FixtureSuite

Run the passed test function with a fixture created by this method.

This method should create the fixture object needed by the tests of the current suite, invoke the test function (passing in the fixture object), and if needed, perform any clean up needed after the test completes. For more detail and examples, see the main documentation for this trait.

Assert that an Option[String] is None. If the condition is None, this method returns normally. Else, it throws TestFailedException with the String value of the Some included in the TestFailedException's detail message.

This form of assert is usually called in conjunction with an implicit conversion to Equalizer, using a === comparison, as in:

assert(a === b)

For more information on how this mechanism works, see the documentation for Equalizer.

Assert that an Option[String] is None. If the condition is None, this method returns normally. Else, it throws TestFailedException with the String value of the Some, as well as the String obtained by invoking toString on the specified message, included in the TestFailedException's detail message.

This form of assert is usually called in conjunction with an implicit conversion to Equalizer, using a === comparison, as in:

assert(a === b, "extra info reported if assertion fails")

For more information on how this mechanism works, see the documentation for Equalizer.

Assert that a boolean condition, described in String message, is true. If the condition is true, this method returns normally. Else, it throws TestFailedException with the String obtained by invoking toString on the specified message as the exception's detail message.

Assert that a boolean condition is true. If the condition is true, this method returns normally. Else, it throws TestFailedException.

Implicit conversion from Any to Equalizer, used to enable assertions with === comparisons.

For more information on this mechanism, see the documentation for Equalizer.

Because trait Suite mixes in Assertions, this implicit conversion will always be available by default in ScalaTest Suites. This is the only implicit conversion that is in scope by default in every ScalaTest Suite. Other implicit conversions offered by ScalaTest, such as those that support the matchers DSL or invokePrivate, must be explicitly invited into your test code, either by mixing in a trait or importing the members of its companion object. The reason ScalaTest requires you to invite in implicit conversions (with the exception of the implicit conversion for === operator) is because if one of ScalaTest's implicit conversions clashes with an implicit conversion used in the code you are trying to test, your program won't compile. Thus there is a chance that if you are ever trying to use a library or test some code that also offers an implicit conversion involving a === operator, you could run into the problem of a compiler error due to an ambiguous implicit conversion. If that happens, you can turn off the implicit conversion offered by this convertToEqualizer method simply by overriding the method in your Suite subclass, but not marking it as implicit:

// In your Suite subclass
override def convertToEqualizer(left: Any) = new Equalizer(left)

Executes one or more tests in this Suite, printing results to the standard output.

This method invokes run on itself, passing in values that can be configured via the parameters to this method, all of which have default values. This behavior is convenient when working with ScalaTest in the Scala interpreter. Here's a summary of this method's parameters and how you can use them:

The testName parameter

If you leave testName at its default value (of null), this method will pass None to the testName parameter of run, and as a result all the tests in this suite will be executed. If you specify a testName, this method will pass Some(testName) to run, and only that test will be run. Thus to run all tests in a suite from the Scala interpreter, you can write:

scala> (new ExampleSuite).execute()

To run just the test named "my favorite test" in a suite from the Scala interpreter, you would write:

scala> (new ExampleSuite).execute("my favorite test")

Or:

scala> (new ExampleSuite).execute(testName = "my favorite test")

The configMap parameter

If you provide a value for the configMap parameter, this method will pass it to run. If not, the default value of an empty Map will be passed. For more information on how to use a config map to configure your test suites, see the config map section in the main documentation for this trait. Here's an example in which you configure a run with the name of an input file:

scala> (new ExampleSuite).execute(configMap = Map("inputFileName" -> "in.txt")

The color parameter

If you leave the color parameter unspecified, this method will configure the reporter it passes to run to print to the standard output in color (via ansi escape characters). If you don't want color output, specify false for color, like this:

scala> (new ExampleSuite).execute(color = false)

The durations parameter

If you leave the durations parameter unspecified, this method will configure the reporter it passes to run to not print durations for tests and suites to the standard output. If you want durations printed, specify true for durations, like this:

scala> (new ExampleSuite).execute(durations = true)

The shortstacks and fullstacks parameters

If you leave both the shortstacks and fullstacks parameters unspecified, this method will configure the reporter it passes to run to not print stack traces for failed tests if it has a stack depth that identifies the offending line of test code. If you prefer a short stack trace (10 to 15 stack frames) to be printed with any test failure, specify true for shortstacks:

scala> (new ExampleSuite).execute(shortstacks = true)

For full stack traces, set fullstacks to true:

scala> (new ExampleSuite).execute(fullstacks = true)

If you specify true for both shortstacks and fullstacks, you'll get full stack traces.

The stats parameter

If you leave the stats parameter unspecified, this method will not fire RunStarting and either RunCompleted or RunAborted events to the reporter it passes to run. If you specify true for stats, this method will fire the run events to the reporter, and the reporter will print the expected test count before the run, and various statistics after, including the number of suites completed and number of tests that succeeded, failed, were ignored or marked pending. Here's how you get the stats:

scala> (new ExampleSuite).execute(stats = true)

To summarize, this method will pass to run:

• testName - None if this method's testName parameter is left at its default value of null, else Some(testName).
• reporter - a reporter that prints to the standard output
• stopper - a Stopper whose apply method always returns false
• filter - a Filter constructed with None for tagsToInclude and Set() for tagsToExclude
• configMap - the configMap passed to this method
• distributor - None
• tracker - a new Tracker

Note: In ScalaTest, the terms "execute" and "run" basically mean the same thing and can be used interchangably. The reason this method isn't named run is that it takes advantage of default arguments, and you can't mix overloaded methods and default arguments in Scala. (If named run, this method would have the same name but different arguments than the main run method that takes seven arguments. Thus it would overload and couldn't be used with default argument values.)

Design note: This method has two "features" that may seem unidiomatic. First, the default value of testName is null. Normally in Scala the type of testName would be Option[String] and the default value would be None, as it is in this trait's run method. The null value is used here for two reasons. First, in ScalaTest 1.5, execute was changed from four overloaded methods to one method with default values, taking advantage of the default and named parameters feature introduced in Scala 2.8. To not break existing source code, testName needed to have type String, as it did in two of the overloaded execute methods prior to 1.5. The other reason is that execute has always been designed to be called primarily from an interpeter environment, such as the Scala REPL (Read-Evaluate-Print-Loop). In an interpreter environment, minimizing keystrokes is king. A String type with a null default value lets users type suite.execute("my test name") rather than suite.execute(Some("my test name")), saving several keystrokes.

The second non-idiomatic feature is that shortstacks and fullstacks are all lower case rather than camel case. This is done to be consistent with the Shell, which also uses those forms. The reason lower case is used in the Shell is to save keystrokes in an interpreter environment. Most Unix commands, for example, are all lower case, making them easier and quicker to type. In the ScalaTest Shell, methods like shortstacks, fullstacks, and nostats, etc., are designed to be all lower case so they feel more like shell commands than methods.

Expect that the value passed as expected equals the value passed as actual. If the actual value equals the expected value (as determined by ==), expect returns normally. Else, expect throws an TestFailedException whose detail message includes the expected and actual values.

Expect that the value passed as expected equals the value passed as actual. If the actual equals the expected (as determined by ==), expect returns normally. Else, if actual is not equal to expected, expect throws an TestFailedException whose detail message includes the expected and actual values, as well as the String obtained by invoking toString on the passed message.

The total number of tests that are expected to run when this Suite's run method is invoked.

This trait's implementation of this method returns the sum of:

• the size of the testNames List, minus the number of tests marked as ignored
• the sum of the values obtained by invoking expectedTestCount on every nested Suite contained in nestedSuites

Throws TestFailedException, with the passed Throwable cause, to indicate a test failed. The getMessage method of the thrown TestFailedException will return cause.toString().

Throws TestFailedException, with the passed String message as the exception's detail message and Throwable cause, to indicate a test failed.

Throws TestFailedException, with the passed String message as the exception's detail message, to indicate a test failed.

Throws TestFailedException to indicate a test failed.

Intercept and return an exception that's expected to be thrown by the passed function value. The thrown exception must be an instance of the type specified by the type parameter of this method. This method invokes the passed function. If the function throws an exception that's an instance of the specified type, this method returns that exception. Else, whether the passed function returns normally or completes abruptly with a different exception, this method throws TestFailedException.

Note that the type specified as this method's type parameter may represent any subtype of AnyRef, not just Throwable or one of its subclasses. In Scala, exceptions can be caught based on traits they implement, so it may at times make sense to specify a trait that the intercepted exception's class must mix in. If a class instance is passed for a type that could not possibly be used to catch an exception (such as String, for example), this method will complete abruptly with a TestFailedException.

A List of this Suite object's nested Suites. If this Suite contains no nested Suites, this method returns an empty List. This trait's implementation of this method returns an empty List.

Throws TestPendingException to indicate a test is pending.

A pending test is one that has been given a name but is not yet implemented. The purpose of pending tests is to facilitate a style of testing in which documentation of behavior is sketched out before tests are written to verify that behavior (and often, the before the behavior of the system being tested is itself implemented). Such sketches form a kind of specification of what tests and functionality to implement later.

To support this style of testing, a test can be given a name that specifies one bit of behavior required by the system being tested. The test can also include some code that sends more information about the behavior to the reporter when the tests run. At the end of the test, it can call method pending, which will cause it to complete abruptly with TestPendingException. Because tests in ScalaTest can be designated as pending with TestPendingException, both the test name and any information sent to the reporter when running the test can appear in the report of a test run. (In other words, the code of a pending test is executed just like any other test.) However, because the test completes abruptly with TestPendingException, the test will be reported as pending, to indicate the actual test, and possibly the functionality it is intended to test, has not yet been implemented.

Note: This method always completes abruptly with a TestPendingException. Thus it always has a side effect. Methods with side effects are usually invoked with parentheses, as in pending(). This method is defined as a parameterless method, in flagrant contradiction to recommended Scala style, because it forms a kind of DSL for pending tests. It enables tests in suites such as FunSuite or FunSpec to be denoted by placing "(pending)" after the test name, as in:

test("that style rules are not laws") (pending)

Readers of the code see "pending" in parentheses, which looks like a little note attached to the test name to indicate it is pending. Whereas "(pending()) looks more like a method call, "(pending)" lets readers stay at a higher level, forgetting how it is implemented and just focusing on the intent of the programmer who wrote the code.

Execute the passed block of code, and if it completes abruptly, throw TestPendingException, else throw TestFailedException.

This method can be used to temporarily change a failing test into a pending test in such a way that it will automatically turn back into a failing test once the problem originally causing the test to fail has been fixed. At that point, you need only remove the pendingUntilFixed call. In other words, a pendingUntilFixed surrounding a block of code that isn't broken is treated as a test failure. The motivation for this behavior is to encourage people to remove pendingUntilFixed calls when there are no longer needed.

This method facilitates a style of testing in which tests are written before the code they test. Sometimes you may encounter a test failure that requires more functionality than you want to tackle without writing more tests. In this case you can mark the bit of test code causing the failure with pendingUntilFixed. You can then write more tests and functionality that eventually will get your production code to a point where the original test won't fail anymore. At this point the code block marked with pendingUntilFixed will no longer throw an exception (because the problem has been fixed). This will in turn cause pendingUntilFixed to throw TestFailedException with a detail message explaining you need to go back and remove the pendingUntilFixed call as the problem orginally causing your test code to fail has been fixed.

Runs this suite of tests.

If testName is None, this trait's implementation of this method calls these two methods on this object in this order:

• runNestedSuites(report, stopper, tagsToInclude, tagsToExclude, configMap, distributor)
• runTests(testName, report, stopper, tagsToInclude, tagsToExclude, configMap)

If testName is defined, then this trait's implementation of this method calls runTests, but does not call runNestedSuites. This behavior is part of the contract of this method. Subclasses that override run must take care not to call runNestedSuites if testName is defined. (The OneInstancePerTest trait depends on this behavior, for example.)

Subclasses and subtraits that override this run method can implement them without invoking either the runTests or runNestedSuites methods, which are invoked by this trait's implementation of this method. It is recommended, but not required, that subclasses and subtraits that override run in a way that does not invoke runNestedSuites also override runNestedSuites and make it final. Similarly it is recommended, but not required, that subclasses and subtraits that override run in a way that does not invoke runTests also override runTests (and runTest, which this trait's implementation of runTests calls) and make it final. The implementation of these final methods can either invoke the superclass implementation of the method, or throw an UnsupportedOperationException if appropriate. The reason for this recommendation is that ScalaTest includes several traits that override these methods to allow behavior to be mixed into a Suite. For example, trait BeforeAndAfterEach overrides runTestss. In a Suite subclass that no longer invokes runTests from run, the BeforeAndAfterEach trait is not applicable. Mixing it in would have no effect. By making runTests final in such a Suite subtrait, you make the attempt to mix BeforeAndAfterEach into a subclass of your subtrait a compiler error. (It would fail to compile with a complaint that BeforeAndAfterEach is trying to override runTests, which is a final method in your trait.)

Run zero to many of this Suite's nested Suites.

If the passed distributor is None, this trait's implementation of this method invokes run on each nested Suite in the List obtained by invoking nestedSuites. If a nested Suite's run method completes abruptly with an exception, this trait's implementation of this method reports that the Suite aborted and attempts to run the next nested Suite. If the passed distributor is defined, this trait's implementation puts each nested Suite into the Distributor contained in the Some, in the order in which the Suites appear in the List returned by nestedSuites, passing in a new Tracker obtained by invoking nextTracker on the Tracker passed to this method.

Implementations of this method are responsible for ensuring SuiteStarting events are fired to the Reporter before executing any nested Suite, and either SuiteCompleted or SuiteAborted after executing any nested Suite.

Run a test.

This trait's implementation uses Java reflection to invoke on this object the test method identified by the passed testName.

Implementations of this method are responsible for ensuring a TestStarting event is fired to the Reporter before executing any test, and either TestSucceeded, TestFailed, or TestPending after executing any nested Suite. (If a test is marked with the org.scalatest.Ignore tag, the runTests method is responsible for ensuring a TestIgnored event is fired and that this runTest method is not invoked for that ignored test.)

Run zero to many of this Suite's tests.

This method takes a testName parameter that optionally specifies a test to invoke. If testName is defined, this trait's implementation of this method invokes runTest on this object, passing in:

• testName - the String value of the testName Option passed to this method
• reporter - the Reporter passed to this method, or one that wraps and delegates to it
• stopper - the Stopper passed to this method, or one that wraps and delegates to it
• configMap - the configMap Map passed to this method, or one that wraps and delegates to it

This method takes a Filter, which encapsulates an optional Set of tag names that should be included (tagsToInclude) and a Set that should be excluded (tagsToExclude), when deciding which of this Suite's tests to run. If tagsToInclude is None, all tests will be run except those those belonging to tags listed in the tagsToExclude Set. If tagsToInclude is defined, only tests belonging to tags mentioned in the tagsToInclude Set, and not mentioned in the tagsToExcludeSet will be run. However, if testName is defined, tagsToInclude and tagsToExclude are essentially ignored. Only if testName is None will tagsToInclude and tagsToExclude be consulted to determine which of the tests named in the testNames Set should be run. This trait's implementation behaves this way, and it is part of the general contract of this method, so all overridden forms of this method should behave this way as well. For more information on test tags, see the main documentation for this trait and for class Filter. Note that this means that even if a test is marked as ignored, for example a test method in a Suite annotated with org.scalatest.Ignore, if that test name is passed as testName to runTest, it will be invoked despite the Ignore annotation.

If testName is None, this trait's implementation of this method invokes testNames on this Suite to get a Set of names of tests to potentially run. (A testNames value of None essentially acts as a wildcard that means all tests in this Suite that are selected by tagsToInclude and tagsToExclude should be run.) For each test in the testName Set, in the order they appear in the iterator obtained by invoking the elements method on the Set, this trait's implementation of this method checks whether the test should be run based on the Filter. If so, this implementation invokes runTest, passing in:

• testName - the String name of the test to run (which will be one of the names in the testNames Set)
• reporter - the Reporter passed to this method, or one that wraps and delegates to it
• stopper - the Stopper passed to this method, or one that wraps and delegates to it
• configMap - the configMap passed to this method, or one that wraps and delegates to it

If a test is marked with the org.scalatest.Ignore tag, implementations of this method are responsible for ensuring a TestIgnored event is fired for that test and that runTest is not called for that test.

A user-friendly suite name for this Suite.

This trait's implementation of this method returns the simple name of this object's class. This trait's implementation of runNestedSuites calls this method to obtain a name for Reports to pass to the suiteStarting, suiteCompleted, and suiteAborted methods of the Reporter.

A Map whose keys are String tag names with which tests in this Suite are marked, and whose values are the Set of test names marked with each tag. If this Suite contains no tags, this method returns an empty Map.

This trait's implementation of this method uses Java reflection to discover any Java annotations attached to its test methods. The fully qualified name of each unique annotation that extends TagAnnotation is considered a tag. This trait's implementation of this method, therefore, places one key/value pair into to the Map for each unique tag annotation name discovered through reflection. The mapped value for each tag name key will contain the test method name, as provided via the testNames method.

Subclasses may override this method to define and/or discover tags in a custom manner, but overriding method implementations should never return an empty Set as a value. If a tag has no tests, its name should not appear as a key in the returned Map.

A Set of test names. If this Suite contains no tests, this method returns an empty Set.

This trait's implementation of this method uses Java reflection to discover all public methods whose name starts with "test", which take either nothing or a single Informer as parameters. For each discovered test method, it assigns a test name comprised of just the method name if the method takes no parameters, or the method name plus (Informer) if the method takes a Informer. Here are a few method signatures and the names that this trait's implementation assigns them:

def testCat() {}         // test name: "testCat"
def testCat(Informer) {} // test name: "testCat(Informer)"
def testDog() {}         // test name: "testDog"
def testDog(Informer) {} // test name: "testDog(Informer)"
def test() {}            // test name: "test"
def test(Informer) {}    // test name: "test(Informer)"

This trait's implementation of this method returns an immutable Set of all such names, excluding the name testNames. The iterator obtained by invoking elements on this returned Set will produce the test names in their natural order, as determined by String's compareTo method.

This trait's implementation of runTests invokes this method and calls runTest for each test name in the order they appear in the returned Set's iterator. Although this trait's implementation of this method returns a Set whose iterator produces String test names in a well-defined order, the contract of this method does not required a defined order. Subclasses are free to override this method and return test names in an undefined order, or in a defined order that's different from String's natural order.

Subclasses may override this method to produce test names in a custom manner. One potential reason to override testNames is to run tests in a different order, for example, to ensure that tests that depend on other tests are run after those other tests. Another potential reason to override is allow tests to be defined in a different manner, such as methods annotated @Test annotations (as is done in JUnitSuite and TestNGSuite) or test functions registered during construction (as is done in FunSuite and FunSpec).

Executes the block of code passed as the second parameter, and, if it completes abruptly with a ModifiableMessage exception, prepends the "clue" string passed as the first parameter to the beginning of the detail message of that thrown exception, then rethrows it. If clue does not end in a white space character, one space will be added between it and the existing detail message (unless the detail message is not defined).

This method allows you to add more information about what went wrong that will be reported when a test fails. Here's an example:

withClue("(Employee's name was: " + employee.name + ")") {
  intercept[IllegalArgumentException] {
    employee.getTask(-1)
  }
}

If an invocation of intercept completed abruptly with an exception, the resulting message would be something like:

(Employee's name was Bob Jones) Expected IllegalArgumentException to be thrown, but no exception was thrown

Run the passed test function in the context of a fixture established by this method.

This method should set up the fixture needed by the tests of the current suite, invoke the test function, and if needed, perform any clean up needed after the test completes. Because the NoArgTest function passed to this method takes no parameters, preparing the fixture will require side effects, such as reassigning instance vars in this Suite or initializing a globally accessible external database. If you want to avoid reassigning instance vars you can use fixture.Suite.

This trait's implementation of runTest invokes this method for each test, passing in a NoArgTest whose apply method will execute the code of the test.

This trait's implementation of this method simply invokes the passed NoArgTest function.