-
Notifications
You must be signed in to change notification settings - Fork 472
Understanding Swift
This guide covers the main Swift concepts you'll need as you start building apps. It is meant to introduce each topic in a way that provides enough background to get you started and defers to the iBook on Swift for more in depth coverage.
- Apple's Official Swift Guide .
- Download the Swift 5 Playground to experiment with the Swift 5 syntax.
- You can also find more resources here .
It is common to start any tutorial for a new language with the Hello World example so we'll start out by showing how easy this is to do in Swift:
print("Hello World!") // Prints "Hello World!" to the output window in XcodeTo make this a little more personal, let's look at one of the handy features of Swift (called string interpolation) where we can insert values into strings using the \(...) syntax. Below, we've inserted name into the middle of our string:
let name: String = "Turi"
print("Hello \(name)!") // Prints "Hello Turi!"You'll notice in the example above we've used let to create the new variable for the name "Bob". In Swift, you'll choose from the following 2 options when creating a new variable:
-
let=> Useletwhen you are defining a constant (a value that will not change)let numberOfContinents: Int = 7 // Seven continents will not change so we've used "let"
-
var=> Usevarwhen you are defining a variable (a value that might change)var continentsVisited: Int = 2 // Continents visited increases over time so we've used "var"
In general, it is considered a best practice to use constants (let) whenever possible.
The two most common types of numbers you'll use in Swift are integers (Int) and doubles (Double):
Integers are whole numbers with no fractional component:
let minValue: Int = -42
let maxValue: Int = 55 Doubles can have a fractional component:
let pi: Double = 3.14159
let billAmount: Double = 10.25Strings represent a series of characters:
let hello: String = "Hello"
let world: String = "World"You can use string interpolation to construct new strings:
let helloWorld: String = "\(hello) \(world)" // "Hello World"Or you can use the + operator to combine strings:
let helloWorld: String = hello + " " + world // "Hello World"Boolean is a very simple type in Swift as it can only ever be true or false:
let swiftIsCool: Bool = true
let iMissObjectiveC: Bool = false Arrays store a list of values that must be of the same type. Below we've kept track of the previous bill amounts (which is a list of doubles):
// Notice how we've used "var" here since we want to append new items to the array
var previousBillAmounts: [Double] = [10.25, 21.32, 15.54]To add a new bill to the array:
previousBillAmounts.append(52.45) // Result: [10.25, 21.32, 15.54, 52.45]To check on how many bills there are in the array:
let count = previousBillAmounts.count // Result: 4Or to check the first bill amount in the array:
let firstBillAmount = previousBillAmounts[0] // Result: 10.25Much like you might use a real world dictionary to look up the definition for a word, you can use the dictionary data structure in Swift to create associations between keys (the word in the real world dictionary) and values (the definition in the real world dictionary).
This is easiest seen using an example. Let's say we want to keep track of the ages of people that are using our app. This would be best done using the following dictionary:
var people: [String: Int] = [
"Bob": 32,
"Cindy": 25
]Here, the key is the name of the person and the value is the person's age.
Later on if we want to find out Bob's age, we can look it up by doing:
let bobsAge = people["Bob"]If we get a new user we can easily add them to our dictionary using the following syntax:
people["Dan"] = 56In most of the examples in this guide, we are explicit with the types of our constants and variables. Explicit typing looks something like:
let name: String = "Bob" // Explicitly designating "name" to be of type "String"However, Swift is smart enough to infer the type for us in a lot of cases. The previous example can be shortened to:
let name = "Bob" // Swift infers "name" is of type "String" since "Bob" is a StringIn cases where Swift can infer the type, it's not necessary to be explicit with the types for your constants and variables. We just do it in this guide so that the examples are easier to follow.
Swift has two special "catch all types" that come in handy when a more specific type cannot be determined.
-
AnyObjectcan represent an instance of any class type. -
Anycan represent an instance of any type at all.
In general it's a best practice to be as specific with your types as possible and avoid the use of AnyObject and Any, but they become particularly helpful when interacting with Objective-C code that is less strict on typing.
Optionals is a very important concept in Swift and is meant to improve the safety of Swift code. By simply placing a question mark (?) after any type, it declares that variable to be optional. An optional type allows that variable to exist in one of the following two states:
- There is a value and it equals x
- There isn't a value at all
Let's look at an example to make this more clear. Consider the following 2 examples where we are trying to convert a String to an Int:
// Example 1 (Conversion succeeds)
let input: String = "123"
let optionalConvertedInput: Int? = Int(input) // optionalConvertedInput = 123
// Example 2 (Conversion fails - input is not a number)
let input: String = "123abc"
let optionalConvertedInput: Int? = Int(input) // optionalConvertedInput = nilSwift requires optionalConvertedInput to be of type Int? (or "optional Int") so that it is explicit that convertedInput might not contain a value (in the case when the conversion fails). If we were to declare convertedInput as simply Int, we'd get a compile error.
There's a handy syntax in Swift that you'll use quite often when working with optionals. If we wanted to use the value of optionalConvertedInput later on in our code, we'd have to first check to make sure it's not nil. We can do so using the following code:
if let convertedInput = optionalConvertedInput {
// Code that gets executed when optionalConvertedInput is NOT nil
} else {
// OptionalConvertedInput IS nil, do something else
}In other languages like Java, NullPointerException's (exceptions that are fired when a null reference is accessed) are a common source of crashes. Swift's Optionals goes a long way to reduce this type of programming error.
When using optionals in string interpolation, the value with either be Optional(...) or nil without being explicitly unwrapped. When the inner value is needed, the optional must be unwrapped.
let maybeString: String? = "value"
print("\(maybeString)") // prints -> Optional("value")
print("\(maybeString!)") // prints -> "value"Functions in Swift are very similar to other languages. The simplest function in Swift can be written as:
// This prints "Hello!" to the output window
func printHello() {
print("Hello!")
}
// Calls this function
printHello()We can extend this function to be a little more personable by taking in a name and returning the greeting instead of printing it out:
// Takes in a "personName" parameter of type String and returns the greeting as a String
func sayHello(personName: String) -> String {
// Implicitly returns when there is a single statement
"Hello \(personName)!"
}
// Calls this function
let greeting: String = sayHello("Bob")Things get a little more interesting when you start to have multiple parameters as Swift has the concept of external and local parameter names. An external parameter name is used to label arguments passed to a function call. A local parameter name is the name used in the implementation of the function.
// "to" + "and" are the external parameter names
// "person" + "anotherPerson" are the local parameter names
func sayHello(to person: String, and anotherPerson: String) -> String {
"Hello \(person) and \(anotherPerson)!"
}
// Calls this function using the "external" parameter names
let greeting: String = sayHello(to: "Bill", and: "Ted")If statements are very similar to other languages and can be expressed as:
let temperatureInFahrenheit: Int = 90
if temperatureInFahrenheit <= 32 {
print("It's very cold. Consider wearing a scarf.")
} else if temperatureInFahrenheit >= 86 {
print("It's really warm. Don't forget to wear sunscreen.")
} else {
print("It's not that cold. Wear a t-shirt.")
}The two most common types of loops you'll need in Swift are for loops and for-in loops.
For loops work well when you want to do something until a particular condition is met (in the case below until index >= 3):
// Simple for loop that prints "Hello" 3 times
for var index = 0; index < 3; index++ {
print("Hello")
}C-style for loops are deprecated and will be removed in future versions of Swift, more info available on the SE-0007 Proposal. For-in loops are preferred instead. The above example could be re-written as:
for _ in 0..<3 {
print("Hello")
}For-in loops come in really handy when you want to do something to each item in a collection (such as an array):
let names = ["Anna", "Alex", "Brian", "Jack"]
// Loops over each name in "names" and prints out a greeting for each person
for name in names {
print("Hello, \(name)!")
}Sometimes you want to loop over each item in an array and also keep track of the index of the item. Array's enumerated() method can help you achieve this:
let names = ["Anna", "Alex", "Brian", "Jack"]
for (index, value) in names.enumerated() {
print("Item \(index + 1): \(value)")
}Classes are the building blocks of your app’s code. You define properties and methods to add functionality to your classes by using the same syntax as for variables and functions.
Below you can find a Person class that is meant to show an example of the types of things you'll want to do when building your classes.
class Person {
// Custom initializer - takes in firstName and lastName
init(firstName: String, lastName: String) {
self.firstName = firstName
self.lastName = lastName
// Increment type property each time a new person is created
Person.numberOfPeople++
}
// *** Properties ***
// Stored Property - Stored as part of the current instance of the class
var firstName: String
var lastName: String
// Computed Property - computes "fullName" from "firstName" and "lastName"
var fullName: String {
get {
"\(firstName) \(lastName)"
}
}
// Type Property - Single instance for all instances of the class,
// similar to a static property in Java
static var numberOfPeople = 0
// *** Methods ***
// Instance Method
func greet() {
// Notice the use of "self" - self refers to the current instance and
// is similar to "this" in Java
print ("Hello \(self.firstName)")
}
// Type Method
class func printNumberOfPeople() {
print("Number of people = \(Person.numberOfPeople)")
}
}
// ... Using the Person Class ...
// Create a new instance of the Person class
let bob = Person(firstName: "Bob", lastName: "Smith")
// Call instance method
bob.greet() // Prints "Hello Bob"
// Accessing properies
print("Bob's first name is: \(bob.firstName)") // Prints "Bob's first name is: Bob"
print("Bob's full name is: \(bob.fullName)") // Prints "Bob's full name is: Bob Smith"
// Call type method
// Prints "Number of people = 1" (since we've only created one Person)
Person.printNumberOfPeople()Protocols are similar to interfaces in other languages. Think about a protocol as a contract. The contract includes a set of methods that must be implemented. Any classes that choose to implement a protocol sign this contract and implement all the methods that are in the protocol.
Let's say we have the following protocol (MyFriendlyGreeterProtocol) that defines 2 methods sayHello() and sayBye():
protocol MyFriendlyGreeterProtocol {
func sayHello()
func sayBye()
}Any classes that implement this protocol (you implement a protocol by adding its name after the class defintion as shown below), must implement both of these methods:
class MyEnglishPerson: MyFriendlyGreeterProtocol {
func sayHello() {
print("Hello!")
}
func sayBye() {
print("Bye!")
}
// ... other methods for this class ...
}
class MySpanishPerson: MyFriendlyGreeterProtocol {
func sayHello() {
print("Hola!")
}
func sayBye() {
print("Adios!")
}
// ... other methods for this class ...
}There is a lot more you can do with protocols as they form one of the key design patterns in iOS. The code above merely shows how to get started with the syntax for protocols.
Closures are self-contained blocks of code that can be passed around and used in your code. Closures in Swift are similar to blocks in Objective-C and lambdas in other programming languages.
Closures can capture and store references to any constants and variables from the context in which they are defined. This is known as "closing" over those constants and variables. Swift handles all of the memory management of capturing for you.
Here’s an example of a function called makeIncrementer, which contains a nested function called incrementer. The nested incrementer() function captures two values, runningTotal and amount, from its surrounding context. After capturing these values, incrementer is returned by makeIncrementer as a closure that increments runningTotal by amount each time it is called.
func makeIncrementer(forIncrement amount: Int) -> () -> Int {
var runningTotal = 0
func incrementer() -> Int {
runningTotal += amount
return runningTotal
}
return incrementer
}The return type of makeIncrementer is () -> Int. This means that it returns a function, rather than a simple value. The function it returns has no parameters, and returns an Int value each time it is called.
The makeIncrementer(forIncrement:) function defines an integer variable called runningTotal, to store the current running total of the incrementer that will be returned. This variable is initialized with a value of 0.
The makeIncrementer(forIncrement:) function has a single Int parameter with an external name of forIncrement, and a local name of amount. The argument value passed to this parameter specifies how much runningTotal should be incremented by each time the returned incrementer function is called.
makeIncrementer defines a nested function called incrementer, which performs the actual incrementing. This function simply adds amount to runningTotal, and returns the result.
When considered in isolation, the nested incrementer() function might seem unusual:
func incrementer() -> Int {
runningTotal += amount
return runningTotal
}The incrementer() function doesn’t have any parameters, and yet it refers to runningTotal and amount from within its function body. It does this by capturing a reference to runningTotal and amount from the surrounding function and using them within its own function body. Capturing by reference ensures that runningTotal and amount do not disappear when the call to makeIncrementer ends, and also ensures that runningTotal is available the next time the incrementer function is called.
Here’s an example of makeIncrementer in action:
// This example sets a constant called incrementByTen to refer to an incrementer function that
// adds 10 to its runningTotal variable each time it is called.
let incrementByTen = makeIncrementer(forIncrement: 10)
incrementByTen() // returns a value of 10
incrementByTen() // returns a value of 20
incrementByTen() // returns a value of 30Type casting changes the type of a particular instance to another compatible type. There are 3 ways to accomplish this with Swift:
-
Guaranteed conversion with
as=> This is the safest cast. It will never fail since the compiler can guarantee the cast will work. Use this when you are upcasting from a child class to its parent or doing something like1 as Float.// Guaranteed conversion as the compiler can verify this will succeed let myFloat = 1 as Float // Guaranteed conversion as upcasting from a type to its parent type is safe // UIView is a parent of UITableView let myView = myTableView as UIView
-
Conditional conversion with
as?=> This is a cautious cast. If the cast fails, it will returnnil. This is needed when downcasting from a parent type to a child type.// If myView is actually a tableView, the downcast will succeed, otherwise it will fail safely if let myTableView = myView as? UITableView { print("The downcast succeeded!") } else { print("The downcast failed!") }
-
Forced conversion with
as!=> This is a dangerous cast that you should avoid using. If the cast fails, this will crash your app. Use this cast carefully.// DANGEROUS: If myView is actually a tableView, the downcast will succeed // Otherwise it will crash the app let myTableView = myView as! UITableView
There are various places where you'll come across the need for using question marks in your code. The following examples are meant to capture the major types of use cases for the question mark operator that can cause confusion when first learning Swift.
When defining a type as optional:
// Designates that optionalName can either have a value or be nil
var optionalName: String? = "Hello"When safely accessing methods or properties (called optional chaining):
// When myInstance IS nil, x = nil
// When myInstance is NOT nil, x = myInstance.count
// If any part of the chain is nil, it stops evaluating and returns
// nil, ignoring everything to the right.
let x = myInstance?.countWhen type casting (called safe conditional conversion):
// If myView is actually a tableView the downcast will succeed, otherwise it will fail safely
if let myTableView = myView as? UITableView {
print("The downcast succeeded!")
} else {
print("The downcast failed!")
}When using the nil coalescing operator (??) to provide a default value:
let nickName: String? = nil // nickName is an optional that might be nil
let fullName: String = "Bob Smith"
// prints the nickName if it is NOT nil, otherwise prints fullName
let informalGreeting = "Hi \(nickName ?? fullName)" // Prints "Hi Bob Smith"There are various places you might come across an exclamation mark in Swift code. The following examples are meant to capture the major types of use cases for the exclamation mark operator that can cause confusion when first learning Swift.
When getting the actual value out of an optional (called unwrapping an optional):
let possibleString: String? = "An optional string."
// DANGEROUS: possibleString must NOT be nil or this will crash
let forcedString: String = possibleString!
// SAFE: Will only enter the if clause if possibleString is NOT nil
if let actualString = possibleString {
// do something with actualString
}
// SAFE: Generally preferred alternate syntax to "if let" that can exit early
guard let actualString = possibleString else {
// exit early or throw exception
}
// do something with actualStringWhen type casting (called forced conversion):
// DANGEROUS: If myView is actually a tableView, the downcast will succeed
// Otherwise it will crash the app
let myTableView = myView as! UITableViewWhen defining variables that are initially nil but get set soon afterwards and are guaranteed not to be nil after that (called implicitly unwrapped optionals):
let assumedString: String! = "An implicitly unwrapped optional string."
// no need for an exclamation mark since assumedString is an implicitly unwrapped optional
let implicitString: String = assumedStringA very useful inclusion of the Swift language is Property Observers. While the language might sound complex, it's actually a literal explanation of what it does. It allows you to observe properties and respond to impending or finished property changes. It's part of Apple's goal to make Swift a cleaner language because a developer merely has to keep any logic related to updating properties in one convenient place.
Let's look at an example.
Assume you have a Fitness tracking application and whenever a user enters their weight, you calculate and display their BMI to them.
Without property observers, and keeping in good practice (since BMI is calculated and not set in initialization), you might have a class like this -
Swift
class Profile {
let height: Double
var weight: Double
private var BMI: Double
init(weight: Double, height: Double) {
self.height = height
self.weight = weight
self.BMI = weight / (height*height)
}
func getBMI() -> Double {
self.BMI
}
func updateBMI() {
self.BMI = weight / (height*height)
}
} In a view where you ask the user to enter their new weight, you would call profile.updateBMI() after the new weight is set.
Isn't this a little tedious? It's almost a trap for developers. You have to actively remember that you should update the BMI every time you update the weight.
Thankfully, with property observers, you don't have to actively watch every property change as a developer. Your class becomes -
class Profile {
let height: Double
var weight: Double {
didSet {
updateBMI()
}
}
private var BMI: Double
init(weight: Double, height: Double) {
self.height = height
self.weight = weight
self.BMI = weight / (height*height)
}
func getBMI() -> Double {
self.BMI
}
func updateBMI() {
self.BMI = weight / (height*height)
}
} It might not look like much, but now whenever you update BMI, you don't have to worry about calling updateBMI() and to be honest, we can clean this up even more by making BMI private(set), changing Profile to a struct and adding a mutating func.
struct Profile {
let height: Double
var weight: Double {
didSet {
updateBMI()
}
}
private(set) var BMI: Double
mutating private func updateBMI() {
BMI = weight / (height * height)
}
} What is mutating you ask? Well, normally Struct's are read only. This is because in order for them to be fast, they need to take up space on Stack instead of Heap. A simple way to explain this is to imagine that Stack is writing with a Pen and Heap is writing with a Pencil. Some of you might think "well what's the big deal? they both write." But think about all the extra work that comes with a Pencil - You need an eraser, you have to clean the eraser off the paper and can end up using more material if you edit a lot. A pen, on the other hand, you just write (we're assuming you're not using an erasable pen or white-out) and it's set.
Mutating gives you an ability to say, "Hey, this Struct will be read only, but if I make it a variable, I might want to update its properties, so please give me the ability to edit that Struct".
There's a little more to this under the hood, but Structs tend to be pretty thread safe, so you can assume every Struct exists by itself and it's pretty safe to edit like this.
Other Property Observers
There are other property observers for every stage of a state change - willSet, set/get are included. willSet has the added ability of being able to compare newValue to the original value (called the property name).
A few hints about where didSet can be quite powerful is when dealing with network data and data sources. When a user has the ability to get objects from a server with a regular network call, a search and tapping a keyword, it can be useful if the var dataSource: [Models] has didSet to reload the data of the view. That way you won't have to do it manually.
A lot of the time when working with REST API's (like Instagram, Twitter, etc), the data that comes back will be JSON. JSON is a human readable data format (very similar to XML).
Below is an example of JSON that simulates the type of JSON you might get back when using an endpoint that returns movies and their ratings:
- An open curly brace means the start of a dictionary
- An open bracket means the start of an array
{
"status": "OK",
"movies": [
{
"title": "Whiplash",
"rating": 8.5
},
{
"title": "Feast",
"rating": 5.2
},
{
"title": "Kung Fury",
"rating": 7.1
}
]
}Below is an example of how'd you extract the movies and ratings out of that response:
// data returned from the network response will typically be of type
// NSData (which is a buffer of bytes)
let responseData: NSData = // ... some value retrieved from the network response ...
// Wrap our code in a do catch as our code might throw an exception which we need to handle
do {
// Start by converting the NSData to a dictionary - a dictionary for the entire response
if let responseDictionary = try NSJSONSerialization.JSONObjectWithData(responseData,
options:NSJSONReadingOptions(rawValue:0)) as? [String:AnyObject] {
// Dip inside the response to find the "movies" key and get the array of movies
if let movies = responseDictionary["movies"] as? [AnyObject] {
// Get each movie dictionary from the array of movies
for movie in movies {
// Use the movie "title" key and "rating" key to get their values
if let title = movie["title"] as? String {
if let rating = movie["rating"] as? Double {
print("Title:\(title), rating:\(rating)")
}
}
}
}
}
} catch {
print("Error parsing JSON")
}Xcode includes a very useful tool for learning Swift called "Playgrounds". It's very easy to create a new playground through Xcode.
Once inside of a playground, you can write Swift code and see it run immediately (without needing to build and run a project each time). This allows you to try out different syntax and test out your code before including it into your app.
We highly recommend checking out Playgrounds while you are learning Swift.