Reactor Flux 操作符

雨点打透心脏的1/2处 2021-09-19 20:22 841阅读 0赞

# Reactor Flux 操作符 #

Flux常见操作符。

## 1 buffer ##

//每次缓存一定数量的元素到List buckets里，并push出去
            Flux.range(1, 30).buffer(20).subscribe(System.out::println);
            System.out.println("---------- 分割线1 ----------");
    
            //每次缓存一定数量，并跳过一定数量的元素到List buckets里，并push出去
            Flux.range(1, 30).buffer(10,10).subscribe(System.out::println);
            System.out.println("---------- 分割线2 ----------");
    
            //每次缓存一定数量，并跳过一定数量的元素到指定的Set buckets里，并push出去
            Flux.range(1, 30).buffer(20,20, HashSet::new).subscribe(System.out::println);
            System.out.println("---------- 分割线3 ----------");
    
            //指定时间内，每次缓存一定数量的元素到List buckets里，并push出去
            Flux.intervalMillis(100).bufferMillis(500).take(3).toStream().forEach(System.out::println);
            System.out.println("---------- 分割线4 ----------");
    
            //缓存元素到List buckets里当符合条件时，把元素push出去
            Flux.range(1, 10).bufferUntil(i -> i % 2 == 0).subscribe(System.out::println);
            System.out.println("---------- 分割线5 ----------");
    
            //把符合条件的缓存元素到List buckets里并把元素push出去
            Flux.range(1, 10).bufferWhile(i -> i % 2 == 0).subscribe(System.out::println);

执行结果：

---------- buffer ----------
    11:02:21.193 [main] DEBUG reactor.util.Loggers$LoggerFactory - Using Slf4j logging framework
    [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20]
    [21, 22, 23, 24, 25, 26, 27, 28, 29, 30]
    ---------- 分割线1 ----------
    [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
    [11, 12, 13, 14, 15, 16, 17, 18, 19, 20]
    [21, 22, 23, 24, 25, 26, 27, 28, 29, 30]
    ---------- 分割线2 ----------
    [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20]
    [21, 22, 23, 24, 25, 26, 27, 28, 29, 30]
    ---------- 分割线3 ----------
    [0, 1, 2, 3]
    [4, 5, 6, 7, 8]
    [9, 10, 11, 12, 13]
    ---------- 分割线4 ----------
    [1, 2]
    [3, 4]
    [5, 6]
    [7, 8]
    [9, 10]
    ---------- 分割线5 ----------
    [2]
    [4]
    [6]
    [8]
    [10]

## 2 filter ##

Flux.range(1, 10).filter(i -> i % 2 == 0).subscribe(System.out::println);

执行结果：

2
    4
    6
    8
    10

## 3 window ##

*  public final Flux<Flux> window(int maxSize)

//从起始位置开始将source sequence拆分成固定大小的flux
    Flux<Flux<Integer>> window = Flux.range(1, 10)
            .window(3);
    //subscribe1
    window.subscribe(integerFlux -> { 
        System.out.println("+++++分隔符+++++");
        integerFlux.subscribe(System.out::println);
    });
    System.out.println("---------- 分割线1 ----------");
    
    //subscribe2
    window.toStream().forEach(integerFlux -> { 
        System.out.println("+++++分隔符+++++");
        integerFlux.subscribe(System.out::println);
    });
    System.out.println("---------- 分割线2 ----------");

执行结果：

+++++分隔符+++++
    1
    2
    3
    +++++分隔符+++++
    4
    5
    6
    +++++分隔符+++++
    7
    8
    9
    +++++分隔符+++++
    10

*  public final Flux<Flux> windowMillis(long timespan)

//在一定时间内将flux sequence拆分成连接的不同的flux
    Flux.intervalMillis(1)
            .windowMillis(3)
            .take(3)
            .toStream()
            .forEach(longFlux -> { 
                System.out.println("+++++分隔符+++++");
                longFlux.subscribe(System.out::println);
            });

执行结果：

+++++分隔符+++++
    0
    1
    +++++分隔符+++++
    2
    3
    4
    +++++分隔符+++++
    5
    6

## 4 zipWith ##

> zipWith 将多个元素合并成元组，并从每个source产生元素直至结束

> public final Flux<Tuple2<T, T2>> zipWith(Publisher<? extends T2> source2)

final AtomicInteger index = new AtomicInteger();
    Flux<Tuple2<String, String>> tuple2Flux = Flux./just/(“a”, “b”)
            .zipWith(Flux./just/(“c”, “d”));
    //subscribe1
    tuple2Flux.subscribe(System./out/::println);
    System./out/.println(“————— 分割线1 —————“);
    
    //subscribe2
    tuple2Flux.subscribe(tupleFlux -> { 
        System./out/.println(“t1—————“ + index + “—————>” + tupleFlux.getT1());
        System./out/.println(“t2—————“ + index + “—————>” + tupleFlux.getT2());
        index.incrementAndGet();
    });
    System./out/.println(“————— 分割线2 —————“);

执行结果：

[a,c]
    [b,d]
    ---------- 分割线1 ----------
    t1--0-->a
    t2--0-->c
    t1--1-->b
    t2--1-->d
    ---------- 分割线2 ----------

> public final <T2, V> Flux zipWith(Publisher<? extends T2> source2,  
> final BiFunction<? super T, ? super T2, ? extends V> combinator)

Flux.just("a", "b")
            .zipWith(Flux.just("c", "d"), (s1, s2) -> String.format("%s-%s", s1, s2))
            .subscribe(System.out::println);

执行结果：

a-c
    b-d

## 5 take ##

Flux.range(1, 20).take(10).subscribe(System.out::println);
    Flux.range(1, 20).takeLast(10).subscribe(System.out::println);
    Flux.range(1, 20).takeWhile(i -> i < 10).subscribe(System.out::println);
    Flux.range(1, 20).takeUntil(i -> i == 10).subscribe(System.out::println);

## 6 reduce ##

Flux.just(1, 2, 3).reduce((x, y) -> x + y).subscribe(System.out::println);
    Flux.just(1, 2, 3).reduceWith(() -> 4, (x, y) -> x + y).subscribe(System.out::println);

## 7 merge ##

//将多个源publisher中的元素合并到一个新的publisher序列中 重复的元素不会覆盖
    Flux.merge(Flux.just(0, 1, 2, 3), Flux.just(7, 5, 6), Flux.just(4, 7), Flux.just(4, 7))
            .toStream()
            .forEach(System.out::print);
    System.out.println();
    //将多个源publisher中的元素合并到一个新的publisher序列中 重复的元素不会覆盖
    //新publisher中元素的顺序为老元素的订阅顺序
    Flux<Integer> flux = Flux.mergeSequential(Flux.just(9, 8, 7), Flux.just(0, 1, 2, 3),
            Flux.just(6, 5, 4));
    System.out.println();
    flux.sort().subscribe(System.out::print);
    System.out.println();
    flux.subscribe(System.out::print);

01237564747
    0123456789
    9870123654

## 8 flatmap ##

Flux.just(1, 2)
            .flatMap(x -> Flux.just(x * 10, 100).take(x))
            .toStream()
            .forEach(System.out::println);

## 9 concat ##

Flux.just(1, 2)
            .concat(Flux.just(5, 8))
            .toStream()
            .forEach(System.out::println);
    
    Flux.concat(Mono.just(3), Mono.just(4), Flux.just(1, 2))
            .subscribe(System.out::println);

## 10 concatMap ##

//绑定给定输入序列的动态序列，如flatMap，但保留顺序和串联发射，而不是合并(没有交织)
    Flux.just(5, 10,100)
            .concatMap(x -> Flux.just(x * 10, 100))
            .toStream()
            .forEach(System.out::println);

执行结果：

50
    100
    100
    100
    1000
    100

## 11 combineLatest ##

//构建一个flux，其数据是由所有发布者最近发布的值组合而生成的
    Flux.combineLatest(
            Arrays::toString,
            Flux.just(1, 3, 1, 3, 2),
            Flux.just(1, 2, 3, 4),
            Mono.just(9),
            Flux.just(5, 6),
            Flux.just(7, 8, 9)
    ).toStream().forEach(System.out::println);

执行结果：

[2, 4, 9, 6, 7]
    [2, 4, 9, 6, 8]
    [2, 4, 9, 6, 9]

## 12 综合应用 ##

private static void compositeOperate() throws InterruptedException { 
        Set<Integer> set = new HashSet<>();
        set.add(1);
        set.add(2);
        set.add(3);
        Flux.fromIterable(set)
                .delayElements(Duration.ofMillis(1))
                .doOnNext(FluxOperators::doSomeThing) //the callback to call on {@link Subscriber#onNext}
                .publishOn(Schedulers.elastic())
                .map(d -> d * 2)
                .take(3)
                .doOnCancel(() -> System.out.println("操作被取消了"))
                .doOnComplete(() -> System.out.println("正常结束了"))
                .onErrorResumeWith(e -> e instanceof RuntimeException, e -> { 
                    System.out.println("出现了 RuntimeException");
                    return Flux.error(e);
                })//某种异常时才处理
                .doOnError(e -> System.out.println("出现了 Exception"))
                .doAfterTerminate(() -> System.out.println("不管正常异常，反正是结束了"))
                .onErrorReturn(444)
                .doFinally(signalType -> System.out.println("finally--->" + signalType))
                .subscribe(System.out::println);
    
        TimeUnit.SECONDS.sleep(5);
    }
    
    
    public static Integer doSomeThing(Integer integer) { 
        //自行调节看效果
        if (1 != 1)
            throw new RuntimeException("a RuntimeException error");
        else if (2 != 2)
            throw new NullPointerException("a NullPointerException error");
        return integer + 1;
    }

执行结果：

2
    4
    6
    正常结束了
    finally--->onComplete
    不管正常异常，反正是结束了