTTPRequestOperationManager()
        manager.GET(resourceUrl, parameters: params, success: { (operation ,responseObject) -> Void in
            if let results = responseObject["results"] as? NSArray {
                var stories: [Story] = []
                for result in results as [NSDictionary] {
                    stories.append(Story(jsonResult: result))
                }
                successCallback(stories)
            }
        }, failure: { (operation, requestError) -> Void in
            if let errorCallback = error? {
                errorCallback(requestError)
            }
        })
    }
}


5. AlamoFire

AlamoFire is another networking library by the same author as AFNetworking. It is written in Swift.

  • Swift only
  • many of the same features as AFNetworking
  • easy to use/read syntax for making common requests
  • no integration with UIKit


Example

This example demonstrates that how to use Alamofire library using Swift Code in iOS.

import UIKit
import Alamofire
class ViewController: UIViewController {
 
    @IBOutlet weak var tableView: UITableView!
    var jsonArray:NSMutableArray?
    var newArray: Array<String> = []
 
    override func viewDidLoad() {
        super.viewDidLoad()
        
        Alamofire.request(.GET, "https://rocky-meadow-1164.herokuapp.com/todo") .responseJSON { response in
                print(response.request)  // original URL request
                print(response.response) // URL response
                print(response.data)     // server data
                print(response.result)   // result of response serialization
                
                if let JSON = response.result.value {
                    self.jsonArray = JSON as? NSMutableArray
                    for item in self.jsonArray! {
                        print(item["name"]!)
                        let string = item["name"]!
                        print("String is \(string!)")
                        
                        self.newArray.append(string! as! String)
                    }
                    
                    print("New array is \(self.newArray)")
                    
                    self.tableView.reloadData()
                }
        }
        
        
        
        // Do any additional setup after loading the view, typically from a nib.
    }
}

That’s all about in this article.


Conclusion

In this article, We understood about 5 Popular Ways Of Making Network Requests In Swift. We have discussed the common ways to accomplish tasks associated with making HTTP requests and handling responses in Swift.

Thanks for reading ! I hope you enjoyed and learned about the Overview of popular network programming methods in Swift. Reading is one thing, but the only way to master it is to do it yourself.

Please follow and subscribe us on this blog and and support us in any way possible. Also like and share the article with others for spread valuable knowledge.

If you have any comments, questions, or think I missed something, feel free to leave them below in the comment box.

Thanks again Reading. HAPPY READING !!???

iOS – How To Solve ARC Strong Reference Cycle In Swift ?

Hello Readers, CoolMonkTechie heartily welcomes you in this article.

In this article, We will learn about How to work Automatic Reference Counting (ARC) in Swift. We will discuss about Strong Reference Cycles problem and solution between Class Instances and for closures in Swift. Memory management is the core concept in any programming language. For more information about Memory Management Concepts, you can explore the below articles :

iOS – Why Is Advanced IOS Memory Management Valuable In Swift ?

A famous quote about learning is :

Education is not the filling of a pot but the lighting of a fire.

So Let’s begin.

Swift uses Automatic Reference Counting (ARC) to track and manage your app’s memory usage. In most cases, this means that memory management “just works” in Swift, and we do not need to think about memory management ourself. ARC automatically frees up the memory used by class instances when those instances are no longer needed.

However, in a few cases ARC requires more information about the relationships between parts of our code in order to manage memory for us. We describes those situations and shows how we enable ARC to manage all of our app’s memory. 

Reference counting applies only to instances of classes. Structures and enumerations are value types, not reference types, and are not stored and passed by reference.


How To Work ARC ?

Every time we create a new instance of a class, ARC allocates a chunk of memory to store information about that instance. This memory holds information about the type of the instance, together with the values of any stored properties associated with that instance.

Additionally, when an instance is no longer needed, ARC frees up the memory used by that instance so that the memory can be used for other purposes instead. This ensures that class instances do not take up space in memory when they are no longer needed.

However, if ARC were to deallocate an instance that was still in use, it would no longer be possible to access that instance’s properties, or call that instance’s methods. Indeed, if we tried to access the instance, our app would most likely crash.

To make sure that instances don’t disappear while they are still needed, ARC tracks how many properties, constants, and variables are currently referring to each class instance. ARC will not deallocate an instance as long as at least one active reference to that instance still exists.

To make this possible, whenever we assign a class instance to a property, constant, or variable, that property, constant, or variable makes a strong reference to the instance. The reference is called a “strong” reference because it keeps a firm hold on that instance, and does not allow it to be deallocated for as long as that strong reference remains.

Example :

Here’s an example of how Automatic Reference Counting works. This example starts with a simple class called Person, which defines a stored constant property called name:

class Person {
    let name: String
    init(name: String) {
        self.name = name
        print("\(name) is being initialized")
    }
    deinit {
        print("\(name) is being deinitialized")
    }
}

The Person class has an initializer that sets the instance’s name property and prints a message to indicate that initialization is underway. The Person class also has a deinitializer that prints a message when an instance of the class is deallocated.

The next code snippet defines three variables of type Person?, which are used to set up multiple references to a new Person instance in subsequent code snippets. Because these variables are of an optional type (Person?, not Person), they are automatically initialized with a value of nil, and do not currently reference a Person instance.

var reference1: Person?
var reference2: Person?
var reference3: Person?

We can now create a new Person instance and assign it to one of these three variables:

reference1 = Person(name: "John Appleseed")
// Prints "John Appleseed is being initialized"

Note that the message "John Appleseed is being initialized" is printed at the point that we call the Person class’s initializer. This confirms that initialization has taken place.

Because the new Person instance has been assigned to the reference1 variable, there’s now a strong reference from reference1 to the new Person instance. Because there’s at least one strong reference, ARC makes sure that this Person is kept in memory and is not deallocated.

If we assign the same Person instance to two more variables, two more strong references to that instance are established:

reference2 = reference1
reference3 = reference1

There are now three strong references to this single Person instance.

If we break two of these strong references (including the original reference) by assigning nil to two of the variables, a single strong reference remains, and the Person instance is not deallocated:

reference1 = nil
reference2 = nil

ARC does not deallocate the Person instance until the third and final strong reference is broken, at which point it’s clear that you are no longer using the Person instance:

reference3 = nil
// Prints "John Appleseed is being deinitialized"


How To Cause Strong Reference Cycles Between Class Instances ?

In the examples above, ARC is able to track the number of references to the new Person instance we create and to deallocate that Person instance when it’s no longer needed.

However, it’s possible to write code in which an instance of a class never gets to a point where it has zero strong references. This can happen if two class instances hold a strong reference to each other, such that each instance keeps the other alive. This is known as a Strong Reference Cycle.

We resolve strong reference cycles by defining some of the relationships between classes as weak or unowned references instead of as strong references. However, before we learn how to resolve a strong reference cycle, it’s useful to understand how such a cycle is caused.

Here’s an example of how a strong reference cycle can be created by accident. This example defines two classes called Person and Apartment, which model a block of apartments and its residents:

class Person {
    let name: String
    init(name: String) { self.name = name }
    var apartment: Apartment?
    deinit { print("\(name) is being deinitialized") }
}

class Apartment {
    let unit: String
    init(unit: String) { self.unit = unit }
    var tenant: Person?
    deinit { print("Apartment \(unit) is being deinitialized") }
}

Every Person instance has a name property of type String and an optional apartment property that is initially nil. The apartment property is optional, because a person may not always have an apartment.

Similarly, every Apartment instance has a unit property of type String and has an optional tenant property that is initially nil. The tenant property is optional because an apartment may not always have a tenant.

Both of these classes also define a deinitializer, which prints the fact that an instance of that class is being deinitialized. This enables us to see whether instances of Person and Apartment are being deallocated as expected.

This next code snippet defines two variables of optional type called john and unit4A, which will be set to a specific Apartment and Person instance below. Both of these variables have an initial value of nil, by virtue of being optional:

var john: Person?
var unit4A: Apartment?

We can now create a specific Person instance and Apartment instance and assign these new instances to the john and unit4A variables:

john = Person(name: "John Appleseed")
unit4A = Apartment(unit: "4A")

Here’s how the strong references look after creating and assigning these two instances. The john variable now has a strong reference to the new Person instance, and the unit4A variable has a strong reference to the new Apartment instance:

We can now link the two instances together so that the person has an apartment, and the apartment has a tenant. Note that an exclamation point (!) is used to unwrap and access the instances stored inside the john and unit4A optional variables, so that the properties of those instances can be set:

john!.apartment = unit4A
unit4A!.tenant = john

Here’s how the strong references look after we link the two instances together:

Unfortunately, linking these two instances creates a strong reference cycle between them. The Person instance now has a strong reference to the Apartment instance, and the Apartment instance has a strong reference to the Person instance. Therefore, when you break the strong references held by the john and unit4A variables, the reference counts do not drop to zero, and the instances are not deallocated by ARC:

john = nil
unit4A = nil

Note that neither deinitializer was called when we set these two variables to nil. The strong reference cycle prevents the Person and Apartment instances from ever being deallocated, causing a memory leak in your app.

Here’s how the strong references look after we set the john and unit4A variables to nil:

The strong references between the Person instance and the Apartment instance remain and cannot be broken.


How To Solve Strong Reference Cycles Between Class Instances ?

Swift provides two ways to resolve strong reference cycles when we work with properties of class type: weak references and unowned references.

Weak and unowned references enable one instance in a reference cycle to refer to the other instance without keeping a strong hold on it. The instances can then refer to each other without creating a strong reference cycle.

Use a weak reference when the other instance has a shorter lifetime—that is, when the other instance can be deallocated first. In the Apartment example above, it’s appropriate for an apartment to be able to have no tenant at some point in its lifetime, and so a weak reference is an appropriate way to break the reference cycle in this case. In contrast, use an unowned reference when the other instance has the same lifetime or a longer lifetime.


Weak References

weak reference is a reference that does not keep a strong hold on the instance it refers to, and so does not stop ARC from disposing of the referenced instance. This behavior prevents the reference from becoming part of a strong reference cycle. We indicate a weak reference by placing the weak keyword before a property or variable declaration.

Because a weak reference does not keep a strong hold on the instance it refers to, it’s possible for that instance to be deallocated while the weak reference is still referring to it. Therefore, ARC automatically sets a weak reference to nil when the instance that it refers to is deallocated. And, because weak references need to allow their value to be changed to nil at runtime, they are always declared as variables, rather than constants, of an optional type.

We can check for the existence of a value in the weak reference, just like any other optional value, and you will never end up with a reference to an invalid instance that no longer exists.

“Property observers aren’t called when ARC sets a weak reference to nil.”

The example below is identical to the Person and Apartment example from above, with one important difference. This time around, the Apartment type’s tenant property is declared as a weak reference:

class Person {
    let name: String
    init(name: String) { self.name = name }
    var apartment: Apartment?
    deinit { print("\(name) is being deinitialized") }
}

class Apartment {
    let unit: String
    init(unit: String) { self.unit = unit }
    weak var tenant: Person?
    deinit { print("Apartment \(unit) is being deinitialized") }
}

The strong references from the two variables (john and unit4A) and the links between the two instances are created as before:

var john: Person?
var unit4A: Apartment?

john = Person(name: "John Appleseed")
unit4A = Apartment(unit: "4A")

john!.apartment = unit4A
unit4A!.tenant = john

Here’s how the references look now that you’ve linked the two instances together:

The Person instance still has a strong reference to the Apartment instance, but the Apartment instance now has a weak reference to the Person instance. This means that when we break the strong reference held by the john variable by setting it to nil, there are no more strong references to the Person instance:

john = nil
// Prints "John Appleseed is being deinitialized"

Because there are no more strong references to the Person instance, it’s deallocated and the tenant property is set to nil:

The only remaining strong reference to the Apartment instance is from the unit4A variable. If we break that strong reference, there are no more strong references to the Apartment instance:

unit4A = nil
// Prints "Apartment 4A is being deinitialized"

Because there are no more strong references to the Apartment instance, it too is deallocated:

“In systems that use garbage collection, weak pointers are sometimes used to implement a simple caching mechanism because objects with no strong references are deallocated only when memory pressure triggers garbage collection. However, with ARC, values are deallocated as soon as their last strong reference is removed, making weak references unsuitable for such a purpose.”


Unowned References

Like a weak reference, an unowned reference does not keep a strong hold on the instance it refers to. Unlike a weak reference, however, an unowned reference is used when the other instance has the same lifetime or a longer lifetime. We indicate an unowned reference by placing the unowned keyword before a property or variable declaration.

Unlike a weak reference, an unowned reference is expected to always have a value. As a result, marking a value as unowned doesn’t make it optional, and ARC never sets an unowned reference’s value to nil.

Use an unowned reference only when you are sure that the reference always refers to an instance that has not been deallocated. If we try to access the value of an unowned reference after that instance has been deallocated, we’ll get a runtime error.

The following example defines two classes, Customer and CreditCard, which model a bank customer and a possible credit card for that customer. These two classes each store an instance of the other class as a property. This relationship has the potential to create a strong reference cycle.

The relationship between Customer and CreditCard is slightly different from the relationship between Apartment and Person seen in the weak reference example above. In this data model, a customer may or may not have a credit card, but a credit card will always be associated with a customer. A CreditCard instance never outlives the Customer that it refers to. To represent this, the Customer class has an optional card property, but the CreditCard class has an unowned (and non-optional) customer property.

Furthermore, a new CreditCard instance can only be created by passing a number value and a customer instance to a custom CreditCard initializer. This ensures that a CreditCard instance always has a customer instance associated with it when the CreditCard instance is created.

Because a credit card will always have a customer, we define its customer property as an unowned reference, to avoid a strong reference cycle:

class Customer {
    let name: String
    var card: CreditCard?
    init(name: String) {
        self.name = name
    }
    deinit { print("\(name) is being deinitialized") }
}

class CreditCard {
    let number: UInt64
    unowned let customer: Customer
    init(number: UInt64, customer: Customer) {
        self.number = number
        self.customer = customer
    }
    deinit { print("Card #\(number) is being deinitialized") }
}

“The number property of the CreditCard class is defined with a type of UInt64 rather than Int, to ensure that the number property’s capacity is large enough to store a 16-digit card number on both 32-bit and 64-bit systems.”

This next code snippet defines an optional Customer variable called john, which will be used to store a reference to a specific customer. This variable has an initial value of nil, by virtue of being optional:

var john: Customer?

We can now create a Customer instance, and use it to initialize and assign a new CreditCard instance as that customer’s card property:

john = Customer(name: "John Appleseed")
john!.card = CreditCard(number: 1234_5678_9012_3456, customer: john!)

Here’s how the references look, now that we’ve linked the two instances:

The Customer instance now has a strong reference to the CreditCard instance, and the CreditCard instance has an unowned reference to the Customer instance.

Because of the unowned customer reference, when we break the strong reference held by the john variable, there are no more strong references to the Customer instance:

Because there are no more strong references to the Customer instance, it’s deallocated. After this happens, there are no more strong references to the CreditCard instance, and it too is deallocated:

john = nil
// Prints "John Appleseed is being deinitialized"
// Prints "Card #1234567890123456 is being deinitialized"

The final code snippet above shows that the deinitializers for the Customer instance and CreditCard instance both print their “deinitialized” messages after the john variable is set to nil.

“The examples above show how to use safe unowned references. Swift also provides unsafe unowned references for cases where you need to disable runtime safety checks—for example, for performance reasons. As with all unsafe operations, we take on the responsibility for checking that code for safety. We indicate an unsafe unowned reference by writing unowned(unsafe). If we try to access an unsafe unowned reference after the instance that it refers to is deallocated, our program will try to access the memory location where the instance used to be, which is an unsafe operation.”


Unowned Optional References

We can mark an optional reference to a class as unowned. In terms of the ARC ownership model, an unowned optional reference and a weak reference can both be used in the same contexts. The difference is that when we use an unowned optional reference, we’re responsible for making sure it always refers to a valid object or is set to nil.

Here’s an example that keeps track of the courses offered by a particular department at a school:

class Department {
    var name: String
    var courses: [Course]
    init(name: String) {
        self.name = name
        self.courses = []
    }
}

class Course {
    var name: String
    unowned var department: Department
    unowned var nextCourse: Course?
    init(name: String, in department: Department) {
        self.name = name
        self.department = department
        self.nextCourse = nil
    }
}

Department maintains a strong reference to each course that the department offers. In the ARC ownership model, a department owns its courses. Course has two unowned references, one to the department and one to the next course a student should take; a course doesn’t own either of these objects. Every course is part of some department so the department property isn’t an optional. However, because some courses don’t have a recommended follow-on course, the nextCourse property is an optional. Here’s an example of using these classes:

let department = Department(name: "Horticulture")

let intro = Course(name: "Survey of Plants", in: department)
let intermediate = Course(name: "Growing Common Herbs", in: department)
let advanced = Course(name: "Caring for Tropical Plants", in: department)

intro.nextCourse = intermediate
intermediate.nextCourse = advanced
department.courses = [intro, intermediate, advanced]

The code above creates a department and its three courses. The intro and intermediate courses both have a suggested next course stored in their nextCourse property, which maintains an unowned optional reference to the course a student should take after after completing this one.

An unowned optional reference doesn’t keep a strong hold on the instance of the class that it wraps, and so it doesn’t prevent ARC from deallocating the instance. It behaves the same as an unowned reference does under ARC, except that an unowned optional reference can be nil.

Like non-optional unowned references, we’re responsible for ensuring that nextCourse always refers to a course that hasn’t been deallocated. In this case, for example, when we delete a course from department.courses we also need to remove any references to it that other courses might have.

” The underlying type of an optional value is Optional, which is an enumeration in the Swift standard library. However, optionals are an exception to the rule that value types can’t be marked with unowned. The optional that wraps the class doesn’t use reference counting, so we don’t need to maintain a strong reference to the optional.”


Unowned References and Implicitly Unwrapped Optional Properties

The examples for weak and unowned references above cover two of the more common scenarios in which it’s necessary to break a strong reference cycle.

However, there’s a scenario, in which both properties should always have a value, and neither property should ever be nil once initialization is complete. In this scenario, it’s useful to combine an unowned property on one class with an implicitly unwrapped optional property on the other class.

This enables both properties to be accessed directly (without optional unwrapping) once initialization is complete, while still avoiding a reference cycle. This section shows you how to set up such a relationship.

The example below defines two classes, Country and City, each of which stores an instance of the other class as a property. In this data model, every country must always have a capital city, and every city must always belong to a country. To represent this, the Country class has a capitalCity property, and the City class has a country property:

class Country {
    let name: String
    var capitalCity: City!
    init(name: String, capitalName: String) {
        self.name = name
        self.capitalCity = City(name: capitalName, country: self)
    }
}

class City {
    let name: String
    unowned let country: Country
    init(name: String, country: Country) {
        self.name = name
        self.country = country
    }
}

To set up the interdependency between the two classes, the initializer for City takes a Country instance, and stores this instance in its country property.

The initializer for City is called from within the initializer for Country. However, the initializer for Country cannot pass self to the City initializer until a new Country instance is fully initialized, 

To cope with this requirement, we declare the capitalCity property of Country as an implicitly unwrapped optional property, indicated by the exclamation point at the end of its type annotation (City!). This means that the capitalCity property has a default value of nil, like any other optional, but can be accessed without the need to unwrap its value.

Because capitalCity has a default nil value, a new Country instance is considered fully initialized as soon as the Country instance sets its name property within its initializer. This means that the Country initializer can start to reference and pass around the implicit self property as soon as the name property is set. The Country initializer can therefore pass self as one of the parameters for the City initializer when the Country initializer is setting its own capitalCity property.

All of this means that we can create the Country and City instances in a single statement, without creating a strong reference cycle, and the capitalCity property can be accessed directly, without needing to use an exclamation point to unwrap its optional value:

var country = Country(name: "Canada", capitalName: "Ottawa")
print("\(country.name)'s capital city is called \(country.capitalCity.name)")
// Prints "Canada's capital city is called Ottawa"

In the example above, the use of an implicitly unwrapped optional means that all of the two-phase class initializer requirements are satisfied. The capitalCity property can be used and accessed like a non-optional value once initialization is complete, while still avoiding a strong reference cycle.


How To Cause Strong Reference Cycles for Closures ?

A strong reference cycle can also occur if we assign a closure to a property of a class instance, and the body of that closure captures the instance. This capture might occur because the closure’s body accesses a property of the instance, such as self.someProperty, or because the closure calls a method on the instance, such as self.someMethod(). In either case, these accesses cause the closure to “capture” self, creating a strong reference cycle.

This strong reference cycle occurs because closures, like classes, are reference types. When we assign a closure to a property, we are assigning a reference to that closure. In essence, it’s the same problem as above—two strong references are keeping each other alive. However, rather than two class instances, this time it’s a class instance and a closure that are keeping each other alive.

Swift provides an elegant solution to this problem, known as a closure capture list. However, before we learn how to break a strong reference cycle with a closure capture list, it’s useful to understand how such a cycle can be caused.

The example below shows how we can create a strong reference cycle when using a closure that references self. This example defines a class called HTMLElement, which provides a simple model for an individual element within an HTML document:

class HTMLElement {

    let name: String
    let text: String?

    lazy var asHTML: () -> String = {
        if let text = self.text {
            return "<\(self.name)>\(text)</\(self.name)>"
        } else {
            return "<\(self.name) />"
        }
    }

    init(name: String, text: String? = nil) {
        self.name = name
        self.text = text
    }

    deinit {
        print("\(name) is being deinitialized")
    }

}

The HTMLElement class defines a name property, which indicates the name of the element, such as "h1" for a heading element, "p" for a paragraph element, or "br" for a line break element. HTMLElement also defines an optional text property, which we can set to a string that represents the text to be rendered within that HTML element.

In addition to these two simple properties, the HTMLElement class defines a lazy property called asHTML. This property references a closure that combines name and text into an HTML string fragment. The asHTML property is of type () -> String, or “a function that takes no parameters, and returns a String value”.

By default, the asHTML property is assigned a closure that returns a string representation of an HTML tag. This tag contains the optional text value if it exists, or no text content if text does not exist. For a paragraph element, the closure would return "<p>some text</p>" or "<p />", depending on whether the text property equals "some text" or nil.

The asHTML property is named and used somewhat like an instance method. However, because asHTML is a closure property rather than an instance method, we can replace the default value of the asHTML property with a custom closure, if we want to change the HTML rendering for a particular HTML element.

For example, the asHTML property could be set to a closure that defaults to some text if the text property is nil, in order to prevent the representation from returning an empty HTML tag:

let heading = HTMLElement(name: "h1")
let defaultText = "some default text"
heading.asHTML = {
    return "<\(heading.name)>\(heading.text ?? defaultText)</\(heading.name)>"
}
print(heading.asHTML())
// Prints "<h1>some default text</h1>"

” The asHTML property is declared as a lazy property, because it’s only needed if and when the element actually needs to be rendered as a string value for some HTML output target. The fact that asHTML is a lazy property means that you can refer to self within the default closure, because the lazy property will not be accessed until after initialization has been completed and self is known to exist.”

The HTMLElement class provides a single initializer, which takes a name argument and (if desired) a text argument to initialize a new element. The class also defines a deinitializer, which prints a message to show when an HTMLElement instance is deallocated.

Here’s how you use the HTMLElement class to create and print a new instance:

var paragraph: HTMLElement? = HTMLElement(name: "p", text: "hello, world")
print(paragraph!.asHTML())
// Prints "<p>hello, world</p>"

” The paragraph variable above is defined as an optional HTMLElement, so that it can be set to nil below to demonstrate the presence of a strong reference cycle.”

Unfortunately, the HTMLElement class, as written above, creates a strong reference cycle between an HTMLElement instance and the closure used for its default asHTML value. Here’s how the cycle looks:

The instance’s asHTML property holds a strong reference to its closure. However, because the closure refers to self within its body (as a way to reference self.name and self.text), the closure captures self, which means that it holds a strong reference back to the HTMLElement instance. A strong reference cycle is created between the two. 

” Even though the closure refers to self multiple times, it only captures one strong reference to the HTMLElement instance.”

If we set the paragraph variable to nil and break its strong reference to the HTMLElement instance, neither the HTMLElement instance nor its closure are deallocated, because of the strong reference cycle:

paragraph = nil

Note that the message in the HTMLElement deinitializer is not printed, which shows that the HTMLElement instance is not deallocated.


How To Resolve Strong Reference Cycles for Closures ?

We resolve a strong reference cycle between a closure and a class instance by defining a capture list as part of the closure’s definition. A capture list defines the rules to use when capturing one or more reference types within the closure’s body. As with strong reference cycles between two class instances, we declare each captured reference to be a weak or unowned reference rather than a strong reference. The appropriate choice of weak or unowned depends on the relationships between the different parts of our code.

” Swift requires us to write self.someProperty or self.someMethod() (rather than just someProperty or someMethod()) whenever we refer to a member of self within a closure. This helps us remember that it’s possible to capture self by accident.”


Defining a Capture List

Each item in a capture list is a pairing of the weak or unowned keyword with a reference to a class instance (such as self) or a variable initialized with some value (such as delegate = self.delegate). These pairings are written within a pair of square braces, separated by commas.

Place the capture list before a closure’s parameter list and return type if they are provided:

lazy var someClosure = {
    [unowned self, weak delegate = self.delegate]
    (index: Int, stringToProcess: String) -> String in
    // closure body goes here
}

If a closure does not specify a parameter list or return type because they can be inferred from context, place the capture list at the very start of the closure, followed by the in keyword:

lazy var someClosure = {
    [unowned self, weak delegate = self.delegate] in
    // closure body goes here
}


Weak and Unowned References

Define a capture in a closure as an unowned reference when the closure and the instance it captures will always refer to each other, and will always be deallocated at the same time.

Conversely, define a capture as a weak reference when the captured reference may become nil at some point in the future. Weak references are always of an optional type, and automatically become nil when the instance they reference is deallocated. This enables you to check for their existence within the closure’s body.

” If the captured reference will never become nil, it should always be captured as an unowned reference, rather than a weak reference.”

An unowned reference is the appropriate capture method to use to resolve the strong reference cycle in the HTMLElement example from Strong Reference Cycles for Closures above. Here’s how we write the HTMLElement class to avoid the cycle:

class HTMLElement {

    let name: String
    let text: String?

    lazy var asHTML: () -> String = {
        [unowned self] in
        if let text = self.text {
            return "<\(self.name)>\(text)</\(self.name)>"
        } else {
            return "<\(self.name) />"
        }
    }

    init(name: String, text: String? = nil) {
        self.name = name
        self.text = text
    }

    deinit {
        print("\(name) is being deinitialized")
    }

}

This implementation of HTMLElement is identical to the previous implementation, apart from the addition of a capture list within the asHTML closure. In this case, the capture list is [unowned self], which means “capture self as an unowned reference rather than a strong reference”.

We can create and print an HTMLElement instance as before:

var paragraph: HTMLElement? = HTMLElement(name: "p", text: "hello, world")
print(paragraph!.asHTML())
// Prints "<p>hello, world</p>"

Here’s how the references look with the capture list in place:

This time, the capture of self by the closure is an unowned reference, and does not keep a strong hold on the HTMLElement instance it has captured. If you set the strong reference from the paragraph variable to nil, the HTMLElement instance is deallocated, as can be seen from the printing of its deinitializer message in the example below:

paragraph = nil
// Prints "p is being deinitialized"

That’s all about in this article.


Conclusion

In this article, We understood that How to work Automatic Reference Counting (ARC) in Swift. We also discussed about Strong Reference Cycles problem and solution between Class Instances and for closures in Swift. We saw how a strong reference cycle can be created when two class instance properties hold a strong reference to each other, and how to use weak and unowned references to break these strong reference cycles.

We understood Strong Reference Cycles with different scenarios below :

  • The Person and Apartment example shows a situation where two properties, both of which are allowed to be nil, have the potential to cause a strong reference cycle. This scenario is best resolved with a weak reference.
  • The Customer and CreditCard example shows a situation where one property that is allowed to be nil and another property that cannot be nil have the potential to cause a strong reference cycle. This scenario is best resolved with an unowned reference.
  • In a third scenario, in which both properties should always have a value, and neither property should ever be nil once initialization is complete. In this scenario, it’s useful to combine an unowned property on one class with an implicitly unwrapped optional property on the other class.
  • A strong reference cycle can also occur if you assign a closure to a property of a class instance, and the body of that closure captures the instance. Swift provides an elegant solution to this problem, known as a closure capture list.

Thanks for reading ! I hope you enjoyed and learned about the Automatic Reference Counting (ARC) and Strong Reference Cycles problem and solutions with different scenarios in Swift. Reading is one thing, but the only way to master it is to do it yourself.

Please follow and subscribe us on this blog and and support us in any way possible. Also like and share the article with others for spread valuable knowledge.

If you have any comments, questions, or think I missed something, feel free to leave them below in the comment box.

Thanks again Reading. HAPPY READING !!???

iOS – Why Is Advanced IOS Memory Management Valuable In Swift ?

Hello Readers, CoolMonkTechie heartily welcomes you in this article.

In this article, We will learn about what’s beyond the basics of iOS memory management, reference counting and object life cycle. Memory management is the core concept in any programming language. Memory management in iOS was initially non-ARC (Automatic Reference Counting), where we have to retain and release the objects. Now, it supports ARC and we don’t have to retain and release the objects. Xcode takes care of the job automatically in compile time. We will explain Memory management in swift from the compiler perspective. We will discuss the fundamentals and gradually make our way to the internals of ARC and Swift Runtime, answering below questions:

  • What is Memory Management?
  • What is Memory Management Issues?
  • What is Memory Management Rules ?
  • How Swift compiler implements Automatic Reference Counting?
  • How ARC Works ?
  • How to handle Memory in ARC ?
  • How strong, weak and unowned references are implemented?
  • What is Swift Runtime ?
  • What are Side Tables?
  • What is the life cycle of Swift objects?
  • What is Reference Count Invariants during Swift object lifecycle ?

A famous quote about learning is :

An investment in knowledge pays the best interest.”

So, Let’s begin.


What is Memory Management ?

At hardware level, memory is just a long list of bytes. We organized into three virtual parts:

  • Stack, where all local variables go.
  • Global data, where static variables, constants and type metadata go.
  • Heap, where all dynamically allocated objects go. Basically, everything that has a lifetime is stored here.

We’ll continue saying ‘objects’ and ‘dynamically allocated objects’ interchangeably. These are Swift reference types and some special cases of value types.

So We can define Memory Management :

Memory Management is the process of controlling program’s memory. It is critical to understand how it works, otherwise you are likely to run across random crashes and subtle bugs.”


What is Memory Management Issues?

As per Apple documentation, the two major issues in memory management are:

  • Freeing or overwriting data that is still in use. It causes memory corruption and typically results in your application crashing, or worse, corrupted user data.
  • Not freeing data that is no longer in use causes memory leaks. When allocated memory is not freed even though it is never going to be used again, it is known as memory leak. Leaks cause your application to use ever-increasing amounts of memory, which in turn may result in poor system performance or (in iOS) your application being terminated.


What is Memory Management Rules ?

Memory Management Rules are :

  • We own the objects we create, and we have to subsequently release them when they are no longer needed.
  • Use Retain to gain ownership of an object that you did not create. You have to release these objects too when they are not needed.
  • Don’t release the objects that you don’t own.


How Swift compiler implements automatic reference counting?

Memory management is tightly connected with the concept of OwnershipOwnership is the responsibility of some piece of code to eventually cause an object to be destroyed. Any language with a concept of destruction has a concept of ownership. In some languages, like C and non-ARC Objective-C, ownership is managed explicitly by programmers. In other languages, like C++ (in part), ownership is managed by the language. Even languages with implicit memory management still have libraries with concepts of ownership, because there are other program resources besides memory, and it is important to understand what code has the responsibility to release those resources.

Swift already has an ownership system, but it’s “under the covers”: it’s an implementation detail that programmers have little ability to influence.

Automatic reference counting (ARC) is Swift ownership system, which implicitly imposes a set of conventions for managing and transferring ownership.

Swift uses Automatic Reference Counting (ARC) to track and manage your app’s memory usage. In most cases, this means that memory management “just works” in Swift, and you do not need to think about memory management yourself. ARC automatically frees up the memory used by class instances when those instances are no longer needed.

However, in a few cases ARC requires more information about the relationships between parts of your code in order to manage memory for you.

The name by which an object can be pointed is called a reference. Swift references have two levels of strength: strong and weak. Additionally, weak references have a flavour, called unowned.

The essence of Swift memory management is: Swift preserves an object if it is strongly referenced and deallocates it otherwise. The rest is just an implementation detail.”


How ARC Works ?

Every time you create a new instance of a class, ARC allocates a chunk of memory to store information about that instance. This memory holds information about the type of the instance, together with the values of any stored properties associated with that instance.

Additionally, when an instance is no longer needed, ARC frees up the memory used by that instance so that the memory can be used for other purposes instead. This ensures that class instances do not take up space in memory when they are no longer needed.

However, if ARC were to deallocate an instance that was still in use, it would no longer be possible to access that instance’s properties, or call that instance’s methods. Indeed, if you tried to access the instance, your app would most likely crash.

To make sure that instances don’t disappear while they are still needed, ARC tracks how many properties, constants, and variables are currently referring to each class instance. ARC will not deallocate an instance as long as at least one active reference to that instance still exists.

To make this possible, whenever you assign a class instance to a property, constant, or variable, that property, constant, or variable makes a strong reference to the instance. The reference is called a “strong” reference because it keeps a firm hold on that instance, and does not allow it to be deallocated for as long as that strong reference remains.

Example :

Here’s an example of how Automatic Reference Counting works. This example starts with a simple class called Person, which defines a stored constant property called name:

class Person {
    let name: String
    init(name: String) {
        self.name = name
        print("\(name) is being initialized")
    }
    deinit {
        print("\(name) is being deinitialized")
    }
}

The Person class has an initializer that sets the instance’s name property and prints a message to indicate that initialization is underway. The Person class also has a deinitializer that prints a message when an instance of the class is deallocated.

The next code snippet defines three variables of type Person?, which are used to set up multiple references to a new Person instance in subsequent code snippets. Because these variables are of an optional type (Person?, not Person), they are automatically initialized with a value of nil, and do not currently reference a Person instance.

var reference1: Person?
var reference2: Person?
var reference3: Person?

You can now create a new Person instance and assign it to one of these three variables:

reference1 = Person(name: "John Appleseed")
// Prints "John Appleseed is being initialized"

Note that the message "John Appleseed is being initialized" is printed at the point that you call the Person class’s initializer. This confirms that initialization has taken place.

Because the new Person instance has been assigned to the reference1 variable, there is now a strong reference from reference1 to the new Person instance. Because there is at least one strong reference, ARC makes sure that this Person is kept in memory and is not deallocated.

If you assign the same Person instance to two more variables, two more strong references to that instance are established:

reference2 = reference1
reference3 = reference1

There are now three strong references to this single Person instance.

If you break two of these strong references (including the original reference) by assigning nil to two of the variables, a single strong reference remains, and the Person instance is not deallocated:

reference1 = nil
reference2 = nil

ARC does not deallocate the Person instance until the third and final strong reference is broken, at which point it’s clear that you are no longer using the Person instance:

reference3 = nil
// Prints "John Appleseed is being deinitialized"


How to handle Memory in ARC ?

You don’t need to use release and retain in ARC. So, all the view controller’s objects will be released when the view controller is removed. Similarly, any object’s sub-objects will be released when they are released. Note that if other classes have a strong reference to an object of a class, then the whole class won’t be released. So, it is recommended to use weak properties for delegates.


How strong, weak and unowned references are implemented?

The purpose of a strong reference is to keep an object alive. Strong referencing might result in several non-trivial problems.

  • Retain cycles. Considering that Swift language is not cycle-collecting, a reference R to an object which holds a strong reference to the object R (possibly indirectly), results in a reference cycle. We must write lots of boilerplate code to explicitly break the cycle.
  • It is not always possible to make strong references valid immediately on object construction, e.g. with delegates.

Weak references address the problem of back references. An object can be destroyed if there are weak references pointing to it. A weak reference returns nil, when an object it points to is no longer alive. This is called zeroing.

Unowned references are different flavor of weak, designed for tight validity invariants. Unowned references are non-zeroing. When trying to read a non-existent object by an unowned reference, a program will crash with assertion error. They are useful to track down and fix consistency bugs.


What is Swift Runtime ?

The mechanism of ARC is implemented in a library called Swift Runtime. It implements such core features as the runtime type system, including dynamic casting, generics, and protocol conformance registration.

Swift Runtime represents every dynamically allocated object with HeapObject struct. It contains all the pieces of data which make up an object in Swift: reference counts and type metadata.

Internally every Swift object has three reference counts: one for each kind of reference. At the SIL generation phase, swiftc compiler inserts calls to the methods swift_retain() and swift_release(), wherever it’s appropriate. This is done by intercepting initialization and destruction of HeapObjects.

Compilation is one of the steps of Xcode Build System.


What are Side Tables?

Side Tables are mechanism for implementing Swift weak references.

Typically objects don’t have any weak references, hence it is wasteful to reserve space for weak reference count in every object. This information is stored externally in side tables, so that it can be allocated only when it’s really needed.

Instead of directly pointing to an object, weak reference points to the side table, which in its turn points to the object. This solves two problems:

  • saves memory for weak reference count, until an object really needs it.
  • allows to safely zero out weak reference, since it does not directly point to an object, and no longer a subject to race conditions.

Side table is just a reference count + a pointer to an object. They are declared in Swift Runtime as follows (C++ code).

class HeapObjectSideTableEntry {
  std::atomic<HeapObject*> object;
  SideTableRefCounts refCounts;
  // Operations to increment and decrement reference counts
}


What is the life cycle of Swift objects?

Swift objects have their own life cycle, represented by a finite state machine on the figure below. Square brackets indicate a condition that triggers transition from state to state. We will discuss the finite state machines in Eliminating Degenerate View Controller States.

In live state an object is alive. Its reference counts are initialized to 1 strong, 1 unowned and 1 weak (side table starts at +1). Strong and unowned reference access work normally. Once there is a weak reference to the object, the side table is created. The weak reference points to the side table instead of the object.

From the live state, the object moves into the deiniting state once strong reference count reaches zero. The deiniting state means that deinit() is in progress. At this point strong ref operations have no effect. Weak reference reads return nil, if there is an associated side table (otherwise there are no weak refs). Unowned reads trigger assertion failure. New unowned references can still be stored. From this state, the object can take two routes:

  • A shortcut in case there no weak, unowned references and the side table. The object transitions to the dead state and is removed from memory immediately.
  • Otherwise, the object moves to deinited state.

In the deinited state deinit() has been completed and the object has outstanding unowned references (at least the initial +1). Strong and weak stores and reads cannot happen at this point. Unowned stores also cannot happen. Unowned reads trigger assertion error. The object can take two routes from here:

  • In case there are no weak references, the object can be deallocated immediately. It transitions into the dead state.
  • Otherwise, there is still a side table to be removed and the object moves into the freed state.

In the freed state the object is fully deallocated, but its side table is still alive. During this phase the weak reference count reaches zero and the side table is destroyed. The object transitions into its final state.

In the dead state there is nothing left from the object, except for the pointer to it. The pointer to the HeapObject is freed from the Heap, leaving no traces of the object in memory.


What is Reference Count Invariants during Swift object lifecycle ?

During their life cycle, the objects maintain following invariants:

  • When the strong reference count becomes zero, the object is deinitedUnowned reference reads raise assertion errors, weak reference reads become nil.
  • The unowned reference count adds +1 to the strong one, which is decremented after object’s deinit completes.
  • The weak reference count adds +1 to the unowned reference count. It is decremented after the object is freed from memory.


Conclusion

In this article, We understood about Advanced iOS Memory management in Swift. Automatic reference counting is no magic and the better we understand how it works internally, the less our code is prone to memory management errors. Here are the key points to remember:

  • Weak references point to side a table. Unowned and strong references point to an object.
  • Automatic referencing count is implemented on the compiler level. The swiftc compiler inserts calls to release and retain wherever appropriate.
  • Swift objects are not destroyed immediately. Instead, they undergo 5 phases in their life cycle: live -> deiniting -> deinited -> freed -> dead.

Thanks for reading ! I hope you enjoyed and learned about the Advanced memory management concepts in Swift. Reading is one thing, but the only way to master it is to do it yourself.

Please follow and subscribe us on this blog and and support us in any way possible. Also like and share the article with others for spread valuable knowledge.

If you have any comments, questions, or think I missed something, feel free to leave them below in the comment box.

Thanks again Reading. HAPPY READING !!???

iOS – Is Awesome Design Pattern Valuable In Swift ?

Hello Readers, CoolMonkTechie heartily welcomes you in this article.

In this article, We will learn about Why design patterns are important and which one is the most popular frequently used design patterns in Swift. Swift is a programming language that allows developers to create versatile applications for multiple operating systems (though it is most frequently used to write applications for iOS). When we are new in programming languages, we don’t know which design patterns we should use with it and how to implement them.

Being able to use a relevant design pattern is a prerequisite to creating functional, high-quality, and secure applications. 

We’ve decided to help by taking an in-depth look at the design patterns most widely used in Swift and showing different approaches to solving common problems in mobile development with them.

A famous quote about learning is :

“ Anyone who stops learning is old, whether at twenty or eighty. Anyone who keeps learning stays young. ” 


So Let’s begin.


Design Patterns: What they are and why you should know them ?

A software design pattern is a solution to a particular problem you might face when designing an app’s architecture. But unlike out-of-the-box services or open-source libraries, we can’t simply paste a design pattern into our application because it isn’t a piece of code. Rather, it’s a general concept for how to solve a problem. A design pattern is a template that tells you how to write code, but it’s up to you to fit our code to this template.

Design patterns bring several benefits:

  • Tested solutions. We don’t need to waste time and reinvent the wheel trying to solve a particular software development problem, as design patterns already provide the best solution and tell us how to implement it.
  • Code unification. Design patterns provide us with typical solutions that have been tested for drawbacks and bugs, helping us make fewer mistakes when designing our app architecture.
  • Common vocabulary. Instead of providing in-depth explanations of how to solve this or that software development problem, we can simply say what design pattern we used and other developers will immediately understand what solutions we implemented.


Types of Software Design Patterns

Before we describe the most common architecture patterns in Swift, you should first learn the three types of software design patterns and how they differ:

  • Creational Design Patterns
  • Structural Design Patterns
  • Behavioral Design Patterns


1. Creational Design Patterns

Creational software design patterns deal with object creation mechanisms, which increase flexibility and reuse of existing code. They try to instantiate objects in a manner suitable for the particular situation. Here are several creational design patterns:

  • Factory Method
  • Abstract Factory
  • Builder
  • Singleton
  • Prototype


2. Structural Design Patterns

Structural design patterns aim to simplify the design by finding an easy way of realizing relationships between classes and objects. These patterns explain how to assemble objects and classes into larger structures while keeping these structures flexible and efficient.These are some structural architecture patterns:

  • Adapter
  • Bridge
  • Facade
  • Decorator
  • Composite
  • Flyweight
  • Proxy


3. Behavioral Design Patterns

Behaviour design patterns identify common communication patterns between entities and implement these patterns. 

These patterns are concerned with algorithms and the assignment of responsibilities between objects. Behavioral design patterns include:

  • Chain of Responsibility
  • Template Method
  • Command
  • Iterator
  • Mediator
  • Memento
  • Observer
  • Strategy
  • State
  • Visitor

Most of these design patterns, however, are rarely used, and you’re likely to forget how they work before you even need them. So we’ve handpicked the five design patterns most frequently used in Swift to develop applications for iOS and other operating systems.


Most frequently used design patterns in Swift

We’re going to provide only the essential information about each software design pattern – namely, how it works from the technical point of view and when it should be applied. We’ll also give an illustrative example in the Swift programming language.


1. Builder

The Builder pattern is a creational design pattern that allows us to create complex objects from simple objects step by step. This design pattern helps us use the same code for creating different object views.

Imagine a complex object that requires incremental initialization of multiple fields and nested objects. Typically, the initialization code for such objects is hidden inside a mammoth constructor with dozens of parameters. Or even worse, it can be scattered all over the client code.

The Builder design pattern calls for separating the construction of an object from its own class. The construction of this object is instead assigned to special objects called builders and split into multiple steps. To create an object, you successively call builder methods. And you don’t need to go through all the steps – only those required for creating an object with a particular configuration.

You should apply the Builder design pattern :

  • when you want to avoid using a telescopic constructor (when a constructor has too many parameters, it gets difficult to read and manage);
  • when your code needs to create different views of a specific object;
  • when you need to compose complex objects.

Example:

Suppose you’re developing an iOS application for a restaurant and you need to implement ordering functionality. You can introduce two structures, Dish and Order, and with the help of the OrderBuilder object, you can compose orders with different sets of dishes.

// Design Patterns: Builder

import Foundation

// Models

enum DishCategory: Int {
    case firstCourses, mainCourses, garnishes, drinks
}

struct Dish {
    var name: String
    var price: Float
}

struct OrderItem {
    var dish: Dish
    var count: Int
}

struct Order {
    var firstCourses: [OrderItem] = []
    var mainCourses: [OrderItem] = []
    var garnishes: [OrderItem] = []
    var drinks: [OrderItem] = []
    
    var price: Float {
        let items = firstCourses + mainCourses + garnishes + drinks
        return items.reduce(Float(0), { $0 + $1.dish.price * Float($1.count) })
    }
}

// Builder

class OrderBuilder {
    private var order: Order?
    
    func reset() {
        order = Order()
    }
    
    func setFirstCourse(_ dish: Dish) {
        set(dish, at: order?.firstCourses, withCategory: .firstCourses)
    }
    
    func setMainCourse(_ dish: Dish) {
        set(dish, at: order?.mainCourses, withCategory: .mainCourses)
    }
    
    func setGarnish(_ dish: Dish) {
        set(dish, at: order?.garnishes, withCategory: .garnishes)
    }
    
    func setDrink(_ dish: Dish) {
        set(dish, at: order?.drinks, withCategory: .drinks)
    }
    
    func getResult() -> Order? {
        return order ?? nil
    }
    
    private func set(_ dish: Dish, at orderCategory: [OrderItem]?, withCategory dishCategory: DishCategory) {
        guard let orderCategory = orderCategory else {
            return
        }
        
        var item: OrderItem! = orderCategory.filter( { $0.dish.name == dish.name } ).first
        
        guard item == nil else {
            item.count += 1
            return
        }
        
        item = OrderItem(dish: dish, count: 1)
        
        switch dishCategory {
        case .firstCourses:
            order?.firstCourses.append(item)
        case .mainCourses:
            order?.mainCourses.append(item)
        case .garnishes:
            order?.garnishes.append(item)
        case .drinks:
            order?.drinks.append(item)
        }
    }
}

// Usage

let steak = Dish(name: "Steak", price: 2.30)
let chips = Dish(name: "Chips", price: 1.20)
let coffee = Dish(name: "Coffee", price: 0.80)

let builder = OrderBuilder()
builder.reset()
builder.setMainCourse(steak)
builder.setGarnish(chips)
builder.setDrink(coffee)

let order = builder.getResult()
order?.price

// Result:
// 4.30


2. Adapter

Adapter is a structural design pattern that allows objects with incompatible interfaces to work together. In other words, it transforms the interface of an object to adapt it to a different object.

An adapter wraps an object, therefore concealing it completely from another object. For example, you could wrap an object that handles meters with an adapter that converts data into feet.

You should use the Adapter design pattern:

  • when you want to use a third-party class but its interface doesn’t match the rest of your application’s code;
  • when you need to use several existing subclasses but they lack particular functionality and, on top of that, you can’t extend the superclass.

Example:

Suppose you want to implement a calendar and event management functionality in your iOS application. To do this, you should integrate the EventKit framework and adapt the Event model from the framework to the model in your application. An Adapter can wrap the model of the framework and make it compatible with the model in your application.

// Design Patterns: Adapter

import EventKit

// Models

protocol Event: class {
    var title: String { get }
    var startDate: String { get }
    var endDate: String { get }
}

extension Event {
    var description: String {
        return "Name: \(title)\nEvent start: \(startDate)\nEvent end: \(endDate)"
    }
}

class LocalEvent: Event {
    var title: String
    var startDate: String
    var endDate: String
    
    init(title: String, startDate: String, endDate: String) {
        self.title = title
        self.startDate = startDate
        self.endDate = endDate
    }
}

// Adapter

class EKEventAdapter: Event {
    private var event: EKEvent
    
    private lazy var dateFormatter: DateFormatter = {
        let dateFormatter = DateFormatter()
        dateFormatter.dateFormat = "MM-dd-yyyy HH:mm"
        return dateFormatter
    }()
    
    var title: String {
        return event.title
    }
    var startDate: String {
        return dateFormatter.string(from: event.startDate)
    }
    var endDate: String {
        return dateFormatter.string(from: event.endDate)
    }
    
    init(event: EKEvent) {
        self.event = event
    }
}

// Usage

let dateFormatter = DateFormatter()
dateFormatter.dateFormat = "MM/dd/yyyy HH:mm"

let eventStore = EKEventStore()
let event = EKEvent(eventStore: eventStore)
event.title = "Design Pattern Meetup"
event.startDate = dateFormatter.date(from: "06/29/2018 18:00")
event.endDate = dateFormatter.date(from: "06/29/2018 19:30")

let adapter = EKEventAdapter(event: event)
adapter.description

// Result:
// Name: Design Pattern Meetup
// Event start: 06-29-2018 18:00
// Event end: 06-29-2018 19:30


3. Decorator

The Decorator pattern is a structural design pattern that allows you to dynamically attach new functionalities to an object by wrapping them in useful wrappers.

No wonder this design pattern is also called the Wrapper design pattern. This name describes more precisely the core idea behind this pattern: you place a target object inside another wrapper object that triggers the basic behavior of the target object and adds its own behavior to the result.

Both objects share the same interface, so it doesn’t matter for a user which of the objects they interact with − clean or wrapped. You can use several wrappers simultaneously and get the combined behavior of all these wrappers.

You should use the Decorator design pattern :

  • when you want to add responsibilities to objects dynamically and conceal those objects from the code that uses them;
  • when it’s impossible to extend responsibilities of an object through inheritance.

Example :

Imagine you need to implement data management in your iOS application. You could create two decorators: EncryptionDecorator for encrypting and decrypting data and EncodingDecorator for encoding and decoding.

// Design Patterns: Decorator

import Foundation

// Helpers (may be not include in blog post)

func encryptString(_ string: String, with encryptionKey: String) -> String {
    let stringBytes = [UInt8](string.utf8)
    let keyBytes = [UInt8](encryptionKey.utf8)
    var encryptedBytes: [UInt8] = []
    
    for stringByte in stringBytes.enumerated() {
        encryptedBytes.append(stringByte.element ^ keyBytes[stringByte.offset % encryptionKey.count])
    }
    
    return String(bytes: encryptedBytes, encoding: .utf8)!
}

func decryptString(_ string: String, with encryptionKey: String) -> String {
    let stringBytes = [UInt8](string.utf8)
    let keyBytes = [UInt8](encryptionKey.utf8)
    var decryptedBytes: [UInt8] = []
    
    for stringByte in stringBytes.enumerated() {
        decryptedBytes.append(stringByte.element ^ keyBytes[stringByte.offset % encryptionKey.count])
    }
    
    return String(bytes: decryptedBytes, encoding: .utf8)!
}

// Services

protocol DataSource: class {
    func writeData(_ data: Any)
    func readData() -> Any
}

class UserDefaultsDataSource: DataSource {
    private let userDefaultsKey: String
    
    init(userDefaultsKey: String) {
        self.userDefaultsKey = userDefaultsKey
    }
    
    func writeData(_ data: Any) {
        UserDefaults.standard.set(data, forKey: userDefaultsKey)
    }
    
    func readData() -> Any {
        return UserDefaults.standard.value(forKey: userDefaultsKey)!
    }
}

// Decorators

class DataSourceDecorator: DataSource {
    let wrappee: DataSource
    
    init(wrappee: DataSource) {
        self.wrappee = wrappee
    }
    
    func writeData(_ data: Any) {
        wrappee.writeData(data)
    }
    
    func readData() -> Any {
        return wrappee.readData()
    }
}

class EncodingDecorator: DataSourceDecorator {
    private let encoding: String.Encoding
    
    init(wrappee: DataSource, encoding: String.Encoding) {
        self.encoding = encoding
        super.init(wrappee: wrappee)
    }
    
    override func writeData(_ data: Any) {
        let stringData = (data as! String).data(using: encoding)!
        wrappee.writeData(stringData)
    }
    
    override func readData() -> Any {
        let data = wrappee.readData() as! Data
        return String(data: data, encoding: encoding)!
    }
}

class EncryptionDecorator: DataSourceDecorator {
    private let encryptionKey: String
    
    init(wrappee: DataSource, encryptionKey: String) {
        self.encryptionKey = encryptionKey
        super.init(wrappee: wrappee)
    }
    
    override func writeData(_ data: Any) {
        let encryptedString = encryptString(data as! String, with: encryptionKey)
        wrappee.writeData(encryptedString)
    }
    
    override func readData() -> Any {
        let encryptedString = wrappee.readData() as! String
        return decryptString(encryptedString, with: encryptionKey)
    }
}

// Usage

var source: DataSource = UserDefaultsDataSource(userDefaultsKey: "decorator")
source = EncodingDecorator(wrappee: source, encoding: .utf8)
source = EncryptionDecorator(wrappee: source, encryptionKey: "secret")
source.writeData("Design Patterns")
source.readData() as! String

// Result:
// Design Patterns


4. Facade

Facade is a structural design pattern that provides a simplified interface to a library, a framework, or any other complex set of classes.

Imagine that your code has to deal with multiple objects of a complex library or framework. You need to initialize all these objects, keep track of the right order of dependencies, and so on. As a result, the business logic of your classes gets intertwined with implementation details of other classes. Such code is difficult to read and maintain.

The Facade pattern provides a simple interface for working with complex subsystems containing lots of classes. The Facade pattern offers a simplified interface with limited functionality that you can extend by using a complex subsystem directly. This simplified interface provides only the features a client needs while concealing all others.

You should use the Facade design pattern:

  • when you want to provide a simple or unified interface to a complex subsystem;
  • when you need to decompose a subsystem into separate layers.

Example:

Lots of modern mobile applications support audio recording and playback, so let’s suppose you need to implement this functionality. You could use the Facade pattern to hide the implementation of services responsible for the file system (FileService), audio sessions (AudioSessionService), audio recording (RecorderService), and audio playback (PlayerService). The Facade provides a simplified interface for this rather complex system of classes.

// Design Patterns: Facade

import AVFoundation

// Services (may be not include in blog post)

struct FileService {
    private var documentDirectory: URL {
        return FileManager.default.urls(for: .documentDirectory, in: .userDomainMask).first!
    }
    
    var contentsOfDocumentDirectory: [URL] {
        return try! FileManager.default.contentsOfDirectory(at: documentDirectory, includingPropertiesForKeys: nil)
    }
    
    func path(withPathComponent component: String) -> URL {
        return documentDirectory.appendingPathComponent(component)
    }
    
    func removeItem(at index: Int) {
        let url = contentsOfDocumentDirectory[index]
        try! FileManager.default.removeItem(at: url)
    }
}

protocol AudioSessionServiceDelegate: class {
    func audioSessionService(_ audioSessionService: AudioSessionService, recordPermissionDidAllow allowed: Bool)
}

class AudioSessionService {
    weak var delegate: AudioSessionServiceDelegate?
    
    func setupSession() {
        try! AVAudioSession.sharedInstance().setCategory(AVAudioSessionCategoryPlayAndRecord, with: [.defaultToSpeaker])
        try! AVAudioSession.sharedInstance().setActive(true)
        
        AVAudioSession.sharedInstance().requestRecordPermission { [weak self] allowed in
            DispatchQueue.main.async {
                guard let strongSelf = self, let delegate = strongSelf.delegate else {
                    return
                }
                
                delegate.audioSessionService(strongSelf, recordPermissionDidAllow: allowed)
            }
        }
    }
    
    func deactivateSession() {
        try! AVAudioSession.sharedInstance().setActive(false)
    }
}

struct RecorderService {
    private var isRecording = false
    private var recorder: AVAudioRecorder!
    private var url: URL
    
    init(url: URL) {
        self.url = url
    }
    
    mutating func startRecord() {
        guard !isRecording else {
            return
        }
        
        isRecording = !isRecording
        recorder = try! AVAudioRecorder(url: url, settings: [AVFormatIDKey: kAudioFormatMPEG4AAC])
        recorder.record()
    }
    
    mutating func stopRecord() {
        guard isRecording else {
            return
        }
        
        isRecording = !isRecording
        recorder.stop()
    }
}

protocol PlayerServiceDelegate: class {
    func playerService(_ playerService: PlayerService, playingDidFinish success: Bool)
}

class PlayerService: NSObject, AVAudioPlayerDelegate {
    private var player: AVAudioPlayer!
    private var url: URL
    weak var delegate: PlayerServiceDelegate?
    
    init(url: URL) {
        self.url = url
    }
    
    func startPlay() {
        player = try! AVAudioPlayer(contentsOf: url)
        player.delegate = self
        player.play()
    }
    
    func stopPlay() {
        player.stop()
    }
    
    func audioPlayerDidFinishPlaying(_ player: AVAudioPlayer, successfully flag: Bool) {
        delegate?.playerService(self, playingDidFinish: flag)
    }
}

// Facade

protocol AudioFacadeDelegate: class {
    func audioFacadePlayingDidFinish(_ audioFacade: AudioFacade)
}

class AudioFacade: PlayerServiceDelegate {
    private let audioSessionService = AudioSessionService()
    private let fileService = FileService()
    private let fileFormat = ".m4a"
    private var playerService: PlayerService!
    private var recorderService: RecorderService!
    weak var delegate: AudioFacadeDelegate?
    
    private lazy var dateFormatter: DateFormatter = {
        let dateFormatter = DateFormatter()
        dateFormatter.dateFormat = "yyyy-MM-dd_HH:mm:ss"
        return dateFormatter
    }()
    
    init() {
        audioSessionService.setupSession()
    }
    
    deinit {
        audioSessionService.deactivateSession()
    }
    
    func startRecord() {
        let fileName = dateFormatter.string(from: Date()).appending(fileFormat)
        let url = fileService.path(withPathComponent: fileName)
        recorderService = RecorderService(url: url)
        recorderService.startRecord()
    }
    
    func stopRecord() {
        recorderService.stopRecord()
    }
    
    func numberOfRecords() -> Int {
        return fileService.contentsOfDocumentDirectory.count
    }
    
    func nameOfRecord(at index: Int) -> String {
        let url = fileService.contentsOfDocumentDirectory[index]
        return url.lastPathComponent
    }
    
    func removeRecord(at index: Int) {
        fileService.removeItem(at: index)
    }
    
    func playRecord(at index: Int) {
        let url = fileService.contentsOfDocumentDirectory[index]
        playerService = PlayerService(url: url)
        playerService.delegate = self
        playerService.startPlay()
    }
    
    func stopPlayRecord() {
        playerService.stopPlay()
    }
    
    func playerService(_ playerService: PlayerService, playingDidFinish success: Bool) {
        if success {
            delegate?.audioFacadePlayingDidFinish(self)
        }
    }
}

// Usage

let audioFacade = AudioFacade()
audioFacade.numberOfRecords()

// Result:
// 0


5. Template Method

The Template Method pattern is a behavioral design pattern that defines a skeleton for an algorithm and delegates responsibility for some steps to subclasses. This pattern allows subclasses to redefine certain steps of an algorithm without changing its overall structure.

This design pattern splits an algorithm into a sequence of steps, describes these steps in separate methods, and calls them consecutively with the help of a single template method.

You should use the Template Method design pattern:

  • when subclasses need to extend a basic algorithm without modifying its structure;
  • when you have several classes responsible for quite similar actions (meaning that whenever you modify one class, you need to change the other classes).

Example:

Suppose you’re working on an iOS app that must be able to take and save pictures. Therefore, your application needs to get permissions to use the iPhone (or iPad) camera and image gallery. To do this, you can use the PermissionService base class that has a specific algorithm.

To get permission to use the camera and gallery, you can create two subclasses, CameraPermissionService and PhotoPermissionService, that redefine certain steps of the algorithm while keeping other steps the same.

// Design Patterns: Template Method

import AVFoundation
import Photos

// Services

typealias AuthorizationCompletion = (status: Bool, message: String)

class PermissionService: NSObject {
    private var message: String = ""
    
    func authorize(_ completion: @escaping (AuthorizationCompletion) -> Void) {
        let status = checkStatus()
        
        guard !status else {
            complete(with: status, completion)
            return
        }
        
        requestAuthorization { [weak self] status in
            self?.complete(with: status, completion)
        }
    }

    func checkStatus() -> Bool {
        return false
    }
    
    func requestAuthorization(_ completion: @escaping (Bool) -> Void) {
        completion(false)
    }
    
    func formMessage(with status: Bool) {
        let messagePrefix = status ? "You have access to " : "You haven't access to "
        let nameOfCurrentPermissionService = String(describing: type(of: self))
        let nameOfBasePermissionService = String(describing: type(of: PermissionService.self))
        let messageSuffix = nameOfCurrentPermissionService.components(separatedBy: nameOfBasePermissionService).first!
        message = messagePrefix + messageSuffix
    }
    
    private func complete(with status: Bool, _ completion: @escaping (AuthorizationCompletion) -> Void) {
        formMessage(with: status)
        
        let result = (status: status, message: message)
        completion(result)
    }
}

class CameraPermissionService: PermissionService {
    override func checkStatus() -> Bool {
        let status = AVCaptureDevice.authorizationStatus(for: .video).rawValue
        return status == AVAuthorizationStatus.authorized.rawValue
    }
    
    override func requestAuthorization(_ completion: @escaping (Bool) -> Void) {
        AVCaptureDevice.requestAccess(for: .video) { status in
            completion(status)
        }
    }
}

class PhotoPermissionService: PermissionService {
    override func checkStatus() -> Bool {
        let status = PHPhotoLibrary.authorizationStatus().rawValue
        return status == PHAuthorizationStatus.authorized.rawValue
    }
    
    override func requestAuthorization(_ completion: @escaping (Bool) -> Void) {
        PHPhotoLibrary.requestAuthorization { status in
            completion(status.rawValue == PHAuthorizationStatus.authorized.rawValue)
        }
    }
}

// Usage

let permissionServices = [CameraPermissionService(), PhotoPermissionService()]

for permissionService in permissionServices {
    permissionService.authorize { (_, message) in
        print(message)
    }
}

// Result:
// You have access to Camera
// You have access to Photo

That’s all about in this article.


Conclusion

In this article, We understood about the five design patterns most frequently used in Swift. The ability to pick a design pattern in Swift that’s relevant for building a particular project allows you to build fully functional and secure applications that are easy to maintain and upgrade. You should certainly have design patterns in your skillset, as they not only simplify software development but also optimize the whole process and ensure high code quality.

Thanks for reading ! I hope you enjoyed and learned about the most frequently used Design Patterns in Swift. Reading is one thing, but the only way to master it is to do it yourself.

Please follow and subscribe us on this blog and and support us in any way possible. Also like and share the article with others for spread valuable knowledge.

If you have any comments, questions, or think I missed something, feel free to leave them below in the comment box.

Thanks again Reading. HAPPY READING !!???

iOS – Is SOLID Valuable Design Principles Applicable For Swift ?

Hello Readers, CoolMonkTechie heartily welcomes you in this article.

In this article, We will learn about most popular design principles SOLID in Swift. We will see that how SOLID is applicable for Swift. Now a days , a Maintainable and Reusable component is just a dream. Maybe not. SOLID principles, may be the way.

A famous quote about learning is :

Change is the end result of all true learning.

So Let’s begin.

Origin of the acronym SOLID

SOLID is an acronym named by Robert C. Martin (Uncle Bob). It represents 5 principles of object-oriented programming :

  • Single responsibility Principle
  • Open/Closed Principle
  • Liskov Substitution Principle
  • Interface Segregation Principle
  • Dependency Inversion Principle

If we apply these five principles:

  • We will have flexible code, which we can easily change and that will be both reusable and maintainable.
  • The software developed will be robust, stable and scalable (we can easily add new features).
  • Together with the use of the Design Patterns, it will allow us to create software that is highly cohesive (that is, the elements of the system are closely related) and loosely coupled (the degree of dependence between elements is low).

So, SOLID can solve the main problems of a bad architecture:

  • Fragility: A change may break unexpected parts—it is very difficult to detect if you don’t have a good test coverage.
  • Immobility: A component is difficult to reuse in another project—or in multiple places of the same project—because it has too many coupled dependencies.
  • Rigidity: A change requires a lot of efforts because affects several parts of the project.

Of course, as Uncle Bob pointed out in a his article, these are not strict rules, but just guidelines to improve the quality of your architecture.

Principles will not turn a bad programmer into a good programmer. Principles have to be applied with judgement. If they are applied by rote it is just as bad as if they are not applied at all.

Principles

The Single Responsibility Principle (SRP)

According to this principle, a class should have a reason, and only one, to change. That is, a class should only have one responsibility.

Now let’s describe Single Responsibility Principle says :

THERE SHOULD NEVER BE MORE THAN ONE REASON FOR A CLASS TO CHANGE.

Every time you create/change a class, you should ask yourself: How many responsibilities does this class have?

Let’s take a look into Swifty communication program.

import Foundation

class InterPlanetMessageReceiver {

    func receiveMessage() {
        print("Received the Message!")
    }

    func displayMessageOnGUI() {
      print("Displaying Message on Screen!")
    }
}

Now let’s understand what is Single Responsibility Principle (SRP) and how the above program doesn’t obey it.

SRP says, “Just because you can implement all the features in a single device, you shouldn’t”.

In Object Oriented terms it means: There should never be more than one reason for a class to change. It doesn’t mean you can’t have multiple methods but the only condition is that they should have one single purpose.

Why? Because it adds a lot of manageability problems for you in the long run.

Here, the InterPlanetMessageReceiver class does the following:

  • It receives the message.
  • It renders it on UI.

And, two applications are using this InterPlanetMessageReceiver class:

  • A messaging application uses this class to receive the message
  • A graphical application uses this class to draw the message on the UI

Do you think it is violating the SRP?

YES, The InterPlanetMessageReceiver class is actually performing two different things. First, it handles the messaging, and second, displaying the message on GUI. This causes some interesting problems:

  • Swifty must include the GUI into the messaging application and also while deploying the messaging application, we must include the GUI library.
  • A change to the InterPlanetMessageReceiver class for the graphical application may lead to a change, build, and test for the messaging application, and vice-versa.

Swifty got frustrated with the amount of changes it required. He thought it would be a minute job but now he has already spent hours on it. So he decided do make a change into his program and fix this dependency.

This is what Swifty came up with

import Foundation

// Handles received message
class InterPlanetMessageReceiver {

    func receive() {
        print("Received the Message!")
    }
}

// Handles the display part
class InterPlanetMessageDisplay {

    func displayMessageOnGUI() {
      print("Displaying Message on Screen!")
    }
}

Here’s how Swifty explained this:

InterPlanetMessageReceiver class will be used by the messaging application, and the InterPlanetMessageDisplay class will be used by the graphical application. We could even separate the classes into two separate files, and that will allow us not to touch the other in case a change is needed to be implemented in one.

Finally, Swifty noted down :Why we need SRP?

  • Each responsibility is an agent of change.
  • Code becomes coupled if classes have more than one responsibility.

Open/Closed Principle

According to this principle, we must be able to extend the a class without changing its behaviour. This is achieved by abstraction.

Now let’s describe Open/Closed Principle says :

SOFTWARE ENTITIES (CLASSES, MODULES, FUNCTIONS, ETC.) SHOULD BE OPEN FOR EXTENSION, BUT CLOSED FOR MODIFICATION.

If you want to create a class easy to maintain, it must have two important characteristics:

  • Open for extension: You should be able to extend or change the behaviours of a class without efforts.
  • Closed for modification: You must extend a class without changing the implementation.

Let’s see our swifty example. Swifty was quite happy with these change and later he celebrated it with a drink in Swiftzen’s best pub and there his eyes fell upon an artifact hanging on the front wall and he found all the symbols he received in the message. Quickly, he opened his diary and completed deciphering all those shapes.

Next day when he returned back, he thought why not fix the DrawGraphic class which draws only circle shape, to include the rest of the shapes and display the message correctly.

// This is the DrawGraphic
class DrawGraphic {

  func drawShape() {
     print("Circle is drawn!")
  }
}

// Updated Class code

enum Shape {
  case circle
  case rectangle
  case square
  case triangle
  case pentagon
  case semicircle
}
class circle {

}

// This is the DrawGraphic
class DrawGraphic {

  func drawShape(shape: Shape) {
     switch shape {
        case .circle:
          print("Circle is drawn")
        case .rectangle:
          print("Rectangle is drawn")
        case square:
          print("Square is drawn")
        case triangle:
          print("Triangle is drawn")
        case pentagon:
          print("Pentagon is drawn")
        case semicircle:
          print("Semicircle is drawn")
        default:
          print("Shape not provided")
     }
  }
}

Swifty was not happy with these changes, what if in future a new shape shows up, after all he saw in the artifacts that there were around 123 shapes. This class will become one fat class. Also, DrawGraphics class is used by other applications and so they also have to adapt to this change. it was nightmare for Swifty.

Open Closed Principle solves nightmare for Swifty. At the most basic level, this means, you should be able to extend a class behavior without modifying it. It’s just like I should be able to put on a dress without doing any change to my body. Imagine what would happen if for every dress I have to change my body.

After hours of thinking, Swifty came up with below implementation of DrawGraphic class.

protocol Draw {
    func draw()
}

class Circle: Draw {
    func draw() {
        print("Circle is drawn!")
    }
}

class Rectangle: Draw {
    func draw() {
        print("Rectangle is drawn!")
    }
}

class DrawGraphic {
    func drawShape(shape: Draw) {
        shape.draw()
    }
}

let circle = Circle()
let rectangle = Rectangle()
let drawGraphic = DrawGraphic()

drawGraphic.drawShape(shape: circle)  // Circle is drawn!
drawGraphic.drawShape(shape: rectangle) // Rectangle is drawn!

Since the DrawGraphic is responsible for drawing all the shapes, and because the shape design is unique to each individual shape, it seems only logical to move the drawing for each shape into its respective class.

That means the DrawGraphic still have to know about all the shapes, right? Because how does it know that the object it’s iterating over has a draw method? Sure, this could be solved with having each of the shape classes inherit from a protocol: the Draw protocol (this can be an abstract class too).

Circle and Rectangle classes holds a reference to the protocol, and the concrete DrawGraphic class implements the protocol Draw class. So, if for any reason the DrawGraphic implementation is changed, the Circle and Rectangle classes are not likely to require any change or vice-versa.

Liskov Subsitution Principle

This principle, introduced by Barbara Liskov in 1987, states that in a program any class should be able to be replaced by one of its subclasses without affecting its functioning.

Now let’s describe Liskov Substitution Principle says :

FUNCTIONS THAT USE POINTERS OR REFERENCES TO BASE CLASSES MUST BE ABLE TO USE OBJECTS OF DERIVED CLASSES WITHOUT KNOWING IT.

Inheritance may be dangerous and you should use composition over inheritance to avoid a messy codebase. Even more if you use inheritance in an improper way.

This principle can help you to use inheritance without messing it up.

Let’s see our swifty example. Swifty was implementing the SenderOrigin class to know whether the sender is from a Planet or not.

The Sender class looked something like this

Class Planet {
  func orbitAroundSun() {
  }
}

class Earth: Planet {
  func description() {
    print("It is Earth!")
  }
}

class Pluto: Planet {
  func description() {
    print("It is Pluto!")
  }
}

class Sender {
  func senderOrigin(planet: Planet) {
    planet.description()
  }
}

In the class design, Pluto should not inherit the Planet class because it is a dwarf planet, and there should be a separate class for Planet that has not cleared the neighborhood around its orbit and Pluto should inherit that.

So the principle says that Objects in a program should be replaceable with instances of their subtypes without altering the correctness of that program.

Swifty whispered it is the polymorphism. Yes it is. “Inheritance” is usually described as an “is a” relationship. If a “Planet” is a “Dwarf”, then the “Planet” class should inherit the “Dwarf” class. Such “Is a” relationships are very important in class designs, but it’s easy to get carried away and end up with a wrong design and a bad inheritance.

The “Liskov’s Substitution Principle” is just a way of ensuring that inheritance is used correctly.

In the above case, both Earth and Pluto can orbit around the Sun but Pluto is not a planet. It has not cleared the neighborhood around its orbit. Swifty understood this and changed the program.

class Planet {
    func oribitAroundSun() {
        print("This planet Orbit around Sun!")
    }
}

class Earth: Planet {
    func description() {
        print("Earth")
    }
}

class DwarfPlanet: Planet {
    func notClearedNeighbourhoodOrbit() {

    }
}

class Pluto: DwarfPlanet {
  func description() {
        print("Pluto")
    }
}

class Sender {
    func senderOrigin(from: Planet) {
        from.description()
    }
}

let pluto = Pluto()
let earth = Earth()

let sender = Sender()
sender.senderOrigin(from: pluto) // Pluto
sender.senderOrigin(from: earth) // Earth

Here, Pluto inherited the planet but added the notClearedNeigbourhood method which distinguishes a dwarf and regular planet.

  • If LSP is not maintained, class hierarchies would be a mess, and if a subclass instance was passed as parameter to methods, strange behavior might occur.
  • If LSP is not maintained, unit tests for the base classes would never succeed for the subclass.

Swifty can design objects and apply LSP as a verification tool to test the hierarchy whether inheritance is properly done.

Interface Segregation Principle

The Principle of segregation of the interface indicates that it is better to have different interfaces (protocols) that are specific to each client, than to have a general interface. In addition, it indicates that a client would not have to implement methods that he does not use.

Now let’s describe Interface Segragation Principle says :

CLIENTS SHOULD NOT BE FORCED TO DEPEND UPON INTERFACES THAT THEY DO NOT USE.

This principle introduces one of the problems of object-oriented programming: the fat interface.

An interface is called “fat” when has too many members/methods, which are not cohesive and contains more information than we really want. This problem can affect both classes and protocols.

Let’s continue our swifty example. Swifty was quite astonished with the improvement in his program. All the changes were making more sense. Now, it was time to share this code with different planet. Swiftzen 50% GDP was dependent on selling softwares and many planet has requested and signed MOU for the Inter Planet communication system.

Swifty was ready to sell the program and but he was not satisfied with current client interface. Let’s us look into it.

protocol interPlanetCommunication {
  func switchOnAntenna()
  func setAntennaAngle()
  func transmitMessage()
  func receivedMessage()
  func displayMessageOnGUI()
}

Now for anyone who want to use interPlanetCommunication, he has to implement all the five methods even-though they might not required.

So the principle says that Many client-specific interfaces are better than one general purpose interface. The principle ensures that Interfaces are developed so that each of them have their own responsibility and thus they are specific, easily understandable, and re-usable.

Swifty quickly made changes to his program interface:

protocol antenna {
  func switchOnAntenna()
  func setAntennaAngle()
}

protocol message {
  func transmitMessage()
  func receivedMessage()
}

protocol dispaly {
  func displayMessageOnGUI()
}

Dependency Inversion Principle

According to the Dependency inversion principle:

“HIGH LEVEL MODULES SHOULD NOT DEPEND UPON LOW LEVEL MODULES. BOTH SHOULD DEPEND UPON ABSTRACTIONS.”

“ABSTRACTIONS SHOULD NOT DEPEND UPON DETAILS. DETAILS SHOULD DEPEND UPON ABSTRACTIONS.”

This principle tries to reduce the dependencies between modules, and thus achieve a lower coupling between classes.

This principle is the right one to follow if you believe in reusable components.

DIP is very similar to Open-Closed Principle: the approach to use, to have a clean architecture, is decoupling the dependencies. You can achieve it thanks to abstract layers.

Let’s continue our swifty example. Before finally shipping the program to all the clients, Swifty was trying to fix the password reminder class which looks like this.

class PasswordReminder {
  func connectToDatabase(db: SwiftZenDB) {
    print("Database Connected to SwiftzenDB")
  }

  func sendReminder() {
    print("Send Reminder")
  }
}

PasswordReminder class is dependent on a lower level module i.e. database connection. Now, let suppose that you want to change the database connection from Swiftzen to Objective-Czen, you will have to edit the PasswordReminder class.

Finally the last principle states that Entities must depend on abstractions not on concretions.

To fix above program Swifty made these changes:

protocol DBConnection {
  func connection()
}

class SwiftzenDB: DBConnection {
  func connection() {
    print("Connected to SwiftzenDB")
  }
}

class PasswordReminder {
  func connectToDatabase(db: DBConnection) {
    db.connection()
  }

  func sendReminder() {
    print("Send Reminder")
  }
}

The DBConnection protocol has a connection method and the SwiftzenDB class implements this protocol, also instead of directly type-hinting SwiftzenDB class in PasswordReminder, Swifty instead type-hint the protocol and no matter the type of database your application uses, the PasswordReminder class can easily connect to the database without any problems and OCP is not violated.

The point is rather than directly depending on the SwiftzenDB, the passwordReminder depends on the abstraction of some specification of Database so that if any the Database conforms to the abstraction, it can be connection with the PasswordReminder and it will work.

That’s all about in this article.

Conclusion

In this article, We understood about SOLID principles in Swift. We learnt that how SOLID is application for Swift. If we follow SOLID principles judiciously, we can increase the quality of our code. Moreover, our components can become more maintainable and reusable.

The mastering of these principles is not the last step to become a perfect developer, actually, it’s just the beginning. We will have to deal with different problems in our projects, understand the best approach and, finally, check if we are breaking some principles.

We have 3 enemies to defeat: Fragility, Immobility and Rigidity. SOLID principles are our weapons. We tried to explain the SOLID concepts in Swift easy way with examples.

Thanks for reading ! I hope you enjoyed and learned about SOLID Principles in Swift. Reading is one thing, but the only way to master it is to do it yourself.

Please follow and subscribe us on this blog and and support us in any way possible. Also like and share the article with others for spread valuable knowledge.

If you have any comments, questions, or think I missed something, feel free to leave them below in the comment box.

Thanks again Reading. HAPPY READING !!???

iOS – Understanding KVO In Swift

Hello Readers, CoolMonkTechie heartily welcomes you in this article.

In this article, We will learn about KVO in Swift. KVO, which stands for Key-Value Observing, is one of the techniques for observing the program state changes available in Objective-C and Swift. This article demonstrates the KVO importance in Objective-C and Swift with practical example.

A famous quote about learning is :

” One learns from books and example only that certain things can be done. Actual learning requires that you do those things.”

So Let’s begin.


Overview

The KVO concept is simple:

” When we have an object with some instance variables, KVO allows other objects to establish surveillance on changes for any of those instance variables. “

KVO is a practical example of the Observer pattern. What makes Objective-C (and Obj-C bridged Swift) unique is that every instance variable that you add to the class becomes observable through KVO right away!

But in the majority of other programming languages, such a tool doesn’t come out of the box – we usually need to write additional code in the variable’s setter to notify the observers about the value changes.

Swift has inherited KVO from Objective-C, so for a full picture we need to understand how KVO works in Objective-C.


KVO in Objective-C

Consider we have a class named Person with properties name and age

@interface Person: NSObject
 
@property (nonatomic, strong) NSString *name;
@property (nonatomic, assign) NSInteger age;
 
@end

The objects of this class now are able to communicate the changes of the properties through KVO, but with no additional code – this feature comes for free !

So the only thing we need to do is to start the observation in another class:

@implementation SomeOtherClass

- (void)observeChanges:(Person *)person {
    [person addObserver:self
             forKeyPath:@"age"
                options:NSKeyValueObservingOptionNew
                context:nil];
}

- (void)observeValueForKeyPath:(NSString *)keyPath
                      ofObject:(id)object
                        change:(NSDictionary<NSKeyValueChangeKey,id> *)change
                       context:(void *)context {
    if ([keyPath isEqualToString:@"age"]) {
        NSNumber *ageNumber = change[NSKeyValueChangeNewKey];
        NSInteger age = [ageNumber integerValue];
        NSLog(@"New age is: %@", age);
    }
}

@end

That’s it! Now every time the age property changes on the Person we’ll have New age is: ... printed to the log from the observer’s side.

As we can see, there are two methods involved in KVO communication.

The first is addObserver:forKeyPath:options:context:, which can be called on any NSObject, including Person. This method attaches the observer to an object.

The second is observeValueForKeyPath:ofObject:change:context: which is another standard method in NSObject that we have to override in our observer’s class. This method is used for handling the observation notifications.

There is a third method, removeObserver:forKeyPath:context:, which allows us to stop the observation. It’s important to unsubscribe from notifications if the observed object outlives the observer. So the subscription just has to be removed in the observer’s dealloc method.

Now, let’s talk about the parameters of the methods used in KVO.

- (void)addObserver:(NSObject *)observer 
         forKeyPath:(NSString *)keyPath
            options:(NSKeyValueObservingOptions)options
            context:(nullable void *)context;

Here, The method have the following parameters:

  • observer is the object that will be receiving the change notifications. Usually, we provide self in this parameter, as the addObserver: is called from inside own instance method.
  • keyPath is a string parameter that in simplest case is just the name of the property we want to observe. If the property references a complex object hierarchy it can be a set of property names for digging into that hierarchy: "person.father.age"
  • options is an enum that allows for customizing what information is delivered with notification and when it should be sent. Available options are NSKeyValueObservingOptionNew and NSKeyValueObservingOptionOld, which control whether to include the most recent and the previous values respectively. There is also NSKeyValueObservingOptionInitial for triggering the notification right after the subscription, and NSKeyValueObservingOptionPrior for diffing the changes in a collection, such as insertions of deletions in NSArray.
  • context is a reference to object of an arbitrary class, which can be helpful for identifying the subscription in certain complex use cases, such as when working with CoreData. In most other cases we simply provide nil here.

The method we used for handling the update notifications

- (void)observeValueForKeyPath:(NSString *)keyPath
                      ofObject:(id)object
                        change:(NSDictionary<NSKeyValueChangeKey,id> *)change
                       context:(void *)context

Here, The method have the following parameters :

  • keyPath is the same string value we provided when attaching the observer. We may ask why it is provided here as well. The reason is that we may be observing multiple properties at once, so this parameter can be used to distinguish the notifications for one property from another.
  • object is the observed object. Since we can observe changes on more than one object, this parameter allows us to identify who’s property has changed.
  • change is the dictionary with information about the changed value. Based on the NSKeyValueObservingOptions we provided upon subscription, this dictionary may contain the current value under key NSKeyValueChangeNewKey, previous value for NSKeyValueChangeOldKey, and the “diff” information when observing changes in a collection: NSKeyValueChangeIndexesKey and NSKeyValueChangeKindKey
  • context is the reference provided upon subscription. Again, used for proper observation identification and in most cases can be ignored.


When KVO does not work

Even though KVO looks like magic, there is nothing extraordinary behind it. In fact, we can have direct access to its internals, which are hidden by default.

The trick is how Objective-C generates setter for properties. When we declare a property like

@property (nonatomic, assign) NSInteger age;

The factual setter generated by Objective-C is equivalent to the following:

- (void)setAge:(NSInteger)age {
    [self willChangeValueForKey:@"age"];
    _age = age;
    [self didChangeValueForKey:@"age"];
}

And if we explicitly define the setter without calling these willChangeValueForKey and didChangeValueForKey.

- (void)setAge:(NSInteger)age {
    _age = age;
}

… the KVO will stop working for this property.

So basically, these two methods willChangeValueForKey and didChangeValueForKey allow KVO to deliver the updates to the subscribers, and the developer can opt-out by omitting those calls from the setter.

It is important to understand that every @property synthesised by Objective-C adds a hidden instance variable with _ prefix.

For example, @property NSInteger age; generates an instance variable with the name _age that can be accessed just like the property:

self.age = 25;
self._age = 25;

The difference is that self.age = 25; triggers setter setAge:, while self._age = 25; changes the stored variable directly.

This means that even if the KVO is enabled for the age property, the KVO communication will work correctly for self.age = 25; and won’t deliver an update for self._age = 25;

Another way to break free from KVO is to not use @property in the first place, but instead store the instance variable in the anonymous category of the class:

@interface Person () {
    NSInteger _privateVariable;
}
@end

For such variables, Objective-C does not generate setter and getter, thus not enabling KVO.


KVO in Swift

Swift has inherited the support for the KVO from Objective-C, but unlike the latter, KVO is disabled in Swift classes by default.

Objective-C classes used in Swift keep KVO enabled, but for a Swift class we need to set the base class to NSObject plus add @objc dynamic attributes to the variables:

class Person: NSObject {
    @objc dynamic var age: Int
    @objc dynamic var name: String
}

There are two APIs available in Swift for Key-Value Observing: the old one, which came from Objective-C, and the new one, which is more flexible, safe and Swift-friendly.

Let’s start with the new one:

class PersonObserver {

    var kvoToken: NSKeyValueObservation?
    
    func observe(person: Person) {
        kvoToken = person.observe(\.age, options: .new) { (person, change) in
            guard let age = change.new else { return }
            print("New age is: \(age)")
        }
    }
    
    deinit {
        kvoToken?.invalidate()
    }
}

As we can see, the new API is using a closure callback for delivering the change notification right in the place where the subscription started.

This is more convenient and safe because we no longer need to check the keyPathobject or context, – no other notifications are delivered in that closure, just the one we’ve subscribed on.

There is a new way for managing the observation lifetime – the act of subscribing returns a token of type NSKeyValueObservation which has to be stored somewhere, for example, in an instance variable of the observer class.

Later on, we can call invalidate() on that token to stop the observation, like in the deinit method above.

The final change is related to the keyPath. String was error-prone because when we rename a variable the compiler won’t be able to tell us that the keyPath now leads to nowhere. Instead, this new API is using Swift’s special type for keyPath, which allows the compiler to verify the path is valid.

The options parameter has just the same set of options as in Objective-C. If we need to provide more than one option, we just bundle them in an array: options: [.new, .old]

The old API is also available, although it maintained all its disadvantages, so we encourage you to use the new API instead.

Here is the old one:

class PersonObserver: NSObject {
    
    func observe(person: Person) {
        person.addObserver(self, forKeyPath: "age",
                           options: .new, context: nil)
    }
    
    override func observeValue(forKeyPath keyPath: String?,
                               of object: Any?,
                               change: [NSKeyValueChangeKey : Any]?,
                               context: UnsafeMutableRawPointer?) {
        if keyPath == "age",
           let age = change?[.newKey] {
             print("New age is: \(age)")
        }
    }
}

The old API requires the observer to be an NSObject descendant as well. We also need to verify the keyPathobject, and context, since other notifications are also delivered in this method, just like in Objective-C.

That’s all about in this article.


Conclusion

In this article, We understood about KVO in Swift. This article described about the KVO importance in Objective-C and Swift with practical example.

Thanks for reading ! I hope you enjoyed and learned about the KVO Concepts in iOS. Reading is one thing, but the only way to master it is to do it yourself.

Please follow and subscribe us on this blog and and support us in any way possible. Also like and share the article with others for spread valuable knowledge.

If you have any comments, questions, or think I missed something, feel free to leave them below in the comment box.

Thanks again Reading. HAPPY READING!!???

Exit mobile version