Swift Testing is a package with expressive and intuitive APIs that make testing your Swift code a breeze.
Swift Testing has a clear and expressive API built using macros, so you can
declare complex behaviors with a small amount of code. The #expect
API uses
Swift expressions and operators, and captures the evaluated values so you can
quickly understand what went wrong when a test fails.
@Test func helloWorld() {
let greeting = "Hello, world!"
#expect(greeting == "Hello") // Expectation failed: (greeting → "Hello, world!") == "Hello"
}
You can customize the behavior of tests or test suites using traits specified in your code. Traits can describe the runtime conditions for a test, like which device a test should run on, or limit a test to certain operating system versions. Traits can also help you use continuous integration effectively by specifying execution time limits for your tests.
@Test(.enabled(if: AppFeatures.isCommentingEnabled))
func videoCommenting() async throws {
let video = try #require(await videoLibrary.video(named: "A Beach"))
#expect(video.comments.contains("So picturesque!"))
}
Swift Testing provides many ways to keep your tests organized. Structure related tests using a hierarchy of groups and subgroups. Apply tags to flexibly manage, edit, and run tests with common characteristics across your test suite, like tests that target a specific device or use a specific module. You can also give tests a descriptive name so you know what they’re doing at a glance.
@Test("Check video metadata",
.tags(.metadata))
func videoMetadata() {
let video = Video(fileName: "By the Lake.mov")
let expectedMetadata = Metadata(duration: .seconds(90))
#expect(video.metadata == expectedMetadata)
}
Parameterized tests help you run the same test over a sequence of values so you can write less code. And all tests integrate seamlessly with Swift Concurrency and run in parallel by default.
@Test("Continents mentioned in videos", arguments: [
"A Beach",
"By the Lake",
"Camping in the Woods"
])
func mentionedContinents(videoName: String) async throws {
let videoLibrary = try await VideoLibrary()
let video = try #require(await videoLibrary.video(named: videoName))
#expect(video.mentionedContinents.count <= 3)
}
Swift Testing works on all major platforms supported by Swift, including Apple platforms, Linux, and Windows, so your tests can behave more consistently when moving between platforms. It’s developed as open source and discussed on the Swift Forums so the very best ideas, from anywhere, can help shape the future of testing in Swift.
The table below describes the current level of support that Swift Testing has for various platforms:
Platform | CI Status (6.0) | CI Status (main) | Support Status |
---|---|---|---|
macOS | Supported | ||
iOS | Supported | ||
watchOS | Supported | ||
tvOS | Supported | ||
visionOS | Supported | ||
Ubuntu 22.04 | Supported | ||
Windows | Supported | ||
Wasm | Experimental |
If you already have tests written using XCTest, you can run them side-by-side with newer tests written using Swift Testing. This helps you migrate tests incrementally, at your own pace.
Important
This package is under active, ongoing development and requires a recent 6.0 development snapshot toolchain. Its contents and interfaces are still considered experimental at this time and may change. See this Forum post for details about stable release plans.
Detailed documentation for Swift Testing can be found on the Swift Package Index. There, you can delve into comprehensive guides, tutorials, and API references to make the most out of this package. To try it yourself, see Getting Started.
Other documentation resources for this project can be found in the
README
of the Documentation/
subdirectory.