Testing Kotlin with Spock Part 3 – Interface default method

Kotlin allows you to put method implementation in an interface. The same mechanism can be found in Java interfaces as default methods (and also Groovy or Scala traits). Let’s see the difference between the Kotlin and Java default methods in interface by testing it with Groovy and Spock.

What do we want to test?

We often have an interface for access object from the database. In domain, they might look similar to this KotlinOrderRepository:

interface KotlinOrderRepository {
    fun save(order: Order)

    fun find(orderId: OrderId): Order?

    fun get(orderId: OrderId): Order =
            find(orderId) ?: throw NotFound()
}

How to fake it with Groovy?

When we want to use such interface in tests, we can, of course, mock it. However, it is far better to fake repositories with a simple, in-memory implementation. Let’s create FakeKotlinOrderRepository in Groovy:

class FakeKotlinOrderRepository implements KotlinOrderRepository {
    private Map<OrderId, Order> data = [:]

    @Override
    void save(Order order) {
        data[order.id] = order
    }

    @Override
    Order find(OrderId orderId) {
        return data[orderId]
    }
}

Unfortunately, this causes a compilation error

/testing-kotlin-in-spock/src/test/groovy/com/github/alien11689/testingkotlinwithspock/defaultmethod/FakeKotlinOrderRepository.groovy: 3: Can't have an abstract method in a non-abstract class. The class 'com.github.alien11689.testingkotlinwithspock.defaultmethod.FakeKotlinOrderRepository' must be declared abstract or the method 'com.github.alien11689.testingkotlinwithspock.defaultmethod.Order get(com.github.alien11689.testingkotlinwithspock.defaultmethod.OrderId)' must be implemented.
 @ line 3, column 1.
   class FakeKotlinOrderRepository implements KotlinOrderRepository {
   ^

1 error

The compiler doesn’t see the implementation of the get method in the Kotlin interface. We have to use some magic to make it work in groovy.

Solution

To solve the problem, let’s look into the generated classes:

$ ls build/classes/main/com/github/alien11689/testingkotlinwithspock/defaultmethod/
JavaOrderRepository.class
KotlinOrderRepository.class
KotlinOrderRepository$DefaultImpls.class
NotFound.class
Order.class
OrderId.class

The KotlinOrderRepository$DefaultImpls class is the one we’re looking for as we can use it in Groovy to implement the missing operation.

class FakeKotlinOrderRepository implements KotlinOrderRepository {

    // ...

    Order get(OrderId orderId) {
        return KotlinOrderRepository.DefaultImpls.get(this, orderId)
    }
}

Now the code compiles and tests pass:

class KotlinRepositoryWithDefaultMethodTest extends Specification {
    OrderId orderId = new OrderId(UUID.randomUUID() as String)
    Order order = new Order(orderId, 'data')
    KotlinOrderRepository kotlinOrderRepository = new FakeKotlinOrderRepository()

    def 'should get order from kotlin repository'() {
        given:
            kotlinOrderRepository.save(order)
        expect:
            kotlinOrderRepository.get(orderId) == order
    }

    def 'should throw NotFound when order does not exist in kotlin repository'() {
        when:
            kotlinOrderRepository.get(orderId)
        then:
            thrown(NotFound)
    }
}

Is there the same problem with Java?

Let’s have a quick look at how this works with Java interfaces. If we write a similar repository in Java:

public interface JavaOrderRepository {
    void save(Order order);

    Optional<Order> find(OrderId orderId);

    default Order get(OrderId orderId) {
        return find(orderId).orElseThrow(NotFound::new);
    }
}

and create a fake implementation in Groovy:

class FakeJavaOrderRepository implements JavaOrderRepository {
    private Map<OrderId, Order> data = [:]

    @Override
    void save(Order order) {
        data[order.id] = order
    }

    @Override
    Optional<Order> find(OrderId orderId) {
        return Optional.ofNullable(data[orderId])
    }
}

there is no compilation error and the tests pass:

class JavaRepositoryWithDefaultMethodTest extends Specification {
    OrderId orderId = new OrderId(UUID.randomUUID() as String)
    Order order = new Order(orderId, 'data')
    JavaOrderRepository javaOrderRepository = new FakeJavaOrderRepository()

    def 'should get order from java repository'() {
        given:
            javaOrderRepository.save(order)
        expect:
            javaOrderRepository.get(orderId) == order
    }

    def 'should throw NotFound when order does not exist in java repository'() {
        when:
            javaOrderRepository.get(orderId)
        then:
            thrown(NotFound)
    }
}

Groovy can implement Java interfaces with the default methods without any problems.

Show me the code

Code is available here.

You May Also Like

Recently at storm-users

I've been reading through storm-users Google Group recently. This resolution was heavily inspired by Adam Kawa's post "Football zero, Apache Pig hero". Since I've encountered a lot of insightful and very interesting information I've decided to describe some of those in this post.

  • nimbus will work in HA mode - There's a pull request open for it already... but some recent work (distributing topology files via Bittorrent) will greatly simplify the implementation. Once the Bittorrent work is done we'll look at reworking the HA pull request. (storm’s pull request)

  • pig on storm - Pig on Trident would be a cool and welcome project. Join and groupBy have very clear semantics there, as those concepts exist directly in Trident. The extensions needed to Pig are the concept of incremental, persistent state across batches (mirroring those concepts in Trident). You can read a complete proposal.

  • implementing topologies in pure python with petrel looks like this:

class Bolt(storm.BasicBolt):
    def initialize(self, conf, context):
       ''' This method executed only once '''
        storm.log('initializing bolt')

    def process(self, tup):
       ''' This method executed every time a new tuple arrived '''       
       msg = tup.values[0]
       storm.log('Got tuple %s' %msg)

if __name__ == "__main__":
    Bolt().run()
  • Fliptop is happy with storm - see their presentation here

  • topology metrics in 0.9.0: The new metrics feature allows you to collect arbitrarily custom metrics over fixed windows. Those metrics are exported to a metrics stream that you can consume by implementing IMetricsConsumer and configure with Config.java#L473. Use TopologyContext#registerMetric to register new metrics.

  • storm vs flume - some users' point of view: I use Storm and Flume and find that they are better at different things - it really depends on your use case as to which one is better suited. First and foremost, they were originally designed to do different things: Flume is a reliable service for collecting, aggregating, and moving large amounts of data from source to destination (e.g. log data from many web servers to HDFS). Storm is more for real-time computation (e.g. streaming analytics) where you analyse data in flight and don't necessarily land it anywhere. Having said that, Storm is also fault-tolerant and can write to external data stores (e.g. HBase) and you can do real-time computation in Flume (using interceptors)

That's all for this day - however, I'll keep on reading through storm-users, so watch this space for more info on storm development.

I've been reading through storm-users Google Group recently. This resolution was heavily inspired by Adam Kawa's post "Football zero, Apache Pig hero". Since I've encountered a lot of insightful and very interesting information I've decided to describe some of those in this post.

  • nimbus will work in HA mode - There's a pull request open for it already... but some recent work (distributing topology files via Bittorrent) will greatly simplify the implementation. Once the Bittorrent work is done we'll look at reworking the HA pull request. (storm’s pull request)

  • pig on storm - Pig on Trident would be a cool and welcome project. Join and groupBy have very clear semantics there, as those concepts exist directly in Trident. The extensions needed to Pig are the concept of incremental, persistent state across batches (mirroring those concepts in Trident). You can read a complete proposal.

  • implementing topologies in pure python with petrel looks like this:

class Bolt(storm.BasicBolt):
    def initialize(self, conf, context):
       ''' This method executed only once '''
        storm.log('initializing bolt')

    def process(self, tup):
       ''' This method executed every time a new tuple arrived '''       
       msg = tup.values[0]
       storm.log('Got tuple %s' %msg)

if __name__ == "__main__":
    Bolt().run()
  • Fliptop is happy with storm - see their presentation here

  • topology metrics in 0.9.0: The new metrics feature allows you to collect arbitrarily custom metrics over fixed windows. Those metrics are exported to a metrics stream that you can consume by implementing IMetricsConsumer and configure with Config.java#L473. Use TopologyContext#registerMetric to register new metrics.

  • storm vs flume - some users' point of view: I use Storm and Flume and find that they are better at different things - it really depends on your use case as to which one is better suited. First and foremost, they were originally designed to do different things: Flume is a reliable service for collecting, aggregating, and moving large amounts of data from source to destination (e.g. log data from many web servers to HDFS). Storm is more for real-time computation (e.g. streaming analytics) where you analyse data in flight and don't necessarily land it anywhere. Having said that, Storm is also fault-tolerant and can write to external data stores (e.g. HBase) and you can do real-time computation in Flume (using interceptors)

That's all for this day - however, I'll keep on reading through storm-users, so watch this space for more info on storm development.

Distributed scans with HBase

HBase is by design a columnar store, that is optimized for random reads. You just ask for a row using rowId as an identifier and you get your data instantaneously. Performing a scan on part or whole table is a completely different thing. First of all, it is sequential. Meaning it is rather slow, because it doesn't use all the RegionServers at the same time. It is implemented that way to realize the contract of Scan command - which has to return results sorted by key. So, how to do this efficiently?HBase is by design a columnar store, that is optimized for random reads. You just ask for a row using rowId as an identifier and you get your data instantaneously. Performing a scan on part or whole table is a completely different thing. First of all, it is sequential. Meaning it is rather slow, because it doesn't use all the RegionServers at the same time. It is implemented that way to realize the contract of Scan command - which has to return results sorted by key. So, how to do this efficiently?

Grails with Spock unit test + IntelliJ IDEA = No thread-bound request found

During my work with Grails project using Spock test in IntelliJ IDEA I've encountered this error:

java.lang.IllegalStateException: No thread-bound request found: Are you referring to request attributes outside of an actual web request, or processing a request outside of the originally receiving thread? If you are actually operating within a web request and still receive this message, your code is probably running outside of DispatcherServlet/DispatcherPortlet: In this case, use RequestContextListener or RequestContextFilter to expose the current request.
at org.springframework.web.context.request.RequestContextHolder.currentRequestAttributes(RequestContextHolder.java:131)
at org.codehaus.groovy.grails.plugins.web.api.CommonWebApi.currentRequestAttributes(CommonWebApi.java:205)
at org.codehaus.groovy.grails.plugins.web.api.CommonWebApi.getParams(CommonWebApi.java:65)
... // and few more lines of stacktrace ;)

It occurred when I tried to debug one of test from IDEA level. What is interesting, this error does not happen when I'm running all test using grails test-app for instance.

So what was the issue? With little of reading and tip from Tomek Kalkosiński (http://refaktor.blogspot.com/) it turned out that our test was missing @TestFor annotation and adding it solved all problems.

This annotation, according to Grails docs (link), indicates Spock what class is being tested and implicitly creates field with given type in test class. It is somehow strange as problematic test had explicitly and "manually" created field with proper controller type. Maybe there is a problem with mocking servlet requests?