The purpose of Reactive Streams is to provide a standard for asynchronous stream processing with non-blocking backpressure.
The latest preview release is available on Maven Central as
<dependency>
<groupId>org.reactivestreams</groupId>
<artifactId>reactive-streams</artifactId>
<version>1.0.0</version>
</dependency>
<dependency>
<groupId>org.reactivestreams</groupId>
<artifactId>reactive-streams-tck</artifactId>
<version>1.0.0</version>
<scope>test</scope>
</dependency>Handling streams of data—especially “live” data whose volume is not predetermined—requires special care in an asynchronous system. The most prominent issue is that resource consumption needs to be carefully controlled such that a fast data source does not overwhelm the stream destination. Asynchrony is needed in order to enable the parallel use of computing resources, on collaborating network hosts or multiple CPU cores within a single machine.
The main goal of Reactive Streams is to govern the exchange of stream data across an asynchronous boundary – think passing elements on to another thread or thread-pool — while ensuring that the receiving side is not forced to buffer arbitrary amounts of data. In other words, backpressure is an integral part of this model in order to allow the queues which mediate between threads to be bounded. The benefits of asynchronous processing would be negated if the communication of backpressure were synchronous (see also the Reactive Manifesto), therefore care has been taken to mandate fully non-blocking and asynchronous behavior of all aspects of a Reactive Streams implementation.
It is the intention of this specification to allow the creation of many conforming implementations, which by virtue of abiding by the rules will be able to interoperate smoothly, preserving the aforementioned benefits and characteristics across the whole processing graph of a stream application.
It should be noted that the precise nature of stream manipulations (transformation, splitting, merging, etc.) is not covered by this specification. Reactive Streams are only concerned with mediating the stream of data between different processing elements. In their development care has been taken to ensure that all basic ways of combining streams can be expressed.
In summary, Reactive Streams is a standard and specification for Stream-oriented libraries for the JVM that
- process a potentially unbounded number of elements
- in sequence,
- asynchronously passing elements between components,
- with mandatory non-blocking backpressure.
The Reactive Streams specification consists of the following parts:
The API specifies the types to implement Reactive Streams and achieve interoperability between different implementations.
The Technology Compatibility Kit (TCK) is a standard test suite for conformance testing of implementations.
Implementations are free to implement additional features not covered by the specification as long as they conform to the API requirements and pass the tests in the TCK.
The API consists of the following components that are required to be provided by Reactive Stream implementations:
- Publisher
- Subscriber
- Subscription
- Processor
A Publisher is a provider of a potentially unbounded number of sequenced elements, publishing them according to the demand received from its Subscriber(s).
In response to a call to Publisher.subscribe(Subscriber) the possible invocation sequences for methods on the Subscriber are given by the following protocol:
onSubscribe onNext* (onError | onComplete)?
This means that onSubscribe is always signalled,
followed by a possibly unbounded number of onNext signals (as requested by Subscriber) followed by an onError signal if there is a failure, or an onComplete signal when no more elements are available—all as long as the Subscription is not cancelled.
- The specifications below use binding words in capital letters from https://www.ietf.org/rfc/rfc2119.txt
- The terms
emit,signalorsendare interchangeable. The specifications below will usesignal. - The terms
synchronouslyorsynchronousrefer to executing in the callingThread. - The term "return normally" means "only throws exceptions that are explicitly allowed by the rule".
1. Publisher (Code)
public interface Publisher<T> {
public void subscribe(Subscriber<? super T> s);
}| ID | Rule |
|---|---|
| 1 | The total number of onNext signals sent by a Publisher to a Subscriber MUST be less than or equal to the total number of elements requested by that Subscriber´s Subscription at all times. |
| 2 | A Publisher MAY signal less onNext than requested and terminate the Subscription by calling onComplete or onError. |
| 3 | onSubscribe, onNext, onError and onComplete signaled to a Subscriber MUST be signaled sequentially (no concurrent notifications). |
| 4 | If a Publisher fails it MUST signal an onError. |
| 5 | If a Publisher terminates successfully (finite stream) it MUST signal an onComplete. |
| 6 | If a Publisher signals either onError or onComplete on a Subscriber, that Subscriber’s Subscription MUST be considered cancelled. |
| 7 | Once a terminal state has been signaled (onError, onComplete) it is REQUIRED that no further signals occur. |
| 8 | If a Subscription is cancelled its Subscriber MUST eventually stop being signaled. |
| 9 | Publisher.subscribe MUST call onSubscribe on the provided Subscriber prior to any other signals to that Subscriber and MUST return normally, except when the provided Subscriber is null in which case it MUST throw a java.lang.NullPointerException to the caller, for all other situations the only legal way to signal failure (or reject the Subscriber) is by calling onError (after calling onSubscribe). |
| 10 | Publisher.subscribe MAY be called as many times as wanted but MUST be with a different Subscriber each time [see 2.12]. |
| 11 | A Publisher MAY support multiple Subscribers and decides whether each Subscription is unicast or multicast. |
[1] : A stateful Publisher can be overwhelmed, bounded by a finite number of underlying resources, exhausted, shut-down or in a failed state.
2. Subscriber (Code)
public interface Subscriber<T> {
public void onSubscribe(Subscription s);
public void onNext(T t);
public void onError(Throwable t);
public void onComplete();
}| ID | Rule |
|---|---|
| 1 | A Subscriber MUST signal demand via Subscription.request(long n) to receive onNext signals. |
| 2 | If a Subscriber suspects that its processing of signals will negatively impact its Publisher's responsivity, it is RECOMMENDED that it asynchronously dispatches its signals. |
| 3 | Subscriber.onComplete() and Subscriber.onError(Throwable t) MUST NOT call any methods on the Subscription or the Publisher. |
| 4 | Subscriber.onComplete() and Subscriber.onError(Throwable t) MUST consider the Subscription cancelled after having received the signal. |
| 5 | A Subscriber MUST call Subscription.cancel() on the given Subscription after an onSubscribe signal if it already has an active Subscription. |
| 6 | A Subscriber MUST call Subscription.cancel() if it is no longer valid to the Publisher without the Publisher having signaled onError or onComplete. |
| 7 | A Subscriber MUST ensure that all calls on its Subscription take place from the same thread or provide for respective external synchronization. |
| 8 | A Subscriber MUST be prepared to receive one or more onNext signals after having called Subscription.cancel() if there are still requested elements pending [see 3.12]. Subscription.cancel() does not guarantee to perform the underlying cleaning operations immediately. |
| 9 | A Subscriber MUST be prepared to receive an onComplete signal with or without a preceding Subscription.request(long n) call. |
| 10 | A Subscriber MUST be prepared to receive an onError signal with or without a preceding Subscription.request(long n) call. |
| 11 | A Subscriber MUST make sure that all calls on its onXXX methods happen-before [1] the processing of the respective signals. I.e. the Subscriber must take care of properly publishing the signal to its processing logic. |
| 12 | Subscriber.onSubscribe MUST be called at most once for a given Subscriber (based on object equality). |
| 13 | Calling onSubscribe, onNext, onError or onComplete MUST return normally except when any provided parameter is null in which case it MUST throw a java.lang.NullPointerException to the caller, for all other situations the only legal way for a Subscriber to signal failure is by cancelling its Subscription. In the case that this rule is violated, any associated Subscription to the Subscriber MUST be considered as cancelled, and the caller MUST raise this error condition in a fashion that is adequate for the runtime environment. |
[1] : See JMM definition of Happen-Before in section 17.4.5. on http://docs.oracle.com/javase/specs/jls/se7/html/jls-17.html
3. Subscription (Code)
public interface Subscription {
public void request(long n);
public void cancel();
}| ID | Rule |
|---|---|
| 1 | Subscription.request and Subscription.cancel MUST only be called inside of its Subscriber context. A Subscription represents the unique relationship between a Subscriber and a Publisher [see 2.12]. |
| 2 | The Subscription MUST allow the Subscriber to call Subscription.request synchronously from within onNext or onSubscribe. |
| 3 | Subscription.request MUST place an upper bound on possible synchronous recursion between Publisher and Subscriber[1]. |
| 4 | Subscription.request SHOULD respect the responsivity of its caller by returning in a timely manner[2]. |
| 5 | Subscription.cancel MUST respect the responsivity of its caller by returning in a timely manner[2], MUST be idempotent and MUST be thread-safe. |
| 6 | After the Subscription is cancelled, additional Subscription.request(long n) MUST be NOPs. |
| 7 | After the Subscription is cancelled, additional Subscription.cancel() MUST be NOPs. |
| 8 | While the Subscription is not cancelled, Subscription.request(long n) MUST register the given number of additional elements to be produced to the respective subscriber. |
| 9 | While the Subscription is not cancelled, Subscription.request(long n) MUST signal onError with a java.lang.IllegalArgumentException if the argument is <= 0. The cause message MUST include a reference to this rule and/or quote the full rule. |
| 10 | While the Subscription is not cancelled, Subscription.request(long n) MAY synchronously call onNext on this (or other) subscriber(s). |
| 11 | While the Subscription is not cancelled, Subscription.request(long n) MAY synchronously call onComplete or onError on this (or other) subscriber(s). |
| 12 | While the Subscription is not cancelled, Subscription.cancel() MUST request the Publisher to eventually stop signaling its Subscriber. The operation is NOT REQUIRED to affect the Subscription immediately. |
| 13 | While the Subscription is not cancelled, Subscription.cancel() MUST request the Publisher to eventually drop any references to the corresponding subscriber. Re-subscribing with the same Subscriber object is discouraged [see 2.12], but this specification does not mandate that it is disallowed since that would mean having to store previously cancelled subscriptions indefinitely. |
| 14 | While the Subscription is not cancelled, calling Subscription.cancel MAY cause the Publisher, if stateful, to transition into the shut-down state if no other Subscription exists at this point [see 1.9]. |
| 15 | Calling Subscription.cancel MUST return normally. The only legal way to signal failure to a Subscriber is via the onError method. |
| 16 | Calling Subscription.request MUST return normally. The only legal way to signal failure to a Subscriber is via the onError method. |
| 17 | A Subscription MUST support an unbounded number of calls to request and MUST support a demand (sum requested - sum delivered) up to 2^63-1 (java.lang.Long.MAX_VALUE). A demand equal or greater than 2^63-1 (java.lang.Long.MAX_VALUE) MAY be considered by the Publisher as “effectively unbounded”[3]. |
[1] : An example for undesirable synchronous, open recursion would be Subscriber.onNext -> Subscription.request -> Subscriber.onNext -> …, as it very quickly would result in blowing the calling Thread´s stack.
[2] : Avoid heavy computations and other things that would stall the caller´s thread of execution
[3] : As it is not feasibly reachable with current or foreseen hardware within a reasonable amount of time (1 element per nanosecond would take 292 years) to fulfill a demand of 2^63-1, it is allowed for a Publisher to stop tracking demand beyond this point.
A Subscription is shared by exactly one Publisher and one Subscriber for the purpose of mediating the data exchange between this pair. This is the reason why the subscribe() method does not return the created Subscription, but instead returns void; the Subscription is only passed to the Subscriber via the onSubscribe callback.
4.Processor (Code)
public interface Processor<T, R> extends Subscriber<T>, Publisher<R> {
}| ID | Rule |
|---|---|
| 1 | A Processor represents a processing stage—which is both a Subscriber and a Publisher and MUST obey the contracts of both. |
| 2 | A Processor MAY choose to recover an onError signal. If it chooses to do so, it MUST consider the Subscription cancelled, otherwise it MUST propagate the onError signal to its Subscribers immediately. |
While not mandated, it can be a good idea to cancel a Processors upstream Subscription when/if its last Subscriber cancels their Subscription,
to let the cancellation signal propagate upstream.
The Reactive Streams API prescribes that all processing of elements (onNext) or termination signals (onError, onComplete) MUST NOT block the Publisher. However, each of the on* handlers can process the events synchronously or asynchronously.
Take this example:
nioSelectorThreadOrigin map(f) filter(p) consumeTo(toNioSelectorOutput)
It has an async origin and an async destination. Let's assume that both origin and destination are selector event loops. The Subscription.request(n) must be chained from the destination to the origin. This is now where each implementation can choose how to do this.
The following uses the pipe | character to signal async boundaries (queue and schedule) and R# to represent resources (possibly threads).
nioSelectorThreadOrigin | map(f) | filter(p) | consumeTo(toNioSelectorOutput)
-------------- R1 ---- | - R2 - | -- R3 --- | ---------- R4 ----------------
In this example each of the 3 consumers, map, filter and consumeTo asynchronously schedule the work. It could be on the same event loop (trampoline), separate threads, whatever.
nioSelectorThreadOrigin map(f) filter(p) | consumeTo(toNioSelectorOutput)
------------------- R1 ----------------- | ---------- R2 ----------------
Here it is only the final step that asynchronously schedules, by adding work to the NioSelectorOutput event loop. The map and filter steps are synchronously performed on the origin thread.
Or another implementation could fuse the operations to the final consumer:
nioSelectorThreadOrigin | map(f) filter(p) consumeTo(toNioSelectorOutput)
--------- R1 ---------- | ------------------ R2 -------------------------
All of these variants are "asynchronous streams". They all have their place and each has different tradeoffs including performance and implementation complexity.
The Reactive Streams contract allows implementations the flexibility to manage resources and scheduling and mix asynchronous and synchronous processing within the bounds of a non-blocking, asynchronous, push-based stream.
In order to allow fully asynchronous implementations of all participating API elements—Publisher/Subscription/Subscriber/Processor—all methods defined by these interfaces return void.
One of the underlying design principles is that all buffer sizes are to be bounded and these bounds must be known and controlled by the subscribers. These bounds are expressed in terms of element count (which in turn translates to the invocation count of onNext). Any implementation that aims to support infinite streams (especially high output rate streams) needs to enforce bounds all along the way to avoid out-of-memory errors and constrain resource usage in general.
Since back-pressure is mandatory the use of unbounded buffers can be avoided. In general, the only time when a queue might grow without bounds is when the publisher side maintains a higher rate than the subscriber for an extended period of time, but this scenario is handled by backpressure instead.
Queue bounds can be controlled by a subscriber signaling demand for the appropriate number of elements. At any point in time the subscriber knows:
- the total number of elements requested:
P - the number of elements that have been processed:
N
Then the maximum number of elements that may arrive—until more demand is signaled to the Publisher—is P - N. In the case that the subscriber also knows the number of elements B in its input buffer then this bound can be refined to P - B - N.
These bounds must be respected by a publisher independent of whether the source it represents can be backpressured or not. In the case of sources whose production rate cannot be influenced—for example clock ticks or mouse movement—the publisher must choose to either buffer or drop elements to obey the imposed bounds.
Subscribers signaling a demand for one element after the reception of an element effectively implement a Stop-and-Wait protocol where the demand signal is equivalent to acknowledgement. By providing demand for multiple elements the cost of acknowledgement is amortized. It is worth noting that the subscriber is allowed to signal demand at any point in time, allowing it to avoid unnecessary delays between the publisher and the subscriber (i.e. keeping its input buffer filled without having to wait for full round-trips).
This project is a collaboration between engineers from Kaazing, Netflix, Pivotal, RedHat, Twitter, Typesafe and many others. The code is offered to the Public Domain in order to allow free use by interested parties who want to create compatible implementations. For details see COPYING.