YMNNALFT: Reactive Dataflow with Project Reactor

Engineering | Josh Long | January 05, 2021 | ...

Welcome to another installment of You May Not Need Another Library For That (YMNNALFT)! I've spent a lot of time since 2016 illuminating (or trying to, anyway!) some of the more enormous opportunities in the Spring ecosystem in my Spring Tips videos. Today, however, I come to you in a different spirit, wanting to focus on the little, sometimes hidden, gems that do fantastic things and that might spare you an additional third-party dependency and its implied complexity.

And it's a good thing we're covering some of these complexity-reducing gems, too, you see, because the world is a confusing, complex, and dizzying place, and nowhere is that more evident than in the world of reactive data stream composition. Life comes at you fast and data, even more so. Data originates from everything (network services, databases, in-memory computation, threaded code, etc.). Data comes in many different sizes and shapes (one record, ten records, infinite records, etc.). Data comes in different volumes, starving or overwhelming its consumers. Data arrives at different cadences and times: now, later, all at once, periodically, etc. It's much more natural to talk about data that's already in memory, in hand, so to speak, than data that hasn't yet materialized. And it's more natural to talk about data in the single-threaded case than to deal with it in the concurrent case.

It's confusing to think about! Historically, it's been even more tedious to address the different data dimensions in code. That is, until reactive programming.

Reactive programming offers a unified world view, allowing us to think about the messy world of (potentially asynchronous and potentially concurrent) integration in terms of an easy to use DSL with operators. The operators support the definition and composition of reactive stream data flow pipelines. Reactive programming offers a structured concurrency paradigm, greatly simplifying writing safe, scalable, resource-efficient code.

There are some great libraries out there (like RxJava and Akka Streams) that work in much the same way. If you haven't got a particular one in mind but want a world-class option and are already using Spring, then you might as well use Project Reactor. It's included in the box!

The Spring team developed Project Reactor to support the reactive efforts in the Spring ecosystem. You don't need Spring to use Project Reactor, but all the reactive APIs in the Spring ecosystem build on Project Reactor for the dataflow options. Microsoft mandates that all their SDK clients and APIs be created using Project Reactor. Facebook developed the Java client for their RSocket protocol using Project Reactor. Project Reactor is mature - it's been around since 2010! - but growing new features all the time. If you're still not getting what you need, it works flawlessly with other reactive data flow libraries through the interoperable Reactive Streams types.

Let's look at an example of how Project Reactor makes short work of the composition of different data flow sources and sinks and all but eliminates any manual threading code. This is a huge win. Remember: only one person truly understands how to write safe, useful, multithreaded Java code.... and it's not you! I don't know who it is. It doesn't matter. Don't tempt fate; let Project Reactor help.

You'll need the following dependencies.

Now, let's look at a sample. This example demonstrates how easy it is to normalize processing given disparate kinds of data. In this example, we look at a Java 8 java.util.Stream<T?> and a CompletableFuture<T>, but the sky's the limit. In most reactive applications, you won't necessarily be in the business of converting non-reactive types to reactive types (like Flux<T> or Mono<T>). Those examples would be even more straightforward. This example assumes that you've got two data sources and need to compose them.

package bootiful.rx;

import org.springframework.boot.SpringApplication;
import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.boot.context.event.ApplicationReadyEvent;
import org.springframework.context.ApplicationListener;
import org.springframework.context.annotation.Bean;
import reactor.core.publisher.Flux;
import reactor.core.publisher.Mono;

import java.util.concurrent.CompletableFuture;
import java.util.stream.Stream;

public class BootifulApplication {

	CompletableFuture<String> returnCompletableFuture(int counter) {
		return CompletableFuture.supplyAsync(() -> {
			var start = System.currentTimeMillis();
			try {
				Thread.sleep((long) (Math.max((Math.random() * 10), 5) * 1000));
			catch (InterruptedException e) {
				// threads smdh
			var stop = System.currentTimeMillis();
			var delta = stop - start;
			return "(" + Thread.currentThread().getName() + ") Hello, #" + counter + "! (after " + delta + " ms.)";

	Stream<Integer> returnStream() {
		return Stream.iterate(0, integer -> integer + 1);

	ApplicationListener<ApplicationReadyEvent> begin() {
		return event -> {

			Flux<String> count = Flux//
					.fromStream(this.returnStream()) //
					.take(10) //
					.flatMap(c -> Flux.zip(Mono.just(c), Mono.fromCompletionStage(this.returnCompletableFuture(c)))) //
					.map(tuple -> tuple.getT2() + " #" + tuple.getT1()); //


	public static void main(String[] args) {
		SpringApplication.run(BootifulApplication.class, args);


Do you know what else you get with Reactor? Operators supporting retries, error handling, timeouts, etc., all of which you would also farm out to yet another third-party library if they weren't included in Project Reactor. Win-win.

I could go on and on about the opportunities that present themselves given something like Project Reactor. Indeed, I did. Check out my book Reactive Spring for (a lot) more.

Well? Did you have fun, and maybe learn something? As always, I'm keen on hearing from you, so please sound off on Twitter (@starbuxman) ! I'll be back with another installment of YMNNALFT later this week, so be sure not to miss that. I've got installments on, among other things, Easy RPC, The Garden of *Utils objects, Dimensional Metrics with Micrometer, and many, many more topics.

Get the Spring newsletter

Thank you!

Get ahead

VMware offers training and certification to turbo-charge your progress.

Learn more

Get support

Spring Runtime offers support and binaries for OpenJDK™, Spring, and Apache Tomcat® in one simple subscription.

Learn more

Upcoming events

Check out all the upcoming events in the Spring community.

View all