API Design Guidelines

Table of Contents

• Introduction
• Fundamentals
• Naming Promote Clear Usage
• Strive for Fluent Usage
• Use Terminology Well

Conventions

• General Conventions
• Parameters
• Argument Labels

Special Instructions

Introduction

Delivering a clear, consistent developer experience when writing Swift code is largely defined by the names and idioms that appear in APIs. These design guidelines explain how to make sure that your code feels like a part of the larger Swift ecosystem.

Fundamentals

• Clarity at the point of use is your most important goal. Entities such as methods and properties are declared only once but used repeatedly. Design APIs to make those uses clear and concise. When evaluating a design, reading a declaration is seldom sufficient; always examine a use case to make sure it looks clear in context.
• Clarity is more important than brevity. Although Swift code can be compact, it is a non-goal to enable the smallest possible code with the fewest characters. Brevity in Swift code, where it occurs, is a side-effect of the strong type system and features that naturally reduce boilerplate.
• Write a documentation comment for every declaration. Insights gained by writing documentation can have a profound impact on your design, so don’t put it off. If you are having trouble describing your API’s functionality in simple terms, you may have designed the wrong API. Use Swift’s dialect of Markdown.
• Begin with a summary that describes the entity being declared. Often, an API can be completely understood from its declaration and its summary. @@FENCE /// Returns a "view" of self containing the same elements in /// reverse order. func reversed() -> ReverseCollection @@/FENCE Focus on the summary; it’s the most important part. Many excellent documentation comments consist of nothing more than a great summary.
• Use a single sentence fragment if possible, ending with a period. Do not use a complete sentence.
• Describe what a function or method does and what it returns, omitting null effects and Void returns: @@FENCE /// Inserts newHead at the beginning of self. mutating func prepend( newHead: Int) /// Returns a List containing head followed by the elements /// of self. func prepending( head: Element) -> List /// Removes and returns the first element of self if non-empty; /// returns nil otherwise. mutating func popFirst() -> Element? @@/FENCE Note: in rare cases like popFirst above, the summary is formed of multiple sentence fragments separated by semicolons.
• Describe what a subscript accesses: @@FENCE /// Accesses the indexth element. subscript(index: Int) -> Element { get set } @@/FENCE
• Describe what an initializer creates: @@FENCE /// Creates an instance containing n repetitions of x. init(count n: Int, repeatedElement x: Element) @@/FENCE
• For all other declarations, describe what the declared entity is. @@FENCE /// A collection that supports equally efficient insertion/removal /// at any position. struct List { /// The element at the beginning of self, or nil if self is /// empty. var first: Element? ... @@/FENCE

Optionally, continue with one or more paragraphs and bullet items. Paragraphs are separated by blank lines and use complete sentences.

/// Writes the textual representation of each     ←   Summary 
/// element of `items` to the standard output.
///                                               ←   Blank line 
/// The textual representation for each item `x`  ←   Additional discussion 
/// is generated by the expression `String(x)`.
///
/// - **Parameter separator**: text to be printed     ⎫ 
///   between items.                              ⎟ 
/// - **Parameter terminator**: text to be printed    ⎬   [Parameters section](https://developer.apple.com/library/prerelease/mac/documentation/Xcode/Reference/xcode_markup_formatting_ref/SymbolDocumentation.html#//apple_ref/doc/uid/TP40016497-CH51-SW14) 
///   at the end.                                 ⎟ 
///                                               ⎭ 
/// - **Note**: To print without a trailing           ⎫ 
///   newline, pass `terminator: ""`              ⎟ 
///                                               ⎬   [Symbol commands](https://developer.apple.com/library/prerelease/mac/documentation/Xcode/Reference/xcode_markup_formatting_ref/SymbolDocumentation.html#//apple_ref/doc/uid/TP40016497-CH51-SW13) 
/// - **SeeAlso**: `CustomDebugStringConvertible`,    ⎟ 
///   `CustomStringConvertible`, `debugPrint`.    ⎭ 
public func print (
  _ items: Any..., separator: String = " ", terminator: String = "\n")

• Use recognized symbol documentation markup elements to add information beyond the summary, whenever appropriate.
• Know and use recognized bullet items with symbol command syntax. Popular development tools such as Xcode give special treatment to bullet items that start with the following keywords: Attention Author Authors Bug Complexity Copyright Date Experiment Important Invariant Note Parameter Parameters Postcondition Precondition Remark Requires Returns SeeAlso Since Throws ToDo Version Warning

Naming

Promote Clear Usage

• Include all the words needed to avoid ambiguity for a person reading code where the name is used. For example, consider a method that removes the element at a given position within a collection. @@FENCE extension List { public mutating func remove(at position: Index) -> Element } employees.remove(at: x) @@/FENCE If we were to omit the word at from the method signature, it could imply to the reader that the method searches for and removes an element equal to x, rather than using x to indicate the position of the element to remove. @@FENCE employees.remove(x) // unclear: are we removing x? @@/FENCE
• Omit needless words. Every word in a name should convey salient information at the use site. More words may be needed to clarify intent or disambiguate meaning, but those that are redundant with information the reader already possesses should be omitted. In particular, omit words that merely repeat type information. @@FENCE public mutating func removeElement( member: Element) -> Element? allViews.removeElement(cancelButton) @@/FENCE In this case, the word Element adds nothing salient at the call site. This API would be better: @@FENCE public mutating func remove( member: Element) -> Element? allViews.remove(cancelButton) // clearer @@/FENCE Occasionally, repeating type information is necessary to avoid ambiguity, but in general it is better to use a word that describes a parameter’s role rather than its type. See the next item for details.
• Name variables, parameters, and associated types according to their roles, rather than their type constraints. @@FENCE var string = "Hello" protocol ViewController { associatedtype ViewType : View } class ProductionLine { func restock(from widgetFactory: WidgetFactory) } @@/FENCE Repurposing a type name in this way fails to optimize clarity and expressivity. Instead, strive to choose a name that expresses the entity’s role. @@FENCE var greeting = "Hello" protocol ViewController { associatedtype ContentView : View } class ProductionLine { func restock(from supplier: WidgetFactory) } @@/FENCE If an associated type is so tightly bound to its protocol constraint that the protocol name is the role, avoid collision by appending Protocol to the protocol name: @@FENCE protocol Sequence { associatedtype Iterator : IteratorProtocol } protocol IteratorProtocol { ... } @@/FENCE
• Compensate for weak type information to clarify a parameter’s role. Especially when a parameter type is NSObject, Any, AnyObject, or a fundamental type such as Int or String, type information and context at the point of use may not fully convey intent. In this example, the declaration may be clear, but the use site is vague. @@FENCE func add( observer: NSObject, for keyPath: String) grid.add(self, for: graphics) // vague @@/FENCE To restore clarity, precede each weakly typed parameter with a noun describing its role: @@FENCE func addObserver( observer: NSObject, forKeyPath path: String) grid.addObserver(self, forKeyPath: graphics) // clear @@/FENCE

Strive for Fluent Usage

• Prefer method and function names that make use sites form grammatical English phrases. @@FENCE x.insert(y, at: z) “x, insert y at z” x.subviews(havingColor: y) “x's subviews having color y” x.capitalizingNouns() “x, capitalizing nouns” @@/FENCE @@FENCE x.insert(y, position: z) x.subviews(color: y) x.nounCapitalize() @@/FENCE It is acceptable for fluency to degrade after the first argument or two when those arguments are not central to the call’s meaning: @@FENCE AudioUnit.instantiate( with: description, options: [.inProcess], completionHandler: stopProgressBar) @@/FENCE
• Begin names of factory methods with “make”, e.g. x.makeIterator().
• The first argument to initializer and factory methods calls should not form a phrase starting with the base name, e.g. x.makeWidget(cogCount: 47) For example, the first arguments to these calls do not read as part of the same phrase as the base name: @@FENCE let foreground = Color(red: 32, green: 64, blue: 128) let newPart = factory.makeWidget(gears: 42, spindles: 14) let ref = Link(target: destination) @@/FENCE In the following, the API author has tried to create grammatical continuity with the first argument. @@FENCE let foreground = Color(havingRGBValuesRed: 32, green: 64, andBlue: 128) let newPart = factory.makeWidget(havingGearCount: 42, andSpindleCount: 14) let ref = Link(to: destination) @@/FENCE In practice, this guideline along with those for argument labels means the first argument will have a label unless the call is performing a value preserving type conversion. @@FENCE let rgbForeground = RGBColor(cmykForeground) @@/FENCE
• Name functions and methods according to their side-effects Those without side-effects should read as noun phrases, e.g. x.distance(to: y), i.successor().
• Those with side-effects should read as imperative verb phrases, e.g., print(x), x.sort(), x.append(y).
• Name Mutating/nonmutating method pairs consistently. A mutating method will often have a nonmutating variant with similar semantics, but that returns a new value rather than updating an instance in-place. When the operation is naturally described by a verb, use the verb’s imperative for the mutating method and apply the “ed” or “ing” suffix to name its nonmutating counterpart. Mutating Nonmutating x.sort() z = x.sorted() x.append(y) z = x.appending(y) Prefer to name the nonmutating variant using the verb’s past participle (usually appending “ed”): @@FENCE /// Reverses self in-place. mutating func reverse() /// Returns a reversed copy of self. func reversed() -> Self ... x.reverse() let y = x.reversed() @@/FENCE
• When adding “ed” is not grammatical because the verb has a direct object, name the nonmutating variant using the verb’s present participle, by appending “ing.” @@FENCE /// Strips all the newlines from self mutating func stripNewlines() /// Returns a copy of self with all the newlines stripped. func strippingNewlines() -> String ... s.stripNewlines() let oneLine = t.strippingNewlines() @@/FENCE

When the operation is naturally described by a noun, use the noun for the nonmutating method and apply the “form” prefix to name its mutating counterpart.

Nonmutating Mutating

x = y.union(z) y.formUnion(z)

j = c.successor(i) c.formSuccessor(&i)

Uses of Boolean methods and properties should read as assertions about the receiver when the use is nonmutating, e.g. x.isEmpty, line1.intersects(line2).

Protocols that describe what something is should read as nouns (e.g. Collection).

Protocols that describe a capability should be named using the suffixes able, ible, or ing (e.g. Equatable, ProgressReporting).

The names of other types, properties, variables, and constants should read as nouns.

Use Terminology Well

Term of Art — noun - a word or phrase that has a precise, specialized meaning within a particular field or profession.

• Avoid obscure terms if a more common word conveys meaning just as well. Don’t say “epidermis” if “skin” will serve your purpose. Terms of art are an essential communication tool, but should only be used to capture crucial meaning that would otherwise be lost.
• Stick to the established meaning if you do use a term of art. The only reason to use a technical term rather than a more common word is that it precisely expresses something that would otherwise be ambiguous or unclear. Therefore, an API should use the term strictly in accordance with its accepted meaning. Don’t surprise an expert: anyone already familiar with the term will be surprised and probably angered if we appear to have invented a new meaning for it.
• Don’t confuse a beginner: anyone trying to learn the term is likely to do a web search and find its traditional meaning.

Avoid abbreviations. Abbreviations, especially non-standard ones, are effectively terms-of-art, because understanding depends on correctly translating them into their non-abbreviated forms.

The intended meaning for any abbreviation you use should be easily found by a web search.

Embrace precedent. Don’t optimize terms for the total beginner at the expense of conformance to existing culture.

It is better to name a contiguous data structure Array than to use a simplified term such as List, even though a beginner might grasp the meaning of List more easily. Arrays are fundamental in modern computing, so every programmer knows—or will soon learn—what an array is. Use a term that most programmers are familiar with, and their web searches and questions will be rewarded.

Within a particular programming domain, such as mathematics, a widely precedented term such as sin(x) is preferable to an explanatory phrase such as verticalPositionOnUnitCircleAtOriginOfEndOfRadiusWithAngle(x). Note that in this case, precedent outweighs the guideline to avoid abbreviations: although the complete word is sine, “sin(x)” has been in common use among programmers for decades, and among mathematicians for centuries.

Conventions

General Conventions

• Document the complexity of any computed property that is not O(1). People often assume that property access involves no significant computation, because they have stored properties as a mental model. Be sure to alert them when that assumption may be violated.
• Prefer methods and properties to free functions. Free functions are used only in special cases: When there’s no obvious self: @@FENCE min(x, y, z) @@/FENCE
• When the function is an unconstrained generic: @@FENCE print(x) @@/FENCE
• When function syntax is part of the established domain notation: @@FENCE sin(x) @@/FENCE
• Follow case conventions. Names of types and protocols are UpperCamelCase. Everything else is lowerCamelCase. Acronyms and initialisms that commonly appear as all upper case in American English should be uniformly up- or down-cased according to case conventions: @@FENCE var utf8Bytes: [UTF8.CodeUnit] var isRepresentableAsASCII = true var userSMTPServer: SecureSMTPServer @@/FENCE Other acronyms should be treated as ordinary words: @@FENCE var radarDetector: RadarScanner var enjoysScubaDiving = true @@/FENCE
• Methods can share a base name when they share the same basic meaning or when they operate in distinct domains. For example, the following is encouraged, since the methods do essentially the same things: @@FENCE extension Shape { /// Returns true if other is within the area of self; /// otherwise, false. func contains( other: Point) -> Bool { ... } /// Returns true if other is entirely within the area of self; /// otherwise, false. func contains( other: Shape) -> Bool { ... } /// Returns true if other is within the area of self; /// otherwise, false. func contains( other: LineSegment) -> Bool { ... } } @@/FENCE And since geometric types and collections are separate domains, this is also fine in the same program: @@FENCE extension Collection where Element : Equatable { /// Returns true if self contains an element equal to /// sought; otherwise, false. func contains( sought: Element) -> Bool { ... } } @@/FENCE However, these index methods have different semantics, and should have been named differently: @@FENCE extension Database { /// Rebuilds the database's search index func index() { ... } /// Returns the nth row in the given table. func index(_ n: Int, inTable: TableID) -> TableRow { ... } } @@/FENCE Lastly, avoid “overloading on return type” because it causes ambiguities in the presence of type inference. @@FENCE extension Box { /// Returns the Int stored in self, if any, and /// nil otherwise. func value() -> Int? { ... } /// Returns the String stored in self, if any, and /// nil otherwise. func value() -> String? { ... } } @@/FENCE

Parameters

func move(from **start**: Point, to **end**: Point)

• Choose parameter names to serve documentation. Even though parameter names do not appear at a function or method’s point of use, they play an important explanatory role. Choose these names to make documentation easy to read. For example, these names make documentation read naturally: @@FENCE /// Return an Array containing the elements of self /// that satisfy predicate. func filter( predicate: (Element) -> Bool) -> [Generator.Element] /// Replace the given subRange of elements with newElements. mutating func replaceRange( subRange: Range, with newElements: [E]) @@/FENCE These, however, make the documentation awkward and ungrammatical: @@FENCE /// Return an Array containing the elements of self /// that satisfy includedInResult. func filter( includedInResult: (Element) -> Bool) -> [Generator.Element] /// Replace the range of elements indicated by r with /// the contents of with. mutating func replaceRange( r: Range, with: [E]) @@/FENCE
• Take advantage of defaulted parameters when it simplifies common uses. Any parameter with a single commonly-used value is a candidate for a default. Default arguments improve readability by hiding irrelevant information. For example: @@FENCE let order = lastName.compare( royalFamilyName, options: , range: nil, locale: nil) @@/FENCE can become the much simpler: @@FENCE let order = lastName.compare(royalFamilyName) @@/FENCE Default arguments are generally preferable to the use of method families, because they impose a lower cognitive burden on anyone trying to understand the API. @@FENCE extension String { /// ...description... public func compare( other: String, options: CompareOptions = , range: Range? = nil, locale: Locale? = nil ) -> Ordering } @@/FENCE The above may not be simple, but it is much simpler than: @@FENCE extension String { /// ...description 1... public func compare( other: String) -> Ordering /// ...description 2... public func compare( other: String, options: CompareOptions) -> Ordering /// ...description 3... public func compare( other: String, options: CompareOptions, range: Range) -> Ordering /// ...description 4... public func compare( _ other: String, options: StringCompareOptions, range: Range, locale: Locale) -> Ordering } @@/FENCE Every member of a method family needs to be separately documented and understood by users. To decide among them, a user needs to understand all of them, and occasional surprising relationships—for example, foo(bar: nil) and foo() aren’t always synonyms—make this a tedious process of ferreting out minor differences in mostly identical documentation. Using a single method with defaults provides a vastly superior programmer experience.
• Prefer to locate parameters with defaults toward the end of the parameter list. Parameters without defaults are usually more essential to the semantics of a method, and provide a stable initial pattern of use where methods are invoked.
• If your API will run in production, prefer #fileID over alternatives. #fileID saves space and protects developers’ privacy. Use #filePath in APIs that are never run by end users (such as test helpers and scripts) if the full path will simplify development workflows or be used for file I/O. Use #file to preserve source compatibility with Swift 5.2 or earlier.

Argument Labels

func move(**from** start: Point, **to** end: Point)
x.move(**from:** x, **to:** y)

• Omit all labels when arguments can’t be usefully distinguished, e.g. min(number1, number2), zip(sequence1, sequence2).
• In initializers that perform value preserving type conversions, omit the first argument label, e.g. Int64(someUInt32) The first argument should always be the source of the conversion. @@FENCE extension String { // Convert x into its textual representation in the given radix init( x: BigInt, radix: Int = 10) ← Note the initial underscore } text = "The value is: " text += String(veryLargeNumber) text += " and in hexadecimal, it's" text += String(veryLargeNumber, radix: 16) @@/FENCE In “narrowing” type conversions, though, a label that describes the narrowing is recommended. @@FENCE extension UInt32 { /// Creates an instance having the specified value. init( value: Int16) ← Widening, so no label /// Creates an instance having the lowest 32 bits of source. init(truncating source: UInt64) /// Creates an instance having the nearest representable /// approximation of valueToApproximate. init(saturating valueToApproximate: UInt64) } @@/FENCE A value preserving type conversion is a monomorphism, i.e. every difference in the value of the source results in a difference in the value of the result. For example, conversion from Int8 to Int64 is value preserving because every distinct Int8 value is converted to a distinct Int64 value. Conversion in the other direction, however, cannot be value preserving: Int64 has more possible values than can be represented in an Int8. Note: the ability to retrieve the original value has no bearing on whether a conversion is value preserving.
• When the first argument forms part of a prepositional phrase, give it an argument label. The argument label should normally begin at the preposition, e.g. x.removeBoxes(havingLength: 12). An exception arises when the first two arguments represent parts of a single abstraction. @@FENCE a.move(toX: b, y: c) a.fade(fromRed: b, green: c, blue: d) @@/FENCE In such cases, begin the argument label after the preposition, to keep the abstraction clear. @@FENCE a.moveTo(x: b, y: c) a.fadeFrom(red: b, green: c, blue: d) @@/FENCE
• Otherwise, if the first argument forms part of a grammatical phrase, omit its label, appending any preceding words to the base name, e.g. x.addSubview(y) This guideline implies that if the first argument doesn’t form part of a grammatical phrase, it should have a label. @@FENCE view.dismiss(animated: false) let text = words.split(maxSplits: 12) let studentsByName = students.sorted(isOrderedBefore: Student.namePrecedes) @@/FENCE Note that it’s important that the phrase convey the correct meaning. The following would be grammatical but would express the wrong thing. @@FENCE view.dismiss(false) Don't dismiss? Dismiss a Bool? words.split(12) Split the number 12? @@/FENCE Note also that arguments with default values can be omitted, and in that case do not form part of a grammatical phrase, so they should always have labels.
• Label all other arguments.

Special Instructions

• Label tuple members and name closure parameters where they appear in your API. These names have explanatory power, can be referenced from documentation comments, and provide expressive access to tuple members. @@FENCE /// Ensure that we hold uniquely-referenced storage for at least /// requestedCapacity elements. /// /// If more storage is needed, allocate is called with /// byteCount equal to the number of maximally-aligned /// bytes to allocate. /// /// - Returns: /// - reallocated: true if a new block of memory /// was allocated; otherwise, false. /// - capacityChanged: true if capacity was updated; /// otherwise, false. mutating func ensureUniqueStorage( minimumCapacity requestedCapacity: Int, allocate: (_ byteCount: Int) -> UnsafePointer ) -> (reallocated: Bool, capacityChanged: Bool) @@/FENCE Names used for closure parameters should be chosen like parameter names for top-level functions. Labels for closure arguments that appear at the call site are not supported.
• Take extra care with unconstrained polymorphism (e.g. Any, AnyObject, and unconstrained generic parameters) to avoid ambiguities in overload sets. For example, consider this overload set: @@FENCE struct Array { /// Inserts newElement at self.endIndex. public mutating func append( newElement: Element) /// Inserts the contents of newElements, in order, at /// self.endIndex. public mutating func append( newElements: S) where S.Generator.Element == Element } @@/FENCE These methods form a semantic family, and the argument types appear at first to be sharply distinct. However, when Element is Any, a single element can have the same type as a sequence of elements. @@FENCE var values: [Any] = [1, "a"] values.append([2, 3, 4]) // [1, "a", [2, 3, 4]] or [1, "a", 2, 3, 4]? @@/FENCE To eliminate the ambiguity, name the second overload more explicitly. @@FENCE struct Array { /// Inserts newElement at self.endIndex. public mutating func append(_ newElement: Element) /// Inserts the contents of newElements, in order, at /// self.endIndex. public mutating func append(contentsOf newElements: S) where S.Generator.Element == Element } @@/FENCE Notice how the new name better matches the documentation comment. In this case, the act of writing the documentation comment actually brought the issue to the API author’s attention.