Working with Functions in Swift
This chapter is from the book
This chapter is from the book
This chapter is from the book
This chapter discusses functions. You will find that Swift functions are based on the best implementations functions in other languages. Swift functions provide you with lots of flexibility to create parameters that are both “internal” and “external,” which jives well with Objective-C. Internal and external parameters allow you to have functions that are easy to read. You’ll be able to quickly read the name of a function and know exactly what it does. This is one excellent feature of Objective-C that has made its way into Swift.
A function itself can also be passed as a parameter of another function, also known as anonymous functions. This makes it easy to pass parameters around to different contexts. Functions can contain other functions; these are called closures and are discussed in Chapter 6, “Reusable Code: Closures.” Closures and functions go hand in hand.
A function groups commonly used code together so that it can be reused as many times as is needed. Say that you have a game in which a character jumps in the air, which is a super common functionality in a game. You would need to write the code to make the character jump. Jumping in games gets rather complicated, and you wouldn’t want to rewrite that code every time you wanted the character to jump; handling the jumping character that way would be messy and error prone. Instead, you could create a function to wrap all that jumping code into a nice little package of goodness. Then, instead of writing all that code again, you could just use your jump() function. This is like using a real-life button press to make something work. That button connects to all the functionality contained within the component you’re activating. You don’t necessarily have to know how it works; you just know that pressing the button will make it work.
When you think about writing Swift code, you have to realize that there is a lot of functionality that just works. You may never know how it was written or how many lines of code it took to write it. You just know that when you call it, it will work.
For example, calling countElements on the string "Hi" returns 2, which is the number of characters in the string. You didn’t have to write that function. It came with Swift. With Swift, you can write your own functions and then forget how you made them work. Once you’ve written jump(), you can call it to have your character jump.
Defining Functions
In Swift, a function is made up of three components: a name, parameters, and a return type. The syntax for this type of declaration is as follows:
func functionName(parameterName: parameterType) -> returnType {
//code
}
This syntax is very different from Objective-C method declarations. However, if you have ever used JavaScript, Python, C, or many other languages, then this syntax will be pretty familiar. You will find that while the structure of functions is different, there are parts that make it compatible with Objective-C.
Let’s look at some examples, starting with a function that takes no arguments and has no return values:
func sayHello() {
println("Hello!")
}
Here you write the keyword func and then name the function sayHello. You use () to house parameters, when you need them; for now you can leave these parentheses empty. You use curly brackets to contain the code that needs to run when the function is called. To call this function, you simply use its name followed by parentheses. You would call sayHello like this:
sayHello()
This is about as basic a function as you can create. You can go a step further and add an argument that allows the function to “say hello” to a specific person. To do that, you need to allow the function to take a single argument of type String that represents a name. That type of declaration might look like this:
func sayHello(name: String) {
println("Hello, \(name)!")
}
Now you’ve added a parameter to the function. That parameter is of type String and called name.
Next, you’ll create another implementation of your sayHello function that says “hello” to someone a certain number of times. This will give you a chance to look at how to declare a function with multiple parameters:
func sayHello(name: String, numberOfTimes: Int) {
for _ in 1...numberOfTimes {
sayHello(name)
}
}
This function declaration can be read as “a function named sayHello that takes two arguments of type String and type Int with no return value.” The syntax is almost identical to that of the single argument function, just with an extra comma added to separate the arguments. You are even using the same name for the function. In fact, you are calling the first implementation of sayHello within the new declaration. Now, if you wanted to use this function, here’s how it would look:
sayHello("Skip", 5)
//Hello Skip!
//Hello Skip!
//Hello Skip!
//Hello Skip!
//Hello Skip!
We’ll elaborate a bit more on how Swift differentiates between these declarations when we discuss function types later in the chapter in “Functions as Types,” but for now, we’re going to move on to adding a return type to the function implementations.
To add a return argument in the function declaration, you simply include the pointer arrow, ->, followed by the return type.
Return Types
Next you’ll create a function that returns its sum, which will also be of return type Int:
func sum(a: Int, b: Int) -> Int {
return a + b
}
This declaration can be read as “a function, sum, that takes two arguments of type Int and has a return value of type Int.” If you wanted to call the new function, it could look something like this:
let total = sum(14, 52) // total = 66
Returning a single value is just fine, but sometimes you want to return multiple values. In Objective-C, this problem is usually solved by creating an object class or by returning some sort of collection. These solutions would work in Swift as well, but there is a better way: You can return multiple arguments in Swift by using tuples, as described in the next section.
Multiple Return Values
Like Objective-C functions, Swift functions can return only one value. Unlike Objective-C, though, Swift lets you use tuples as values, which can be useful for packaging together multiple return values so that you can pass around multiple values as one value. Consider a situation where a function is required to return the sum of two numbers as well as the higher of the two. This means that you need the function to returns two values, both of type Int encapsulated in a tuple. Here’s what it looks like:
func sumAndCeiling(a: Int, b: Int) -> (Int, Int) {
let ceiling = a > b ? a : b
let sum = a + b
return (sum, ceiling)
}
You can declare multiple return values by encapsulating them in parentheses, separated by a comma. This is the syntax for a tuple. The preceding function can be read “a function named sumAndCeiling that takes two arguments of type Int and returns a tuple of type (Int, Int).” You can grab values from the returned tuple, from its indexes, like so:
let result = sumAndCeiling(4, 52) let sum = result.0 let ceiling = result.1
This is a good way to return multiple values within one function, but using indexes to access values can be confusing. It’s also not very pretty, and it’s hard to remember which is which. Imagine if someone decided to change the order of the tuples without reading how they were used. It wouldn’t be very smart or very nice, and it would severely mess things up. It’s more helpful to name the values within a tuple.
Here’s how you can modify the sumAndCeiling function with named values within the tuple:
func sumAndCeiling(a: Int, b: Int) -> (sum: Int, ceiling: Int) {
let ceiling = a > b ? a : b
let sum = a + b
return (sum, ceiling)
}
The syntax for a named tuple is almost identical to the syntax of a parameter. Adding named tuples is an easy way to create more readable code while dealing with fewer errors. Here’s a new implementation:
let result = sumAndCeiling(16, 103) let sum = result.sum // sum = 119 // result.sum == result.0 let ceiling = result.ceiling // ceiling = 103 // result.ceiling == result.1