Spock Framework Reference Documentation

Spock has built-in support for implementing each of the conceptual phases of a feature method. To this end, feature methods are structured into so-called blocks. Blocks start with a label, and extend to the beginning of the next block, or the end of the method. There are six kinds of blocks: given, when, then, expect, cleanup, and where blocks. Any statements between the beginning of the method and the first explicit block belong to an implicit given block.

A feature method must have at least one explicit (i.e. labelled) block - in fact, the presence of an explicit block is what makes a method a feature method. Blocks divide a method into distinct sections, and cannot be nested.

Let’s take a closer look at the then: block. It contains two interactions, each of which has four distinct parts: a cardinality, a target constraint, a method constraint, and an argument constraint:

The cardinality of an interaction describes how often a method call is expected. It can either be a fixed number or a range:

The target constraint of an interaction describes which mock object is expected to receive the method call:

The method constraint of an interaction describes which method is expected to be called:

When expecting a call to a getter method, Groovy property syntax can be used instead of method syntax:

When expecting a call to a setter method, only method syntax can be used:

The argument constraints of an interaction describe which method arguments are expected:

Argument constraints work as expected for methods with multiple arguments:

When dealing with vararg methods, vararg syntax can also be used in the corresponding interactions:

Sometimes it can be useful to match "anything", in some sense of the word:

Now, when would matching any method call be useful? A good example is strict mocking, a style of mocking where no interactions other than those explicitly declared are allowed:

0 * only makes sense as the last interaction of a then: block or method. Note the use of _ * (any number of calls), which allows any interaction with the auditing component.

So far, we declared all our interactions in a then: block. This often results in a spec that reads naturally. However, it is also permissible to put interactions anywhere before the when: block that is supposed to satisfy them. In particular, this means that interactions can be declared in a setup method. Interactions can also be declared in any "helper" instance method of the same specification class.

When an invocation on a mock object occurs, it is matched against interactions in the interactions' declared order. If an invocation matches multiple interactions, the earliest declared interaction that hasn’t reached its upper invocation limit will win. There is one exception to this rule: Interactions declared in a then: block are matched against before any other interactions. This allows to override interactions declared in, say, a setup method with interactions declared in a then: block.

If a mock has a set of "base" interactions that don’t vary, they can be declared right at mock creation time:

This feature is particularly attractive for Stubbing and with dedicated Stubs. Note that the interactions don’t (and cannot ^[5]) have a target constraint; it’s clear from the context which mock object they belong to.

Interactions can also be declared when initializing an instance field with a mock:

Interactions sharing the same target can be grouped in a Specification.with block. Similar to Declaring Interactions at Mock Creation Time, this makes it unnecessary to repeat the target constraint:

A with block can also be used for grouping conditions with the same target.

A then: block may contain both interactions and conditions. Although not strictly required, it is customary to declare interactions before conditions:

Read out aloud: "When the publisher sends a 'hello' message, then the subscriber should receive the message exactly once, and the publisher’s message count should be one."

Internally, Spock must have full information about expected interactions before they take place. So how is it possible for interactions to be declared in a then: block? The answer is that under the hood, Spock moves interactions declared in a then: block to immediately before the preceding when: block. In most cases this works out just fine, but sometimes it can lead to problems:

Here we have introduced a variable for the expected argument. (Likewise, we could have introduced a variable for the cardinality.) However, Spock isn’t smart enough (huh?) to tell that the interaction is intrinsically linked to the variable declaration. Hence it will just move the interaction, which will cause a MissingPropertyException at runtime.

One way to solve this problem is to move (at least) the variable declaration to before the when: block. (Fans of Data Driven Testing might move the variable into a where: block.) In our example, this would have the added benefit that we could use the same variable for sending the message.

Another solution is to be explicit about the fact that variable declaration and interaction belong together:

Since an MockingApi.interaction block is always moved in its entirety, the code now works as intended.

Interactions declared in a then: block are scoped to the preceding when: block:

This makes sure that subscriber receives "message1" during execution of the first when: block, and "message2" during execution of the second when: block.

Interactions declared outside a then: block are active from their declaration until the end of the containing feature method.

Interactions are always scoped to a particular feature method. Hence they cannot be declared in a static method, setupSpec method, or cleanupSpec method. Likewise, mock objects should not be stored in static or @Shared fields.

There are two main ways in which a mock-based test can fail: An interaction can match more invocations than allowed, or it can match fewer invocations than required. The former case is detected right when the invocation happens, and causes a TooManyInvocationsError:

To make it easier to diagnose why too many invocations matched, Spock will show all invocations matching the interaction in question:

According to this output, one of the receive("hello") calls triggered the TooManyInvocationsError. Note that because indistinguishable calls like the two invocations of subscriber.receive("hello") are aggregated into a single line of output, the first receive("hello") may well have occurred before the receive("goodbye").

The second case (fewer invocations than required) can only be detected once execution of the when block has completed. (Until then, further invocations may still occur.) It causes a TooFewInvocationsError:

Note that it doesn’t matter whether the method was not called at all, the same method was called with different arguments, the same method was called on a different mock object, or a different method was called "instead" of this one; in either case, a TooFewInvocationsError error will occur.

To make it easier to diagnose what happened "instead" of a missing invocation, Spock will show all invocations that didn’t match any interaction, ordered by their similarity with the interaction in question. In particular, invocations that match everything but the interaction’s arguments will be shown first:

Often, the exact method invocation order isn’t relevant and may change over time. To avoid over-specification, Spock defaults to allowing any invocation order, provided that the specified interactions are eventually satisfied:

Here, any of the invocation sequences "hello" "hello" "goodbye", "hello" "goodbye" "hello", and "goodbye" "hello" "hello" will satisfy the specified interactions.

In those cases where invocation order matters, you can impose an order by splitting up interactions into multiple then: blocks:

Now Spock will verify that both "hello"'s are received before the "goodbye". In other words, invocation order is enforced between but not within then: blocks.

Besides interfaces, Spock also supports mocking of classes. Mocking classes works just like mocking interfaces; the only additional requirement is to put cglib-nodep-2.2 or higher and objenesis-1.2 or higher on the class path. If either of these libraries is missing from the class path, Spock will gently let you know.

We have already seen the use of the right-shift (>>) operator to return a fixed value:

To return different values for different invocations, use multiple interactions:

This will return "ok" whenever "message1" is received, and "fail" whenever "message2" is received. There is no limit as to which values can be returned, provided they are compatible with the method’s declared return type.

To return different values on successive invocations, use the triple-right-shift (>>>) operator:

This will return "ok" for the first invocation, "error" for the second and third invocation, and "ok" for all remaining invocations. The right-hand side must be a value that Groovy knows how to iterate over; in this example, we’ve used a plain list.

To compute a return value based on the method’s argument, use the the right-shift (>>) operator together with a closure. If the closure declares a single untyped parameter, it gets passed the method’s argument list:

Here "ok" gets returned if the message is more than three characters long, and "fail" otherwise.

In most cases it would be more convenient to have direct access to the method’s arguments. If the closure declares more than one parameter or a single typed parameter, method arguments will be mapped one-by-one to closure parameters:^[6]

This response generator behaves the same as the previous one, but is arguably more readable.

If you find yourself in need of more information about a method invocation than its arguments, have a look at org.spockframework.mock.IMockInvocation. All methods declared in this interface are available inside the closure, without a need to prefix them. (In Groovy terminology, the closure delegates to an instance of IMockInvocation.)

Sometimes you may want to do more than just computing a return value. A typical example is throwing an exception. Again, closures come to the rescue:

Of course, the closure can contain more code, for example a println statement. It will get executed every time an incoming invocation matches the interaction.

Method responses can be chained:

This will return "ok", "fail", "ok" for the first three invocations, throw InternalError for the fourth invocations, and return ok for any further invocation.

A stub is created with the MockingApi.Stub factory method:

Whereas a mock can be used both for stubbing and mocking, a stub can only be used for stubbing. Limiting a collaborator to a stub communicates its role to the readers of the specification.

Like a mock, a stub allows unexpected invocations. However, the values returned by a stub in such cases are more ambitious:

A stub often has a fixed set of interactions, which makes declaring interactions at mock creation time particularly attractive:

(Think twice before using this feature. It might be better to change the design of the code under specification.)

A spy is created with the MockingApi.Spy factory method:

A spy is always based on a real object. Hence you must provide a class type rather than an interface type, along with any constructor arguments for the type. If no constructor arguments are provided, the type’s default constructor will be used.

You may also create a spy from an instantiated object. This may be useful in cases where you do not have full control over the instatiation of types you are interested in spying. (For example when testing within a Dependency Injection framework such as Spring or Guice.)

Method calls on a spy are automatically delegated to the real object. Likewise, values returned from the real object’s methods are passed back to the caller via the spy.

After creating a spy, you can listen in on the conversation between the caller and the real object underlying the spy:

Apart from making sure that receive gets called exactly once, the conversation between the publisher and the SubscriberImpl instance underlying the spy remains unaltered.

When stubbing a method on a spy, the real method no longer gets called:

Instead of calling SubscriberImpl.receive, the receive method will now simply return "ok".

Sometimes, it is desirable to both execute some code and delegate to the real method:

Here we use callRealMethod() to delegate the method invocation to the real object. Note that we don’t have to pass the message argument along; this is taken care of automatically. callRealMethod() returns the real invocation’s result, but in this example we opted to return our own result instead. If we had wanted to pass a different message to the real method, we could have used callRealMethodWithArgs("changed message").

(Think twice before using this feature. It might be better to change the design of the code under specification.)

Spies can also be used as partial mocks:

All Groovy mocks implement the GroovyObject interface. They support the mocking and stubbing of dynamic methods as if they were physically declared methods:

(Think twice before using this feature. It might be better to change the design of the code under specification.)

Usually, Groovy mocks need to be injected into the code under specification just like regular mocks. However, when a Groovy mock is created as global, it automagically replaces all real instances of the mocked type for the duration of the feature method:^[7]

Here, we set up the publisher with two instances of a real subscriber implementation. Then we create a global mock of the same type. This reroutes all method calls on the real subscribers to the mock object. The mock object’s instance isn’t ever passed to the publisher; it is only used to describe the interaction.

Since global mocks have a somewhat, well, global effect, it’s often convenient to use them together with GroovySpy. This leads to the real code getting executed unless an interaction matches, allowing you to selectively listen in on objects and change their behavior just where needed.

(Think twice before using this feature. It might be better to change the design of the code under specification.)

Global mocks support mocking of constructors:

Since we are using a spy, the object returned from the constructor call remains unchanged. To change which object gets constructed, we can stub the constructor:

Now, whenever some code tries to construct a subscriber named Fred, we’ll construct a subscriber named Barney instead.

(Think twice before using this feature. It might be better to change the design of the code under specification.)

Global mocks support mocking and stubbing of static methods:

The same works for dynamic static methods.

When a global mock is used solely for mocking constructors and static methods, the mock’s instance isn’t really needed. In such a case one can just write:

At the end of the day, the Mock(), Stub(), and Spy() factory methods are just canned ways to create mock objects with a certain configuration. If you want more fine-grained control over a mock’s configuration, have a look at the org.spockframework.mock.IMockConfiguration interface. All properties of this interface ^[8] can be passed as named arguments to the Mock() method. For example:

Here, we create a mock whose default return values match those of a Mock(), but whose invocations aren’t verified (as for a Stub()). Instead of passing ZeroOrNullResponse, we could have supplied our own custom org.spockframework.mock.IDefaultResponse for responding to unexpected method invocations.

To find out whether a particular object is a Spock mock object, use a org.spockframework.mock.MockUtil:

An util can also be used to get more information about a mock object:

You can configure Spock whether it should filter stack traces or not by using the configuration file. The default value is true.

To temporarily prevent a feature method from getting executed, annotate it with spock.lang.Ignore:

For documentation purposes, a reason can be provided:

To ignore a whole specification, annotate its class:

In most execution environments, ignored feature methods and specs will be reported as "skipped".

Care should be taken when ignoring feature methods in a spec class annotated with spock.lang.Stepwise since later feature methods may depend on earlier feature methods having executed.

To ignore all but a (typically) small subset of methods, annotate the latter with spock.lang.IgnoreRest:

@IgnoreRest is especially handy in execution environments that don’t provide an (easy) way to run a subset of methods.

Care should be taken when ignoring feature methods in a spec class annotated with spock.lang.Stepwise since later feature methods may depend on earlier feature methods having executed.

To ignore a feature method under certain conditions, annotate it with spock.lang.IgnoreIf, followed by a predicate:

To make predicates easier to read and write, the following properties are available inside the closure:

Using the os property, the previous example can be rewritten as:

Care should be taken when ignoring feature methods in a spec class annotated with spock.lang.Stepwise since later feature methods may depend on earlier feature methods having executed.

To execute a feature method under certain conditions, annotate it with spock.lang.Requires, followed by a predicate:

Requires works exactly like IgnoreIf, except that the predicate is inverted. In general, it is preferable to state the conditions under which a method gets executed, rather than the conditions under which it gets ignored.

To indicate that the feature is not fully implemented yet and should not be reported as error, annotate it with spock.lang.PendingFeature.

The use case is to annotate tests that can not yet run but should already be committed. The main difference to Ignore is that the test are executed, but test failures are ignored. If the test passes without an error, then it will be reported as failure since the PendingFeature annotation should be removed. This way the tests will become part of the normal tests instead of being ignored forever.

Groovy has the groovy.transform.NotYetImplemented annotation which is similar but behaves a differently.

PendingFeature:

To execute features in the order that they are declared, annotate a spec class with spock.lang.Stepwise:

Stepwise only affects the class carrying the annotation; not sub or super classes. Features after the first failure are skipped.

Stepwise does not override the behaviour of annotations such as Ignore, IgnoreRest, and IgnoreIf, so care should be taken when ignoring feature methods in spec classes annotated with Stepwise.

To fail a feature method, fixture, or class that exceeds a given execution duration, use spock.lang.Timeout, followed by a duration, and optionally a time unit. The default time unit is seconds.

When applied to a feature method, the timeout is per execution of one iteration, excluding time spent in fixture methods:

Applying Timeout to a spec class has the same effect as applying it to each feature that is not already annotated with Timeout, excluding time spent in fixtures:

When applied to a fixture method, the timeout is per execution of the fixture method.

When a timeout is reported to the user, the stack trace shown reflects the execution stack of the test framework when the timeout was exceeded.

The @Retry extensions can be used for flaky integration tests, where remote systems can fail sometimes. By default it retries an iteration 3 times with 0 delay if either an Exception or AssertionError has been thrown, all this is configurable. In addition, an optional condition closure can be used to determine if a feature should be retried. It also provides special support for data driven features, offering to either retry all iterations or just the failing ones.

Retries can also be applied to spec classes which has the same effect as applying it to each feature method that isn’t already annotated with {@code Retry}.

A {@code @Retry} annotation that is declared on a spec class is applied to all features in all subclasses as well, unless a subclass declares its own annotation. If so, the retries defined in the subclass are applied to all feature methods declared in the subclass as well as inherited ones.

Given the following example, running FooIntegrationSpec will execute both inherited and foo with one retry. Running BarIntegrationSpec will execute inherited and bar with two retries.

Check RetryFeatureExtensionSpec for more examples.

To activate one or more Groovy categories within the scope of a feature method or spec, use spock.util.mop.Use:

This can be useful for stubbing of dynamic methods, which are usually provided by the runtime environment (e.g. Grails). It has no effect when applied to a helper method. However, when applied to a spec class, it will also affect its helper methods.

To confine meta class changes to the scope of a feature method or spec class, use spock.util.mop.ConfineMetaClassChanges:

When applied to a spec class, the meta classes are restored to the state that they were in before setupSpec was executed, after cleanupSpec is executed.

When applied to a feature method, the meta classes are restored to as they were after setup was executed, before cleanup is executed.

Saves system properties before the annotated feature method (including any setup and cleanup methods) gets run, and restores them afterwards.

Applying this annotation to a spec class has the same effect as applying it to all its feature methods.

Automatically attaches a detached mock to the current Specification. Use this if there is no direct framework support available. Spring and Guice dependency injection is automatically handled by the Spring Module and Guice Module respectively.

Automatically clean up a field or property at the end of its lifetime by using spock.lang.AutoCleanup.

By default, an object is cleaned up by invoking its parameterless close() method. If some other method should be called instead, override the annotation’s value attribute:

If multiple fields or properties are annotated with AutoCleanup, their objects are cleaned up sequentially, in reverse field/property declaration order, starting from the most derived class class and walking up the inheritance chain.

If a cleanup operation fails with an exception, the exception is reported by default, and cleanup proceeds with the next annotated object. To prevent cleanup exceptions from being reported, override the annotation’s quiet attribute:

To attach a natural-language name to a spec, use spock.lang.Title:

Similarly, to attach a natural-language description to a spec, use spock.lang.Narrative:

To link to one or more references to external information related to a specification or feature, use spock.lang.See:

To indicate that a feature or spec relates to one or more issues in an external tracking system, use spock.lang.Issue:

If you have a common prefix URL for all issues in a project, you can use the Spock Configuration File to set it up for all at once. If it is set, it is prepended to the value of the @Issue annotation when building the URL.

If the issueNamePrefix is set, it is prepended to the value of the @Issue annotation when building the name for the issue.

To indicate one or more subjects of a spec, use spock.lang.Subject:

Additionally, Subject can be applied to fields and local variables:

Subject currently has only informational purposes.

Spock understands @org.junit.Rule annotations on non-@Shared instance fields. The according rules are run at the iteration interception point in the Spock lifecycle. This means that the rules before-actions are done before the execution of setup methods and the after-actions are done after the execution of cleanup methods.

Spock understands @org.junit.ClassRule annotations on @Shared fields. The according rules are run at the specification interception point in the Spock lifecycle. This means that the rules before-actions are done before the execution of setupSpec methods and the after-actions are done after the execution of cleanupSpec methods.

Spock is capable of including and excluding specifications according to their classes, super-classes and interfaces and according to annotations that are applied to the specification. Spock is also capable of including and excluding individual features according to annotations that are applied to the feature method. The configuration for what to include or exclude is done according to the Spock Configuration File section.

Spock can remember which features last failed and how often successively and also how long a feature needed to be tested. For successive runs Spock will then first run features that failed at last run and first features that failed more often successively. Within the previously failed or non-failed features Spock will run the fastest tests first. This behaviour can be enabled according to the Spock Configuration File section. The default value is false.

Spock can create a report log of the executed tests in JSON format. This report contains also things like @Title, @Narrative, @See and @Issue values or block descriptors. This report can be enabled according to the Spock Configuration File section. The default is to not generate this report.

For the report to be generated, you have to enable it and set at least the logFileDir and logFileName. enabled can also be set via the system property spock.logEnabled, logFileDir can also be set via the system property spock.logFileDir and logFileName can also be set via the system property spock.logFileName.

If a logFileSuffix is set (or the system property spock.logFileSuffix), it is appended to the base filename, separated by a dash. If the suffix contains the string #timestamp, this is replaced by the current date and time in UTC automatically. If you instead want to have your local date and time, you can use the setting from the example below.

To create a global extension you need to create a class that implements the interface IGlobalExtension and put its fully-qualified class name in a file META-INF/services/org.spockframework.runtime.extension.IGlobalExtension in the class path. As soon as these two conditions are satisfied, the extension is automatically loaded and used when Spock is running. For convenience there is also the class AbstractGlobalExtension, which provides empty implementations for all methods in the interface, so that only the needed ones need to be overridden.

IGlobalExtension has the following three methods:

To create an annotation driven local extension you need to create a class that implements the interface IAnnotationDrivenExtension. As type argument to the interface you need to supply an annotation class that has @Retention set to RUNTIME, @Target set to one or more of FIELD, METHOD and TYPE - depending on where you want your annotation to be applicable - and @ExtensionAnnotation applied, with the IAnnotationDrivenExtension class as argument. Of course the annotation class can have some attributes with which the user can further configure the behaviour of the extension for each annotation application. For convenience there is also the class AbstractAnnotationDrivenExtension, which provides empty implementations for all methods in the interface, so that only the needed ones need to be overridden.

Your annotation can be applied to a specification, a feature method, a fixture method or a field. On all other places like helper methods or other places if the @Target is set accordingly, the annotation will be ignored and has no effect other than being visible in the source code.

IAnnotationDrivenExtension has the following five methods, where in each you can prepare a specification with your extension magic, like attaching interceptors to various interception points as described in the chapter Interceptors:

You can add own sections in the Spock Configuration File for your extension by creating POJOs or POGOs that are annotated with @ConfigurationObject and have a default constructor (either implicitly or explicitly). The argument to the annotation is the name of the top-level section that is added to the Spock configuration file syntax. The default values for the configuration options are defined in the class by initializing the fields at declaration time or in the constructor. In the Spock configuration file those values can then be edited by the user of your extension.

To use the values of the configuration object in your extension, just define an uninitialized instance field of that type. Spock will then automatically create exactly one instance of the configuration object per Spock run, apply the settings from the configuration file to it (before the start() methods of global extensions are called) and inject that instance into the extension class instances.

A configuration object cannot be used exclusively in an annotation driven local extension, but it has to be used in at least one global extension to properly get initialized and populated with the settings from the configuration file. But if the configuration object is used in a global extension, you can also use it just fine in an annotation driven local extension. If the configuration object is only used in an annotation driven local extension, you will get an exception when then configuration object is to be injected into the extension and you will also get an error when the configuration file is evaluated and it contains the section, as the configuration object is not properly registered yet.

For applying the magic of your extension, there are various interception points, where you can attach interceptors from the extension methods described above to hook into the Spock lifecycle. For each interception point there can of course be multiple interceptors added by arbitrary Spock extensions (shipped or 3rd party). Their order is currently depending on the order they are added, but there should not be made any order assumptions within one interception point.

An ellipsis in the figure means that the block before it can be repeated an arbitrary amount of times.

The …​ method interceptors are of course only run if there are actual methods of this type to be executed (the white boxes) and those can inject parameters to be given to the method that will be run.

The difference between shared initializer interceptor and shared initializer method interceptor and between initializer interceptor and initializer method interceptor - as there can be at most one of those methods each - is, that there are only the two methods if there are @Shared, respectively non-@Shared, fields that get values assigned at declaration time. The compiler will put those initializations in a generated method and call it at the proper place in the lifecycle. So if there are no such initializations, no method is generated and thus the method interceptor is never called. The non-method interceptors are always called at the proper place in the lifecycle to do work that has to be done at that time.

To create an interceptor to be attached to an interception point, you need to create a class that implements the interface IMethodInterceptor. This interface has the sole method intercept(IMethodInvocation invocation). The invocation parameter can be used to get and modify the current state of execution. Each interceptor must call the method invocation.proceed(), which will go on in the lifecycle, except you really want to prevent further execution of the nested elements like shown in the figure above. But this should be a very rare use case.

If you write an interceptor that can be used at different interception points and should do different work at different interception points, there is also the convenience class AbstractMethodInterceptor, which you can extend and which provides various methods for overriding that are called for the various interception points. Most of these methods have a double meaning, like interceptSetupMethod which is called for the setup interceptor and the setup method interceptor. If you attach your interceptor to both of them and need a differentiation, you can check for invocation.method.reflection, which will be set in the method interceptor case and null otherwise. Alternatively you can of course build two different interceptors or add a parameter to your interceptor and create two instances, telling each at addition time whether it is attached to the method interceptor or the other one.

Spock 1.1 introduced the DetachedMockFactory and the SpockMockFactoryBean which allow the creation of Spock mocks outside of a specification.

Spock ignores bean that is not a singleton (in the singleton scope) by default. To enable mocks to work for scoped beans you need to add @ScanScopedBeans to the spec and make sure that the scope allows access to the bean during the setup phase.

You can limit the scanning to certain scopes by using the value property of @ScanScopedBeans.

Thanks to all the contributors to this release: Marc Philipp, Rob Elliot, jochenberger, Jan Papenbrock, Paul King, Marcin Zajączkowski, mrb-twx, Alexander Kazakov, Serban Iordache, Xavier Fournet, timothy-long, John Osberg, AlexElin, Benjamin Muschko, Andreas Neumann, geoand, Burk Hufnagel, signalw, Martin Vseticka, Tilman Ginzel

Thanks to all the contributors to this release: Marc Philipp, Marcin Zajączkowski, signalw

Thanks to all the contributors to this release: Marc Philipp, Tilman Ginzel, Marcin Zajączkowski, Martin Vseticka

Thanks to all the contributors to this release: Rob Elliot, jochenberger, Jan Papenbrock, Paul King, Marcin Zajączkowski, mrb-twx, Alexander Kazakov, Serban Iordache, Xavier Fournet, timothy-long, John Osberg, AlexElin, Benjamin Muschko, Andreas Neumann, geoand, Burk Hufnagel

Thanks to all the contributors to this release: Taylor Wicksell, Rafael Winterhalter, Marcin Zajączkowski, Eduardo Grajeda, Paul King, Andrii, Björn Kautler, Libor Rysavy

Known issues with groovy 2.4.10 which breaks a smoke test, but should have little impact on normal use (#709).

Thanks to all the contributors to this release: Jochen Berger, Leonard Brünings, Mariusz Gilewicz, Tomasz Juchniewicz, Gamal Mateo, Tobias Schulte, Florian Wilhelm, Kevin Wittek

Special thanks to all the contributors to this release: Dmitry Andreychuk, Aseem Bansal, Daniel Bechler, Fedor Bobin, Leonard Brünings, Leonard Daume, Marcin Erdmann, Jarl Friis, Søren Berg Glasius, Serban Iordache, Michal Kordas, Pap Lőrinc, Vlad Muresan, Etienne Neveu, Glyn Normington, David Norton, Magnus Palmér, Gus Power, Oliver Reissig, Kevin Wittek and Marcin Zajączkowski

With a revamped build/release process and a reforming core team, we hope to release much more frequently from now on. Another big focus will be to better involve the community and their valuable contributions. Last but not least, we are finally shooting for a professional logo and website. Stay tuned for announcements!

Test Long And Prosper,

The Spock Team

17 awesome people contributed to this release:

21 burning issues were fixed:

18 hand-crafted pull requests were merged or cherry-picked:

These awesome extensions have been published or updated:

These great features didn’t make it into this release (but hopefully the next!):

Spock snapshots are now available from https://oss.sonatype.org/content/repositories/snapshots/org/spockframework/.

The new Spock reference documentation is available at http://docs.spockframework.org. It will gradually replace the documentation at http://wiki.spockframework.org. Each Spock version is documented separately (e.g. http://docs.spockframework.org/en/spock-0.7-groovy-1.8). Documentation for the latest Spock snapshot is at http://docs.spockframework.org/en/latest. As of Spock 0.7, the chapters on Data Driven Testing and Interaction Based Testing are complete.

The diagnostic message accompanying a TooManyInvocationsError has been greatly improved. Here is an example:

Reference Documentation

The diagnostic message accompanying a TooFewInvocationsError has been greatly improved. Here is an example:

Reference Documentation

Besides mocks, Spock now has explicit support for stubs:

A stub is a restricted form of mock object that responds to invocations without ever demanding them. Other than not having a cardinality, a stub’s interactions look just like a mock’s interactions. Using a stub over a mock is an effective way to communicate its role to readers of the specification.

Reference Documentation

Besides mocks, Spock now has support for spies:

A spy sits atop a real object, in this example an instance of class Person. All invocations on the spy that don’t match an interaction are delegated to that object. This allows to listen in on and selectively change the behavior of the real object. Furthermore, spies can be used as partial mocks.

Reference Documentation

Interactions can now be declared at mock creation time:

This feature is particularly attractive for Stubs.

Reference Documentation

Spock now offers specialized mock objects for spec’ing Groovy code:

A Groovy mock automatically implements groovy.lang.GroovyObject. It allows stubbing and mocking of dynamic methods just like for statically declared methods. When a Groovy mock is called from Java rather than Groovy code, it behaves like a regular mock.

Reference Documentation

A Groovy mock can be made global:

A global mock can only be created for a class type. It effectively replaces all instances of that type and makes them amenable to stubbing and mocking. (You may know this behavior from Groovy’s MockFor and StubFor facilities.) Furthermore, a global mock allows mocking of the type’s constructors and static methods.

Reference Documentation

Inspired from Groovy’s Object.with method, the Specification.with method allows to group conditions involving the same target object:

The with method can also be used for grouping interactions:

Reference Documentation

spock.util.concurrent.PollingConditions joins AsyncConditions and BlockingVariable(s) as another utility for testing asynchronous code:

Spock now ships with a DSL descriptor that lets Groovy Eclipse better understand certain parts of Spock’s DSL. The descriptor is automatically detected and activated by the IDE. Here is an example:

Another example:

DSL support is activated for Groovy Eclipse 2.7.1 and higher. If necessary, it can be deactivated in the Groovy Eclipse preferences.

Spock now ships with a DSL descriptor that lets Intellij IDEA better understand certain parts of Spock’s DSL. The descriptor is automatically detected and activated by the IDE. Here is an example:

Another example:

DSL support is activated for IntelliJ IDEA 11.1 and higher.

Parts of class spock.lang.Specification were pulled up into two new super classes: spock.lang.MockingApi now contains all mocking-related methods, and org.spockframework.lang.SpecInternals contains internal methods which aren’t meant to be used directly.

Instead of just passing through exceptions, Specification.notThrown and Specification.noExceptionThrown now fail with messages like:

Class spock.util.matcher.HamcrestSupport has a new expect method that makes Hamcrest assertions read better in then-blocks:

Recently introduced classes and methods may be annotated with @Beta, as a sign that they may still undergo incompatible changes. This gives us a chance to incorporate valuable feedback from our users. (Yes, we need your feedback!) Typically, a @Beta annotation is removed within one or two releases.

See the issue tracker for a list of fixed issues.

The mocking framework now provides better diagnostic messages in some cases.

Multiple result declarations can be chained. The following causes method bar to throw an IOException when first called, return the numbers one, two, and three on the next calls, and throw a RuntimeException for all subsequent calls:

It’s now possible to match any argument list (including the empty list) with foo.bar(*_).

Method arguments can now be constrained with Hamcrest matchers:

In addition to rules implementing org.junit.rules.MethodRule (which has been deprecated in JUnit 4.9), Spock now also supports rules implementing the new org.junit.rules.TestRule interface. Also supported is the new @ClassRule annotation. Rule declarations are now verified and can leave off the initialization part. I that case Spock will automatically initialize the rule by calling the default constructor. The @TestName rule, and rules in general, now honor the @Unroll annotation and any defined naming pattern.

See Issue 240 for a known limitation with Spock’s TestRule support.

When two objects are compared with the == operator, they are unequal, but their string representations are the same, Spock will now print the objects' types:

Fixture methods can now be declared with JUnit’s @Before, @After, @BeforeClass, and @AfterClass annotations, as an addition or alternative to Spock’s own fixture methods. This was particularly needed for Grails 2.0 support.

Thanks to a contribution from Howard Lewis Ship, the Tapestry module is now compatible with Tapestry 5.3. Older 5.x versions are still supported.

Spock now runs fine on IBM JDKs, working around a bug in the IBM JDK’s verifier.

org.junit.internal.AssumptionViolatedException is now recognized and handled as known from JUnit. @Unrolled methods no longer cause "yellow" nodes in IDEs.

The @Unroll naming pattern can now be provided in the method name, instead of as an argument to the annotation:

The naming pattern now supports property access and zero-arg method calls:

The @Unroll annotation can now be applied to a spec class. In this case, all data-driven feature methods in the class will be unrolled.

The @Timeout annotation can now be applied to a spec class. In this case, the timeout applies to all feature methods (individually) that aren’t already annotated with @Timeout. Timed methods are now executed on the regular test framework thread. This can be important for tests that rely on thread-local state (like Grails integration tests). Also the interruption behavior has been improved, to increase the chance that a timeout can be enforced.

The failure exception that is thrown when a timeout occurs now contains the stacktrace of test execution, allowing you to see where the test was “stuck” or how far it got in the allocated time.

Table cells can now be separated with double pipes. This can be used to visually set apart expected outputs from provided inputs:

Spock 0.6 ships in three variants for Groovy 1.7, 1.8, and 2.0. Make sure to pick the right version - for example, for Groovy 1.8 you need to use spock-core-0.6-groovy-1.8 (likewise for all other modules). The Groovy 2.0 variant is based on Groovy 2.0-beta-3-SNAPSHOT and only available from http://m2repo.spockframework.org. The Groovy 1.7 and 1.8 variants are also available from Maven Central. The next version of Spock will no longer support Groovy 1.7.

Spock’s Grails plugin was split off into a separate project and now lives at http://github.spockframework.org/spock-grails. The plugin supports both Grails 1.3 and 2.0.

The Spock Grails plugin supports all of the new Grails 2.0 test mixins, effectively deprecating the existing unit testing classes (e.g. UnitSpec). For integration testing, IntegrationSpec must still be used.

The folks from JetBrains have added a few handy features around data tables. Data tables will now be layed out automatically when reformatting code. Data variables are no longer shown as "unknown" and have their types inferred from the values in the table (!).

All source code has moved to http://github.spockframework.org/. The Grails Spock plugin, Spock Example project, and Spock Web Console now have their own GitHub projects. Also available are slides and code for various Spock presentations (such as this one).

Spock is now exclusively built with Gradle. Building Spock yourself is as easy as cloning the Github repo and executing gradlew build. No build tool installation is required; the only prerequisite for building Spock is a JDK installation (1.5 or higher).

See the issue tracker for a list of fixed issues.

Given these specifications:

In 0.5, above assignments happened in the order base1, base2, derived1, derived2. In other words, field initializers were executed right before the setup method in the same class. In 0.6, assignments happen in the order base1, derived1, base2, derived2. This is a more conventional order that solves a few problems that users faced with the previous behavior, and also allows us to support JUnit’s new TestRule. As a result of this change, the following will no longer work:

To overcome this problem, you can either use a field initializer for base, or move the assignment of derived into a setup method.

In 0.5, the naming pattern was string based:

In 0.6-SNAPSHOT, this was changed to a closure returning a GString:

For various reasons, the new syntax didn’t work out as we had hoped, and eventually we decided to go back to the string based syntax. See Improved @Unroll for recent improvements to that syntax.

Spock offers a very neat syntax for using Hamcrest matchers:

Due to changes made between Groovy 1.7 and 1.8, this syntax no longer works in as many cases as it did before. For example, the following will no longer work:

To avoid such problems, use HamcrestSupport.that:

A future version of Spock will likely remove the former syntax and strengthen the latter one.