Test.Tasty.Hspec

hspec and tasty serve similar purposes; consider using one or the other.

However, in a pinch, this module allows you to run an hspec Spec as a tasty TestTree.

Tests

testSpec :: TestName -> Spec -> IO TestTree

Create a tasty TestTree from an hspec Spec.

testSpecs :: Spec -> IO [TestTree]

Create a list of tasty TestTree from an hspec Spec. This returns the same tests as testSpec but doesn't create a tasty test group from them.

Options

data TreatPendingAs

How to treat hspec pending tests.

tasty does not have the concept of pending tests, so we must map them to either successes or failures. By default, they are treated as failures.

Set via the command line flag --treat-pending-as (success|failure).

Constructors

TreatPendingAsFailure

Default.

TreatPendingAsSuccess

Instances

IsOption TreatPendingAs

Defined in Test.Tasty.Hspec

Methods

defaultValue :: TreatPendingAs

parseValue :: String -> Maybe TreatPendingAs

optionName :: Tagged TreatPendingAs String

optionHelp :: Tagged TreatPendingAs String

showDefaultValue :: TreatPendingAs -> Maybe String

optionCLParser :: Parser TreatPendingAs

Examples

The simplest usage of this library involves first creating a TestTree in IO, then running it with defaultMain.

main = do
  spec <- testSpec "spec" mySpec
  defaultMain
    (testGroup "tests"
      [ spec
      , ...
      ])

You can treat an pending/pendingWith test as a success instead of a failure (the default):

tests :: TestTree
tests =
  localOption TreatPendingAsSuccess $ testGroup "My Hspec TestTree"
    [ unsafePerformIO (testSpec "My first Hspec test" spec_firstHspecTest)
    , ...
    ]

If you don't do any IO during Spec creation, or the IO need not be performed at any particular time relative to other IO actions, it's perfectly fine to use unsafePerformIO.

main = do
  defaultMain
    (testGroup "tests"
      [ unsafePerformIO (testSpec "spec" mySpec)
      , ...
      ])