1.5.0

The Date.random(in:using:) functions now use BinaryFloatingPoint.randomPrintable(in:using:) to generate time intervals. This should lead to greater reliability when converting the time intervals of generated dates into strings.

1.4.0

Stub and ThrowingStub now conform to Observable. The only property that is tracked is calls. Changes to dependent properties like callArguments and callResults can also be tracked, but changes to resultQueue and defaultResult are not.

1.3.0

Adds functions for randomly generating dates within a specified range.

• Date has been extended with two static functions, both spelled random(in:using:), that generate a random date in either a closed or half-open range.
• RandomValueGenerating has been extended with a new property requirement and two new functions.
  • The new property is a ClosedRange<Date> spelled defaultClosedDateRange. This property provides a default closed date range to use when generating random dates. A default implementation is provided.
  • The two new functions, both spelled randomDate(in:), generate a random date in either a closed or half-open range using the aforementioned Date extension. The closed range version’s range parameter is optional. When nil is used (the default), defaultClosedDateRange is used.

1.2.0

This version updates DevTesting to require version 26 of Apple’s OSes.

1.1.0

Adds two initializers to ThrowingStub: init(defaultReturnValue:resultQueue:) and init(defaultError:resultQueue:). These initializers enable cleaner call sites, as in the following example:

// Before
mock.doSomethingSuccessfullyStub = ThrowingStub(defaultResult: .success(value))
mock.doSomethingUnsuccessfullyStub = ThrowingStub(defaultResult: .failure(error))

// After
mock.doSomethingSuccessfullyStub = ThrowingStub(defaultReturnValue: value)
mock.doSomethingUnsuccessfullyStub = ThrowingStub(defaultError: error)

1.0.0

This is the first release of DevTesting. The initial feature set includes

• SeedableRandomNumberGenerator, a random number generator that can be seeded to enable repeatable tests with random data.
• RandomValueGenerating, a protocol that your test suite can conform to add random value generation. It uses a seedable random number generator, and logs the seed before every test so that you can debug tests that failed. It also includes convenient functions for generating random data:
  • Booleans
  • Case iterables
  • Data
  • Floating point types
  • Integer types
  • Optionals
  • Strings
  • UUIDs
  • URLs, URL components, and URL query items
• Stub and ThrowingStub provide stubbing and spying functionality for your mock objects.