Facilitates a “behavior-driven” style of development (BDD), in which tests are combined with text that specifies the behavior the tests verify.
Implementation trait for class AnyFlatSpec
, which facilitates a
“behavior-driven” style of development (BDD), in which tests
are combined with text that specifies the behavior the tests verify.
Implementation trait for class AnyFlatSpec
, which facilitates a
“behavior-driven” style of development (BDD), in which tests
are combined with text that specifies the behavior the tests verify.
AnyFlatSpec
is a class, not a trait,
to minimize compile time given there is a slight compiler overhead to
mixing in traits compared to extending classes. If you need to mix the
behavior of AnyFlatSpec
into some other class, you can use this
trait instead, because class AnyFlatSpec
does nothing more than
extend this trait and add a nice toString
implementation.
See the documentation of the class for a detailed
overview of AnyFlatSpec
.
Enables testing of asynchronous code without blocking,
using a style consistent with traditional AsyncFlatSpec
tests.
Enables testing of asynchronous code without blocking,
using a style consistent with traditional AsyncFlatSpec
tests.
Recommended Usage:
AsyncFlatSpec is intended to enable users of AnyFlatSpec
to write non-blocking asynchronous tests that are consistent with their traditional AnyFlatSpec tests.
Note: AsyncFlatSpec is intended for use in special situations where non-blocking asynchronous
testing is needed, with class AnyFlatSpec used for general needs.
|
Given a Future
returned by the code you are testing,
you need not block until the Future
completes before
performing assertions against its value. You can instead map those
assertions onto the Future
and return the resulting
Future[Assertion]
to ScalaTest. The test will complete
asynchronously, when the Future[Assertion]
completes.
Trait AsyncFlatSpec
is so named because
your specification text and tests line up flat against the left-side indentation level, with no nesting needed.
Here's an example AsyncFlatSpec
:
package org.scalatest.examples.asyncflatspec import org.scalatest.flatspec.AsyncFlatSpec import scala.concurrent.Future class AddSpec extends AsyncFlatSpec { def addSoon(addends: Int*): Future[Int] = Future { addends.sum } behavior of "addSoon" it should "eventually compute a sum of passed Ints" in { val futureSum: Future[Int] = addSoon(1, 2) // You can map assertions onto a Future, then return // the resulting Future[Assertion] to ScalaTest: futureSum map { sum => assert(sum == 3) } } def addNow(addends: Int*): Int = addends.sum "addNow" should "immediately compute a sum of passed Ints" in { val sum: Int = addNow(1, 2) // You can also write synchronous tests. The body // must have result type Assertion: assert(sum == 3) } }
The initial test in this example demonstrates the use of an explicit behavior of
clause, which establishes
addSoon
as the subject. The second test demonstrates the alternate syntax of replacing the first it
with the subject string, in this case, "addNow"
.
As with traditional AnyFlatSpec
s, you can use must
or can
as well as should
.
For example, instead of it should "eventually
..., you could write
it must "eventually
... or it can "eventually
....
You can also write they
instead of it
. See the documentation for AsyncFlatSpec
for
more detail.
Running the above AddSpec
in the Scala interpreter would yield:
addSoon
- should eventually compute a sum of passed Ints
- should immediately compute a sum of passed Ints
Starting with version 3.0.0, ScalaTest assertions and matchers have result type Assertion
.
The result type of the first test in the example above, therefore, is Future[Assertion]
.
For clarity, here's the relevant code in a REPL session:
scala> import org.scalatest._ import org.scalatest._ scala> import Assertions._ import Assertions._ scala> import scala.concurrent.Future import scala.concurrent.Future scala> import scala.concurrent.ExecutionContext import scala.concurrent.ExecutionContext scala> implicit val executionContext = ExecutionContext.Implicits.global executionContext: scala.concurrent.ExecutionContextExecutor = scala.concurrent.impl.ExecutionContextImpl@26141c5b scala> def addSoon(addends: Int*): Future[Int] = Future { addends.sum } addSoon: (addends: Int*)scala.concurrent.Future[Int] scala> val futureSum: Future[Int] = addSoon(1, 2) futureSum: scala.concurrent.Future[Int] = scala.concurrent.impl.Promise$DefaultPromise@721f47b2 scala> futureSum map { sum => assert(sum == 3) } res0: scala.concurrent.Future[org.scalatest.Assertion] = scala.concurrent.impl.Promise$DefaultPromise@3955cfcb
The second test has result type Assertion
:
scala> def addNow(addends: Int*): Int = addends.sum addNow: (addends: Int*)Int scala> val sum: Int = addNow(1, 2) sum: Int = 3 scala> assert(sum == 3) res1: org.scalatest.Assertion = Succeeded
When AddSpec
is constructed, the second test will be implicitly converted to
Future[Assertion]
and registered. The implicit conversion is from Assertion
to Future[Assertion]
, so you must end synchronous tests in some ScalaTest assertion
or matcher expression. If a test would not otherwise end in type Assertion
, you can
place succeed
at the end of the test. succeed
, a field in trait Assertions
,
returns the Succeeded
singleton:
scala> succeed res2: org.scalatest.Assertion = Succeeded
Thus placing succeed
at the end of a test body will satisfy the type checker:
"addNow" should "immediately compute a sum of passed Ints" in { val sum: Int = addNow(1, 2) assert(sum == 3) println("hi") // println has result type Unit succeed // succeed has result type Assertion }
An AsyncFlatSpec
's lifecycle has two phases: the registration phase and the
ready phase. It starts in registration phase and enters ready phase the first time
run
is called on it. It then remains in ready phase for the remainder of its lifetime.
Tests can only be registered with the it
method while the AsyncFlatSpec
is
in its registration phase. Any attempt to register a test after the AsyncFlatSpec
has
entered its ready phase, i.e., after run
has been invoked on the AsyncFlatSpec
,
will be met with a thrown TestRegistrationClosedException
. The recommended style
of using AsyncFlatSpec
is to register tests during object construction as is done in all
the examples shown here. If you keep to the recommended style, you should never see a
TestRegistrationClosedException
.
AsyncFlatSpec
extends AsyncTestSuite
, which provides an
implicit scala.concurrent.ExecutionContext
named executionContext
. This
execution context is used by AsyncFlatSpec
to
transform the Future[Assertion]
s returned by each test
into the FutureOutcome
returned by the test
function
passed to withFixture
.
This ExecutionContext
is also intended to be used in the tests,
including when you map assertions onto futures.
On both the JVM and Scala.js, the default execution context provided by ScalaTest's asynchronous
testing styles confines execution to a single thread per test. On JavaScript, where single-threaded
execution is the only possibility, the default execution context is
scala.scalajs.concurrent.JSExecutionContext.Implicits.queue
. On the JVM,
the default execution context is a serial execution context provided by ScalaTest itself.
When ScalaTest's serial execution context is called upon to execute a task, that task is recorded
in a queue for later execution. For example, one task that will be placed in this queue is the
task that transforms the Future[Assertion]
returned by an asynchronous test body
to the FutureOutcome
returned from the test
function.
Other tasks that will be queued are any transformations of, or callbacks registered on, Future
s that occur
in your test body, including any assertions you map onto Future
s. Once the test body returns,
the thread that executed the test body will execute the tasks in that queue one after another, in the order they
were enqueued.
ScalaTest provides its serial execution context as the default on the JVM for three reasons. First, most often
running both tests and suites in parallel does not give a significant performance boost compared to
just running suites in parallel. Thus parallel execution of Future
transformations within
individual tests is not generally needed for performance reasons.
Second, if multiple threads are operating in the same suite
concurrently, you'll need to make sure access to any mutable fixture objects by multiple threads is synchronized.
Although access to mutable state along
the same linear chain of Future
transformations need not be synchronized,
this does not hold true for callbacks, and in general it is easy to make a mistake. Simply put: synchronizing access to
shared mutable state is difficult and error prone.
Because ScalaTest's default execution context on the JVM confines execution of Future
transformations
and call backs to a single thread, you need not (by default) worry about synchronizing access to mutable state
in your asynchronous-style tests.
Third, asynchronous-style tests need not be complete when the test body returns, because the test body returns
a Future[Assertion]
. This Future[Assertion]
will often represent a test that has not yet
completed. As a result, when using a more traditional execution context backed by a thread-pool, you could
potentially start many more tests executing concurrently than there are threads in the thread pool. The more
concurrently execute tests you have competing for threads from the same limited thread pool, the more likely it
will be that tests will intermitently fail due to timeouts.
Using ScalaTest's serial execution context on the JVM will ensure the same thread that produced the Future[Assertion]
returned from a test body is also used to execute any tasks given to the execution context while executing the test
body—and that thread will not be allowed to do anything else until the test completes.
If the serial execution context's task queue ever becomes empty while the Future[Assertion]
returned by
that test's body has not yet completed, the thread will block until another task for that test is enqueued. Although
it may seem counter-intuitive, this blocking behavior means the total number of tests allowed to run concurrently will be limited
to the total number of threads executing suites. This fact means you can tune the thread pool such that maximum performance
is reached while avoiding (or at least, reducing the likelihood of) tests that fail due to timeouts because of thread competition.
This thread confinement strategy does mean, however, that when you are using the default execution context on the JVM, you
must be sure to never block in the test body waiting for a task to be completed by the
execution context. If you block, your test will never complete. This kind of problem will be obvious, because the test will
consistently hang every time you run it. (If a test is hanging, and you're not sure which one it is,
enable slowpoke notifications.) If you really do
want to block in your tests, you may wish to just use a
traditional AnyFlatSpec
with
ScalaFutures
instead. Alternatively, you could override
the executionContext
and use a traditional ExecutionContext
backed by a thread pool. This
will enable you to block in an asynchronous-style test on the JVM, but you'll need to worry about synchronizing access to
shared mutable state.
To use a different execution context, just override executionContext
. For example, if you prefer to use
the runNow
execution context on Scala.js instead of the default queue
, you would write:
// on Scala.js implicit override def executionContext = org.scalatest.concurrent.TestExecutionContext.runNow
If you prefer on the JVM to use the global execution context, which is backed by a thread pool, instead of ScalaTest's default serial execution contex, which confines execution to a single thread, you would write:
// on the JVM (and also compiles on Scala.js, giving // you the queue execution context) implicit override def executionContext = scala.concurrent.ExecutionContext.Implicits.global
By default (unless you mix in ParallelTestExecution
), tests in an AsyncFlatSpec
will be executed one after
another, i.e., serially. This is true whether those tests return Assertion
or Future[Assertion]
,
no matter what threads are involved. This default behavior allows
you to re-use a shared fixture, such as an external database that needs to be cleaned
after each test, in multiple tests in async-style suites. This is implemented by registering each test, other than the first test, to run
as a continuation after the previous test completes.
If you want the tests of an AsyncFlatSpec
to be executed in parallel, you
must mix in ParallelTestExecution
and enable parallel execution of tests in your build.
You enable parallel execution in Runner
with the -P
command line flag.
In the ScalaTest Maven Plugin, set parallel
to true
.
In sbt
, parallel execution is the default, but to be explicit you can write:
parallelExecution in Test := true // the default in sbt
On the JVM, if both ParallelTestExecution
is mixed in and
parallel execution is enabled in the build, tests in an async-style suite will be started in parallel, using threads from
the Distributor
, and allowed to complete in parallel, using threads from the
executionContext
. If you are using ScalaTest's serial execution context, the JVM default, asynchronous tests will
run in parallel very much like traditional (such as AnyFlatSpec
) tests run in
parallel: 1) Because ParallelTestExecution
extends
OneInstancePerTest
, each test will run in its own instance of the test class, you need not worry about synchronizing
access to mutable instance state shared by different tests in the same suite.
2) Because the serial execution context will confine the execution of each test to the single thread that executes the test body,
you need not worry about synchronizing access to shared mutable state accessed by transformations and callbacks of Future
s
inside the test.
If ParallelTestExecution
is mixed in but
parallel execution of suites is not enabled, asynchronous tests on the JVM will be started sequentially, by the single thread
that invoked run
, but without waiting for one test to complete before the next test is started. As a result,
asynchronous tests will be allowed to complete in parallel, using threads
from the executionContext
. If you are using the serial execution context, however, you'll see
the same behavior you see when parallel execution is disabled and a traditional suite that mixes in ParallelTestExecution
is executed: the tests will run sequentially. If you use an execution context backed by a thread-pool, such as global
,
however, even though tests will be started sequentially by one thread, they will be allowed to run concurrently using threads from the
execution context's thread pool.
The latter behavior is essentially what you'll see on Scala.js when you execute a suite that mixes in ParallelTestExecution
.
Because only one thread exists when running under JavaScript, you can't "enable parallel execution of suites." However, it may
still be useful to run tests in parallel on Scala.js, because tests can invoke API calls that are truly asynchronous by calling into
external APIs that take advantage of non-JavaScript threads. Thus on Scala.js, ParallelTestExecution
allows asynchronous
tests to run in parallel, even though they must be started sequentially. This may give you better performance when you are using API
calls in your Scala.js tests that are truly asynchronous.
If you need to test for expected exceptions in the context of futures, you can use the
recoverToSucceededIf
and recoverToExceptionIf
methods of trait
RecoverMethods
. Because this trait is mixed into
supertrait AsyncTestSuite
, both of these methods are
available by default in an AsyncFlatSpec
.
If you just want to ensure that a future fails with a particular exception type, and do
not need to inspect the exception further, use recoverToSucceededIf
:
recoverToSucceededIf[IllegalStateException] { // Result type: Future[Assertion] emptyStackActor ? Peek }
The recoverToSucceededIf
method performs a job similar to
assertThrows
, except
in the context of a future. It transforms a Future
of any type into a
Future[Assertion]
that succeeds only if the original future fails with the specified
exception. Here's an example in the REPL:
scala> import org.scalatest.RecoverMethods._ import org.scalatest.RecoverMethods._ scala> import scala.concurrent.Future import scala.concurrent.Future scala> import scala.concurrent.ExecutionContext.Implicits.global import scala.concurrent.ExecutionContext.Implicits.global scala> recoverToSucceededIf[IllegalStateException] { | Future { throw new IllegalStateException } | } res0: scala.concurrent.Future[org.scalatest.Assertion] = ... scala> res0.value res1: Option[scala.util.Try[org.scalatest.Assertion]] = Some(Success(Succeeded))
Otherwise it fails with an error message similar to those given by assertThrows
:
scala> recoverToSucceededIf[IllegalStateException] { | Future { throw new RuntimeException } | } res2: scala.concurrent.Future[org.scalatest.Assertion] = ... scala> res2.value res3: Option[scala.util.Try[org.scalatest.Assertion]] = Some(Failure(org.scalatest.exceptions.TestFailedException: Expected exception java.lang.IllegalStateException to be thrown, but java.lang.RuntimeException was thrown)) scala> recoverToSucceededIf[IllegalStateException] { | Future { 42 } | } res4: scala.concurrent.Future[org.scalatest.Assertion] = ... scala> res4.value res5: Option[scala.util.Try[org.scalatest.Assertion]] = Some(Failure(org.scalatest.exceptions.TestFailedException: Expected exception java.lang.IllegalStateException to be thrown, but no exception was thrown))
The recoverToExceptionIf
method differs from the recoverToSucceededIf
in
its behavior when the assertion succeeds: recoverToSucceededIf
yields a Future[Assertion]
,
whereas recoverToExceptionIf
yields a Future[T]
, where T
is the
expected exception type.
recoverToExceptionIf[IllegalStateException] { // Result type: Future[IllegalStateException] emptyStackActor ? Peek }
In other words, recoverToExpectionIf
is to
intercept
as
recovertToSucceededIf
is to assertThrows
. The first one allows you to
perform further assertions on the expected exception. The second one gives you a result type that will satisfy the type checker
at the end of the test body. Here's an example showing recoverToExceptionIf
in the REPL:
scala> val futureEx = | recoverToExceptionIf[IllegalStateException] { | Future { throw new IllegalStateException("hello") } | } futureEx: scala.concurrent.Future[IllegalStateException] = ... scala> futureEx.value res6: Option[scala.util.Try[IllegalStateException]] = Some(Success(java.lang.IllegalStateException: hello)) scala> futureEx map { ex => assert(ex.getMessage == "world") } res7: scala.concurrent.Future[org.scalatest.Assertion] = ... scala> res7.value res8: Option[scala.util.Try[org.scalatest.Assertion]] = Some(Failure(org.scalatest.exceptions.TestFailedException: "[hello]" did not equal "[world]"))
To support the common use case of temporarily disabling a test, with the
good intention of resurrecting the test at a later time, AsyncFlatSpec
provides two ways
to ignore a test, both demonstrated in the following example:
package org.scalatest.examples.asyncflatspec.ignore import org.scalatest.flatspec.AsyncFlatSpec import scala.concurrent.Future class AddSpec extends AsyncFlatSpec { def addSoon(addends: Int*): Future[Int] = Future { addends.sum } behavior of "addSoon" ignore should "eventually compute a sum of passed Ints" in { val futureSum: Future[Int] = addSoon(1, 2) // You can map assertions onto a Future, then return // the resulting Future[Assertion] to ScalaTest: futureSum map { sum => assert(sum == 3) } } def addNow(addends: Int*): Int = addends.sum "addNow" should "immediately compute a sum of passed Ints" ignore { val sum: Int = addNow(1, 2) // You can also write synchronous tests. The body // must have result type Assertion: assert(sum == 3) } }
In the first test, ignore
is used instead of it
.
In the second test, which uses the shorthand notation, no it
exists to change into ignore
.
To ignore such tests, you must instead change in
to ignore
, as shown in the above example.
If you run this version of AddSpec
with:
scala> org.scalatest.run(new AddSpec)
It will report both tests as ignored:
AddSpec: addSoon - should eventually compute a sum of passed Ints !!! IGNORED !!! addNow - should immediately compute a sum of passed Ints !!! IGNORED !!!
If you wish to temporarily ignore an entire suite of tests, you can (on the JVM, not Scala.js) annotate the test class with @Ignore
, like this:
package org.scalatest.examples.asyncflatspec.ignoreall import org.scalatest.flatspec.AsyncFlatSpec import scala.concurrent.Future import org.scalatest.Ignore @Ignore class AddSpec extends AsyncFlatSpec { def addSoon(addends: Int*): Future[Int] = Future { addends.sum } "addSoon" should "eventually compute a sum of passed Ints" in { val futureSum: Future[Int] = addSoon(1, 2) // You can map assertions onto a Future, then return // the resulting Future[Assertion] to ScalaTest: futureSum map { sum => assert(sum == 3) } } def addNow(addends: Int*): Int = addends.sum "addNow" should "immediately compute a sum of passed Ints" in { val sum: Int = addNow(1, 2) // You can also write synchronous tests. The body // must have result type Assertion: assert(sum == 3) } }
When you mark a test class with a tag annotation, ScalaTest will mark each test defined in that class with that tag.
Thus, marking the AddSpec
in the above example with the @Ignore
tag annotation means that both tests
in the class will be ignored. If you run the above AddSpec
in the Scala interpreter, you'll see:
AddSpec: addSoon - should eventually compute a sum of passed Ints !!! IGNORED !!! addNow - should immediately compute a sum of passed Ints !!! IGNORED !!!
Note that marking a test class as ignored won't prevent it from being discovered by ScalaTest. Ignored classes
will be discovered and run, and all their tests will be reported as ignored. This is intended to keep the ignored
class visible, to encourage the developers to eventually fix and “un-ignore” it. If you want to
prevent a class from being discovered at all (on the JVM, not Scala.js), use the DoNotDiscover
annotation instead.
If you want to ignore all tests of a suite on Scala.js, where annotations can't be inspected at runtime, you'll need
to change it
to ignore
at each test site. To make a suite non-discoverable on Scala.js, ensure it
does not declare a public no-arg constructor. You can either declare a public constructor that takes one or more
arguments, or make the no-arg constructor non-public. Because this technique will also make the suite non-discoverable
on the JVM, it is a good approach for suites you want to run (but not be discoverable) on both Scala.js and the JVM.
One of the parameters to AsyncFlatSpec
's run
method is a Reporter
, which
will collect and report information about the running suite of tests.
Information about suites and tests that were run, whether tests succeeded or failed,
and tests that were ignored will be passed to the Reporter
as the suite runs.
Most often the reporting done by default by AsyncFlatSpec
's methods will be sufficient, but
occasionally you may wish to provide custom information to the Reporter
from a test.
For this purpose, an Informer
that will forward information to the current Reporter
is provided via the info
parameterless method.
You can pass the extra information to the Informer
via its apply
method.
The Informer
will then pass the information to the Reporter
via an InfoProvided
event.
One use case for the Informer
is to pass more information about a specification to the reporter. For example,
the GivenWhenThen
trait provides methods that use the implicit info
provided by AsyncFlatSpec
to pass such information to the reporter. Here's an example:
package org.scalatest.examples.asyncflatspec.info import collection.mutable import org.scalatest._ class SetSpec extends flatspec.AsyncFlatSpec with GivenWhenThen { "A mutable Set" should "allow an element to be added" in { Given("an empty mutable Set") val set = mutable.Set.empty[String] When("an element is added") set += "clarity" Then("the Set should have size 1") assert(set.size === 1) And("the Set should contain the added element") assert(set.contains("clarity")) info("That's all folks!") succeed } }
If you run this AsyncFlatSpec
from the interpreter, you will see the following output:
scala> org.scalatest.run(new SetSpec)
SetSpec:
A mutable Set
- should allow an element to be added
+ Given an empty mutable Set
+ When an element is added
+ Then the Set should have size 1
+ And the Set should contain the added element
+ That's all folks!
AsyncFlatSpec
also provides a markup
method that returns a Documenter
, which allows you to send
to the Reporter
text formatted in Markdown syntax.
You can pass the extra information to the Documenter
via its apply
method.
The Documenter
will then pass the information to the Reporter
via an MarkupProvided
event.
Here's an example AsyncFlatSpec
that uses markup
:
package org.scalatest.examples.asyncflatspec.markup import collection.mutable import org.scalatest._ class SetSpec extends flatspec.AsyncFlatSpec with GivenWhenThen { markup { """ Mutable Set ----------- A set is a collection that contains no duplicate elements. To implement a concrete mutable set, you need to provide implementations of the following methods: def contains(elem: A): Boolean def iterator: Iterator[A] def += (elem: A): this.type def -= (elem: A): this.type If you wish that methods like `take`, `drop`, `filter` return the same kind of set, you should also override: def empty: This It is also good idea to override methods `foreach` and `size` for efficiency. """ } "A mutable Set" should "allow an element to be added" in { Given("an empty mutable Set") val set = mutable.Set.empty[String] When("an element is added") set += "clarity" Then("the Set should have size 1") assert(set.size === 1) And("the Set should contain the added element") assert(set.contains("clarity")) markup("This test finished with a **bold** statement!") succeed } }
Although all of ScalaTest's built-in reporters will display the markup text in some form,
the HTML reporter will format the markup information into HTML. Thus, the main purpose of markup
is to
add nicely formatted text to HTML reports. Here's what the above SetSpec
would look like in the HTML reporter:
ScalaTest records text passed to info
and markup
during tests, and sends the recorded text in the recordedEvents
field of
test completion events like TestSucceeded
and TestFailed
. This allows string reporters (like the standard out reporter) to show
info
and markup
text after the test name in a color determined by the outcome of the test. For example, if the test fails, string
reporters will show the info
and markup
text in red. If a test succeeds, string reporters will show the info
and markup
text in green. While this approach helps the readability of reports, it means that you can't use info
to get status
updates from long running tests.
To get immediate (i.e., non-recorded) notifications from tests, you can use note
(a Notifier
) and alert
(an Alerter
). Here's an example showing the differences:
package org.scalatest.examples.asyncflatspec.note import collection.mutable import org.scalatest._ class SetSpec extends flatspec.AsyncFlatSpec { "A mutable Set" should "allow an element to be added" in { info("info is recorded") markup("markup is *also* recorded") note("notes are sent immediately") alert("alerts are also sent immediately") val set = mutable.Set.empty[String] set += "clarity" assert(set.size === 1) assert(set.contains("clarity")) } }
Because note
and alert
information is sent immediately, it will appear before the test name in string reporters, and its color will
be unrelated to the ultimate outcome of the test: note
text will always appear in green, alert
text will always appear in yellow.
Here's an example:
scala> org.scalatest.run(new SetSpec) SetSpec: A mutable Set + notes are sent immediately + alerts are also sent immediately - should allow an element to be added + info is recorded + markup is *also* recorded
Another example is slowpoke notifications.
If you find a test is taking a long time to complete, but you're not sure which test, you can enable
slowpoke notifications. ScalaTest will use an Alerter
to fire an event whenever a test has been running
longer than a specified amount of time.
In summary, use info
and markup
for text that should form part of the specification output. Use
note
and alert
to send status notifications. (Because the HTML reporter is intended to produce a
readable, printable specification, info
and markup
text will appear in the HTML report, but
note
and alert
text will not.)
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, 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. 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, has not yet been implemented. Here's an example:
package org.scalatest.examples.asyncflatspec.pending import org.scalatest.flatspec.AsyncFlatSpec import scala.concurrent.Future class AddSpec extends AsyncFlatSpec { def addSoon(addends: Int*): Future[Int] = Future { addends.sum } "addSoon" should "eventually compute a sum of passed Ints" in (pending) def addNow(addends: Int*): Int = addends.sum "addNow" should "immediately compute a sum of passed Ints" in { val sum: Int = addNow(1, 2) // You can also write synchronous tests. The body // must have result type Assertion: assert(sum == 3) } }
(Note: "(pending)
" is the body of the test. Thus the test contains just one statement, an invocation
of the pending
method, which throws TestPendingException
.)
If you run this version of AddSpec
with:
scala> org.scalatest.run(new AddSpec)
It will run both tests, but report that first test is pending. You'll see:
AddSpec: addSoon - should eventually compute a sum of passed Ints (pending) addNow - should immediately compute a sum of passed Ints
One difference between an ignored test and a pending one is that an ignored test is intended to be used during significant refactorings of the code under test, when tests break and you don't want to spend the time to fix all of them immediately. You can mark some of those broken tests as ignored temporarily, so that you can focus the red bar on just failing tests you actually want to fix immediately. Later you can go back and fix the ignored tests. In other words, by ignoring some failing tests temporarily, you can more easily notice failed tests that you actually want to fix. By contrast, a pending test is intended to be used before a test and/or the code under test is written. Pending indicates you've decided to write a test for a bit of behavior, but either you haven't written the test yet, or have only written part of it, or perhaps you've written the test but don't want to implement the behavior it tests until after you've implemented a different bit of behavior you realized you need first. Thus ignored tests are designed to facilitate refactoring of existing code whereas pending tests are designed to facilitate the creation of new code.
One other difference between ignored and pending tests is that ignored tests are implemented as a test tag that is
excluded by default. Thus an ignored test is never executed. By contrast, a pending test is implemented as a
test that throws TestPendingException
(which is what calling the pending
method does). Thus
the body of pending tests are executed up until they throw TestPendingException
.
An AsyncFlatSpec
's tests may be classified into groups by tagging them with string names.
As with any suite, when executing an AsyncFlatSpec
, groups of tests can
optionally be included and/or excluded. To tag an AsyncFlatSpec
's tests,
you pass objects that extend class org.scalatest.Tag
to methods
that register tests. Class Tag
takes one parameter, a string name. If you have
created tag annotation interfaces as described in the Tag
documentation, then you
will probably want to use tag names on your test functions that match. To do so, simply
pass the fully qualified names of the tag interfaces to the Tag
constructor. For example, if you've
defined a tag annotation interface with fully qualified name,
com.mycompany.tags.DbTest
, then you could
create a matching tag for AsyncFlatSpec
s like this:
package org.scalatest.examples.asyncflatspec.tagging import org.scalatest.Tag object DbTest extends Tag("com.mycompany.tags.DbTest")
Given these definitions, you could place AsyncFlatSpec
tests into groups with tags like this:
import org.scalatest.flatspec.AsyncFlatSpec import org.scalatest.tagobjects.Slow import scala.concurrent.Future class AddSpec extends AsyncFlatSpec { def addSoon(addends: Int*): Future[Int] = Future { addends.sum } "addSoon" should "eventually compute a sum of passed Ints" taggedAs(Slow) in { val futureSum: Future[Int] = addSoon(1, 2) // You can map assertions onto a Future, then return // the resulting Future[Assertion] to ScalaTest: futureSum map { sum => assert(sum == 3) } } def addNow(addends: Int*): Int = addends.sum "addNow" should "immediately compute a sum of passed Ints" taggedAs(Slow, DbTest) in { val sum: Int = addNow(1, 2) // You can also write synchronous tests. The body // must have result type Assertion: assert(sum == 3) } }
This code marks both tests with the org.scalatest.tags.Slow
tag,
and the second test with the com.mycompany.tags.DbTest
tag.
The run
method takes a Filter
, whose constructor takes an optional
Set[String]
called tagsToInclude
and a Set[String]
called
tagsToExclude
. 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 tagsToExclude
,
will be run.
It is recommended, though not required, that you create a corresponding tag annotation when you
create a Tag
object. A tag annotation (on the JVM, not Scala.js) allows you to tag all the tests of an AsyncFlatSpec
in
one stroke by annotating the class. For more information and examples, see the
documentation for class Tag
. On Scala.js, to tag all tests of a suite, you'll need to
tag each test individually at the test site.
A test fixture is composed of the objects and other artifacts (files, sockets, database connections, etc.) tests use to do their work. When multiple tests need to work with the same fixtures, it is important to try and avoid duplicating the fixture code across those tests. The more code duplication you have in your tests, the greater drag the tests will have on refactoring the actual production code.
ScalaTest recommends three techniques to eliminate such code duplication in async styles:
withFixture
Each technique is geared towards helping you reduce code duplication without introducing
instance var
s, shared mutable objects, or other dependencies between tests. Eliminating shared
mutable state across tests will make your test code easier to reason about and eliminate the need to
synchronize access to shared mutable state on the JVM.
The following sections describe these techniques, including explaining the recommended usage for each. But first, here's a table summarizing the options:
Refactor using Scala when different tests need different fixtures. | |
get-fixture methods | The extract method refactor helps you create a fresh instances of mutable fixture objects in each test that needs them, but doesn't help you clean them up when you're done. |
loan-fixture methods | Factor out dupicate code with the loan pattern when different tests need different fixtures that must be cleaned up afterwards. |
Override withFixture when most or all tests need the same fixture.
|
|
withFixture(NoArgAsyncTest)
|
The recommended default approach when most or all tests need the same fixture treatment. This general technique
allows you, for example, to perform side effects at the beginning and end of all or most tests,
transform the outcome of tests, retry tests, make decisions based on test names, tags, or other test data.
Use this technique unless:
|
withFixture(OneArgAsyncTest)
|
Use when you want to pass the same fixture object or objects as a parameter into all or most tests. |
Mix in a before-and-after trait when you want an aborted suite, not a failed test, if the fixture code fails. | |
BeforeAndAfter
|
Use this boilerplate-buster when you need to perform the same side-effects before and/or after tests, rather than at the beginning or end of tests. |
BeforeAndAfterEach
|
Use when you want to stack traits that perform the same side-effects before and/or after tests, rather than at the beginning or end of tests. |
If you need to create the same mutable fixture objects in multiple tests, and don't need to clean them up after using them, the simplest approach is to write one or more get-fixture methods. A get-fixture method returns a new instance of a needed fixture object (or a holder object containing multiple fixture objects) each time it is called. You can call a get-fixture method at the beginning of each test that needs the fixture, storing the returned object or objects in local variables. Here's an example:
package org.scalatest.examples.asyncflatspec.getfixture import org.scalatest.flatspec.AsyncFlatSpec import scala.concurrent.Future class ExampleSpec extends AsyncFlatSpec { def fixture: Future[String] = Future { "ScalaTest is " } "Testing" should "be easy" in { val future = fixture val result = future map { s => s + "easy!" } result map { s => assert(s == "ScalaTest is easy!") } } it should "be fun" in { val future = fixture val result = future map { s => s + "fun!" } result map { s => assert(s == "ScalaTest is fun!") } } }
If you need to configure fixture objects differently in different tests, you can pass configuration into the get-fixture method. For example, you could pass in an initial value for a fixture object as a parameter to the get-fixture method.
withFixture(NoArgAsyncTest)
Although the get-fixture method approach takes care of setting up a fixture at the beginning of each
test, it doesn't address the problem of cleaning up a fixture at the end of the test. If you just need to perform a side-effect at the beginning or end of
a test, and don't need to actually pass any fixture objects into the test, you can override withFixture(NoArgAsyncTest)
, a
method defined in trait AsyncTestSuite
, a supertrait of AsyncFlatSpec
.
Trait AsyncFlatSpec
's runTest
method passes a no-arg async test function to
withFixture(NoArgAsyncTest)
. It is withFixture
's
responsibility to invoke that test function. The default implementation of withFixture
simply
invokes the function and returns the result, like this:
// Default implementation in trait AsyncTestSuite protected def withFixture(test: NoArgAsyncTest): FutureOutcome = { test() }
You can, therefore, override withFixture
to perform setup before invoking the test function,
and/or perform cleanup after the test completes. The recommended way to ensure cleanup is performed after a test completes is
to use the complete
-lastly
syntax, defined in supertrait CompleteLastly
.
The complete
-lastly
syntax will ensure that
cleanup will occur whether future-producing code completes abruptly by throwing an exception, or returns
normally yielding a future. In the latter case, complete
-lastly
will register the cleanup code
to execute asynchronously when the future completes.
The withFixture
method is designed to be stacked, and to enable this, you should always call the super
implementation
of withFixture
, and let it invoke the test function rather than invoking the test function directly. In other words, instead of writing
“test()
”, you should write “super.withFixture(test)
”, like this:
// Your implementation override def withFixture(test: NoArgAsyncTest) = { // Perform setup here complete { super.withFixture(test) // Invoke the test function } lastly { // Perform cleanup here } }
If you have no cleanup to perform, you can write withFixture
like this instead:
// Your implementation override def withFixture(test: NoArgAsyncTest) = { // Perform setup here super.withFixture(test) // Invoke the test function }
If you want to perform an action only for certain outcomes, you'll need to
register code performing that action as a callback on the Future
using
one of Future
's registration methods: onComplete
, onSuccess
,
or onFailure
. Note that if a test fails, that will be treated as a
scala.util.Success(org.scalatest.Failed)
. So if you want to perform an
action if a test fails, for example, you'd register the callback using onSuccess
.
Here's an example in which withFixture(NoArgAsyncTest)
is used to take a
snapshot of the working directory if a test fails, and
send that information to the standard output stream:
package org.scalatest.examples.asyncflatspec.noargasynctest import java.io.File import org.scalatest._ import scala.concurrent.Future class ExampleSpec extends flatspec.AsyncFlatSpec { override def withFixture(test: NoArgAsyncTest) = { super.withFixture(test) onFailedThen { _ => val currDir = new File(".") val fileNames = currDir.list() info("Dir snapshot: " + fileNames.mkString(", ")) } } def addSoon(addends: Int*): Future[Int] = Future { addends.sum } "This test" should "succeed" in { addSoon(1, 1) map { sum => assert(sum == 2) } } it should "fail" in { addSoon(1, 1) map { sum => assert(sum == 3) } } }
Running this version of ExampleSpec
in the interpreter in a directory with two files, hello.txt
and world.txt
would give the following output:
scala> org.scalatest.run(new ExampleSpec) ExampleSpec: This test - should succeed - should fail *** FAILED *** 2 did not equal 3 (:33)
Note that the NoArgAsyncTest
passed to withFixture
, in addition to
an apply
method that executes the test, also includes the test name and the config
map passed to runTest
. Thus you can also use the test name and configuration objects in your withFixture
implementation.
Lastly, if you want to transform the outcome in some way in withFixture
, you'll need to use either the
map
or transform
methods of Future
, like this:
// Your implementation override def withFixture(test: NoArgAsyncTest) = { // Perform setup here val futureOutcome = super.withFixture(test) // Invoke the test function futureOutcome change { outcome => // transform the outcome into a new outcome here } }
Note that a NoArgAsyncTest
's apply
method will return a scala.util.Failure
only if
the test completes abruptly with a "test-fatal" exception (such as OutOfMemoryError
) that should
cause the suite to abort rather than the test to fail. Thus usually you would use map
to transform future outcomes, not transform
, so that such test-fatal exceptions pass through
unchanged. The suite will abort asynchronously with any exception returned from NoArgAsyncTest
's
apply method in a scala.util.Failure
.
If you need to both pass a fixture object into a test and perform cleanup at the end of the test, you'll need to use the loan pattern. If different tests need different fixtures that require cleanup, you can implement the loan pattern directly by writing loan-fixture methods. A loan-fixture method takes a function whose body forms part or all of a test's code. It creates a fixture, passes it to the test code by invoking the function, then cleans up the fixture after the function returns.
The following example shows three tests that use two fixtures, a database and a file. Both require cleanup after, so each is provided via a
loan-fixture method. (In this example, the database is simulated with a StringBuffer
.)
package org.scalatest.examples.asyncflatspec.loanfixture import java.util.concurrent.ConcurrentHashMap import scala.concurrent.Future import scala.concurrent.ExecutionContext object DbServer { // Simulating a database server type Db = StringBuffer private final val databases = new ConcurrentHashMap[String, Db] def createDb(name: String): Db = { val db = new StringBuffer // java.lang.StringBuffer is thread-safe databases.put(name, db) db } def removeDb(name: String): Unit = { databases.remove(name) } } // Defining actor messages sealed abstract class StringOp case object Clear extends StringOp case class Append(value: String) extends StringOp case object GetValue class StringActor { // Simulating an actor private final val sb = new StringBuilder def !(op: StringOp): Unit = synchronized { op match { case Append(value) => sb.append(value) case Clear => sb.clear() } } def ?(get: GetValue.type)(implicit c: ExecutionContext): Future[String] = Future { synchronized { sb.toString } } } import org.scalatest._ import DbServer._ import java.util.UUID.randomUUID class ExampleSpec extends flatspec.AsyncFlatSpec { def withDatabase(testCode: Future[Db] => Future[Assertion]) = { val dbName = randomUUID.toString // generate a unique db name val futureDb = Future { createDb(dbName) } // create the fixture complete { val futurePopulatedDb = futureDb map { db => db.append("ScalaTest is ") // perform setup } testCode(futurePopulatedDb) // "loan" the fixture to the test code } lastly { removeDb(dbName) // ensure the fixture will be cleaned up } } def withActor(testCode: StringActor => Future[Assertion]) = { val actor = new StringActor complete { actor ! Append("ScalaTest is ") // set up the fixture testCode(actor) // "loan" the fixture to the test code } lastly { actor ! Clear // ensure the fixture will be cleaned up } } // This test needs the actor fixture "Testing" should "be productive" in { withActor { actor => actor ! Append("productive!") val futureString = actor ? GetValue futureString map { s => assert(s == "ScalaTest is productive!") } } } // This test needs the database fixture "Test code" should "be readable" in { withDatabase { futureDb => futureDb map { db => db.append("readable!") assert(db.toString == "ScalaTest is readable!") } } } // This test needs both the actor and the database it should "be clear and concise" in { withDatabase { futureDb => withActor { actor => // loan-fixture methods compose actor ! Append("concise!") val futureString = actor ? GetValue val futurePair: Future[(Db, String)] = futureDb zip futureString futurePair map { case (db, s) => db.append("clear!") assert(db.toString == "ScalaTest is clear!") assert(s == "ScalaTest is concise!") } } } } }
As demonstrated by the last test, loan-fixture methods compose. Not only do loan-fixture methods allow you to give each test the fixture it needs, they allow you to give a test multiple fixtures and clean everything up afterwards.
Also demonstrated in this example is the technique of giving each test its own "fixture sandbox" to play in. When your fixtures involve external side-effects, like creating databases, it is a good idea to give each database a unique name as is done in this example. This keeps tests completely isolated, allowing you to run them in parallel if desired.
withFixture(OneArgTest)
If all or most tests need the same fixture, you can avoid some of the boilerplate of the loan-fixture method approach by using a
fixture.AsyncTestSuite
and overriding withFixture(OneArgAsyncTest)
.
Each test in a fixture.AsyncTestSuite
takes a fixture as a parameter, allowing you to pass the fixture into
the test. You must indicate the type of the fixture parameter by specifying FixtureParam
, and implement a
withFixture
method that takes a OneArgAsyncTest
. This withFixture
method is responsible for
invoking the one-arg async test function, so you can perform fixture set up before invoking and passing
the fixture into the test function, and ensure clean up is performed after the test completes.
To enable the stacking of traits that define withFixture(NoArgAsyncTest)
, it is a good idea to let
withFixture(NoArgAsyncTest)
invoke the test function instead of invoking the test
function directly. To do so, you'll need to convert the OneArgAsyncTest
to a NoArgAsyncTest
. You can do that by passing
the fixture object to the toNoArgAsyncTest
method of OneArgAsyncTest
. In other words, instead of
writing “test(theFixture)
”, you'd delegate responsibility for
invoking the test function to the withFixture(NoArgAsyncTest)
method of the same instance by writing:
withFixture(test.toNoArgAsyncTest(theFixture))
Here's a complete example:
package org.scalatest.examples.asyncflatspec.oneargasynctest import org.scalatest._ import scala.concurrent.Future import scala.concurrent.ExecutionContext // Defining actor messages sealed abstract class StringOp case object Clear extends StringOp case class Append(value: String) extends StringOp case object GetValue class StringActor { // Simulating an actor private final val sb = new StringBuilder def !(op: StringOp): Unit = synchronized { op match { case Append(value) => sb.append(value) case Clear => sb.clear() } } def ?(get: GetValue.type)(implicit c: ExecutionContext): Future[String] = Future { synchronized { sb.toString } } } class ExampleSpec extends FixtureAsyncFlatSpec { type FixtureParam = StringActor def withFixture(test: OneArgAsyncTest): FutureOutcome = { val actor = new StringActor complete { actor ! Append("ScalaTest is ") // set up the fixture withFixture(test.toNoArgAsyncTest(actor)) } lastly { actor ! Clear // ensure the fixture will be cleaned up } } "Testing" should "be easy" in { actor => actor ! Append("easy!") val futureString = actor ? GetValue futureString map { s => assert(s == "ScalaTest is easy!") } } it should "be fun" in { actor => actor ! Append("fun!") val futureString = actor ? GetValue futureString map { s => assert(s == "ScalaTest is fun!") } } }
In this example, the tests required one fixture object, a StringActor
. If your tests need multiple fixture objects, you can
simply define the FixtureParam
type to be a tuple containing the objects or, alternatively, a case class containing
the objects. For more information on the withFixture(OneArgAsyncTest)
technique, see
the documentation for FixtureAsyncFlatSpec
.
BeforeAndAfter
In all the shared fixture examples shown so far, the activities of creating, setting up, and cleaning up the fixture objects have been
performed during the test. This means that if an exception occurs during any of these activities, it will be reported as a test failure.
Sometimes, however, you may want setup to happen before the test starts, and cleanup after the test has completed, so that if an
exception occurs during setup or cleanup, the entire suite aborts and no more tests are attempted. The simplest way to accomplish this in ScalaTest is
to mix in trait BeforeAndAfter
. With this trait you can denote a bit of code to run before each test
with before
and/or after each test each test with after
, like this:
package org.scalatest.examples.asyncflatspec.beforeandafter import org.scalatest.flatspec.AsyncFlatSpec import org.scalatest.BeforeAndAfter import scala.concurrent.Future import scala.concurrent.ExecutionContext // Defining actor messages sealed abstract class StringOp case object Clear extends StringOp case class Append(value: String) extends StringOp case object GetValue class StringActor { // Simulating an actor private final val sb = new StringBuilder def !(op: StringOp): Unit = synchronized { op match { case Append(value) => sb.append(value) case Clear => sb.clear() } } def ?(get: GetValue.type)(implicit c: ExecutionContext): Future[String] = Future { synchronized { sb.toString } } } class ExampleSpec extends AsyncFlatSpec with BeforeAndAfter { final val actor = new StringActor before { actor ! Append("ScalaTest is ") // set up the fixture } after { actor ! Clear // clean up the fixture } "Testing" should "be easy" in { actor ! Append("easy!") val futureString = actor ? GetValue futureString map { s => assert(s == "ScalaTest is easy!") } } it should "be fun" in { actor ! Append("fun!") val futureString = actor ? GetValue futureString map { s => assert(s == "ScalaTest is fun!") } } }
Note that the only way before
and after
code can communicate with test code is via some
side-effecting mechanism, commonly by reassigning instance var
s or by changing the state of mutable
objects held from instance val
s (as in this example). If using instance var
s or
mutable objects held from instance val
s you wouldn't be able to run tests in parallel in the same instance
of the test class (on the JVM, not Scala.js) unless you synchronized access to the shared, mutable state.
Note that on the JVM, if you override ScalaTest's default
serial execution context, you will likely need to
worry about synchronizing access to shared mutable fixture state, because the execution
context may assign different threads to process
different Future
transformations. Although access to mutable state along
the same linear chain of Future
transformations need not be synchronized,
it can be difficult to spot cases where these constraints are violated. The best approach
is to use only immutable objects when transforming Future
s. When that's not
practical, involve only thread-safe mutable objects, as is done in the above example.
On Scala.js, by contrast, you need not worry about thread synchronization, because
in effect only one thread exists.
Although BeforeAndAfter
provides a minimal-boilerplate way to execute code before and after tests, it isn't designed to enable stackable
traits, because the order of execution would be non-obvious. If you want to factor out before and after code that is common to multiple test suites, you
should use trait BeforeAndAfterEach
instead, as shown later in the next section,
composing fixtures by stacking traits.
In larger projects, teams often end up with several different fixtures that test classes need in different combinations,
and possibly initialized (and cleaned up) in different orders. A good way to accomplish this in ScalaTest is to factor the individual
fixtures into traits that can be composed using the stackable trait pattern. This can be done, for example, by placing
withFixture
methods in several traits, each of which call super.withFixture
. Here's an example in
which the StringBuilderActor
and StringBufferActor
fixtures used in the previous examples have been
factored out into two stackable fixture traits named Builder
and Buffer
:
package org.scalatest.examples.asyncflatspec.composingwithasyncfixture import org.scalatest._ import org.scalatest.SuiteMixin import collection.mutable.ListBuffer import scala.concurrent.Future import scala.concurrent.ExecutionContext // Defining actor messages sealed abstract class StringOp case object Clear extends StringOp case class Append(value: String) extends StringOp case object GetValue class StringBuilderActor { // Simulating an actor private final val sb = new StringBuilder def !(op: StringOp): Unit = synchronized { op match { case Append(value) => sb.append(value) case Clear => sb.clear() } } def ?(get: GetValue.type)(implicit c: ExecutionContext): Future[String] = Future { synchronized { sb.toString } } } class StringBufferActor { private final val buf = ListBuffer.empty[String] def !(op: StringOp): Unit = synchronized { op match { case Append(value) => buf += value case Clear => buf.clear() } } def ?(get: GetValue.type)(implicit c: ExecutionContext): Future[List[String]] = Future { synchronized { buf.toList } } } trait Builder extends AsyncTestSuiteMixin { this: AsyncTestSuite => final val builderActor = new StringBuilderActor abstract override def withFixture(test: NoArgAsyncTest) = { builderActor ! Append("ScalaTest is ") complete { super.withFixture(test) // To be stackable, must call super.withFixture } lastly { builderActor ! Clear } } } trait Buffer extends AsyncTestSuiteMixin { this: AsyncTestSuite => final val bufferActor = new StringBufferActor abstract override def withFixture(test: NoArgAsyncTest) = { complete { super.withFixture(test) // To be stackable, must call super.withFixture } lastly { bufferActor ! Clear } } } class ExampleSpec extends flatspec.AsyncFlatSpec with Builder with Buffer { "Testing" should "be easy" in { builderActor ! Append("easy!") val futureString = builderActor ? GetValue val futureList = bufferActor ? GetValue val futurePair: Future[(String, List[String])] = futureString zip futureList futurePair map { case (str, lst) => assert(str == "ScalaTest is easy!") assert(lst.isEmpty) bufferActor ! Append("sweet") succeed } } it should "be fun" in { builderActor ! Append("fun!") val futureString = builderActor ? GetValue val futureList = bufferActor ? GetValue val futurePair: Future[(String, List[String])] = futureString zip futureList futurePair map { case (str, lst) => assert(str == "ScalaTest is fun!") assert(lst.isEmpty) bufferActor ! Append("awesome") succeed } } }
By mixing in both the Builder
and Buffer
traits, ExampleSpec
gets both fixtures, which will be
initialized before each test and cleaned up after. The order the traits are mixed together determines the order of execution.
In this case, Builder
is “super” to Buffer
. If you wanted Buffer
to be “super”
to Builder
, you need only switch the order you mix them together, like this:
class Example2Spec extends flatspec.AsyncFlatSpec with Buffer with Builder
If you only need one fixture you mix in only that trait:
class Example3Spec extends flatspec.AsyncFlatSpec with Builder
Another way to create stackable fixture traits is by extending the BeforeAndAfterEach
and/or BeforeAndAfterAll
traits.
BeforeAndAfterEach
has a beforeEach
method that will be run before each test (like JUnit's setUp
),
and an afterEach
method that will be run after (like JUnit's tearDown
).
Similarly, BeforeAndAfterAll
has a beforeAll
method that will be run before all tests,
and an afterAll
method that will be run after all tests. Here's what the previously shown example would look like if it
were rewritten to use the BeforeAndAfterEach
methods instead of withFixture
:
package org.scalatest.examples.asyncflatspec.composingbeforeandaftereach import org.scalatest._ import org.scalatest.BeforeAndAfterEach import collection.mutable.ListBuffer import scala.concurrent.Future import scala.concurrent.ExecutionContext // Defining actor messages sealed abstract class StringOp case object Clear extends StringOp case class Append(value: String) extends StringOp case object GetValue class StringBuilderActor { // Simulating an actor private final val sb = new StringBuilder def !(op: StringOp): Unit = synchronized { op match { case Append(value) => sb.append(value) case Clear => sb.clear() } } def ?(get: GetValue.type)(implicit c: ExecutionContext): Future[String] = Future { synchronized { sb.toString } } } class StringBufferActor { private final val buf = ListBuffer.empty[String] def !(op: StringOp): Unit = synchronized { op match { case Append(value) => buf += value case Clear => buf.clear() } } def ?(get: GetValue.type)(implicit c: ExecutionContext): Future[List[String]] = Future { synchronized { buf.toList } } } trait Builder extends BeforeAndAfterEach { this: Suite => final val builderActor = new StringBuilderActor override def beforeEach() { builderActor ! Append("ScalaTest is ") super.beforeEach() // To be stackable, must call super.beforeEach } override def afterEach() { try super.afterEach() // To be stackable, must call super.afterEach finally builderActor ! Clear } } trait Buffer extends BeforeAndAfterEach { this: Suite => final val bufferActor = new StringBufferActor override def afterEach() { try super.afterEach() // To be stackable, must call super.afterEach finally bufferActor ! Clear } } class ExampleSpec extends flatspec.AsyncFlatSpec with Builder with Buffer { "Testing" should "be easy" in { builderActor ! Append("easy!") val futureString = builderActor ? GetValue val futureList = bufferActor ? GetValue val futurePair: Future[(String, List[String])] = futureString zip futureList futurePair map { case (str, lst) => assert(str == "ScalaTest is easy!") assert(lst.isEmpty) bufferActor ! Append("sweet") succeed } } it should "be fun" in { builderActor ! Append("fun!") val futureString = builderActor ? GetValue val futureList = bufferActor ? GetValue val futurePair: Future[(String, List[String])] = futureString zip futureList futurePair map { case (str, lst) => assert(str == "ScalaTest is fun!") assert(lst.isEmpty) bufferActor ! Append("awesome") succeed } } }
To get the same ordering as withFixture
, place your super.beforeEach
call at the end of each
beforeEach
method, and the super.afterEach
call at the beginning of each afterEach
method, as shown in the previous example. It is a good idea to invoke super.afterEach
in a try
block and perform cleanup in a finally
clause, as shown in the previous example, because this ensures the
cleanup code is performed even if super.afterEach
throws an exception.
The difference between stacking traits that extend BeforeAndAfterEach
versus traits that implement withFixture
is
that setup and cleanup code happens before and after the test in BeforeAndAfterEach
, but at the beginning and
end of the test in withFixture
. Thus if a withFixture
method completes abruptly with an exception, it is
considered a failed test. By contrast, if any of the beforeEach
or afterEach
methods of BeforeAndAfterEach
complete abruptly, it is considered an aborted suite, which will result in a SuiteAborted
event.
Sometimes you may want to run the same test code on different fixture objects. In other words, you may want to write tests that are "shared"
by different fixture objects.
To accomplish this in an AsyncFlatSpec
, you first place shared tests in
behavior functions. These behavior functions will be
invoked during the construction phase of any AsyncFlatSpec
that uses them, so that the tests they contain will
be registered as tests in that AsyncFlatSpec
.
For example, given this StackActor
class:
package org.scalatest.examples.asyncflatspec.sharedtests import scala.collection.mutable.ListBuffer import scala.concurrent.Future import scala.concurrent.ExecutionContext // Stack operations case class Push[T](value: T) sealed abstract class StackOp case object Pop extends StackOp case object Peek extends StackOp case object Size extends StackOp // Stack info case class StackInfo[T](top: Option[T], size: Int, max: Int) { require(size > 0, "size was less than zero") require(max > size, "max was less than size") val isFull: Boolean = size == max val isEmpty: Boolean = size == 0 } class StackActor[T](Max: Int, name: String) { private final val buf = new ListBuffer[T] def !(push: Push[T]): Unit = synchronized { if (buf.size != Max) buf.prepend(push.value) else throw new IllegalStateException("can't push onto a full stack") } def ?(op: StackOp)(implicit c: ExecutionContext): Future[StackInfo[T]] = synchronized { op match { case Pop => Future { if (buf.size != 0) StackInfo(Some(buf.remove(0)), buf.size, Max) else throw new IllegalStateException("can't pop an empty stack") } case Peek => Future { if (buf.size != 0) StackInfo(Some(buf(0)), buf.size, Max) else throw new IllegalStateException("can't peek an empty stack") } case Size => Future { StackInfo(None, buf.size, Max) } } } override def toString: String = name }
You may want to test the stack represented by the StackActor
class in different states: empty, full, with one item, with one item less than capacity,
etc. You may find you have several tests that make sense any time the stack is non-empty. Thus you'd ideally want to run
those same tests for three stack fixture objects: a full stack, a stack with a one item, and a stack with one item less than
capacity. With shared tests, you can factor these tests out into a behavior function, into which you pass the
stack fixture to use when running the tests. So in your AsyncFlatSpec
for StackActor
, you'd invoke the
behavior function three times, passing in each of the three stack fixtures so that the shared tests are run for all three fixtures.
You can define a behavior function that encapsulates these shared tests inside the AsyncFlatSpec
that uses them. If they are shared
between different AsyncFlatSpec
s, however, you could also define them in a separate trait that is mixed into
each AsyncFlatSpec
that uses them.
For example, here the nonEmptyStackActor
behavior function (in this case, a
behavior method) is defined in a trait along with another
method containing shared tests for non-full stacks:
import org.scalatest.AsyncFlatSpec trait AsyncFlatSpecStackBehaviors { this: AsyncFlatSpec => def nonEmptyStackActor(createNonEmptyStackActor: => StackActor[Int], lastItemAdded: Int, name: String): Unit = { it should ("return non-empty StackInfo when Size is fired at non-empty stack actor: " + name) in { val stackActor = createNonEmptyStackActor val futureStackInfo = stackActor ? Size futureStackInfo map { stackInfo => assert(!stackInfo.isEmpty) } } it should ("return before and after StackInfo that has existing size and lastItemAdded as top when Peek is fired at non-empty stack actor: " + name) in { val stackActor = createNonEmptyStackActor val futurePair: Future[(StackInfo[Int], StackInfo[Int])] = for { beforePeek <- stackActor ? Size afterPeek <- stackActor ? Peek } yield (beforePeek, afterPeek) futurePair map { case (beforePeek, afterPeek) => assert(afterPeek.top == Some(lastItemAdded)) assert(afterPeek.size == beforePeek.size) } } it should ("return before and after StackInfo that has existing size - 1 and lastItemAdded as top when Pop is fired at non-empty stack actor: " + name) in { val stackActor = createNonEmptyStackActor val futurePair: Future[(StackInfo[Int], StackInfo[Int])] = for { beforePop <- stackActor ? Size afterPop <- stackActor ? Pop } yield (beforePop, afterPop) futurePair map { case (beforePop, afterPop) => assert(afterPop.top == Some(lastItemAdded)) assert(afterPop.size == beforePop.size - 1) } } } def nonFullStackActor(createNonFullStackActor: => StackActor[Int], name: String): Unit = { it should ("return non-full StackInfo when Size is fired at non-full stack actor: " + name) in { val stackActor = createNonFullStackActor val futureStackInfo = stackActor ? Size futureStackInfo map { stackInfo => assert(!stackInfo.isFull) } } it should ("return before and after StackInfo that has existing size + 1 and new item as top when Push is fired at non-full stack actor: " + name) in { val stackActor = createNonFullStackActor val futurePair: Future[(StackInfo[Int], StackInfo[Int])] = for { beforePush <- stackActor ? Size afterPush <- { stackActor ! Push(7); stackActor ? Peek } } yield (beforePush, afterPush) futurePair map { case (beforePush, afterPush) => assert(afterPush.top == Some(7)) assert(afterPush.size == beforePush.size + 1) } } } }
Given these behavior functions, you could invoke them directly, but AsyncFlatSpec
offers a DSL for the purpose,
which looks like this:
it should behave like nonEmptyStackActor(almostEmptyStackActor, LastValuePushed, almostEmptyStackActorName) it should behave like nonFullStackActor(almostEmptyStackActor, almostEmptyStackActorName)
Here's an example:
class StackSpec extends AsyncFlatSpec with AsyncFlatSpecStackBehaviors { val Max = 10 val LastValuePushed = Max - 1 // Stack fixture creation methods val emptyStackActorName = "empty stack actor" def emptyStackActor = new StackActor[Int](Max, emptyStackActorName ) val fullStackActorName = "full stack actor" def fullStackActor = { val stackActor = new StackActor[Int](Max, fullStackActorName ) for (i <- 0 until Max) stackActor ! Push(i) stackActor } val almostEmptyStackActorName = "almost empty stack actor" def almostEmptyStackActor = { val stackActor = new StackActor[Int](Max, almostEmptyStackActorName ) stackActor ! Push(LastValuePushed) stackActor } val almostFullStackActorName = "almost full stack actor" def almostFullStackActor = { val stackActor = new StackActor[Int](Max, almostFullStackActorName) for (i <- 1 to LastValuePushed) stackActor ! Push(i) stackActor } "A Stack actor (when empty)" should "return empty StackInfo when Size is fired at it" in { val stackActor = emptyStackActor val futureStackInfo = stackActor ? Size futureStackInfo map { stackInfo => assert(stackInfo.isEmpty) } } it should "complain when Peek is fired at it" in { recoverToSucceededIf[IllegalStateException] { emptyStackActor ? Peek } } it should "complain when Pop is fired at it" in { recoverToSucceededIf[IllegalStateException] { emptyStackActor ? Pop } } "A Stack actor (when non-empty)" should behave like nonEmptyStackActor(almostEmptyStackActor, LastValuePushed, almostEmptyStackActorName) it should behave like nonFullStackActor(almostEmptyStackActor, almostEmptyStackActorName) it should behave like nonEmptyStackActor(almostFullStackActor, LastValuePushed, almostFullStackActorName) it should behave like nonFullStackActor(almostFullStackActor, almostFullStackActorName) "A Stack actor (when full)" should "return full StackInfo when Size is fired at it" in { val stackActor = fullStackActor val futureStackInfo = stackActor ? Size futureStackInfo map { stackInfo => assert(stackInfo.isFull) } } it should behave like nonEmptyStackActor(fullStackActor, LastValuePushed, fullStackActorName) it should "complain when Push is fired at it" in { val stackActor = fullStackActor assertThrows[IllegalStateException] { stackActor ! Push(10) } } }
If you load these classes into the Scala interpreter (with scalatest's JAR file on the class path), and execute it, you'll see:
scala> org.scalatest.run(new StackSpec)
StackSpec:
A Stack actor (when empty)
- should return empty StackInfo when Size is fired at it
- should complain when Peek is fired at it
- should complain when Pop is fired at it
A Stack actor (when non-empty)
- should return non-empty StackInfo when Size is fired at non-empty stack actor: almost empty stack actor
- should return before and after StackInfo that has existing size and lastItemAdded as top when Peek is fired at non-empty stack actor: almost empty stack actor
- should return before and after StackInfo that has existing size - 1 and lastItemAdded as top when Pop is fired at non-empty stack actor: almost empty stack actor
- should return non-full StackInfo when Size is fired at non-full stack actor: almost empty stack actor
- should return before and after StackInfo that has existing size + 1 and new item as top when Push is fired at non-full stack actor: almost empty stack actor
- should return non-empty StackInfo when Size is fired at non-empty stack actor: almost full stack actor
- should return before and after StackInfo that has existing size and lastItemAdded as top when Peek is fired at non-empty stack actor: almost full stack actor
- should return before and after StackInfo that has existing size - 1 and lastItemAdded as top when Pop is fired at non-empty stack actor: almost full stack actor
- should return non-full StackInfo when Size is fired at non-full stack actor: almost full stack actor
- should return before and after StackInfo that has existing size + 1 and new item as top when Push is fired at non-full stack actor: almost full stack actor
A Stack actor (when full)
- should return full StackInfo when Size is fired at it
- should return non-empty StackInfo when Size is fired at non-empty stack actor: full stack actor
- should return before and after StackInfo that has existing size and lastItemAdded as top when Peek is fired at non-empty stack actor: full stack actor
- should return before and after StackInfo that has existing size - 1 and lastItemAdded as top when Pop is fired at non-empty stack actor: full stack actor
- should complain when Push is fired at it
One thing to keep in mind when using shared tests is that in ScalaTest, each test in a suite must have a unique name.
If you register the same tests repeatedly in the same suite, one problem you may encounter is an exception at runtime
complaining that multiple tests are being registered with the same test name.
Although in an AsyncFlatSpec
, the behavior of
clause is a nesting construct analogous to
AsyncFunSpec
's describe
clause, you many sometimes need to do a bit of
extra work to ensure that the test names are unique. If a duplicate test name problem shows up in an
AsyncFlatSpec
, you'll need to pass in a prefix or suffix string to add to each test name. You can call
toString
on the shared fixture object, or pass this string
the same way you pass any other data needed by the shared tests.
This is the approach taken by the previous AsyncFlatSpecStackBehaviors
example.
Given this AsyncFlatSpecStackBehaviors
trait, calling it with the almostEmptyStackActor
fixture, like this:
"A Stack actor (when non-empty)" should behave like nonEmptyStackActor(almostEmptyStackActor, LastValuePushed, almostEmptyStackActorName)
yields test names:
A Stack actor (when non-empty) should return non-empty StackInfo when Size is fired at non-empty stack actor: almost empty stack actor
A Stack actor (when non-empty) should return before and after StackInfo that has existing size and lastItemAdded as top when Peek is fired at non-empty stack actor: almost empty stack actor
A Stack actor (when non-empty) should return before and after StackInfo that has existing size - 1 and lastItemAdded as top when Pop is fired at non-empty stack actor: almost empty stack actor
Whereas calling it with the almostFullStackActor
fixture, like this:
it should behave like nonEmptyStackActor(almostFullStackActor, LastValuePushed, almostFullStackActorName)
yields different test names:
A Stack actor (when non-empty) should return non-empty StackInfo when Size is fired at non-empty stack actor: almost full stack actor
A Stack actor (when non-empty) should return before and after StackInfo that has existing size and lastItemAdded as top when Peek is fired at non-empty stack actor: almost full stack actor
A Stack actor (when non-empty) should return before and after StackInfo that has existing size - 1 and lastItemAdded as top when Pop is fired at non-empty stack actor: almost full stack actor
Implementation trait for class AsyncFlatSpec
, which facilitates a
“behavior-driven” style of development (BDD), in which tests
are combined with text that specifies the behavior the tests verify.
Implementation trait for class AsyncFlatSpec
, which facilitates a
“behavior-driven” style of development (BDD), in which tests
are combined with text that specifies the behavior the tests verify.
AsyncFlatSpec
is a class, not a trait,
to minimize compile time given there is a slight compiler overhead to
mixing in traits compared to extending classes. If you need to mix the
behavior of AsyncFlatSpec
into some other class, you can use this
trait instead, because class AsyncFlatSpec
does nothing more than
extend this trait and add a nice toString
implementation.
See the documentation of the class for a detailed
overview of AsyncFlatSpec
.
A sister class to org.scalatest.flatspec.AnyFlatSpec
that can pass a fixture object into its tests.
A sister class to org.scalatest.flatspec.AnyFlatSpec
that can pass a fixture object into its tests.
Recommended Usage:
Use class FixtureAnyFlatSpec in situations for which AnyFlatSpec
would be a good choice, when all or most tests need the same fixture objects
that must be cleaned up afterwards. Note: FixtureAnyFlatSpec is intended for use in special situations, with class AnyFlatSpec used for general needs. For
more insight into where FixtureAnyFlatSpec fits in the big picture, see the withFixture(OneArgTest) subsection of the Shared fixtures section in the documentation for class AnyFlatSpec .
|
Class FixtureAnyFlatSpec
behaves similarly to class org.scalatest.flatspec.AnyFlatSpec
, except that tests may have a
fixture parameter. The type of the
fixture parameter is defined by the abstract FixtureParam
type, which is a member of this class.
This class also contains an abstract withFixture
method. This withFixture
method
takes a OneArgTest
, which is a nested trait defined as a member of this class.
OneArgTest
has an apply
method that takes a FixtureParam
.
This apply
method is responsible for running a test.
This class's runTest
method delegates the actual running of each test to withFixture(OneArgTest)
, passing
in the test code to run via the OneArgTest
argument. The withFixture(OneArgTest)
method (abstract in this class) is responsible
for creating the fixture argument and passing it to the test function.
Subclasses of this class must, therefore, do three things differently from a plain old org.scalatest.flatspec.AnyFlatSpec
:
FixtureParam
withFixture(OneArgTest)
methodIf the fixture you want to pass into your tests consists of multiple objects, you will need to combine them into one object to use this class. One good approach to passing multiple fixture objects is to encapsulate them in a case class. Here's an example:
case class FixtureParam(file: File, writer: FileWriter)
To enable the stacking of traits that define withFixture(NoArgTest)
, it is a good idea to let
withFixture(NoArgTest)
invoke the test function instead of invoking the test
function directly. To do so, you'll need to convert the OneArgTest
to a NoArgTest
. You can do that by passing
the fixture object to the toNoArgTest
method of OneArgTest
. In other words, instead of
writing “test(theFixture)
”, you'd delegate responsibility for
invoking the test function to the withFixture(NoArgTest)
method of the same instance by writing:
withFixture(test.toNoArgTest(theFixture))
Here's a complete example:
package org.scalatest.examples.flatspec.oneargtest import org.scalatest.flatspec import java.io._ class ExampleSpec extends flatspec.FixtureAnyFlatSpec { case class FixtureParam(file: File, writer: FileWriter) def withFixture(test: OneArgTest) = { // create the fixture val file = File.createTempFile("hello", "world") val writer = new FileWriter(file) val theFixture = FixtureParam(file, writer) try { writer.write("ScalaTest is ") // set up the fixture withFixture(test.toNoArgTest(theFixture)) // "loan" the fixture to the test } finally writer.close() // clean up the fixture } "Testing" should "be easy" in { f => f.writer.write("easy!") f.writer.flush() assert(f.file.length === 18) } it should "be fun" in { f => f.writer.write("fun!") f.writer.flush() assert(f.file.length === 17) } }
If a test fails because of an exception, the OneArgTest
function will result in a Failed wrapping the exception.
To ensure clean up happens even if an exception occurs, you should invoke the test function from inside a try
block and do the cleanup in a
finally
clause, as shown in the previous example.
If multiple test classes need the same fixture, you can define the FixtureParam
and withFixture(OneArgTest)
implementations
in a trait, then mix that trait into the test classes that need it. For example, if your application requires a database and your integration tests
use that database, you will likely have many test classes that need a database fixture. You can create a "database fixture" trait that creates a
database with a unique name, passes the connector into the test, then removes the database once the test completes. This is shown in the following example:
package org.scalatest.examples.fixture.flatspec.sharing import java.util.concurrent.ConcurrentHashMap import org.scalatest._ import DbServer._ import java.util.UUID.randomUUID object DbServer { // Simulating a database server type Db = StringBuffer private val databases = new ConcurrentHashMap[String, Db] def createDb(name: String): Db = { val db = new StringBuffer databases.put(name, db) db } def removeDb(name: String) { databases.remove(name) } } trait DbFixture { this: fixture.Suite => type FixtureParam = Db // Allow clients to populate the database after // it is created def populateDb(db: Db) {} def withFixture(test: OneArgTest) = { val dbName = randomUUID.toString val db = createDb(dbName) // create the fixture try { populateDb(db) // setup the fixture withFixture(test.toNoArgTest(db)) // "loan" the fixture to the test } finally removeDb(dbName) // clean up the fixture } } class ExampleSpec extends flatspec.FixtureAnyFlatSpec with DbFixture { override def populateDb(db: Db) { // setup the fixture db.append("ScalaTest is ") } "Testing" should "be easy" in { db => db.append("easy!") assert(db.toString === "ScalaTest is easy!") } it should "be fun" in { db => db.append("fun!") assert(db.toString === "ScalaTest is fun!") } // This test doesn't need a Db "Test code" should "be clear" in { () => val buf = new StringBuffer buf.append("ScalaTest code is ") buf.append("clear!") assert(buf.toString === "ScalaTest code is clear!") } }
Often when you create fixtures in a trait like DbFixture
, you'll still need to enable individual test classes
to "setup" a newly created fixture before it gets passed into the tests. A good way to accomplish this is to pass the newly
created fixture into a setup method, like populateDb
in the previous example, before passing it to the test
function. Classes that need to perform such setup can override the method, as does ExampleSpec
.
If a test doesn't need the fixture, you can indicate that by providing a no-arg instead of a one-arg function, as is done in the
third test in the previous example, “Test code should be clear
”. In other words, instead of starting your function literal
with something like “db =>
”, you'd start it with “() =>
”. For such tests, runTest
will not invoke withFixture(OneArgTest)
. It will instead directly invoke withFixture(NoArgTest)
.
Both examples shown above demonstrate the technique of giving each test its own "fixture sandbox" to play in. When your fixtures
involve external side-effects, like creating files or databases, it is a good idea to give each file or database a unique name as is
done in these examples. This keeps tests completely isolated, allowing you to run them in parallel if desired. You could mix
ParallelTestExecution
into either of these ExampleSpec
classes, and the tests would run in parallel just fine.
Implementation trait for class FixtureAnyFlatSpec
, which is
a sister class to org.scalatest.flatspec.AnyFlatSpec
that can pass a
fixture object into its tests.
Implementation trait for class FixtureAnyFlatSpec
, which is
a sister class to org.scalatest.flatspec.AnyFlatSpec
that can pass a
fixture object into its tests.
FixtureAnyFlatSpec
is a class,
not a trait, to minimize compile time given there is a slight compiler
overhead to mixing in traits compared to extending classes. If you need
to mix the behavior of FixtureAnyFlatSpec
into some other
class, you can use this trait instead, because class
FixtureAnyFlatSpec
does nothing more than extend this trait and add a nice toString
implementation.
See the documentation of the class for a detailed
overview of FixtureAnyFlatSpec
.
A sister class to org.scalatest.flatspec.AsyncFlatSpec
that can pass a fixture object into its tests.
A sister class to org.scalatest.flatspec.AsyncFlatSpec
that can pass a fixture object into its tests.
Recommended Usage:
Use class FixtureAsyncFlatSpec in situations for which AsyncFlatSpec
would be a good choice, when all or most tests need the same fixture objects
that must be cleaned up afterwards. Note: FixtureAsyncFlatSpec is intended for use in special situations, with class AsyncFlatSpec used for general needs. For
more insight into where FixtureAsyncFlatSpec fits in the big picture, see the withFixture(OneArgAsyncTest) subsection of the Shared fixtures section in the documentation for class AsyncFlatSpec .
|
Class FixtureAsyncFlatSpec
behaves similarly to class org.scalatest.flatspec.AsyncFlatSpec
, except that tests may have a
fixture parameter. The type of the
fixture parameter is defined by the abstract FixtureParam
type, which is a member of this class.
This class also contains an abstract withFixture
method. This withFixture
method
takes a OneArgAsyncTest
, which is a nested trait defined as a member of this class.
OneArgAsyncTest
has an apply
method that takes a FixtureParam
.
This apply
method is responsible for running a test.
This class's runTest
method delegates the actual running of each test to withFixture(OneArgAsyncTest)
, passing
in the test code to run via the OneArgAsyncTest
argument. The withFixture(OneArgAsyncTest)
method (abstract in this class) is responsible
for creating the fixture argument and passing it to the test function.
Subclasses of this class must, therefore, do three things differently from a plain old org.scalatest.flatspec.AsyncFlatSpec
:
FixtureParam
withFixture(OneArgAsyncTest)
methodIf the fixture you want to pass into your tests consists of multiple objects, you will need to combine them into one object to use this class. One good approach to passing multiple fixture objects is to encapsulate them in a case class. Here's an example:
case class FixtureParam(file: File, writer: FileWriter)
To enable the stacking of traits that define withFixture(NoArgAsyncTest)
, it is a good idea to let
withFixture(NoArgAsyncTest)
invoke the test function instead of invoking the test
function directly. To do so, you'll need to convert the OneArgAsyncTest
to a NoArgAsyncTest
. You can do that by passing
the fixture object to the toNoArgAsyncTest
method of OneArgAsyncTest
. In other words, instead of
writing “test(theFixture)
”, you'd delegate responsibility for
invoking the test function to the withFixture(NoArgAsyncTest)
method of the same instance by writing:
withFixture(test.toNoArgAsyncTest(theFixture))
Here's a complete example:
package org.scalatest.examples.asyncflatspec.oneargasynctest import org.scalatest._ import scala.concurrent.Future import scala.concurrent.ExecutionContext // Defining actor messages sealed abstract class StringOp case object Clear extends StringOp case class Append(value: String) extends StringOp case object GetValue class StringActor { // Simulating an actor private final val sb = new StringBuilder def !(op: StringOp): Unit = synchronized { op match { case Append(value) => sb.append(value) case Clear => sb.clear() } } def ?(get: GetValue.type)(implicit c: ExecutionContext): Future[String] = Future { synchronized { sb.toString } } } class ExampleSpec extends flatspec.FixtureAsyncFlatSpec { type FixtureParam = StringActor def withFixture(test: OneArgAsyncTest): FutureOutcome = { val actor = new StringActor complete { actor ! Append("ScalaTest is ") // set up the fixture withFixture(test.toNoArgAsyncTest(actor)) } lastly { actor ! Clear // ensure the fixture will be cleaned up } } "Testing" should "be easy" in { actor => actor ! Append("easy!") val futureString = actor ? GetValue futureString map { s => assert(s == "ScalaTest is easy!") } } it should "be fun" in { actor => actor ! Append("fun!") val futureString = actor ? GetValue futureString map { s => assert(s == "ScalaTest is fun!") } } }
If a test fails, the future returned by the OneArgAsyncTest
function will result in
an org.scalatest.Failed wrapping the exception describing
the failure. To ensure clean up happens even if a test fails, you should invoke the test function and do the cleanup using
complete
-lastly
, as shown in the previous example. The complete
-lastly
syntax, defined in CompleteLastly
, which is extended by AsyncTestSuite
, ensures
the second, cleanup block of code is executed, whether the the first block throws an exception or returns a future. If it returns a
future, the cleanup will be executed when the future completes.
If multiple test classes need the same fixture, you can define the FixtureParam
and withFixture(OneArgAsyncTest)
implementations in a trait, then mix that trait into the test classes that need it. For example, if your application requires a database and your
integration tests use that database, you will likely have many test classes that need a database fixture. You can create a "database fixture" trait
that creates a database with a unique name, passes the connector into the test, then removes the database once the test completes. This is shown in
the following example:
package org.scalatest.examples.fixture.asyncflatspec.sharing import java.util.concurrent.ConcurrentHashMap import org.scalatest._ import DbServer._ import java.util.UUID.randomUUID import scala.concurrent.Future object DbServer { // Simulating a database server type Db = StringBuffer private val databases = new ConcurrentHashMap[String, Db] def createDb(name: String): Db = { val db = new StringBuffer databases.put(name, db) db } def removeDb(name: String) { databases.remove(name) } } trait DbFixture { this: fixture.AsyncTestSuite => type FixtureParam = Db // Allow clients to populate the database after // it is created def populateDb(db: Db) {} def withFixture(test: OneArgAsyncTest): FutureOutcome = { val dbName = randomUUID.toString val db = createDb(dbName) // create the fixture complete { populateDb(db) // setup the fixture withFixture(test.toNoArgAsyncTest(db)) // "loan" the fixture to the test } lastly { removeDb(dbName) // ensure the fixture will be cleaned up } } } class ExampleSpec extends flatspec.FixtureAsyncFlatSpec with DbFixture { override def populateDb(db: Db) { // setup the fixture db.append("ScalaTest is ") } "Testing" should "should be easy" in { db => Future { db.append("easy!") assert(db.toString === "ScalaTest is easy!") } } it should "be fun" in { db => Future { db.append("fun!") assert(db.toString === "ScalaTest is fun!") } } // This test doesn't need a Db "Test code" should "be clear" in { () => Future { val buf = new StringBuffer buf.append("ScalaTest code is ") buf.append("clear!") assert(buf.toString === "ScalaTest code is clear!") } } }
Often when you create fixtures in a trait like DbFixture
, you'll still need to enable individual test classes
to "setup" a newly created fixture before it gets passed into the tests. A good way to accomplish this is to pass the newly
created fixture into a setup method, like populateDb
in the previous example, before passing it to the test
function. Classes that need to perform such setup can override the method, as does ExampleSuite
.
If a test doesn't need the fixture, you can indicate that by providing a no-arg instead of a one-arg function, as is done in the
third test in the previous example, “test code should be clear
”. In other words, instead of starting your function literal
with something like “db =>
”, you'd start it with “() =>
”. For such tests, runTest
will not invoke withFixture(OneArgAsyncTest)
. It will instead directly invoke withFixture(NoArgAsyncTest)
.
Both examples shown above demonstrate the technique of giving each test its own "fixture sandbox" to play in. When your fixtures
involve external side-effects, like creating files or databases, it is a good idea to give each file or database a unique name as is
done in these examples. This keeps tests completely isolated, allowing you to run them in parallel if desired. You could mix
ParallelTestExecution
into either of these ExampleSuite
classes, and the tests would run in parallel just fine.
Implementation trait for class FixtureAsyncFlatSpec
, which is
a sister class to org.scalatest.flatspec.FixtureAsyncFlatSpec
that can pass a
fixture object into its tests.
Implementation trait for class FixtureAsyncFlatSpec
, which is
a sister class to org.scalatest.flatspec.FixtureAsyncFlatSpec
that can pass a
fixture object into its tests.
FixtureAsyncFlatSpec
is a class,
not a trait, to minimize compile time given there is a slight compiler
overhead to mixing in traits compared to extending classes. If you need
to mix the behavior of FixtureAsyncFlatSpec
into some other
class, you can use this trait instead, because class
FixtureAsyncFlatSpec
does nothing more than extend this trait and add a nice toString
implementation.
See the documentation of the class for a detailed
overview of FixtureAsyncFlatSpec
.
Facilitates a “behavior-driven” style of development (BDD), in which tests are combined with text that specifies the behavior the tests verify.
AnyFlatSpec
is a good first step for teams wishing to move from xUnit to BDD, because its structure is flat like xUnit, so simple and familiar, but the test names must be written in a specification style: “X should Y,” “A must B,” etc.Trait
AnyFlatSpec
is so named because your specification text and tests line up flat against the left-side indentation level, with no nesting needed. Here's an exampleAnyFlatSpec
:Note: you can use
must
orcan
as well asshould
in aAnyFlatSpec
. For example, instead ofit should "have
..., you could writeit must "have
... orit can "have
....Instead of using a
behavior of
clause, you can alternatively use a shorthand syntax in which you replace the firstit
with the subject string, like this:Running either of the two previous versions of
SetSpec
in the Scala interpreter would yield:An empty Set - should have size 0 - should produce NoSuchElementException when head is invoked
In a
AnyFlatSpec
you write a one (or more) sentence specification for each bit of behavior you wish to specify and test. Each specification sentence has a "subject," which is sometimes called the system under test (or SUT). The subject is the entity being specified and tested and also serves as the subject of the sentences you write for each test. Often you will want to write multiple tests for the same subject. In aAnyFlatSpec
, you name the subject once, with abehavior of
clause or its shorthand, then write tests for that subject withit should
/must
/can "do something"
phrases. Eachit
refers to the most recently declared subject. For example, the four tests shown in this snippet are all testing a stack that contains one item:The same is true if the tests are written using the shorthand notation:
In a
AnyFlatSpec
, therefore, to figure out what "it
" means, you just scan vertically until you find the most recent use ofbehavior of
or the shorthand notation.Because sometimes the subject could be plural, you can alternatively use
they
instead ofit
:A
AnyFlatSpec
's lifecycle has two phases: the registration phase and the ready phase. It starts in registration phase and enters ready phase the first timerun
is called on it. It then remains in ready phase for the remainder of its lifetime.Tests can only be registered while the
AnyFlatSpec
is in its registration phase. Any attempt to register a test after theAnyFlatSpec
has entered its ready phase, i.e., afterrun
has been invoked on theAnyFlatSpec
, will be met with a thrownTestRegistrationClosedException
. The recommended style of usingAnyFlatSpec
is to register tests during object construction as is done in all the examples shown here. If you keep to the recommended style, you should never see aTestRegistrationClosedException
.Ignored tests
To support the common use case of temporarily disabling a test, with the good intention of resurrecting the test at a later time,
AnyFlatSpec
provides a methodignore
that can be used instead ofit
orthey
to register a test. For example, to temporarily disable the test with the name"An empty Set should produce NoSuchElementException when head is invoked"
, just change “it
” into “ignore
,” like this:If you run this version of
SetSpec
with:It will run only the first test and report that the second test was ignored:
When using shorthand notation, you won't have an
it
to change intoignore
for the first test of each new subject. To ignore such tests, you must instead changein
toignore
. For example, to temporarily disable the test with the name"An empty Set should have size 0"
, change “in
” into “ignore
” like this:If you run this version of
StackSpec
with:It will run only the second test and report that the first test was ignored:
If you wish to temporarily ignore an entire suite of tests, you can (on the JVM, not Scala.js) annotate the test class with
@Ignore
, like this:When you mark a test class with a tag annotation, ScalaTest will mark each test defined in that class with that tag. Thus, marking the
SetSpec
in the above example with the@Ignore
tag annotation means that both tests in the class will be ignored. If you run the aboveSetSpec
in the Scala interpreter, you'll see:Note that marking a test class as ignored won't prevent it from being discovered by ScalaTest. Ignored classes will be discovered and run, and all their tests will be reported as ignored. This is intended to keep the ignored class visible, to encourage the developers to eventually fix and “un-ignore” it. If you want to prevent a class from being discovered at all (on the JVM, not Scala.js), use the
DoNotDiscover
annotation instead.Informers
One of the parameters to
AnyFlatSpec
'srun
method is aReporter
, which will collect and report information about the running suite of tests. Information about suites and tests that were run, whether tests succeeded or failed, and tests that were ignored will be passed to theReporter
as the suite runs. Most often the reporting done by default byAnyFlatSpec
's methods will be sufficient, but occasionally you may wish to provide custom information to theReporter
from a test. For this purpose, anInformer
that will forward information to the currentReporter
is provided via theinfo
parameterless method. You can pass the extra information to theInformer
via itsapply
method. TheInformer
will then pass the information to theReporter
via anInfoProvided
event.One use case for the
Informer
is to pass more information about a specification to the reporter. For example, theGivenWhenThen
trait provides methods that use the implicitinfo
provided byFlatSpec
to pass such information to the reporter. Here's an example:If you run this
AnyFlatSpec
from the interpreter, you will see the following output:scala> org.scalatest.run(new SetSpec) SetSpec: A mutable Set - should allow an element to be added + Given an empty mutable Set + When an element is added + Then the Set should have size 1 + And the Set should contain the added element + That's all folks!
Documenters
AnyFlatSpec
also provides amarkup
method that returns aDocumenter
, which allows you to send to theReporter
text formatted in Markdown syntax. You can pass the extra information to theDocumenter
via itsapply
method. TheDocumenter
will then pass the information to theReporter
via anMarkupProvided
event.Here's an example
AnyFlatSpec
that usesmarkup
:Although all of ScalaTest's built-in reporters will display the markup text in some form, the HTML reporter will format the markup information into HTML. Thus, the main purpose of
markup
is to add nicely formatted text to HTML reports. Here's what the aboveSetSpec
would look like in the HTML reporter:Notifiers and alerters
ScalaTest records text passed to
info
andmarkup
during tests, and sends the recorded text in therecordedEvents
field of test completion events likeTestSucceeded
andTestFailed
. This allows string reporters (like the standard out reporter) to showinfo
andmarkup
text after the test name in a color determined by the outcome of the test. For example, if the test fails, string reporters will show theinfo
andmarkup
text in red. If a test succeeds, string reporters will show theinfo
andmarkup
text in green. While this approach helps the readability of reports, it means that you can't useinfo
to get status updates from long running tests.To get immediate (i.e., non-recorded) notifications from tests, you can use
note
(aNotifier
) andalert
(anAlerter
). Here's an example showing the differences:Because
note
andalert
information is sent immediately, it will appear before the test name in string reporters, and its color will be unrelated to the ultimate outcome of the test:note
text will always appear in green,alert
text will always appear in yellow. Here's an example:Another example is slowpoke notifications. If you find a test is taking a long time to complete, but you're not sure which test, you can enable slowpoke notifications. ScalaTest will use an
Alerter
to fire an event whenever a test has been running longer than a specified amount of time.In summary, use
info
andmarkup
for text that should form part of the specification output. Usenote
andalert
to send status notifications. (Because the HTML reporter is intended to produce a readable, printable specification,info
andmarkup
text will appear in the HTML report, butnote
andalert
text will not.)Pending tests
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, 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 withTestPendingException
.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 withTestPendingException
, 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. You can mark tests as pending inFlatSpec
like this:If you run this version of
AnyFlatSpec
with:It will run both tests but report that
An empty Set should have size 0
is pending. You'll see:One difference between an ignored test and a pending one is that an ignored test is intended to be used during a significant refactorings of the code under test, when tests break and you don't want to spend the time to fix all of them immediately. You can mark some of those broken tests as ignored temporarily, so that you can focus the red bar on just failing tests you actually want to fix immediately. Later you can go back and fix the ignored tests. In other words, by ignoring some failing tests temporarily, you can more easily notice failed tests that you actually want to fix. By contrast, a pending test is intended to be used before a test and/or the code under test is written. Pending indicates you've decided to write a test for a bit of behavior, but either you haven't written the test yet, or have only written part of it, or perhaps you've written the test but don't want to implement the behavior it tests until after you've implemented a different bit of behavior you realized you need first. Thus ignored tests are designed to facilitate refactoring of existing code whereas pending tests are designed to facilitate the creation of new code.
One other difference between ignored and pending tests is that ignored tests are implemented as a test tag that is excluded by default. Thus an ignored test is never executed. By contrast, a pending test is implemented as a test that throws
TestPendingException
(which is what calling thepending
method does). Thus the body of pending tests are executed up until they throwTestPendingException
. The reason for this difference is that it enables your unfinished test to sendInfoProvided
messages to the reporter before it completes abruptly withTestPendingException
, as shown in the previous example onInformer
s that used theGivenWhenThen
trait. For example, the following snippet in aAnyFlatSpec
:Would yield the following output when run in the interpreter:
Tagging tests
A
AnyFlatSpec
's tests may be classified into groups by tagging them with string names. As with any suite, when executing aAnyFlatSpec
, groups of tests can optionally be included and/or excluded. To tag aAnyFlatSpec
's tests, you pass objects that extend classorg.scalatest.Tag
to methods that register tests. ClassTag
takes one parameter, a string name. If you have created tag annotation interfaces as described in theTag
documentation, then you will probably want to use tag names on your test functions that match. To do so, simply pass the fully qualified names of the tag interfaces to theTag
constructor. For example, if you've defined a tag annotation interface with fully qualified name,com.mycompany.tags.DbTest
, then you could create a matching tag forAnyFlatSpec
s like this:Given these definitions, you could place
AnyFlatSpec
tests into groups with tags like this:This code marks both tests with the
org.scalatest.tags.Slow
tag, and the second test with thecom.mycompany.tags.DbTest
tag.The
run
method takes aFilter
, whose constructor takes an optionalSet[String]
calledtagsToInclude
and aSet[String]
calledtagsToExclude
. IftagsToInclude
isNone
, all tests will be run except those those belonging to tags listed in thetagsToExclude
Set
. IftagsToInclude
is defined, only tests belonging to tags mentioned in thetagsToInclude
set, and not mentioned intagsToExclude
, will be run.It is recommended, though not required, that you create a corresponding tag annotation when you create a
Tag
object. A tag annotation (on the JVM, not Scala.js) allows you to tag all the tests of aAnyFlatSpec
in one stroke by annotating the class. For more information and examples, see the documentation for classTag
. On Scala.js, to tag all tests of a suite, you'll need to tag each test individually at the test site.Shared fixtures
A test fixture is composed of the objects and other artifacts (files, sockets, database connections, etc.) tests use to do their work. When multiple tests need to work with the same fixtures, it is important to try and avoid duplicating the fixture code across those tests. The more code duplication you have in your tests, the greater drag the tests will have on refactoring the actual production code.
ScalaTest recommends three techniques to eliminate such code duplication:
withFixture
Each technique is geared towards helping you reduce code duplication without introducing instance
var
s, shared mutable objects, or other dependencies between tests. Eliminating shared mutable state across tests will make your test code easier to reason about and more amenable for parallel test execution.The following sections describe these techniques, including explaining the recommended usage for each. But first, here's a table summarizing the options:
withFixture
when most or all tests need the same fixture.withFixture(NoArgTest)
withFixture(OneArgTest)
instead)withFixture(OneArgTest)
BeforeAndAfter
BeforeAndAfterEach
Calling get-fixture methods
If you need to create the same mutable fixture objects in multiple tests, and don't need to clean them up after using them, the simplest approach is to write one or more get-fixture methods. A get-fixture method returns a new instance of a needed fixture object (or a holder object containing multiple fixture objects) each time it is called. You can call a get-fixture method at the beginning of each test that needs the fixture, storing the returned object or objects in local variables. Here's an example:
The “
f.
” in front of each use of a fixture object provides a visual indication of which objects are part of the fixture, but if you prefer, you can import the the members with “import f._
” and use the names directly.If you need to configure fixture objects differently in different tests, you can pass configuration into the get-fixture method. For example, if you could pass in an initial value for a mutable fixture object as a parameter to the get-fixture method.
Instantiating fixture-context objects
An alternate technique that is especially useful when different tests need different combinations of fixture objects is to define the fixture objects as instance variables of fixture-context objects whose instantiation forms the body of tests. Like get-fixture methods, fixture-context objects are only appropriate if you don't need to clean up the fixtures after using them.
To use this technique, you define instance variables intialized with fixture objects in traits and/or classes, then in each test instantiate an object that contains just the fixture objects needed by the test. Traits allow you to mix together just the fixture objects needed by each test, whereas classes allow you to pass data in via a constructor to configure the fixture objects. Here's an example in which fixture objects are partitioned into two traits and each test just mixes together the traits it needs:
Overriding
withFixture(NoArgTest)
Although the get-fixture method and fixture-context object approaches take care of setting up a fixture at the beginning of each test, they don't address the problem of cleaning up a fixture at the end of the test. If you just need to perform a side-effect at the beginning or end of a test, and don't need to actually pass any fixture objects into the test, you can override
withFixture(NoArgTest)
, one of ScalaTest's lifecycle methods defined in traitSuite
.Trait
Suite
's implementation ofrunTest
passes a no-arg test function towithFixture(NoArgTest)
. It iswithFixture
's responsibility to invoke that test function.Suite
's implementation ofwithFixture
simply invokes the function, like this:You can, therefore, override
withFixture
to perform setup before and/or cleanup after invoking the test function. If you have cleanup to perform, you should invoke the test function inside atry
block and perform the cleanup in afinally
clause, in case an exception propagates back throughwithFixture
. (If a test fails because of an exception, the test function invoked by withFixture will result in aFailed
wrapping the exception. Nevertheless, best practice is to perform cleanup in a finally clause just in case an exception occurs.)The
withFixture
method is designed to be stacked, and to enable this, you should always call thesuper
implementation ofwithFixture
, and let it invoke the test function rather than invoking the test function directly. That is to say, instead of writing “test()
”, you should write “super.withFixture(test)
”, like this:Here's an example in which
withFixture(NoArgTest)
is used to take a snapshot of the working directory if a test fails, and send that information to the reporter:Running this version of
ExampleSuite
in the interpreter in a directory with two files,hello.txt
andworld.txt
would give the following output:Note that the
NoArgTest
passed towithFixture
, in addition to anapply
method that executes the test, also includesTestData
such as the test name and the config map passed torunTest
. Thus you can also use the test name and configuration objects in yourwithFixture
implementation.Calling loan-fixture methods
If you need to both pass a fixture object into a test and perform cleanup at the end of the test, you'll need to use the loan pattern. If different tests need different fixtures that require cleanup, you can implement the loan pattern directly by writing loan-fixture methods. A loan-fixture method takes a function whose body forms part or all of a test's code. It creates a fixture, passes it to the test code by invoking the function, then cleans up the fixture after the function returns.
The following example shows three tests that use two fixtures, a database and a file. Both require cleanup after, so each is provided via a loan-fixture method. (In this example, the database is simulated with a
StringBuffer
.)As demonstrated by the last test, loan-fixture methods compose. Not only do loan-fixture methods allow you to give each test the fixture it needs, they allow you to give a test multiple fixtures and clean everything up afterwards.
Also demonstrated in this example is the technique of giving each test its own "fixture sandbox" to play in. When your fixtures involve external side-effects, like creating files or databases, it is a good idea to give each file or database a unique name as is done in this example. This keeps tests completely isolated, allowing you to run them in parallel if desired.
Overriding
withFixture(OneArgTest)
If all or most tests need the same fixture, you can avoid some of the boilerplate of the loan-fixture method approach by using a
FixtureAnyFlatSpec
and overridingwithFixture(OneArgTest)
. Each test in aFixtureAnyFlatSpec
takes a fixture as a parameter, allowing you to pass the fixture into the test. You must indicate the type of the fixture parameter by specifyingFixtureParam
, and implement awithFixture
method that takes aOneArgTest
. ThiswithFixture
method is responsible for invoking the one-arg test function, so you can perform fixture set up before, and clean up after, invoking and passing the fixture into the test function.To enable the stacking of traits that define
withFixture(NoArgTest)
, it is a good idea to letwithFixture(NoArgTest)
invoke the test function instead of invoking the test function directly. To do so, you'll need to convert theOneArgTest
to aNoArgTest
. You can do that by passing the fixture object to thetoNoArgTest
method ofOneArgTest
. In other words, instead of writing “test(theFixture)
”, you'd delegate responsibility for invoking the test function to thewithFixture(NoArgTest)
method of the same instance by writing:Here's a complete example:
In this example, the tests actually required two fixture objects, a
File
and aFileWriter
. In such situations you can simply define theFixtureParam
type to be a tuple containing the objects, or as is done in this example, a case class containing the objects. For more information on thewithFixture(OneArgTest)
technique, see the documentation forFixtureAnyFlatSpec
.Mixing in
BeforeAndAfter
In all the shared fixture examples shown so far, the activities of creating, setting up, and cleaning up the fixture objects have been performed during the test. This means that if an exception occurs during any of these activities, it will be reported as a test failure. Sometimes, however, you may want setup to happen before the test starts, and cleanup after the test has completed, so that if an exception occurs during setup or cleanup, the entire suite aborts and no more tests are attempted. The simplest way to accomplish this in ScalaTest is to mix in trait
BeforeAndAfter
. With this trait you can denote a bit of code to run before each test withbefore
and/or after each test each test withafter
, like this:Note that the only way
before
andafter
code can communicate with test code is via some side-effecting mechanism, commonly by reassigning instancevar
s or by changing the state of mutable objects held from instanceval
s (as in this example). If using instancevar
s or mutable objects held from instanceval
s you wouldn't be able to run tests in parallel in the same instance of the test class (on the JVM, not Scala.js) unless you synchronized access to the shared, mutable state. This is why ScalaTest'sParallelTestExecution
trait extendsOneInstancePerTest
. By running each test in its own instance of the class, each test has its own copy of the instance variables, so you don't need to synchronize. If you mixedParallelTestExecution
into theExampleSuite
above, the tests would run in parallel just fine without any synchronization needed on the mutableStringBuilder
andListBuffer[String]
objects.Although
BeforeAndAfter
provides a minimal-boilerplate way to execute code before and after tests, it isn't designed to enable stackable traits, because the order of execution would be non-obvious. If you want to factor out before and after code that is common to multiple test suites, you should use traitBeforeAndAfterEach
instead, as shown later in the next section, composing fixtures by stacking traits.Composing fixtures by stacking traits
In larger projects, teams often end up with several different fixtures that test classes need in different combinations, and possibly initialized (and cleaned up) in different orders. A good way to accomplish this in ScalaTest is to factor the individual fixtures into traits that can be composed using the stackable trait pattern. This can be done, for example, by placing
withFixture
methods in several traits, each of which callsuper.withFixture
. Here's an example in which theStringBuilder
andListBuffer[String]
fixtures used in the previous examples have been factored out into two stackable fixture traits namedBuilder
andBuffer
:By mixing in both the
Builder
andBuffer
traits,ExampleSuite
gets both fixtures, which will be initialized before each test and cleaned up after. The order the traits are mixed together determines the order of execution. In this case,Builder
is “super” toBuffer
. If you wantedBuffer
to be “super” toBuilder
, you need only switch the order you mix them together, like this:And if you only need one fixture you mix in only that trait:
Another way to create stackable fixture traits is by extending the
BeforeAndAfterEach
and/orBeforeAndAfterAll
traits.BeforeAndAfterEach
has abeforeEach
method that will be run before each test (like JUnit'ssetUp
), and anafterEach
method that will be run after (like JUnit'stearDown
). Similarly,BeforeAndAfterAll
has abeforeAll
method that will be run before all tests, and anafterAll
method that will be run after all tests. Here's what the previously shown example would look like if it were rewritten to use theBeforeAndAfterEach
methods instead ofwithFixture
:To get the same ordering as
withFixture
, place yoursuper.beforeEach
call at the end of eachbeforeEach
method, and thesuper.afterEach
call at the beginning of eachafterEach
method, as shown in the previous example. It is a good idea to invokesuper.afterEach
in atry
block and perform cleanup in afinally
clause, as shown in the previous example, because this ensures the cleanup code is performed even ifsuper.afterEach
throws an exception.The difference between stacking traits that extend
BeforeAndAfterEach
versus traits that implementwithFixture
is that setup and cleanup code happens before and after the test inBeforeAndAfterEach
, but at the beginning and end of the test inwithFixture
. Thus if awithFixture
method completes abruptly with an exception, it is considered a failed test. By contrast, if any of thebeforeEach
orafterEach
methods ofBeforeAndAfterEach
complete abruptly, it is considered an aborted suite, which will result in aSuiteAborted
event.Shared tests
Sometimes you may want to run the same test code on different fixture objects. In other words, you may want to write tests that are "shared" by different fixture objects. To accomplish this in a
AnyFlatSpec
, you first place shared tests in behavior functions. These behavior functions will be invoked during the construction phase of anyAnyFlatSpec
that uses them, so that the tests they contain will be registered as tests in thatAnyFlatSpec
. For example, given this stack class:You may want to test the
Stack
class in different states: empty, full, with one item, with one item less than capacity, etc. You may find you have several tests that make sense any time the stack is non-empty. Thus you'd ideally want to run those same tests for three stack fixture objects: a full stack, a stack with a one item, and a stack with one item less than capacity. With shared tests, you can factor these tests out into a behavior function, into which you pass the stack fixture to use when running the tests. So in yourAnyFlatSpec
for stack, you'd invoke the behavior function three times, passing in each of the three stack fixtures so that the shared tests are run for all three fixtures. You can define a behavior function that encapsulates these shared tests inside theAnyFlatSpec
that uses them. If they are shared between differentAnyFlatSpec
s, however, you could also define them in a separate trait that is mixed into eachAnyFlatSpec
that uses them.For example, here the
nonEmptyStack
behavior function (in this case, a behavior method) is defined in a trait along with another method containing shared tests for non-full stacks:Given these behavior functions, you could invoke them directly, but
AnyFlatSpec
offers a DSL for the purpose, which looks like this:If you prefer to use an imperative style to change fixtures, for example by mixing in
BeforeAndAfterEach
and reassigning astack
var
inbeforeEach
, you could write your behavior functions in the context of thatvar
, which means you wouldn't need to pass in the stack fixture because it would be in scope already inside the behavior function. In that case, your code would look like this:The recommended style, however, is the functional, pass-all-the-needed-values-in style. Here's an example:
If you load these classes into the Scala interpreter (with scalatest's JAR file on the class path), and execute it, you'll see:
scala> org.scalatest.run(new SharedTestExampleSpec) A Stack (when empty) - should be empty - should complain on peek - should complain on pop A Stack (with one item) - should be non-empty - should return the top item on peek - should not remove the top item on peek - should remove the top item on pop - should not be full - should add to the top on push A Stack (with one item less than capacity) - should be non-empty - should return the top item on peek - should not remove the top item on peek - should remove the top item on pop - should not be full - should add to the top on push A Stack (full) - should be full - should be non-empty - should return the top item on peek - should not remove the top item on peek - should remove the top item on pop - should complain on a push
One thing to keep in mind when using shared tests is that in ScalaTest, each test in a suite must have a unique name. If you register the same tests repeatedly in the same suite, one problem you may encounter is an exception at runtime complaining that multiple tests are being registered with the same test name. A good way to solve this problem in a
AnyFlatSpec
is to make sure each invocation of a behavior function is in the context of a different set ofwhen
, verb (should
,must
, or can), andthat
clauses, which will prepend a string to each test name. For example, the following code in aAnyFlatSpec
would register a test with the name"A Stack (when empty) should be empty"
:Or, using the shorthand notation:
If the
"should be empty"
test was factored out into a behavior function, it could be called repeatedly so long as each invocation of the behavior function is in the context of a different combination ofwhen
, verb, andthat
clauses.