When you want to return a single value from a function, you write an arrow and a data type before the opening brace:
func isShinobi(name: String) -> Bool {
name == "Naruto" || name == "Sasuke" || name == "Sakura"
}
That compares a name against a list of known shinobi names, returning true or false.
But what if you need to return two or more values? Let's explore the options.
๐ซ Attempt 1 โ Using an Array
Say we want a function that sends back a character's first and last name:
func getCharacter() -> [String] {
["Monkey", "D. Luffy"]
}
let character = getCharacter()
print("Name: \(character[0]) \(character[1])")
This is problematic โ it's hard to remember what character[0] and character[1] actually represent. If we ever change the order of the array, character[0] might suddenly become the last name instead of the first, and nothing would warn us.
๐ซ Attempt 2 โ Using a Dictionary
func getCharacter() -> [String: String] {
[
"firstName": "Monkey",
"lastName": "D. Luffy"
]
}
let character = getCharacter()
print("Name: \(character["firstName", default: "Unknown"]) \(character["lastName", default: "Unknown"])")
Now we have meaningful names โ firstName and lastName instead of 0 and 1. But look at that print() call: even though we know both keys will exist, Swift doesn't know that, so we still need to provide default values just in case.
โ Attempt 3 โ Using a Tuple
Tuples let us put multiple pieces of data into a single variable, but unlike arrays and dictionaries, tuples have a fixed size and can hold a mix of different types.
func getCharacter() -> (firstName: String, lastName: String) {
(firstName: "Monkey", lastName: "D. Luffy")
}
let character = getCharacter()
print("Name: \(character.firstName) \(character.lastName)")
Breaking that down:
- The return type is
(firstName: String, lastName: String)โ a tuple containing two strings - Each value in the tuple has a name, and these aren't in quotes โ they're specific labels, not arbitrary dictionary keys
- Inside the function we send back a tuple matching exactly what we promised
- When calling
getCharacter(), we read the values using.firstNameand.lastName
๐ Tuples vs Dictionaries โ What's the Real Difference?
They might look similar, but they behave very differently:
- With a dictionary, Swift can't know ahead of time whether a key exists. We know
character["firstName"]will be there, but Swift doesn't โ so we must provide a default value. - With a tuple, Swift knows exactly what's available because the tuple's type says so.
character.firstNameeither compiles or it doesn't โ there's no chance of a typo likecharacter["First Name"]silently returning nothing. - A dictionary could contain hundreds of other keys alongside
firstNameโ a tuple can't. It must contain exactly what its type says, nothing more and nothing less.
๐ก Three More Things About Tuples
1. You don't need to repeat the names in return
Since Swift already knows the names from the function's return type, this works exactly the same as the named version above:
func getCharacter() -> (firstName: String, lastName: String) {
("Monkey", "D. Luffy")
}
2. Unnamed tuples use numerical indices
If a tuple's elements don't have names, access them with .0, .1, and so on:
func getCharacter() -> (String, String) {
("Monkey", "D. Luffy")
}
let character = getCharacter()
print("Name: \(character.0) \(character.1)")
Numerical indices also work on named tuples, but using names is almost always clearer.
3. You can pull a tuple apart into separate constants
Here's the long way โ store the tuple, then copy each part out individually:
func getCharacter() -> (firstName: String, lastName: String) {
(firstName: "Roronoa", lastName: "Zoro")
}
let character = getCharacter()
let firstName = character.firstName
let lastName = character.lastName
print("Name: \(firstName) \(lastName)")
Or skip the middle step entirely and destructure the tuple straight from the function call:
let (firstName, lastName) = getCharacter()
print("Name: \(firstName) \(lastName)")
And if you only need part of the tuple, use _ to ignore the rest:
let (firstName, _) = getCharacter()
print("Name: \(firstName)")
When Should You Use an Array, a Set, or a Tuple? ๐ค
Now that we've covered arrays, sets, and tuples individually, here's how to pick the right one. Remember the key differences:
- Arrays keep order and can have duplicates
- Sets are unordered and can't have duplicates
- Tuples have a fixed number of values, each with a fixed type
Here's how that plays out with some examples:
| Scenario | Best Choice | Why |
|---|---|---|
| A list of all jutsu names a player has unlocked, where no jutsu can be unlocked twice and order doesn't matter | Set | No duplicates, order irrelevant |
| Episodes of a show a user has watched | Set or Array | Set if you only care whether they watched it; Array if the order watched matters |
| High scores in a game | Array | Order matters, and duplicate scores are possible if two players tie |
| Items on a to-do list | Array | Predictable order is important |
| A character's exact stats โ say, two integers for strength and speed, plus a Boolean for whether they've unlocked their final form | Tuple | A fixed number of values with fixed, specific types |
Wrap Up ๐ฌ
- Returning multiple values from a function? Skip arrays and dictionaries โ reach for a tuple
- Tuples give you named, ordered, fixed-type values without the need for default values or string keys
- Tuple names don't need repeating in
return, unnamed tuples use.0,.1, etc., and tuples can be destructured directly into separate constants withlet (a, b) = ... - When choosing between array, set, and tuple: think about whether order matters, whether duplicates are allowed, and whether you're dealing with a fixed, known number of values
Top comments (0)