How to combine two results of CompletableFuture when one depends on another?

user_14643902 0
Viewed 820

There are 2 services:

  1. Service1 returns list of ids1;
  2. Service2 returns list of ids2 by id1:

How can I group result by id1 -> [id2, id2, id2, ...] using CompletableFuture?
I was thinking to put it all to the HashMap but I don't understand how to pass ids1 to the second Future. I was thinking of using thenCombine(ids1Future, (v1, v2) -> {...}), but it seems quite ugly to me. Are there any "best practices" to do it?

Here is not async way:

Map<ID1, List<ID2>> idsMap = new HashMap();
      
List<IDS1> ids1List = service1.service.find();
      
for (ID1 id1: IDS1) {
    List<IDS2> ids2List = service2.findById(id1);
    idsMap.put(id1, ids2List);
}
service2.process(idsMap);

Trying to do it async:

CompletableFuture<List<IDS1>> ids1Future
        = CompletableFuture.supplyAsync(() -> service.find());

ids1Future.thenApply((ids1) -> {
    CompletableFuture<List<IDS2>> listFutureIds2
            = ids1.stream().map(id1
                    -> CompletableFuture.supplyAsync(()
                    -> service2.getById(id1)))
                    .collect(Collectors.toList());

    CompletableFuture<Void> allFuturesDone
            = CompletableFuture.allOf(
                    listFutureIds2.toArray(
                            new CompletableFuture[listFutureIds2.size()]));
    allFuturesDone.thenApply(v
            -> list.stream()
                    .map(CompletableFuture::join)
                    .collect(toList()));

});
completable-future java.util.concurrent java
2 Answers

This is an example of how to use Fork/Join to convert a list of chars to a list of strings, where each string is a replication of N times of the char. You can adjust this example to your case.

I hope that this is what you are looking for.

import java.util.Collection;
import java.util.List;
import java.util.concurrent.ForkJoinPool;
import java.util.concurrent.ForkJoinTask;
import java.util.concurrent.RecursiveTask;
import java.util.stream.Collectors;

public class ForkJoinExample {
    
    public static class RepeatTask extends RecursiveTask<List<String>> {

        private final List<Character> chars;
        private final int repeatCount;

        public RepeatTask(List<Character> chars, int repeatCount) {
            this.chars = chars;
            this.repeatCount = repeatCount;
        }

        public RepeatTask(Character c, int repeatCount) {
            this.chars = List.of(c);
            this.repeatCount = repeatCount;
        }

        @Override
        protected List<String> compute() {
            if (chars.size() > 1) {
                ForkJoinTask.invokeAll(chars.stream().map(c -> new RepeatTask(c, repeatCount)).collect(Collectors.toList()));
                return ForkJoinTask.invokeAll(chars.stream().map(c -> new RepeatTask(c, repeatCount)).collect(Collectors.toList()))
                        .stream()
                        .map(ForkJoinTask::join)
                        .flatMap(Collection::stream)
                        .collect(Collectors.toList());
            }
            return List.of(String.valueOf(chars.get(0)).repeat(Math.max(0, repeatCount)));
        }

    }

    public static void main(String[] args) {
        ForkJoinPool commonPool = ForkJoinPool.commonPool();
        RepeatTask task = new RepeatTask(List.of('a', 'b', 'c'), 4);
        commonPool.execute(task);
        System.out.println(task.join());
    }
}

The result is this one: [aaaa, bbbb, cccc]

user_17655090

I hope I understood you correctly, so here it is:

   CompletableFuture<List<IDS1>> ids1Future = CompletableFuture.supplyAsync(service::find);

   CompletableFuture<Map<IDS1, List<IDS2>>> map = ids1Future.thenApply(listIDS1 -> {
        List<Entry<IDS1, CompletableFuture<List<IDS2>>>> list =
            listIDS1
                .stream()
                .map(ids1 -> new SimpleEntry<>(ids1, CompletableFuture.supplyAsync(() -> service2.getById(ids1))))
                .collect(Collectors.toList());

        return list.stream()
                   .map(entry -> new SimpleEntry<>(entry.getKey(), entry.getValue().join()))
                   .collect(Collectors.toMap(Entry::getKey, Entry::getValue));

    });

    Map<IDS1, List<IDS2>> result = map.join();
user_10593720
Related