Latest news

• 19/06/2017: Version 3.7.0 is released. Checkout what's new in the change log.
• 05/03/2017: Version 3.6.0 is out with new features and bug fixes. See all details in the change log.

What is Random Beans ?

Random Beans is a library that generates random Java beans. Let's say you have a class Person and you want to generate a random instance of it, here we go:

Person person = random(Person.class);

The static method EnhancedRandom.random is able to generate random instances of a given type.

Let's see another example. If you want to generate a random stream of 10 persons, you can use the following snippet:

Stream<Person> persons = randomStreamOf(10, Person.class);

This second static method of the EnhancedRandom API generates a stream of random instances of a given type.

What is this EnhancedRandom API?

The java.util.Random API provides 7 methods to generate random data: nextInt(), nextLong(), nextDouble(), nextFloat(), nextBytes(), nextBoolean() and nextGaussian(). What if you need to generate a random String? Or say a random instance of your domain object? Random Beans provides the EnhancedRandom API that extends (enhances) java.util.Random with a method called nextObject(Class type). This method is able to generate a random instance of any arbitrary Java bean:

EnhancedRandom enhancedRandom = EnhancedRandomBuilder.aNewEnhancedRandom();
 Person person = enhancedRandom.nextObject(Person.class);

The EnhancedRandomBuilder is the main entry point to configure EnhancedRandom instances. It allows you to set all parameters to control how random data is generated:

EnhancedRandom random = EnhancedRandomBuilder.aNewEnhancedRandomBuilder()
 .seed(123L)
 .objectPoolSize(100)
 .randomizationDepth(3)
 .charset(forName("UTF-8"))
 .timeRange(nine, five)
 .dateRange(today, tomorrow)
 .stringLengthRange(5, 50)
 .collectionSizeRange(1, 10)
 .scanClasspathForConcreteTypes(true)
 .overrideDefaultInitialization(false)
 .build();

For more details about these parameters, please refer to the configuration parameters section.

In most cases, default options are enough and you can use a static import of the EnhancedRandom.random(Class object) as seen previously.

Random beans allows you to control how to generate random data through the Randomizer interface and makes it easy to exclude some fields from the object graph using the fluent FieldDefinition API:

EnhancedRandom enhancedRandom = EnhancedRandomBuilder.aNewEnhancedRandomBuilder()
 .randomize(String.class, (Randomizer<String>) () -> "foo")
 .exclude(field().named("age").ofType(Integer.class).inClass(Person.class).get())
 .build();
  Person person = enhancedRandom.nextObject(Person.class);

Why Random Beans ?

Populating a Java object with random data can look easy at first glance, unless your domain model involves many related classes. In the previous example, let's suppose the Person type is defined as follows:

Without Random Beans, you would write the following code in order to create an instance of the Person class:

Street street = new Street(12, (byte) 1, "Oxford street");
 Address address = new Address(street, "123456", "London", "United Kingdom");
 Person person = new Person("Foo", "Bar", "[email protected]", Gender.MALE, address);

And if these classes do not provide constructors with parameters (may be some legacy beans you don't have the control over), you would write:

Street street = new Street();
 street.setNumber(12);
 street.setType((byte) 1);
 street.setName("Oxford street");
  Address address = new Address();
 address.setStreet(street);
 address.setZipCode("123456");
 address.setCity("London");
 address.setCountry("United Kingdom");
  Person person = new Person();
 person.setFirstName("Foo");
 person.setLastName("Bar");
 person.setEmail("[email protected]");
 person.setGender(Gender.MALE);
 person.setAddress(address);

With Random Beans, generating a random Person object is done with random(Person.class). The library will recursively populate all the object graph. That's a big difference!

How can this be useful ?

Sometimes, the test fixture does not really matter to the test logic. For example, if we want to test the result of a new sorting algorithm, we can generate random input data and assert the output is sorted, regardless of the data itself:

@org.junit.Test public void testSortAlgorithm() {

  // Given
 int[] ints = aNewEnhancedRandom().nextObject(int[].class);

  // When
 int[] sortedInts = myAwesomeSortAlgo.sort(ints);

  // Then
 assertThat(sortedInts).isSorted();
 // fake assertion  
}

Another example is testing the persistence of a domain object, we can generate a random domain object, persist it and assert the database contains the same values:

@org.junit.Test public void testPersistPerson() throws Exception {

 // Given
 Person person = random(Person.class);

  // When
 personDao.persist(person);

  // Then
 assertThat("person_table").column("name").value().isEqualTo(person.getName());
 // assretj db 
}

There are many other uses cases where random beans can be useful, you can find a non exhaustive list in the wiki.

Contribution

You are welcome to contribute to the project with pull requests on GitHub.

If you believe you found a bug, please use the issue tracker.

If you have any question, suggestion, or feedback, do not hesitate to use the Gitter channel of the project.

Core team and contributors

Core team

• Mahmoud Ben Hassine
• Pascal Schumacher

Awesome contributors

• Adriano Machado
• Alberto Lagna
• Dovid Kopel
• huningd
• Eric Taix
• euZebe
• Fred Eckertson
• Jose Manuel Prieto
• kermit-the-frog
• Lucas Andersson
• Nikola Milivojevic
• Oleksandr Shcherbyna
• Petromir Dzhunev
• Ryan Dunckel
• Rebecca McQuary
• Rémi Alvergnat
• Rodrigue Alcazar
• Sam Van Overmeire
• Valters Vingolds
• Vincent Potucek

Thank you all for your contributions!