Reactive Streams With Spring Data and MongoDB

DZone 's Guide to

Reactive Streams With Spring Data and MongoDB

Spring Data and MongoDB have made it easy to include Reactive Streams in your projects. Here we cover the config work and changes to annotations you need to know.

· Java Zone ·
Free Resource

This is a short post looking into Reactive Streams and how they can be used with MongoDB and Spring Data. This post won’t go into the depths of what Reactive Programming and Reactive Streams are, as there have been plenty of posts covering that recently, such as What are Reactive Streams in Java and Reactive Spring 5 and Application Design Impact. Instead, it will simply demonstrate how to use the newer versions (at the time of writing) of Spring Data, which comes equipped with the features necessary to use Reactive Streams. In this post, we will be using MongoDB due to it being one of the few currently available databases with a Reactive implementation with Spring Data. The others include (at the time of writing) Cassandra and Redis.

In terms of getting everything set up, not much is different to using the non-Reactive version of MongoDB with Spring Data. The main difference that you will see is that the word “reactive” pops up into loads of class names, such as ReactiveMongoRepository instead of MongoRepository, hopefully helping you distinguish between them. The other main difference is that, instead of returning a document or a list of documents, you will now receive some slightly different objects. In this post, we will use Reactor as our Reactive Streaming library, meaning that the returned objects are Mono<T> for a singular document and a Flux<T> when multiple documents are being returned. For more information on Flux and Mono from Project Reactor, have a look at this Intro into Reactor Core.

So let's get started. First things first: Let's include the relevant projects as dependencies in our code. Below are the maven dependencies required.

        <name>Spring Snapshot Repository</name>

For this tutorial, Spring was kind enough to have the spring-boot-starter-data-mongodb-reactive dependency ready-made for us that contains everything we need to get going. This includes dependencies such as spring-data-mongodb and reactor-core at the correct versions to allow us to use Reactive Streams. While we’re talking about versioning, it is also important to use the newer version of spring-boot-starter-parent, which needs to be 2.0.0 or above to include the reactive libraries. As the dependency for this is currently a snapshot, the repository needs to be defined to retrieve it.

@EnableReactiveMongoRepositories(basePackageClasses = PersonRepository.class)
public class MongoConfig extends AbstractReactiveMongoConfiguration {

  public MongoClient mongoClient() {
    return MongoClients.create();

  protected String getDatabaseName() {
    return "test";

  public ReactiveMongoTemplate reactiveMongoTemplate() {
    return new ReactiveMongoTemplate(mongoClient(), getDatabaseName());

Here, we have a @Configuration class that extends AbsractReactiveMongoConfiguration to provide some beans to setup the application to use reactive MongoDB. This class is not required if non-Reactive spring-data-mongodb code was being used.

@EnableReactiveMongoRepositories goes back to what I said earlier in the post. It has taken an existing annotation, @EnableMongoRepositories, and added the word “reactive” to it. Through the use of this annotation, implementations of ReactiveCrudRepository can be used. The above example specifies the PersonRepository (which we will look at later), as it is found in a different package to this configuration class. If you wish to still use a mix-and-match of Reactive and non-reactive repositories, you will still need to include the @EnableMongoRepositories annotation. The bean mongoClient calls MongoClients.create() to instantiate a MongoClient with the default connection of mongodb://localhost. Settings or a different connection string could be passed in if desired. This client is used along with the database name to create the ReactiveMongoTemplate bean.

public interface PersonRepository extends ReactiveMongoRepository < Person, String > {
    Flux < Person > findByFirstName(final String firstName);
    Mono < Person > findOneByFirstName(final String firstName);

As described in one of my earlier posts, Getting started with Spring Data and MongoDB, the implementation of PersonRepository is not required, as the executed code is inferred from the name of the methods specified on the interface. Again this another example of where the Reactive version closely resembles its original (MongoRepository in case you're wondering). ReactiveMongoRepository inherits from ReactiveCrudRepository, allowing the @EnableReactiveMongoRepositories to include it when setting up. As mentioned earlier in this post, Mono<Person> and Flux<Person> are returned instead of Person and List<Person> respectively.

Finally, to put it all together, we need to create the main application and, using the CommandLineRunner , we can give the code a quick trial run.

@SpringBootApplication public class Application implements CommandLineRunner {
    @Autowired private PersonRepository personRepository;
    public static void main(String args[]) {
    @Override public void run(String args[]) {
        final Person johnAoe = new Person("john", "aoe", LocalDateTime.now(), "loser", 0);
        final Person johnBoe = new Person("john", "boe", LocalDateTime.now(), "a bit of a loser", 10);
        final Person johnCoe = new Person("john", "coe", LocalDateTime.now(), "average", 100);
        final Person johnDoe = new Person("john", "doe", LocalDateTime.now(), "winner", 1000);
        personRepository.saveAll(Flux.just(johnAoe, johnBoe, johnCoe, johnDoe)).subscribe();

Running this piece of code creates some initial data and then retrieves it. log is called to demonstrate what is going on inside the Reactive streams, and the output of the streams are printed to the console using the subscribe method, along with the Method Reference of System.out::println.

2017-07-16 16:44:09.201 INFO 13476 --- [ main] reactor.Flux.OnErrorResume.1 : onSubscribe(FluxOnErrorResume.ResumeSubscriber)
2017-07-16 16:44:09.208 INFO 13476 --- [ main] reactor.Flux.OnErrorResume.1 : request(unbounded)
2017-07-16 16:44:09.242 INFO 13476 --- [ Thread-4] reactor.Flux.OnErrorResume.1 : onNext(Person(firstName=john, secondName=aoe, profession=loser, salary=0))
2017-07-16 16:44:09.243 INFO 13476 --- [ Thread-4] reactor.Flux.OnErrorResume.1 : onNext(Person(firstName=john, secondName=boe, profession=a bit of a loser, salary=10))
2017-07-16 16:44:09.244 INFO 13476 --- [ Thread-4] reactor.Flux.OnErrorResume.1 : onNext(Person(firstName=john, secondName=coe, profession=average, salary=100))
2017-07-16 16:44:09.245 INFO 13476 --- [ Thread-4] reactor.Flux.OnErrorResume.1 : onNext(Person(firstName=john, secondName=doe, profession=winner, salary=1000))
2017-07-16 16:44:09.247 INFO 13476 --- [ Thread-4] reactor.Flux.OnErrorResume.1 : onComplete()
2017-07-16 16:44:09.254 INFO 13476 --- [ main] reactor.Mono.OnErrorResume.2 : onSubscribe(FluxOnErrorResume.ResumeSubscriber)
2017-07-16 16:44:09.255 INFO 13476 --- [ main] reactor.Mono.OnErrorResume.2 : request(unbounded)
2017-07-16 16:44:09.260 INFO 13476 --- [ Thread-4] reactor.Mono.OnErrorResume.2 : onNext(Person(firstName=john, secondName=aoe, profession=loser, salary=0))
2017-07-16 16:44:09.260 INFO 13476 --- [ Thread-4] reactor.Mono.OnErrorResume.2 : onComplete()

Hopefully, you can get the gist of what is going on by looking at the console output above. In case you want clarification, onSubscribe is output due to calling subscribe onto one of the Reactive streams, triggering a request to retrieve elements from the stream and for each element onNext that is called. Finally, after the last element is received, onComplete is called. Stuck in between these log messages are the print lines that were output from the subscribe method.

In conclusion, getting up a running using Reactive Streams with Spring Data and MongoDB is no harder than using their non-Reactive counterparts. All we need to do is add a small amount of extra configuration and insert the word “reactive” into a few class and interface names and use the Flux and Mono types (from Reactor) instead of directly returning a list or object.

The code used in this post can be found on my GitHub.

java ,mongodb ,reactive streams ,spring data ,tutorial

Published at DZone with permission of Dan Newton , DZone MVB. See the original article here.

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}