reactor/reactor-core

EmitterProcessor queue not cleared on immediate cancellation

Open

#3,359 opened on Feb 20, 2023

View on GitHub
 (0 comments) (0 reactions) (0 assignees)Java (1,164 forks)batch import
area/doOnDiscardhelp wantedtype/bug

Repository metrics

Stars
 (4,748 stars)
PR merge metrics
 (Avg merge 16d 15h) (25 merged PRs in 30d)

Description

EmitterProcessor#remove has support to discard the buffer once the last subscriber is gone. However, it returns early if the subscribers array is already EMPTY. For cases where the subscription is immediately cancelled (e.g. .take(0)) this may leak the buffered elements.

Here is a repro case:

@Test
void shouldClearBufferOnCancellation() {
  BlockingQueue<Integer> queue = new LinkedBlockingQueue<>();

  Many<Integer> sink = Sinks.many().unicast().onBackpressureBuffer(queue);

  sink.tryEmitNext(1);
  sink.tryEmitNext(2);
  sink.tryEmitNext(3);

  assertThat(queue).hasSize(3);

  sink.asFlux().take(0).blockLast();

  assertThat(queue).isEmpty();
}

As a workaround .takeWhile(x -> false) does make the test pass for example.

Contributor guide

Research direction
Examine the EmitterProcessor#remove method, particularly the early return when the subscribers array is EMPTY, and identify how buffer discarding logic is skipped when the last subscriber is cancelled immediately.
Tech stack
java
Domain
backend
Issue type
Bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-3 hours
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
JavaReactive programming concepts
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
70